Add files via upload

arch/arm/Kconfig

@@ -0,0 +1,2286 @@
+config ARM
+	bool
+	default y
+	select ARCH_CLOCKSOURCE_DATA
+	select ARCH_HAS_DEVMEM_IS_ALLOWED
+	select ARCH_HAS_ELF_RANDOMIZE
+	select ARCH_HAS_TICK_BROADCAST if GENERIC_CLOCKEVENTS_BROADCAST
+	select ARCH_HAVE_CUSTOM_GPIO_H
+	select ARCH_HAS_GCOV_PROFILE_ALL
+	select ARCH_MIGHT_HAVE_PC_PARPORT
+	select ARCH_SUPPORTS_ATOMIC_RMW
+	select ARCH_HAS_PTE_SPECIAL if (ARM_LPAE || CPU_V7 || CPU_V7M || CPU_V6 || CPUV6K)
+	select ARCH_USE_BUILTIN_BSWAP
+	select ARCH_USE_CMPXCHG_LOCKREF
+	select ARCH_WANT_IPC_PARSE_VERSION
+	select BUILDTIME_EXTABLE_SORT if MMU
+	select CLONE_BACKWARDS
+	select CPU_PM if (SUSPEND || CPU_IDLE)
+	select DCACHE_WORD_ACCESS if HAVE_EFFICIENT_UNALIGNED_ACCESS
+	select EDAC_SUPPORT
+	select EDAC_ATOMIC_SCRUB
+	select GENERIC_ALLOCATOR
+	select GENERIC_ATOMIC64 if (CPU_V7M || CPU_V6 || !CPU_32v6K || !AEABI)
+	select GENERIC_CLOCKEVENTS_BROADCAST if SMP
+	select GENERIC_EARLY_IOREMAP
+	select GENERIC_IDLE_POLL_SETUP
+	select GENERIC_IRQ_PROBE
+	select GENERIC_IRQ_SHOW
+	select GENERIC_IRQ_SHOW_LEVEL
+	select GENERIC_PCI_IOMAP
+	select GENERIC_SCHED_CLOCK
+	select GENERIC_SMP_IDLE_THREAD
+	select GENERIC_STRNCPY_FROM_USER
+	select GENERIC_STRNLEN_USER
+	select HANDLE_DOMAIN_IRQ
+	select HARDIRQS_SW_RESEND
+	select HAVE_ARCH_AUDITSYSCALL if (AEABI && !OABI_COMPAT)
+	select HAVE_ARCH_BITREVERSE if (CPU_32v7M || CPU_32v7) && !CPU_32v6
+	select HAVE_ARCH_HARDENED_USERCOPY
+	select HAVE_ARCH_JUMP_LABEL if !XIP_KERNEL && !CPU_ENDIAN_BE32 && MMU
+	select HAVE_ARCH_KGDB if !CPU_ENDIAN_BE32 && MMU
+	select HAVE_ARCH_MMAP_RND_BITS if MMU
+	select HAVE_ARCH_SECCOMP_FILTER if (AEABI && !OABI_COMPAT)
+	select HAVE_ARCH_TRACEHOOK
+	select HAVE_ARM_SMCCC if CPU_V7
+	select HAVE_CBPF_JIT
+	select HAVE_CC_STACKPROTECTOR
+	select HAVE_CONTEXT_TRACKING
+	select HAVE_C_RECORDMCOUNT
+	select HAVE_DEBUG_KMEMLEAK
+	select HAVE_DMA_API_DEBUG
+	select HAVE_DMA_CONTIGUOUS if MMU
+	select HAVE_DYNAMIC_FTRACE if (!XIP_KERNEL) && !CPU_ENDIAN_BE32 && MMU
+	select HAVE_EFFICIENT_UNALIGNED_ACCESS if (CPU_V6 || CPU_V6K || CPU_V7) && MMU
+	select HAVE_EXIT_THREAD
+	select HAVE_FTRACE_MCOUNT_RECORD if (!XIP_KERNEL)
+	select HAVE_FUNCTION_GRAPH_TRACER if (!THUMB2_KERNEL)
+	select HAVE_FUNCTION_TRACER if (!XIP_KERNEL)
+	select HAVE_GCC_PLUGINS
+	select HAVE_GENERIC_DMA_COHERENT
+	select HAVE_HW_BREAKPOINT if (PERF_EVENTS && (CPU_V6 || CPU_V6K || CPU_V7))
+	select HAVE_IDE if PCI || ISA || PCMCIA
+	select HAVE_IRQ_TIME_ACCOUNTING
+	select HAVE_KERNEL_GZIP
+	select HAVE_KERNEL_LZ4
+	select HAVE_KERNEL_LZMA
+	select HAVE_KERNEL_LZO
+	select HAVE_KERNEL_XZ
+	select HAVE_KPROBES if !XIP_KERNEL && !CPU_ENDIAN_BE32 && !CPU_V7M
+	select HAVE_KRETPROBES if (HAVE_KPROBES)
+	select HAVE_MEMBLOCK
+	select HAVE_MOD_ARCH_SPECIFIC
+	select HAVE_NMI
+	select HAVE_OPROFILE if (HAVE_PERF_EVENTS)
+	select HAVE_OPTPROBES if !THUMB2_KERNEL
+	select HAVE_PERF_EVENTS
+	select HAVE_PERF_REGS
+	select HAVE_PERF_USER_STACK_DUMP
+	select HAVE_RCU_TABLE_FREE if (SMP && ARM_LPAE)
+	select HAVE_REGS_AND_STACK_ACCESS_API
+	select HAVE_SYSCALL_TRACEPOINTS
+	select HAVE_UID16
+	select HAVE_VIRT_CPU_ACCOUNTING_GEN
+	select IRQ_FORCED_THREADING
+	select MODULES_USE_ELF_REL
+	select NO_BOOTMEM
+	select OF_EARLY_FLATTREE if OF
+	select OF_RESERVED_MEM if OF
+	select OLD_SIGACTION
+	select OLD_SIGSUSPEND3
+	select PERF_USE_VMALLOC
+	select RTC_LIB
+	select SYS_SUPPORTS_APM_EMULATION
+	# Above selects are sorted alphabetically; please add new ones
+	# according to that.  Thanks.
+	help
+	  The ARM series is a line of low-power-consumption RISC chip designs
+	  licensed by ARM Ltd and targeted at embedded applications and
+	  handhelds such as the Compaq IPAQ.  ARM-based PCs are no longer
+	  manufactured, but legacy ARM-based PC hardware remains popular in
+	  Europe.  There is an ARM Linux project with a web page at
+	  <http://www.arm.linux.org.uk/>.
+
+config ARM_HAS_SG_CHAIN
+	select ARCH_HAS_SG_CHAIN
+	bool
+
+config NEED_SG_DMA_LENGTH
+	bool
+
+config ARM_DMA_USE_IOMMU
+	bool
+	select ARM_HAS_SG_CHAIN
+	select NEED_SG_DMA_LENGTH
+
+if ARM_DMA_USE_IOMMU
+
+config ARM_DMA_IOMMU_ALIGNMENT
+	int "Maximum PAGE_SIZE order of alignment for DMA IOMMU buffers"
+	range 4 9
+	default 8
+	help
+	  DMA mapping framework by default aligns all buffers to the smallest
+	  PAGE_SIZE order which is greater than or equal to the requested buffer
+	  size. This works well for buffers up to a few hundreds kilobytes, but
+	  for larger buffers it just a waste of address space. Drivers which has
+	  relatively small addressing window (like 64Mib) might run out of
+	  virtual space with just a few allocations.
+
+	  With this parameter you can specify the maximum PAGE_SIZE order for
+	  DMA IOMMU buffers. Larger buffers will be aligned only to this
+	  specified order. The order is expressed as a power of two multiplied
+	  by the PAGE_SIZE.
+
+endif
+
+config MIGHT_HAVE_PCI
+	bool
+
+config SYS_SUPPORTS_APM_EMULATION
+	bool
+
+config HAVE_TCM
+	bool
+	select GENERIC_ALLOCATOR
+
+config HAVE_PROC_CPU
+	bool
+
+config NO_IOPORT_MAP
+	bool
+
+config EISA
+	bool
+	---help---
+	  The Extended Industry Standard Architecture (EISA) bus was
+	  developed as an open alternative to the IBM MicroChannel bus.
+
+	  The EISA bus provided some of the features of the IBM MicroChannel
+	  bus while maintaining backward compatibility with cards made for
+	  the older ISA bus.  The EISA bus saw limited use between 1988 and
+	  1995 when it was made obsolete by the PCI bus.
+
+	  Say Y here if you are building a kernel for an EISA-based machine.
+
+	  Otherwise, say N.
+
+config SBUS
+	bool
+
+config STACKTRACE_SUPPORT
+	bool
+	default y
+
+config LOCKDEP_SUPPORT
+	bool
+	default y
+
+config TRACE_IRQFLAGS_SUPPORT
+	bool
+	default !CPU_V7M
+
+config RWSEM_XCHGADD_ALGORITHM
+	bool
+	default y
+
+config ARCH_HAS_ILOG2_U32
+	bool
+
+config ARCH_HAS_ILOG2_U64
+	bool
+
+config ARCH_HAS_BANDGAP
+	bool
+
+config FIX_EARLYCON_MEM
+	def_bool y if MMU
+
+config GENERIC_HWEIGHT
+	bool
+	default y
+
+config GENERIC_CALIBRATE_DELAY
+	bool
+	default y
+
+config ARCH_MAY_HAVE_PC_FDC
+	bool
+
+config ZONE_DMA
+	bool
+
+config NEED_DMA_MAP_STATE
+       def_bool y
+
+config ARCH_SUPPORTS_UPROBES
+	def_bool y
+
+config ARCH_HAS_DMA_SET_COHERENT_MASK
+	bool
+
+config GENERIC_ISA_DMA
+	bool
+
+config FIQ
+	bool
+
+config NEED_RET_TO_USER
+	bool
+
+config ARCH_MTD_XIP
+	bool
+
+config VECTORS_BASE
+	hex
+	default 0xffff0000 if MMU || CPU_HIGH_VECTOR
+	default DRAM_BASE if REMAP_VECTORS_TO_RAM
+	default 0x00000000
+	help
+	  The base address of exception vectors.  This must be two pages
+	  in size.
+
+config ARM_PATCH_PHYS_VIRT
+	bool "Patch physical to virtual translations at runtime" if EMBEDDED
+	default y
+	depends on !XIP_KERNEL && MMU
+	help
+	  Patch phys-to-virt and virt-to-phys translation functions at
+	  boot and module load time according to the position of the
+	  kernel in system memory.
+
+	  This can only be used with non-XIP MMU kernels where the base
+	  of physical memory is at a 16MB boundary.
+
+	  Only disable this option if you know that you do not require
+	  this feature (eg, building a kernel for a single machine) and
+	  you need to shrink the kernel to the minimal size.
+
+config NEED_MACH_IO_H
+	bool
+	help
+	  Select this when mach/io.h is required to provide special
+	  definitions for this platform.  The need for mach/io.h should
+	  be avoided when possible.
+
+config NEED_MACH_MEMORY_H
+	bool
+	help
+	  Select this when mach/memory.h is required to provide special
+	  definitions for this platform.  The need for mach/memory.h should
+	  be avoided when possible.
+
+config PHYS_OFFSET
+	hex "Physical address of main memory" if MMU
+	depends on !ARM_PATCH_PHYS_VIRT
+	default DRAM_BASE if !MMU
+	default 0x00000000 if ARCH_EBSA110 || \
+			ARCH_FOOTBRIDGE || \
+			ARCH_INTEGRATOR || \
+			ARCH_IOP13XX || \
+			ARCH_KS8695 || \
+			ARCH_REALVIEW
+	default 0x10000000 if ARCH_OMAP1 || ARCH_RPC
+	default 0x20000000 if ARCH_S5PV210
+	default 0xc0000000 if ARCH_SA1100
+	help
+	  Please provide the physical address corresponding to the
+	  location of main memory in your system.
+
+config GENERIC_BUG
+	def_bool y
+	depends on BUG
+
+config PGTABLE_LEVELS
+	int
+	default 3 if ARM_LPAE
+	default 2
+
+source "init/Kconfig"
+
+source "kernel/Kconfig.freezer"
+
+menu "System Type"
+
+config MMU
+	bool "MMU-based Paged Memory Management Support"
+	default y
+	help
+	  Select if you want MMU-based virtualised addressing space
+	  support by paged memory management. If unsure, say 'Y'.
+
+config ARCH_MMAP_RND_BITS_MIN
+	default 8
+
+config ARCH_MMAP_RND_BITS_MAX
+	default 14 if PAGE_OFFSET=0x40000000
+	default 15 if PAGE_OFFSET=0x80000000
+	default 16
+
+#
+# The "ARM system type" choice list is ordered alphabetically by option
+# text.  Please add new entries in the option alphabetic order.
+#
+choice
+	prompt "ARM system type"
+	default ARM_SINGLE_ARMV7M if !MMU
+	default ARCH_MULTIPLATFORM if MMU
+
+config ARCH_MULTIPLATFORM
+	bool "Allow multiple platforms to be selected"
+	depends on MMU
+	select ARM_HAS_SG_CHAIN
+	select ARM_PATCH_PHYS_VIRT
+	select AUTO_ZRELADDR
+	select CLKSRC_OF
+	select COMMON_CLK
+	select GENERIC_CLOCKEVENTS
+	select MIGHT_HAVE_PCI
+	select MULTI_IRQ_HANDLER
+	select PCI_DOMAINS if PCI
+	select SPARSE_IRQ
+	select USE_OF
+
+config ARM_SINGLE_ARMV7M
+	bool "ARMv7-M based platforms (Cortex-M0/M3/M4)"
+	depends on !MMU
+	select ARM_NVIC
+	select AUTO_ZRELADDR
+	select CLKSRC_OF
+	select COMMON_CLK
+	select CPU_V7M
+	select GENERIC_CLOCKEVENTS
+	select NO_IOPORT_MAP
+	select SPARSE_IRQ
+	select USE_OF
+
+config ARCH_GEMINI
+	bool "Cortina Systems Gemini"
+	select CLKSRC_MMIO
+	select CPU_FA526
+	select GENERIC_CLOCKEVENTS
+	select GPIOLIB
+	help
+	  Support for the Cortina Systems Gemini family SoCs
+
+config ARCH_EBSA110
+	bool "EBSA-110"
+	select ARCH_USES_GETTIMEOFFSET
+	select CPU_SA110
+	select ISA
+	select NEED_MACH_IO_H
+	select NEED_MACH_MEMORY_H
+	select NO_IOPORT_MAP
+	help
+	  This is an evaluation board for the StrongARM processor available
+	  from Digital. It has limited hardware on-board, including an
+	  Ethernet interface, two PCMCIA sockets, two serial ports and a
+	  parallel port.
+
+config ARCH_EP93XX
+	bool "EP93xx-based"
+	select ARCH_HAS_HOLES_MEMORYMODEL
+	select ARM_AMBA
+	select ARM_PATCH_PHYS_VIRT
+	select ARM_VIC
+	select AUTO_ZRELADDR
+	select CLKDEV_LOOKUP
+	select CLKSRC_MMIO
+	select CPU_ARM920T
+	select GENERIC_CLOCKEVENTS
+	select GPIOLIB
+	help
+	  This enables support for the Cirrus EP93xx series of CPUs.
+
+config ARCH_FOOTBRIDGE
+	bool "FootBridge"
+	select CPU_SA110
+	select FOOTBRIDGE
+	select GENERIC_CLOCKEVENTS
+	select HAVE_IDE
+	select NEED_MACH_IO_H if !MMU
+	select NEED_MACH_MEMORY_H
+	help
+	  Support for systems based on the DC21285 companion chip
+	  ("FootBridge"), such as the Simtec CATS and the Rebel NetWinder.
+
+config ARCH_NETX
+	bool "Hilscher NetX based"
+	select ARM_VIC
+	select CLKSRC_MMIO
+	select CPU_ARM926T
+	select GENERIC_CLOCKEVENTS
+	help
+	  This enables support for systems based on the Hilscher NetX Soc
+
+config ARCH_IOP13XX
+	bool "IOP13xx-based"
+	depends on MMU
+	select CPU_XSC3
+	select NEED_MACH_MEMORY_H
+	select NEED_RET_TO_USER
+	select PCI
+	select PLAT_IOP
+	select VMSPLIT_1G
+	select SPARSE_IRQ
+	help
+	  Support for Intel's IOP13XX (XScale) family of processors.
+
+config ARCH_IOP32X
+	bool "IOP32x-based"
+	depends on MMU
+	select CPU_XSCALE
+	select GPIO_IOP
+	select GPIOLIB
+	select NEED_RET_TO_USER
+	select PCI
+	select PLAT_IOP
+	help
+	  Support for Intel's 80219 and IOP32X (XScale) family of
+	  processors.
+
+config ARCH_IOP33X
+	bool "IOP33x-based"
+	depends on MMU
+	select CPU_XSCALE
+	select GPIO_IOP
+	select GPIOLIB
+	select NEED_RET_TO_USER
+	select PCI
+	select PLAT_IOP
+	help
+	  Support for Intel's IOP33X (XScale) family of processors.
+
+config ARCH_IXP4XX
+	bool "IXP4xx-based"
+	depends on MMU
+	select ARCH_HAS_DMA_SET_COHERENT_MASK
+	select ARCH_SUPPORTS_BIG_ENDIAN
+	select CLKSRC_MMIO
+	select CPU_XSCALE
+	select DMABOUNCE if PCI
+	select GENERIC_CLOCKEVENTS
+	select GPIOLIB
+	select MIGHT_HAVE_PCI
+	select NEED_MACH_IO_H
+	select USB_EHCI_BIG_ENDIAN_DESC
+	select USB_EHCI_BIG_ENDIAN_MMIO
+	help
+	  Support for Intel's IXP4XX (XScale) family of processors.
+
+config ARCH_DOVE
+	bool "Marvell Dove"
+	select CPU_PJ4
+	select GENERIC_CLOCKEVENTS
+	select GPIOLIB
+	select MIGHT_HAVE_PCI
+	select MULTI_IRQ_HANDLER
+	select MVEBU_MBUS
+	select PINCTRL
+	select PINCTRL_DOVE
+	select PLAT_ORION_LEGACY
+	select SPARSE_IRQ
+	select PM_GENERIC_DOMAINS if PM
+	help
+	  Support for the Marvell Dove SoC 88AP510
+
+config ARCH_KS8695
+	bool "Micrel/Kendin KS8695"
+	select CLKSRC_MMIO
+	select CPU_ARM922T
+	select GENERIC_CLOCKEVENTS
+	select GPIOLIB
+	select NEED_MACH_MEMORY_H
+	help
+	  Support for Micrel/Kendin KS8695 "Centaur" (ARM922T) based
+	  System-on-Chip devices.
+
+config ARCH_W90X900
+	bool "Nuvoton W90X900 CPU"
+	select CLKDEV_LOOKUP
+	select CLKSRC_MMIO
+	select CPU_ARM926T
+	select GENERIC_CLOCKEVENTS
+	select GPIOLIB
+	help
+	  Support for Nuvoton (Winbond logic dept.) ARM9 processor,
+	  At present, the w90x900 has been renamed nuc900, regarding
+	  the ARM series product line, you can login the following
+	  link address to know more.
+
+	  <http://www.nuvoton.com/hq/enu/ProductAndSales/ProductLines/
+		ConsumerElectronicsIC/ARMMicrocontroller/ARMMicrocontroller>
+
+config ARCH_LPC32XX
+	bool "NXP LPC32XX"
+	select ARM_AMBA
+	select CLKDEV_LOOKUP
+	select CLKSRC_LPC32XX
+	select COMMON_CLK
+	select CPU_ARM926T
+	select GENERIC_CLOCKEVENTS
+	select GPIOLIB
+	select MULTI_IRQ_HANDLER
+	select SPARSE_IRQ
+	select USE_OF
+	help
+	  Support for the NXP LPC32XX family of processors
+
+config ARCH_PXA
+	bool "PXA2xx/PXA3xx-based"
+	depends on MMU
+	select ARCH_MTD_XIP
+	select ARM_CPU_SUSPEND if PM
+	select AUTO_ZRELADDR
+	select COMMON_CLK
+	select CLKDEV_LOOKUP
+	select CLKSRC_PXA
+	select CLKSRC_MMIO
+	select CLKSRC_OF
+	select CPU_XSCALE if !CPU_XSC3
+	select GENERIC_CLOCKEVENTS
+	select GPIO_PXA
+	select GPIOLIB
+	select HAVE_IDE
+	select IRQ_DOMAIN
+	select MULTI_IRQ_HANDLER
+	select PLAT_PXA
+	select SPARSE_IRQ
+	help
+	  Support for Intel/Marvell's PXA2xx/PXA3xx processor line.
+
+config ARCH_RPC
+	bool "RiscPC"
+	depends on MMU
+	select ARCH_ACORN
+	select ARCH_MAY_HAVE_PC_FDC
+	select ARCH_SPARSEMEM_ENABLE
+	select ARCH_USES_GETTIMEOFFSET
+	select CPU_SA110
+	select FIQ
+	select HAVE_IDE
+	select HAVE_PATA_PLATFORM
+	select ISA_DMA_API
+	select NEED_MACH_IO_H
+	select NEED_MACH_MEMORY_H
+	select NO_IOPORT_MAP
+	help
+	  On the Acorn Risc-PC, Linux can support the internal IDE disk and
+	  CD-ROM interface, serial and parallel port, and the floppy drive.
+
+config ARCH_SA1100
+	bool "SA1100-based"
+	select ARCH_MTD_XIP
+	select ARCH_SPARSEMEM_ENABLE
+	select CLKDEV_LOOKUP
+	select CLKSRC_MMIO
+	select CLKSRC_PXA
+	select CLKSRC_OF if OF
+	select CPU_FREQ
+	select CPU_SA1100
+	select GENERIC_CLOCKEVENTS
+	select GPIOLIB
+	select HAVE_IDE
+	select IRQ_DOMAIN
+	select ISA
+	select MULTI_IRQ_HANDLER
+	select NEED_MACH_MEMORY_H
+	select SPARSE_IRQ
+	help
+	  Support for StrongARM 11x0 based boards.
+
+config ARCH_S3C24XX
+	bool "Samsung S3C24XX SoCs"
+	select ATAGS
+	select CLKDEV_LOOKUP
+	select CLKSRC_SAMSUNG_PWM
+	select GENERIC_CLOCKEVENTS
+	select GPIO_SAMSUNG
+	select GPIOLIB
+	select HAVE_S3C2410_I2C if I2C
+	select HAVE_S3C2410_WATCHDOG if WATCHDOG
+	select HAVE_S3C_RTC if RTC_CLASS
+	select MULTI_IRQ_HANDLER
+	select NEED_MACH_IO_H
+	select SAMSUNG_ATAGS
+	help
+	  Samsung S3C2410, S3C2412, S3C2413, S3C2416, S3C2440, S3C2442, S3C2443
+	  and S3C2450 SoCs based systems, such as the Simtec Electronics BAST
+	  (<http://www.simtec.co.uk/products/EB110ITX/>), the IPAQ 1940 or the
+	  Samsung SMDK2410 development board (and derivatives).
+
+config ARCH_DAVINCI
+	bool "TI DaVinci"
+	select ARCH_HAS_HOLES_MEMORYMODEL
+	select CLKDEV_LOOKUP
+	select CPU_ARM926T
+	select GENERIC_ALLOCATOR
+	select GENERIC_CLOCKEVENTS
+	select GENERIC_IRQ_CHIP
+	select GPIOLIB
+	select HAVE_IDE
+	select USE_OF
+	select ZONE_DMA
+	help
+	  Support for TI's DaVinci platform.
+
+config ARCH_OMAP1
+	bool "TI OMAP1"
+	depends on MMU
+	select ARCH_HAS_HOLES_MEMORYMODEL
+	select ARCH_OMAP
+	select CLKDEV_LOOKUP
+	select CLKSRC_MMIO
+	select GENERIC_CLOCKEVENTS
+	select GENERIC_IRQ_CHIP
+	select GPIOLIB
+	select HAVE_IDE
+	select IRQ_DOMAIN
+	select MULTI_IRQ_HANDLER
+	select NEED_MACH_IO_H if PCCARD
+	select NEED_MACH_MEMORY_H
+	select SPARSE_IRQ
+	help
+	  Support for older TI OMAP1 (omap7xx, omap15xx or omap16xx)
+
+endchoice
+
+menu "Multiple platform selection"
+	depends on ARCH_MULTIPLATFORM
+
+comment "CPU Core family selection"
+
+config ARCH_MULTI_V4
+	bool "ARMv4 based platforms (FA526)"
+	depends on !ARCH_MULTI_V6_V7
+	select ARCH_MULTI_V4_V5
+	select CPU_FA526
+
+config ARCH_MULTI_V4T
+	bool "ARMv4T based platforms (ARM720T, ARM920T, ...)"
+	depends on !ARCH_MULTI_V6_V7
+	select ARCH_MULTI_V4_V5
+	select CPU_ARM920T if !(CPU_ARM7TDMI || CPU_ARM720T || \
+		CPU_ARM740T || CPU_ARM9TDMI || CPU_ARM922T || \
+		CPU_ARM925T || CPU_ARM940T)
+
+config ARCH_MULTI_V5
+	bool "ARMv5 based platforms (ARM926T, XSCALE, PJ1, ...)"
+	depends on !ARCH_MULTI_V6_V7
+	select ARCH_MULTI_V4_V5
+	select CPU_ARM926T if !(CPU_ARM946E || CPU_ARM1020 || \
+		CPU_ARM1020E || CPU_ARM1022 || CPU_ARM1026 || \
+		CPU_XSCALE || CPU_XSC3 || CPU_MOHAWK || CPU_FEROCEON)
+
+config ARCH_MULTI_V4_V5
+	bool
+
+config ARCH_MULTI_V6
+	bool "ARMv6 based platforms (ARM11)"
+	select ARCH_MULTI_V6_V7
+	select CPU_V6K
+
+config ARCH_MULTI_V7
+	bool "ARMv7 based platforms (Cortex-A, PJ4, Scorpion, Krait)"
+	default y
+	select ARCH_MULTI_V6_V7
+	select CPU_V7
+	select HAVE_SMP
+
+config ARCH_MULTI_V6_V7
+	bool
+	select MIGHT_HAVE_CACHE_L2X0
+
+config ARCH_MULTI_CPU_AUTO
+	def_bool !(ARCH_MULTI_V4 || ARCH_MULTI_V4T || ARCH_MULTI_V6_V7)
+	select ARCH_MULTI_V5
+
+endmenu
+
+config ARCH_VIRT
+	bool "Dummy Virtual Machine"
+	depends on ARCH_MULTI_V7
+	select ARM_AMBA
+	select ARM_GIC
+	select ARM_GIC_V2M if PCI
+	select ARM_GIC_V3
+	select ARM_PSCI
+	select HAVE_ARM_ARCH_TIMER
+
+#
+# This is sorted alphabetically by mach-* pathname.  However, plat-*
+# Kconfigs may be included either alphabetically (according to the
+# plat- suffix) or along side the corresponding mach-* source.
+#
+source "arch/arm/mach-mvebu/Kconfig"
+
+source "arch/arm/mach-alpine/Kconfig"
+
+source "arch/arm/mach-artpec/Kconfig"
+
+source "arch/arm/mach-asm9260/Kconfig"
+
+source "arch/arm/mach-at91/Kconfig"
+
+source "arch/arm/mach-axxia/Kconfig"
+
+source "arch/arm/mach-bcm/Kconfig"
+
+source "arch/arm/mach-berlin/Kconfig"
+
+source "arch/arm/mach-clps711x/Kconfig"
+
+source "arch/arm/mach-cns3xxx/Kconfig"
+
+source "arch/arm/mach-davinci/Kconfig"
+
+source "arch/arm/mach-digicolor/Kconfig"
+
+source "arch/arm/mach-dove/Kconfig"
+
+source "arch/arm/mach-ep93xx/Kconfig"
+
+source "arch/arm/mach-footbridge/Kconfig"
+
+source "arch/arm/mach-gemini/Kconfig"
+
+source "arch/arm/mach-highbank/Kconfig"
+
+source "arch/arm/mach-hisi/Kconfig"
+
+source "arch/arm/mach-integrator/Kconfig"
+
+source "arch/arm/mach-iop32x/Kconfig"
+
+source "arch/arm/mach-iop33x/Kconfig"
+
+source "arch/arm/mach-iop13xx/Kconfig"
+
+source "arch/arm/mach-ixp4xx/Kconfig"
+
+source "arch/arm/mach-keystone/Kconfig"
+
+source "arch/arm/mach-ks8695/Kconfig"
+
+source "arch/arm/mach-meson/Kconfig"
+
+source "arch/arm/mach-moxart/Kconfig"
+
+source "arch/arm/mach-aspeed/Kconfig"
+
+source "arch/arm/mach-mv78xx0/Kconfig"
+
+source "arch/arm/mach-imx/Kconfig"
+
+source "arch/arm/mach-mediatek/Kconfig"
+
+source "arch/arm/mach-mxs/Kconfig"
+
+source "arch/arm/mach-netx/Kconfig"
+
+source "arch/arm/mach-nomadik/Kconfig"
+
+source "arch/arm/mach-nspire/Kconfig"
+
+source "arch/arm/plat-omap/Kconfig"
+
+source "arch/arm/mach-omap1/Kconfig"
+
+source "arch/arm/mach-omap2/Kconfig"
+
+source "arch/arm/mach-orion5x/Kconfig"
+
+source "arch/arm/mach-picoxcell/Kconfig"
+
+source "arch/arm/mach-pxa/Kconfig"
+source "arch/arm/plat-pxa/Kconfig"
+
+source "arch/arm/mach-mmp/Kconfig"
+
+source "arch/arm/mach-oxnas/Kconfig"
+
+source "arch/arm/mach-qcom/Kconfig"
+
+source "arch/arm/mach-realview/Kconfig"
+
+source "arch/arm/mach-rockchip/Kconfig"
+
+source "arch/arm/mach-sa1100/Kconfig"
+
+source "arch/arm/mach-socfpga/Kconfig"
+
+source "arch/arm/mach-spear/Kconfig"
+
+source "arch/arm/mach-sti/Kconfig"
+
+source "arch/arm/mach-s3c24xx/Kconfig"
+
+source "arch/arm/mach-s3c64xx/Kconfig"
+
+source "arch/arm/mach-s5pv210/Kconfig"
+
+source "arch/arm/mach-exynos/Kconfig"
+source "arch/arm/plat-samsung/Kconfig"
+
+source "arch/arm/mach-shmobile/Kconfig"
+
+source "arch/arm/mach-sunxi/Kconfig"
+
+source "arch/arm/mach-prima2/Kconfig"
+
+source "arch/arm/mach-tango/Kconfig"
+
+source "arch/arm/mach-tegra/Kconfig"
+
+source "arch/arm/mach-u300/Kconfig"
+
+source "arch/arm/mach-uniphier/Kconfig"
+
+source "arch/arm/mach-ux500/Kconfig"
+
+source "arch/arm/mach-versatile/Kconfig"
+
+source "arch/arm/mach-vexpress/Kconfig"
+source "arch/arm/plat-versatile/Kconfig"
+
+source "arch/arm/mach-vt8500/Kconfig"
+
+source "arch/arm/mach-w90x900/Kconfig"
+
+source "arch/arm/mach-zx/Kconfig"
+
+source "arch/arm/mach-zynq/Kconfig"
+
+# ARMv7-M architecture
+config ARCH_EFM32
+	bool "Energy Micro efm32"
+	depends on ARM_SINGLE_ARMV7M
+	select GPIOLIB
+	help
+	  Support for Energy Micro's (now Silicon Labs) efm32 Giant Gecko
+	  processors.
+
+config ARCH_LPC18XX
+	bool "NXP LPC18xx/LPC43xx"
+	depends on ARM_SINGLE_ARMV7M
+	select ARCH_HAS_RESET_CONTROLLER
+	select ARM_AMBA
+	select CLKSRC_LPC32XX
+	select PINCTRL
+	help
+	  Support for NXP's LPC18xx Cortex-M3 and LPC43xx Cortex-M4
+	  high performance microcontrollers.
+
+config ARCH_STM32
+	bool "STMicrolectronics STM32"
+	depends on ARM_SINGLE_ARMV7M
+	select ARCH_HAS_RESET_CONTROLLER
+	select ARMV7M_SYSTICK
+	select CLKSRC_STM32
+	select PINCTRL
+	select RESET_CONTROLLER
+	select STM32_EXTI
+	help
+	  Support for STMicroelectronics STM32 processors.
+
+config MACH_STM32F429
+	bool "STMicrolectronics STM32F429"
+	depends on ARCH_STM32
+	default y
+
+config ARCH_MPS2
+	bool "ARM MPS2 platform"
+	depends on ARM_SINGLE_ARMV7M
+	select ARM_AMBA
+	select CLKSRC_MPS2
+	help
+	  Support for Cortex-M Prototyping System (or V2M-MPS2) which comes
+	  with a range of available cores like Cortex-M3/M4/M7.
+
+	  Please, note that depends which Application Note is used memory map
+	  for the platform may vary, so adjustment of RAM base might be needed.
+
+# Definitions to make life easier
+config ARCH_ACORN
+	bool
+
+config PLAT_IOP
+	bool
+	select GENERIC_CLOCKEVENTS
+
+config PLAT_ORION
+	bool
+	select CLKSRC_MMIO
+	select COMMON_CLK
+	select GENERIC_IRQ_CHIP
+	select IRQ_DOMAIN
+
+config PLAT_ORION_LEGACY
+	bool
+	select PLAT_ORION
+
+config PLAT_PXA
+	bool
+
+config PLAT_VERSATILE
+	bool
+
+source "arch/arm/firmware/Kconfig"
+
+source arch/arm/mm/Kconfig
+
+config IWMMXT
+	bool "Enable iWMMXt support"
+	depends on CPU_XSCALE || CPU_XSC3 || CPU_MOHAWK || CPU_PJ4 || CPU_PJ4B
+	default y if PXA27x || PXA3xx || ARCH_MMP || CPU_PJ4 || CPU_PJ4B
+	help
+	  Enable support for iWMMXt context switching at run time if
+	  running on a CPU that supports it.
+
+config MULTI_IRQ_HANDLER
+	bool
+	help
+	  Allow each machine to specify it's own IRQ handler at run time.
+
+if !MMU
+source "arch/arm/Kconfig-nommu"
+endif
+
+config PJ4B_ERRATA_4742
+	bool "PJ4B Errata 4742: IDLE Wake Up Commands can Cause the CPU Core to Cease Operation"
+	depends on CPU_PJ4B && MACH_ARMADA_370
+	default y
+	help
+	  When coming out of either a Wait for Interrupt (WFI) or a Wait for
+	  Event (WFE) IDLE states, a specific timing sensitivity exists between
+	  the retiring WFI/WFE instructions and the newly issued subsequent
+	  instructions.  This sensitivity can result in a CPU hang scenario.
+	  Workaround:
+	  The software must insert either a Data Synchronization Barrier (DSB)
+	  or Data Memory Barrier (DMB) command immediately after the WFI/WFE
+	  instruction
+
+config ARM_ERRATA_326103
+	bool "ARM errata: FSR write bit incorrect on a SWP to read-only memory"
+	depends on CPU_V6
+	help
+	  Executing a SWP instruction to read-only memory does not set bit 11
+	  of the FSR on the ARM 1136 prior to r1p0. This causes the kernel to
+	  treat the access as a read, preventing a COW from occurring and
+	  causing the faulting task to livelock.
+
+config ARM_ERRATA_411920
+	bool "ARM errata: Invalidation of the Instruction Cache operation can fail"
+	depends on CPU_V6 || CPU_V6K
+	help
+	  Invalidation of the Instruction Cache operation can
+	  fail. This erratum is present in 1136 (before r1p4), 1156 and 1176.
+	  It does not affect the MPCore. This option enables the ARM Ltd.
+	  recommended workaround.
+
+config ARM_ERRATA_430973
+	bool "ARM errata: Stale prediction on replaced interworking branch"
+	depends on CPU_V7
+	help
+	  This option enables the workaround for the 430973 Cortex-A8
+	  r1p* erratum. If a code sequence containing an ARM/Thumb
+	  interworking branch is replaced with another code sequence at the
+	  same virtual address, whether due to self-modifying code or virtual
+	  to physical address re-mapping, Cortex-A8 does not recover from the
+	  stale interworking branch prediction. This results in Cortex-A8
+	  executing the new code sequence in the incorrect ARM or Thumb state.
+	  The workaround enables the BTB/BTAC operations by setting ACTLR.IBE
+	  and also flushes the branch target cache at every context switch.
+	  Note that setting specific bits in the ACTLR register may not be
+	  available in non-secure mode.
+
+config ARM_ERRATA_458693
+	bool "ARM errata: Processor deadlock when a false hazard is created"
+	depends on CPU_V7
+	depends on !ARCH_MULTIPLATFORM
+	help
+	  This option enables the workaround for the 458693 Cortex-A8 (r2p0)
+	  erratum. For very specific sequences of memory operations, it is
+	  possible for a hazard condition intended for a cache line to instead
+	  be incorrectly associated with a different cache line. This false
+	  hazard might then cause a processor deadlock. The workaround enables
+	  the L1 caching of the NEON accesses and disables the PLD instruction
+	  in the ACTLR register. Note that setting specific bits in the ACTLR
+	  register may not be available in non-secure mode.
+
+config ARM_ERRATA_460075
+	bool "ARM errata: Data written to the L2 cache can be overwritten with stale data"
+	depends on CPU_V7
+	depends on !ARCH_MULTIPLATFORM
+	help
+	  This option enables the workaround for the 460075 Cortex-A8 (r2p0)
+	  erratum. Any asynchronous access to the L2 cache may encounter a
+	  situation in which recent store transactions to the L2 cache are lost
+	  and overwritten with stale memory contents from external memory. The
+	  workaround disables the write-allocate mode for the L2 cache via the
+	  ACTLR register. Note that setting specific bits in the ACTLR register
+	  may not be available in non-secure mode.
+
+config ARM_ERRATA_742230
+	bool "ARM errata: DMB operation may be faulty"
+	depends on CPU_V7 && SMP
+	depends on !ARCH_MULTIPLATFORM
+	help
+	  This option enables the workaround for the 742230 Cortex-A9
+	  (r1p0..r2p2) erratum. Under rare circumstances, a DMB instruction
+	  between two write operations may not ensure the correct visibility
+	  ordering of the two writes. This workaround sets a specific bit in
+	  the diagnostic register of the Cortex-A9 which causes the DMB
+	  instruction to behave as a DSB, ensuring the correct behaviour of
+	  the two writes.
+
+config ARM_ERRATA_742231
+	bool "ARM errata: Incorrect hazard handling in the SCU may lead to data corruption"
+	depends on CPU_V7 && SMP
+	depends on !ARCH_MULTIPLATFORM
+	help
+	  This option enables the workaround for the 742231 Cortex-A9
+	  (r2p0..r2p2) erratum. Under certain conditions, specific to the
+	  Cortex-A9 MPCore micro-architecture, two CPUs working in SMP mode,
+	  accessing some data located in the same cache line, may get corrupted
+	  data due to bad handling of the address hazard when the line gets
+	  replaced from one of the CPUs at the same time as another CPU is
+	  accessing it. This workaround sets specific bits in the diagnostic
+	  register of the Cortex-A9 which reduces the linefill issuing
+	  capabilities of the processor.
+
+config ARM_ERRATA_643719
+	bool "ARM errata: LoUIS bit field in CLIDR register is incorrect"
+	depends on CPU_V7 && SMP
+	default y
+	help
+	  This option enables the workaround for the 643719 Cortex-A9 (prior to
+	  r1p0) erratum. On affected cores the LoUIS bit field of the CLIDR
+	  register returns zero when it should return one. The workaround
+	  corrects this value, ensuring cache maintenance operations which use
+	  it behave as intended and avoiding data corruption.
+
+config ARM_ERRATA_720789
+	bool "ARM errata: TLBIASIDIS and TLBIMVAIS operations can broadcast a faulty ASID"
+	depends on CPU_V7
+	help
+	  This option enables the workaround for the 720789 Cortex-A9 (prior to
+	  r2p0) erratum. A faulty ASID can be sent to the other CPUs for the
+	  broadcasted CP15 TLB maintenance operations TLBIASIDIS and TLBIMVAIS.
+	  As a consequence of this erratum, some TLB entries which should be
+	  invalidated are not, resulting in an incoherency in the system page
+	  tables. The workaround changes the TLB flushing routines to invalidate
+	  entries regardless of the ASID.
+
+config ARM_ERRATA_743622
+	bool "ARM errata: Faulty hazard checking in the Store Buffer may lead to data corruption"
+	depends on CPU_V7
+	depends on !ARCH_MULTIPLATFORM
+	help
+	  This option enables the workaround for the 743622 Cortex-A9
+	  (r2p*) erratum. Under very rare conditions, a faulty
+	  optimisation in the Cortex-A9 Store Buffer may lead to data
+	  corruption. This workaround sets a specific bit in the diagnostic
+	  register of the Cortex-A9 which disables the Store Buffer
+	  optimisation, preventing the defect from occurring. This has no
+	  visible impact on the overall performance or power consumption of the
+	  processor.
+
+config ARM_ERRATA_751472
+	bool "ARM errata: Interrupted ICIALLUIS may prevent completion of broadcasted operation"
+	depends on CPU_V7
+	depends on !ARCH_MULTIPLATFORM
+	help
+	  This option enables the workaround for the 751472 Cortex-A9 (prior
+	  to r3p0) erratum. An interrupted ICIALLUIS operation may prevent the
+	  completion of a following broadcasted operation if the second
+	  operation is received by a CPU before the ICIALLUIS has completed,
+	  potentially leading to corrupted entries in the cache or TLB.
+
+config ARM_ERRATA_754322
+	bool "ARM errata: possible faulty MMU translations following an ASID switch"
+	depends on CPU_V7
+	help
+	  This option enables the workaround for the 754322 Cortex-A9 (r2p*,
+	  r3p*) erratum. A speculative memory access may cause a page table walk
+	  which starts prior to an ASID switch but completes afterwards. This
+	  can populate the micro-TLB with a stale entry which may be hit with
+	  the new ASID. This workaround places two dsb instructions in the mm
+	  switching code so that no page table walks can cross the ASID switch.
+
+config ARM_ERRATA_754327
+	bool "ARM errata: no automatic Store Buffer drain"
+	depends on CPU_V7 && SMP
+	help
+	  This option enables the workaround for the 754327 Cortex-A9 (prior to
+	  r2p0) erratum. The Store Buffer does not have any automatic draining
+	  mechanism and therefore a livelock may occur if an external agent
+	  continuously polls a memory location waiting to observe an update.
+	  This workaround defines cpu_relax() as smp_mb(), preventing correctly
+	  written polling loops from denying visibility of updates to memory.
+
+config ARM_ERRATA_364296
+	bool "ARM errata: Possible cache data corruption with hit-under-miss enabled"
+	depends on CPU_V6
+	help
+	  This options enables the workaround for the 364296 ARM1136
+	  r0p2 erratum (possible cache data corruption with
+	  hit-under-miss enabled). It sets the undocumented bit 31 in
+	  the auxiliary control register and the FI bit in the control
+	  register, thus disabling hit-under-miss without putting the
+	  processor into full low interrupt latency mode. ARM11MPCore
+	  is not affected.
+
+config ARM_ERRATA_764369
+	bool "ARM errata: Data cache line maintenance operation by MVA may not succeed"
+	depends on CPU_V7 && SMP
+	help
+	  This option enables the workaround for erratum 764369
+	  affecting Cortex-A9 MPCore with two or more processors (all
+	  current revisions). Under certain timing circumstances, a data
+	  cache line maintenance operation by MVA targeting an Inner
+	  Shareable memory region may fail to proceed up to either the
+	  Point of Coherency or to the Point of Unification of the
+	  system. This workaround adds a DSB instruction before the
+	  relevant cache maintenance functions and sets a specific bit
+	  in the diagnostic control register of the SCU.
+
+config ARM_ERRATA_775420
+       bool "ARM errata: A data cache maintenance operation which aborts, might lead to deadlock"
+       depends on CPU_V7
+       help
+	 This option enables the workaround for the 775420 Cortex-A9 (r2p2,
+	 r2p6,r2p8,r2p10,r3p0) erratum. In case a date cache maintenance
+	 operation aborts with MMU exception, it might cause the processor
+	 to deadlock. This workaround puts DSB before executing ISB if
+	 an abort may occur on cache maintenance.
+
+config ARM_ERRATA_798181
+	bool "ARM errata: TLBI/DSB failure on Cortex-A15"
+	depends on CPU_V7 && SMP
+	help
+	  On Cortex-A15 (r0p0..r3p2) the TLBI*IS/DSB operations are not
+	  adequately shooting down all use of the old entries. This
+	  option enables the Linux kernel workaround for this erratum
+	  which sends an IPI to the CPUs that are running the same ASID
+	  as the one being invalidated.
+
+config ARM_ERRATA_773022
+	bool "ARM errata: incorrect instructions may be executed from loop buffer"
+	depends on CPU_V7
+	help
+	  This option enables the workaround for the 773022 Cortex-A15
+	  (up to r0p4) erratum. In certain rare sequences of code, the
+	  loop buffer may deliver incorrect instructions. This
+	  workaround disables the loop buffer to avoid the erratum.
+
+config ARM_ERRATA_814220
+	bool "ARM errata: L2 cache maintenance operation can overtake a L1 cache maintenance operation"
+	depends on CPU_V7
+	help
+	  This option enables the workaround for erratum 814220
+	  Code that intends to clean dirty data from L1 to L2 and then
+	  from L2 to L3 using set/way operations might not behave as
+	  expected. The L2 to L3 operation might happen first and
+	  result in dirty data remaining in L2 after the L1 to L2
+	  operation has completed.
+
+	  This workaround put a DSB before changing cache levels
+
+config ARM_ERRATA_818325_852422
+	bool "ARM errata: A12: some seqs of opposed cond code instrs => deadlock or corruption"
+	depends on CPU_V7
+	help
+	  This option enables the workaround for:
+	  - Cortex-A12 818325: Execution of an UNPREDICTABLE STR or STM
+	    instruction might deadlock.  Fixed in r0p1.
+	  - Cortex-A12 852422: Execution of a sequence of instructions might
+	    lead to either a data corruption or a CPU deadlock.  Not fixed in
+	    any Cortex-A12 cores yet.
+	  This workaround for all both errata involves setting bit[12] of the
+	  Feature Register. This bit disables an optimisation applied to a
+	  sequence of 2 instructions that use opposing condition codes.
+
+config ARM_ERRATA_821420
+	bool "ARM errata: A12: sequence of VMOV to core registers might lead to a dead lock"
+	depends on CPU_V7
+	help
+	  This option enables the workaround for the 821420 Cortex-A12
+	  (all revs) erratum. In very rare timing conditions, a sequence
+	  of VMOV to Core registers instructions, for which the second
+	  one is in the shadow of a branch or abort, can lead to a
+	  deadlock when the VMOV instructions are issued out-of-order.
+
+config ARM_ERRATA_825619
+	bool "ARM errata: A12: DMB NSHST/ISHST mixed ... might cause deadlock"
+	depends on CPU_V7
+	help
+	  This option enables the workaround for the 825619 Cortex-A12
+	  (all revs) erratum. Within rare timing constraints, executing a
+	  DMB NSHST or DMB ISHST instruction followed by a mix of Cacheable
+	  and Device/Strongly-Ordered loads and stores might cause deadlock
+
+config ARM_ERRATA_852421
+	bool "ARM errata: A17: DMB ST might fail to create order between stores"
+	depends on CPU_V7
+	help
+	  This option enables the workaround for the 852421 Cortex-A17
+	  (r1p0, r1p1, r1p2) erratum. Under very rare timing conditions,
+	  execution of a DMB ST instruction might fail to properly order
+	  stores from GroupA and stores from GroupB.
+
+config ARM_ERRATA_852423
+	bool "ARM errata: A17: some seqs of opposed cond code instrs => deadlock or corruption"
+	depends on CPU_V7
+	help
+	  This option enables the workaround for:
+	  - Cortex-A17 852423: Execution of a sequence of instructions might
+	    lead to either a data corruption or a CPU deadlock.  Not fixed in
+	    any Cortex-A17 cores yet.
+	  This is identical to Cortex-A12 erratum 852422.  It is a separate
+	  config option from the A12 erratum due to the way errata are checked
+	  for and handled.
+
+config ARM_ERRATA_855873
+	bool "ARM errata: An eviction might overtake a cache clean operation"
+	help
+	  This option enables the workaround for the 855873 Cortex-A53
+	  (up to r0p4) erratum. Because of this erratum, the processor
+	  might issue an eviction and an L2 cache clean operation to
+	  the interconnect in the wrong order.
+
+config ARM_ERRATA_824069
+	bool "ARM errata: Cache line might not be marked as clean after a CleanShared snoop"
+	help
+	  This option enables the workaround for the 824069 Cortex-A53
+	  (up to r0p2) erratum. If a Cortex-A53 processor is executing
+	  a store or PLDW instruction at the same time as a processor
+	  in another cluster is executing a cache maintenance operation
+	  to the same address, then this erratum might cause
+	  a clean cache line to be incorrectly marked as dirty.
+
+endmenu
+
+source "arch/arm/common/Kconfig"
+
+menu "Bus support"
+
+config ISA
+	bool
+	help
+	  Find out whether you have ISA slots on your motherboard.  ISA is the
+	  name of a bus system, i.e. the way the CPU talks to the other stuff
+	  inside your box.  Other bus systems are PCI, EISA, MicroChannel
+	  (MCA) or VESA.  ISA is an older system, now being displaced by PCI;
+	  newer boards don't support it.  If you have ISA, say Y, otherwise N.
+
+# Select ISA DMA controller support
+config ISA_DMA
+	bool
+	select ISA_DMA_API
+
+# Select ISA DMA interface
+config ISA_DMA_API
+	bool
+
+config PCI
+	bool "PCI support" if MIGHT_HAVE_PCI
+	help
+	  Find out whether you have a PCI motherboard. PCI is the name of a
+	  bus system, i.e. the way the CPU talks to the other stuff inside
+	  your box. Other bus systems are ISA, EISA, MicroChannel (MCA) or
+	  VESA. If you have PCI, say Y, otherwise N.
+
+config PCI_DOMAINS
+	bool
+	depends on PCI
+
+config PCI_DOMAINS_GENERIC
+	def_bool PCI_DOMAINS
+
+config PCI_NANOENGINE
+	bool "BSE nanoEngine PCI support"
+	depends on SA1100_NANOENGINE
+	help
+	  Enable PCI on the BSE nanoEngine board.
+
+config PCI_SYSCALL
+	def_bool PCI
+
+config PCI_HOST_ITE8152
+	bool
+	depends on PCI && MACH_ARMCORE
+	default y
+	select DMABOUNCE
+
+source "drivers/pci/Kconfig"
+
+source "drivers/pcmcia/Kconfig"
+
+endmenu
+
+menu "Kernel Features"
+
+config HAVE_SMP
+	bool
+	help
+	  This option should be selected by machines which have an SMP-
+	  capable CPU.
+
+	  The only effect of this option is to make the SMP-related
+	  options available to the user for configuration.
+
+config SMP
+	bool "Symmetric Multi-Processing"
+	depends on CPU_V6K || CPU_V7
+	depends on GENERIC_CLOCKEVENTS
+	depends on HAVE_SMP
+	depends on MMU || ARM_MPU
+	select IRQ_WORK
+	help
+	  This enables support for systems with more than one CPU. If you have
+	  a system with only one CPU, say N. If you have a system with more
+	  than one CPU, say Y.
+
+	  If you say N here, the kernel will run on uni- and multiprocessor
+	  machines, but will use only one CPU of a multiprocessor machine. If
+	  you say Y here, the kernel will run on many, but not all,
+	  uniprocessor machines. On a uniprocessor machine, the kernel
+	  will run faster if you say N here.
+
+	  See also <file:Documentation/x86/i386/IO-APIC.txt>,
+	  <file:Documentation/nmi_watchdog.txt> and the SMP-HOWTO available at
+	  <http://tldp.org/HOWTO/SMP-HOWTO.html>.
+
+	  If you don't know what to do here, say N.
+
+config SMP_ON_UP
+	bool "Allow booting SMP kernel on uniprocessor systems"
+	depends on SMP && !XIP_KERNEL && MMU
+	default y
+	help
+	  SMP kernels contain instructions which fail on non-SMP processors.
+	  Enabling this option allows the kernel to modify itself to make
+	  these instructions safe.  Disabling it allows about 1K of space
+	  savings.
+
+	  If you don't know what to do here, say Y.
+
+config ARM_CPU_TOPOLOGY
+	bool "Support cpu topology definition"
+	depends on SMP && CPU_V7
+	default y
+	help
+	  Support ARM cpu topology definition. The MPIDR register defines
+	  affinity between processors which is then used to describe the cpu
+	  topology of an ARM System.
+
+config SCHED_MC
+	bool "Multi-core scheduler support"
+	depends on ARM_CPU_TOPOLOGY
+	help
+	  Multi-core scheduler support improves the CPU scheduler's decision
+	  making when dealing with multi-core CPU chips at a cost of slightly
+	  increased overhead in some places. If unsure say N here.
+
+config SCHED_SMT
+	bool "SMT scheduler support"
+	depends on ARM_CPU_TOPOLOGY
+	help
+	  Improves the CPU scheduler's decision making when dealing with
+	  MultiThreading at a cost of slightly increased overhead in some
+	  places. If unsure say N here.
+
+config HAVE_ARM_SCU
+	bool
+	help
+	  This option enables support for the ARM system coherency unit
+
+config HAVE_ARM_ARCH_TIMER
+	bool "Architected timer support"
+	depends on CPU_V7
+	select ARM_ARCH_TIMER
+	select GENERIC_CLOCKEVENTS
+	help
+	  This option enables support for the ARM architected timer
+
+config HAVE_ARM_TWD
+	bool
+	select CLKSRC_OF if OF
+	help
+	  This options enables support for the ARM timer and watchdog unit
+
+config MCPM
+	bool "Multi-Cluster Power Management"
+	depends on CPU_V7 && SMP
+	help
+	  This option provides the common power management infrastructure
+	  for (multi-)cluster based systems, such as big.LITTLE based
+	  systems.
+
+config MCPM_QUAD_CLUSTER
+	bool
+	depends on MCPM
+	help
+	  To avoid wasting resources unnecessarily, MCPM only supports up
+	  to 2 clusters by default.
+	  Platforms with 3 or 4 clusters that use MCPM must select this
+	  option to allow the additional clusters to be managed.
+
+config BIG_LITTLE
+	bool "big.LITTLE support (Experimental)"
+	depends on CPU_V7 && SMP
+	select MCPM
+	help
+	  This option enables support selections for the big.LITTLE
+	  system architecture.
+
+config BL_SWITCHER
+	bool "big.LITTLE switcher support"
+	depends on BIG_LITTLE && MCPM && HOTPLUG_CPU && ARM_GIC
+	select CPU_PM
+	help
+	  The big.LITTLE "switcher" provides the core functionality to
+	  transparently handle transition between a cluster of A15's
+	  and a cluster of A7's in a big.LITTLE system.
+
+config BL_SWITCHER_DUMMY_IF
+	tristate "Simple big.LITTLE switcher user interface"
+	depends on BL_SWITCHER && DEBUG_KERNEL
+	help
+	  This is a simple and dummy char dev interface to control
+	  the big.LITTLE switcher core code.  It is meant for
+	  debugging purposes only.
+
+choice
+	prompt "Memory split"
+	depends on MMU
+	default VMSPLIT_3G
+	help
+	  Select the desired split between kernel and user memory.
+
+	  If you are not absolutely sure what you are doing, leave this
+	  option alone!
+
+	config VMSPLIT_3G
+		bool "3G/1G user/kernel split"
+	config VMSPLIT_3G_OPT
+		bool "3G/1G user/kernel split (for full 1G low memory)"
+	config VMSPLIT_2G
+		bool "2G/2G user/kernel split"
+	config VMSPLIT_1G
+		bool "1G/3G user/kernel split"
+endchoice
+
+config PAGE_OFFSET
+	hex
+	default PHYS_OFFSET if !MMU
+	default 0x40000000 if VMSPLIT_1G
+	default 0x80000000 if VMSPLIT_2G
+	default 0xB0000000 if VMSPLIT_3G_OPT
+	default 0xC0000000
+
+config NR_CPUS
+	int "Maximum number of CPUs (2-32)"
+	range 2 32
+	depends on SMP
+	default "4"
+
+config HOTPLUG_CPU
+	bool "Support for hot-pluggable CPUs"
+	depends on SMP
+	help
+	  Say Y here to experiment with turning CPUs off and on.  CPUs
+	  can be controlled through /sys/devices/system/cpu.
+
+config ARM_PSCI
+	bool "Support for the ARM Power State Coordination Interface (PSCI)"
+	depends on HAVE_ARM_SMCCC
+	select ARM_PSCI_FW
+	help
+	  Say Y here if you want Linux to communicate with system firmware
+	  implementing the PSCI specification for CPU-centric power
+	  management operations described in ARM document number ARM DEN
+	  0022A ("Power State Coordination Interface System Software on
+	  ARM processors").
+
+# The GPIO number here must be sorted by descending number. In case of
+# a multiplatform kernel, we just want the highest value required by the
+# selected platforms.
+config ARCH_NR_GPIO
+	int
+	default 1024 if ARCH_BRCMSTB || ARCH_SHMOBILE || ARCH_TEGRA || \
+		ARCH_ZYNQ
+	default 512 if ARCH_EXYNOS || ARCH_KEYSTONE || SOC_OMAP5 || \
+		SOC_DRA7XX || ARCH_S3C24XX || ARCH_S3C64XX || ARCH_S5PV210
+	default 416 if ARCH_SUNXI
+	default 392 if ARCH_U8500
+	default 352 if ARCH_VT8500
+	default 288 if ARCH_ROCKCHIP
+	default 264 if MACH_H4700
+	default 0
+	help
+	  Maximum number of GPIOs in the system.
+
+	  If unsure, leave the default value.
+
+source kernel/Kconfig.preempt
+
+config HZ_FIXED
+	int
+	default 200 if ARCH_EBSA110 || ARCH_S3C24XX || \
+		ARCH_S5PV210 || ARCH_EXYNOS4
+	default 128 if SOC_AT91RM9200
+	default 0
+
+choice
+	depends on HZ_FIXED = 0
+	prompt "Timer frequency"
+
+config HZ_100
+	depends on !MTK_ENABLE_AGO
+	bool "100 Hz"
+
+config HZ_200
+	depends on !MTK_ENABLE_AGO
+	bool "200 Hz"
+
+config HZ_250
+	depends on !MTK_ENABLE_AGO
+	bool "250 Hz"
+
+config HZ_300
+	bool "300 Hz"
+
+config HZ_500
+	bool "500 Hz"
+
+config HZ_1000
+	bool "1000 Hz"
+
+endchoice
+
+config HZ
+	int
+	default HZ_FIXED if HZ_FIXED != 0
+	default 100 if HZ_100
+	default 200 if HZ_200
+	default 250 if HZ_250
+	default 300 if HZ_300
+	default 500 if HZ_500
+	default 1000
+
+config SCHED_HRTICK
+	def_bool HIGH_RES_TIMERS
+
+config THUMB2_KERNEL
+	bool "Compile the kernel in Thumb-2 mode" if !CPU_THUMBONLY
+	depends on (CPU_V7 || CPU_V7M) && !CPU_V6 && !CPU_V6K
+	default y if CPU_THUMBONLY
+	select AEABI
+	select ARM_ASM_UNIFIED
+	select ARM_UNWIND
+	help
+	  By enabling this option, the kernel will be compiled in
+	  Thumb-2 mode. A compiler/assembler that understand the unified
+	  ARM-Thumb syntax is needed.
+
+	  If unsure, say N.
+
+config THUMB2_AVOID_R_ARM_THM_JUMP11
+	bool "Work around buggy Thumb-2 short branch relocations in gas"
+	depends on THUMB2_KERNEL && MODULES
+	default y
+	help
+	  Various binutils versions can resolve Thumb-2 branches to
+	  locally-defined, preemptible global symbols as short-range "b.n"
+	  branch instructions.
+
+	  This is a problem, because there's no guarantee the final
+	  destination of the symbol, or any candidate locations for a
+	  trampoline, are within range of the branch.  For this reason, the
+	  kernel does not support fixing up the R_ARM_THM_JUMP11 (102)
+	  relocation in modules at all, and it makes little sense to add
+	  support.
+
+	  The symptom is that the kernel fails with an "unsupported
+	  relocation" error when loading some modules.
+
+	  Until fixed tools are available, passing
+	  -fno-optimize-sibling-calls to gcc should prevent gcc generating
+	  code which hits this problem, at the cost of a bit of extra runtime
+	  stack usage in some cases.
+
+	  The problem is described in more detail at:
+	      https://bugs.launchpad.net/binutils-linaro/+bug/725126
+
+	  Only Thumb-2 kernels are affected.
+
+	  Unless you are sure your tools don't have this problem, say Y.
+
+config ARM_ASM_UNIFIED
+	bool
+
+config ARM_PATCH_IDIV
+	bool "Runtime patch udiv/sdiv instructions into __aeabi_{u}idiv()"
+	depends on CPU_32v7 && !XIP_KERNEL
+	default y
+	help
+	  The ARM compiler inserts calls to __aeabi_idiv() and
+	  __aeabi_uidiv() when it needs to perform division on signed
+	  and unsigned integers. Some v7 CPUs have support for the sdiv
+	  and udiv instructions that can be used to implement those
+	  functions.
+
+	  Enabling this option allows the kernel to modify itself to
+	  replace the first two instructions of these library functions
+	  with the sdiv or udiv plus "bx lr" instructions when the CPU
+	  it is running on supports them. Typically this will be faster
+	  and less power intensive than running the original library
+	  code to do integer division.
+
+config AEABI
+	bool "Use the ARM EABI to compile the kernel"
+	help
+	  This option allows for the kernel to be compiled using the latest
+	  ARM ABI (aka EABI).  This is only useful if you are using a user
+	  space environment that is also compiled with EABI.
+
+	  Since there are major incompatibilities between the legacy ABI and
+	  EABI, especially with regard to structure member alignment, this
+	  option also changes the kernel syscall calling convention to
+	  disambiguate both ABIs and allow for backward compatibility support
+	  (selected with CONFIG_OABI_COMPAT).
+
+	  To use this you need GCC version 4.0.0 or later.
+
+config OABI_COMPAT
+	bool "Allow old ABI binaries to run with this kernel (EXPERIMENTAL)"
+	depends on AEABI && !THUMB2_KERNEL
+	help
+	  This option preserves the old syscall interface along with the
+	  new (ARM EABI) one. It also provides a compatibility layer to
+	  intercept syscalls that have structure arguments which layout
+	  in memory differs between the legacy ABI and the new ARM EABI
+	  (only for non "thumb" binaries). This option adds a tiny
+	  overhead to all syscalls and produces a slightly larger kernel.
+
+	  The seccomp filter system will not be available when this is
+	  selected, since there is no way yet to sensibly distinguish
+	  between calling conventions during filtering.
+
+	  If you know you'll be using only pure EABI user space then you
+	  can say N here. If this option is not selected and you attempt
+	  to execute a legacy ABI binary then the result will be
+	  UNPREDICTABLE (in fact it can be predicted that it won't work
+	  at all). If in doubt say N.
+
+config ARCH_HAS_HOLES_MEMORYMODEL
+	bool
+
+config ARCH_SPARSEMEM_ENABLE
+	bool
+
+config ARCH_SPARSEMEM_DEFAULT
+	def_bool ARCH_SPARSEMEM_ENABLE
+
+config ARCH_SELECT_MEMORY_MODEL
+	def_bool ARCH_SPARSEMEM_ENABLE
+
+config HAVE_ARCH_PFN_VALID
+	def_bool ARCH_HAS_HOLES_MEMORYMODEL || !SPARSEMEM
+
+config HAVE_GENERIC_RCU_GUP
+	def_bool y
+	depends on ARM_LPAE
+
+config HIGHMEM
+	bool "High Memory Support"
+	depends on MMU
+	help
+	  The address space of ARM processors is only 4 Gigabytes large
+	  and it has to accommodate user address space, kernel address
+	  space as well as some memory mapped IO. That means that, if you
+	  have a large amount of physical memory and/or IO, not all of the
+	  memory can be "permanently mapped" by the kernel. The physical
+	  memory that is not permanently mapped is called "high memory".
+
+	  Depending on the selected kernel/user memory split, minimum
+	  vmalloc space and actual amount of RAM, you may not need this
+	  option which should result in a slightly faster kernel.
+
+	  If unsure, say n.
+
+config HIGHPTE
+	bool "Allocate 2nd-level pagetables from highmem" if EXPERT
+	depends on HIGHMEM
+	default y
+	help
+	  The VM uses one page of physical memory for each page table.
+	  For systems with a lot of processes, this can use a lot of
+	  precious low memory, eventually leading to low memory being
+	  consumed by page tables.  Setting this option will allow
+	  user-space 2nd level page tables to reside in high memory.
+
+config CPU_SW_DOMAIN_PAN
+	bool "Enable use of CPU domains to implement privileged no-access"
+	depends on MMU && !ARM_LPAE
+	default y
+	help
+	  Increase kernel security by ensuring that normal kernel accesses
+	  are unable to access userspace addresses.  This can help prevent
+	  use-after-free bugs becoming an exploitable privilege escalation
+	  by ensuring that magic values (such as LIST_POISON) will always
+	  fault when dereferenced.
+
+	  CPUs with low-vector mappings use a best-efforts implementation.
+	  Their lower 1MB needs to remain accessible for the vectors, but
+	  the remainder of userspace will become appropriately inaccessible.
+
+config HW_PERF_EVENTS
+	def_bool y
+	depends on ARM_PMU
+
+config SYS_SUPPORTS_HUGETLBFS
+       def_bool y
+       depends on ARM_LPAE
+
+config HAVE_ARCH_TRANSPARENT_HUGEPAGE
+       def_bool y
+       depends on ARM_LPAE
+
+config ARCH_WANT_GENERAL_HUGETLB
+	def_bool y
+
+config ARM_MODULE_PLTS
+	bool "Use PLTs to allow module memory to spill over into vmalloc area"
+	depends on MODULES
+	help
+	  Allocate PLTs when loading modules so that jumps and calls whose
+	  targets are too far away for their relative offsets to be encoded
+	  in the instructions themselves can be bounced via veneers in the
+	  module's PLT. This allows modules to be allocated in the generic
+	  vmalloc area after the dedicated module memory area has been
+	  exhausted. The modules will use slightly more memory, but after
+	  rounding up to page size, the actual memory footprint is usually
+	  the same.
+
+	  Say y if you are getting out of memory errors while loading modules
+
+source "mm/Kconfig"
+
+config FORCE_MAX_ZONEORDER
+	int "Maximum zone order"
+	default "12" if SOC_AM33XX
+	default "9" if SA1111 || ARCH_EFM32
+	default "11"
+	help
+	  The kernel memory allocator divides physically contiguous memory
+	  blocks into "zones", where each zone is a power of two number of
+	  pages.  This option selects the largest power of two that the kernel
+	  keeps in the memory allocator.  If you need to allocate very large
+	  blocks of physically contiguous memory, then you may need to
+	  increase this value.
+
+	  This config option is actually maximum order plus one. For example,
+	  a value of 11 means that the largest free memory block is 2^10 pages.
+
+config ALIGNMENT_TRAP
+	bool
+	depends on CPU_CP15_MMU
+	default y if !ARCH_EBSA110
+	select HAVE_PROC_CPU if PROC_FS
+	help
+	  ARM processors cannot fetch/store information which is not
+	  naturally aligned on the bus, i.e., a 4 byte fetch must start at an
+	  address divisible by 4. On 32-bit ARM processors, these non-aligned
+	  fetch/store instructions will be emulated in software if you say
+	  here, which has a severe performance impact. This is necessary for
+	  correct operation of some network protocols. With an IP-only
+	  configuration it is safe to say N, otherwise say Y.
+
+config UACCESS_WITH_MEMCPY
+	bool "Use kernel mem{cpy,set}() for {copy_to,clear}_user()"
+	depends on MMU
+	default y if CPU_FEROCEON
+	help
+	  Implement faster copy_to_user and clear_user methods for CPU
+	  cores where a 8-word STM instruction give significantly higher
+	  memory write throughput than a sequence of individual 32bit stores.
+
+	  A possible side effect is a slight increase in scheduling latency
+	  between threads sharing the same address space if they invoke
+	  such copy operations with large buffers.
+
+	  However, if the CPU data cache is using a write-allocate mode,
+	  this option is unlikely to provide any performance gain.
+
+config SECCOMP
+	bool
+	prompt "Enable seccomp to safely compute untrusted bytecode"
+	---help---
+	  This kernel feature is useful for number crunching applications
+	  that may need to compute untrusted bytecode during their
+	  execution. By using pipes or other transports made available to
+	  the process as file descriptors supporting the read/write
+	  syscalls, it's possible to isolate those applications in
+	  their own address space using seccomp. Once seccomp is
+	  enabled via prctl(PR_SET_SECCOMP), it cannot be disabled
+	  and the task is only allowed to execute a few safe syscalls
+	  defined by each seccomp mode.
+
+config SWIOTLB
+	def_bool y
+
+config IOMMU_HELPER
+	def_bool SWIOTLB
+
+config PARAVIRT
+	bool "Enable paravirtualization code"
+	help
+	  This changes the kernel so it can modify itself when it is run
+	  under a hypervisor, potentially improving performance significantly
+	  over full virtualization.
+
+config PARAVIRT_TIME_ACCOUNTING
+	bool "Paravirtual steal time accounting"
+	select PARAVIRT
+	default n
+	help
+	  Select this option to enable fine granularity task steal time
+	  accounting. Time spent executing other tasks in parallel with
+	  the current vCPU is discounted from the vCPU power. To account for
+	  that, there can be a small performance impact.
+
+	  If in doubt, say N here.
+
+config XEN_DOM0
+	def_bool y
+	depends on XEN
+
+config XEN
+	bool "Xen guest support on ARM"
+	depends on ARM && AEABI && OF
+	depends on CPU_V7 && !CPU_V6
+	depends on !GENERIC_ATOMIC64
+	depends on MMU
+	select ARCH_DMA_ADDR_T_64BIT
+	select ARM_PSCI
+	select SWIOTLB_XEN
+	select PARAVIRT
+	help
+	  Say Y if you want to run Linux in a Virtual Machine on Xen on ARM.
+
+config ARM_FLUSH_CONSOLE_ON_RESTART
+	bool "Force flush the console on restart"
+	help
+	  If the console is locked while the system is rebooted, the messages
+	  in the temporary logbuffer would not have propogated to all the
+	  console drivers. This option forces the console lock to be
+	  released if it failed to be acquired, which will cause all the
+	  pending messages to be flushed.
+
+endmenu
+
+menu "Boot options"
+
+config USE_OF
+	bool "Flattened Device Tree support"
+	select IRQ_DOMAIN
+	select OF
+	help
+	  Include support for flattened device tree machine descriptions.
+
+config ATAGS
+	bool "Support for the traditional ATAGS boot data passing" if USE_OF
+	default y
+	help
+	  This is the traditional way of passing data to the kernel at boot
+	  time. If you are solely relying on the flattened device tree (or
+	  the ARM_ATAG_DTB_COMPAT option) then you may unselect this option
+	  to remove ATAGS support from your kernel binary.  If unsure,
+	  leave this to y.
+
+config DEPRECATED_PARAM_STRUCT
+	bool "Provide old way to pass kernel parameters"
+	depends on ATAGS
+	help
+	  This was deprecated in 2001 and announced to live on for 5 years.
+	  Some old boot loaders still use this way.
+
+config BUILD_ARM_APPENDED_DTB_IMAGE
+	bool "Build a concatenated zImage/dtb by default"
+	depends on OF
+	help
+	  Enabling this option will cause a concatenated zImage and list of
+	  DTBs to be built by default (instead of a standalone zImage.)
+	  The image will built in arch/arm/boot/zImage-dtb
+
+config BUILD_ARM_APPENDED_DTB_IMAGE_NAMES
+	string "Default dtb names"
+	depends on BUILD_ARM_APPENDED_DTB_IMAGE
+	help
+	  Space separated list of names of dtbs to append when
+	  building a concatenated zImage-dtb.
+
+config BUILD_ARM_DTB_OVERLAY_IMAGE
+	bool "Build a device tree overlay image"
+	depends on OF
+	help
+	  Enabling this option will casue overlay device tree
+	  to be built by default.
+	  Overlay DT contains device-specific configurations.
+	  Please be aware that bootloader supports DT merging.
+
+config BUILD_ARM_DTB_OVERLAY_IMAGE_NAMES
+	string "Default dtb overlay names"
+	depends on BUILD_ARM_DTB_OVERLAY_IMAGE
+	help
+	  Space separated list of names of dtbs to append when
+	  building a concatenated overlay image
+
+# Compressed boot loader in ROM.  Yes, we really want to ask about
+# TEXT and BSS so we preserve their values in the config files.
+config ZBOOT_ROM_TEXT
+	hex "Compressed ROM boot loader base address"
+	default "0"
+	help
+	  The physical address at which the ROM-able zImage is to be
+	  placed in the target.  Platforms which normally make use of
+	  ROM-able zImage formats normally set this to a suitable
+	  value in their defconfig file.
+
+	  If ZBOOT_ROM is not enabled, this has no effect.
+
+config ZBOOT_ROM_BSS
+	hex "Compressed ROM boot loader BSS address"
+	default "0"
+	help
+	  The base address of an area of read/write memory in the target
+	  for the ROM-able zImage which must be available while the
+	  decompressor is running. It must be large enough to hold the
+	  entire decompressed kernel plus an additional 128 KiB.
+	  Platforms which normally make use of ROM-able zImage formats
+	  normally set this to a suitable value in their defconfig file.
+
+	  If ZBOOT_ROM is not enabled, this has no effect.
+
+config ZBOOT_ROM
+	bool "Compressed boot loader in ROM/flash"
+	depends on ZBOOT_ROM_TEXT != ZBOOT_ROM_BSS
+	depends on !ARM_APPENDED_DTB && !XIP_KERNEL && !AUTO_ZRELADDR
+	help
+	  Say Y here if you intend to execute your compressed kernel image
+	  (zImage) directly from ROM or flash.  If unsure, say N.
+
+config ARM_APPENDED_DTB
+	bool "Use appended device tree blob to zImage (EXPERIMENTAL)"
+	depends on OF
+	help
+	  With this option, the boot code will look for a device tree binary
+	  (DTB) appended to zImage
+	  (e.g. cat zImage <filename>.dtb > zImage_w_dtb).
+
+	  This is meant as a backward compatibility convenience for those
+	  systems with a bootloader that can't be upgraded to accommodate
+	  the documented boot protocol using a device tree.
+
+	  Beware that there is very little in terms of protection against
+	  this option being confused by leftover garbage in memory that might
+	  look like a DTB header after a reboot if no actual DTB is appended
+	  to zImage.  Do not leave this option active in a production kernel
+	  if you don't intend to always append a DTB.  Proper passing of the
+	  location into r2 of a bootloader provided DTB is always preferable
+	  to this option.
+
+config ARM_ATAG_DTB_COMPAT
+	bool "Supplement the appended DTB with traditional ATAG information"
+	depends on ARM_APPENDED_DTB
+	help
+	  Some old bootloaders can't be updated to a DTB capable one, yet
+	  they provide ATAGs with memory configuration, the ramdisk address,
+	  the kernel cmdline string, etc.  Such information is dynamically
+	  provided by the bootloader and can't always be stored in a static
+	  DTB.  To allow a device tree enabled kernel to be used with such
+	  bootloaders, this option allows zImage to extract the information
+	  from the ATAG list and store it at run time into the appended DTB.
+
+choice
+	prompt "Kernel command line type" if ARM_ATAG_DTB_COMPAT
+	default ARM_ATAG_DTB_COMPAT_CMDLINE_FROM_BOOTLOADER
+
+config ARM_ATAG_DTB_COMPAT_CMDLINE_FROM_BOOTLOADER
+	bool "Use bootloader kernel arguments if available"
+	help
+	  Uses the command-line options passed by the boot loader instead of
+	  the device tree bootargs property. If the boot loader doesn't provide
+	  any, the device tree bootargs property will be used.
+
+config ARM_ATAG_DTB_COMPAT_CMDLINE_EXTEND
+	bool "Extend with bootloader kernel arguments"
+	help
+	  The command-line arguments provided by the boot loader will be
+	  appended to the the device tree bootargs property.
+
+endchoice
+
+config CMDLINE
+	string "Default kernel command string"
+	default ""
+	help
+	  On some architectures (EBSA110 and CATS), there is currently no way
+	  for the boot loader to pass arguments to the kernel. For these
+	  architectures, you should supply some command-line options at build
+	  time by entering them here. As a minimum, you should specify the
+	  memory size and the root device (e.g., mem=64M root=/dev/nfs).
+
+choice
+	prompt "Kernel command line type" if CMDLINE != ""
+	default CMDLINE_FROM_BOOTLOADER
+	depends on ATAGS
+
+config CMDLINE_FROM_BOOTLOADER
+	bool "Use bootloader kernel arguments if available"
+	help
+	  Uses the command-line options passed by the boot loader. If
+	  the boot loader doesn't provide any, the default kernel command
+	  string provided in CMDLINE will be used.
+
+config CMDLINE_EXTEND
+	bool "Extend bootloader kernel arguments"
+	help
+	  The command-line arguments provided by the boot loader will be
+	  appended to the default kernel command string.
+
+config CMDLINE_FORCE
+	bool "Always use the default kernel command string"
+	help
+	  Always use the default kernel command string, even if the boot
+	  loader passes other arguments to the kernel.
+	  This is useful if you cannot or don't want to change the
+	  command-line options your boot loader passes to the kernel.
+endchoice
+
+config XIP_KERNEL
+	bool "Kernel Execute-In-Place from ROM"
+	depends on !ARM_LPAE && !ARCH_MULTIPLATFORM
+	help
+	  Execute-In-Place allows the kernel to run from non-volatile storage
+	  directly addressable by the CPU, such as NOR flash. This saves RAM
+	  space since the text section of the kernel is not loaded from flash
+	  to RAM.  Read-write sections, such as the data section and stack,
+	  are still copied to RAM.  The XIP kernel is not compressed since
+	  it has to run directly from flash, so it will take more space to
+	  store it.  The flash address used to link the kernel object files,
+	  and for storing it, is configuration dependent. Therefore, if you
+	  say Y here, you must know the proper physical address where to
+	  store the kernel image depending on your own flash memory usage.
+
+	  Also note that the make target becomes "make xipImage" rather than
+	  "make zImage" or "make Image".  The final kernel binary to put in
+	  ROM memory will be arch/arm/boot/xipImage.
+
+	  If unsure, say N.
+
+config XIP_PHYS_ADDR
+	hex "XIP Kernel Physical Location"
+	depends on XIP_KERNEL
+	default "0x00080000"
+	help
+	  This is the physical address in your flash memory the kernel will
+	  be linked for and stored to.  This address is dependent on your
+	  own flash usage.
+
+config KEXEC
+	bool "Kexec system call (EXPERIMENTAL)"
+	depends on (!SMP || PM_SLEEP_SMP)
+	depends on !CPU_V7M
+	select KEXEC_CORE
+	help
+	  kexec is a system call that implements the ability to shutdown your
+	  current kernel, and to start another kernel.  It is like a reboot
+	  but it is independent of the system firmware.   And like a reboot
+	  you can start any kernel with it, not just Linux.
+
+	  It is an ongoing process to be certain the hardware in a machine
+	  is properly shutdown, so do not be surprised if this code does not
+	  initially work for you.
+
+config ATAGS_PROC
+	bool "Export atags in procfs"
+	depends on ATAGS && KEXEC
+	default y
+	help
+	  Should the atags used to boot the kernel be exported in an "atags"
+	  file in procfs. Useful with kexec.
+
+config CRASH_DUMP
+	bool "Build kdump crash kernel (EXPERIMENTAL)"
+	help
+	  Generate crash dump after being started by kexec. This should
+	  be normally only set in special crash dump kernels which are
+	  loaded in the main kernel with kexec-tools into a specially
+	  reserved region and then later executed after a crash by
+	  kdump/kexec. The crash dump kernel must be compiled to a
+	  memory address not used by the main kernel
+
+	  For more details see Documentation/kdump/kdump.txt
+
+config AUTO_ZRELADDR
+	bool "Auto calculation of the decompressed kernel image address"
+	help
+	  ZRELADDR is the physical address where the decompressed kernel
+	  image will be placed. If AUTO_ZRELADDR is selected, the address
+	  will be determined at run-time by masking the current IP with
+	  0xf8000000. This assumes the zImage being placed in the first 128MB
+	  from start of memory.
+
+config EFI_STUB
+	bool
+
+config EFI
+	bool "UEFI runtime support"
+	depends on OF && !CPU_BIG_ENDIAN && MMU && AUTO_ZRELADDR && !XIP_KERNEL
+	select UCS2_STRING
+	select EFI_PARAMS_FROM_FDT
+	select EFI_STUB
+	select EFI_ARMSTUB
+	select EFI_RUNTIME_WRAPPERS
+	---help---
+	  This option provides support for runtime services provided
+	  by UEFI firmware (such as non-volatile variables, realtime
+	  clock, and platform reset). A UEFI stub is also provided to
+	  allow the kernel to be booted as an EFI application. This
+	  is only useful for kernels that may run on systems that have
+	  UEFI firmware.
+
+endmenu
+
+menu "CPU Power Management"
+
+source "drivers/cpufreq/Kconfig"
+
+source "drivers/cpuidle/Kconfig"
+
+endmenu
+
+menu "Floating point emulation"
+
+comment "At least one emulation must be selected"
+
+config FPE_NWFPE
+	bool "NWFPE math emulation"
+	depends on (!AEABI || OABI_COMPAT) && !THUMB2_KERNEL
+	---help---
+	  Say Y to include the NWFPE floating point emulator in the kernel.
+	  This is necessary to run most binaries. Linux does not currently
+	  support floating point hardware so you need to say Y here even if
+	  your machine has an FPA or floating point co-processor podule.
+
+	  You may say N here if you are going to load the Acorn FPEmulator
+	  early in the bootup.
+
+config FPE_NWFPE_XP
+	bool "Support extended precision"
+	depends on FPE_NWFPE
+	help
+	  Say Y to include 80-bit support in the kernel floating-point
+	  emulator.  Otherwise, only 32 and 64-bit support is compiled in.
+	  Note that gcc does not generate 80-bit operations by default,
+	  so in most cases this option only enlarges the size of the
+	  floating point emulator without any good reason.
+
+	  You almost surely want to say N here.
+
+config FPE_FASTFPE
+	bool "FastFPE math emulation (EXPERIMENTAL)"
+	depends on (!AEABI || OABI_COMPAT) && !CPU_32v3
+	---help---
+	  Say Y here to include the FAST floating point emulator in the kernel.
+	  This is an experimental much faster emulator which now also has full
+	  precision for the mantissa.  It does not support any exceptions.
+	  It is very simple, and approximately 3-6 times faster than NWFPE.
+
+	  It should be sufficient for most programs.  It may be not suitable
+	  for scientific calculations, but you have to check this for yourself.
+	  If you do not feel you need a faster FP emulation you should better
+	  choose NWFPE.
+
+config VFP
+	bool "VFP-format floating point maths"
+	depends on CPU_V6 || CPU_V6K || CPU_ARM926T || CPU_V7 || CPU_FEROCEON
+	help
+	  Say Y to include VFP support code in the kernel. This is needed
+	  if your hardware includes a VFP unit.
+
+	  Please see <file:Documentation/arm/VFP/release-notes.txt> for
+	  release notes and additional status information.
+
+	  Say N if your target does not have VFP hardware.
+
+config VFPv3
+	bool
+	depends on VFP
+	default y if CPU_V7
+
+config NEON
+	bool "Advanced SIMD (NEON) Extension support"
+	depends on VFPv3 && CPU_V7
+	help
+	  Say Y to include support code for NEON, the ARMv7 Advanced SIMD
+	  Extension.
+
+config KERNEL_MODE_NEON
+	bool "Support for NEON in kernel mode"
+	depends on NEON && AEABI
+	help
+	  Say Y to include support for NEON in kernel mode.
+
+endmenu
+
+menu "Userspace binary formats"
+
+source "fs/Kconfig.binfmt"
+
+endmenu
+
+menu "Power management options"
+
+source "kernel/power/Kconfig"
+
+config ARCH_SUSPEND_POSSIBLE
+	depends on CPU_ARM920T || CPU_ARM926T || CPU_FEROCEON || CPU_SA1100 || \
+		CPU_V6 || CPU_V6K || CPU_V7 || CPU_V7M || CPU_XSC3 || CPU_XSCALE || CPU_MOHAWK
+	def_bool y
+
+config ARM_CPU_SUSPEND
+	def_bool PM_SLEEP || BL_SWITCHER || ARM_PSCI_FW
+	depends on ARCH_SUSPEND_POSSIBLE
+
+config ARCH_HIBERNATION_POSSIBLE
+	bool
+	depends on MMU
+	default y if ARCH_SUSPEND_POSSIBLE
+
+endmenu
+
+source "net/Kconfig"
+
+source "drivers/Kconfig"
+
+source "drivers/firmware/Kconfig"
+
+source "fs/Kconfig"
+
+source "arch/arm/Kconfig.debug"
+
+source "security/Kconfig"
+
+source "crypto/Kconfig"
+if CRYPTO
+source "arch/arm/crypto/Kconfig"
+endif
+
+source "lib/Kconfig"
+
+source "arch/arm/kvm/Kconfig"

arch/arm/Kconfig-nommu

@@ -0,0 +1,63 @@
+#
+# Kconfig for uClinux(non-paged MM) depend configurations
+# Hyok S. Choi <hyok.choi@samsung.com>
+# 
+
+config SET_MEM_PARAM
+	bool "Set flash/sdram size and base addr"
+	help
+	 Say Y to manually set the base addresses and sizes.
+	 otherwise, the default values are assigned.
+
+config DRAM_BASE
+	hex '(S)DRAM Base Address' if SET_MEM_PARAM
+	default 0x00800000
+
+config DRAM_SIZE
+	hex '(S)DRAM SIZE' if SET_MEM_PARAM
+	default 0x00800000
+
+config FLASH_MEM_BASE
+	hex 'FLASH Base Address' if SET_MEM_PARAM
+	default 0x00400000
+
+config FLASH_SIZE
+	hex 'FLASH Size' if SET_MEM_PARAM
+	default 0x00400000
+
+config PROCESSOR_ID
+	hex 'Hard wire the processor ID'
+	default 0x00007700
+	depends on !(CPU_CP15 || CPU_V7M)
+	help
+	  If processor has no CP15 register, this processor ID is
+	  used instead of the auto-probing which utilizes the register.
+
+config REMAP_VECTORS_TO_RAM
+	bool 'Install vectors to the beginning of RAM'
+	help
+	  The kernel needs to change the hardware exception vectors.
+	  In nommu mode, the hardware exception vectors are normally
+	  placed at address 0x00000000. However, this region may be
+	  occupied by read-only memory depending on H/W design.
+
+	  If the region contains read-write memory, say 'n' here.
+
+	  If your CPU provides a remap facility which allows the exception
+	  vectors to be mapped to writable memory, say 'n' here.
+
+	  Otherwise, say 'y' here.  In this case, the kernel will require
+	  external support to redirect the hardware exception vectors to
+	  the writable versions located at DRAM_BASE.
+
+config ARM_MPU
+       bool 'Use the ARM v7 PMSA Compliant MPU'
+       depends on CPU_V7
+       default y
+       help
+         Some ARM systems without an MMU have instead a Memory Protection
+         Unit (MPU) that defines the type and permissions for regions of
+         memory.
+
+         If your CPU has an MPU then you should choose 'y' here unless you
+         know that you do not want to use the MPU.

arch/arm/Kconfig.debug

@@ -0,0 +1,1762 @@
+menu "Kernel hacking"
+
+source "lib/Kconfig.debug"
+
+config ARM_PTDUMP
+	bool "Export kernel pagetable layout to userspace via debugfs"
+	depends on DEBUG_KERNEL
+	depends on MMU
+	select DEBUG_FS
+	---help---
+	  Say Y here if you want to show the kernel pagetable layout in a
+	  debugfs file. This information is only useful for kernel developers
+	  who are working in architecture specific areas of the kernel.
+	  It is probably not a good idea to enable this feature in a production
+	  kernel.
+	  If in doubt, say "N"
+
+# RMK wants arm kernels compiled with frame pointers or stack unwinding.
+# If you know what you are doing and are willing to live without stack
+# traces, you can get a slightly smaller kernel by setting this option to
+# n, but then RMK will have to kill you ;).
+config FRAME_POINTER
+	bool
+	depends on !THUMB2_KERNEL
+	default y if !ARM_UNWIND || FUNCTION_GRAPH_TRACER
+	help
+	  If you say N here, the resulting kernel will be slightly smaller and
+	  faster. However, if neither FRAME_POINTER nor ARM_UNWIND are enabled,
+	  when a problem occurs with the kernel, the information that is
+	  reported is severely limited.
+
+config ARM_UNWIND
+	bool "Enable stack unwinding support (EXPERIMENTAL)"
+	depends on AEABI
+	default y
+	help
+	  This option enables stack unwinding support in the kernel
+	  using the information automatically generated by the
+	  compiler. The resulting kernel image is slightly bigger but
+	  the performance is not affected. Currently, this feature
+	  only works with EABI compilers. If unsure say Y.
+
+config OLD_MCOUNT
+	bool
+	depends on FUNCTION_TRACER && FRAME_POINTER
+	default y
+
+config DEBUG_USER
+	bool "Verbose user fault messages"
+	help
+	  When a user program crashes due to an exception, the kernel can
+	  print a brief message explaining what the problem was. This is
+	  sometimes helpful for debugging but serves no purpose on a
+	  production system. Most people should say N here.
+
+	  In addition, you need to pass user_debug=N on the kernel command
+	  line to enable this feature.  N consists of the sum of:
+
+	      1 - undefined instruction events
+	      2 - system calls
+	      4 - invalid data aborts
+	      8 - SIGSEGV faults
+	     16 - SIGBUS faults
+
+# These options are only for real kernel hackers who want to get their hands dirty.
+config DEBUG_LL
+	bool "Kernel low-level debugging functions (read help!)"
+	depends on DEBUG_KERNEL
+	help
+	  Say Y here to include definitions of printascii, printch, printhex
+	  in the kernel.  This is helpful if you are debugging code that
+	  executes before the console is initialized.
+
+	  Note that selecting this option will limit the kernel to a single
+	  UART definition, as specified below. Attempting to boot the kernel
+	  image on a different platform *will not work*, so this option should
+	  not be enabled for kernels that are intended to be portable.
+
+choice
+	prompt "Kernel low-level debugging port"
+	depends on DEBUG_LL
+
+	config DEBUG_ALPINE_UART0
+		bool "Kernel low-level debugging messages via Alpine UART0"
+		depends on ARCH_ALPINE
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Alpine based platforms.
+
+	config DEBUG_ASM9260_UART
+		bool "Kernel low-level debugging via asm9260 UART"
+		depends on MACH_ASM9260
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to an UART or USART port on asm9260 based
+		  machines.
+
+		    DEBUG_UART_PHYS | DEBUG_UART_VIRT
+
+		    0x80000000      | 0xf0000000     | UART0
+		    0x80004000      | 0xf0004000     | UART1
+		    0x80008000      | 0xf0008000     | UART2
+		    0x8000c000      | 0xf000c000     | UART3
+		    0x80010000      | 0xf0010000     | UART4
+		    0x80014000      | 0xf0014000     | UART5
+		    0x80018000      | 0xf0018000     | UART6
+		    0x8001c000      | 0xf001c000     | UART7
+		    0x80020000      | 0xf0020000     | UART8
+		    0x80024000      | 0xf0024000     | UART9
+
+	config DEBUG_AT91_RM9200_DBGU
+		bool "Kernel low-level debugging on AT91RM9200, AT91SAM9 DBGU"
+		select DEBUG_AT91_UART
+		depends on SOC_AT91RM9200 || SOC_AT91SAM9
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on the DBGU port of:
+		    at91rm9200, at91sam9260, at91sam9g20, at91sam9261,
+		    at91sam9g10, at91sam9n12, at91sam9rl64, at91sam9x5
+
+	config DEBUG_AT91_SAM9263_DBGU
+		bool "Kernel low-level debugging on AT91SAM{9263,9G45,A5D3} DBGU"
+		select DEBUG_AT91_UART
+		depends on SOC_AT91SAM9 || SOC_SAMA5D3
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on the DBGU port of:
+		    at91sam9263, at91sam9g45, at91sam9m10,
+		    sama5d3
+
+	config DEBUG_AT91_SAMA5D2_UART1
+		bool "Kernel low-level debugging on SAMA5D2 UART1"
+		select DEBUG_AT91_UART
+		depends on SOC_SAMA5D2
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on the UART1 port of sama5d2.
+
+	config DEBUG_AT91_SAMA5D4_USART3
+		bool "Kernel low-level debugging on SAMA5D4 USART3"
+		select DEBUG_AT91_UART
+		depends on SOC_SAMA5D4
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on the USART3 port of sama5d4.
+
+	config DEBUG_BCM2835
+		bool "Kernel low-level debugging on BCM2835 PL011 UART"
+		depends on ARCH_BCM2835 && ARCH_MULTI_V6
+		select DEBUG_UART_PL01X
+
+	config DEBUG_BCM2836
+		bool "Kernel low-level debugging on BCM2836 PL011 UART"
+		depends on ARCH_BCM2835 && ARCH_MULTI_V7
+		select DEBUG_UART_PL01X
+
+	config DEBUG_BCM_5301X
+		bool "Kernel low-level debugging on BCM5301X/NSP UART1"
+		depends on ARCH_BCM_5301X || ARCH_BCM_NSP
+		select DEBUG_UART_8250
+
+	config DEBUG_BCM_KONA_UART
+		bool "Kernel low-level debugging messages via BCM KONA UART"
+		depends on ARCH_BCM_MOBILE
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Broadcom SoC platforms.
+		  This low level debug works for Broadcom
+		  mobile SoCs in the Kona family of chips (e.g. bcm28155,
+		  bcm11351, etc...)
+
+	config DEBUG_BCM63XX_UART
+		bool "Kernel low-level debugging on BCM63XX UART"
+		depends on ARCH_BCM_63XX
+
+	config DEBUG_BERLIN_UART
+		bool "Marvell Berlin SoC Debug UART"
+		depends on ARCH_BERLIN
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Marvell Berlin SoC based platforms.
+
+	config DEBUG_BRCMSTB_UART
+		bool "Use BRCMSTB UART for low-level debug"
+		depends on ARCH_BRCMSTB
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the first serial port on these devices. The
+		  UART physical and virtual address is automatically provided
+		  based on the chip identification register value.
+
+		  If you have a Broadcom STB chip and would like early print
+		  messages to appear over the UART, select this option.
+
+	config DEBUG_CLPS711X_UART1
+		bool "Kernel low-level debugging messages via UART1"
+		depends on ARCH_CLPS711X
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the first serial port on these devices.
+
+	config DEBUG_CLPS711X_UART2
+		bool "Kernel low-level debugging messages via UART2"
+		depends on ARCH_CLPS711X
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the second serial port on these devices.
+
+	config DEBUG_CNS3XXX
+		bool "Kernel Kernel low-level debugging on Cavium Networks CNS3xxx"
+		depends on ARCH_CNS3XXX
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want the debug print routines to direct
+                  their output to the CNS3xxx UART0.
+
+	config DEBUG_DAVINCI_DA8XX_UART1
+		bool "Kernel low-level debugging on DaVinci DA8XX using UART1"
+		depends on ARCH_DAVINCI_DA8XX
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to UART1 serial port on DaVinci DA8XX devices.
+
+	config DEBUG_DAVINCI_DA8XX_UART2
+		bool "Kernel low-level debugging on DaVinci DA8XX using UART2"
+		depends on ARCH_DAVINCI_DA8XX
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to UART2 serial port on DaVinci DA8XX devices.
+
+	config DEBUG_DAVINCI_DMx_UART0
+		bool "Kernel low-level debugging on DaVinci DMx using UART0"
+		depends on ARCH_DAVINCI_DMx
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to UART0 serial port on DaVinci DMx devices.
+
+	config DEBUG_DC21285_PORT
+		bool "Kernel low-level debugging messages via footbridge serial port"
+		depends on FOOTBRIDGE
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the serial port in the DC21285 (Footbridge).
+
+	config DEBUG_DIGICOLOR_UA0
+		bool "Kernel low-level debugging messages via Digicolor UA0"
+		depends on ARCH_DIGICOLOR
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the UA0 serial port in the CX92755.
+
+	config DEBUG_EP93XX
+		bool "Kernel low-level debugging messages via ep93xx UART"
+		depends on ARCH_EP93XX
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Cirrus Logic EP93xx based platforms.
+
+	config DEBUG_FOOTBRIDGE_COM1
+		bool "Kernel low-level debugging messages via footbridge 8250 at PCI COM1"
+		depends on FOOTBRIDGE
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the 8250 at PCI COM1.
+
+	config DEBUG_GEMINI
+		bool "Kernel low-level debugging messages via Cortina Systems Gemini UART"
+		depends on ARCH_GEMINI
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Cortina Gemini based platforms.
+
+	config DEBUG_HI3620_UART
+		bool "Hisilicon HI3620 Debug UART"
+		depends on ARCH_HI3xxx
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on HI3620 UART.
+
+	config DEBUG_HIGHBANK_UART
+		bool "Kernel low-level debugging messages via Highbank UART"
+		depends on ARCH_HIGHBANK
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the UART on Highbank based devices.
+
+	config DEBUG_HIP01_UART
+		bool "Hisilicon Hip01 Debug UART"
+		depends on ARCH_HIP01
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on HIP01 UART.
+
+	config DEBUG_HIP04_UART
+		bool "Hisilicon HiP04 Debug UART"
+		depends on ARCH_HIP04
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on HIP04 UART.
+
+	config DEBUG_HIX5HD2_UART
+		bool "Hisilicon Hix5hd2 Debug UART"
+		depends on ARCH_HIX5HD2
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Hix5hd2 UART.
+
+	config DEBUG_IMX1_UART
+		bool "i.MX1 Debug UART"
+		depends on SOC_IMX1
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX1.
+
+	config DEBUG_IMX23_UART
+		bool "i.MX23 Debug UART"
+		depends on SOC_IMX23
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX23.
+
+	config DEBUG_IMX25_UART
+		bool "i.MX25 Debug UART"
+		depends on SOC_IMX25
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX25.
+
+	config DEBUG_IMX21_IMX27_UART
+		bool "i.MX21 and i.MX27 Debug UART"
+		depends on SOC_IMX21 || SOC_IMX27
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX21 or i.MX27.
+
+	config DEBUG_IMX28_UART
+		bool "i.MX28 Debug UART"
+		depends on SOC_IMX28
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX28.
+
+	config DEBUG_IMX31_UART
+		bool "i.MX31 Debug UART"
+		depends on SOC_IMX31
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX31.
+
+	config DEBUG_IMX35_UART
+		bool "i.MX35 Debug UART"
+		depends on SOC_IMX35
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX35.
+
+	config DEBUG_IMX50_UART
+		bool "i.MX50 Debug UART"
+		depends on SOC_IMX50
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX50.
+
+	config DEBUG_IMX51_UART
+		bool "i.MX51 Debug UART"
+		depends on SOC_IMX51
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX51.
+
+	config DEBUG_IMX53_UART
+		bool "i.MX53 Debug UART"
+		depends on SOC_IMX53
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX53.
+
+	config DEBUG_IMX6Q_UART
+		bool "i.MX6Q/DL Debug UART"
+		depends on SOC_IMX6Q
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX6Q/DL.
+
+	config DEBUG_IMX6SL_UART
+		bool "i.MX6SL Debug UART"
+		depends on SOC_IMX6SL
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX6SL.
+
+	config DEBUG_IMX6SX_UART
+		bool "i.MX6SX Debug UART"
+		depends on SOC_IMX6SX
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX6SX.
+
+	config DEBUG_IMX6UL_UART
+		bool "i.MX6UL Debug UART"
+		depends on SOC_IMX6UL
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX6UL.
+
+	config DEBUG_IMX7D_UART
+		bool "i.MX7D Debug UART"
+		depends on SOC_IMX7D
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on i.MX7D.
+
+	config DEBUG_INTEGRATOR
+		bool "Kernel low-level debugging messages via ARM Integrator UART"
+		depends on ARCH_INTEGRATOR
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on ARM Integrator platforms.
+
+	config DEBUG_KEYSTONE_UART0
+		bool "Kernel low-level debugging on KEYSTONE2 using UART0"
+		depends on ARCH_KEYSTONE
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to UART0 serial port on KEYSTONE2 devices.
+
+	config DEBUG_KEYSTONE_UART1
+		bool "Kernel low-level debugging on KEYSTONE2 using UART1"
+		depends on ARCH_KEYSTONE
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to UART1 serial port on KEYSTONE2 devices.
+
+	config DEBUG_KS8695_UART
+		bool "KS8695 Debug UART"
+		depends on ARCH_KS8695
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on KS8695.
+
+	config DEBUG_LPC18XX_UART0
+		bool "Kernel low-level debugging via LPC18xx/43xx UART0"
+		depends on ARCH_LPC18XX
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on NXP LPC18xx/43xx UART0.
+
+	config DEBUG_LPC32XX
+		bool "Kernel low-level debugging messages via NXP LPC32xx UART"
+		depends on ARCH_LPC32XX
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on NXP LPC32xx based platforms.
+
+	config DEBUG_MESON_UARTAO
+		bool "Kernel low-level debugging via Meson6 UARTAO"
+		depends on ARCH_MESON
+		help
+		  Say Y here if you want kernel low-lever debugging support
+		  on Amlogic Meson6 based platforms on the UARTAO.
+
+	config DEBUG_MMP_UART2
+		bool "Kernel low-level debugging message via MMP UART2"
+		depends on ARCH_MMP
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on MMP UART2.
+
+	config DEBUG_MMP_UART3
+		bool "Kernel low-level debugging message via MMP UART3"
+		depends on ARCH_MMP
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on MMP UART3.
+
+	config DEBUG_MVEBU_UART0
+		bool "Kernel low-level debugging messages via MVEBU UART0 (old bootloaders)"
+		depends on ARCH_MVEBU
+		depends on ARCH_MVEBU && CPU_V7
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on MVEBU based platforms on UART0.
+
+		  This option should be used with the old bootloaders
+		  that left the internal registers mapped at
+		  0xd0000000. As of today, this is the case on
+		  platforms such as the Globalscale Mirabox or the
+		  Plathome OpenBlocks AX3, when using the original
+		  bootloader.
+
+		  This option will not work on older Marvell platforms
+		  (Kirkwood, Dove, MV78xx0, Orion5x), which should pick
+		  the "new bootloader" variant.
+
+		  If the wrong DEBUG_MVEBU_UART* option is selected,
+		  when u-boot hands over to the kernel, the system
+		  silently crashes, with no serial output at all.
+
+	config DEBUG_MVEBU_UART0_ALTERNATE
+		bool "Kernel low-level debugging messages via MVEBU UART0 (new bootloaders)"
+		depends on ARCH_MVEBU || ARCH_DOVE || ARCH_MV78XX0 || ARCH_ORION5X
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on MVEBU based platforms on UART0. (Armada XP, Armada 3xx,
+		  Kirkwood, Dove, MV78xx0, Orion5x).
+
+
+		  This option should be used with the new bootloaders
+		  that remap the internal registers at 0xf1000000.
+
+		  If the wrong DEBUG_MVEBU_UART* option is selected,
+		  when u-boot hands over to the kernel, the system
+		  silently crashes, with no serial output at all.
+
+	config DEBUG_MVEBU_UART1_ALTERNATE
+		bool "Kernel low-level debugging messages via MVEBU UART1 (new bootloaders)"
+		depends on ARCH_MVEBU
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on MVEBU based platforms on UART1. (Armada XP, Armada 3xx,
+		  Kirkwood, Dove, MV78xx0, Orion5x).
+
+		  This option should be used with the new bootloaders
+		  that remap the internal registers at 0xf1000000.
+		  All of the older (pre Armada XP/370) platforms also use
+		  this address, regardless of the boot loader version.
+
+		  If the wrong DEBUG_MVEBU_UART* option is selected,
+		  when u-boot hands over to the kernel, the system
+		  silently crashes, with no serial output at all.
+
+	config DEBUG_MT6589_UART0
+		bool "Mediatek mt6589 UART0"
+		depends on ARCH_MEDIATEK
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  for Mediatek mt6589 based platforms on UART0.
+
+	config DEBUG_MT8127_UART0
+		bool "Mediatek mt8127/mt6592 UART0"
+		depends on ARCH_MEDIATEK
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  for Mediatek mt8127 based platforms on UART0.
+
+	config DEBUG_MT8135_UART3
+		bool "Mediatek mt8135 UART3"
+		depends on ARCH_MEDIATEK
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  for Mediatek mt8135 based platforms on UART3.
+
+	config DEBUG_NETX_UART
+		bool "Kernel low-level debugging messages via NetX UART"
+		depends on ARCH_NETX
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Hilscher NetX based platforms.
+
+	config DEBUG_NOMADIK_UART
+		bool "Kernel low-level debugging messages via NOMADIK UART"
+		depends on ARCH_NOMADIK
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on NOMADIK based platforms.
+
+	config DEBUG_NSPIRE_CLASSIC_UART
+		bool "Kernel low-level debugging via TI-NSPIRE 8250 UART"
+		depends on ARCH_NSPIRE
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on TI-NSPIRE classic models.
+
+	config DEBUG_NSPIRE_CX_UART
+		bool "Kernel low-level debugging via TI-NSPIRE PL011 UART"
+		depends on ARCH_NSPIRE
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on TI-NSPIRE CX models.
+
+	config DEBUG_OMAP1UART1
+		bool "Kernel low-level debugging via OMAP1 UART1"
+		depends on ARCH_OMAP1
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on OMAP1 based platforms (except OMAP730) on the UART1.
+
+	config DEBUG_OMAP1UART2
+		bool "Kernel low-level debugging via OMAP1 UART2"
+		depends on ARCH_OMAP1
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on OMAP1 based platforms (except OMAP730) on the UART2.
+
+	config DEBUG_OMAP1UART3
+		bool "Kernel low-level debugging via OMAP1 UART3"
+		depends on ARCH_OMAP1
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on OMAP1 based platforms (except OMAP730) on the UART3.
+
+	config DEBUG_OMAP2UART1
+		bool "OMAP2/3/4 UART1 (omap2/3 sdp boards and some omap3 boards)"
+		depends on ARCH_OMAP2PLUS
+		select DEBUG_OMAP2PLUS_UART
+		help
+		  This covers at least h4, 2430sdp, 3430sdp, 3630sdp,
+		  omap3 torpedo and 3530 lv som.
+
+	config DEBUG_OMAP2UART2
+		bool "Kernel low-level debugging messages via OMAP2/3/4 UART2"
+		depends on ARCH_OMAP2PLUS
+		select DEBUG_OMAP2PLUS_UART
+
+	config DEBUG_OMAP2UART3
+		bool "Kernel low-level debugging messages via OMAP2 UART3 (n8x0)"
+		depends on ARCH_OMAP2PLUS
+		select DEBUG_OMAP2PLUS_UART
+
+	config DEBUG_OMAP3UART3
+		bool "Kernel low-level debugging messages via OMAP3 UART3 (most omap3 boards)"
+		depends on ARCH_OMAP2PLUS
+		select DEBUG_OMAP2PLUS_UART
+		help
+		  This covers at least cm_t3x, beagle, crane, devkit8000,
+		  igep00x0, ldp, n900, n9(50), pandora, overo, touchbook,
+		  and 3517evm.
+
+	config DEBUG_OMAP4UART3
+		bool "Kernel low-level debugging messages via OMAP4/5 UART3 (omap4 blaze, panda, omap5 sevm)"
+		depends on ARCH_OMAP2PLUS
+		select DEBUG_OMAP2PLUS_UART
+
+	config DEBUG_OMAP3UART4
+		bool "Kernel low-level debugging messages via OMAP36XX UART4"
+		depends on ARCH_OMAP2PLUS
+		select DEBUG_OMAP2PLUS_UART
+
+	config DEBUG_OMAP4UART4
+		bool "Kernel low-level debugging messages via OMAP4/5 UART4"
+		depends on ARCH_OMAP2PLUS
+		select DEBUG_OMAP2PLUS_UART
+
+	config DEBUG_OMAP7XXUART1
+		bool "Kernel low-level debugging via OMAP730 UART1"
+		depends on ARCH_OMAP730
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on OMAP730 based platforms on the UART1.
+
+	config DEBUG_OMAP7XXUART2
+		bool "Kernel low-level debugging via OMAP730 UART2"
+		depends on ARCH_OMAP730
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on OMAP730 based platforms on the UART2.
+
+	config DEBUG_OMAP7XXUART3
+		bool "Kernel low-level debugging via OMAP730 UART3"
+		depends on ARCH_OMAP730
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on OMAP730 based platforms on the UART3.
+
+	config DEBUG_TI81XXUART1
+		bool "Kernel low-level debugging messages via TI81XX UART1 (ti8148evm)"
+		depends on ARCH_OMAP2PLUS
+		select DEBUG_OMAP2PLUS_UART
+
+	config DEBUG_TI81XXUART2
+		bool "Kernel low-level debugging messages via TI81XX UART2"
+		depends on ARCH_OMAP2PLUS
+		select DEBUG_OMAP2PLUS_UART
+
+	config DEBUG_TI81XXUART3
+		bool "Kernel low-level debugging messages via TI81XX UART3 (ti8168evm)"
+		depends on ARCH_OMAP2PLUS
+		select DEBUG_OMAP2PLUS_UART
+
+	config DEBUG_AM33XXUART1
+		bool "Kernel low-level debugging messages via AM33XX UART1"
+		depends on ARCH_OMAP2PLUS
+		select DEBUG_OMAP2PLUS_UART
+
+	config DEBUG_ZOOM_UART
+		bool "Kernel low-level debugging messages via Zoom2/3 UART"
+		depends on ARCH_OMAP2PLUS
+		select DEBUG_OMAP2PLUS_UART
+
+	config DEBUG_PICOXCELL_UART
+		depends on ARCH_PICOXCELL
+		bool "Use PicoXcell UART for low-level debug"
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on PicoXcell based platforms.
+
+	config DEBUG_PXA_UART1
+		depends on ARCH_PXA
+		bool "Use PXA UART1 for low-level debug"
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on PXA UART1.
+
+	config DEBUG_QCOM_UARTDM
+		bool "Kernel low-level debugging messages via QCOM UARTDM"
+		depends on ARCH_QCOM
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the serial port on Qualcomm devices.
+
+		  ARCH      DEBUG_UART_PHYS   DEBUG_UART_VIRT
+		  APQ8064   0x16640000        0xf0040000
+		  APQ8084   0xf995e000        0xfa75e000
+		  MSM8X60   0x19c40000        0xf0040000
+		  MSM8960   0x16440000        0xf0040000
+		  MSM8974   0xf991e000        0xfa71e000
+
+		  Please adjust DEBUG_UART_PHYS and DEBUG_UART_BASE configuration
+		  options based on your needs.
+
+	config DEBUG_REALVIEW_STD_PORT
+		bool "RealView Default UART"
+		depends on ARCH_REALVIEW
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the serial port on RealView EB, PB11MP, PBA8
+		  and PBX platforms.
+
+	config DEBUG_REALVIEW_PB1176_PORT
+		bool "RealView PB1176 UART"
+		depends on MACH_REALVIEW_PB1176
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the standard serial port on the RealView
+		  PB1176 platform.
+
+	config DEBUG_RK29_UART0
+		bool "Kernel low-level debugging messages via Rockchip RK29 UART0"
+		depends on ARCH_ROCKCHIP
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Rockchip based platforms.
+
+	config DEBUG_RK29_UART1
+		bool "Kernel low-level debugging messages via Rockchip RK29 UART1"
+		depends on ARCH_ROCKCHIP
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Rockchip based platforms.
+
+	config DEBUG_RK29_UART2
+		bool "Kernel low-level debugging messages via Rockchip RK29 UART2"
+		depends on ARCH_ROCKCHIP
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Rockchip based platforms.
+
+	config DEBUG_RK3X_UART0
+		bool "Kernel low-level debugging messages via Rockchip RK30/RK31 UART0"
+		depends on ARCH_ROCKCHIP
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Rockchip based platforms.
+
+	config DEBUG_RK3X_UART1
+		bool "Kernel low-level debugging messages via Rockchip RK30/RK31 UART1"
+		depends on ARCH_ROCKCHIP
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Rockchip based platforms.
+
+	config DEBUG_RK3X_UART2
+		bool "Kernel low-level debugging messages via Rockchip RK30/RK31 UART2"
+		depends on ARCH_ROCKCHIP
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Rockchip based platforms.
+
+	config DEBUG_RK3X_UART3
+		bool "Kernel low-level debugging messages via Rockchip RK30/RK31 UART3"
+		depends on ARCH_ROCKCHIP
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Rockchip based platforms.
+
+	config DEBUG_RK32_UART2
+		bool "Kernel low-level debugging messages via Rockchip RK32 UART2"
+		depends on ARCH_ROCKCHIP
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Rockchip RK32xx based platforms.
+
+	config DEBUG_R7S72100_SCIF2
+		bool "Kernel low-level debugging messages via SCIF2 on R7S72100"
+		depends on ARCH_R7S72100
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  via SCIF2 on Renesas RZ/A1H (R7S72100).
+
+	config DEBUG_RCAR_GEN1_SCIF0
+		bool "Kernel low-level debugging messages via SCIF0 on R8A7778"
+		depends on ARCH_R8A7778
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  via SCIF0 on Renesas R-Car M1A (R8A7778).
+
+	config DEBUG_RCAR_GEN1_SCIF2
+		bool "Kernel low-level debugging messages via SCIF2 on R8A7779"
+		depends on ARCH_R8A7779
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  via SCIF2 on Renesas R-Car H1 (R8A7779).
+
+	config DEBUG_RCAR_GEN2_SCIF0
+		bool "Kernel low-level debugging messages via SCIF0 on R8A7790/R8A7791/R8A7792/R8A7793"
+		depends on ARCH_R8A7790 || ARCH_R8A7791 || ARCH_R8A7792 || ARCH_R8A7793
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  via SCIF0 on Renesas R-Car H2 (R8A7790), M2-W (R8A7791), V2H
+		  (R8A7792), or M2-N (R8A7793).
+
+	config DEBUG_RCAR_GEN2_SCIF2
+		bool "Kernel low-level debugging messages via SCIF2 on R8A7794"
+		depends on ARCH_R8A7794
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  via SCIF2 on Renesas R-Car E2 (R8A7794).
+
+	config DEBUG_RMOBILE_SCIFA0
+		bool "Kernel low-level debugging messages via SCIFA0 on R8A73A4"
+		depends on ARCH_R8A73A4
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  via SCIFA0 on Renesas R-Mobile APE6 (R8A73A4).
+
+	config DEBUG_RMOBILE_SCIFA1
+		bool "Kernel low-level debugging messages via SCIFA1 on R8A7740"
+		depends on ARCH_R8A7740
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  via SCIFA1 on Renesas R-Mobile A1 (R8A7740).
+
+	config DEBUG_RMOBILE_SCIFA4
+		bool "Kernel low-level debugging messages via SCIFA4 on SH73A0"
+		depends on ARCH_SH73A0
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  via SCIFA4 on Renesas SH-Mobile AG5 (SH73A0).
+
+	config DEBUG_S3C_UART0
+		depends on PLAT_SAMSUNG
+		select DEBUG_EXYNOS_UART if ARCH_EXYNOS
+		select DEBUG_S3C24XX_UART if ARCH_S3C24XX
+		select DEBUG_S3C64XX_UART if ARCH_S3C64XX
+		select DEBUG_S5PV210_UART if ARCH_S5PV210
+		bool "Use Samsung S3C UART 0 for low-level debug"
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to UART 0. The port must have been initialised
+		  by the boot-loader before use.
+
+	config DEBUG_S3C_UART1
+		depends on PLAT_SAMSUNG
+		select DEBUG_EXYNOS_UART if ARCH_EXYNOS
+		select DEBUG_S3C24XX_UART if ARCH_S3C24XX
+		select DEBUG_S3C64XX_UART if ARCH_S3C64XX
+		select DEBUG_S5PV210_UART if ARCH_S5PV210
+		bool "Use Samsung S3C UART 1 for low-level debug"
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to UART 1. The port must have been initialised
+		  by the boot-loader before use.
+
+	config DEBUG_S3C_UART2
+		depends on PLAT_SAMSUNG
+		select DEBUG_EXYNOS_UART if ARCH_EXYNOS
+		select DEBUG_S3C24XX_UART if ARCH_S3C24XX
+		select DEBUG_S3C64XX_UART if ARCH_S3C64XX
+		select DEBUG_S5PV210_UART if ARCH_S5PV210
+		bool "Use Samsung S3C UART 2 for low-level debug"
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to UART 2. The port must have been initialised
+		  by the boot-loader before use.
+
+	config DEBUG_S3C_UART3
+		depends on PLAT_SAMSUNG && (ARCH_EXYNOS || ARCH_S5PV210)
+		select DEBUG_EXYNOS_UART if ARCH_EXYNOS
+		select DEBUG_S3C64XX_UART if ARCH_S3C64XX
+		select DEBUG_S5PV210_UART if ARCH_S5PV210
+		bool "Use Samsung S3C UART 3 for low-level debug"
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to UART 3. The port must have been initialised
+		  by the boot-loader before use.
+
+	config DEBUG_S3C2410_UART0
+		depends on ARCH_S3C24XX
+		select DEBUG_S3C2410_UART
+		bool "Use S3C2410/S3C2412 UART 0 for low-level debug"
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to UART 0. The port must have been initialised
+		  by the boot-loader before use.
+
+	config DEBUG_S3C2410_UART1
+		depends on ARCH_S3C24XX
+		select DEBUG_S3C2410_UART
+		bool "Use S3C2410/S3C2412 UART 1 for low-level debug"
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to UART 1. The port must have been initialised
+		  by the boot-loader before use.
+
+	config DEBUG_S3C2410_UART2
+		depends on ARCH_S3C24XX
+		select DEBUG_S3C2410_UART
+		bool "Use S3C2410/S3C2412 UART 2 for low-level debug"
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to UART 2. The port must have been initialised
+		  by the boot-loader before use.
+
+	config DEBUG_SA1100
+		depends on ARCH_SA1100
+		bool "Use SA1100 UARTs for low-level debug"
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on SA-11x0 UART ports. The kernel will check for the first
+		  enabled UART in a sequence 3-1-2.
+
+	config DEBUG_SOCFPGA_UART0
+		depends on ARCH_SOCFPGA
+		bool "Use SOCFPGA UART0 for low-level debug"
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on SOCFPGA(Cyclone 5 and Arria 5) based platforms.
+
+	config DEBUG_SOCFPGA_UART1
+		depends on ARCH_SOCFPGA
+		bool "Use SOCFPGA UART1 for low-level debug"
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on SOCFPGA(Arria 10) based platforms.
+
+
+	config DEBUG_SUN9I_UART0
+		bool "Kernel low-level debugging messages via sun9i UART0"
+		depends on MACH_SUN9I
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Allwinner A80 based platforms on the UART0.
+
+	config DEBUG_SUNXI_UART0
+		bool "Kernel low-level debugging messages via sunXi UART0"
+		depends on ARCH_SUNXI
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Allwinner A1X based platforms on the UART0.
+
+	config DEBUG_SUNXI_UART1
+		bool "Kernel low-level debugging messages via sunXi UART1"
+		depends on ARCH_SUNXI
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Allwinner A1X based platforms on the UART1.
+
+	config DEBUG_SUNXI_R_UART
+		bool "Kernel low-level debugging messages via sunXi R_UART"
+		depends on MACH_SUN6I || MACH_SUN8I
+		select DEBUG_UART_8250
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Allwinner A31/A23 based platforms on the R_UART.
+
+	config DEBUG_SIRFPRIMA2_UART1
+		bool "Kernel low-level debugging messages via SiRFprimaII UART1"
+		depends on ARCH_PRIMA2
+		select DEBUG_SIRFSOC_UART
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the uart1 port on SiRFprimaII devices.
+
+	config DEBUG_SIRFATLAS7_UART0
+		bool "Kernel low-level debugging messages via SiRFatlas7 UART0"
+		depends on ARCH_ATLAS7
+		select DEBUG_SIRFSOC_UART
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the uart0 port on SiRFATLAS7 devices.The uart0
+		  is used on SiRFATLAS7 as a extra debug port.sometimes an extra
+		  debug port can be very useful.
+
+	config DEBUG_SIRFATLAS7_UART1
+		bool "Kernel low-level debugging messages via SiRFatlas7 UART1"
+		depends on ARCH_ATLAS7
+		select DEBUG_SIRFSOC_UART
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the uart1 port on SiRFATLAS7 devices.
+
+	config DEBUG_SPEAR3XX
+		bool "Kernel low-level debugging messages via ST SPEAr 3xx/6xx UART"
+		depends on ARCH_SPEAR3XX || ARCH_SPEAR6XX
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on ST SPEAr based platforms.
+
+	config DEBUG_SPEAR13XX
+		bool "Kernel low-level debugging messages via ST SPEAr 13xx UART"
+		depends on ARCH_SPEAR13XX
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on ST SPEAr13xx based platforms.
+
+	config STIH41X_DEBUG_ASC2
+		bool "Use StiH415/416 ASC2 UART for low-level debug"
+		depends on ARCH_STI
+		select DEBUG_STI_UART
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on STiH415/416 based platforms like b2000, which has
+		  default UART wired up to ASC2.
+
+		  If unsure, say N.
+
+	config STIH41X_DEBUG_SBC_ASC1
+		bool "Use StiH415/416 SBC ASC1 UART for low-level debug"
+		depends on ARCH_STI
+		select DEBUG_STI_UART
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on STiH415/416 based platforms like b2020. which has
+		  default UART wired up to SBC ASC1.
+
+		  If unsure, say N.
+
+	config TEGRA_DEBUG_UART_AUTO_ODMDATA
+		bool "Kernel low-level debugging messages via Tegra UART via ODMDATA"
+		depends on ARCH_TEGRA
+		select DEBUG_TEGRA_UART
+		help
+		  Automatically determines which UART to use for low-level
+		  debug based on the ODMDATA value. This value is part of
+		  the BCT, and is written to the boot memory device using
+		  nvflash, or other flashing tool.  When bits 19:18 are 3,
+		  then bits 17:15 indicate which UART to use; 0/1/2/3/4
+		  are UART A/B/C/D/E.
+
+	config TEGRA_DEBUG_UARTA
+		bool "Kernel low-level debugging messages via Tegra UART A"
+		depends on ARCH_TEGRA
+		select DEBUG_TEGRA_UART
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Tegra based platforms.
+
+	config TEGRA_DEBUG_UARTB
+		bool "Kernel low-level debugging messages via Tegra UART B"
+		depends on ARCH_TEGRA
+		select DEBUG_TEGRA_UART
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Tegra based platforms.
+
+	config TEGRA_DEBUG_UARTC
+		bool "Kernel low-level debugging messages via Tegra UART C"
+		depends on ARCH_TEGRA
+		select DEBUG_TEGRA_UART
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Tegra based platforms.
+
+	config TEGRA_DEBUG_UARTD
+		bool "Kernel low-level debugging messages via Tegra UART D"
+		depends on ARCH_TEGRA
+		select DEBUG_TEGRA_UART
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Tegra based platforms.
+
+	config TEGRA_DEBUG_UARTE
+		bool "Kernel low-level debugging messages via Tegra UART E"
+		depends on ARCH_TEGRA
+		select DEBUG_TEGRA_UART
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Tegra based platforms.
+
+	config DEBUG_U300_UART
+		bool "Kernel low-level debugging messages via U300 UART0"
+		depends on ARCH_U300
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the uart port on U300 devices.
+
+	config DEBUG_UX500_UART
+		depends on ARCH_U8500
+		bool "Use Ux500 UART for low-level debug"
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Ux500 based platforms.
+
+	config DEBUG_VERSATILE
+		bool "Kernel low-level debugging messages via ARM Versatile UART"
+		depends on ARCH_VERSATILE
+		select DEBUG_UART_PL01X
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on ARM Versatile platforms.
+
+	config DEBUG_VEXPRESS_UART0_DETECT
+		bool "Autodetect UART0 on Versatile Express Cortex-A core tiles"
+		depends on ARCH_VEXPRESS && CPU_CP15_MMU
+		help
+		  This option enables a simple heuristic which tries to determine
+		  the motherboard's memory map variant (original or RS1) and then
+		  choose the relevant UART0 base address.
+
+		  Note that this will only work with standard A-class core tiles,
+		  and may fail with non-standard SMM or custom software models.
+
+	config DEBUG_VEXPRESS_UART0_CA9
+		bool "Use PL011 UART0 at 0x10009000 (V2P-CA9 core tile)"
+		depends on ARCH_VEXPRESS
+		select DEBUG_UART_PL01X
+		help
+		  This option selects UART0 at 0x10009000. Except for custom models,
+		  this applies only to the V2P-CA9 tile.
+
+	config DEBUG_VEXPRESS_UART0_RS1
+		bool "Use PL011 UART0 at 0x1c090000 (RS1 complaint tiles)"
+		depends on ARCH_VEXPRESS
+		select DEBUG_UART_PL01X
+		help
+		  This option selects UART0 at 0x1c090000. This applies to most
+		  of the tiles using the RS1 memory map, including all new A-class
+		  core tiles, FPGA-based SMMs and software models.
+
+	config DEBUG_VEXPRESS_UART0_CRX
+		bool "Use PL011 UART0 at 0xb0090000 (Cortex-R compliant tiles)"
+		depends on ARCH_VEXPRESS && !MMU
+		select DEBUG_UART_PL01X
+		help
+		  This option selects UART0 at 0xb0090000. This is appropriate for
+		  Cortex-R series tiles and SMMs, such as Cortex-R5 and Cortex-R7
+
+	config DEBUG_VF_UART
+		bool "Vybrid UART"
+		depends on SOC_VF610
+		help
+		  Say Y here if you want kernel low-level debugging support
+		  on Vybrid based platforms.
+
+	config DEBUG_VT8500_UART0
+		bool "Use UART0 on VIA/Wondermedia SoCs"
+		depends on ARCH_VT8500
+		help
+		  This option selects UART0 on VIA/Wondermedia System-on-a-chip
+		  devices, including VT8500, WM8505, WM8650 and WM8850.
+
+	config DEBUG_ZTE_ZX
+		bool "Use ZTE ZX UART"
+		select DEBUG_UART_PL01X
+		depends on ARCH_ZX
+		help
+		  Say Y here if you are enabling ZTE ZX296702 SOC and need
+		  debug uart support.
+
+		  This option is preferred over the platform specific
+		  options; the platform specific options are deprecated
+		  and will be soon removed.
+
+	config DEBUG_ZYNQ_UART0
+		bool "Kernel low-level debugging on Xilinx Zynq using UART0"
+		depends on ARCH_ZYNQ
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to UART0 on the Zynq platform.
+
+	config DEBUG_ZYNQ_UART1
+		bool "Kernel low-level debugging on Xilinx Zynq using UART1"
+		depends on ARCH_ZYNQ
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to UART1 on the Zynq platform.
+
+		  If you have a ZC702 board and want early boot messages to
+		  appear on the USB serial adaptor, select this option.
+
+	config DEBUG_ICEDCC
+		bool "Kernel low-level debugging via EmbeddedICE DCC channel"
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to the EmbeddedICE macrocell's DCC channel using
+		  co-processor 14. This is known to work on the ARM9 style ICE
+		  channel and on the XScale with the PEEDI.
+
+		  Note that the system will appear to hang during boot if there
+		  is nothing connected to read from the DCC.
+
+	config DEBUG_SEMIHOSTING
+		bool "Kernel low-level debug output via semihosting I/O"
+		help
+		  Semihosting enables code running on an ARM target to use
+		  the I/O facilities on a host debugger/emulator through a
+		  simple SVC call. The host debugger or emulator must have
+		  semihosting enabled for the special svc call to be trapped
+		  otherwise the kernel will crash.
+
+		  This is known to work with OpenOCD, as well as
+		  ARM's Fast Models, or any other controlling environment
+		  that implements semihosting.
+
+		  For more details about semihosting, please see
+		  chapter 8 of DUI0203I_rvct_developer_guide.pdf from ARM Ltd.
+
+	config DEBUG_LL_UART_8250
+		bool "Kernel low-level debugging via 8250 UART"
+		help
+		  Say Y here if you wish the debug print routes to direct
+		  their output to an 8250 UART.  You can use this option
+		  to provide the parameters for the 8250 UART rather than
+		  selecting one of the platform specific options above if
+		  you know the parameters for the port.
+
+		  This option is preferred over the platform specific
+		  options; the platform specific options are deprecated
+		  and will be soon removed.
+
+	config DEBUG_LL_UART_EFM32
+		bool "Kernel low-level debugging via efm32 UART"
+		depends on ARCH_EFM32
+		help
+		  Say Y here if you want the debug print routines to direct
+		  their output to an UART or USART port on efm32 based
+		  machines. Use the following addresses for DEBUG_UART_PHYS:
+
+		    0x4000c000 | USART0
+		    0x4000c400 | USART1
+		    0x4000c800 | USART2
+		    0x4000e000 | UART0
+		    0x4000e400 | UART1
+
+	config DEBUG_LL_UART_PL01X
+		bool "Kernel low-level debugging via ARM Ltd PL01x Primecell UART"
+		help
+		  Say Y here if you wish the debug print routes to direct
+		  their output to a PL01x Primecell UART.  You can use
+		  this option to provide the parameters for the UART
+		  rather than selecting one of the platform specific
+		  options above if you know the parameters for the port.
+
+		  This option is preferred over the platform specific
+		  options; the platform specific options are deprecated
+		  and will be soon removed.
+
+endchoice
+
+config DEBUG_AT91_UART
+	bool
+	depends on ARCH_AT91
+
+config DEBUG_EXYNOS_UART
+	bool
+
+config DEBUG_S3C2410_UART
+	bool
+	select DEBUG_S3C24XX_UART
+
+config DEBUG_S3C24XX_UART
+	bool
+
+config DEBUG_S3C64XX_UART
+	bool
+
+config DEBUG_S5PV210_UART
+	bool
+
+config DEBUG_OMAP2PLUS_UART
+	bool
+	depends on ARCH_OMAP2PLUS
+
+config DEBUG_IMX_UART_PORT
+	int "i.MX Debug UART Port Selection" if DEBUG_IMX1_UART || \
+						DEBUG_IMX25_UART || \
+						DEBUG_IMX21_IMX27_UART || \
+						DEBUG_IMX31_UART || \
+						DEBUG_IMX35_UART || \
+						DEBUG_IMX50_UART || \
+						DEBUG_IMX51_UART || \
+						DEBUG_IMX53_UART || \
+						DEBUG_IMX6Q_UART || \
+						DEBUG_IMX6SL_UART || \
+						DEBUG_IMX6SX_UART || \
+						DEBUG_IMX6UL_UART || \
+						DEBUG_IMX7D_UART
+	default 1
+	depends on ARCH_MXC
+	help
+	  Choose UART port on which kernel low-level debug messages
+	  should be output.
+
+config DEBUG_VF_UART_PORT
+	int "Vybrid Debug UART Port Selection" if DEBUG_VF_UART
+	default 1
+	range 0 3
+	depends on SOC_VF610
+	help
+	  Choose UART port on which kernel low-level debug messages
+	  should be output.
+
+config DEBUG_TEGRA_UART
+	bool
+	depends on ARCH_TEGRA
+
+config DEBUG_STI_UART
+	bool
+	depends on ARCH_STI
+
+config DEBUG_SIRFSOC_UART
+	bool
+	depends on ARCH_SIRF
+
+config DEBUG_LL_INCLUDE
+	string
+	default "debug/sa1100.S" if DEBUG_SA1100
+	default "debug/palmchip.S" if DEBUG_UART_8250_PALMCHIP
+	default "debug/8250.S" if DEBUG_LL_UART_8250 || DEBUG_UART_8250
+	default "debug/at91.S" if DEBUG_AT91_UART
+	default "debug/asm9260.S" if DEBUG_ASM9260_UART
+	default "debug/clps711x.S" if DEBUG_CLPS711X_UART1 || DEBUG_CLPS711X_UART2
+	default "debug/dc21285.S" if DEBUG_DC21285_PORT
+	default "debug/meson.S" if DEBUG_MESON_UARTAO
+	default "debug/pl01x.S" if DEBUG_LL_UART_PL01X || DEBUG_UART_PL01X
+	default "debug/exynos.S" if DEBUG_EXYNOS_UART
+	default "debug/efm32.S" if DEBUG_LL_UART_EFM32
+	default "debug/icedcc.S" if DEBUG_ICEDCC
+	default "debug/imx.S" if DEBUG_IMX1_UART || \
+				 DEBUG_IMX25_UART || \
+				 DEBUG_IMX21_IMX27_UART || \
+				 DEBUG_IMX31_UART || \
+				 DEBUG_IMX35_UART || \
+				 DEBUG_IMX50_UART || \
+				 DEBUG_IMX51_UART || \
+				 DEBUG_IMX53_UART ||\
+				 DEBUG_IMX6Q_UART || \
+				 DEBUG_IMX6SL_UART || \
+				 DEBUG_IMX6SX_UART || \
+				 DEBUG_IMX6UL_UART || \
+				 DEBUG_IMX7D_UART
+	default "debug/ks8695.S" if DEBUG_KS8695_UART
+	default "debug/msm.S" if DEBUG_QCOM_UARTDM
+	default "debug/netx.S" if DEBUG_NETX_UART
+	default "debug/omap2plus.S" if DEBUG_OMAP2PLUS_UART
+	default "debug/renesas-scif.S" if DEBUG_R7S72100_SCIF2
+	default "debug/renesas-scif.S" if DEBUG_RCAR_GEN1_SCIF0
+	default "debug/renesas-scif.S" if DEBUG_RCAR_GEN1_SCIF2
+	default "debug/renesas-scif.S" if DEBUG_RCAR_GEN2_SCIF0
+	default "debug/renesas-scif.S" if DEBUG_RCAR_GEN2_SCIF2
+	default "debug/renesas-scif.S" if DEBUG_RMOBILE_SCIFA0
+	default "debug/renesas-scif.S" if DEBUG_RMOBILE_SCIFA1
+	default "debug/renesas-scif.S" if DEBUG_RMOBILE_SCIFA4
+	default "debug/s3c24xx.S" if DEBUG_S3C24XX_UART || DEBUG_S3C64XX_UART
+	default "debug/s5pv210.S" if DEBUG_S5PV210_UART
+	default "debug/sirf.S" if DEBUG_SIRFSOC_UART
+	default "debug/sti.S" if DEBUG_STI_UART
+	default "debug/tegra.S" if DEBUG_TEGRA_UART
+	default "debug/ux500.S" if DEBUG_UX500_UART
+	default "debug/vexpress.S" if DEBUG_VEXPRESS_UART0_DETECT
+	default "debug/vf.S" if DEBUG_VF_UART
+	default "debug/vt8500.S" if DEBUG_VT8500_UART0
+	default "debug/zynq.S" if DEBUG_ZYNQ_UART0 || DEBUG_ZYNQ_UART1
+	default "debug/bcm63xx.S" if DEBUG_BCM63XX_UART
+	default "debug/digicolor.S" if DEBUG_DIGICOLOR_UA0
+	default "debug/brcmstb.S" if DEBUG_BRCMSTB_UART
+	default "mach/debug-macro.S"
+
+# Compatibility options for PL01x
+config DEBUG_UART_PL01X
+	bool
+
+# Compatibility options for 8250
+config DEBUG_UART_8250
+	def_bool ARCH_EBSA110 || \
+		ARCH_IOP13XX || ARCH_IOP32X || ARCH_IOP33X || ARCH_IXP4XX || \
+		ARCH_RPC
+
+config DEBUG_UART_PHYS
+	hex "Physical base address of debug UART"
+	default 0x00100a00 if DEBUG_NETX_UART
+	default 0x01c20000 if DEBUG_DAVINCI_DMx_UART0
+	default 0x01c28000 if DEBUG_SUNXI_UART0
+	default 0x01c28400 if DEBUG_SUNXI_UART1
+	default 0x01d0c000 if DEBUG_DAVINCI_DA8XX_UART1
+	default 0x01d0d000 if DEBUG_DAVINCI_DA8XX_UART2
+	default 0x01f02800 if DEBUG_SUNXI_R_UART
+	default 0x02530c00 if DEBUG_KEYSTONE_UART0
+	default 0x02531000 if DEBUG_KEYSTONE_UART1
+	default 0x03010fe0 if ARCH_RPC
+	default 0x07000000 if DEBUG_SUN9I_UART0
+	default 0x09405000 if DEBUG_ZTE_ZX
+	default 0x10009000 if DEBUG_REALVIEW_STD_PORT || \
+				DEBUG_VEXPRESS_UART0_CA9
+	default 0x1010c000 if DEBUG_REALVIEW_PB1176_PORT
+	default 0x10124000 if DEBUG_RK3X_UART0
+	default 0x10126000 if DEBUG_RK3X_UART1
+	default 0x101f1000 if DEBUG_VERSATILE
+	default 0x101fb000 if DEBUG_NOMADIK_UART
+	default 0x11002000 if DEBUG_MT8127_UART0
+	default 0x11006000 if DEBUG_MT6589_UART0
+	default 0x11009000 if DEBUG_MT8135_UART3
+	default 0x16000000 if DEBUG_INTEGRATOR
+	default 0x18000300 if DEBUG_BCM_5301X
+	default 0x18010000 if DEBUG_SIRFATLAS7_UART0
+	default 0x18020000 if DEBUG_SIRFATLAS7_UART1
+	default 0x1c090000 if DEBUG_VEXPRESS_UART0_RS1
+	default 0x20001000 if DEBUG_HIP01_UART
+	default 0x20060000 if DEBUG_RK29_UART0
+	default 0x20064000 if DEBUG_RK29_UART1 || DEBUG_RK3X_UART2
+	default 0x20068000 if DEBUG_RK29_UART2 || DEBUG_RK3X_UART3
+	default 0x20201000 if DEBUG_BCM2835
+	default 0x3f201000 if DEBUG_BCM2836
+	default 0x3e000000 if DEBUG_BCM_KONA_UART
+	default 0x4000e400 if DEBUG_LL_UART_EFM32
+	default 0x40081000 if DEBUG_LPC18XX_UART0
+	default 0x40090000 if DEBUG_LPC32XX
+	default 0x40100000 if DEBUG_PXA_UART1
+	default 0x42000000 if DEBUG_GEMINI
+	default 0x50000000 if DEBUG_S3C24XX_UART && (DEBUG_S3C_UART0 || \
+				DEBUG_S3C2410_UART0)
+	default 0x50004000 if DEBUG_S3C24XX_UART && (DEBUG_S3C_UART1 || \
+				DEBUG_S3C2410_UART1)
+	default 0x50008000 if DEBUG_S3C24XX_UART && (DEBUG_S3C_UART2 || \
+				DEBUG_S3C2410_UART2)
+	default 0x78000000 if DEBUG_CNS3XXX
+	default 0x7c0003f8 if DEBUG_FOOTBRIDGE_COM1
+	default 0x7f005000 if DEBUG_S3C64XX_UART && DEBUG_S3C_UART0
+	default 0x7f005400 if DEBUG_S3C64XX_UART && DEBUG_S3C_UART1
+	default 0x7f005800 if DEBUG_S3C64XX_UART && DEBUG_S3C_UART2
+	default 0x7f005c00 if DEBUG_S3C64XX_UART && DEBUG_S3C_UART3
+	default 0x80010000 if DEBUG_ASM9260_UART
+	default 0x80070000 if DEBUG_IMX23_UART
+	default 0x80074000 if DEBUG_IMX28_UART
+	default 0x80230000 if DEBUG_PICOXCELL_UART
+	default 0x808c0000 if DEBUG_EP93XX || ARCH_EP93XX
+	default 0x90020000 if DEBUG_NSPIRE_CLASSIC_UART || DEBUG_NSPIRE_CX_UART
+	default 0xb0060000 if DEBUG_SIRFPRIMA2_UART1
+	default 0xb0090000 if DEBUG_VEXPRESS_UART0_CRX
+	default 0xc0013000 if DEBUG_U300_UART
+	default 0xc8000000 if ARCH_IXP4XX && !CPU_BIG_ENDIAN
+	default 0xc8000003 if ARCH_IXP4XX && CPU_BIG_ENDIAN
+	default 0xd0000000 if DEBUG_SPEAR3XX
+	default 0xd0012000 if DEBUG_MVEBU_UART0
+	default 0xc81004c0 if DEBUG_MESON_UARTAO
+	default 0xd4017000 if DEBUG_MMP_UART2
+	default 0xd4018000 if DEBUG_MMP_UART3
+	default 0xe0000000 if DEBUG_SPEAR13XX
+	default 0xe4007000 if DEBUG_HIP04_UART
+	default 0xe6c40000 if DEBUG_RMOBILE_SCIFA0
+	default 0xe6c50000 if DEBUG_RMOBILE_SCIFA1
+	default 0xe6c80000 if DEBUG_RMOBILE_SCIFA4
+	default 0xe6e58000 if DEBUG_RCAR_GEN2_SCIF2
+	default 0xe6e60000 if DEBUG_RCAR_GEN2_SCIF0
+	default 0xe8008000 if DEBUG_R7S72100_SCIF2
+	default 0xf0000be0 if ARCH_EBSA110
+	default 0xf1012000 if DEBUG_MVEBU_UART0_ALTERNATE
+	default 0xf1012100 if DEBUG_MVEBU_UART1_ALTERNATE
+	default 0xf7fc9000 if DEBUG_BERLIN_UART
+	default 0xf8020000 if DEBUG_AT91_SAMA5D2_UART1
+	default 0xf8b00000 if DEBUG_HIX5HD2_UART
+	default 0xf991e000 if DEBUG_QCOM_UARTDM
+	default 0xfc00c000 if DEBUG_AT91_SAMA5D4_USART3
+	default 0xfcb00000 if DEBUG_HI3620_UART
+	default 0xfd883000 if DEBUG_ALPINE_UART0
+	default 0xfe800000 if ARCH_IOP32X
+	default 0xff690000 if DEBUG_RK32_UART2
+	default 0xffc02000 if DEBUG_SOCFPGA_UART0
+	default 0xffc02100 if DEBUG_SOCFPGA_UART1
+	default 0xffd82340 if ARCH_IOP13XX
+	default 0xffe40000 if DEBUG_RCAR_GEN1_SCIF0
+	default 0xffe42000 if DEBUG_RCAR_GEN1_SCIF2
+	default 0xfff36000 if DEBUG_HIGHBANK_UART
+	default 0xfffb0000 if DEBUG_OMAP1UART1 || DEBUG_OMAP7XXUART1
+	default 0xfffb0800 if DEBUG_OMAP1UART2 || DEBUG_OMAP7XXUART2
+	default 0xfffb9800 if DEBUG_OMAP1UART3 || DEBUG_OMAP7XXUART3
+	default 0xfffe8600 if DEBUG_BCM63XX_UART
+	default 0xffffee00 if DEBUG_AT91_SAM9263_DBGU
+	default 0xfffff200 if DEBUG_AT91_RM9200_DBGU
+	default 0xfffff700 if ARCH_IOP33X
+	depends on ARCH_EP93XX || \
+	        DEBUG_LL_UART_8250 || DEBUG_LL_UART_PL01X || \
+		DEBUG_LL_UART_EFM32 || \
+		DEBUG_UART_8250 || DEBUG_UART_PL01X || DEBUG_MESON_UARTAO || \
+		DEBUG_NETX_UART || \
+		DEBUG_QCOM_UARTDM || DEBUG_R7S72100_SCIF2 || \
+		DEBUG_RCAR_GEN1_SCIF0 || DEBUG_RCAR_GEN1_SCIF2 || \
+		DEBUG_RCAR_GEN2_SCIF0 || DEBUG_RCAR_GEN2_SCIF2 || \
+		DEBUG_RMOBILE_SCIFA0 || DEBUG_RMOBILE_SCIFA1 || \
+		DEBUG_RMOBILE_SCIFA4 || DEBUG_S3C24XX_UART || \
+		DEBUG_S3C64XX_UART || \
+		DEBUG_BCM63XX_UART || DEBUG_ASM9260_UART || \
+		DEBUG_SIRFSOC_UART || DEBUG_DIGICOLOR_UA0 || \
+		DEBUG_AT91_UART
+
+config DEBUG_UART_VIRT
+	hex "Virtual base address of debug UART"
+	default 0xe0000a00 if DEBUG_NETX_UART
+	default 0xe0010fe0 if ARCH_RPC
+	default 0xf0000be0 if ARCH_EBSA110
+	default 0xf0010000 if DEBUG_ASM9260_UART
+	default 0xf01fb000 if DEBUG_NOMADIK_UART
+	default 0xf0201000 if DEBUG_BCM2835 || DEBUG_BCM2836
+	default 0xf1000300 if DEBUG_BCM_5301X
+	default 0xf1002000 if DEBUG_MT8127_UART0
+	default 0xf1006000 if DEBUG_MT6589_UART0
+	default 0xf1009000 if DEBUG_MT8135_UART3
+	default 0xf11f1000 if DEBUG_VERSATILE
+	default 0xf1600000 if DEBUG_INTEGRATOR
+	default 0xf1c28000 if DEBUG_SUNXI_UART0
+	default 0xf1c28400 if DEBUG_SUNXI_UART1
+	default 0xf1f02800 if DEBUG_SUNXI_R_UART
+	default 0xf31004c0 if DEBUG_MESON_UARTAO
+	default 0xf4090000 if DEBUG_LPC32XX
+	default 0xf4200000 if DEBUG_GEMINI
+	default 0xf6200000 if DEBUG_PXA_UART1
+	default 0xf7000000 if DEBUG_SUN9I_UART0
+	default 0xf7000000 if DEBUG_S3C64XX_UART && DEBUG_S3C_UART0
+	default 0xf7000000 if DEBUG_S3C24XX_UART && (DEBUG_S3C_UART0 || \
+				DEBUG_S3C2410_UART0)
+	default 0xf7000400 if DEBUG_S3C64XX_UART && DEBUG_S3C_UART1
+	default 0xf7000800 if DEBUG_S3C64XX_UART && DEBUG_S3C_UART2
+	default 0xf7000c00 if DEBUG_S3C64XX_UART && DEBUG_S3C_UART3
+	default 0xf7004000 if DEBUG_S3C24XX_UART && (DEBUG_S3C_UART1 || \
+				DEBUG_S3C2410_UART1)
+	default 0xf7008000 if DEBUG_S3C24XX_UART && (DEBUG_S3C_UART2 || \
+				DEBUG_S3C2410_UART2)
+	default 0xf7020000 if DEBUG_AT91_SAMA5D2_UART1
+	default 0xf7fc9000 if DEBUG_BERLIN_UART
+	default 0xf8007000 if DEBUG_HIP04_UART
+	default 0xf8009000 if DEBUG_VEXPRESS_UART0_CA9
+	default 0xf8090000 if DEBUG_VEXPRESS_UART0_RS1
+	default 0xf8ffee00 if DEBUG_AT91_SAM9263_DBGU
+	default 0xf8fff200 if DEBUG_AT91_RM9200_DBGU
+	default 0xfa71e000 if DEBUG_QCOM_UARTDM
+	default 0xfb002000 if DEBUG_CNS3XXX
+	default 0xfb009000 if DEBUG_REALVIEW_STD_PORT
+	default 0xfb00c000 if DEBUG_AT91_SAMA5D4_USART3
+	default 0xfb10c000 if DEBUG_REALVIEW_PB1176_PORT
+	default 0xfc705000 if DEBUG_ZTE_ZX
+	default 0xfcfe8600 if DEBUG_BCM63XX_UART
+	default 0xfd000000 if DEBUG_SPEAR3XX || DEBUG_SPEAR13XX
+	default 0xfd012000 if DEBUG_MVEBU_UART0_ALTERNATE && ARCH_MV78XX0
+	default 0xfd883000 if DEBUG_ALPINE_UART0
+	default 0xfde12000 if DEBUG_MVEBU_UART0_ALTERNATE && ARCH_DOVE
+	default 0xfe012000 if DEBUG_MVEBU_UART0_ALTERNATE && ARCH_ORION5X
+	default 0xfe017000 if DEBUG_MMP_UART2
+	default 0xfe018000 if DEBUG_MMP_UART3
+	default 0xfe100000 if DEBUG_IMX23_UART || DEBUG_IMX28_UART
+	default 0xfe230000 if DEBUG_PICOXCELL_UART
+	default 0xfe300000 if DEBUG_BCM_KONA_UART
+	default 0xfe800000 if ARCH_IOP32X
+	default 0xfeb00000 if DEBUG_HI3620_UART || DEBUG_HIX5HD2_UART
+	default 0xfeb24000 if DEBUG_RK3X_UART0
+	default 0xfeb26000 if DEBUG_RK3X_UART1
+	default 0xfeb30c00 if DEBUG_KEYSTONE_UART0
+	default 0xfeb31000 if DEBUG_KEYSTONE_UART1
+	default 0xfec02000 if DEBUG_SOCFPGA_UART0
+	default 0xfec02100 if DEBUG_SOCFPGA_UART1
+	default 0xfec12000 if (DEBUG_MVEBU_UART0 || DEBUG_MVEBU_UART0_ALTERNATE) && ARCH_MVEBU
+	default 0xfec12100 if DEBUG_MVEBU_UART1_ALTERNATE
+	default 0xfec10000 if DEBUG_SIRFATLAS7_UART0
+	default 0xfec20000 if DEBUG_DAVINCI_DMx_UART0
+	default 0xfec20000 if DEBUG_SIRFATLAS7_UART1
+	default 0xfec60000 if DEBUG_SIRFPRIMA2_UART1
+	default 0xfec90000 if DEBUG_RK32_UART2
+	default 0xfed0c000 if DEBUG_DAVINCI_DA8XX_UART1
+	default 0xfed0d000 if DEBUG_DAVINCI_DA8XX_UART2
+	default 0xfed60000 if DEBUG_RK29_UART0
+	default 0xfed64000 if DEBUG_RK29_UART1 || DEBUG_RK3X_UART2
+	default 0xfed68000 if DEBUG_RK29_UART2 || DEBUG_RK3X_UART3
+	default 0xfedc0000 if DEBUG_EP93XX
+	default 0xfee003f8 if DEBUG_FOOTBRIDGE_COM1
+	default 0xfee20000 if DEBUG_NSPIRE_CLASSIC_UART || DEBUG_NSPIRE_CX_UART
+	default 0xfee82340 if ARCH_IOP13XX
+	default 0xfef00000 if ARCH_IXP4XX && !CPU_BIG_ENDIAN
+	default 0xfef00003 if ARCH_IXP4XX && CPU_BIG_ENDIAN
+	default 0xfef36000 if DEBUG_HIGHBANK_UART
+	default 0xfefb0000 if DEBUG_OMAP1UART1 || DEBUG_OMAP7XXUART1
+	default 0xfefb0800 if DEBUG_OMAP1UART2 || DEBUG_OMAP7XXUART2
+	default 0xfefb9800 if DEBUG_OMAP1UART3 || DEBUG_OMAP7XXUART3
+	default 0xfefff700 if ARCH_IOP33X
+	default 0xff003000 if DEBUG_U300_UART
+	default 0xffd01000 if DEBUG_HIP01_UART
+	default DEBUG_UART_PHYS if !MMU
+	depends on DEBUG_LL_UART_8250 || DEBUG_LL_UART_PL01X || \
+		DEBUG_UART_8250 || DEBUG_UART_PL01X || DEBUG_MESON_UARTAO || \
+		DEBUG_NETX_UART || \
+		DEBUG_QCOM_UARTDM || DEBUG_S3C24XX_UART || \
+		DEBUG_S3C64XX_UART || \
+		DEBUG_BCM63XX_UART || DEBUG_ASM9260_UART || \
+		DEBUG_SIRFSOC_UART || DEBUG_DIGICOLOR_UA0 || \
+		DEBUG_AT91_UART
+
+config DEBUG_UART_8250_SHIFT
+	int "Register offset shift for the 8250 debug UART"
+	depends on DEBUG_LL_UART_8250 || DEBUG_UART_8250
+	default 0 if DEBUG_FOOTBRIDGE_COM1 || ARCH_IOP32X || DEBUG_BCM_5301X || \
+		DEBUG_OMAP7XXUART1 || DEBUG_OMAP7XXUART2 || DEBUG_OMAP7XXUART3
+	default 2
+
+config DEBUG_UART_8250_WORD
+	bool "Use 32-bit accesses for 8250 UART"
+	depends on DEBUG_LL_UART_8250 || DEBUG_UART_8250
+	depends on DEBUG_UART_8250_SHIFT >= 2
+	default y if DEBUG_PICOXCELL_UART || \
+		DEBUG_SOCFPGA_UART0 || DEBUG_SOCFPGA_UART1 || \
+		DEBUG_KEYSTONE_UART0 || DEBUG_KEYSTONE_UART1 || \
+		DEBUG_ALPINE_UART0 || \
+		DEBUG_DAVINCI_DMx_UART0 || DEBUG_DAVINCI_DA8XX_UART1 || \
+		DEBUG_DAVINCI_DA8XX_UART2 || \
+		DEBUG_BCM_KONA_UART || DEBUG_RK32_UART2
+
+config DEBUG_UART_8250_PALMCHIP
+	bool "8250 UART is Palmchip BK-310x"
+	depends on DEBUG_LL_UART_8250 || DEBUG_UART_8250
+	help
+	  Palmchip provides a UART implementation compatible with 16550
+	  except for having a different register layout.  Say Y here if
+	  the debug UART is of this type.
+
+config DEBUG_UART_8250_FLOW_CONTROL
+	bool "Enable flow control for 8250 UART"
+	depends on DEBUG_LL_UART_8250 || DEBUG_UART_8250
+	default y if ARCH_EBSA110 || DEBUG_FOOTBRIDGE_COM1 || DEBUG_GEMINI || ARCH_RPC
+
+config DEBUG_UNCOMPRESS
+	bool
+	depends on ARCH_MULTIPLATFORM || PLAT_SAMSUNG || ARM_SINGLE_ARMV7M
+	default y if DEBUG_LL && !DEBUG_OMAP2PLUS_UART && \
+		     (!DEBUG_TEGRA_UART || !ZBOOT_ROM) && \
+		     !DEBUG_BRCMSTB_UART
+	help
+	  This option influences the normal decompressor output for
+	  multiplatform kernels.  Normally, multiplatform kernels disable
+	  decompressor output because it is not possible to know where to
+	  send the decompressor output.
+
+	  When this option is set, the selected DEBUG_LL output method
+	  will be re-used for normal decompressor output on multiplatform
+	  kernels.
+	  
+
+config UNCOMPRESS_INCLUDE
+	string
+	default "debug/uncompress.h" if ARCH_MULTIPLATFORM || ARCH_MSM || \
+					PLAT_SAMSUNG || ARM_SINGLE_ARMV7M
+	default "mach/uncompress.h"
+
+config EARLY_PRINTK
+	bool "Early printk"
+	depends on DEBUG_LL
+	help
+	  Say Y here if you want to have an early console using the
+	  kernel low-level debugging functions. Add earlyprintk to your
+	  kernel parameters to enable this console.
+
+config EARLY_PRINTK_DIRECT
+	bool "Early printk direct"
+	depends on DEBUG_LL
+	help
+	  Say Y here if you want to have an early console using the
+	  kernel low-level debugging functions and EARLY_PRINTK is
+	  not early enough.
+
+config ARM_KPROBES_TEST
+	tristate "Kprobes test module"
+	depends on KPROBES && MODULES
+	help
+	  Perform tests of kprobes API and instruction set simulation.
+
+config PID_IN_CONTEXTIDR
+	bool "Write the current PID to the CONTEXTIDR register"
+	depends on CPU_COPY_V6
+	help
+	  Enabling this option causes the kernel to write the current PID to
+	  the PROCID field of the CONTEXTIDR register, at the expense of some
+	  additional instructions during context switch. Say Y here only if you
+	  are planning to use hardware trace tools with this kernel.
+
+config DEBUG_SET_MODULE_RONX
+	bool "Set loadable kernel module data as NX and text as RO"
+	depends on MODULES && MMU
+	---help---
+	  This option helps catch unintended modifications to loadable
+	  kernel module's text and read-only data. It also prevents execution
+	  of module data. Such protection may interfere with run-time code
+	  patching and dynamic kernel tracing - and they might also protect
+	  against certain classes of kernel exploits.
+	  If in doubt, say "N".
+
+source "drivers/hwtracing/coresight/Kconfig"
+
+endmenu

arch/arm/Makefile

@@ -0,0 +1,388 @@
+#
+# arch/arm/Makefile
+#
+# This file is included by the global makefile so that you can add your own
+# architecture-specific flags and dependencies.
+#
+# This file is subject to the terms and conditions of the GNU General Public
+# License.  See the file "COPYING" in the main directory of this archive
+# for more details.
+#
+# Copyright (C) 1995-2001 by Russell King
+
+# Ensure linker flags are correct
+LDFLAGS		:=
+
+LDFLAGS_vmlinux	:=-p --no-undefined -X --pic-veneer
+ifeq ($(CONFIG_CPU_ENDIAN_BE8),y)
+LDFLAGS_vmlinux	+= --be8
+LDFLAGS_MODULE	+= --be8
+endif
+
+ifeq ($(CONFIG_ARM_MODULE_PLTS),y)
+LDFLAGS_MODULE	+= -T $(srctree)/arch/arm/kernel/module.lds
+endif
+
+GZFLAGS		:=-9
+#KBUILD_CFLAGS	+=-pipe
+
+# Never generate .eh_frame
+KBUILD_CFLAGS	+= $(call cc-option,-fno-dwarf2-cfi-asm)
+
+# This should work on most of the modern platforms
+KBUILD_DEFCONFIG := multi_v7_defconfig
+
+# defines filename extension depending memory management type.
+ifeq ($(CONFIG_MMU),)
+MMUEXT		:= -nommu
+KBUILD_CFLAGS	+= $(call cc-option,-mno-unaligned-access)
+endif
+
+ifeq ($(CONFIG_FRAME_POINTER),y)
+KBUILD_CFLAGS	+=-fno-omit-frame-pointer -mapcs -mno-sched-prolog
+endif
+
+ifeq ($(CONFIG_CPU_BIG_ENDIAN),y)
+KBUILD_CPPFLAGS	+= -mbig-endian
+AS		+= -EB
+LD		+= -EB
+else
+KBUILD_CPPFLAGS	+= -mlittle-endian
+AS		+= -EL
+LD		+= -EL
+endif
+
+#
+# The Scalar Replacement of Aggregates (SRA) optimization pass in GCC 4.9 and
+# later may result in code being generated that handles signed short and signed
+# char struct members incorrectly. So disable it.
+# (https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65932)
+#
+KBUILD_CFLAGS	+= $(call cc-option,-fno-ipa-sra)
+
+# This selects which instruction set is used.
+# Note that GCC does not numerically define an architecture version
+# macro, but instead defines a whole series of macros which makes
+# testing for a specific architecture or later rather impossible.
+arch-$(CONFIG_CPU_32v7M)	=-D__LINUX_ARM_ARCH__=7 -march=armv7-m -Wa,-march=armv7-m
+arch-$(CONFIG_CPU_32v7)		=-D__LINUX_ARM_ARCH__=7 $(call cc-option,-march=armv7-a,-march=armv5t -Wa$(comma)-march=armv7-a)
+arch-$(CONFIG_CPU_32v6)		=-D__LINUX_ARM_ARCH__=6 $(call cc-option,-march=armv6,-march=armv5t -Wa$(comma)-march=armv6)
+# Only override the compiler option if ARMv6. The ARMv6K extensions are
+# always available in ARMv7
+ifeq ($(CONFIG_CPU_32v6),y)
+arch-$(CONFIG_CPU_32v6K)	=-D__LINUX_ARM_ARCH__=6 $(call cc-option,-march=armv6k,-march=armv5t -Wa$(comma)-march=armv6k)
+endif
+arch-$(CONFIG_CPU_32v5)		=-D__LINUX_ARM_ARCH__=5 $(call cc-option,-march=armv5te,-march=armv4t)
+arch-$(CONFIG_CPU_32v4T)	=-D__LINUX_ARM_ARCH__=4 -march=armv4t
+arch-$(CONFIG_CPU_32v4)		=-D__LINUX_ARM_ARCH__=4 -march=armv4
+arch-$(CONFIG_CPU_32v3)		=-D__LINUX_ARM_ARCH__=3 -march=armv3
+
+# Evaluate arch cc-option calls now
+arch-y := $(arch-y)
+
+# This selects how we optimise for the processor.
+tune-$(CONFIG_CPU_ARM7TDMI)	=-mtune=arm7tdmi
+tune-$(CONFIG_CPU_ARM720T)	=-mtune=arm7tdmi
+tune-$(CONFIG_CPU_ARM740T)	=-mtune=arm7tdmi
+tune-$(CONFIG_CPU_ARM9TDMI)	=-mtune=arm9tdmi
+tune-$(CONFIG_CPU_ARM940T)	=-mtune=arm9tdmi
+tune-$(CONFIG_CPU_ARM946E)	=$(call cc-option,-mtune=arm9e,-mtune=arm9tdmi)
+tune-$(CONFIG_CPU_ARM920T)	=-mtune=arm9tdmi
+tune-$(CONFIG_CPU_ARM922T)	=-mtune=arm9tdmi
+tune-$(CONFIG_CPU_ARM925T)	=-mtune=arm9tdmi
+tune-$(CONFIG_CPU_ARM926T)	=-mtune=arm9tdmi
+tune-$(CONFIG_CPU_FA526)	=-mtune=arm9tdmi
+tune-$(CONFIG_CPU_SA110)	=-mtune=strongarm110
+tune-$(CONFIG_CPU_SA1100)	=-mtune=strongarm1100
+tune-$(CONFIG_CPU_XSCALE)	=$(call cc-option,-mtune=xscale,-mtune=strongarm110) -Wa,-mcpu=xscale
+tune-$(CONFIG_CPU_XSC3)		=$(call cc-option,-mtune=xscale,-mtune=strongarm110) -Wa,-mcpu=xscale
+tune-$(CONFIG_CPU_FEROCEON)	=$(call cc-option,-mtune=marvell-f,-mtune=xscale)
+tune-$(CONFIG_CPU_V6)		=$(call cc-option,-mtune=arm1136j-s,-mtune=strongarm)
+tune-$(CONFIG_CPU_V6K)		=$(call cc-option,-mtune=arm1136j-s,-mtune=strongarm)
+
+# Evaluate tune cc-option calls now
+tune-y := $(tune-y)
+
+ifeq ($(CONFIG_AEABI),y)
+CFLAGS_ABI	:=-mabi=aapcs-linux -mno-thumb-interwork -mfpu=vfp
+else
+CFLAGS_ABI	:=$(call cc-option,-mapcs-32,-mabi=apcs-gnu) $(call cc-option,-mno-thumb-interwork,)
+endif
+
+ifeq ($(CONFIG_ARM_UNWIND),y)
+CFLAGS_ABI	+=-funwind-tables
+endif
+
+ifeq ($(CONFIG_THUMB2_KERNEL),y)
+AFLAGS_AUTOIT	:=$(call as-option,-Wa$(comma)-mimplicit-it=always,-Wa$(comma)-mauto-it)
+AFLAGS_NOWARN	:=$(call as-option,-Wa$(comma)-mno-warn-deprecated,-Wa$(comma)-W)
+CFLAGS_ISA	:=-mthumb $(AFLAGS_AUTOIT) $(AFLAGS_NOWARN)
+AFLAGS_ISA	:=$(CFLAGS_ISA) -Wa$(comma)-mthumb
+# Work around buggy relocation from gas if requested:
+ifeq ($(CONFIG_THUMB2_AVOID_R_ARM_THM_JUMP11),y)
+CFLAGS_MODULE	+=-fno-optimize-sibling-calls
+endif
+else
+CFLAGS_ISA	:=$(call cc-option,-marm,)
+AFLAGS_ISA	:=$(CFLAGS_ISA)
+endif
+
+# Need -Uarm for gcc < 3.x
+KBUILD_CFLAGS	+=$(CFLAGS_ABI) $(CFLAGS_ISA) $(arch-y) $(tune-y) $(call cc-option,-mshort-load-bytes,$(call cc-option,-malignment-traps,)) -msoft-float -Uarm
+KBUILD_AFLAGS	+=$(CFLAGS_ABI) $(AFLAGS_ISA) $(arch-y) $(tune-y) -include asm/unified.h -msoft-float
+
+CHECKFLAGS	+= -D__arm__
+
+#Default value
+head-y		:= arch/arm/kernel/head$(MMUEXT).o
+
+# Text offset. This list is sorted numerically by address in order to
+# provide a means to avoid/resolve conflicts in multi-arch kernels.
+textofs-y	:= 0x00008000
+# We don't want the htc bootloader to corrupt kernel during resume
+textofs-$(CONFIG_PM_H1940)      := 0x00108000
+# SA1111 DMA bug: we don't want the kernel to live in precious DMA-able memory
+ifeq ($(CONFIG_ARCH_SA1100),y)
+textofs-$(CONFIG_SA1111) := 0x00208000
+endif
+textofs-$(CONFIG_ARCH_MSM8X60) := 0x00208000
+textofs-$(CONFIG_ARCH_MSM8960) := 0x00208000
+textofs-$(CONFIG_ARCH_AXXIA) := 0x00308000
+
+# Machine directory name.  This list is sorted alphanumerically
+# by CONFIG_* macro name.
+machine-$(CONFIG_ARCH_ALPINE)		+= alpine
+machine-$(CONFIG_ARCH_ARTPEC)		+= artpec
+machine-$(CONFIG_ARCH_AT91)		+= at91
+machine-$(CONFIG_ARCH_AXXIA)		+= axxia
+machine-$(CONFIG_ARCH_BCM)		+= bcm
+machine-$(CONFIG_ARCH_BERLIN)		+= berlin
+machine-$(CONFIG_ARCH_CLPS711X)		+= clps711x
+machine-$(CONFIG_ARCH_CNS3XXX)		+= cns3xxx
+machine-$(CONFIG_ARCH_DAVINCI)		+= davinci
+machine-$(CONFIG_ARCH_DIGICOLOR)	+= digicolor
+machine-$(CONFIG_ARCH_DOVE)		+= dove
+machine-$(CONFIG_ARCH_EBSA110)		+= ebsa110
+machine-$(CONFIG_ARCH_EFM32)		+= efm32
+machine-$(CONFIG_ARCH_EP93XX)		+= ep93xx
+machine-$(CONFIG_ARCH_EXYNOS)		+= exynos
+machine-$(CONFIG_ARCH_FOOTBRIDGE)	+= footbridge
+machine-$(CONFIG_ARCH_GEMINI)		+= gemini
+machine-$(CONFIG_ARCH_HIGHBANK)		+= highbank
+machine-$(CONFIG_ARCH_HISI)		+= hisi
+machine-$(CONFIG_ARCH_INTEGRATOR)	+= integrator
+machine-$(CONFIG_ARCH_IOP13XX)		+= iop13xx
+machine-$(CONFIG_ARCH_IOP32X)		+= iop32x
+machine-$(CONFIG_ARCH_IOP33X)		+= iop33x
+machine-$(CONFIG_ARCH_IXP4XX)		+= ixp4xx
+machine-$(CONFIG_ARCH_KEYSTONE)		+= keystone
+machine-$(CONFIG_ARCH_KS8695)		+= ks8695
+machine-$(CONFIG_ARCH_LPC18XX)		+= lpc18xx
+machine-$(CONFIG_ARCH_LPC32XX)		+= lpc32xx
+machine-$(CONFIG_ARCH_MESON)		+= meson
+machine-$(CONFIG_ARCH_MMP)		+= mmp
+machine-$(CONFIG_ARCH_MPS2)		+= vexpress
+machine-$(CONFIG_ARCH_MOXART)		+= moxart
+machine-$(CONFIG_ARCH_MV78XX0)		+= mv78xx0
+machine-$(CONFIG_ARCH_MVEBU)		+= mvebu
+machine-$(CONFIG_ARCH_MXC)		+= imx
+machine-$(CONFIG_ARCH_MEDIATEK)		+= mediatek
+machine-$(CONFIG_ARCH_MXS)		+= mxs
+machine-$(CONFIG_ARCH_NETX)		+= netx
+machine-$(CONFIG_ARCH_NOMADIK)		+= nomadik
+machine-$(CONFIG_ARCH_NSPIRE)		+= nspire
+machine-$(CONFIG_ARCH_OMAP1)		+= omap1
+machine-$(CONFIG_ARCH_OMAP2PLUS)	+= omap2
+machine-$(CONFIG_ARCH_ORION5X)		+= orion5x
+machine-$(CONFIG_ARCH_PICOXCELL)	+= picoxcell
+machine-$(CONFIG_ARCH_PXA)		+= pxa
+machine-$(CONFIG_ARCH_QCOM)		+= qcom
+machine-$(CONFIG_ARCH_REALVIEW)		+= realview
+machine-$(CONFIG_ARCH_ROCKCHIP)		+= rockchip
+machine-$(CONFIG_ARCH_RPC)		+= rpc
+machine-$(CONFIG_ARCH_S3C24XX)		+= s3c24xx
+machine-$(CONFIG_ARCH_S3C64XX)		+= s3c64xx
+machine-$(CONFIG_ARCH_S5PV210)		+= s5pv210
+machine-$(CONFIG_ARCH_SA1100)		+= sa1100
+machine-$(CONFIG_ARCH_SHMOBILE) 	+= shmobile
+machine-$(CONFIG_ARCH_SIRF)		+= prima2
+machine-$(CONFIG_ARCH_SOCFPGA)		+= socfpga
+machine-$(CONFIG_ARCH_STI)		+= sti
+machine-$(CONFIG_ARCH_STM32)		+= stm32
+machine-$(CONFIG_ARCH_SUNXI)		+= sunxi
+machine-$(CONFIG_ARCH_TANGO)		+= tango
+machine-$(CONFIG_ARCH_TEGRA)		+= tegra
+machine-$(CONFIG_ARCH_U300)		+= u300
+machine-$(CONFIG_ARCH_U8500)		+= ux500
+machine-$(CONFIG_ARCH_UNIPHIER)		+= uniphier
+machine-$(CONFIG_ARCH_VERSATILE)	+= versatile
+machine-$(CONFIG_ARCH_VEXPRESS)		+= vexpress
+machine-$(CONFIG_ARCH_VT8500)		+= vt8500
+machine-$(CONFIG_ARCH_W90X900)		+= w90x900
+machine-$(CONFIG_ARCH_ZX)		+= zx
+machine-$(CONFIG_ARCH_ZYNQ)		+= zynq
+machine-$(CONFIG_PLAT_SPEAR)		+= spear
+
+# Platform directory name.  This list is sorted alphanumerically
+# by CONFIG_* macro name.
+plat-$(CONFIG_ARCH_EXYNOS)	+= samsung
+plat-$(CONFIG_ARCH_OMAP)	+= omap
+plat-$(CONFIG_ARCH_S3C64XX)	+= samsung
+plat-$(CONFIG_ARCH_S5PV210)	+= samsung
+plat-$(CONFIG_PLAT_IOP)		+= iop
+plat-$(CONFIG_PLAT_ORION)	+= orion
+plat-$(CONFIG_PLAT_PXA)		+= pxa
+plat-$(CONFIG_PLAT_S3C24XX)	+= samsung
+plat-$(CONFIG_PLAT_VERSATILE)	+= versatile
+
+ifeq ($(CONFIG_ARCH_EBSA110),y)
+# This is what happens if you forget the IOCS16 line.
+# PCMCIA cards stop working.
+CFLAGS_3c589_cs.o :=-DISA_SIXTEEN_BIT_PERIPHERAL
+export CFLAGS_3c589_cs.o
+endif
+
+# The byte offset of the kernel image in RAM from the start of RAM.
+TEXT_OFFSET := $(textofs-y)
+
+# The first directory contains additional information for the boot setup code
+ifneq ($(machine-y),)
+MACHINE  := arch/arm/mach-$(word 1,$(machine-y))/
+else
+MACHINE  :=
+endif
+ifeq ($(CONFIG_ARCH_MULTIPLATFORM),y)
+MACHINE  :=
+endif
+
+machdirs := $(patsubst %,arch/arm/mach-%/,$(machine-y))
+platdirs := $(patsubst %,arch/arm/plat-%/,$(sort $(plat-y)))
+
+ifneq ($(CONFIG_ARCH_MULTIPLATFORM),y)
+ifneq ($(CONFIG_ARM_SINGLE_ARMV7M),y)
+ifeq ($(KBUILD_SRC),)
+KBUILD_CPPFLAGS += $(patsubst %,-I%include,$(machdirs) $(platdirs))
+else
+KBUILD_CPPFLAGS += $(patsubst %,-I$(srctree)/%include,$(machdirs) $(platdirs))
+endif
+endif
+endif
+
+export	TEXT_OFFSET GZFLAGS MMUEXT
+
+# Do we have FASTFPE?
+FASTFPE		:=arch/arm/fastfpe
+ifeq ($(FASTFPE),$(wildcard $(FASTFPE)))
+FASTFPE_OBJ	:=$(FASTFPE)/
+endif
+
+subdir-ccflags-y += -Werror
+
+core-$(CONFIG_FPE_NWFPE)	+= arch/arm/nwfpe/
+core-$(CONFIG_FPE_FASTFPE)	+= $(FASTFPE_OBJ)
+core-$(CONFIG_VFP)		+= arch/arm/vfp/
+core-$(CONFIG_XEN)		+= arch/arm/xen/
+core-$(CONFIG_KVM_ARM_HOST) 	+= arch/arm/kvm/
+core-$(CONFIG_VDSO)		+= arch/arm/vdso/
+
+# If we have a machine-specific directory, then include it in the build.
+core-y				+= arch/arm/kernel/ arch/arm/mm/ arch/arm/common/
+core-y				+= arch/arm/probes/
+core-y				+= arch/arm/net/
+core-y				+= arch/arm/crypto/
+core-y				+= arch/arm/firmware/
+core-y				+= $(machdirs) $(platdirs)
+
+drivers-$(CONFIG_OPROFILE)      += arch/arm/oprofile/
+
+libs-y				:= arch/arm/lib/ $(libs-y)
+
+# Default target when executing plain make
+ifeq ($(CONFIG_XIP_KERNEL),y)
+KBUILD_IMAGE := xipImage
+else ifeq ($(CONFIG_BUILD_ARM_APPENDED_DTB_IMAGE),y)
+KBUILD_IMAGE := zImage-dtb
+else
+KBUILD_IMAGE := zImage
+endif
+
+# Build the DT binary blobs if we have OF configured
+ifeq ($(CONFIG_USE_OF),y)
+KBUILD_DTBS := dtbs
+endif
+
+all:	$(KBUILD_IMAGE) $(KBUILD_DTBS)
+
+MTK_PLATFORM := $(CONFIG_MTK_PLATFORM:"%"=%)
+MTK_PROJECT := $(CONFIG_ARCH_MTK_PROJECT:"%"=%)
+export MTK_PLATFORM MTK_PROJECT MTK_TARGET_PROJECT
+-include $(srctree)/scripts/drvgen/drvgen.mk
+
+boot := arch/arm/boot
+
+archprepare:
+	$(Q)$(MAKE) $(build)=arch/arm/tools include/generated/mach-types.h
+
+# Convert bzImage to zImage
+bzImage: zImage
+
+BOOT_TARGETS	= zImage Image xipImage bootpImage uImage
+INSTALL_TARGETS	= zinstall uinstall install
+
+PHONY += bzImage $(BOOT_TARGETS) $(INSTALL_TARGETS)
+
+bootpImage uImage: zImage
+zImage: Image
+
+$(BOOT_TARGETS): vmlinux
+	$(Q)$(MAKE) $(build)=$(boot) MACHINE=$(MACHINE) $(boot)/$@
+	@$(kecho) '  Kernel: $(boot)/$@ is ready'
+
+$(INSTALL_TARGETS):
+	$(Q)$(MAKE) $(build)=$(boot) MACHINE=$(MACHINE) $@
+
+%.dtb: | scripts $(DRVGEN_FILE_LIST)
+	$(Q)$(MAKE) $(build)=$(boot)/dts MACHINE=$(MACHINE) $(boot)/dts/$@
+
+PHONY += dtbs dtbs_install
+
+dtbs: prepare scripts dtbo_check
+	$(Q)$(MAKE) $(build)=$(boot)/dts
+
+dtbs_install:
+	$(Q)$(MAKE) $(dtbinst)=$(boot)/dts
+
+PHONY += vdso_install
+vdso_install:
+ifeq ($(CONFIG_VDSO),y)
+	$(Q)$(MAKE) $(build)=arch/arm/vdso $@
+endif
+
+zImage-dtb: vmlinux scripts dtbs
+	$(Q)$(MAKE) $(build)=$(boot) MACHINE=$(MACHINE) $(boot)/$@
+
+# We use MRPROPER_FILES and CLEAN_FILES now
+archclean:
+	$(Q)$(MAKE) $(clean)=$(boot)
+
+# My testing targets (bypasses dependencies)
+bp:;	$(Q)$(MAKE) $(build)=$(boot) MACHINE=$(MACHINE) $(boot)/bootpImage
+
+
+define archhelp
+  echo  '* zImage        - Compressed kernel image (arch/$(ARCH)/boot/zImage)'
+  echo  '  Image         - Uncompressed kernel image (arch/$(ARCH)/boot/Image)'
+  echo  '* xipImage      - XIP kernel image, if configured (arch/$(ARCH)/boot/xipImage)'
+  echo  '  uImage        - U-Boot wrapped zImage'
+  echo  '  bootpImage    - Combined zImage and initial RAM disk'
+  echo  '                  (supply initrd image via make variable INITRD=<path>)'
+  echo  '* dtbs          - Build device tree blobs for enabled boards'
+  echo  '  dtbs_install  - Install dtbs to $(INSTALL_DTBS_PATH)'
+  echo  '  install       - Install uncompressed kernel'
+  echo  '  zinstall      - Install compressed kernel'
+  echo  '  uinstall      - Install U-Boot wrapped compressed kernel'
+  echo  '                  Install using (your) ~/bin/$(INSTALLKERNEL) or'
+  echo  '                  (distribution) /sbin/$(INSTALLKERNEL) or'
+  echo  '                  install to $$(INSTALL_PATH) and run lilo'
+  echo  '  vdso_install  - Install unstripped vdso.so to $$(INSTALL_MOD_PATH)/vdso'
+endef

arch/arm/plat-iop/Makefile

@@ -0,0 +1,27 @@
+#
+# Makefile for the linux kernel.
+#
+
+# IOP32X
+obj-$(CONFIG_ARCH_IOP32X) += i2c.o
+obj-$(CONFIG_ARCH_IOP32X) += pci.o
+obj-$(CONFIG_ARCH_IOP32X) += setup.o
+obj-$(CONFIG_ARCH_IOP32X) += time.o
+obj-$(CONFIG_ARCH_IOP32X) += cp6.o
+obj-$(CONFIG_ARCH_IOP32X) += adma.o
+obj-$(CONFIG_ARCH_IOP32X) += pmu.o
+obj-$(CONFIG_ARCH_IOP32X) += restart.o
+
+# IOP33X
+obj-$(CONFIG_ARCH_IOP33X) += i2c.o
+obj-$(CONFIG_ARCH_IOP33X) += pci.o
+obj-$(CONFIG_ARCH_IOP33X) += setup.o
+obj-$(CONFIG_ARCH_IOP33X) += time.o
+obj-$(CONFIG_ARCH_IOP33X) += cp6.o
+obj-$(CONFIG_ARCH_IOP33X) += adma.o
+obj-$(CONFIG_ARCH_IOP33X) += pmu.o
+obj-$(CONFIG_ARCH_IOP33X) += restart.o
+
+# IOP13XX
+obj-$(CONFIG_ARCH_IOP13XX) += cp6.o
+obj-$(CONFIG_ARCH_IOP13XX) += time.o

arch/arm/plat-iop/adma.c

@@ -0,0 +1,205 @@
+/*
+ * platform device definitions for the iop3xx dma/xor engines
+ * Copyright © 2006, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ */
+#include <linux/platform_device.h>
+#include <asm/hardware/iop3xx.h>
+#include <linux/dma-mapping.h>
+#include <mach/adma.h>
+#include <asm/hardware/iop_adma.h>
+
+#ifdef CONFIG_ARCH_IOP32X
+#define IRQ_DMA0_EOT IRQ_IOP32X_DMA0_EOT
+#define IRQ_DMA0_EOC IRQ_IOP32X_DMA0_EOC
+#define IRQ_DMA0_ERR IRQ_IOP32X_DMA0_ERR
+
+#define IRQ_DMA1_EOT IRQ_IOP32X_DMA1_EOT
+#define IRQ_DMA1_EOC IRQ_IOP32X_DMA1_EOC
+#define IRQ_DMA1_ERR IRQ_IOP32X_DMA1_ERR
+
+#define IRQ_AA_EOT IRQ_IOP32X_AA_EOT
+#define IRQ_AA_EOC IRQ_IOP32X_AA_EOC
+#define IRQ_AA_ERR IRQ_IOP32X_AA_ERR
+#endif
+#ifdef CONFIG_ARCH_IOP33X
+#define IRQ_DMA0_EOT IRQ_IOP33X_DMA0_EOT
+#define IRQ_DMA0_EOC IRQ_IOP33X_DMA0_EOC
+#define IRQ_DMA0_ERR IRQ_IOP33X_DMA0_ERR
+
+#define IRQ_DMA1_EOT IRQ_IOP33X_DMA1_EOT
+#define IRQ_DMA1_EOC IRQ_IOP33X_DMA1_EOC
+#define IRQ_DMA1_ERR IRQ_IOP33X_DMA1_ERR
+
+#define IRQ_AA_EOT IRQ_IOP33X_AA_EOT
+#define IRQ_AA_EOC IRQ_IOP33X_AA_EOC
+#define IRQ_AA_ERR IRQ_IOP33X_AA_ERR
+#endif
+/* AAU and DMA Channels */
+static struct resource iop3xx_dma_0_resources[] = {
+	[0] = {
+		.start = IOP3XX_DMA_PHYS_BASE(0),
+		.end = IOP3XX_DMA_UPPER_PA(0),
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_DMA0_EOT,
+		.end = IRQ_DMA0_EOT,
+		.flags = IORESOURCE_IRQ
+	},
+	[2] = {
+		.start = IRQ_DMA0_EOC,
+		.end = IRQ_DMA0_EOC,
+		.flags = IORESOURCE_IRQ
+	},
+	[3] = {
+		.start = IRQ_DMA0_ERR,
+		.end = IRQ_DMA0_ERR,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+static struct resource iop3xx_dma_1_resources[] = {
+	[0] = {
+		.start = IOP3XX_DMA_PHYS_BASE(1),
+		.end = IOP3XX_DMA_UPPER_PA(1),
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_DMA1_EOT,
+		.end = IRQ_DMA1_EOT,
+		.flags = IORESOURCE_IRQ
+	},
+	[2] = {
+		.start = IRQ_DMA1_EOC,
+		.end = IRQ_DMA1_EOC,
+		.flags = IORESOURCE_IRQ
+	},
+	[3] = {
+		.start = IRQ_DMA1_ERR,
+		.end = IRQ_DMA1_ERR,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+
+static struct resource iop3xx_aau_resources[] = {
+	[0] = {
+		.start = IOP3XX_AAU_PHYS_BASE,
+		.end = IOP3XX_AAU_UPPER_PA,
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_AA_EOT,
+		.end = IRQ_AA_EOT,
+		.flags = IORESOURCE_IRQ
+	},
+	[2] = {
+		.start = IRQ_AA_EOC,
+		.end = IRQ_AA_EOC,
+		.flags = IORESOURCE_IRQ
+	},
+	[3] = {
+		.start = IRQ_AA_ERR,
+		.end = IRQ_AA_ERR,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+static u64 iop3xx_adma_dmamask = DMA_BIT_MASK(32);
+
+static struct iop_adma_platform_data iop3xx_dma_0_data = {
+	.hw_id = DMA0_ID,
+	.pool_size = PAGE_SIZE,
+};
+
+static struct iop_adma_platform_data iop3xx_dma_1_data = {
+	.hw_id = DMA1_ID,
+	.pool_size = PAGE_SIZE,
+};
+
+static struct iop_adma_platform_data iop3xx_aau_data = {
+	.hw_id = AAU_ID,
+	.pool_size = 3 * PAGE_SIZE,
+};
+
+struct platform_device iop3xx_dma_0_channel = {
+	.name = "iop-adma",
+	.id = 0,
+	.num_resources = 4,
+	.resource = iop3xx_dma_0_resources,
+	.dev = {
+		.dma_mask = &iop3xx_adma_dmamask,
+		.coherent_dma_mask = DMA_BIT_MASK(64),
+		.platform_data = (void *) &iop3xx_dma_0_data,
+	},
+};
+
+struct platform_device iop3xx_dma_1_channel = {
+	.name = "iop-adma",
+	.id = 1,
+	.num_resources = 4,
+	.resource = iop3xx_dma_1_resources,
+	.dev = {
+		.dma_mask = &iop3xx_adma_dmamask,
+		.coherent_dma_mask = DMA_BIT_MASK(64),
+		.platform_data = (void *) &iop3xx_dma_1_data,
+	},
+};
+
+struct platform_device iop3xx_aau_channel = {
+	.name = "iop-adma",
+	.id = 2,
+	.num_resources = 4,
+	.resource = iop3xx_aau_resources,
+	.dev = {
+		.dma_mask = &iop3xx_adma_dmamask,
+		.coherent_dma_mask = DMA_BIT_MASK(64),
+		.platform_data = (void *) &iop3xx_aau_data,
+	},
+};
+
+static int __init iop3xx_adma_cap_init(void)
+{
+	#ifdef CONFIG_ARCH_IOP32X /* the 32x DMA does not perform CRC32C */
+	dma_cap_set(DMA_MEMCPY, iop3xx_dma_0_data.cap_mask);
+	dma_cap_set(DMA_INTERRUPT, iop3xx_dma_0_data.cap_mask);
+	#else
+	dma_cap_set(DMA_MEMCPY, iop3xx_dma_0_data.cap_mask);
+	dma_cap_set(DMA_INTERRUPT, iop3xx_dma_0_data.cap_mask);
+	#endif
+
+	#ifdef CONFIG_ARCH_IOP32X /* the 32x DMA does not perform CRC32C */
+	dma_cap_set(DMA_MEMCPY, iop3xx_dma_1_data.cap_mask);
+	dma_cap_set(DMA_INTERRUPT, iop3xx_dma_1_data.cap_mask);
+	#else
+	dma_cap_set(DMA_MEMCPY, iop3xx_dma_1_data.cap_mask);
+	dma_cap_set(DMA_INTERRUPT, iop3xx_dma_1_data.cap_mask);
+	#endif
+
+	#ifdef CONFIG_ARCH_IOP32X /* the 32x AAU does not perform zero sum */
+	dma_cap_set(DMA_XOR, iop3xx_aau_data.cap_mask);
+	dma_cap_set(DMA_INTERRUPT, iop3xx_aau_data.cap_mask);
+	#else
+	dma_cap_set(DMA_XOR, iop3xx_aau_data.cap_mask);
+	dma_cap_set(DMA_XOR_VAL, iop3xx_aau_data.cap_mask);
+	dma_cap_set(DMA_INTERRUPT, iop3xx_aau_data.cap_mask);
+	#endif
+
+	return 0;
+}
+
+arch_initcall(iop3xx_adma_cap_init);

arch/arm/plat-iop/cp6.c

@@ -0,0 +1,51 @@
+/*
+ * IOP Coprocessor-6 access handler
+ * Copyright (c) 2006, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
+ * Place - Suite 330, Boston, MA 02111-1307 USA.
+ *
+ */
+#include <linux/init.h>
+#include <asm/traps.h>
+#include <asm/ptrace.h>
+
+static int cp6_trap(struct pt_regs *regs, unsigned int instr)
+{
+	u32 temp;
+
+        /* enable cp6 access */
+        asm volatile (
+		"mrc	p15, 0, %0, c15, c1, 0\n\t"
+		"orr	%0, %0, #(1 << 6)\n\t"
+		"mcr	p15, 0, %0, c15, c1, 0\n\t"
+		: "=r"(temp));
+
+	return 0;
+}
+
+/* permit kernel space cp6 access
+ * deny user space cp6 access
+ */
+static struct undef_hook cp6_hook = {
+	.instr_mask     = 0x0f000ff0,
+	.instr_val      = 0x0e000610,
+	.cpsr_mask      = MODE_MASK,
+	.cpsr_val       = SVC_MODE,
+	.fn             = cp6_trap,
+};
+
+void __init iop_init_cp6_handler(void)
+{
+	register_undef_hook(&cp6_hook);
+}

arch/arm/plat-iop/i2c.c

@@ -0,0 +1,79 @@
+/*
+ * arch/arm/plat-iop/i2c.c
+ *
+ * Author: Nicolas Pitre <nico@cam.org>
+ * Copyright (C) 2001 MontaVista Software, Inc.
+ * Copyright (C) 2004 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/major.h>
+#include <linux/fs.h>
+#include <linux/platform_device.h>
+#include <linux/serial.h>
+#include <linux/tty.h>
+#include <linux/serial_core.h>
+#include <linux/io.h>
+#include <asm/pgtable.h>
+#include <asm/page.h>
+#include <asm/mach/map.h>
+#include <asm/setup.h>
+#include <asm/memory.h>
+#include <mach/hardware.h>
+#include <asm/hardware/iop3xx.h>
+#include <asm/mach/arch.h>
+
+#ifdef CONFIG_ARCH_IOP32X
+#define IRQ_IOP3XX_I2C_0	IRQ_IOP32X_I2C_0
+#define IRQ_IOP3XX_I2C_1	IRQ_IOP32X_I2C_1
+#endif
+#ifdef CONFIG_ARCH_IOP33X
+#define IRQ_IOP3XX_I2C_0	IRQ_IOP33X_I2C_0
+#define IRQ_IOP3XX_I2C_1	IRQ_IOP33X_I2C_1
+#endif
+
+static struct resource iop3xx_i2c0_resources[] = {
+	[0] = {
+		.start	= 0xfffff680,
+		.end	= 0xfffff697,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= IRQ_IOP3XX_I2C_0,
+		.end	= IRQ_IOP3XX_I2C_0,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+struct platform_device iop3xx_i2c0_device = {
+	.name		= "IOP3xx-I2C",
+	.id		= 0,
+	.num_resources	= 2,
+	.resource	= iop3xx_i2c0_resources,
+};
+
+
+static struct resource iop3xx_i2c1_resources[] = {
+	[0] = {
+		.start	= 0xfffff6a0,
+		.end	= 0xfffff6b7,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= IRQ_IOP3XX_I2C_1,
+		.end	= IRQ_IOP3XX_I2C_1,
+		.flags	= IORESOURCE_IRQ,
+	}
+};
+
+struct platform_device iop3xx_i2c1_device = {
+	.name		= "IOP3xx-I2C",
+	.id		= 1,
+	.num_resources	= 2,
+	.resource	= iop3xx_i2c1_resources,
+};

arch/arm/plat-iop/pci.c

@@ -0,0 +1,404 @@
+/*
+ * arch/arm/plat-iop/pci.c
+ *
+ * PCI support for the Intel IOP32X and IOP33X processors
+ *
+ * Author: Rory Bolt <rorybolt@pacbell.net>
+ * Copyright (C) 2002 Rory Bolt
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/ioport.h>
+#include <linux/io.h>
+#include <asm/irq.h>
+#include <asm/signal.h>
+#include <mach/hardware.h>
+#include <asm/mach/pci.h>
+#include <asm/hardware/iop3xx.h>
+
+// #define DEBUG
+
+#ifdef DEBUG
+#define  DBG(x...) printk(x)
+#else
+#define  DBG(x...) do { } while (0)
+#endif
+
+/*
+ * This routine builds either a type0 or type1 configuration command.  If the
+ * bus is on the 803xx then a type0 made, else a type1 is created.
+ */
+static u32 iop3xx_cfg_address(struct pci_bus *bus, int devfn, int where)
+{
+	struct pci_sys_data *sys = bus->sysdata;
+	u32 addr;
+
+	if (sys->busnr == bus->number)
+		addr = 1 << (PCI_SLOT(devfn) + 16) | (PCI_SLOT(devfn) << 11);
+	else
+		addr = bus->number << 16 | PCI_SLOT(devfn) << 11 | 1;
+
+	addr |=	PCI_FUNC(devfn) << 8 | (where & ~3);
+
+	return addr;
+}
+
+/*
+ * This routine checks the status of the last configuration cycle.  If an error
+ * was detected it returns a 1, else it returns a 0.  The errors being checked
+ * are parity, master abort, target abort (master and target).  These types of
+ * errors occur during a config cycle where there is no device, like during
+ * the discovery stage.
+ */
+static int iop3xx_pci_status(void)
+{
+	unsigned int status;
+	int ret = 0;
+
+	/*
+	 * Check the status registers.
+	 */
+	status = *IOP3XX_ATUSR;
+	if (status & 0xf900) {
+		DBG("\t\t\tPCI: P0 - status = 0x%08x\n", status);
+		*IOP3XX_ATUSR = status & 0xf900;
+		ret = 1;
+	}
+
+	status = *IOP3XX_ATUISR;
+	if (status & 0x679f) {
+		DBG("\t\t\tPCI: P1 - status = 0x%08x\n", status);
+		*IOP3XX_ATUISR = status & 0x679f;
+		ret = 1;
+	}
+
+	return ret;
+}
+
+/*
+ * Simply write the address register and read the configuration
+ * data.  Note that the 4 nops ensure that we are able to handle
+ * a delayed abort (in theory.)
+ */
+static u32 iop3xx_read(unsigned long addr)
+{
+	u32 val;
+
+	__asm__ __volatile__(
+		"str	%1, [%2]\n\t"
+		"ldr	%0, [%3]\n\t"
+		"nop\n\t"
+		"nop\n\t"
+		"nop\n\t"
+		"nop\n\t"
+		: "=r" (val)
+		: "r" (addr), "r" (IOP3XX_OCCAR), "r" (IOP3XX_OCCDR));
+
+	return val;
+}
+
+/*
+ * The read routines must check the error status of the last configuration
+ * cycle.  If there was an error, the routine returns all hex f's.
+ */
+static int
+iop3xx_read_config(struct pci_bus *bus, unsigned int devfn, int where,
+		int size, u32 *value)
+{
+	unsigned long addr = iop3xx_cfg_address(bus, devfn, where);
+	u32 val = iop3xx_read(addr) >> ((where & 3) * 8);
+
+	if (iop3xx_pci_status())
+		val = 0xffffffff;
+
+	*value = val;
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int
+iop3xx_write_config(struct pci_bus *bus, unsigned int devfn, int where,
+		int size, u32 value)
+{
+	unsigned long addr = iop3xx_cfg_address(bus, devfn, where);
+	u32 val;
+
+	if (size != 4) {
+		val = iop3xx_read(addr);
+		if (iop3xx_pci_status())
+			return PCIBIOS_SUCCESSFUL;
+
+		where = (where & 3) * 8;
+
+		if (size == 1)
+			val &= ~(0xff << where);
+		else
+			val &= ~(0xffff << where);
+
+		*IOP3XX_OCCDR = val | value << where;
+	} else {
+		asm volatile(
+			"str	%1, [%2]\n\t"
+			"str	%0, [%3]\n\t"
+			"nop\n\t"
+			"nop\n\t"
+			"nop\n\t"
+			"nop\n\t"
+			:
+			: "r" (value), "r" (addr),
+			  "r" (IOP3XX_OCCAR), "r" (IOP3XX_OCCDR));
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+struct pci_ops iop3xx_ops = {
+	.read	= iop3xx_read_config,
+	.write	= iop3xx_write_config,
+};
+
+/*
+ * When a PCI device does not exist during config cycles, the 80200 gets a
+ * bus error instead of returning 0xffffffff. This handler simply returns.
+ */
+static int
+iop3xx_pci_abort(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
+{
+	DBG("PCI abort: address = 0x%08lx fsr = 0x%03x PC = 0x%08lx LR = 0x%08lx\n",
+		addr, fsr, regs->ARM_pc, regs->ARM_lr);
+
+	/*
+	 * If it was an imprecise abort, then we need to correct the
+	 * return address to be _after_ the instruction.
+	 */
+	if (fsr & (1 << 10))
+		regs->ARM_pc += 4;
+
+	return 0;
+}
+
+int iop3xx_pci_setup(int nr, struct pci_sys_data *sys)
+{
+	struct resource *res;
+
+	if (nr != 0)
+		return 0;
+
+	res = kzalloc(sizeof(struct resource), GFP_KERNEL);
+	if (!res)
+		panic("PCI: unable to alloc resources");
+
+	res->start = IOP3XX_PCI_LOWER_MEM_PA;
+	res->end   = IOP3XX_PCI_LOWER_MEM_PA + IOP3XX_PCI_MEM_WINDOW_SIZE - 1;
+	res->name  = "IOP3XX PCI Memory Space";
+	res->flags = IORESOURCE_MEM;
+	request_resource(&iomem_resource, res);
+
+	/*
+	 * Use whatever translation is already setup.
+	 */
+	sys->mem_offset = IOP3XX_PCI_LOWER_MEM_PA - *IOP3XX_OMWTVR0;
+
+	pci_add_resource_offset(&sys->resources, res, sys->mem_offset);
+
+	pci_ioremap_io(0, IOP3XX_PCI_LOWER_IO_PA);
+
+	return 1;
+}
+
+void __init iop3xx_atu_setup(void)
+{
+	/* BAR 0 ( Disabled ) */
+	*IOP3XX_IAUBAR0 = 0x0;
+	*IOP3XX_IABAR0  = 0x0;
+	*IOP3XX_IATVR0  = 0x0;
+	*IOP3XX_IALR0   = 0x0;
+
+	/* BAR 1 ( Disabled ) */
+	*IOP3XX_IAUBAR1 = 0x0;
+	*IOP3XX_IABAR1  = 0x0;
+	*IOP3XX_IALR1   = 0x0;
+
+	/* BAR 2 (1:1 mapping with Physical RAM) */
+	/* Set limit and enable */
+	*IOP3XX_IALR2 = ~((u32)IOP3XX_MAX_RAM_SIZE - 1) & ~0x1;
+	*IOP3XX_IAUBAR2 = 0x0;
+
+	/* Align the inbound bar with the base of memory */
+	*IOP3XX_IABAR2 = PHYS_OFFSET |
+			       PCI_BASE_ADDRESS_MEM_TYPE_64 |
+			       PCI_BASE_ADDRESS_MEM_PREFETCH;
+
+	*IOP3XX_IATVR2 = PHYS_OFFSET;
+
+	/* Outbound window 0 */
+	*IOP3XX_OMWTVR0 = IOP3XX_PCI_LOWER_MEM_BA;
+	*IOP3XX_OUMWTVR0 = 0;
+
+	/* Outbound window 1 */
+	*IOP3XX_OMWTVR1 = IOP3XX_PCI_LOWER_MEM_BA +
+			  IOP3XX_PCI_MEM_WINDOW_SIZE / 2;
+	*IOP3XX_OUMWTVR1 = 0;
+
+	/* BAR 3 ( Disabled ) */
+	*IOP3XX_IAUBAR3 = 0x0;
+	*IOP3XX_IABAR3  = 0x0;
+	*IOP3XX_IATVR3  = 0x0;
+	*IOP3XX_IALR3   = 0x0;
+
+	/* Setup the I/O Bar
+	 */
+	*IOP3XX_OIOWTVR = IOP3XX_PCI_LOWER_IO_BA;
+
+	/* Enable inbound and outbound cycles
+	 */
+	*IOP3XX_ATUCMD |= PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER |
+			       PCI_COMMAND_PARITY | PCI_COMMAND_SERR;
+	*IOP3XX_ATUCR |= IOP3XX_ATUCR_OUT_EN;
+}
+
+void __init iop3xx_atu_disable(void)
+{
+	*IOP3XX_ATUCMD = 0;
+	*IOP3XX_ATUCR = 0;
+
+	/* wait for cycles to quiesce */
+	while (*IOP3XX_PCSR & (IOP3XX_PCSR_OUT_Q_BUSY |
+				     IOP3XX_PCSR_IN_Q_BUSY))
+		cpu_relax();
+
+	/* BAR 0 ( Disabled ) */
+	*IOP3XX_IAUBAR0 = 0x0;
+	*IOP3XX_IABAR0  = 0x0;
+	*IOP3XX_IATVR0  = 0x0;
+	*IOP3XX_IALR0   = 0x0;
+
+	/* BAR 1 ( Disabled ) */
+	*IOP3XX_IAUBAR1 = 0x0;
+	*IOP3XX_IABAR1  = 0x0;
+	*IOP3XX_IALR1   = 0x0;
+
+	/* BAR 2 ( Disabled ) */
+	*IOP3XX_IAUBAR2 = 0x0;
+	*IOP3XX_IABAR2  = 0x0;
+	*IOP3XX_IATVR2  = 0x0;
+	*IOP3XX_IALR2   = 0x0;
+
+	/* BAR 3 ( Disabled ) */
+	*IOP3XX_IAUBAR3 = 0x0;
+	*IOP3XX_IABAR3  = 0x0;
+	*IOP3XX_IATVR3  = 0x0;
+	*IOP3XX_IALR3   = 0x0;
+
+	/* Clear the outbound windows */
+	*IOP3XX_OIOWTVR  = 0;
+
+	/* Outbound window 0 */
+	*IOP3XX_OMWTVR0 = 0;
+	*IOP3XX_OUMWTVR0 = 0;
+
+	/* Outbound window 1 */
+	*IOP3XX_OMWTVR1 = 0;
+	*IOP3XX_OUMWTVR1 = 0;
+}
+
+/* Flag to determine whether the ATU is initialized and the PCI bus scanned */
+int init_atu;
+
+int iop3xx_get_init_atu(void) {
+	/* check if default has been overridden */
+	if (init_atu != IOP3XX_INIT_ATU_DEFAULT)
+		return init_atu;
+	else
+		return IOP3XX_INIT_ATU_DISABLE;
+}
+
+static void __init iop3xx_atu_debug(void)
+{
+	DBG("PCI: Intel IOP3xx PCI init.\n");
+	DBG("PCI: Outbound memory window 0: PCI 0x%08x%08x\n",
+		*IOP3XX_OUMWTVR0, *IOP3XX_OMWTVR0);
+	DBG("PCI: Outbound memory window 1: PCI 0x%08x%08x\n",
+		*IOP3XX_OUMWTVR1, *IOP3XX_OMWTVR1);
+	DBG("PCI: Outbound IO window: PCI 0x%08x\n",
+		*IOP3XX_OIOWTVR);
+
+	DBG("PCI: Inbound memory window 0: PCI 0x%08x%08x 0x%08x -> 0x%08x\n",
+		*IOP3XX_IAUBAR0, *IOP3XX_IABAR0, *IOP3XX_IALR0, *IOP3XX_IATVR0);
+	DBG("PCI: Inbound memory window 1: PCI 0x%08x%08x 0x%08x\n",
+		*IOP3XX_IAUBAR1, *IOP3XX_IABAR1, *IOP3XX_IALR1);
+	DBG("PCI: Inbound memory window 2: PCI 0x%08x%08x 0x%08x -> 0x%08x\n",
+		*IOP3XX_IAUBAR2, *IOP3XX_IABAR2, *IOP3XX_IALR2, *IOP3XX_IATVR2);
+	DBG("PCI: Inbound memory window 3: PCI 0x%08x%08x 0x%08x -> 0x%08x\n",
+		*IOP3XX_IAUBAR3, *IOP3XX_IABAR3, *IOP3XX_IALR3, *IOP3XX_IATVR3);
+
+	DBG("PCI: Expansion ROM window: PCI 0x%08x%08x 0x%08x -> 0x%08x\n",
+		0, *IOP3XX_ERBAR, *IOP3XX_ERLR, *IOP3XX_ERTVR);
+
+	DBG("ATU: IOP3XX_ATUCMD=0x%04x\n", *IOP3XX_ATUCMD);
+	DBG("ATU: IOP3XX_ATUCR=0x%08x\n", *IOP3XX_ATUCR);
+
+	hook_fault_code(16+6, iop3xx_pci_abort, SIGBUS, 0, "imprecise external abort");
+}
+
+/* for platforms that might be host-bus-adapters */
+void __init iop3xx_pci_preinit_cond(void)
+{
+	if (iop3xx_get_init_atu() == IOP3XX_INIT_ATU_ENABLE) {
+		iop3xx_atu_disable();
+		iop3xx_atu_setup();
+		iop3xx_atu_debug();
+	}
+}
+
+void __init iop3xx_pci_preinit(void)
+{
+	pcibios_min_mem = 0;
+
+	iop3xx_atu_disable();
+	iop3xx_atu_setup();
+	iop3xx_atu_debug();
+}
+
+/* allow init_atu to be user overridden */
+static int __init iop3xx_init_atu_setup(char *str)
+{
+	init_atu = IOP3XX_INIT_ATU_DEFAULT;
+	if (str) {
+		while (*str != '\0') {
+			switch (*str) {
+			case 'y':
+			case 'Y':
+				init_atu = IOP3XX_INIT_ATU_ENABLE;
+				break;
+			case 'n':
+			case 'N':
+				init_atu = IOP3XX_INIT_ATU_DISABLE;
+				break;
+			case ',':
+			case '=':
+				break;
+			default:
+				printk(KERN_DEBUG "\"%s\" malformed at "
+					    "character: \'%c\'",
+					    __func__,
+					    *str);
+				*(str + 1) = '\0';
+			}
+			str++;
+		}
+	}
+
+	return 1;
+}
+
+__setup("iop3xx_init_atu", iop3xx_init_atu_setup);
+

arch/arm/plat-iop/pmu.c

@@ -0,0 +1,39 @@
+/*
+ * PMU IRQ registration for the iop3xx xscale PMU families.
+ * Copyright (C) 2010 Will Deacon, ARM Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/platform_device.h>
+#include <mach/irqs.h>
+
+static struct resource pmu_resource = {
+#ifdef CONFIG_ARCH_IOP32X
+	.start	= IRQ_IOP32X_CORE_PMU,
+	.end	= IRQ_IOP32X_CORE_PMU,
+#endif
+#ifdef CONFIG_ARCH_IOP33X
+	.start	= IRQ_IOP33X_CORE_PMU,
+	.end	= IRQ_IOP33X_CORE_PMU,
+#endif
+	.flags	= IORESOURCE_IRQ,
+};
+
+static struct platform_device pmu_device = {
+	.name		= "xscale-pmu",
+	.id		= -1,
+	.resource	= &pmu_resource,
+	.num_resources	= 1,
+};
+
+static int __init iop3xx_pmu_init(void)
+{
+	platform_device_register(&pmu_device);
+	return 0;
+}
+
+arch_initcall(iop3xx_pmu_init);

arch/arm/plat-iop/restart.c

@@ -0,0 +1,20 @@
+/*
+ * restart.c
+ *
+ * Copyright (C) 2001 MontaVista Software, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <asm/hardware/iop3xx.h>
+#include <asm/system_misc.h>
+#include <mach/hardware.h>
+
+void iop3xx_restart(enum reboot_mode mode, const char *cmd)
+{
+	*IOP3XX_PCSR = 0x30;
+
+	/* Jump into ROM at address 0 */
+	soft_restart(0);
+}

arch/arm/plat-iop/setup.c

@@ -0,0 +1,34 @@
+/*
+ * arch/arm/plat-iop/setup.c
+ *
+ * Author: Nicolas Pitre <nico@fluxnic.net>
+ * Copyright (C) 2001 MontaVista Software, Inc.
+ * Copyright (C) 2004 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <asm/mach/map.h>
+#include <asm/hardware/iop3xx.h>
+
+/*
+ * Standard IO mapping for all IOP3xx based systems.  Note that
+ * the IOP3xx OCCDR must be mapped uncached and unbuffered.
+ */
+static struct map_desc iop3xx_std_desc[] __initdata = {
+	{	/* mem mapped registers */
+		.virtual	= IOP3XX_PERIPHERAL_VIRT_BASE,
+		.pfn		= __phys_to_pfn(IOP3XX_PERIPHERAL_PHYS_BASE),
+		.length		= IOP3XX_PERIPHERAL_SIZE,
+		.type		= MT_UNCACHED,
+	},
+};
+
+void __init iop3xx_map_io(void)
+{
+	iotable_init(iop3xx_std_desc, ARRAY_SIZE(iop3xx_std_desc));
+}

arch/arm/plat-iop/time.c

@@ -0,0 +1,186 @@
+/*
+ * arch/arm/plat-iop/time.c
+ *
+ * Timer code for IOP32x and IOP33x based systems
+ *
+ * Author: Deepak Saxena <dsaxena@mvista.com>
+ *
+ * Copyright 2002-2003 MontaVista Software Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/time.h>
+#include <linux/init.h>
+#include <linux/timex.h>
+#include <linux/io.h>
+#include <linux/clocksource.h>
+#include <linux/clockchips.h>
+#include <linux/export.h>
+#include <linux/sched_clock.h>
+#include <mach/hardware.h>
+#include <asm/irq.h>
+#include <asm/uaccess.h>
+#include <asm/mach/irq.h>
+#include <asm/mach/time.h>
+#include <mach/time.h>
+
+/*
+ * Minimum clocksource/clockevent timer range in seconds
+ */
+#define IOP_MIN_RANGE 4
+
+/*
+ * IOP clocksource (free-running timer 1).
+ */
+static cycle_t notrace iop_clocksource_read(struct clocksource *unused)
+{
+	return 0xffffffffu - read_tcr1();
+}
+
+static struct clocksource iop_clocksource = {
+	.name 		= "iop_timer1",
+	.rating		= 300,
+	.read		= iop_clocksource_read,
+	.mask		= CLOCKSOURCE_MASK(32),
+	.flags		= CLOCK_SOURCE_IS_CONTINUOUS,
+};
+
+/*
+ * IOP sched_clock() implementation via its clocksource.
+ */
+static u64 notrace iop_read_sched_clock(void)
+{
+	return 0xffffffffu - read_tcr1();
+}
+
+/*
+ * IOP clockevents (interrupting timer 0).
+ */
+static int iop_set_next_event(unsigned long delta,
+			      struct clock_event_device *unused)
+{
+	u32 tmr = IOP_TMR_PRIVILEGED | IOP_TMR_RATIO_1_1;
+
+	BUG_ON(delta == 0);
+	write_tmr0(tmr & ~(IOP_TMR_EN | IOP_TMR_RELOAD));
+	write_tcr0(delta);
+	write_tmr0((tmr & ~IOP_TMR_RELOAD) | IOP_TMR_EN);
+
+	return 0;
+}
+
+static unsigned long ticks_per_jiffy;
+
+static int iop_set_periodic(struct clock_event_device *evt)
+{
+	u32 tmr = read_tmr0();
+
+	write_tmr0(tmr & ~IOP_TMR_EN);
+	write_tcr0(ticks_per_jiffy - 1);
+	write_trr0(ticks_per_jiffy - 1);
+	tmr |= (IOP_TMR_RELOAD | IOP_TMR_EN);
+
+	write_tmr0(tmr);
+	return 0;
+}
+
+static int iop_set_oneshot(struct clock_event_device *evt)
+{
+	u32 tmr = read_tmr0();
+
+	/* ->set_next_event sets period and enables timer */
+	tmr &= ~(IOP_TMR_RELOAD | IOP_TMR_EN);
+	write_tmr0(tmr);
+	return 0;
+}
+
+static int iop_shutdown(struct clock_event_device *evt)
+{
+	u32 tmr = read_tmr0();
+
+	tmr &= ~IOP_TMR_EN;
+	write_tmr0(tmr);
+	return 0;
+}
+
+static int iop_resume(struct clock_event_device *evt)
+{
+	u32 tmr = read_tmr0();
+
+	tmr |= IOP_TMR_EN;
+	write_tmr0(tmr);
+	return 0;
+}
+
+static struct clock_event_device iop_clockevent = {
+	.name			= "iop_timer0",
+	.features		= CLOCK_EVT_FEAT_PERIODIC |
+				  CLOCK_EVT_FEAT_ONESHOT,
+	.rating			= 300,
+	.set_next_event		= iop_set_next_event,
+	.set_state_shutdown	= iop_shutdown,
+	.set_state_periodic	= iop_set_periodic,
+	.tick_resume		= iop_resume,
+	.set_state_oneshot	= iop_set_oneshot,
+};
+
+static irqreturn_t
+iop_timer_interrupt(int irq, void *dev_id)
+{
+	struct clock_event_device *evt = dev_id;
+
+	write_tisr(1);
+	evt->event_handler(evt);
+	return IRQ_HANDLED;
+}
+
+static struct irqaction iop_timer_irq = {
+	.name		= "IOP Timer Tick",
+	.handler	= iop_timer_interrupt,
+	.flags		= IRQF_TIMER | IRQF_IRQPOLL,
+	.dev_id		= &iop_clockevent,
+};
+
+static unsigned long iop_tick_rate;
+unsigned long get_iop_tick_rate(void)
+{
+	return iop_tick_rate;
+}
+EXPORT_SYMBOL(get_iop_tick_rate);
+
+void __init iop_init_time(unsigned long tick_rate)
+{
+	u32 timer_ctl;
+
+	sched_clock_register(iop_read_sched_clock, 32, tick_rate);
+
+	ticks_per_jiffy = DIV_ROUND_CLOSEST(tick_rate, HZ);
+	iop_tick_rate = tick_rate;
+
+	timer_ctl = IOP_TMR_EN | IOP_TMR_PRIVILEGED |
+			IOP_TMR_RELOAD | IOP_TMR_RATIO_1_1;
+
+	/*
+	 * Set up interrupting clockevent timer 0.
+	 */
+	write_tmr0(timer_ctl & ~IOP_TMR_EN);
+	write_tisr(1);
+	setup_irq(IRQ_IOP_TIMER0, &iop_timer_irq);
+	iop_clockevent.cpumask = cpumask_of(0);
+	clockevents_config_and_register(&iop_clockevent, tick_rate,
+					0xf, 0xfffffffe);
+
+	/*
+	 * Set up free-running clocksource timer 1.
+	 */
+	write_trr1(0xffffffff);
+	write_tcr1(0xffffffff);
+	write_tmr1(timer_ctl);
+	clocksource_register_hz(&iop_clocksource, tick_rate);
+}

arch/arm/plat-orion/Makefile

@@ -0,0 +1,8 @@
+#
+# Makefile for the linux kernel.
+#
+ccflags-$(CONFIG_ARCH_MULTIPLATFORM) := -I$(srctree)/$(src)/include
+
+orion-gpio-$(CONFIG_GPIOLIB)      += gpio.o
+obj-$(CONFIG_PLAT_ORION_LEGACY)   += irq.o pcie.o time.o common.o mpp.o
+obj-$(CONFIG_PLAT_ORION_LEGACY)   += $(orion-gpio-y)

arch/arm/plat-orion/common.c

@@ -0,0 +1,840 @@
+/*
+ * arch/arm/plat-orion/common.c
+ *
+ * Marvell Orion SoC common setup code used by multiple mach-/common.c
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/dma-mapping.h>
+#include <linux/serial_8250.h>
+#include <linux/ata_platform.h>
+#include <linux/clk.h>
+#include <linux/clkdev.h>
+#include <linux/mv643xx_eth.h>
+#include <linux/mv643xx_i2c.h>
+#include <net/dsa.h>
+#include <linux/platform_data/dma-mv_xor.h>
+#include <linux/platform_data/usb-ehci-orion.h>
+#include <plat/common.h>
+
+/* Create a clkdev entry for a given device/clk */
+void __init orion_clkdev_add(const char *con_id, const char *dev_id,
+			     struct clk *clk)
+{
+	clkdev_create(clk, con_id, "%s", dev_id);
+}
+
+/* Create clkdev entries for all orion platforms except kirkwood.
+   Kirkwood has gated clocks for some of its peripherals, so creates
+   its own clkdev entries. For all the other orion devices, create
+   clkdev entries to the tclk. */
+void __init orion_clkdev_init(struct clk *tclk)
+{
+	orion_clkdev_add(NULL, "orion_spi.0", tclk);
+	orion_clkdev_add(NULL, "orion_spi.1", tclk);
+	orion_clkdev_add(NULL, MV643XX_ETH_NAME ".0", tclk);
+	orion_clkdev_add(NULL, MV643XX_ETH_NAME ".1", tclk);
+	orion_clkdev_add(NULL, MV643XX_ETH_NAME ".2", tclk);
+	orion_clkdev_add(NULL, MV643XX_ETH_NAME ".3", tclk);
+	orion_clkdev_add(NULL, "orion_wdt", tclk);
+	orion_clkdev_add(NULL, MV64XXX_I2C_CTLR_NAME ".0", tclk);
+}
+
+/* Fill in the resources structure and link it into the platform
+   device structure. There is always a memory region, and nearly
+   always an interrupt.*/
+static void fill_resources(struct platform_device *device,
+			   struct resource *resources,
+			   resource_size_t mapbase,
+			   resource_size_t size)
+{
+	device->resource = resources;
+	device->num_resources = 1;
+	resources[0].flags = IORESOURCE_MEM;
+	resources[0].start = mapbase;
+	resources[0].end = mapbase + size;
+}
+
+static void fill_resources_irq(struct platform_device *device,
+			       struct resource *resources,
+			       resource_size_t mapbase,
+			       resource_size_t size,
+			       unsigned int irq)
+{
+	fill_resources(device, resources, mapbase, size);
+
+	device->num_resources++;
+	resources[1].flags = IORESOURCE_IRQ;
+	resources[1].start = irq;
+	resources[1].end = irq;
+}
+
+/*****************************************************************************
+ * UART
+ ****************************************************************************/
+static unsigned long __init uart_get_clk_rate(struct clk *clk)
+{
+	clk_prepare_enable(clk);
+	return clk_get_rate(clk);
+}
+
+static void __init uart_complete(
+	struct platform_device *orion_uart,
+	struct plat_serial8250_port *data,
+	struct resource *resources,
+	void __iomem *membase,
+	resource_size_t mapbase,
+	unsigned int irq,
+	struct clk *clk)
+{
+	data->mapbase = mapbase;
+	data->membase = membase;
+	data->irq = irq;
+	data->uartclk = uart_get_clk_rate(clk);
+	orion_uart->dev.platform_data = data;
+
+	fill_resources_irq(orion_uart, resources, mapbase, 0xff, irq);
+	platform_device_register(orion_uart);
+}
+
+/*****************************************************************************
+ * UART0
+ ****************************************************************************/
+static struct plat_serial8250_port orion_uart0_data[] = {
+	{
+		.flags		= UPF_SKIP_TEST | UPF_BOOT_AUTOCONF,
+		.iotype		= UPIO_MEM,
+		.regshift	= 2,
+	}, {
+	},
+};
+
+static struct resource orion_uart0_resources[2];
+
+static struct platform_device orion_uart0 = {
+	.name			= "serial8250",
+	.id			= PLAT8250_DEV_PLATFORM,
+};
+
+void __init orion_uart0_init(void __iomem *membase,
+			     resource_size_t mapbase,
+			     unsigned int irq,
+			     struct clk *clk)
+{
+	uart_complete(&orion_uart0, orion_uart0_data, orion_uart0_resources,
+		      membase, mapbase, irq, clk);
+}
+
+/*****************************************************************************
+ * UART1
+ ****************************************************************************/
+static struct plat_serial8250_port orion_uart1_data[] = {
+	{
+		.flags		= UPF_SKIP_TEST | UPF_BOOT_AUTOCONF,
+		.iotype		= UPIO_MEM,
+		.regshift	= 2,
+	}, {
+	},
+};
+
+static struct resource orion_uart1_resources[2];
+
+static struct platform_device orion_uart1 = {
+	.name			= "serial8250",
+	.id			= PLAT8250_DEV_PLATFORM1,
+};
+
+void __init orion_uart1_init(void __iomem *membase,
+			     resource_size_t mapbase,
+			     unsigned int irq,
+			     struct clk *clk)
+{
+	uart_complete(&orion_uart1, orion_uart1_data, orion_uart1_resources,
+		      membase, mapbase, irq, clk);
+}
+
+/*****************************************************************************
+ * UART2
+ ****************************************************************************/
+static struct plat_serial8250_port orion_uart2_data[] = {
+	{
+		.flags		= UPF_SKIP_TEST | UPF_BOOT_AUTOCONF,
+		.iotype		= UPIO_MEM,
+		.regshift	= 2,
+	}, {
+	},
+};
+
+static struct resource orion_uart2_resources[2];
+
+static struct platform_device orion_uart2 = {
+	.name			= "serial8250",
+	.id			= PLAT8250_DEV_PLATFORM2,
+};
+
+void __init orion_uart2_init(void __iomem *membase,
+			     resource_size_t mapbase,
+			     unsigned int irq,
+			     struct clk *clk)
+{
+	uart_complete(&orion_uart2, orion_uart2_data, orion_uart2_resources,
+		      membase, mapbase, irq, clk);
+}
+
+/*****************************************************************************
+ * UART3
+ ****************************************************************************/
+static struct plat_serial8250_port orion_uart3_data[] = {
+	{
+		.flags		= UPF_SKIP_TEST | UPF_BOOT_AUTOCONF,
+		.iotype		= UPIO_MEM,
+		.regshift	= 2,
+	}, {
+	},
+};
+
+static struct resource orion_uart3_resources[2];
+
+static struct platform_device orion_uart3 = {
+	.name			= "serial8250",
+	.id			= 3,
+};
+
+void __init orion_uart3_init(void __iomem *membase,
+			     resource_size_t mapbase,
+			     unsigned int irq,
+			     struct clk *clk)
+{
+	uart_complete(&orion_uart3, orion_uart3_data, orion_uart3_resources,
+		      membase, mapbase, irq, clk);
+}
+
+/*****************************************************************************
+ * SoC RTC
+ ****************************************************************************/
+static struct resource orion_rtc_resource[2];
+
+void __init orion_rtc_init(unsigned long mapbase,
+			   unsigned long irq)
+{
+	orion_rtc_resource[0].start = mapbase;
+	orion_rtc_resource[0].end = mapbase + SZ_32 - 1;
+	orion_rtc_resource[0].flags = IORESOURCE_MEM;
+	orion_rtc_resource[1].start = irq;
+	orion_rtc_resource[1].end = irq;
+	orion_rtc_resource[1].flags = IORESOURCE_IRQ;
+
+	platform_device_register_simple("rtc-mv", -1, orion_rtc_resource, 2);
+}
+
+/*****************************************************************************
+ * GE
+ ****************************************************************************/
+static __init void ge_complete(
+	struct mv643xx_eth_shared_platform_data *orion_ge_shared_data,
+	struct resource *orion_ge_resource, unsigned long irq,
+	struct platform_device *orion_ge_shared,
+	struct platform_device *orion_ge_mvmdio,
+	struct mv643xx_eth_platform_data *eth_data,
+	struct platform_device *orion_ge)
+{
+	orion_ge_resource->start = irq;
+	orion_ge_resource->end = irq;
+	eth_data->shared = orion_ge_shared;
+	orion_ge->dev.platform_data = eth_data;
+
+	platform_device_register(orion_ge_shared);
+	if (orion_ge_mvmdio)
+		platform_device_register(orion_ge_mvmdio);
+	platform_device_register(orion_ge);
+}
+
+/*****************************************************************************
+ * GE00
+ ****************************************************************************/
+static struct mv643xx_eth_shared_platform_data orion_ge00_shared_data;
+
+static struct resource orion_ge00_shared_resources[] = {
+	{
+		.name	= "ge00 base",
+	},
+};
+
+static struct platform_device orion_ge00_shared = {
+	.name		= MV643XX_ETH_SHARED_NAME,
+	.id		= 0,
+	.dev		= {
+		.platform_data	= &orion_ge00_shared_data,
+	},
+};
+
+static struct resource orion_ge_mvmdio_resources[] = {
+	{
+		.name	= "ge00 mvmdio base",
+	}, {
+		.name	= "ge00 mvmdio err irq",
+	},
+};
+
+static struct platform_device orion_ge_mvmdio = {
+	.name		= "orion-mdio",
+	.id		= -1,
+};
+
+static struct resource orion_ge00_resources[] = {
+	{
+		.name	= "ge00 irq",
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device orion_ge00 = {
+	.name		= MV643XX_ETH_NAME,
+	.id		= 0,
+	.num_resources	= 1,
+	.resource	= orion_ge00_resources,
+	.dev		= {
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+};
+
+void __init orion_ge00_init(struct mv643xx_eth_platform_data *eth_data,
+			    unsigned long mapbase,
+			    unsigned long irq,
+			    unsigned long irq_err,
+			    unsigned int tx_csum_limit)
+{
+	fill_resources(&orion_ge00_shared, orion_ge00_shared_resources,
+		       mapbase + 0x2000, SZ_16K - 1);
+	fill_resources_irq(&orion_ge_mvmdio, orion_ge_mvmdio_resources,
+			mapbase + 0x2004, 0x84 - 1, irq_err);
+	orion_ge00_shared_data.tx_csum_limit = tx_csum_limit;
+	ge_complete(&orion_ge00_shared_data,
+		    orion_ge00_resources, irq, &orion_ge00_shared,
+		    &orion_ge_mvmdio,
+		    eth_data, &orion_ge00);
+}
+
+/*****************************************************************************
+ * GE01
+ ****************************************************************************/
+static struct mv643xx_eth_shared_platform_data orion_ge01_shared_data;
+
+static struct resource orion_ge01_shared_resources[] = {
+	{
+		.name	= "ge01 base",
+	}
+};
+
+static struct platform_device orion_ge01_shared = {
+	.name		= MV643XX_ETH_SHARED_NAME,
+	.id		= 1,
+	.dev		= {
+		.platform_data	= &orion_ge01_shared_data,
+	},
+};
+
+static struct resource orion_ge01_resources[] = {
+	{
+		.name	= "ge01 irq",
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device orion_ge01 = {
+	.name		= MV643XX_ETH_NAME,
+	.id		= 1,
+	.num_resources	= 1,
+	.resource	= orion_ge01_resources,
+	.dev		= {
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+};
+
+void __init orion_ge01_init(struct mv643xx_eth_platform_data *eth_data,
+			    unsigned long mapbase,
+			    unsigned long irq,
+			    unsigned int tx_csum_limit)
+{
+	fill_resources(&orion_ge01_shared, orion_ge01_shared_resources,
+		       mapbase + 0x2000, SZ_16K - 1);
+	orion_ge01_shared_data.tx_csum_limit = tx_csum_limit;
+	ge_complete(&orion_ge01_shared_data,
+		    orion_ge01_resources, irq, &orion_ge01_shared,
+		    NULL,
+		    eth_data, &orion_ge01);
+}
+
+/*****************************************************************************
+ * GE10
+ ****************************************************************************/
+static struct mv643xx_eth_shared_platform_data orion_ge10_shared_data;
+
+static struct resource orion_ge10_shared_resources[] = {
+	{
+		.name	= "ge10 base",
+	}
+};
+
+static struct platform_device orion_ge10_shared = {
+	.name		= MV643XX_ETH_SHARED_NAME,
+	.id		= 2,
+	.dev		= {
+		.platform_data	= &orion_ge10_shared_data,
+	},
+};
+
+static struct resource orion_ge10_resources[] = {
+	{
+		.name	= "ge10 irq",
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device orion_ge10 = {
+	.name		= MV643XX_ETH_NAME,
+	.id		= 2,
+	.num_resources	= 1,
+	.resource	= orion_ge10_resources,
+	.dev		= {
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+};
+
+void __init orion_ge10_init(struct mv643xx_eth_platform_data *eth_data,
+			    unsigned long mapbase,
+			    unsigned long irq)
+{
+	fill_resources(&orion_ge10_shared, orion_ge10_shared_resources,
+		       mapbase + 0x2000, SZ_16K - 1);
+	ge_complete(&orion_ge10_shared_data,
+		    orion_ge10_resources, irq, &orion_ge10_shared,
+		    NULL,
+		    eth_data, &orion_ge10);
+}
+
+/*****************************************************************************
+ * GE11
+ ****************************************************************************/
+static struct mv643xx_eth_shared_platform_data orion_ge11_shared_data;
+
+static struct resource orion_ge11_shared_resources[] = {
+	{
+		.name	= "ge11 base",
+	},
+};
+
+static struct platform_device orion_ge11_shared = {
+	.name		= MV643XX_ETH_SHARED_NAME,
+	.id		= 3,
+	.dev		= {
+		.platform_data	= &orion_ge11_shared_data,
+	},
+};
+
+static struct resource orion_ge11_resources[] = {
+	{
+		.name	= "ge11 irq",
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device orion_ge11 = {
+	.name		= MV643XX_ETH_NAME,
+	.id		= 3,
+	.num_resources	= 1,
+	.resource	= orion_ge11_resources,
+	.dev		= {
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+};
+
+void __init orion_ge11_init(struct mv643xx_eth_platform_data *eth_data,
+			    unsigned long mapbase,
+			    unsigned long irq)
+{
+	fill_resources(&orion_ge11_shared, orion_ge11_shared_resources,
+		       mapbase + 0x2000, SZ_16K - 1);
+	ge_complete(&orion_ge11_shared_data,
+		    orion_ge11_resources, irq, &orion_ge11_shared,
+		    NULL,
+		    eth_data, &orion_ge11);
+}
+
+/*****************************************************************************
+ * Ethernet switch
+ ****************************************************************************/
+void __init orion_ge00_switch_init(struct dsa_platform_data *d)
+{
+	int i;
+
+	d->netdev = &orion_ge00.dev;
+	for (i = 0; i < d->nr_chips; i++)
+		d->chip[i].host_dev = &orion_ge_mvmdio.dev;
+
+	platform_device_register_data(NULL, "dsa", 0, d, sizeof(d));
+}
+
+/*****************************************************************************
+ * I2C
+ ****************************************************************************/
+static struct mv64xxx_i2c_pdata orion_i2c_pdata = {
+	.freq_n		= 3,
+	.timeout	= 1000, /* Default timeout of 1 second */
+};
+
+static struct resource orion_i2c_resources[2];
+
+static struct platform_device orion_i2c = {
+	.name		= MV64XXX_I2C_CTLR_NAME,
+	.id		= 0,
+	.dev		= {
+		.platform_data	= &orion_i2c_pdata,
+	},
+};
+
+static struct mv64xxx_i2c_pdata orion_i2c_1_pdata = {
+	.freq_n		= 3,
+	.timeout	= 1000, /* Default timeout of 1 second */
+};
+
+static struct resource orion_i2c_1_resources[2];
+
+static struct platform_device orion_i2c_1 = {
+	.name		= MV64XXX_I2C_CTLR_NAME,
+	.id		= 1,
+	.dev		= {
+		.platform_data	= &orion_i2c_1_pdata,
+	},
+};
+
+void __init orion_i2c_init(unsigned long mapbase,
+			   unsigned long irq,
+			   unsigned long freq_m)
+{
+	orion_i2c_pdata.freq_m = freq_m;
+	fill_resources_irq(&orion_i2c, orion_i2c_resources, mapbase,
+		       SZ_32 - 1, irq);
+	platform_device_register(&orion_i2c);
+}
+
+void __init orion_i2c_1_init(unsigned long mapbase,
+			     unsigned long irq,
+			     unsigned long freq_m)
+{
+	orion_i2c_1_pdata.freq_m = freq_m;
+	fill_resources_irq(&orion_i2c_1, orion_i2c_1_resources, mapbase,
+		       SZ_32 - 1, irq);
+	platform_device_register(&orion_i2c_1);
+}
+
+/*****************************************************************************
+ * SPI
+ ****************************************************************************/
+static struct resource orion_spi_resources;
+
+static struct platform_device orion_spi = {
+	.name		= "orion_spi",
+	.id		= 0,
+};
+
+static struct resource orion_spi_1_resources;
+
+static struct platform_device orion_spi_1 = {
+	.name		= "orion_spi",
+	.id		= 1,
+};
+
+/* Note: The SPI silicon core does have interrupts. However the
+ * current Linux software driver does not use interrupts. */
+
+void __init orion_spi_init(unsigned long mapbase)
+{
+	fill_resources(&orion_spi, &orion_spi_resources,
+		       mapbase, SZ_512 - 1);
+	platform_device_register(&orion_spi);
+}
+
+void __init orion_spi_1_init(unsigned long mapbase)
+{
+	fill_resources(&orion_spi_1, &orion_spi_1_resources,
+		       mapbase, SZ_512 - 1);
+	platform_device_register(&orion_spi_1);
+}
+
+/*****************************************************************************
+ * XOR
+ ****************************************************************************/
+static u64 orion_xor_dmamask = DMA_BIT_MASK(32);
+
+/*****************************************************************************
+ * XOR0
+ ****************************************************************************/
+static struct resource orion_xor0_shared_resources[] = {
+	{
+		.name	= "xor 0 low",
+		.flags	= IORESOURCE_MEM,
+	}, {
+		.name	= "xor 0 high",
+		.flags	= IORESOURCE_MEM,
+	}, {
+		.name   = "irq channel 0",
+		.flags  = IORESOURCE_IRQ,
+	}, {
+		.name   = "irq channel 1",
+		.flags  = IORESOURCE_IRQ,
+	},
+};
+
+static struct mv_xor_channel_data orion_xor0_channels_data[2];
+
+static struct mv_xor_platform_data orion_xor0_pdata = {
+	.channels = orion_xor0_channels_data,
+};
+
+static struct platform_device orion_xor0_shared = {
+	.name		= MV_XOR_NAME,
+	.id		= 0,
+	.num_resources	= ARRAY_SIZE(orion_xor0_shared_resources),
+	.resource	= orion_xor0_shared_resources,
+	.dev            = {
+		.dma_mask               = &orion_xor_dmamask,
+		.coherent_dma_mask      = DMA_BIT_MASK(64),
+		.platform_data          = &orion_xor0_pdata,
+	},
+};
+
+void __init orion_xor0_init(unsigned long mapbase_low,
+			    unsigned long mapbase_high,
+			    unsigned long irq_0,
+			    unsigned long irq_1)
+{
+	orion_xor0_shared_resources[0].start = mapbase_low;
+	orion_xor0_shared_resources[0].end = mapbase_low + 0xff;
+	orion_xor0_shared_resources[1].start = mapbase_high;
+	orion_xor0_shared_resources[1].end = mapbase_high + 0xff;
+
+	orion_xor0_shared_resources[2].start = irq_0;
+	orion_xor0_shared_resources[2].end = irq_0;
+	orion_xor0_shared_resources[3].start = irq_1;
+	orion_xor0_shared_resources[3].end = irq_1;
+
+	dma_cap_set(DMA_MEMCPY, orion_xor0_channels_data[0].cap_mask);
+	dma_cap_set(DMA_XOR, orion_xor0_channels_data[0].cap_mask);
+
+	dma_cap_set(DMA_MEMCPY, orion_xor0_channels_data[1].cap_mask);
+	dma_cap_set(DMA_XOR, orion_xor0_channels_data[1].cap_mask);
+
+	platform_device_register(&orion_xor0_shared);
+}
+
+/*****************************************************************************
+ * XOR1
+ ****************************************************************************/
+static struct resource orion_xor1_shared_resources[] = {
+	{
+		.name	= "xor 1 low",
+		.flags	= IORESOURCE_MEM,
+	}, {
+		.name	= "xor 1 high",
+		.flags	= IORESOURCE_MEM,
+	}, {
+		.name   = "irq channel 0",
+		.flags  = IORESOURCE_IRQ,
+	}, {
+		.name   = "irq channel 1",
+		.flags  = IORESOURCE_IRQ,
+	},
+};
+
+static struct mv_xor_channel_data orion_xor1_channels_data[2];
+
+static struct mv_xor_platform_data orion_xor1_pdata = {
+	.channels = orion_xor1_channels_data,
+};
+
+static struct platform_device orion_xor1_shared = {
+	.name		= MV_XOR_NAME,
+	.id		= 1,
+	.num_resources	= ARRAY_SIZE(orion_xor1_shared_resources),
+	.resource	= orion_xor1_shared_resources,
+	.dev            = {
+		.dma_mask               = &orion_xor_dmamask,
+		.coherent_dma_mask      = DMA_BIT_MASK(64),
+		.platform_data          = &orion_xor1_pdata,
+	},
+};
+
+void __init orion_xor1_init(unsigned long mapbase_low,
+			    unsigned long mapbase_high,
+			    unsigned long irq_0,
+			    unsigned long irq_1)
+{
+	orion_xor1_shared_resources[0].start = mapbase_low;
+	orion_xor1_shared_resources[0].end = mapbase_low + 0xff;
+	orion_xor1_shared_resources[1].start = mapbase_high;
+	orion_xor1_shared_resources[1].end = mapbase_high + 0xff;
+
+	orion_xor1_shared_resources[2].start = irq_0;
+	orion_xor1_shared_resources[2].end = irq_0;
+	orion_xor1_shared_resources[3].start = irq_1;
+	orion_xor1_shared_resources[3].end = irq_1;
+
+	dma_cap_set(DMA_MEMCPY, orion_xor1_channels_data[0].cap_mask);
+	dma_cap_set(DMA_XOR, orion_xor1_channels_data[0].cap_mask);
+
+	dma_cap_set(DMA_MEMCPY, orion_xor1_channels_data[1].cap_mask);
+	dma_cap_set(DMA_XOR, orion_xor1_channels_data[1].cap_mask);
+
+	platform_device_register(&orion_xor1_shared);
+}
+
+/*****************************************************************************
+ * EHCI
+ ****************************************************************************/
+static struct orion_ehci_data orion_ehci_data;
+static u64 ehci_dmamask = DMA_BIT_MASK(32);
+
+
+/*****************************************************************************
+ * EHCI0
+ ****************************************************************************/
+static struct resource orion_ehci_resources[2];
+
+static struct platform_device orion_ehci = {
+	.name		= "orion-ehci",
+	.id		= 0,
+	.dev		= {
+		.dma_mask		= &ehci_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+		.platform_data		= &orion_ehci_data,
+	},
+};
+
+void __init orion_ehci_init(unsigned long mapbase,
+			    unsigned long irq,
+			    enum orion_ehci_phy_ver phy_version)
+{
+	orion_ehci_data.phy_version = phy_version;
+	fill_resources_irq(&orion_ehci, orion_ehci_resources, mapbase, SZ_4K - 1,
+		       irq);
+
+	platform_device_register(&orion_ehci);
+}
+
+/*****************************************************************************
+ * EHCI1
+ ****************************************************************************/
+static struct resource orion_ehci_1_resources[2];
+
+static struct platform_device orion_ehci_1 = {
+	.name		= "orion-ehci",
+	.id		= 1,
+	.dev		= {
+		.dma_mask		= &ehci_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+		.platform_data		= &orion_ehci_data,
+	},
+};
+
+void __init orion_ehci_1_init(unsigned long mapbase,
+			      unsigned long irq)
+{
+	fill_resources_irq(&orion_ehci_1, orion_ehci_1_resources,
+		       mapbase, SZ_4K - 1, irq);
+
+	platform_device_register(&orion_ehci_1);
+}
+
+/*****************************************************************************
+ * EHCI2
+ ****************************************************************************/
+static struct resource orion_ehci_2_resources[2];
+
+static struct platform_device orion_ehci_2 = {
+	.name		= "orion-ehci",
+	.id		= 2,
+	.dev		= {
+		.dma_mask		= &ehci_dmamask,
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+		.platform_data		= &orion_ehci_data,
+	},
+};
+
+void __init orion_ehci_2_init(unsigned long mapbase,
+			      unsigned long irq)
+{
+	fill_resources_irq(&orion_ehci_2, orion_ehci_2_resources,
+		       mapbase, SZ_4K - 1, irq);
+
+	platform_device_register(&orion_ehci_2);
+}
+
+/*****************************************************************************
+ * SATA
+ ****************************************************************************/
+static struct resource orion_sata_resources[2] = {
+	{
+		.name	= "sata base",
+	}, {
+		.name	= "sata irq",
+	},
+};
+
+static struct platform_device orion_sata = {
+	.name		= "sata_mv",
+	.id		= 0,
+	.dev		= {
+		.coherent_dma_mask	= DMA_BIT_MASK(32),
+	},
+};
+
+void __init orion_sata_init(struct mv_sata_platform_data *sata_data,
+			    unsigned long mapbase,
+			    unsigned long irq)
+{
+	orion_sata.dev.platform_data = sata_data;
+	fill_resources_irq(&orion_sata, orion_sata_resources,
+		       mapbase, 0x5000 - 1, irq);
+
+	platform_device_register(&orion_sata);
+}
+
+/*****************************************************************************
+ * Cryptographic Engines and Security Accelerator (CESA)
+ ****************************************************************************/
+static struct resource orion_crypto_resources[] = {
+	{
+		.name   = "regs",
+	}, {
+		.name   = "crypto interrupt",
+	}, {
+		.name   = "sram",
+		.flags  = IORESOURCE_MEM,
+	},
+};
+
+static struct platform_device orion_crypto = {
+	.name           = "mv_crypto",
+	.id             = -1,
+};
+
+void __init orion_crypto_init(unsigned long mapbase,
+			      unsigned long srambase,
+			      unsigned long sram_size,
+			      unsigned long irq)
+{
+	fill_resources_irq(&orion_crypto, orion_crypto_resources,
+		       mapbase, 0xffff, irq);
+	orion_crypto.num_resources = 3;
+	orion_crypto_resources[2].start = srambase;
+	orion_crypto_resources[2].end = srambase + sram_size - 1;
+
+	platform_device_register(&orion_crypto);
+}

arch/arm/plat-orion/gpio.c

@@ -0,0 +1,623 @@
+/*
+ * arch/arm/plat-orion/gpio.c
+ *
+ * Marvell Orion SoC GPIO handling.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#define DEBUG
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <linux/irqdomain.h>
+#include <linux/module.h>
+#include <linux/spinlock.h>
+#include <linux/bitops.h>
+#include <linux/io.h>
+#include <linux/gpio.h>
+#include <linux/leds.h>
+#include <linux/of.h>
+#include <linux/of_irq.h>
+#include <linux/of_address.h>
+#include <plat/orion-gpio.h>
+
+/*
+ * GPIO unit register offsets.
+ */
+#define GPIO_OUT_OFF		0x0000
+#define GPIO_IO_CONF_OFF	0x0004
+#define GPIO_BLINK_EN_OFF	0x0008
+#define GPIO_IN_POL_OFF		0x000c
+#define GPIO_DATA_IN_OFF	0x0010
+#define GPIO_EDGE_CAUSE_OFF	0x0014
+#define GPIO_EDGE_MASK_OFF	0x0018
+#define GPIO_LEVEL_MASK_OFF	0x001c
+
+struct orion_gpio_chip {
+	struct gpio_chip	chip;
+	spinlock_t		lock;
+	void __iomem		*base;
+	unsigned long		valid_input;
+	unsigned long		valid_output;
+	int			mask_offset;
+	int			secondary_irq_base;
+	struct irq_domain       *domain;
+};
+
+static void __iomem *GPIO_OUT(struct orion_gpio_chip *ochip)
+{
+	return ochip->base + GPIO_OUT_OFF;
+}
+
+static void __iomem *GPIO_IO_CONF(struct orion_gpio_chip *ochip)
+{
+	return ochip->base + GPIO_IO_CONF_OFF;
+}
+
+static void __iomem *GPIO_BLINK_EN(struct orion_gpio_chip *ochip)
+{
+	return ochip->base + GPIO_BLINK_EN_OFF;
+}
+
+static void __iomem *GPIO_IN_POL(struct orion_gpio_chip *ochip)
+{
+	return ochip->base + GPIO_IN_POL_OFF;
+}
+
+static void __iomem *GPIO_DATA_IN(struct orion_gpio_chip *ochip)
+{
+	return ochip->base + GPIO_DATA_IN_OFF;
+}
+
+static void __iomem *GPIO_EDGE_CAUSE(struct orion_gpio_chip *ochip)
+{
+	return ochip->base + GPIO_EDGE_CAUSE_OFF;
+}
+
+static void __iomem *GPIO_EDGE_MASK(struct orion_gpio_chip *ochip)
+{
+	return ochip->base + ochip->mask_offset + GPIO_EDGE_MASK_OFF;
+}
+
+static void __iomem *GPIO_LEVEL_MASK(struct orion_gpio_chip *ochip)
+{
+	return ochip->base + ochip->mask_offset + GPIO_LEVEL_MASK_OFF;
+}
+
+
+static struct orion_gpio_chip orion_gpio_chips[2];
+static int orion_gpio_chip_count;
+
+static inline void
+__set_direction(struct orion_gpio_chip *ochip, unsigned pin, int input)
+{
+	u32 u;
+
+	u = readl(GPIO_IO_CONF(ochip));
+	if (input)
+		u |= 1 << pin;
+	else
+		u &= ~(1 << pin);
+	writel(u, GPIO_IO_CONF(ochip));
+}
+
+static void __set_level(struct orion_gpio_chip *ochip, unsigned pin, int high)
+{
+	u32 u;
+
+	u = readl(GPIO_OUT(ochip));
+	if (high)
+		u |= 1 << pin;
+	else
+		u &= ~(1 << pin);
+	writel(u, GPIO_OUT(ochip));
+}
+
+static inline void
+__set_blinking(struct orion_gpio_chip *ochip, unsigned pin, int blink)
+{
+	u32 u;
+
+	u = readl(GPIO_BLINK_EN(ochip));
+	if (blink)
+		u |= 1 << pin;
+	else
+		u &= ~(1 << pin);
+	writel(u, GPIO_BLINK_EN(ochip));
+}
+
+static inline int
+orion_gpio_is_valid(struct orion_gpio_chip *ochip, unsigned pin, int mode)
+{
+	if (pin >= ochip->chip.ngpio)
+		goto err_out;
+
+	if ((mode & GPIO_INPUT_OK) && !test_bit(pin, &ochip->valid_input))
+		goto err_out;
+
+	if ((mode & GPIO_OUTPUT_OK) && !test_bit(pin, &ochip->valid_output))
+		goto err_out;
+
+	return 1;
+
+err_out:
+	pr_debug("%s: invalid GPIO %d\n", __func__, pin);
+	return false;
+}
+
+/*
+ * GPIO primitives.
+ */
+static int orion_gpio_request(struct gpio_chip *chip, unsigned pin)
+{
+	struct orion_gpio_chip *ochip = gpiochip_get_data(chip);
+
+	if (orion_gpio_is_valid(ochip, pin, GPIO_INPUT_OK) ||
+	    orion_gpio_is_valid(ochip, pin, GPIO_OUTPUT_OK))
+		return 0;
+
+	return -EINVAL;
+}
+
+static int orion_gpio_direction_input(struct gpio_chip *chip, unsigned pin)
+{
+	struct orion_gpio_chip *ochip = gpiochip_get_data(chip);
+	unsigned long flags;
+
+	if (!orion_gpio_is_valid(ochip, pin, GPIO_INPUT_OK))
+		return -EINVAL;
+
+	spin_lock_irqsave(&ochip->lock, flags);
+	__set_direction(ochip, pin, 1);
+	spin_unlock_irqrestore(&ochip->lock, flags);
+
+	return 0;
+}
+
+static int orion_gpio_get(struct gpio_chip *chip, unsigned pin)
+{
+	struct orion_gpio_chip *ochip = gpiochip_get_data(chip);
+	int val;
+
+	if (readl(GPIO_IO_CONF(ochip)) & (1 << pin)) {
+		val = readl(GPIO_DATA_IN(ochip)) ^ readl(GPIO_IN_POL(ochip));
+	} else {
+		val = readl(GPIO_OUT(ochip));
+	}
+
+	return (val >> pin) & 1;
+}
+
+static int
+orion_gpio_direction_output(struct gpio_chip *chip, unsigned pin, int value)
+{
+	struct orion_gpio_chip *ochip = gpiochip_get_data(chip);
+	unsigned long flags;
+
+	if (!orion_gpio_is_valid(ochip, pin, GPIO_OUTPUT_OK))
+		return -EINVAL;
+
+	spin_lock_irqsave(&ochip->lock, flags);
+	__set_blinking(ochip, pin, 0);
+	__set_level(ochip, pin, value);
+	__set_direction(ochip, pin, 0);
+	spin_unlock_irqrestore(&ochip->lock, flags);
+
+	return 0;
+}
+
+static void orion_gpio_set(struct gpio_chip *chip, unsigned pin, int value)
+{
+	struct orion_gpio_chip *ochip = gpiochip_get_data(chip);
+	unsigned long flags;
+
+	spin_lock_irqsave(&ochip->lock, flags);
+	__set_level(ochip, pin, value);
+	spin_unlock_irqrestore(&ochip->lock, flags);
+}
+
+static int orion_gpio_to_irq(struct gpio_chip *chip, unsigned pin)
+{
+	struct orion_gpio_chip *ochip = gpiochip_get_data(chip);
+
+	return irq_create_mapping(ochip->domain,
+				  ochip->secondary_irq_base + pin);
+}
+
+/*
+ * Orion-specific GPIO API extensions.
+ */
+static struct orion_gpio_chip *orion_gpio_chip_find(int pin)
+{
+	int i;
+
+	for (i = 0; i < orion_gpio_chip_count; i++) {
+		struct orion_gpio_chip *ochip = orion_gpio_chips + i;
+		struct gpio_chip *chip = &ochip->chip;
+
+		if (pin >= chip->base && pin < chip->base + chip->ngpio)
+			return ochip;
+	}
+
+	return NULL;
+}
+
+void __init orion_gpio_set_unused(unsigned pin)
+{
+	struct orion_gpio_chip *ochip = orion_gpio_chip_find(pin);
+
+	if (ochip == NULL)
+		return;
+
+	pin -= ochip->chip.base;
+
+	/* Configure as output, drive low. */
+	__set_level(ochip, pin, 0);
+	__set_direction(ochip, pin, 0);
+}
+
+void __init orion_gpio_set_valid(unsigned pin, int mode)
+{
+	struct orion_gpio_chip *ochip = orion_gpio_chip_find(pin);
+
+	if (ochip == NULL)
+		return;
+
+	pin -= ochip->chip.base;
+
+	if (mode == 1)
+		mode = GPIO_INPUT_OK | GPIO_OUTPUT_OK;
+
+	if (mode & GPIO_INPUT_OK)
+		__set_bit(pin, &ochip->valid_input);
+	else
+		__clear_bit(pin, &ochip->valid_input);
+
+	if (mode & GPIO_OUTPUT_OK)
+		__set_bit(pin, &ochip->valid_output);
+	else
+		__clear_bit(pin, &ochip->valid_output);
+}
+
+void orion_gpio_set_blink(unsigned pin, int blink)
+{
+	struct orion_gpio_chip *ochip = orion_gpio_chip_find(pin);
+	unsigned long flags;
+
+	if (ochip == NULL)
+		return;
+
+	spin_lock_irqsave(&ochip->lock, flags);
+	__set_level(ochip, pin & 31, 0);
+	__set_blinking(ochip, pin & 31, blink);
+	spin_unlock_irqrestore(&ochip->lock, flags);
+}
+EXPORT_SYMBOL(orion_gpio_set_blink);
+
+#define ORION_BLINK_HALF_PERIOD 100 /* ms */
+
+int orion_gpio_led_blink_set(struct gpio_desc *desc, int state,
+	unsigned long *delay_on, unsigned long *delay_off)
+{
+	unsigned gpio = desc_to_gpio(desc);
+
+	if (delay_on && delay_off && !*delay_on && !*delay_off)
+		*delay_on = *delay_off = ORION_BLINK_HALF_PERIOD;
+
+	switch (state) {
+	case GPIO_LED_NO_BLINK_LOW:
+	case GPIO_LED_NO_BLINK_HIGH:
+		orion_gpio_set_blink(gpio, 0);
+		gpio_set_value(gpio, state);
+		break;
+	case GPIO_LED_BLINK:
+		orion_gpio_set_blink(gpio, 1);
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(orion_gpio_led_blink_set);
+
+
+/*****************************************************************************
+ * Orion GPIO IRQ
+ *
+ * GPIO_IN_POL register controls whether GPIO_DATA_IN will hold the same
+ * value of the line or the opposite value.
+ *
+ * Level IRQ handlers: DATA_IN is used directly as cause register.
+ *                     Interrupt are masked by LEVEL_MASK registers.
+ * Edge IRQ handlers:  Change in DATA_IN are latched in EDGE_CAUSE.
+ *                     Interrupt are masked by EDGE_MASK registers.
+ * Both-edge handlers: Similar to regular Edge handlers, but also swaps
+ *                     the polarity to catch the next line transaction.
+ *                     This is a race condition that might not perfectly
+ *                     work on some use cases.
+ *
+ * Every eight GPIO lines are grouped (OR'ed) before going up to main
+ * cause register.
+ *
+ *                    EDGE  cause    mask
+ *        data-in   /--------| |-----| |----\
+ *     -----| |-----                         ---- to main cause reg
+ *           X      \----------------| |----/
+ *        polarity    LEVEL          mask
+ *
+ ****************************************************************************/
+
+static int gpio_irq_set_type(struct irq_data *d, u32 type)
+{
+	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d);
+	struct irq_chip_type *ct = irq_data_get_chip_type(d);
+	struct orion_gpio_chip *ochip = gc->private;
+	int pin;
+	u32 u;
+
+	pin = d->hwirq - ochip->secondary_irq_base;
+
+	u = readl(GPIO_IO_CONF(ochip)) & (1 << pin);
+	if (!u) {
+		return -EINVAL;
+	}
+
+	type &= IRQ_TYPE_SENSE_MASK;
+	if (type == IRQ_TYPE_NONE)
+		return -EINVAL;
+
+	/* Check if we need to change chip and handler */
+	if (!(ct->type & type))
+		if (irq_setup_alt_chip(d, type))
+			return -EINVAL;
+
+	/*
+	 * Configure interrupt polarity.
+	 */
+	if (type == IRQ_TYPE_EDGE_RISING || type == IRQ_TYPE_LEVEL_HIGH) {
+		u = readl(GPIO_IN_POL(ochip));
+		u &= ~(1 << pin);
+		writel(u, GPIO_IN_POL(ochip));
+	} else if (type == IRQ_TYPE_EDGE_FALLING || type == IRQ_TYPE_LEVEL_LOW) {
+		u = readl(GPIO_IN_POL(ochip));
+		u |= 1 << pin;
+		writel(u, GPIO_IN_POL(ochip));
+	} else if (type == IRQ_TYPE_EDGE_BOTH) {
+		u32 v;
+
+		v = readl(GPIO_IN_POL(ochip)) ^ readl(GPIO_DATA_IN(ochip));
+
+		/*
+		 * set initial polarity based on current input level
+		 */
+		u = readl(GPIO_IN_POL(ochip));
+		if (v & (1 << pin))
+			u |= 1 << pin;		/* falling */
+		else
+			u &= ~(1 << pin);	/* rising */
+		writel(u, GPIO_IN_POL(ochip));
+	}
+	return 0;
+}
+
+static void gpio_irq_handler(struct irq_desc *desc)
+{
+	struct orion_gpio_chip *ochip = irq_desc_get_handler_data(desc);
+	u32 cause, type;
+	int i;
+
+	if (ochip == NULL)
+		return;
+
+	cause = readl(GPIO_DATA_IN(ochip)) & readl(GPIO_LEVEL_MASK(ochip));
+	cause |= readl(GPIO_EDGE_CAUSE(ochip)) & readl(GPIO_EDGE_MASK(ochip));
+
+	for (i = 0; i < ochip->chip.ngpio; i++) {
+		int irq;
+
+		irq = ochip->secondary_irq_base + i;
+
+		if (!(cause & (1 << i)))
+			continue;
+
+		type = irq_get_trigger_type(irq);
+		if ((type & IRQ_TYPE_SENSE_MASK) == IRQ_TYPE_EDGE_BOTH) {
+			/* Swap polarity (race with GPIO line) */
+			u32 polarity;
+
+			polarity = readl(GPIO_IN_POL(ochip));
+			polarity ^= 1 << i;
+			writel(polarity, GPIO_IN_POL(ochip));
+		}
+		generic_handle_irq(irq);
+	}
+}
+
+#ifdef CONFIG_DEBUG_FS
+#include <linux/seq_file.h>
+
+static void orion_gpio_dbg_show(struct seq_file *s, struct gpio_chip *chip)
+{
+
+	struct orion_gpio_chip *ochip = gpiochip_get_data(chip);
+	u32 out, io_conf, blink, in_pol, data_in, cause, edg_msk, lvl_msk;
+	int i;
+
+	out	= readl_relaxed(GPIO_OUT(ochip));
+	io_conf	= readl_relaxed(GPIO_IO_CONF(ochip));
+	blink	= readl_relaxed(GPIO_BLINK_EN(ochip));
+	in_pol	= readl_relaxed(GPIO_IN_POL(ochip));
+	data_in	= readl_relaxed(GPIO_DATA_IN(ochip));
+	cause	= readl_relaxed(GPIO_EDGE_CAUSE(ochip));
+	edg_msk	= readl_relaxed(GPIO_EDGE_MASK(ochip));
+	lvl_msk	= readl_relaxed(GPIO_LEVEL_MASK(ochip));
+
+	for (i = 0; i < chip->ngpio; i++) {
+		const char *label;
+		u32 msk;
+		bool is_out;
+
+		label = gpiochip_is_requested(chip, i);
+		if (!label)
+			continue;
+
+		msk = 1 << i;
+		is_out = !(io_conf & msk);
+
+		seq_printf(s, " gpio-%-3d (%-20.20s)", chip->base + i, label);
+
+		if (is_out) {
+			seq_printf(s, " out %s %s\n",
+				   out & msk ? "hi" : "lo",
+				   blink & msk ? "(blink )" : "");
+			continue;
+		}
+
+		seq_printf(s, " in  %s (act %s) - IRQ",
+			   (data_in ^ in_pol) & msk  ? "hi" : "lo",
+			   in_pol & msk ? "lo" : "hi");
+		if (!((edg_msk | lvl_msk) & msk)) {
+			seq_printf(s, " disabled\n");
+			continue;
+		}
+		if (edg_msk & msk)
+			seq_printf(s, " edge ");
+		if (lvl_msk & msk)
+			seq_printf(s, " level");
+		seq_printf(s, " (%s)\n", cause & msk ? "pending" : "clear  ");
+	}
+}
+#else
+#define orion_gpio_dbg_show NULL
+#endif
+
+static void orion_gpio_unmask_irq(struct irq_data *d)
+{
+	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d);
+	struct irq_chip_type *ct = irq_data_get_chip_type(d);
+	u32 reg_val;
+	u32 mask = d->mask;
+
+	irq_gc_lock(gc);
+	reg_val = irq_reg_readl(gc, ct->regs.mask);
+	reg_val |= mask;
+	irq_reg_writel(gc, reg_val, ct->regs.mask);
+	irq_gc_unlock(gc);
+}
+
+static void orion_gpio_mask_irq(struct irq_data *d)
+{
+	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d);
+	struct irq_chip_type *ct = irq_data_get_chip_type(d);
+	u32 mask = d->mask;
+	u32 reg_val;
+
+	irq_gc_lock(gc);
+	reg_val = irq_reg_readl(gc, ct->regs.mask);
+	reg_val &= ~mask;
+	irq_reg_writel(gc, reg_val, ct->regs.mask);
+	irq_gc_unlock(gc);
+}
+
+void __init orion_gpio_init(struct device_node *np,
+			    int gpio_base, int ngpio,
+			    void __iomem *base, int mask_offset,
+			    int secondary_irq_base,
+			    int irqs[4])
+{
+	struct orion_gpio_chip *ochip;
+	struct irq_chip_generic *gc;
+	struct irq_chip_type *ct;
+	char gc_label[16];
+	int i;
+
+	if (orion_gpio_chip_count == ARRAY_SIZE(orion_gpio_chips))
+		return;
+
+	snprintf(gc_label, sizeof(gc_label), "orion_gpio%d",
+		orion_gpio_chip_count);
+
+	ochip = orion_gpio_chips + orion_gpio_chip_count;
+	ochip->chip.label = kstrdup(gc_label, GFP_KERNEL);
+	ochip->chip.request = orion_gpio_request;
+	ochip->chip.direction_input = orion_gpio_direction_input;
+	ochip->chip.get = orion_gpio_get;
+	ochip->chip.direction_output = orion_gpio_direction_output;
+	ochip->chip.set = orion_gpio_set;
+	ochip->chip.to_irq = orion_gpio_to_irq;
+	ochip->chip.base = gpio_base;
+	ochip->chip.ngpio = ngpio;
+	ochip->chip.can_sleep = 0;
+#ifdef CONFIG_OF
+	ochip->chip.of_node = np;
+#endif
+	ochip->chip.dbg_show = orion_gpio_dbg_show;
+
+	spin_lock_init(&ochip->lock);
+	ochip->base = (void __iomem *)base;
+	ochip->valid_input = 0;
+	ochip->valid_output = 0;
+	ochip->mask_offset = mask_offset;
+	ochip->secondary_irq_base = secondary_irq_base;
+
+	gpiochip_add_data(&ochip->chip, ochip);
+
+	/*
+	 * Mask and clear GPIO interrupts.
+	 */
+	writel(0, GPIO_EDGE_CAUSE(ochip));
+	writel(0, GPIO_EDGE_MASK(ochip));
+	writel(0, GPIO_LEVEL_MASK(ochip));
+
+	/* Setup the interrupt handlers. Each chip can have up to 4
+	 * interrupt handlers, with each handler dealing with 8 GPIO
+	 * pins. */
+
+	for (i = 0; i < 4; i++) {
+		if (irqs[i]) {
+			irq_set_chained_handler_and_data(irqs[i],
+							 gpio_irq_handler,
+							 ochip);
+		}
+	}
+
+	gc = irq_alloc_generic_chip("orion_gpio_irq", 2,
+				    secondary_irq_base,
+				    ochip->base, handle_level_irq);
+	gc->private = ochip;
+	ct = gc->chip_types;
+	ct->regs.mask = ochip->mask_offset + GPIO_LEVEL_MASK_OFF;
+	ct->type = IRQ_TYPE_LEVEL_HIGH | IRQ_TYPE_LEVEL_LOW;
+	ct->chip.irq_mask = orion_gpio_mask_irq;
+	ct->chip.irq_unmask = orion_gpio_unmask_irq;
+	ct->chip.irq_set_type = gpio_irq_set_type;
+	ct->chip.name = ochip->chip.label;
+
+	ct++;
+	ct->regs.mask = ochip->mask_offset + GPIO_EDGE_MASK_OFF;
+	ct->regs.ack = GPIO_EDGE_CAUSE_OFF;
+	ct->type = IRQ_TYPE_EDGE_RISING | IRQ_TYPE_EDGE_FALLING;
+	ct->chip.irq_ack = irq_gc_ack_clr_bit;
+	ct->chip.irq_mask = orion_gpio_mask_irq;
+	ct->chip.irq_unmask = orion_gpio_unmask_irq;
+	ct->chip.irq_set_type = gpio_irq_set_type;
+	ct->handler = handle_edge_irq;
+	ct->chip.name = ochip->chip.label;
+
+	irq_setup_generic_chip(gc, IRQ_MSK(ngpio), IRQ_GC_INIT_MASK_CACHE,
+			       IRQ_NOREQUEST, IRQ_LEVEL | IRQ_NOPROBE);
+
+	/* Setup irq domain on top of the generic chip. */
+	ochip->domain = irq_domain_add_legacy(np,
+					      ochip->chip.ngpio,
+					      ochip->secondary_irq_base,
+					      ochip->secondary_irq_base,
+					      &irq_domain_simple_ops,
+					      ochip);
+	if (!ochip->domain)
+		panic("%s: couldn't allocate irq domain (DT).\n",
+		      ochip->chip.label);
+
+	orion_gpio_chip_count++;
+}

arch/arm/plat-orion/include/plat/addr-map.h

@@ -0,0 +1,54 @@
+/*
+ * arch/arm/plat-orion/include/plat/addr-map.h
+ *
+ * Marvell Orion SoC address map handling.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef __PLAT_ADDR_MAP_H
+#define __PLAT_ADDR_MAP_H
+
+extern struct mbus_dram_target_info orion_mbus_dram_info;
+
+struct orion_addr_map_cfg {
+	const int num_wins;	/* Total number of windows */
+	const int remappable_wins;
+	void __iomem *bridge_virt_base;
+	int hw_io_coherency;
+
+	/* If NULL, the default cpu_win_can_remap will be used, using
+	   the value in remappable_wins */
+	int (*cpu_win_can_remap) (const struct orion_addr_map_cfg *cfg,
+				  const int win);
+	/* If NULL, the default win_cfg_base will be used, using the
+	   value in bridge_virt_base */
+	void __iomem *(*win_cfg_base) (const struct orion_addr_map_cfg *cfg,
+				 const int win);
+};
+
+/*
+ * Information needed to setup one address mapping.
+ */
+struct orion_addr_map_info {
+	const int win;
+	const u32 base;
+	const u32 size;
+	const u8 target;
+	const u8 attr;
+	const int remap;
+};
+
+void __init orion_config_wins(struct orion_addr_map_cfg *cfg,
+			      const struct orion_addr_map_info *info);
+
+void __init orion_setup_cpu_win(const struct orion_addr_map_cfg *cfg,
+				const int win, const u32 base,
+				const u32 size, const u8 target,
+				const u8 attr, const int remap);
+
+void __init orion_setup_cpu_mbus_target(const struct orion_addr_map_cfg *cfg,
+					const void __iomem *ddr_window_cpu_base);
+#endif

arch/arm/plat-orion/include/plat/common.h

@@ -0,0 +1,107 @@
+/*
+ * arch/arm/plat-orion/include/plat/common.h
+ *
+ * Marvell Orion SoC common setup code used by different mach-/common.c
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef __PLAT_COMMON_H
+#include <linux/mv643xx_eth.h>
+#include <linux/platform_data/usb-ehci-orion.h>
+
+struct dsa_platform_data;
+struct mv_sata_platform_data;
+
+void __init orion_uart0_init(void __iomem *membase,
+			     resource_size_t mapbase,
+			     unsigned int irq,
+			     struct clk *clk);
+
+void __init orion_uart1_init(void __iomem *membase,
+			     resource_size_t mapbase,
+			     unsigned int irq,
+			     struct clk *clk);
+
+void __init orion_uart2_init(void __iomem *membase,
+			     resource_size_t mapbase,
+			     unsigned int irq,
+			     struct clk *clk);
+
+void __init orion_uart3_init(void __iomem *membase,
+			     resource_size_t mapbase,
+			     unsigned int irq,
+			     struct clk *clk);
+
+void __init orion_rtc_init(unsigned long mapbase,
+			   unsigned long irq);
+
+void __init orion_ge00_init(struct mv643xx_eth_platform_data *eth_data,
+			    unsigned long mapbase,
+			    unsigned long irq,
+			    unsigned long irq_err,
+			    unsigned int tx_csum_limit);
+
+void __init orion_ge01_init(struct mv643xx_eth_platform_data *eth_data,
+			    unsigned long mapbase,
+			    unsigned long irq,
+			    unsigned int tx_csum_limit);
+
+void __init orion_ge10_init(struct mv643xx_eth_platform_data *eth_data,
+			    unsigned long mapbase,
+			    unsigned long irq);
+
+void __init orion_ge11_init(struct mv643xx_eth_platform_data *eth_data,
+			    unsigned long mapbase,
+			    unsigned long irq);
+
+void __init orion_ge00_switch_init(struct dsa_platform_data *d);
+
+void __init orion_i2c_init(unsigned long mapbase,
+			   unsigned long irq,
+			   unsigned long freq_m);
+
+void __init orion_i2c_1_init(unsigned long mapbase,
+			     unsigned long irq,
+			     unsigned long freq_m);
+
+void __init orion_spi_init(unsigned long mapbase);
+
+void __init orion_spi_1_init(unsigned long mapbase);
+
+void __init orion_xor0_init(unsigned long mapbase_low,
+			    unsigned long mapbase_high,
+			    unsigned long irq_0,
+			    unsigned long irq_1);
+
+void __init orion_xor1_init(unsigned long mapbase_low,
+			    unsigned long mapbase_high,
+			    unsigned long irq_0,
+			    unsigned long irq_1);
+
+void __init orion_ehci_init(unsigned long mapbase,
+			    unsigned long irq,
+			    enum orion_ehci_phy_ver phy_version);
+
+void __init orion_ehci_1_init(unsigned long mapbase,
+			      unsigned long irq);
+
+void __init orion_ehci_2_init(unsigned long mapbase,
+			      unsigned long irq);
+
+void __init orion_sata_init(struct mv_sata_platform_data *sata_data,
+			    unsigned long mapbase,
+			    unsigned long irq);
+
+void __init orion_crypto_init(unsigned long mapbase,
+			      unsigned long srambase,
+			      unsigned long sram_size,
+			      unsigned long irq);
+
+void __init orion_clkdev_add(const char *con_id, const char *dev_id,
+			     struct clk *clk);
+
+void __init orion_clkdev_init(struct clk *tclk);
+#endif

arch/arm/plat-orion/include/plat/irq.h

@@ -0,0 +1,15 @@
+/*
+ * arch/arm/plat-orion/include/plat/irq.h
+ *
+ * Marvell Orion SoC IRQ handling.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef __PLAT_IRQ_H
+#define __PLAT_IRQ_H
+
+void orion_irq_init(unsigned int irq_start, void __iomem *maskaddr);
+#endif

arch/arm/plat-orion/include/plat/mpp.h

@@ -0,0 +1,34 @@
+/*
+ * arch/arm/plat-orion/include/plat/mpp.h
+ *
+ * Marvell Orion SoC MPP handling.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef __PLAT_MPP_H
+#define __PLAT_MPP_H
+
+#define MPP_NUM(x)	((x) & 0xff)
+#define MPP_SEL(x)	(((x) >> 8) & 0xf)
+
+/* This is the generic MPP macro, without any variant information.
+   Each machine architecture is expected to extend this with further
+   bit fields indicating which MPP configurations are valid for a
+   specific variant. */
+
+#define GENERIC_MPP(_num, _sel, _in, _out) ( \
+	/* MPP number */		((_num) & 0xff) | \
+	/* MPP select value */		(((_sel) & 0xf) << 8) | \
+	/* may be input signal */	((!!(_in)) << 12) | \
+	/* may be output signal */	((!!(_out)) << 13))
+
+#define MPP_INPUT_MASK		GENERIC_MPP(0, 0x0, 1, 0)
+#define MPP_OUTPUT_MASK		GENERIC_MPP(0, 0x0, 0, 1)
+
+void __init orion_mpp_conf(unsigned int *mpp_list, unsigned int variant_mask,
+			   unsigned int mpp_max, void __iomem *dev_bus);
+
+#endif

arch/arm/plat-orion/include/plat/orion-gpio.h

@@ -0,0 +1,39 @@
+/*
+ * arch/arm/plat-orion/include/plat/orion-gpio.h
+ *
+ * Marvell Orion SoC GPIO handling.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef __PLAT_GPIO_H
+#define __PLAT_GPIO_H
+
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/irqdomain.h>
+
+struct gpio_desc;
+
+/*
+ * Orion-specific GPIO API extensions.
+ */
+void orion_gpio_set_unused(unsigned pin);
+void orion_gpio_set_blink(unsigned pin, int blink);
+int orion_gpio_led_blink_set(struct gpio_desc *desc, int state,
+	unsigned long *delay_on, unsigned long *delay_off);
+
+#define GPIO_INPUT_OK		(1 << 0)
+#define GPIO_OUTPUT_OK		(1 << 1)
+void orion_gpio_set_valid(unsigned pin, int mode);
+
+/* Initialize gpiolib. */
+void __init orion_gpio_init(struct device_node *np,
+			    int gpio_base, int ngpio,
+			    void __iomem *base, int mask_offset,
+			    int secondary_irq_base,
+			    int irq[4]);
+
+#endif

arch/arm/plat-orion/include/plat/pcie.h

@@ -0,0 +1,34 @@
+/*
+ * arch/arm/plat-orion/include/plat/pcie.h
+ *
+ * Marvell Orion SoC PCIe handling.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef __PLAT_PCIE_H
+#define __PLAT_PCIE_H
+
+struct pci_bus;
+
+u32 orion_pcie_dev_id(void __iomem *base);
+u32 orion_pcie_rev(void __iomem *base);
+int orion_pcie_link_up(void __iomem *base);
+int orion_pcie_x4_mode(void __iomem *base);
+int orion_pcie_get_local_bus_nr(void __iomem *base);
+void orion_pcie_set_local_bus_nr(void __iomem *base, int nr);
+void orion_pcie_reset(void __iomem *base);
+void orion_pcie_setup(void __iomem *base);
+int orion_pcie_rd_conf(void __iomem *base, struct pci_bus *bus,
+		       u32 devfn, int where, int size, u32 *val);
+int orion_pcie_rd_conf_tlp(void __iomem *base, struct pci_bus *bus,
+			   u32 devfn, int where, int size, u32 *val);
+int orion_pcie_rd_conf_wa(void __iomem *wa_base, struct pci_bus *bus,
+			  u32 devfn, int where, int size, u32 *val);
+int orion_pcie_wr_conf(void __iomem *base, struct pci_bus *bus,
+		       u32 devfn, int where, int size, u32 val);
+
+
+#endif

arch/arm/plat-orion/include/plat/time.h

@@ -0,0 +1,20 @@
+/*
+ * arch/arm/plat-orion/include/plat/time.h
+ *
+ * Marvell Orion SoC time handling.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef __PLAT_TIME_H
+#define __PLAT_TIME_H
+
+void orion_time_set_base(void __iomem *timer_base);
+
+void orion_time_init(void __iomem *bridge_base, u32 bridge_timer1_clr_mask,
+		     unsigned int irq, unsigned int tclk);
+
+
+#endif

arch/arm/plat-orion/irq.c

@@ -0,0 +1,39 @@
+/*
+ * arch/arm/plat-orion/irq.c
+ *
+ * Marvell Orion SoC IRQ handling.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <linux/irqdomain.h>
+#include <linux/io.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+#include <asm/exception.h>
+#include <plat/irq.h>
+#include <plat/orion-gpio.h>
+
+void __init orion_irq_init(unsigned int irq_start, void __iomem *maskaddr)
+{
+	struct irq_chip_generic *gc;
+	struct irq_chip_type *ct;
+
+	/*
+	 * Mask all interrupts initially.
+	 */
+	writel(0, maskaddr);
+
+	gc = irq_alloc_generic_chip("orion_irq", 1, irq_start, maskaddr,
+				    handle_level_irq);
+	ct = gc->chip_types;
+	ct->chip.irq_mask = irq_gc_mask_clr_bit;
+	ct->chip.irq_unmask = irq_gc_mask_set_bit;
+	irq_setup_generic_chip(gc, IRQ_MSK(32), IRQ_GC_INIT_MASK_CACHE,
+			       IRQ_NOREQUEST, IRQ_LEVEL | IRQ_NOPROBE);
+}

arch/arm/plat-orion/mpp.c

@@ -0,0 +1,77 @@
+/*
+ * arch/arm/plat-orion/mpp.c
+ *
+ * MPP functions for Marvell orion SoCs
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/mbus.h>
+#include <linux/io.h>
+#include <linux/gpio.h>
+#include <plat/orion-gpio.h>
+#include <plat/mpp.h>
+
+/* Address of the ith MPP control register */
+static __init void __iomem *mpp_ctrl_addr(unsigned int i,
+					  void __iomem *dev_bus)
+{
+	return dev_bus + (i) * 4;
+}
+
+
+void __init orion_mpp_conf(unsigned int *mpp_list, unsigned int variant_mask,
+			   unsigned int mpp_max, void __iomem *dev_bus)
+{
+	unsigned int mpp_nr_regs = (1 + mpp_max/8);
+	u32 mpp_ctrl[mpp_nr_regs];
+	int i;
+
+	printk(KERN_DEBUG "initial MPP regs:");
+	for (i = 0; i < mpp_nr_regs; i++) {
+		mpp_ctrl[i] = readl(mpp_ctrl_addr(i, dev_bus));
+		printk(" %08x", mpp_ctrl[i]);
+	}
+	printk("\n");
+
+	for ( ; *mpp_list; mpp_list++) {
+		unsigned int num = MPP_NUM(*mpp_list);
+		unsigned int sel = MPP_SEL(*mpp_list);
+		int shift, gpio_mode;
+
+		if (num > mpp_max) {
+			printk(KERN_ERR "orion_mpp_conf: invalid MPP "
+					"number (%u)\n", num);
+			continue;
+		}
+		if (variant_mask && !(*mpp_list & variant_mask)) {
+			printk(KERN_WARNING
+			       "orion_mpp_conf: requested MPP%u config "
+			       "unavailable on this hardware\n", num);
+			continue;
+		}
+
+		shift = (num & 7) << 2;
+		mpp_ctrl[num / 8] &= ~(0xf << shift);
+		mpp_ctrl[num / 8] |= sel << shift;
+
+		gpio_mode = 0;
+		if (*mpp_list & MPP_INPUT_MASK)
+			gpio_mode |= GPIO_INPUT_OK;
+		if (*mpp_list & MPP_OUTPUT_MASK)
+			gpio_mode |= GPIO_OUTPUT_OK;
+
+		orion_gpio_set_valid(num, gpio_mode);
+	}
+
+	printk(KERN_DEBUG "  final MPP regs:");
+	for (i = 0; i < mpp_nr_regs; i++) {
+		writel(mpp_ctrl[i], mpp_ctrl_addr(i, dev_bus));
+		printk(" %08x", mpp_ctrl[i]);
+	}
+	printk("\n");
+}

arch/arm/plat-orion/pcie.c

@@ -0,0 +1,288 @@
+/*
+ * arch/arm/plat-orion/pcie.c
+ *
+ * Marvell Orion SoC PCIe handling.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/mbus.h>
+#include <asm/mach/pci.h>
+#include <plat/pcie.h>
+#include <plat/addr-map.h>
+#include <linux/delay.h>
+
+/*
+ * PCIe unit register offsets.
+ */
+#define PCIE_DEV_ID_OFF		0x0000
+#define PCIE_CMD_OFF		0x0004
+#define PCIE_DEV_REV_OFF	0x0008
+#define PCIE_BAR_LO_OFF(n)	(0x0010 + ((n) << 3))
+#define PCIE_BAR_HI_OFF(n)	(0x0014 + ((n) << 3))
+#define PCIE_HEADER_LOG_4_OFF	0x0128
+#define PCIE_BAR_CTRL_OFF(n)	(0x1804 + ((n - 1) * 4))
+#define PCIE_WIN04_CTRL_OFF(n)	(0x1820 + ((n) << 4))
+#define PCIE_WIN04_BASE_OFF(n)	(0x1824 + ((n) << 4))
+#define PCIE_WIN04_REMAP_OFF(n)	(0x182c + ((n) << 4))
+#define PCIE_WIN5_CTRL_OFF	0x1880
+#define PCIE_WIN5_BASE_OFF	0x1884
+#define PCIE_WIN5_REMAP_OFF	0x188c
+#define PCIE_CONF_ADDR_OFF	0x18f8
+#define  PCIE_CONF_ADDR_EN		0x80000000
+#define  PCIE_CONF_REG(r)		((((r) & 0xf00) << 16) | ((r) & 0xfc))
+#define  PCIE_CONF_BUS(b)		(((b) & 0xff) << 16)
+#define  PCIE_CONF_DEV(d)		(((d) & 0x1f) << 11)
+#define  PCIE_CONF_FUNC(f)		(((f) & 0x7) << 8)
+#define PCIE_CONF_DATA_OFF	0x18fc
+#define PCIE_MASK_OFF		0x1910
+#define PCIE_CTRL_OFF		0x1a00
+#define  PCIE_CTRL_X1_MODE		0x0001
+#define PCIE_STAT_OFF		0x1a04
+#define  PCIE_STAT_DEV_OFFS		20
+#define  PCIE_STAT_DEV_MASK		0x1f
+#define  PCIE_STAT_BUS_OFFS		8
+#define  PCIE_STAT_BUS_MASK		0xff
+#define  PCIE_STAT_LINK_DOWN		1
+#define PCIE_DEBUG_CTRL         0x1a60
+#define  PCIE_DEBUG_SOFT_RESET		(1<<20)
+
+
+u32 orion_pcie_dev_id(void __iomem *base)
+{
+	return readl(base + PCIE_DEV_ID_OFF) >> 16;
+}
+
+u32 orion_pcie_rev(void __iomem *base)
+{
+	return readl(base + PCIE_DEV_REV_OFF) & 0xff;
+}
+
+int orion_pcie_link_up(void __iomem *base)
+{
+	return !(readl(base + PCIE_STAT_OFF) & PCIE_STAT_LINK_DOWN);
+}
+
+int __init orion_pcie_x4_mode(void __iomem *base)
+{
+	return !(readl(base + PCIE_CTRL_OFF) & PCIE_CTRL_X1_MODE);
+}
+
+int orion_pcie_get_local_bus_nr(void __iomem *base)
+{
+	u32 stat = readl(base + PCIE_STAT_OFF);
+
+	return (stat >> PCIE_STAT_BUS_OFFS) & PCIE_STAT_BUS_MASK;
+}
+
+void __init orion_pcie_set_local_bus_nr(void __iomem *base, int nr)
+{
+	u32 stat;
+
+	stat = readl(base + PCIE_STAT_OFF);
+	stat &= ~(PCIE_STAT_BUS_MASK << PCIE_STAT_BUS_OFFS);
+	stat |= nr << PCIE_STAT_BUS_OFFS;
+	writel(stat, base + PCIE_STAT_OFF);
+}
+
+void __init orion_pcie_reset(void __iomem *base)
+{
+	u32 reg;
+	int i;
+
+	/*
+	 * MV-S104860-U0, Rev. C:
+	 * PCI Express Unit Soft Reset
+	 * When set, generates an internal reset in the PCI Express unit.
+	 * This bit should be cleared after the link is re-established.
+	 */
+	reg = readl(base + PCIE_DEBUG_CTRL);
+	reg |= PCIE_DEBUG_SOFT_RESET;
+	writel(reg, base + PCIE_DEBUG_CTRL);
+
+	for (i = 0; i < 20; i++) {
+		mdelay(10);
+
+		if (orion_pcie_link_up(base))
+			break;
+	}
+
+	reg &= ~(PCIE_DEBUG_SOFT_RESET);
+	writel(reg, base + PCIE_DEBUG_CTRL);
+}
+
+/*
+ * Setup PCIE BARs and Address Decode Wins:
+ * BAR[0,2] -> disabled, BAR[1] -> covers all DRAM banks
+ * WIN[0-3] -> DRAM bank[0-3]
+ */
+static void __init orion_pcie_setup_wins(void __iomem *base)
+{
+	const struct mbus_dram_target_info *dram;
+	u32 size;
+	int i;
+
+	dram = mv_mbus_dram_info();
+
+	/*
+	 * First, disable and clear BARs and windows.
+	 */
+	for (i = 1; i <= 2; i++) {
+		writel(0, base + PCIE_BAR_CTRL_OFF(i));
+		writel(0, base + PCIE_BAR_LO_OFF(i));
+		writel(0, base + PCIE_BAR_HI_OFF(i));
+	}
+
+	for (i = 0; i < 5; i++) {
+		writel(0, base + PCIE_WIN04_CTRL_OFF(i));
+		writel(0, base + PCIE_WIN04_BASE_OFF(i));
+		writel(0, base + PCIE_WIN04_REMAP_OFF(i));
+	}
+
+	writel(0, base + PCIE_WIN5_CTRL_OFF);
+	writel(0, base + PCIE_WIN5_BASE_OFF);
+	writel(0, base + PCIE_WIN5_REMAP_OFF);
+
+	/*
+	 * Setup windows for DDR banks.  Count total DDR size on the fly.
+	 */
+	size = 0;
+	for (i = 0; i < dram->num_cs; i++) {
+		const struct mbus_dram_window *cs = dram->cs + i;
+
+		writel(cs->base & 0xffff0000, base + PCIE_WIN04_BASE_OFF(i));
+		writel(0, base + PCIE_WIN04_REMAP_OFF(i));
+		writel(((cs->size - 1) & 0xffff0000) |
+			(cs->mbus_attr << 8) |
+			(dram->mbus_dram_target_id << 4) | 1,
+				base + PCIE_WIN04_CTRL_OFF(i));
+
+		size += cs->size;
+	}
+
+	/*
+	 * Round up 'size' to the nearest power of two.
+	 */
+	if ((size & (size - 1)) != 0)
+		size = 1 << fls(size);
+
+	/*
+	 * Setup BAR[1] to all DRAM banks.
+	 */
+	writel(dram->cs[0].base, base + PCIE_BAR_LO_OFF(1));
+	writel(0, base + PCIE_BAR_HI_OFF(1));
+	writel(((size - 1) & 0xffff0000) | 1, base + PCIE_BAR_CTRL_OFF(1));
+}
+
+void __init orion_pcie_setup(void __iomem *base)
+{
+	u16 cmd;
+	u32 mask;
+
+	/*
+	 * Point PCIe unit MBUS decode windows to DRAM space.
+	 */
+	orion_pcie_setup_wins(base);
+
+	/*
+	 * Master + slave enable.
+	 */
+	cmd = readw(base + PCIE_CMD_OFF);
+	cmd |= PCI_COMMAND_IO;
+	cmd |= PCI_COMMAND_MEMORY;
+	cmd |= PCI_COMMAND_MASTER;
+	writew(cmd, base + PCIE_CMD_OFF);
+
+	/*
+	 * Enable interrupt lines A-D.
+	 */
+	mask = readl(base + PCIE_MASK_OFF);
+	mask |= 0x0f000000;
+	writel(mask, base + PCIE_MASK_OFF);
+}
+
+int orion_pcie_rd_conf(void __iomem *base, struct pci_bus *bus,
+		       u32 devfn, int where, int size, u32 *val)
+{
+	writel(PCIE_CONF_BUS(bus->number) |
+		PCIE_CONF_DEV(PCI_SLOT(devfn)) |
+		PCIE_CONF_FUNC(PCI_FUNC(devfn)) |
+		PCIE_CONF_REG(where) | PCIE_CONF_ADDR_EN,
+			base + PCIE_CONF_ADDR_OFF);
+
+	*val = readl(base + PCIE_CONF_DATA_OFF);
+
+	if (size == 1)
+		*val = (*val >> (8 * (where & 3))) & 0xff;
+	else if (size == 2)
+		*val = (*val >> (8 * (where & 3))) & 0xffff;
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+int orion_pcie_rd_conf_tlp(void __iomem *base, struct pci_bus *bus,
+			   u32 devfn, int where, int size, u32 *val)
+{
+	writel(PCIE_CONF_BUS(bus->number) |
+		PCIE_CONF_DEV(PCI_SLOT(devfn)) |
+		PCIE_CONF_FUNC(PCI_FUNC(devfn)) |
+		PCIE_CONF_REG(where) | PCIE_CONF_ADDR_EN,
+			base + PCIE_CONF_ADDR_OFF);
+
+	*val = readl(base + PCIE_CONF_DATA_OFF);
+
+	if (bus->number != orion_pcie_get_local_bus_nr(base) ||
+	    PCI_FUNC(devfn) != 0)
+		*val = readl(base + PCIE_HEADER_LOG_4_OFF);
+
+	if (size == 1)
+		*val = (*val >> (8 * (where & 3))) & 0xff;
+	else if (size == 2)
+		*val = (*val >> (8 * (where & 3))) & 0xffff;
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+int orion_pcie_rd_conf_wa(void __iomem *wa_base, struct pci_bus *bus,
+			  u32 devfn, int where, int size, u32 *val)
+{
+	*val = readl(wa_base + (PCIE_CONF_BUS(bus->number) |
+				PCIE_CONF_DEV(PCI_SLOT(devfn)) |
+				PCIE_CONF_FUNC(PCI_FUNC(devfn)) |
+				PCIE_CONF_REG(where)));
+
+	if (size == 1)
+		*val = (*val >> (8 * (where & 3))) & 0xff;
+	else if (size == 2)
+		*val = (*val >> (8 * (where & 3))) & 0xffff;
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+int orion_pcie_wr_conf(void __iomem *base, struct pci_bus *bus,
+		       u32 devfn, int where, int size, u32 val)
+{
+	int ret = PCIBIOS_SUCCESSFUL;
+
+	writel(PCIE_CONF_BUS(bus->number) |
+		PCIE_CONF_DEV(PCI_SLOT(devfn)) |
+		PCIE_CONF_FUNC(PCI_FUNC(devfn)) |
+		PCIE_CONF_REG(where) | PCIE_CONF_ADDR_EN,
+			base + PCIE_CONF_ADDR_OFF);
+
+	if (size == 4) {
+		writel(val, base + PCIE_CONF_DATA_OFF);
+	} else if (size == 2) {
+		writew(val, base + PCIE_CONF_DATA_OFF + (where & 3));
+	} else if (size == 1) {
+		writeb(val, base + PCIE_CONF_DATA_OFF + (where & 3));
+	} else {
+		ret = PCIBIOS_BAD_REGISTER_NUMBER;
+	}
+
+	return ret;
+}

arch/arm/plat-orion/time.c

@@ -0,0 +1,242 @@
+/*
+ * arch/arm/plat-orion/time.c
+ *
+ * Marvell Orion SoC timer handling.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ *
+ * Timer 0 is used as free-running clocksource, while timer 1 is
+ * used as clock_event_device.
+ */
+
+#include <linux/kernel.h>
+#include <linux/timer.h>
+#include <linux/clockchips.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/sched_clock.h>
+#include <plat/time.h>
+#include <asm/delay.h>
+
+/*
+ * MBus bridge block registers.
+ */
+#define BRIDGE_CAUSE_OFF	0x0110
+#define BRIDGE_MASK_OFF		0x0114
+#define  BRIDGE_INT_TIMER0	 0x0002
+#define  BRIDGE_INT_TIMER1	 0x0004
+
+
+/*
+ * Timer block registers.
+ */
+#define TIMER_CTRL_OFF		0x0000
+#define  TIMER0_EN		 0x0001
+#define  TIMER0_RELOAD_EN	 0x0002
+#define  TIMER1_EN		 0x0004
+#define  TIMER1_RELOAD_EN	 0x0008
+#define TIMER0_RELOAD_OFF	0x0010
+#define TIMER0_VAL_OFF		0x0014
+#define TIMER1_RELOAD_OFF	0x0018
+#define TIMER1_VAL_OFF		0x001c
+
+
+/*
+ * SoC-specific data.
+ */
+static void __iomem *bridge_base;
+static u32 bridge_timer1_clr_mask;
+static void __iomem *timer_base;
+
+
+/*
+ * Number of timer ticks per jiffy.
+ */
+static u32 ticks_per_jiffy;
+
+
+/*
+ * Orion's sched_clock implementation. It has a resolution of
+ * at least 7.5ns (133MHz TCLK).
+ */
+
+static u64 notrace orion_read_sched_clock(void)
+{
+	return ~readl(timer_base + TIMER0_VAL_OFF);
+}
+
+/*
+ * Clockevent handling.
+ */
+static int
+orion_clkevt_next_event(unsigned long delta, struct clock_event_device *dev)
+{
+	unsigned long flags;
+	u32 u;
+
+	if (delta == 0)
+		return -ETIME;
+
+	local_irq_save(flags);
+
+	/*
+	 * Clear and enable clockevent timer interrupt.
+	 */
+	writel(bridge_timer1_clr_mask, bridge_base + BRIDGE_CAUSE_OFF);
+
+	u = readl(bridge_base + BRIDGE_MASK_OFF);
+	u |= BRIDGE_INT_TIMER1;
+	writel(u, bridge_base + BRIDGE_MASK_OFF);
+
+	/*
+	 * Setup new clockevent timer value.
+	 */
+	writel(delta, timer_base + TIMER1_VAL_OFF);
+
+	/*
+	 * Enable the timer.
+	 */
+	u = readl(timer_base + TIMER_CTRL_OFF);
+	u = (u & ~TIMER1_RELOAD_EN) | TIMER1_EN;
+	writel(u, timer_base + TIMER_CTRL_OFF);
+
+	local_irq_restore(flags);
+
+	return 0;
+}
+
+static int orion_clkevt_shutdown(struct clock_event_device *evt)
+{
+	unsigned long flags;
+	u32 u;
+
+	local_irq_save(flags);
+
+	/* Disable timer */
+	u = readl(timer_base + TIMER_CTRL_OFF);
+	writel(u & ~TIMER1_EN, timer_base + TIMER_CTRL_OFF);
+
+	/* Disable timer interrupt */
+	u = readl(bridge_base + BRIDGE_MASK_OFF);
+	writel(u & ~BRIDGE_INT_TIMER1, bridge_base + BRIDGE_MASK_OFF);
+
+	/* ACK pending timer interrupt */
+	writel(bridge_timer1_clr_mask, bridge_base + BRIDGE_CAUSE_OFF);
+
+	local_irq_restore(flags);
+
+	return 0;
+}
+
+static int orion_clkevt_set_periodic(struct clock_event_device *evt)
+{
+	unsigned long flags;
+	u32 u;
+
+	local_irq_save(flags);
+
+	/* Setup timer to fire at 1/HZ intervals */
+	writel(ticks_per_jiffy - 1, timer_base + TIMER1_RELOAD_OFF);
+	writel(ticks_per_jiffy - 1, timer_base + TIMER1_VAL_OFF);
+
+	/* Enable timer interrupt */
+	u = readl(bridge_base + BRIDGE_MASK_OFF);
+	writel(u | BRIDGE_INT_TIMER1, bridge_base + BRIDGE_MASK_OFF);
+
+	/* Enable timer */
+	u = readl(timer_base + TIMER_CTRL_OFF);
+	writel(u | TIMER1_EN | TIMER1_RELOAD_EN, timer_base + TIMER_CTRL_OFF);
+
+	local_irq_restore(flags);
+
+	return 0;
+}
+
+static struct clock_event_device orion_clkevt = {
+	.name			= "orion_tick",
+	.features		= CLOCK_EVT_FEAT_ONESHOT |
+				  CLOCK_EVT_FEAT_PERIODIC,
+	.rating			= 300,
+	.set_next_event		= orion_clkevt_next_event,
+	.set_state_shutdown	= orion_clkevt_shutdown,
+	.set_state_periodic	= orion_clkevt_set_periodic,
+	.set_state_oneshot	= orion_clkevt_shutdown,
+	.tick_resume		= orion_clkevt_shutdown,
+};
+
+static irqreturn_t orion_timer_interrupt(int irq, void *dev_id)
+{
+	/*
+	 * ACK timer interrupt and call event handler.
+	 */
+	writel(bridge_timer1_clr_mask, bridge_base + BRIDGE_CAUSE_OFF);
+	orion_clkevt.event_handler(&orion_clkevt);
+
+	return IRQ_HANDLED;
+}
+
+static struct irqaction orion_timer_irq = {
+	.name		= "orion_tick",
+	.flags		= IRQF_TIMER,
+	.handler	= orion_timer_interrupt
+};
+
+void __init
+orion_time_set_base(void __iomem *_timer_base)
+{
+	timer_base = _timer_base;
+}
+
+static unsigned long orion_delay_timer_read(void)
+{
+	return ~readl(timer_base + TIMER0_VAL_OFF);
+}
+
+static struct delay_timer orion_delay_timer = {
+	.read_current_timer = orion_delay_timer_read,
+};
+
+void __init
+orion_time_init(void __iomem *_bridge_base, u32 _bridge_timer1_clr_mask,
+		unsigned int irq, unsigned int tclk)
+{
+	u32 u;
+
+	/*
+	 * Set SoC-specific data.
+	 */
+	bridge_base = _bridge_base;
+	bridge_timer1_clr_mask = _bridge_timer1_clr_mask;
+
+	ticks_per_jiffy = (tclk + HZ/2) / HZ;
+
+	orion_delay_timer.freq = tclk;
+	register_current_timer_delay(&orion_delay_timer);
+
+	/*
+	 * Set scale and timer for sched_clock.
+	 */
+	sched_clock_register(orion_read_sched_clock, 32, tclk);
+
+	/*
+	 * Setup free-running clocksource timer (interrupts
+	 * disabled).
+	 */
+	writel(0xffffffff, timer_base + TIMER0_VAL_OFF);
+	writel(0xffffffff, timer_base + TIMER0_RELOAD_OFF);
+	u = readl(bridge_base + BRIDGE_MASK_OFF);
+	writel(u & ~BRIDGE_INT_TIMER0, bridge_base + BRIDGE_MASK_OFF);
+	u = readl(timer_base + TIMER_CTRL_OFF);
+	writel(u | TIMER0_EN | TIMER0_RELOAD_EN, timer_base + TIMER_CTRL_OFF);
+	clocksource_mmio_init(timer_base + TIMER0_VAL_OFF, "orion_clocksource",
+		tclk, 300, 32, clocksource_mmio_readl_down);
+
+	/*
+	 * Setup clockevent timer (interrupt-driven).
+	 */
+	setup_irq(irq, &orion_timer_irq);
+	orion_clkevt.cpumask = cpumask_of(0);
+	clockevents_config_and_register(&orion_clkevt, tclk, 1, 0xfffffffe);
+}

arch/arm/plat-pxa/Kconfig

@@ -0,0 +1,8 @@
+if PLAT_PXA
+
+config PXA_SSP
+	tristate
+	help
+	  Enable support for PXA2xx SSP ports
+
+endif

arch/arm/plat-pxa/Makefile

@@ -0,0 +1,9 @@
+#
+# Makefile for code common across different PXA processor families
+#
+ccflags-$(CONFIG_ARCH_MMP) := -I$(srctree)/$(src)/include
+
+obj-$(CONFIG_PXA3xx)		+= mfp.o
+obj-$(CONFIG_ARCH_MMP)		+= mfp.o
+
+obj-$(CONFIG_PXA_SSP)		+= ssp.o

arch/arm/plat-pxa/include/plat/mfp.h

@@ -0,0 +1,475 @@
+/*
+ * arch/arm/plat-pxa/include/plat/mfp.h
+ *
+ *   Common Multi-Function Pin Definitions
+ *
+ * Copyright (C) 2007 Marvell International Ltd.
+ *
+ * 2007-8-21: eric miao <eric.miao@marvell.com>
+ *            initial version
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_PLAT_MFP_H
+#define __ASM_PLAT_MFP_H
+
+#define mfp_to_gpio(m)	((m) % 256)
+
+/* list of all the configurable MFP pins */
+enum {
+	MFP_PIN_INVALID = -1,
+
+	MFP_PIN_GPIO0 = 0,
+	MFP_PIN_GPIO1,
+	MFP_PIN_GPIO2,
+	MFP_PIN_GPIO3,
+	MFP_PIN_GPIO4,
+	MFP_PIN_GPIO5,
+	MFP_PIN_GPIO6,
+	MFP_PIN_GPIO7,
+	MFP_PIN_GPIO8,
+	MFP_PIN_GPIO9,
+	MFP_PIN_GPIO10,
+	MFP_PIN_GPIO11,
+	MFP_PIN_GPIO12,
+	MFP_PIN_GPIO13,
+	MFP_PIN_GPIO14,
+	MFP_PIN_GPIO15,
+	MFP_PIN_GPIO16,
+	MFP_PIN_GPIO17,
+	MFP_PIN_GPIO18,
+	MFP_PIN_GPIO19,
+	MFP_PIN_GPIO20,
+	MFP_PIN_GPIO21,
+	MFP_PIN_GPIO22,
+	MFP_PIN_GPIO23,
+	MFP_PIN_GPIO24,
+	MFP_PIN_GPIO25,
+	MFP_PIN_GPIO26,
+	MFP_PIN_GPIO27,
+	MFP_PIN_GPIO28,
+	MFP_PIN_GPIO29,
+	MFP_PIN_GPIO30,
+	MFP_PIN_GPIO31,
+	MFP_PIN_GPIO32,
+	MFP_PIN_GPIO33,
+	MFP_PIN_GPIO34,
+	MFP_PIN_GPIO35,
+	MFP_PIN_GPIO36,
+	MFP_PIN_GPIO37,
+	MFP_PIN_GPIO38,
+	MFP_PIN_GPIO39,
+	MFP_PIN_GPIO40,
+	MFP_PIN_GPIO41,
+	MFP_PIN_GPIO42,
+	MFP_PIN_GPIO43,
+	MFP_PIN_GPIO44,
+	MFP_PIN_GPIO45,
+	MFP_PIN_GPIO46,
+	MFP_PIN_GPIO47,
+	MFP_PIN_GPIO48,
+	MFP_PIN_GPIO49,
+	MFP_PIN_GPIO50,
+	MFP_PIN_GPIO51,
+	MFP_PIN_GPIO52,
+	MFP_PIN_GPIO53,
+	MFP_PIN_GPIO54,
+	MFP_PIN_GPIO55,
+	MFP_PIN_GPIO56,
+	MFP_PIN_GPIO57,
+	MFP_PIN_GPIO58,
+	MFP_PIN_GPIO59,
+	MFP_PIN_GPIO60,
+	MFP_PIN_GPIO61,
+	MFP_PIN_GPIO62,
+	MFP_PIN_GPIO63,
+	MFP_PIN_GPIO64,
+	MFP_PIN_GPIO65,
+	MFP_PIN_GPIO66,
+	MFP_PIN_GPIO67,
+	MFP_PIN_GPIO68,
+	MFP_PIN_GPIO69,
+	MFP_PIN_GPIO70,
+	MFP_PIN_GPIO71,
+	MFP_PIN_GPIO72,
+	MFP_PIN_GPIO73,
+	MFP_PIN_GPIO74,
+	MFP_PIN_GPIO75,
+	MFP_PIN_GPIO76,
+	MFP_PIN_GPIO77,
+	MFP_PIN_GPIO78,
+	MFP_PIN_GPIO79,
+	MFP_PIN_GPIO80,
+	MFP_PIN_GPIO81,
+	MFP_PIN_GPIO82,
+	MFP_PIN_GPIO83,
+	MFP_PIN_GPIO84,
+	MFP_PIN_GPIO85,
+	MFP_PIN_GPIO86,
+	MFP_PIN_GPIO87,
+	MFP_PIN_GPIO88,
+	MFP_PIN_GPIO89,
+	MFP_PIN_GPIO90,
+	MFP_PIN_GPIO91,
+	MFP_PIN_GPIO92,
+	MFP_PIN_GPIO93,
+	MFP_PIN_GPIO94,
+	MFP_PIN_GPIO95,
+	MFP_PIN_GPIO96,
+	MFP_PIN_GPIO97,
+	MFP_PIN_GPIO98,
+	MFP_PIN_GPIO99,
+	MFP_PIN_GPIO100,
+	MFP_PIN_GPIO101,
+	MFP_PIN_GPIO102,
+	MFP_PIN_GPIO103,
+	MFP_PIN_GPIO104,
+	MFP_PIN_GPIO105,
+	MFP_PIN_GPIO106,
+	MFP_PIN_GPIO107,
+	MFP_PIN_GPIO108,
+	MFP_PIN_GPIO109,
+	MFP_PIN_GPIO110,
+	MFP_PIN_GPIO111,
+	MFP_PIN_GPIO112,
+	MFP_PIN_GPIO113,
+	MFP_PIN_GPIO114,
+	MFP_PIN_GPIO115,
+	MFP_PIN_GPIO116,
+	MFP_PIN_GPIO117,
+	MFP_PIN_GPIO118,
+	MFP_PIN_GPIO119,
+	MFP_PIN_GPIO120,
+	MFP_PIN_GPIO121,
+	MFP_PIN_GPIO122,
+	MFP_PIN_GPIO123,
+	MFP_PIN_GPIO124,
+	MFP_PIN_GPIO125,
+	MFP_PIN_GPIO126,
+	MFP_PIN_GPIO127,
+
+	MFP_PIN_GPIO128,
+	MFP_PIN_GPIO129,
+	MFP_PIN_GPIO130,
+	MFP_PIN_GPIO131,
+	MFP_PIN_GPIO132,
+	MFP_PIN_GPIO133,
+	MFP_PIN_GPIO134,
+	MFP_PIN_GPIO135,
+	MFP_PIN_GPIO136,
+	MFP_PIN_GPIO137,
+	MFP_PIN_GPIO138,
+	MFP_PIN_GPIO139,
+	MFP_PIN_GPIO140,
+	MFP_PIN_GPIO141,
+	MFP_PIN_GPIO142,
+	MFP_PIN_GPIO143,
+	MFP_PIN_GPIO144,
+	MFP_PIN_GPIO145,
+	MFP_PIN_GPIO146,
+	MFP_PIN_GPIO147,
+	MFP_PIN_GPIO148,
+	MFP_PIN_GPIO149,
+	MFP_PIN_GPIO150,
+	MFP_PIN_GPIO151,
+	MFP_PIN_GPIO152,
+	MFP_PIN_GPIO153,
+	MFP_PIN_GPIO154,
+	MFP_PIN_GPIO155,
+	MFP_PIN_GPIO156,
+	MFP_PIN_GPIO157,
+	MFP_PIN_GPIO158,
+	MFP_PIN_GPIO159,
+	MFP_PIN_GPIO160,
+	MFP_PIN_GPIO161,
+	MFP_PIN_GPIO162,
+	MFP_PIN_GPIO163,
+	MFP_PIN_GPIO164,
+	MFP_PIN_GPIO165,
+	MFP_PIN_GPIO166,
+	MFP_PIN_GPIO167,
+	MFP_PIN_GPIO168,
+	MFP_PIN_GPIO169,
+	MFP_PIN_GPIO170,
+	MFP_PIN_GPIO171,
+	MFP_PIN_GPIO172,
+	MFP_PIN_GPIO173,
+	MFP_PIN_GPIO174,
+	MFP_PIN_GPIO175,
+	MFP_PIN_GPIO176,
+	MFP_PIN_GPIO177,
+	MFP_PIN_GPIO178,
+	MFP_PIN_GPIO179,
+	MFP_PIN_GPIO180,
+	MFP_PIN_GPIO181,
+	MFP_PIN_GPIO182,
+	MFP_PIN_GPIO183,
+	MFP_PIN_GPIO184,
+	MFP_PIN_GPIO185,
+	MFP_PIN_GPIO186,
+	MFP_PIN_GPIO187,
+	MFP_PIN_GPIO188,
+	MFP_PIN_GPIO189,
+	MFP_PIN_GPIO190,
+	MFP_PIN_GPIO191,
+
+	MFP_PIN_GPIO255 = 255,
+
+	MFP_PIN_GPIO0_2,
+	MFP_PIN_GPIO1_2,
+	MFP_PIN_GPIO2_2,
+	MFP_PIN_GPIO3_2,
+	MFP_PIN_GPIO4_2,
+	MFP_PIN_GPIO5_2,
+	MFP_PIN_GPIO6_2,
+	MFP_PIN_GPIO7_2,
+	MFP_PIN_GPIO8_2,
+	MFP_PIN_GPIO9_2,
+	MFP_PIN_GPIO10_2,
+	MFP_PIN_GPIO11_2,
+	MFP_PIN_GPIO12_2,
+	MFP_PIN_GPIO13_2,
+	MFP_PIN_GPIO14_2,
+	MFP_PIN_GPIO15_2,
+	MFP_PIN_GPIO16_2,
+	MFP_PIN_GPIO17_2,
+
+	MFP_PIN_ULPI_STP,
+	MFP_PIN_ULPI_NXT,
+	MFP_PIN_ULPI_DIR,
+
+	MFP_PIN_nXCVREN,
+	MFP_PIN_DF_CLE_nOE,
+	MFP_PIN_DF_nADV1_ALE,
+	MFP_PIN_DF_SCLK_E,
+	MFP_PIN_DF_SCLK_S,
+	MFP_PIN_nBE0,
+	MFP_PIN_nBE1,
+	MFP_PIN_DF_nADV2_ALE,
+	MFP_PIN_DF_INT_RnB,
+	MFP_PIN_DF_nCS0,
+	MFP_PIN_DF_nCS1,
+	MFP_PIN_nLUA,
+	MFP_PIN_nLLA,
+	MFP_PIN_DF_nWE,
+	MFP_PIN_DF_ALE_nWE,
+	MFP_PIN_DF_nRE_nOE,
+	MFP_PIN_DF_ADDR0,
+	MFP_PIN_DF_ADDR1,
+	MFP_PIN_DF_ADDR2,
+	MFP_PIN_DF_ADDR3,
+	MFP_PIN_DF_IO0,
+	MFP_PIN_DF_IO1,
+	MFP_PIN_DF_IO2,
+	MFP_PIN_DF_IO3,
+	MFP_PIN_DF_IO4,
+	MFP_PIN_DF_IO5,
+	MFP_PIN_DF_IO6,
+	MFP_PIN_DF_IO7,
+	MFP_PIN_DF_IO8,
+	MFP_PIN_DF_IO9,
+	MFP_PIN_DF_IO10,
+	MFP_PIN_DF_IO11,
+	MFP_PIN_DF_IO12,
+	MFP_PIN_DF_IO13,
+	MFP_PIN_DF_IO14,
+	MFP_PIN_DF_IO15,
+	MFP_PIN_DF_nCS0_SM_nCS2,
+	MFP_PIN_DF_nCS1_SM_nCS3,
+	MFP_PIN_SM_nCS0,
+	MFP_PIN_SM_nCS1,
+	MFP_PIN_DF_WEn,
+	MFP_PIN_DF_REn,
+	MFP_PIN_DF_CLE_SM_OEn,
+	MFP_PIN_DF_ALE_SM_WEn,
+	MFP_PIN_DF_RDY0,
+	MFP_PIN_DF_RDY1,
+
+	MFP_PIN_SM_SCLK,
+	MFP_PIN_SM_BE0,
+	MFP_PIN_SM_BE1,
+	MFP_PIN_SM_ADV,
+	MFP_PIN_SM_ADVMUX,
+	MFP_PIN_SM_RDY,
+
+	MFP_PIN_MMC1_DAT7,
+	MFP_PIN_MMC1_DAT6,
+	MFP_PIN_MMC1_DAT5,
+	MFP_PIN_MMC1_DAT4,
+	MFP_PIN_MMC1_DAT3,
+	MFP_PIN_MMC1_DAT2,
+	MFP_PIN_MMC1_DAT1,
+	MFP_PIN_MMC1_DAT0,
+	MFP_PIN_MMC1_CMD,
+	MFP_PIN_MMC1_CLK,
+	MFP_PIN_MMC1_CD,
+	MFP_PIN_MMC1_WP,
+
+	/* additional pins on PXA930 */
+	MFP_PIN_GSIM_UIO,
+	MFP_PIN_GSIM_UCLK,
+	MFP_PIN_GSIM_UDET,
+	MFP_PIN_GSIM_nURST,
+	MFP_PIN_PMIC_INT,
+	MFP_PIN_RDY,
+
+	/* additional pins on MMP2 */
+	MFP_PIN_TWSI1_SCL,
+	MFP_PIN_TWSI1_SDA,
+	MFP_PIN_TWSI4_SCL,
+	MFP_PIN_TWSI4_SDA,
+	MFP_PIN_CLK_REQ,
+
+	MFP_PIN_MAX,
+};
+
+/*
+ * a possible MFP configuration is represented by a 32-bit integer
+ *
+ * bit  0.. 9 - MFP Pin Number (1024 Pins Maximum)
+ * bit 10..12 - Alternate Function Selection
+ * bit 13..15 - Drive Strength
+ * bit 16..18 - Low Power Mode State
+ * bit 19..20 - Low Power Mode Edge Detection
+ * bit 21..22 - Run Mode Pull State
+ *
+ * to facilitate the definition, the following macros are provided
+ *
+ * MFP_CFG_DEFAULT - default MFP configuration value, with
+ * 		  alternate function = 0,
+ * 		  drive strength = fast 3mA (MFP_DS03X)
+ * 		  low power mode = default
+ * 		  edge detection = none
+ *
+ * MFP_CFG	- default MFPR value with alternate function
+ * MFP_CFG_DRV	- default MFPR value with alternate function and
+ * 		  pin drive strength
+ * MFP_CFG_LPM	- default MFPR value with alternate function and
+ * 		  low power mode
+ * MFP_CFG_X	- default MFPR value with alternate function,
+ * 		  pin drive strength and low power mode
+ */
+
+typedef unsigned long mfp_cfg_t;
+
+#define MFP_PIN(x)		((x) & 0x3ff)
+
+#define MFP_AF0			(0x0 << 10)
+#define MFP_AF1			(0x1 << 10)
+#define MFP_AF2			(0x2 << 10)
+#define MFP_AF3			(0x3 << 10)
+#define MFP_AF4			(0x4 << 10)
+#define MFP_AF5			(0x5 << 10)
+#define MFP_AF6			(0x6 << 10)
+#define MFP_AF7			(0x7 << 10)
+#define MFP_AF_MASK		(0x7 << 10)
+#define MFP_AF(x)		(((x) >> 10) & 0x7)
+
+#define MFP_DS01X		(0x0 << 13)
+#define MFP_DS02X		(0x1 << 13)
+#define MFP_DS03X		(0x2 << 13)
+#define MFP_DS04X		(0x3 << 13)
+#define MFP_DS06X		(0x4 << 13)
+#define MFP_DS08X		(0x5 << 13)
+#define MFP_DS10X		(0x6 << 13)
+#define MFP_DS13X		(0x7 << 13)
+#define MFP_DS_MASK		(0x7 << 13)
+#define MFP_DS(x)		(((x) >> 13) & 0x7)
+
+#define MFP_LPM_DEFAULT		(0x0 << 16)
+#define MFP_LPM_DRIVE_LOW	(0x1 << 16)
+#define MFP_LPM_DRIVE_HIGH	(0x2 << 16)
+#define MFP_LPM_PULL_LOW	(0x3 << 16)
+#define MFP_LPM_PULL_HIGH	(0x4 << 16)
+#define MFP_LPM_FLOAT		(0x5 << 16)
+#define MFP_LPM_INPUT		(0x6 << 16)
+#define MFP_LPM_STATE_MASK	(0x7 << 16)
+#define MFP_LPM_STATE(x)	(((x) >> 16) & 0x7)
+
+#define MFP_LPM_EDGE_NONE	(0x0 << 19)
+#define MFP_LPM_EDGE_RISE	(0x1 << 19)
+#define MFP_LPM_EDGE_FALL	(0x2 << 19)
+#define MFP_LPM_EDGE_BOTH	(0x3 << 19)
+#define MFP_LPM_EDGE_MASK	(0x3 << 19)
+#define MFP_LPM_EDGE(x)		(((x) >> 19) & 0x3)
+
+#define MFP_PULL_NONE		(0x0 << 21)
+#define MFP_PULL_LOW		(0x1 << 21)
+#define MFP_PULL_HIGH		(0x2 << 21)
+#define MFP_PULL_BOTH		(0x3 << 21)
+#define MFP_PULL_FLOAT		(0x4 << 21)
+#define MFP_PULL_MASK		(0x7 << 21)
+#define MFP_PULL(x)		(((x) >> 21) & 0x7)
+
+#define MFP_CFG_DEFAULT		(MFP_AF0 | MFP_DS03X | MFP_LPM_DEFAULT |\
+				 MFP_LPM_EDGE_NONE | MFP_PULL_NONE)
+
+#define MFP_CFG(pin, af)		\
+	((MFP_CFG_DEFAULT & ~MFP_AF_MASK) |\
+	 (MFP_PIN(MFP_PIN_##pin) | MFP_##af))
+
+#define MFP_CFG_DRV(pin, af, drv)	\
+	((MFP_CFG_DEFAULT & ~(MFP_AF_MASK | MFP_DS_MASK)) |\
+	 (MFP_PIN(MFP_PIN_##pin) | MFP_##af | MFP_##drv))
+
+#define MFP_CFG_LPM(pin, af, lpm)	\
+	((MFP_CFG_DEFAULT & ~(MFP_AF_MASK | MFP_LPM_STATE_MASK)) |\
+	 (MFP_PIN(MFP_PIN_##pin) | MFP_##af | MFP_LPM_##lpm))
+
+#define MFP_CFG_X(pin, af, drv, lpm)	\
+	((MFP_CFG_DEFAULT & ~(MFP_AF_MASK | MFP_DS_MASK | MFP_LPM_STATE_MASK)) |\
+	 (MFP_PIN(MFP_PIN_##pin) | MFP_##af | MFP_##drv | MFP_LPM_##lpm))
+
+#if defined(CONFIG_PXA3xx) || defined(CONFIG_ARCH_MMP)
+/*
+ * each MFP pin will have a MFPR register, since the offset of the
+ * register varies between processors, the processor specific code
+ * should initialize the pin offsets by mfp_init()
+ *
+ * mfp_init_base() - accepts a virtual base for all MFPR registers and
+ * initialize the MFP table to a default state
+ *
+ * mfp_init_addr() - accepts a table of "mfp_addr_map" structure, which
+ * represents a range of MFP pins from "start" to "end", with the offset
+ * beginning at "offset", to define a single pin, let "end" = -1.
+ *
+ * use
+ *
+ * MFP_ADDR_X() to define a range of pins
+ * MFP_ADDR()   to define a single pin
+ * MFP_ADDR_END to signal the end of pin offset definitions
+ */
+struct mfp_addr_map {
+	unsigned int	start;
+	unsigned int	end;
+	unsigned long	offset;
+};
+
+#define MFP_ADDR_X(start, end, offset) \
+	{ MFP_PIN_##start, MFP_PIN_##end, offset }
+
+#define MFP_ADDR(pin, offset) \
+	{ MFP_PIN_##pin, -1, offset }
+
+#define MFP_ADDR_END	{ MFP_PIN_INVALID, 0 }
+
+void __init mfp_init_base(void __iomem *mfpr_base);
+void __init mfp_init_addr(struct mfp_addr_map *map);
+
+/*
+ * mfp_{read, write}()	- for direct read/write access to the MFPR register
+ * mfp_config()		- for configuring a group of MFPR registers
+ * mfp_config_lpm()	- configuring all low power MFPR registers for suspend
+ * mfp_config_run()	- configuring all run time  MFPR registers after resume
+ */
+unsigned long mfp_read(int mfp);
+void mfp_write(int mfp, unsigned long mfpr_val);
+void mfp_config(unsigned long *mfp_cfgs, int num);
+void mfp_config_run(void);
+void mfp_config_lpm(void);
+#endif /* CONFIG_PXA3xx || CONFIG_ARCH_MMP */
+
+#endif /* __ASM_PLAT_MFP_H */

arch/arm/plat-pxa/mfp.c

@@ -0,0 +1,285 @@
+/*
+ * linux/arch/arm/plat-pxa/mfp.c
+ *
+ *   Multi-Function Pin Support
+ *
+ * Copyright (C) 2007 Marvell Internation Ltd.
+ *
+ * 2007-08-21: eric miao <eric.miao@marvell.com>
+ *             initial version
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/io.h>
+
+#include <plat/mfp.h>
+
+#define MFPR_SIZE	(PAGE_SIZE)
+
+/* MFPR register bit definitions */
+#define MFPR_PULL_SEL		(0x1 << 15)
+#define MFPR_PULLUP_EN		(0x1 << 14)
+#define MFPR_PULLDOWN_EN	(0x1 << 13)
+#define MFPR_SLEEP_SEL		(0x1 << 9)
+#define MFPR_SLEEP_OE_N		(0x1 << 7)
+#define MFPR_EDGE_CLEAR		(0x1 << 6)
+#define MFPR_EDGE_FALL_EN	(0x1 << 5)
+#define MFPR_EDGE_RISE_EN	(0x1 << 4)
+
+#define MFPR_SLEEP_DATA(x)	((x) << 8)
+#define MFPR_DRIVE(x)		(((x) & 0x7) << 10)
+#define MFPR_AF_SEL(x)		(((x) & 0x7) << 0)
+
+#define MFPR_EDGE_NONE		(0)
+#define MFPR_EDGE_RISE		(MFPR_EDGE_RISE_EN)
+#define MFPR_EDGE_FALL		(MFPR_EDGE_FALL_EN)
+#define MFPR_EDGE_BOTH		(MFPR_EDGE_RISE | MFPR_EDGE_FALL)
+
+/*
+ * Table that determines the low power modes outputs, with actual settings
+ * used in parentheses for don't-care values. Except for the float output,
+ * the configured driven and pulled levels match, so if there is a need for
+ * non-LPM pulled output, the same configuration could probably be used.
+ *
+ * Output value  sleep_oe_n  sleep_data  pullup_en  pulldown_en  pull_sel
+ *                 (bit 7)    (bit 8)    (bit 14)     (bit 13)   (bit 15)
+ *
+ * Input            0          X(0)        X(0)        X(0)       0
+ * Drive 0          0          0           0           X(1)       0
+ * Drive 1          0          1           X(1)        0	  0
+ * Pull hi (1)      1          X(1)        1           0	  0
+ * Pull lo (0)      1          X(0)        0           1	  0
+ * Z (float)        1          X(0)        0           0	  0
+ */
+#define MFPR_LPM_INPUT		(0)
+#define MFPR_LPM_DRIVE_LOW	(MFPR_SLEEP_DATA(0) | MFPR_PULLDOWN_EN)
+#define MFPR_LPM_DRIVE_HIGH    	(MFPR_SLEEP_DATA(1) | MFPR_PULLUP_EN)
+#define MFPR_LPM_PULL_LOW      	(MFPR_LPM_DRIVE_LOW  | MFPR_SLEEP_OE_N)
+#define MFPR_LPM_PULL_HIGH     	(MFPR_LPM_DRIVE_HIGH | MFPR_SLEEP_OE_N)
+#define MFPR_LPM_FLOAT         	(MFPR_SLEEP_OE_N)
+#define MFPR_LPM_MASK		(0xe080)
+
+/*
+ * The pullup and pulldown state of the MFP pin at run mode is by default
+ * determined by the selected alternate function. In case that some buggy
+ * devices need to override this default behavior,  the definitions below
+ * indicates the setting of corresponding MFPR bits
+ *
+ * Definition       pull_sel  pullup_en  pulldown_en
+ * MFPR_PULL_NONE       0         0        0
+ * MFPR_PULL_LOW        1         0        1
+ * MFPR_PULL_HIGH       1         1        0
+ * MFPR_PULL_BOTH       1         1        1
+ * MFPR_PULL_FLOAT	1         0        0
+ */
+#define MFPR_PULL_NONE		(0)
+#define MFPR_PULL_LOW		(MFPR_PULL_SEL | MFPR_PULLDOWN_EN)
+#define MFPR_PULL_BOTH		(MFPR_PULL_LOW | MFPR_PULLUP_EN)
+#define MFPR_PULL_HIGH		(MFPR_PULL_SEL | MFPR_PULLUP_EN)
+#define MFPR_PULL_FLOAT		(MFPR_PULL_SEL)
+
+/* mfp_spin_lock is used to ensure that MFP register configuration
+ * (most likely a read-modify-write operation) is atomic, and that
+ * mfp_table[] is consistent
+ */
+static DEFINE_SPINLOCK(mfp_spin_lock);
+
+static void __iomem *mfpr_mmio_base;
+
+struct mfp_pin {
+	unsigned long	config;		/* -1 for not configured */
+	unsigned long	mfpr_off;	/* MFPRxx Register offset */
+	unsigned long	mfpr_run;	/* Run-Mode Register Value */
+	unsigned long	mfpr_lpm;	/* Low Power Mode Register Value */
+};
+
+static struct mfp_pin mfp_table[MFP_PIN_MAX];
+
+/* mapping of MFP_LPM_* definitions to MFPR_LPM_* register bits */
+static const unsigned long mfpr_lpm[] = {
+	MFPR_LPM_INPUT,
+	MFPR_LPM_DRIVE_LOW,
+	MFPR_LPM_DRIVE_HIGH,
+	MFPR_LPM_PULL_LOW,
+	MFPR_LPM_PULL_HIGH,
+	MFPR_LPM_FLOAT,
+	MFPR_LPM_INPUT,
+};
+
+/* mapping of MFP_PULL_* definitions to MFPR_PULL_* register bits */
+static const unsigned long mfpr_pull[] = {
+	MFPR_PULL_NONE,
+	MFPR_PULL_LOW,
+	MFPR_PULL_HIGH,
+	MFPR_PULL_BOTH,
+	MFPR_PULL_FLOAT,
+};
+
+/* mapping of MFP_LPM_EDGE_* definitions to MFPR_EDGE_* register bits */
+static const unsigned long mfpr_edge[] = {
+	MFPR_EDGE_NONE,
+	MFPR_EDGE_RISE,
+	MFPR_EDGE_FALL,
+	MFPR_EDGE_BOTH,
+};
+
+#define mfpr_readl(off)			\
+	__raw_readl(mfpr_mmio_base + (off))
+
+#define mfpr_writel(off, val)		\
+	__raw_writel(val, mfpr_mmio_base + (off))
+
+#define mfp_configured(p)	((p)->config != -1)
+
+/*
+ * perform a read-back of any valid MFPR register to make sure the
+ * previous writings are finished
+ */
+static unsigned long mfpr_off_readback;
+#define mfpr_sync()	(void)__raw_readl(mfpr_mmio_base + mfpr_off_readback)
+
+static inline void __mfp_config_run(struct mfp_pin *p)
+{
+	if (mfp_configured(p))
+		mfpr_writel(p->mfpr_off, p->mfpr_run);
+}
+
+static inline void __mfp_config_lpm(struct mfp_pin *p)
+{
+	if (mfp_configured(p)) {
+		unsigned long mfpr_clr = (p->mfpr_run & ~MFPR_EDGE_BOTH) | MFPR_EDGE_CLEAR;
+		if (mfpr_clr != p->mfpr_run)
+			mfpr_writel(p->mfpr_off, mfpr_clr);
+		if (p->mfpr_lpm != mfpr_clr)
+			mfpr_writel(p->mfpr_off, p->mfpr_lpm);
+	}
+}
+
+void mfp_config(unsigned long *mfp_cfgs, int num)
+{
+	unsigned long flags;
+	int i;
+
+	spin_lock_irqsave(&mfp_spin_lock, flags);
+
+	for (i = 0; i < num; i++, mfp_cfgs++) {
+		unsigned long tmp, c = *mfp_cfgs;
+		struct mfp_pin *p;
+		int pin, af, drv, lpm, edge, pull;
+
+		pin = MFP_PIN(c);
+		BUG_ON(pin >= MFP_PIN_MAX);
+		p = &mfp_table[pin];
+
+		af  = MFP_AF(c);
+		drv = MFP_DS(c);
+		lpm = MFP_LPM_STATE(c);
+		edge = MFP_LPM_EDGE(c);
+		pull = MFP_PULL(c);
+
+		/* run-mode pull settings will conflict with MFPR bits of
+		 * low power mode state,  calculate mfpr_run and mfpr_lpm
+		 * individually if pull != MFP_PULL_NONE
+		 */
+		tmp = MFPR_AF_SEL(af) | MFPR_DRIVE(drv);
+
+		if (likely(pull == MFP_PULL_NONE)) {
+			p->mfpr_run = tmp | mfpr_lpm[lpm] | mfpr_edge[edge];
+			p->mfpr_lpm = p->mfpr_run;
+		} else {
+			p->mfpr_lpm = tmp | mfpr_lpm[lpm] | mfpr_edge[edge];
+			p->mfpr_run = tmp | mfpr_pull[pull];
+		}
+
+		p->config = c; __mfp_config_run(p);
+	}
+
+	mfpr_sync();
+	spin_unlock_irqrestore(&mfp_spin_lock, flags);
+}
+
+unsigned long mfp_read(int mfp)
+{
+	unsigned long val, flags;
+
+	BUG_ON(mfp < 0 || mfp >= MFP_PIN_MAX);
+
+	spin_lock_irqsave(&mfp_spin_lock, flags);
+	val = mfpr_readl(mfp_table[mfp].mfpr_off);
+	spin_unlock_irqrestore(&mfp_spin_lock, flags);
+
+	return val;
+}
+
+void mfp_write(int mfp, unsigned long val)
+{
+	unsigned long flags;
+
+	BUG_ON(mfp < 0 || mfp >= MFP_PIN_MAX);
+
+	spin_lock_irqsave(&mfp_spin_lock, flags);
+	mfpr_writel(mfp_table[mfp].mfpr_off, val);
+	mfpr_sync();
+	spin_unlock_irqrestore(&mfp_spin_lock, flags);
+}
+
+void __init mfp_init_base(void __iomem *mfpr_base)
+{
+	int i;
+
+	/* initialize the table with default - unconfigured */
+	for (i = 0; i < ARRAY_SIZE(mfp_table); i++)
+		mfp_table[i].config = -1;
+
+	mfpr_mmio_base = mfpr_base;
+}
+
+void __init mfp_init_addr(struct mfp_addr_map *map)
+{
+	struct mfp_addr_map *p;
+	unsigned long offset, flags;
+	int i;
+
+	spin_lock_irqsave(&mfp_spin_lock, flags);
+
+	/* mfp offset for readback */
+	mfpr_off_readback = map[0].offset;
+
+	for (p = map; p->start != MFP_PIN_INVALID; p++) {
+		offset = p->offset;
+		i = p->start;
+
+		do {
+			mfp_table[i].mfpr_off = offset;
+			mfp_table[i].mfpr_run = 0;
+			mfp_table[i].mfpr_lpm = 0;
+			offset += 4; i++;
+		} while ((i <= p->end) && (p->end != -1));
+	}
+
+	spin_unlock_irqrestore(&mfp_spin_lock, flags);
+}
+
+void mfp_config_lpm(void)
+{
+	struct mfp_pin *p = &mfp_table[0];
+	int pin;
+
+	for (pin = 0; pin < ARRAY_SIZE(mfp_table); pin++, p++)
+		__mfp_config_lpm(p);
+}
+
+void mfp_config_run(void)
+{
+	struct mfp_pin *p = &mfp_table[0];
+	int pin;
+
+	for (pin = 0; pin < ARRAY_SIZE(mfp_table); pin++, p++)
+		__mfp_config_run(p);
+}

arch/arm/plat-pxa/ssp.c

@@ -0,0 +1,290 @@
+/*
+ *  linux/arch/arm/mach-pxa/ssp.c
+ *
+ *  based on linux/arch/arm/mach-sa1100/ssp.c by Russell King
+ *
+ *  Copyright (C) 2003 Russell King.
+ *  Copyright (C) 2003 Wolfson Microelectronics PLC
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ *  PXA2xx SSP driver.  This provides the generic core for simple
+ *  IO-based SSP applications and allows easy port setup for DMA access.
+ *
+ *  Author: Liam Girdwood <liam.girdwood@wolfsonmicro.com>
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/errno.h>
+#include <linux/interrupt.h>
+#include <linux/ioport.h>
+#include <linux/init.h>
+#include <linux/mutex.h>
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/platform_device.h>
+#include <linux/spi/pxa2xx_spi.h>
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+
+#include <asm/irq.h>
+
+static DEFINE_MUTEX(ssp_lock);
+static LIST_HEAD(ssp_list);
+
+struct ssp_device *pxa_ssp_request(int port, const char *label)
+{
+	struct ssp_device *ssp = NULL;
+
+	mutex_lock(&ssp_lock);
+
+	list_for_each_entry(ssp, &ssp_list, node) {
+		if (ssp->port_id == port && ssp->use_count == 0) {
+			ssp->use_count++;
+			ssp->label = label;
+			break;
+		}
+	}
+
+	mutex_unlock(&ssp_lock);
+
+	if (&ssp->node == &ssp_list)
+		return NULL;
+
+	return ssp;
+}
+EXPORT_SYMBOL(pxa_ssp_request);
+
+struct ssp_device *pxa_ssp_request_of(const struct device_node *of_node,
+				      const char *label)
+{
+	struct ssp_device *ssp = NULL;
+
+	mutex_lock(&ssp_lock);
+
+	list_for_each_entry(ssp, &ssp_list, node) {
+		if (ssp->of_node == of_node && ssp->use_count == 0) {
+			ssp->use_count++;
+			ssp->label = label;
+			break;
+		}
+	}
+
+	mutex_unlock(&ssp_lock);
+
+	if (&ssp->node == &ssp_list)
+		return NULL;
+
+	return ssp;
+}
+EXPORT_SYMBOL(pxa_ssp_request_of);
+
+void pxa_ssp_free(struct ssp_device *ssp)
+{
+	mutex_lock(&ssp_lock);
+	if (ssp->use_count) {
+		ssp->use_count--;
+		ssp->label = NULL;
+	} else
+		dev_err(&ssp->pdev->dev, "device already free\n");
+	mutex_unlock(&ssp_lock);
+}
+EXPORT_SYMBOL(pxa_ssp_free);
+
+#ifdef CONFIG_OF
+static const struct of_device_id pxa_ssp_of_ids[] = {
+	{ .compatible = "mrvl,pxa25x-ssp",	.data = (void *) PXA25x_SSP },
+	{ .compatible = "mvrl,pxa25x-nssp",	.data = (void *) PXA25x_NSSP },
+	{ .compatible = "mrvl,pxa27x-ssp",	.data = (void *) PXA27x_SSP },
+	{ .compatible = "mrvl,pxa3xx-ssp",	.data = (void *) PXA3xx_SSP },
+	{ .compatible = "mvrl,pxa168-ssp",	.data = (void *) PXA168_SSP },
+	{ .compatible = "mrvl,pxa910-ssp",	.data = (void *) PXA910_SSP },
+	{ .compatible = "mrvl,ce4100-ssp",	.data = (void *) CE4100_SSP },
+	{ },
+};
+MODULE_DEVICE_TABLE(of, pxa_ssp_of_ids);
+#endif
+
+static int pxa_ssp_probe(struct platform_device *pdev)
+{
+	struct resource *res;
+	struct ssp_device *ssp;
+	struct device *dev = &pdev->dev;
+
+	ssp = devm_kzalloc(dev, sizeof(struct ssp_device), GFP_KERNEL);
+	if (ssp == NULL)
+		return -ENOMEM;
+
+	ssp->pdev = pdev;
+
+	ssp->clk = devm_clk_get(dev, NULL);
+	if (IS_ERR(ssp->clk))
+		return PTR_ERR(ssp->clk);
+
+	if (dev->of_node) {
+		struct of_phandle_args dma_spec;
+		struct device_node *np = dev->of_node;
+		int ret;
+
+		/*
+		 * FIXME: we should allocate the DMA channel from this
+		 * context and pass the channel down to the ssp users.
+		 * For now, we lookup the rx and tx indices manually
+		 */
+
+		/* rx */
+		ret = of_parse_phandle_with_args(np, "dmas", "#dma-cells",
+						 0, &dma_spec);
+
+		if (ret) {
+			dev_err(dev, "Can't parse dmas property\n");
+			return -ENODEV;
+		}
+		ssp->drcmr_rx = dma_spec.args[0];
+		of_node_put(dma_spec.np);
+
+		/* tx */
+		ret = of_parse_phandle_with_args(np, "dmas", "#dma-cells",
+						 1, &dma_spec);
+		if (ret) {
+			dev_err(dev, "Can't parse dmas property\n");
+			return -ENODEV;
+		}
+		ssp->drcmr_tx = dma_spec.args[0];
+		of_node_put(dma_spec.np);
+	} else {
+		res = platform_get_resource(pdev, IORESOURCE_DMA, 0);
+		if (res == NULL) {
+			dev_err(dev, "no SSP RX DRCMR defined\n");
+			return -ENODEV;
+		}
+		ssp->drcmr_rx = res->start;
+
+		res = platform_get_resource(pdev, IORESOURCE_DMA, 1);
+		if (res == NULL) {
+			dev_err(dev, "no SSP TX DRCMR defined\n");
+			return -ENODEV;
+		}
+		ssp->drcmr_tx = res->start;
+	}
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (res == NULL) {
+		dev_err(dev, "no memory resource defined\n");
+		return -ENODEV;
+	}
+
+	res = devm_request_mem_region(dev, res->start, resource_size(res),
+				      pdev->name);
+	if (res == NULL) {
+		dev_err(dev, "failed to request memory resource\n");
+		return -EBUSY;
+	}
+
+	ssp->phys_base = res->start;
+
+	ssp->mmio_base = devm_ioremap(dev, res->start, resource_size(res));
+	if (ssp->mmio_base == NULL) {
+		dev_err(dev, "failed to ioremap() registers\n");
+		return -ENODEV;
+	}
+
+	ssp->irq = platform_get_irq(pdev, 0);
+	if (ssp->irq < 0) {
+		dev_err(dev, "no IRQ resource defined\n");
+		return -ENODEV;
+	}
+
+	if (dev->of_node) {
+		const struct of_device_id *id =
+			of_match_device(of_match_ptr(pxa_ssp_of_ids), dev);
+		ssp->type = (int) id->data;
+	} else {
+		const struct platform_device_id *id =
+			platform_get_device_id(pdev);
+		ssp->type = (int) id->driver_data;
+
+		/* PXA2xx/3xx SSP ports starts from 1 and the internal pdev->id
+		 * starts from 0, do a translation here
+		 */
+		ssp->port_id = pdev->id + 1;
+	}
+
+	ssp->use_count = 0;
+	ssp->of_node = dev->of_node;
+
+	mutex_lock(&ssp_lock);
+	list_add(&ssp->node, &ssp_list);
+	mutex_unlock(&ssp_lock);
+
+	platform_set_drvdata(pdev, ssp);
+
+	return 0;
+}
+
+static int pxa_ssp_remove(struct platform_device *pdev)
+{
+	struct resource *res;
+	struct ssp_device *ssp;
+
+	ssp = platform_get_drvdata(pdev);
+	if (ssp == NULL)
+		return -ENODEV;
+
+	iounmap(ssp->mmio_base);
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	release_mem_region(res->start, resource_size(res));
+
+	clk_put(ssp->clk);
+
+	mutex_lock(&ssp_lock);
+	list_del(&ssp->node);
+	mutex_unlock(&ssp_lock);
+
+	kfree(ssp);
+	return 0;
+}
+
+static const struct platform_device_id ssp_id_table[] = {
+	{ "pxa25x-ssp",		PXA25x_SSP },
+	{ "pxa25x-nssp",	PXA25x_NSSP },
+	{ "pxa27x-ssp",		PXA27x_SSP },
+	{ "pxa3xx-ssp",		PXA3xx_SSP },
+	{ "pxa168-ssp",		PXA168_SSP },
+	{ "pxa910-ssp",		PXA910_SSP },
+	{ },
+};
+
+static struct platform_driver pxa_ssp_driver = {
+	.probe		= pxa_ssp_probe,
+	.remove		= pxa_ssp_remove,
+	.driver		= {
+		.name		= "pxa2xx-ssp",
+		.of_match_table	= of_match_ptr(pxa_ssp_of_ids),
+	},
+	.id_table	= ssp_id_table,
+};
+
+static int __init pxa_ssp_init(void)
+{
+	return platform_driver_register(&pxa_ssp_driver);
+}
+
+static void __exit pxa_ssp_exit(void)
+{
+	platform_driver_unregister(&pxa_ssp_driver);
+}
+
+arch_initcall(pxa_ssp_init);
+module_exit(pxa_ssp_exit);
+
+MODULE_DESCRIPTION("PXA SSP driver");
+MODULE_AUTHOR("Liam Girdwood");
+MODULE_LICENSE("GPL");

arch/arm/plat-versatile/Kconfig

@@ -0,0 +1,6 @@
+if PLAT_VERSATILE
+
+config PLAT_VERSATILE_SCHED_CLOCK
+	bool
+
+endif

arch/arm/plat-versatile/Makefile

@@ -0,0 +1,4 @@
+ccflags-$(CONFIG_ARCH_MULTIPLATFORM) := -I$(srctree)/$(src)/include
+
+obj-$(CONFIG_PLAT_VERSATILE_SCHED_CLOCK) += sched-clock.o
+obj-$(CONFIG_SMP) += headsmp.o platsmp.o

arch/arm/plat-versatile/headsmp.S

@@ -0,0 +1,41 @@
+/*
+ *  linux/arch/arm/plat-versatile/headsmp.S
+ *
+ *  Copyright (c) 2003 ARM Limited
+ *  All Rights Reserved
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <linux/init.h>
+#include <asm/assembler.h>
+
+/*
+ * Realview/Versatile Express specific entry point for secondary CPUs.
+ * This provides a "holding pen" into which all secondary cores are held
+ * until we're ready for them to initialise.
+ */
+ENTRY(versatile_secondary_startup)
+ ARM_BE8(setend	be)
+	mrc	p15, 0, r0, c0, c0, 5
+	bic	r0, #0xff000000
+	adr	r4, 1f
+	ldmia	r4, {r5, r6}
+	sub	r4, r4, r5
+	add	r6, r6, r4
+pen:	ldr	r7, [r6]
+	cmp	r7, r0
+	bne	pen
+
+	/*
+	 * we've been released from the holding pen: secondary_stack
+	 * should now contain the SVC stack for this core
+	 */
+	b	secondary_startup
+
+	.align
+1:	.long	.
+	.long	pen_release
+ENDPROC(versatile_secondary_startup)

arch/arm/plat-versatile/include/plat/clock.h

@@ -0,0 +1,15 @@
+#ifndef PLAT_CLOCK_H
+#define PLAT_CLOCK_H
+
+#include <asm/hardware/icst.h>
+
+struct clk_ops {
+	long	(*round)(struct clk *, unsigned long);
+	int	(*set)(struct clk *, unsigned long);
+	void	(*setvco)(struct clk *, struct icst_vco);
+};
+
+int icst_clk_set(struct clk *, unsigned long);
+long icst_clk_round(struct clk *, unsigned long);
+
+#endif

arch/arm/plat-versatile/include/plat/platsmp.h

@@ -0,0 +1,14 @@
+/*
+ *  linux/arch/arm/plat-versatile/include/plat/platsmp.h
+ *
+ *  Copyright (C) 2011 ARM Ltd.
+ *  All Rights Reserved
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+extern void versatile_secondary_startup(void);
+extern void versatile_secondary_init(unsigned int cpu);
+extern int  versatile_boot_secondary(unsigned int cpu, struct task_struct *idle);

arch/arm/plat-versatile/include/plat/sched_clock.h

@@ -0,0 +1,6 @@
+#ifndef ARM_PLAT_SCHED_CLOCK_H
+#define ARM_PLAT_SCHED_CLOCK_H
+
+void versatile_sched_clock_init(void __iomem *, unsigned long);
+
+#endif

arch/arm/plat-versatile/platsmp.c

@@ -0,0 +1,93 @@
+/*
+ *  linux/arch/arm/plat-versatile/platsmp.c
+ *
+ *  Copyright (C) 2002 ARM Ltd.
+ *  All Rights Reserved
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <linux/jiffies.h>
+#include <linux/smp.h>
+
+#include <asm/cacheflush.h>
+#include <asm/smp_plat.h>
+
+#include <plat/platsmp.h>
+
+/*
+ * Write pen_release in a way that is guaranteed to be visible to all
+ * observers, irrespective of whether they're taking part in coherency
+ * or not.  This is necessary for the hotplug code to work reliably.
+ */
+static void write_pen_release(int val)
+{
+	pen_release = val;
+	smp_wmb();
+	sync_cache_w(&pen_release);
+}
+
+static DEFINE_SPINLOCK(boot_lock);
+
+void versatile_secondary_init(unsigned int cpu)
+{
+	/*
+	 * let the primary processor know we're out of the
+	 * pen, then head off into the C entry point
+	 */
+	write_pen_release(-1);
+
+	/*
+	 * Synchronise with the boot thread.
+	 */
+	spin_lock(&boot_lock);
+	spin_unlock(&boot_lock);
+}
+
+int versatile_boot_secondary(unsigned int cpu, struct task_struct *idle)
+{
+	unsigned long timeout;
+
+	/*
+	 * Set synchronisation state between this boot processor
+	 * and the secondary one
+	 */
+	spin_lock(&boot_lock);
+
+	/*
+	 * This is really belt and braces; we hold unintended secondary
+	 * CPUs in the holding pen until we're ready for them.  However,
+	 * since we haven't sent them a soft interrupt, they shouldn't
+	 * be there.
+	 */
+	write_pen_release(cpu_logical_map(cpu));
+
+	/*
+	 * Send the secondary CPU a soft interrupt, thereby causing
+	 * the boot monitor to read the system wide flags register,
+	 * and branch to the address found there.
+	 */
+	arch_send_wakeup_ipi_mask(cpumask_of(cpu));
+
+	timeout = jiffies + (1 * HZ);
+	while (time_before(jiffies, timeout)) {
+		smp_rmb();
+		if (pen_release == -1)
+			break;
+
+		udelay(10);
+	}
+
+	/*
+	 * now the secondary core is starting up let it run its
+	 * calibrations, then wait for it to finish
+	 */
+	spin_unlock(&boot_lock);
+
+	return pen_release != -1 ? -ENOSYS : 0;
+}

arch/arm/plat-versatile/sched-clock.c

@@ -0,0 +1,41 @@
+/*
+ *  linux/arch/arm/plat-versatile/sched-clock.c
+ *
+ *  Copyright (C) 1999 - 2003 ARM Limited
+ *  Copyright (C) 2000 Deep Blue Solutions Ltd
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+#include <linux/kernel.h>
+#include <linux/io.h>
+#include <linux/sched_clock.h>
+
+#include <plat/sched_clock.h>
+
+static void __iomem *ctr;
+
+static u64 notrace versatile_read_sched_clock(void)
+{
+	if (ctr)
+		return readl(ctr);
+
+	return 0;
+}
+
+void __init versatile_sched_clock_init(void __iomem *reg, unsigned long rate)
+{
+	ctr = reg;
+	sched_clock_register(versatile_read_sched_clock, 32, rate);
+}

arch/arm/probes/Makefile

@@ -0,0 +1,7 @@
+obj-$(CONFIG_UPROBES)		+= decode.o decode-arm.o uprobes/
+obj-$(CONFIG_KPROBES)		+= decode.o kprobes/
+ifdef CONFIG_THUMB2_KERNEL
+obj-$(CONFIG_KPROBES)		+= decode-thumb.o
+else
+obj-$(CONFIG_KPROBES)		+= decode-arm.o
+endif

arch/arm/probes/decode-arm.c

@@ -0,0 +1,736 @@
+/*
+ *
+ * arch/arm/probes/decode-arm.c
+ *
+ * Some code moved here from arch/arm/kernel/kprobes-arm.c
+ *
+ * Copyright (C) 2006, 2007 Motorola Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/stddef.h>
+#include <linux/ptrace.h>
+
+#include "decode.h"
+#include "decode-arm.h"
+
+#define sign_extend(x, signbit) ((x) | (0 - ((x) & (1 << (signbit)))))
+
+#define branch_displacement(insn) sign_extend(((insn) & 0xffffff) << 2, 25)
+
+/*
+ * To avoid the complications of mimicing single-stepping on a
+ * processor without a Next-PC or a single-step mode, and to
+ * avoid having to deal with the side-effects of boosting, we
+ * simulate or emulate (almost) all ARM instructions.
+ *
+ * "Simulation" is where the instruction's behavior is duplicated in
+ * C code.  "Emulation" is where the original instruction is rewritten
+ * and executed, often by altering its registers.
+ *
+ * By having all behavior of the kprobe'd instruction completed before
+ * returning from the kprobe_handler(), all locks (scheduler and
+ * interrupt) can safely be released.  There is no need for secondary
+ * breakpoints, no race with MP or preemptable kernels, nor having to
+ * clean up resources counts at a later time impacting overall system
+ * performance.  By rewriting the instruction, only the minimum registers
+ * need to be loaded and saved back optimizing performance.
+ *
+ * Calling the insnslot_*_rwflags version of a function doesn't hurt
+ * anything even when the CPSR flags aren't updated by the
+ * instruction.  It's just a little slower in return for saving
+ * a little space by not having a duplicate function that doesn't
+ * update the flags.  (The same optimization can be said for
+ * instructions that do or don't perform register writeback)
+ * Also, instructions can either read the flags, only write the
+ * flags, or read and write the flags.  To save combinations
+ * rather than for sheer performance, flag functions just assume
+ * read and write of flags.
+ */
+
+void __kprobes simulate_bbl(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	long iaddr = (long) regs->ARM_pc - 4;
+	int disp  = branch_displacement(insn);
+
+	if (insn & (1 << 24))
+		regs->ARM_lr = iaddr + 4;
+
+	regs->ARM_pc = iaddr + 8 + disp;
+}
+
+void __kprobes simulate_blx1(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	long iaddr = (long) regs->ARM_pc - 4;
+	int disp = branch_displacement(insn);
+
+	regs->ARM_lr = iaddr + 4;
+	regs->ARM_pc = iaddr + 8 + disp + ((insn >> 23) & 0x2);
+	regs->ARM_cpsr |= PSR_T_BIT;
+}
+
+void __kprobes simulate_blx2bx(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	int rm = insn & 0xf;
+	long rmv = regs->uregs[rm];
+
+	if (insn & (1 << 5))
+		regs->ARM_lr = (long) regs->ARM_pc;
+
+	regs->ARM_pc = rmv & ~0x1;
+	regs->ARM_cpsr &= ~PSR_T_BIT;
+	if (rmv & 0x1)
+		regs->ARM_cpsr |= PSR_T_BIT;
+}
+
+void __kprobes simulate_mrs(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	int rd = (insn >> 12) & 0xf;
+	unsigned long mask = 0xf8ff03df; /* Mask out execution state */
+	regs->uregs[rd] = regs->ARM_cpsr & mask;
+}
+
+void __kprobes simulate_mov_ipsp(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	regs->uregs[12] = regs->uregs[13];
+}
+
+/*
+ * For the instruction masking and comparisons in all the "space_*"
+ * functions below, Do _not_ rearrange the order of tests unless
+ * you're very, very sure of what you are doing.  For the sake of
+ * efficiency, the masks for some tests sometimes assume other test
+ * have been done prior to them so the number of patterns to test
+ * for an instruction set can be as broad as possible to reduce the
+ * number of tests needed.
+ */
+
+static const union decode_item arm_1111_table[] = {
+	/* Unconditional instructions					*/
+
+	/* memory hint		1111 0100 x001 xxxx xxxx xxxx xxxx xxxx */
+	/* PLDI (immediate)	1111 0100 x101 xxxx xxxx xxxx xxxx xxxx */
+	/* PLDW (immediate)	1111 0101 x001 xxxx xxxx xxxx xxxx xxxx */
+	/* PLD (immediate)	1111 0101 x101 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_SIMULATE	(0xfe300000, 0xf4100000, PROBES_PRELOAD_IMM),
+
+	/* memory hint		1111 0110 x001 xxxx xxxx xxxx xxx0 xxxx */
+	/* PLDI (register)	1111 0110 x101 xxxx xxxx xxxx xxx0 xxxx */
+	/* PLDW (register)	1111 0111 x001 xxxx xxxx xxxx xxx0 xxxx */
+	/* PLD (register)	1111 0111 x101 xxxx xxxx xxxx xxx0 xxxx */
+	DECODE_SIMULATE	(0xfe300010, 0xf6100000, PROBES_PRELOAD_REG),
+
+	/* BLX (immediate)	1111 101x xxxx xxxx xxxx xxxx xxxx xxxx */
+	DECODE_SIMULATE	(0xfe000000, 0xfa000000, PROBES_BRANCH_IMM),
+
+	/* CPS			1111 0001 0000 xxx0 xxxx xxxx xx0x xxxx */
+	/* SETEND		1111 0001 0000 0001 xxxx xxxx 0000 xxxx */
+	/* SRS			1111 100x x1x0 xxxx xxxx xxxx xxxx xxxx */
+	/* RFE			1111 100x x0x1 xxxx xxxx xxxx xxxx xxxx */
+
+	/* Coprocessor instructions... */
+	/* MCRR2		1111 1100 0100 xxxx xxxx xxxx xxxx xxxx */
+	/* MRRC2		1111 1100 0101 xxxx xxxx xxxx xxxx xxxx */
+	/* LDC2			1111 110x xxx1 xxxx xxxx xxxx xxxx xxxx */
+	/* STC2			1111 110x xxx0 xxxx xxxx xxxx xxxx xxxx */
+	/* CDP2			1111 1110 xxxx xxxx xxxx xxxx xxx0 xxxx */
+	/* MCR2			1111 1110 xxx0 xxxx xxxx xxxx xxx1 xxxx */
+	/* MRC2			1111 1110 xxx1 xxxx xxxx xxxx xxx1 xxxx */
+
+	/* Other unallocated instructions...				*/
+	DECODE_END
+};
+
+static const union decode_item arm_cccc_0001_0xx0____0xxx_table[] = {
+	/* Miscellaneous instructions					*/
+
+	/* MRS cpsr		cccc 0001 0000 xxxx xxxx xxxx 0000 xxxx */
+	DECODE_SIMULATEX(0x0ff000f0, 0x01000000, PROBES_MRS,
+						 REGS(0, NOPC, 0, 0, 0)),
+
+	/* BX			cccc 0001 0010 xxxx xxxx xxxx 0001 xxxx */
+	DECODE_SIMULATE	(0x0ff000f0, 0x01200010, PROBES_BRANCH_REG),
+
+	/* BLX (register)	cccc 0001 0010 xxxx xxxx xxxx 0011 xxxx */
+	DECODE_SIMULATEX(0x0ff000f0, 0x01200030, PROBES_BRANCH_REG,
+						 REGS(0, 0, 0, 0, NOPC)),
+
+	/* CLZ			cccc 0001 0110 xxxx xxxx xxxx 0001 xxxx */
+	DECODE_EMULATEX	(0x0ff000f0, 0x01600010, PROBES_CLZ,
+						 REGS(0, NOPC, 0, 0, NOPC)),
+
+	/* QADD			cccc 0001 0000 xxxx xxxx xxxx 0101 xxxx */
+	/* QSUB			cccc 0001 0010 xxxx xxxx xxxx 0101 xxxx */
+	/* QDADD		cccc 0001 0100 xxxx xxxx xxxx 0101 xxxx */
+	/* QDSUB		cccc 0001 0110 xxxx xxxx xxxx 0101 xxxx */
+	DECODE_EMULATEX	(0x0f9000f0, 0x01000050, PROBES_SATURATING_ARITHMETIC,
+						 REGS(NOPC, NOPC, 0, 0, NOPC)),
+
+	/* BXJ			cccc 0001 0010 xxxx xxxx xxxx 0010 xxxx */
+	/* MSR			cccc 0001 0x10 xxxx xxxx xxxx 0000 xxxx */
+	/* MRS spsr		cccc 0001 0100 xxxx xxxx xxxx 0000 xxxx */
+	/* BKPT			1110 0001 0010 xxxx xxxx xxxx 0111 xxxx */
+	/* SMC			cccc 0001 0110 xxxx xxxx xxxx 0111 xxxx */
+	/* And unallocated instructions...				*/
+	DECODE_END
+};
+
+static const union decode_item arm_cccc_0001_0xx0____1xx0_table[] = {
+	/* Halfword multiply and multiply-accumulate			*/
+
+	/* SMLALxy		cccc 0001 0100 xxxx xxxx xxxx 1xx0 xxxx */
+	DECODE_EMULATEX	(0x0ff00090, 0x01400080, PROBES_MUL1,
+						 REGS(NOPC, NOPC, NOPC, 0, NOPC)),
+
+	/* SMULWy		cccc 0001 0010 xxxx xxxx xxxx 1x10 xxxx */
+	DECODE_OR	(0x0ff000b0, 0x012000a0),
+	/* SMULxy		cccc 0001 0110 xxxx xxxx xxxx 1xx0 xxxx */
+	DECODE_EMULATEX	(0x0ff00090, 0x01600080, PROBES_MUL2,
+						 REGS(NOPC, 0, NOPC, 0, NOPC)),
+
+	/* SMLAxy		cccc 0001 0000 xxxx xxxx xxxx 1xx0 xxxx */
+	DECODE_OR	(0x0ff00090, 0x01000080),
+	/* SMLAWy		cccc 0001 0010 xxxx xxxx xxxx 1x00 xxxx */
+	DECODE_EMULATEX	(0x0ff000b0, 0x01200080, PROBES_MUL2,
+						 REGS(NOPC, NOPC, NOPC, 0, NOPC)),
+
+	DECODE_END
+};
+
+static const union decode_item arm_cccc_0000_____1001_table[] = {
+	/* Multiply and multiply-accumulate				*/
+
+	/* MUL			cccc 0000 0000 xxxx xxxx xxxx 1001 xxxx */
+	/* MULS			cccc 0000 0001 xxxx xxxx xxxx 1001 xxxx */
+	DECODE_EMULATEX	(0x0fe000f0, 0x00000090, PROBES_MUL2,
+						 REGS(NOPC, 0, NOPC, 0, NOPC)),
+
+	/* MLA			cccc 0000 0010 xxxx xxxx xxxx 1001 xxxx */
+	/* MLAS			cccc 0000 0011 xxxx xxxx xxxx 1001 xxxx */
+	DECODE_OR	(0x0fe000f0, 0x00200090),
+	/* MLS			cccc 0000 0110 xxxx xxxx xxxx 1001 xxxx */
+	DECODE_EMULATEX	(0x0ff000f0, 0x00600090, PROBES_MUL2,
+						 REGS(NOPC, NOPC, NOPC, 0, NOPC)),
+
+	/* UMAAL		cccc 0000 0100 xxxx xxxx xxxx 1001 xxxx */
+	DECODE_OR	(0x0ff000f0, 0x00400090),
+	/* UMULL		cccc 0000 1000 xxxx xxxx xxxx 1001 xxxx */
+	/* UMULLS		cccc 0000 1001 xxxx xxxx xxxx 1001 xxxx */
+	/* UMLAL		cccc 0000 1010 xxxx xxxx xxxx 1001 xxxx */
+	/* UMLALS		cccc 0000 1011 xxxx xxxx xxxx 1001 xxxx */
+	/* SMULL		cccc 0000 1100 xxxx xxxx xxxx 1001 xxxx */
+	/* SMULLS		cccc 0000 1101 xxxx xxxx xxxx 1001 xxxx */
+	/* SMLAL		cccc 0000 1110 xxxx xxxx xxxx 1001 xxxx */
+	/* SMLALS		cccc 0000 1111 xxxx xxxx xxxx 1001 xxxx */
+	DECODE_EMULATEX	(0x0f8000f0, 0x00800090, PROBES_MUL1,
+						 REGS(NOPC, NOPC, NOPC, 0, NOPC)),
+
+	DECODE_END
+};
+
+static const union decode_item arm_cccc_0001_____1001_table[] = {
+	/* Synchronization primitives					*/
+
+#if __LINUX_ARM_ARCH__ < 6
+	/* Deprecated on ARMv6 and may be UNDEFINED on v7		*/
+	/* SMP/SWPB		cccc 0001 0x00 xxxx xxxx xxxx 1001 xxxx */
+	DECODE_EMULATEX	(0x0fb000f0, 0x01000090, PROBES_SWP,
+						 REGS(NOPC, NOPC, 0, 0, NOPC)),
+#endif
+	/* LDREX/STREX{,D,B,H}	cccc 0001 1xxx xxxx xxxx xxxx 1001 xxxx */
+	/* And unallocated instructions...				*/
+	DECODE_END
+};
+
+static const union decode_item arm_cccc_000x_____1xx1_table[] = {
+	/* Extra load/store instructions				*/
+
+	/* STRHT		cccc 0000 xx10 xxxx xxxx xxxx 1011 xxxx */
+	/* ???			cccc 0000 xx10 xxxx xxxx xxxx 11x1 xxxx */
+	/* LDRHT		cccc 0000 xx11 xxxx xxxx xxxx 1011 xxxx */
+	/* LDRSBT		cccc 0000 xx11 xxxx xxxx xxxx 1101 xxxx */
+	/* LDRSHT		cccc 0000 xx11 xxxx xxxx xxxx 1111 xxxx */
+	DECODE_REJECT	(0x0f200090, 0x00200090),
+
+	/* LDRD/STRD lr,pc,{...	cccc 000x x0x0 xxxx 111x xxxx 1101 xxxx */
+	DECODE_REJECT	(0x0e10e0d0, 0x0000e0d0),
+
+	/* LDRD (register)	cccc 000x x0x0 xxxx xxxx xxxx 1101 xxxx */
+	/* STRD (register)	cccc 000x x0x0 xxxx xxxx xxxx 1111 xxxx */
+	DECODE_EMULATEX	(0x0e5000d0, 0x000000d0, PROBES_LDRSTRD,
+						 REGS(NOPCWB, NOPCX, 0, 0, NOPC)),
+
+	/* LDRD (immediate)	cccc 000x x1x0 xxxx xxxx xxxx 1101 xxxx */
+	/* STRD (immediate)	cccc 000x x1x0 xxxx xxxx xxxx 1111 xxxx */
+	DECODE_EMULATEX	(0x0e5000d0, 0x004000d0, PROBES_LDRSTRD,
+						 REGS(NOPCWB, NOPCX, 0, 0, 0)),
+
+	/* STRH (register)	cccc 000x x0x0 xxxx xxxx xxxx 1011 xxxx */
+	DECODE_EMULATEX	(0x0e5000f0, 0x000000b0, PROBES_STORE_EXTRA,
+						 REGS(NOPCWB, NOPC, 0, 0, NOPC)),
+
+	/* LDRH (register)	cccc 000x x0x1 xxxx xxxx xxxx 1011 xxxx */
+	/* LDRSB (register)	cccc 000x x0x1 xxxx xxxx xxxx 1101 xxxx */
+	/* LDRSH (register)	cccc 000x x0x1 xxxx xxxx xxxx 1111 xxxx */
+	DECODE_EMULATEX	(0x0e500090, 0x00100090, PROBES_LOAD_EXTRA,
+						 REGS(NOPCWB, NOPC, 0, 0, NOPC)),
+
+	/* STRH (immediate)	cccc 000x x1x0 xxxx xxxx xxxx 1011 xxxx */
+	DECODE_EMULATEX	(0x0e5000f0, 0x004000b0, PROBES_STORE_EXTRA,
+						 REGS(NOPCWB, NOPC, 0, 0, 0)),
+
+	/* LDRH (immediate)	cccc 000x x1x1 xxxx xxxx xxxx 1011 xxxx */
+	/* LDRSB (immediate)	cccc 000x x1x1 xxxx xxxx xxxx 1101 xxxx */
+	/* LDRSH (immediate)	cccc 000x x1x1 xxxx xxxx xxxx 1111 xxxx */
+	DECODE_EMULATEX	(0x0e500090, 0x00500090, PROBES_LOAD_EXTRA,
+						 REGS(NOPCWB, NOPC, 0, 0, 0)),
+
+	DECODE_END
+};
+
+static const union decode_item arm_cccc_000x_table[] = {
+	/* Data-processing (register)					*/
+
+	/* <op>S PC, ...	cccc 000x xxx1 xxxx 1111 xxxx xxxx xxxx */
+	DECODE_REJECT	(0x0e10f000, 0x0010f000),
+
+	/* MOV IP, SP		1110 0001 1010 0000 1100 0000 0000 1101 */
+	DECODE_SIMULATE	(0xffffffff, 0xe1a0c00d, PROBES_MOV_IP_SP),
+
+	/* TST (register)	cccc 0001 0001 xxxx xxxx xxxx xxx0 xxxx */
+	/* TEQ (register)	cccc 0001 0011 xxxx xxxx xxxx xxx0 xxxx */
+	/* CMP (register)	cccc 0001 0101 xxxx xxxx xxxx xxx0 xxxx */
+	/* CMN (register)	cccc 0001 0111 xxxx xxxx xxxx xxx0 xxxx */
+	DECODE_EMULATEX	(0x0f900010, 0x01100000, PROBES_DATA_PROCESSING_REG,
+						 REGS(ANY, 0, 0, 0, ANY)),
+
+	/* MOV (register)	cccc 0001 101x xxxx xxxx xxxx xxx0 xxxx */
+	/* MVN (register)	cccc 0001 111x xxxx xxxx xxxx xxx0 xxxx */
+	DECODE_EMULATEX	(0x0fa00010, 0x01a00000, PROBES_DATA_PROCESSING_REG,
+						 REGS(0, ANY, 0, 0, ANY)),
+
+	/* AND (register)	cccc 0000 000x xxxx xxxx xxxx xxx0 xxxx */
+	/* EOR (register)	cccc 0000 001x xxxx xxxx xxxx xxx0 xxxx */
+	/* SUB (register)	cccc 0000 010x xxxx xxxx xxxx xxx0 xxxx */
+	/* RSB (register)	cccc 0000 011x xxxx xxxx xxxx xxx0 xxxx */
+	/* ADD (register)	cccc 0000 100x xxxx xxxx xxxx xxx0 xxxx */
+	/* ADC (register)	cccc 0000 101x xxxx xxxx xxxx xxx0 xxxx */
+	/* SBC (register)	cccc 0000 110x xxxx xxxx xxxx xxx0 xxxx */
+	/* RSC (register)	cccc 0000 111x xxxx xxxx xxxx xxx0 xxxx */
+	/* ORR (register)	cccc 0001 100x xxxx xxxx xxxx xxx0 xxxx */
+	/* BIC (register)	cccc 0001 110x xxxx xxxx xxxx xxx0 xxxx */
+	DECODE_EMULATEX	(0x0e000010, 0x00000000, PROBES_DATA_PROCESSING_REG,
+						 REGS(ANY, ANY, 0, 0, ANY)),
+
+	/* TST (reg-shift reg)	cccc 0001 0001 xxxx xxxx xxxx 0xx1 xxxx */
+	/* TEQ (reg-shift reg)	cccc 0001 0011 xxxx xxxx xxxx 0xx1 xxxx */
+	/* CMP (reg-shift reg)	cccc 0001 0101 xxxx xxxx xxxx 0xx1 xxxx */
+	/* CMN (reg-shift reg)	cccc 0001 0111 xxxx xxxx xxxx 0xx1 xxxx */
+	DECODE_EMULATEX	(0x0f900090, 0x01100010, PROBES_DATA_PROCESSING_REG,
+						 REGS(NOPC, 0, NOPC, 0, NOPC)),
+
+	/* MOV (reg-shift reg)	cccc 0001 101x xxxx xxxx xxxx 0xx1 xxxx */
+	/* MVN (reg-shift reg)	cccc 0001 111x xxxx xxxx xxxx 0xx1 xxxx */
+	DECODE_EMULATEX	(0x0fa00090, 0x01a00010, PROBES_DATA_PROCESSING_REG,
+						 REGS(0, NOPC, NOPC, 0, NOPC)),
+
+	/* AND (reg-shift reg)	cccc 0000 000x xxxx xxxx xxxx 0xx1 xxxx */
+	/* EOR (reg-shift reg)	cccc 0000 001x xxxx xxxx xxxx 0xx1 xxxx */
+	/* SUB (reg-shift reg)	cccc 0000 010x xxxx xxxx xxxx 0xx1 xxxx */
+	/* RSB (reg-shift reg)	cccc 0000 011x xxxx xxxx xxxx 0xx1 xxxx */
+	/* ADD (reg-shift reg)	cccc 0000 100x xxxx xxxx xxxx 0xx1 xxxx */
+	/* ADC (reg-shift reg)	cccc 0000 101x xxxx xxxx xxxx 0xx1 xxxx */
+	/* SBC (reg-shift reg)	cccc 0000 110x xxxx xxxx xxxx 0xx1 xxxx */
+	/* RSC (reg-shift reg)	cccc 0000 111x xxxx xxxx xxxx 0xx1 xxxx */
+	/* ORR (reg-shift reg)	cccc 0001 100x xxxx xxxx xxxx 0xx1 xxxx */
+	/* BIC (reg-shift reg)	cccc 0001 110x xxxx xxxx xxxx 0xx1 xxxx */
+	DECODE_EMULATEX	(0x0e000090, 0x00000010, PROBES_DATA_PROCESSING_REG,
+						 REGS(NOPC, NOPC, NOPC, 0, NOPC)),
+
+	DECODE_END
+};
+
+static const union decode_item arm_cccc_001x_table[] = {
+	/* Data-processing (immediate)					*/
+
+	/* MOVW			cccc 0011 0000 xxxx xxxx xxxx xxxx xxxx */
+	/* MOVT			cccc 0011 0100 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0x0fb00000, 0x03000000, PROBES_MOV_HALFWORD,
+						 REGS(0, NOPC, 0, 0, 0)),
+
+	/* YIELD		cccc 0011 0010 0000 xxxx xxxx 0000 0001 */
+	DECODE_OR	(0x0fff00ff, 0x03200001),
+	/* SEV			cccc 0011 0010 0000 xxxx xxxx 0000 0100 */
+	DECODE_EMULATE	(0x0fff00ff, 0x03200004, PROBES_SEV),
+	/* NOP			cccc 0011 0010 0000 xxxx xxxx 0000 0000 */
+	/* WFE			cccc 0011 0010 0000 xxxx xxxx 0000 0010 */
+	/* WFI			cccc 0011 0010 0000 xxxx xxxx 0000 0011 */
+	DECODE_SIMULATE	(0x0fff00fc, 0x03200000, PROBES_WFE),
+	/* DBG			cccc 0011 0010 0000 xxxx xxxx ffff xxxx */
+	/* unallocated hints	cccc 0011 0010 0000 xxxx xxxx xxxx xxxx */
+	/* MSR (immediate)	cccc 0011 0x10 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0x0fb00000, 0x03200000),
+
+	/* <op>S PC, ...	cccc 001x xxx1 xxxx 1111 xxxx xxxx xxxx */
+	DECODE_REJECT	(0x0e10f000, 0x0210f000),
+
+	/* TST (immediate)	cccc 0011 0001 xxxx xxxx xxxx xxxx xxxx */
+	/* TEQ (immediate)	cccc 0011 0011 xxxx xxxx xxxx xxxx xxxx */
+	/* CMP (immediate)	cccc 0011 0101 xxxx xxxx xxxx xxxx xxxx */
+	/* CMN (immediate)	cccc 0011 0111 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0x0f900000, 0x03100000, PROBES_DATA_PROCESSING_IMM,
+						 REGS(ANY, 0, 0, 0, 0)),
+
+	/* MOV (immediate)	cccc 0011 101x xxxx xxxx xxxx xxxx xxxx */
+	/* MVN (immediate)	cccc 0011 111x xxxx xxxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0x0fa00000, 0x03a00000, PROBES_DATA_PROCESSING_IMM,
+						 REGS(0, ANY, 0, 0, 0)),
+
+	/* AND (immediate)	cccc 0010 000x xxxx xxxx xxxx xxxx xxxx */
+	/* EOR (immediate)	cccc 0010 001x xxxx xxxx xxxx xxxx xxxx */
+	/* SUB (immediate)	cccc 0010 010x xxxx xxxx xxxx xxxx xxxx */
+	/* RSB (immediate)	cccc 0010 011x xxxx xxxx xxxx xxxx xxxx */
+	/* ADD (immediate)	cccc 0010 100x xxxx xxxx xxxx xxxx xxxx */
+	/* ADC (immediate)	cccc 0010 101x xxxx xxxx xxxx xxxx xxxx */
+	/* SBC (immediate)	cccc 0010 110x xxxx xxxx xxxx xxxx xxxx */
+	/* RSC (immediate)	cccc 0010 111x xxxx xxxx xxxx xxxx xxxx */
+	/* ORR (immediate)	cccc 0011 100x xxxx xxxx xxxx xxxx xxxx */
+	/* BIC (immediate)	cccc 0011 110x xxxx xxxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0x0e000000, 0x02000000, PROBES_DATA_PROCESSING_IMM,
+						 REGS(ANY, ANY, 0, 0, 0)),
+
+	DECODE_END
+};
+
+static const union decode_item arm_cccc_0110_____xxx1_table[] = {
+	/* Media instructions						*/
+
+	/* SEL			cccc 0110 1000 xxxx xxxx xxxx 1011 xxxx */
+	DECODE_EMULATEX	(0x0ff000f0, 0x068000b0, PROBES_SATURATE,
+						 REGS(NOPC, NOPC, 0, 0, NOPC)),
+
+	/* SSAT			cccc 0110 101x xxxx xxxx xxxx xx01 xxxx */
+	/* USAT			cccc 0110 111x xxxx xxxx xxxx xx01 xxxx */
+	DECODE_OR(0x0fa00030, 0x06a00010),
+	/* SSAT16		cccc 0110 1010 xxxx xxxx xxxx 0011 xxxx */
+	/* USAT16		cccc 0110 1110 xxxx xxxx xxxx 0011 xxxx */
+	DECODE_EMULATEX	(0x0fb000f0, 0x06a00030, PROBES_SATURATE,
+						 REGS(0, NOPC, 0, 0, NOPC)),
+
+	/* REV			cccc 0110 1011 xxxx xxxx xxxx 0011 xxxx */
+	/* REV16		cccc 0110 1011 xxxx xxxx xxxx 1011 xxxx */
+	/* RBIT			cccc 0110 1111 xxxx xxxx xxxx 0011 xxxx */
+	/* REVSH		cccc 0110 1111 xxxx xxxx xxxx 1011 xxxx */
+	DECODE_EMULATEX	(0x0fb00070, 0x06b00030, PROBES_REV,
+						 REGS(0, NOPC, 0, 0, NOPC)),
+
+	/* ???			cccc 0110 0x00 xxxx xxxx xxxx xxx1 xxxx */
+	DECODE_REJECT	(0x0fb00010, 0x06000010),
+	/* ???			cccc 0110 0xxx xxxx xxxx xxxx 1011 xxxx */
+	DECODE_REJECT	(0x0f8000f0, 0x060000b0),
+	/* ???			cccc 0110 0xxx xxxx xxxx xxxx 1101 xxxx */
+	DECODE_REJECT	(0x0f8000f0, 0x060000d0),
+	/* SADD16		cccc 0110 0001 xxxx xxxx xxxx 0001 xxxx */
+	/* SADDSUBX		cccc 0110 0001 xxxx xxxx xxxx 0011 xxxx */
+	/* SSUBADDX		cccc 0110 0001 xxxx xxxx xxxx 0101 xxxx */
+	/* SSUB16		cccc 0110 0001 xxxx xxxx xxxx 0111 xxxx */
+	/* SADD8		cccc 0110 0001 xxxx xxxx xxxx 1001 xxxx */
+	/* SSUB8		cccc 0110 0001 xxxx xxxx xxxx 1111 xxxx */
+	/* QADD16		cccc 0110 0010 xxxx xxxx xxxx 0001 xxxx */
+	/* QADDSUBX		cccc 0110 0010 xxxx xxxx xxxx 0011 xxxx */
+	/* QSUBADDX		cccc 0110 0010 xxxx xxxx xxxx 0101 xxxx */
+	/* QSUB16		cccc 0110 0010 xxxx xxxx xxxx 0111 xxxx */
+	/* QADD8		cccc 0110 0010 xxxx xxxx xxxx 1001 xxxx */
+	/* QSUB8		cccc 0110 0010 xxxx xxxx xxxx 1111 xxxx */
+	/* SHADD16		cccc 0110 0011 xxxx xxxx xxxx 0001 xxxx */
+	/* SHADDSUBX		cccc 0110 0011 xxxx xxxx xxxx 0011 xxxx */
+	/* SHSUBADDX		cccc 0110 0011 xxxx xxxx xxxx 0101 xxxx */
+	/* SHSUB16		cccc 0110 0011 xxxx xxxx xxxx 0111 xxxx */
+	/* SHADD8		cccc 0110 0011 xxxx xxxx xxxx 1001 xxxx */
+	/* SHSUB8		cccc 0110 0011 xxxx xxxx xxxx 1111 xxxx */
+	/* UADD16		cccc 0110 0101 xxxx xxxx xxxx 0001 xxxx */
+	/* UADDSUBX		cccc 0110 0101 xxxx xxxx xxxx 0011 xxxx */
+	/* USUBADDX		cccc 0110 0101 xxxx xxxx xxxx 0101 xxxx */
+	/* USUB16		cccc 0110 0101 xxxx xxxx xxxx 0111 xxxx */
+	/* UADD8		cccc 0110 0101 xxxx xxxx xxxx 1001 xxxx */
+	/* USUB8		cccc 0110 0101 xxxx xxxx xxxx 1111 xxxx */
+	/* UQADD16		cccc 0110 0110 xxxx xxxx xxxx 0001 xxxx */
+	/* UQADDSUBX		cccc 0110 0110 xxxx xxxx xxxx 0011 xxxx */
+	/* UQSUBADDX		cccc 0110 0110 xxxx xxxx xxxx 0101 xxxx */
+	/* UQSUB16		cccc 0110 0110 xxxx xxxx xxxx 0111 xxxx */
+	/* UQADD8		cccc 0110 0110 xxxx xxxx xxxx 1001 xxxx */
+	/* UQSUB8		cccc 0110 0110 xxxx xxxx xxxx 1111 xxxx */
+	/* UHADD16		cccc 0110 0111 xxxx xxxx xxxx 0001 xxxx */
+	/* UHADDSUBX		cccc 0110 0111 xxxx xxxx xxxx 0011 xxxx */
+	/* UHSUBADDX		cccc 0110 0111 xxxx xxxx xxxx 0101 xxxx */
+	/* UHSUB16		cccc 0110 0111 xxxx xxxx xxxx 0111 xxxx */
+	/* UHADD8		cccc 0110 0111 xxxx xxxx xxxx 1001 xxxx */
+	/* UHSUB8		cccc 0110 0111 xxxx xxxx xxxx 1111 xxxx */
+	DECODE_EMULATEX	(0x0f800010, 0x06000010, PROBES_MMI,
+						 REGS(NOPC, NOPC, 0, 0, NOPC)),
+
+	/* PKHBT		cccc 0110 1000 xxxx xxxx xxxx x001 xxxx */
+	/* PKHTB		cccc 0110 1000 xxxx xxxx xxxx x101 xxxx */
+	DECODE_EMULATEX	(0x0ff00030, 0x06800010, PROBES_PACK,
+						 REGS(NOPC, NOPC, 0, 0, NOPC)),
+
+	/* ???			cccc 0110 1001 xxxx xxxx xxxx 0111 xxxx */
+	/* ???			cccc 0110 1101 xxxx xxxx xxxx 0111 xxxx */
+	DECODE_REJECT	(0x0fb000f0, 0x06900070),
+
+	/* SXTB16		cccc 0110 1000 1111 xxxx xxxx 0111 xxxx */
+	/* SXTB			cccc 0110 1010 1111 xxxx xxxx 0111 xxxx */
+	/* SXTH			cccc 0110 1011 1111 xxxx xxxx 0111 xxxx */
+	/* UXTB16		cccc 0110 1100 1111 xxxx xxxx 0111 xxxx */
+	/* UXTB			cccc 0110 1110 1111 xxxx xxxx 0111 xxxx */
+	/* UXTH			cccc 0110 1111 1111 xxxx xxxx 0111 xxxx */
+	DECODE_EMULATEX	(0x0f8f00f0, 0x068f0070, PROBES_EXTEND,
+						 REGS(0, NOPC, 0, 0, NOPC)),
+
+	/* SXTAB16		cccc 0110 1000 xxxx xxxx xxxx 0111 xxxx */
+	/* SXTAB		cccc 0110 1010 xxxx xxxx xxxx 0111 xxxx */
+	/* SXTAH		cccc 0110 1011 xxxx xxxx xxxx 0111 xxxx */
+	/* UXTAB16		cccc 0110 1100 xxxx xxxx xxxx 0111 xxxx */
+	/* UXTAB		cccc 0110 1110 xxxx xxxx xxxx 0111 xxxx */
+	/* UXTAH		cccc 0110 1111 xxxx xxxx xxxx 0111 xxxx */
+	DECODE_EMULATEX	(0x0f8000f0, 0x06800070, PROBES_EXTEND_ADD,
+						 REGS(NOPCX, NOPC, 0, 0, NOPC)),
+
+	DECODE_END
+};
+
+static const union decode_item arm_cccc_0111_____xxx1_table[] = {
+	/* Media instructions						*/
+
+	/* UNDEFINED		cccc 0111 1111 xxxx xxxx xxxx 1111 xxxx */
+	DECODE_REJECT	(0x0ff000f0, 0x07f000f0),
+
+	/* SMLALD		cccc 0111 0100 xxxx xxxx xxxx 00x1 xxxx */
+	/* SMLSLD		cccc 0111 0100 xxxx xxxx xxxx 01x1 xxxx */
+	DECODE_EMULATEX	(0x0ff00090, 0x07400010, PROBES_MUL_ADD_LONG,
+						 REGS(NOPC, NOPC, NOPC, 0, NOPC)),
+
+	/* SMUAD		cccc 0111 0000 xxxx 1111 xxxx 00x1 xxxx */
+	/* SMUSD		cccc 0111 0000 xxxx 1111 xxxx 01x1 xxxx */
+	DECODE_OR	(0x0ff0f090, 0x0700f010),
+	/* SMMUL		cccc 0111 0101 xxxx 1111 xxxx 00x1 xxxx */
+	DECODE_OR	(0x0ff0f0d0, 0x0750f010),
+	/* USAD8		cccc 0111 1000 xxxx 1111 xxxx 0001 xxxx */
+	DECODE_EMULATEX	(0x0ff0f0f0, 0x0780f010, PROBES_MUL_ADD,
+						 REGS(NOPC, 0, NOPC, 0, NOPC)),
+
+	/* SMLAD		cccc 0111 0000 xxxx xxxx xxxx 00x1 xxxx */
+	/* SMLSD		cccc 0111 0000 xxxx xxxx xxxx 01x1 xxxx */
+	DECODE_OR	(0x0ff00090, 0x07000010),
+	/* SMMLA		cccc 0111 0101 xxxx xxxx xxxx 00x1 xxxx */
+	DECODE_OR	(0x0ff000d0, 0x07500010),
+	/* USADA8		cccc 0111 1000 xxxx xxxx xxxx 0001 xxxx */
+	DECODE_EMULATEX	(0x0ff000f0, 0x07800010, PROBES_MUL_ADD,
+						 REGS(NOPC, NOPCX, NOPC, 0, NOPC)),
+
+	/* SMMLS		cccc 0111 0101 xxxx xxxx xxxx 11x1 xxxx */
+	DECODE_EMULATEX	(0x0ff000d0, 0x075000d0, PROBES_MUL_ADD,
+						 REGS(NOPC, NOPC, NOPC, 0, NOPC)),
+
+	/* SBFX			cccc 0111 101x xxxx xxxx xxxx x101 xxxx */
+	/* UBFX			cccc 0111 111x xxxx xxxx xxxx x101 xxxx */
+	DECODE_EMULATEX	(0x0fa00070, 0x07a00050, PROBES_BITFIELD,
+						 REGS(0, NOPC, 0, 0, NOPC)),
+
+	/* BFC			cccc 0111 110x xxxx xxxx xxxx x001 1111 */
+	DECODE_EMULATEX	(0x0fe0007f, 0x07c0001f, PROBES_BITFIELD,
+						 REGS(0, NOPC, 0, 0, 0)),
+
+	/* BFI			cccc 0111 110x xxxx xxxx xxxx x001 xxxx */
+	DECODE_EMULATEX	(0x0fe00070, 0x07c00010, PROBES_BITFIELD,
+						 REGS(0, NOPC, 0, 0, NOPCX)),
+
+	DECODE_END
+};
+
+static const union decode_item arm_cccc_01xx_table[] = {
+	/* Load/store word and unsigned byte				*/
+
+	/* LDRB/STRB pc,[...]	cccc 01xx x0xx xxxx xxxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0x0c40f000, 0x0440f000),
+
+	/* STRT			cccc 01x0 x010 xxxx xxxx xxxx xxxx xxxx */
+	/* LDRT			cccc 01x0 x011 xxxx xxxx xxxx xxxx xxxx */
+	/* STRBT		cccc 01x0 x110 xxxx xxxx xxxx xxxx xxxx */
+	/* LDRBT		cccc 01x0 x111 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0x0d200000, 0x04200000),
+
+	/* STR (immediate)	cccc 010x x0x0 xxxx xxxx xxxx xxxx xxxx */
+	/* STRB (immediate)	cccc 010x x1x0 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0x0e100000, 0x04000000, PROBES_STORE,
+						 REGS(NOPCWB, ANY, 0, 0, 0)),
+
+	/* LDR (immediate)	cccc 010x x0x1 xxxx xxxx xxxx xxxx xxxx */
+	/* LDRB (immediate)	cccc 010x x1x1 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0x0e100000, 0x04100000, PROBES_LOAD,
+						 REGS(NOPCWB, ANY, 0, 0, 0)),
+
+	/* STR (register)	cccc 011x x0x0 xxxx xxxx xxxx xxxx xxxx */
+	/* STRB (register)	cccc 011x x1x0 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0x0e100000, 0x06000000, PROBES_STORE,
+						 REGS(NOPCWB, ANY, 0, 0, NOPC)),
+
+	/* LDR (register)	cccc 011x x0x1 xxxx xxxx xxxx xxxx xxxx */
+	/* LDRB (register)	cccc 011x x1x1 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0x0e100000, 0x06100000, PROBES_LOAD,
+						 REGS(NOPCWB, ANY, 0, 0, NOPC)),
+
+	DECODE_END
+};
+
+static const union decode_item arm_cccc_100x_table[] = {
+	/* Block data transfer instructions				*/
+
+	/* LDM			cccc 100x x0x1 xxxx xxxx xxxx xxxx xxxx */
+	/* STM			cccc 100x x0x0 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_CUSTOM	(0x0e400000, 0x08000000, PROBES_LDMSTM),
+
+	/* STM (user registers)	cccc 100x x1x0 xxxx xxxx xxxx xxxx xxxx */
+	/* LDM (user registers)	cccc 100x x1x1 xxxx 0xxx xxxx xxxx xxxx */
+	/* LDM (exception ret)	cccc 100x x1x1 xxxx 1xxx xxxx xxxx xxxx */
+	DECODE_END
+};
+
+const union decode_item probes_decode_arm_table[] = {
+	/*
+	 * Unconditional instructions
+	 *			1111 xxxx xxxx xxxx xxxx xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0xf0000000, 0xf0000000, arm_1111_table),
+
+	/*
+	 * Miscellaneous instructions
+	 *			cccc 0001 0xx0 xxxx xxxx xxxx 0xxx xxxx
+	 */
+	DECODE_TABLE	(0x0f900080, 0x01000000, arm_cccc_0001_0xx0____0xxx_table),
+
+	/*
+	 * Halfword multiply and multiply-accumulate
+	 *			cccc 0001 0xx0 xxxx xxxx xxxx 1xx0 xxxx
+	 */
+	DECODE_TABLE	(0x0f900090, 0x01000080, arm_cccc_0001_0xx0____1xx0_table),
+
+	/*
+	 * Multiply and multiply-accumulate
+	 *			cccc 0000 xxxx xxxx xxxx xxxx 1001 xxxx
+	 */
+	DECODE_TABLE	(0x0f0000f0, 0x00000090, arm_cccc_0000_____1001_table),
+
+	/*
+	 * Synchronization primitives
+	 *			cccc 0001 xxxx xxxx xxxx xxxx 1001 xxxx
+	 */
+	DECODE_TABLE	(0x0f0000f0, 0x01000090, arm_cccc_0001_____1001_table),
+
+	/*
+	 * Extra load/store instructions
+	 *			cccc 000x xxxx xxxx xxxx xxxx 1xx1 xxxx
+	 */
+	DECODE_TABLE	(0x0e000090, 0x00000090, arm_cccc_000x_____1xx1_table),
+
+	/*
+	 * Data-processing (register)
+	 *			cccc 000x xxxx xxxx xxxx xxxx xxx0 xxxx
+	 * Data-processing (register-shifted register)
+	 *			cccc 000x xxxx xxxx xxxx xxxx 0xx1 xxxx
+	 */
+	DECODE_TABLE	(0x0e000000, 0x00000000, arm_cccc_000x_table),
+
+	/*
+	 * Data-processing (immediate)
+	 *			cccc 001x xxxx xxxx xxxx xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0x0e000000, 0x02000000, arm_cccc_001x_table),
+
+	/*
+	 * Media instructions
+	 *			cccc 011x xxxx xxxx xxxx xxxx xxx1 xxxx
+	 */
+	DECODE_TABLE	(0x0f000010, 0x06000010, arm_cccc_0110_____xxx1_table),
+	DECODE_TABLE	(0x0f000010, 0x07000010, arm_cccc_0111_____xxx1_table),
+
+	/*
+	 * Load/store word and unsigned byte
+	 *			cccc 01xx xxxx xxxx xxxx xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0x0c000000, 0x04000000, arm_cccc_01xx_table),
+
+	/*
+	 * Block data transfer instructions
+	 *			cccc 100x xxxx xxxx xxxx xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0x0e000000, 0x08000000, arm_cccc_100x_table),
+
+	/* B			cccc 1010 xxxx xxxx xxxx xxxx xxxx xxxx */
+	/* BL			cccc 1011 xxxx xxxx xxxx xxxx xxxx xxxx */
+	DECODE_SIMULATE	(0x0e000000, 0x0a000000, PROBES_BRANCH),
+
+	/*
+	 * Supervisor Call, and coprocessor instructions
+	 */
+
+	/* MCRR			cccc 1100 0100 xxxx xxxx xxxx xxxx xxxx */
+	/* MRRC			cccc 1100 0101 xxxx xxxx xxxx xxxx xxxx */
+	/* LDC			cccc 110x xxx1 xxxx xxxx xxxx xxxx xxxx */
+	/* STC			cccc 110x xxx0 xxxx xxxx xxxx xxxx xxxx */
+	/* CDP			cccc 1110 xxxx xxxx xxxx xxxx xxx0 xxxx */
+	/* MCR			cccc 1110 xxx0 xxxx xxxx xxxx xxx1 xxxx */
+	/* MRC			cccc 1110 xxx1 xxxx xxxx xxxx xxx1 xxxx */
+	/* SVC			cccc 1111 xxxx xxxx xxxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0x0c000000, 0x0c000000),
+
+	DECODE_END
+};
+#ifdef CONFIG_ARM_KPROBES_TEST_MODULE
+EXPORT_SYMBOL_GPL(probes_decode_arm_table);
+#endif
+
+static void __kprobes arm_singlestep(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	regs->ARM_pc += 4;
+	asi->insn_handler(insn, asi, regs);
+}
+
+/* Return:
+ *   INSN_REJECTED     If instruction is one not allowed to kprobe,
+ *   INSN_GOOD         If instruction is supported and uses instruction slot,
+ *   INSN_GOOD_NO_SLOT If instruction is supported but doesn't use its slot.
+ *
+ * For instructions we don't want to kprobe (INSN_REJECTED return result):
+ *   These are generally ones that modify the processor state making
+ *   them "hard" to simulate such as switches processor modes or
+ *   make accesses in alternate modes.  Any of these could be simulated
+ *   if the work was put into it, but low return considering they
+ *   should also be very rare.
+ */
+enum probes_insn __kprobes
+arm_probes_decode_insn(probes_opcode_t insn, struct arch_probes_insn *asi,
+		       bool emulate, const union decode_action *actions,
+		       const struct decode_checker *checkers[])
+{
+	asi->insn_singlestep = arm_singlestep;
+	asi->insn_check_cc = probes_condition_checks[insn>>28];
+	return probes_decode_insn(insn, asi, probes_decode_arm_table, false,
+				  emulate, actions, checkers);
+}

arch/arm/probes/decode-arm.h

@@ -0,0 +1,74 @@
+/*
+ * arch/arm/probes/decode-arm.h
+ *
+ * Copyright 2013 Linaro Ltd.
+ * Written by: David A. Long
+ *
+ * The code contained herein is licensed under the GNU General Public
+ * License. You may obtain a copy of the GNU General Public License
+ * Version 2 or later at the following locations:
+ *
+ * http://www.opensource.org/licenses/gpl-license.html
+ * http://www.gnu.org/copyleft/gpl.html
+ */
+
+#ifndef _ARM_KERNEL_PROBES_ARM_H
+#define  _ARM_KERNEL_PROBES_ARM_H
+
+#include "decode.h"
+
+enum probes_arm_action {
+	PROBES_PRELOAD_IMM,
+	PROBES_PRELOAD_REG,
+	PROBES_BRANCH_IMM,
+	PROBES_BRANCH_REG,
+	PROBES_MRS,
+	PROBES_CLZ,
+	PROBES_SATURATING_ARITHMETIC,
+	PROBES_MUL1,
+	PROBES_MUL2,
+	PROBES_SWP,
+	PROBES_LDRSTRD,
+	PROBES_LOAD,
+	PROBES_STORE,
+	PROBES_LOAD_EXTRA,
+	PROBES_STORE_EXTRA,
+	PROBES_MOV_IP_SP,
+	PROBES_DATA_PROCESSING_REG,
+	PROBES_DATA_PROCESSING_IMM,
+	PROBES_MOV_HALFWORD,
+	PROBES_SEV,
+	PROBES_WFE,
+	PROBES_SATURATE,
+	PROBES_REV,
+	PROBES_MMI,
+	PROBES_PACK,
+	PROBES_EXTEND,
+	PROBES_EXTEND_ADD,
+	PROBES_MUL_ADD_LONG,
+	PROBES_MUL_ADD,
+	PROBES_BITFIELD,
+	PROBES_BRANCH,
+	PROBES_LDMSTM,
+	NUM_PROBES_ARM_ACTIONS
+};
+
+void __kprobes simulate_bbl(probes_opcode_t opcode,
+	struct arch_probes_insn *asi, struct pt_regs *regs);
+void __kprobes simulate_blx1(probes_opcode_t opcode,
+	struct arch_probes_insn *asi, struct pt_regs *regs);
+void __kprobes simulate_blx2bx(probes_opcode_t opcode,
+	struct arch_probes_insn *asi, struct pt_regs *regs);
+void __kprobes simulate_mrs(probes_opcode_t opcode,
+	struct arch_probes_insn *asi, struct pt_regs *regs);
+void __kprobes simulate_mov_ipsp(probes_opcode_t opcode,
+	struct arch_probes_insn *asi, struct pt_regs *regs);
+
+extern const union decode_item probes_decode_arm_table[];
+
+enum probes_insn arm_probes_decode_insn(probes_opcode_t,
+		struct arch_probes_insn *, bool emulate,
+		const union decode_action *actions,
+		const struct decode_checker *checkers[]);
+
+#endif

arch/arm/probes/decode-thumb.c

@@ -0,0 +1,884 @@
+/*
+ * arch/arm/probes/decode-thumb.c
+ *
+ * Copyright (C) 2011 Jon Medhurst <tixy@yxit.co.uk>.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/stddef.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+
+#include "decode.h"
+#include "decode-thumb.h"
+
+
+static const union decode_item t32_table_1110_100x_x0xx[] = {
+	/* Load/store multiple instructions */
+
+	/* Rn is PC		1110 100x x0xx 1111 xxxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xfe4f0000, 0xe80f0000),
+
+	/* SRS			1110 1000 00x0 xxxx xxxx xxxx xxxx xxxx */
+	/* RFE			1110 1000 00x1 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xffc00000, 0xe8000000),
+	/* SRS			1110 1001 10x0 xxxx xxxx xxxx xxxx xxxx */
+	/* RFE			1110 1001 10x1 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xffc00000, 0xe9800000),
+
+	/* STM Rn, {...pc}	1110 100x x0x0 xxxx 1xxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xfe508000, 0xe8008000),
+	/* LDM Rn, {...lr,pc}	1110 100x x0x1 xxxx 11xx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xfe50c000, 0xe810c000),
+	/* LDM/STM Rn, {...sp}	1110 100x x0xx xxxx xx1x xxxx xxxx xxxx */
+	DECODE_REJECT	(0xfe402000, 0xe8002000),
+
+	/* STMIA		1110 1000 10x0 xxxx xxxx xxxx xxxx xxxx */
+	/* LDMIA		1110 1000 10x1 xxxx xxxx xxxx xxxx xxxx */
+	/* STMDB		1110 1001 00x0 xxxx xxxx xxxx xxxx xxxx */
+	/* LDMDB		1110 1001 00x1 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_CUSTOM	(0xfe400000, 0xe8000000, PROBES_T32_LDMSTM),
+
+	DECODE_END
+};
+
+static const union decode_item t32_table_1110_100x_x1xx[] = {
+	/* Load/store dual, load/store exclusive, table branch */
+
+	/* STRD (immediate)	1110 1000 x110 xxxx xxxx xxxx xxxx xxxx */
+	/* LDRD (immediate)	1110 1000 x111 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_OR	(0xff600000, 0xe8600000),
+	/* STRD (immediate)	1110 1001 x1x0 xxxx xxxx xxxx xxxx xxxx */
+	/* LDRD (immediate)	1110 1001 x1x1 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xff400000, 0xe9400000, PROBES_T32_LDRDSTRD,
+						 REGS(NOPCWB, NOSPPC, NOSPPC, 0, 0)),
+
+	/* TBB			1110 1000 1101 xxxx xxxx xxxx 0000 xxxx */
+	/* TBH			1110 1000 1101 xxxx xxxx xxxx 0001 xxxx */
+	DECODE_SIMULATEX(0xfff000e0, 0xe8d00000, PROBES_T32_TABLE_BRANCH,
+						 REGS(NOSP, 0, 0, 0, NOSPPC)),
+
+	/* STREX		1110 1000 0100 xxxx xxxx xxxx xxxx xxxx */
+	/* LDREX		1110 1000 0101 xxxx xxxx xxxx xxxx xxxx */
+	/* STREXB		1110 1000 1100 xxxx xxxx xxxx 0100 xxxx */
+	/* STREXH		1110 1000 1100 xxxx xxxx xxxx 0101 xxxx */
+	/* STREXD		1110 1000 1100 xxxx xxxx xxxx 0111 xxxx */
+	/* LDREXB		1110 1000 1101 xxxx xxxx xxxx 0100 xxxx */
+	/* LDREXH		1110 1000 1101 xxxx xxxx xxxx 0101 xxxx */
+	/* LDREXD		1110 1000 1101 xxxx xxxx xxxx 0111 xxxx */
+	/* And unallocated instructions...				*/
+	DECODE_END
+};
+
+static const union decode_item t32_table_1110_101x[] = {
+	/* Data-processing (shifted register)				*/
+
+	/* TST			1110 1010 0001 xxxx xxxx 1111 xxxx xxxx */
+	/* TEQ			1110 1010 1001 xxxx xxxx 1111 xxxx xxxx */
+	DECODE_EMULATEX	(0xff700f00, 0xea100f00, PROBES_T32_TST,
+						 REGS(NOSPPC, 0, 0, 0, NOSPPC)),
+
+	/* CMN			1110 1011 0001 xxxx xxxx 1111 xxxx xxxx */
+	DECODE_OR	(0xfff00f00, 0xeb100f00),
+	/* CMP			1110 1011 1011 xxxx xxxx 1111 xxxx xxxx */
+	DECODE_EMULATEX	(0xfff00f00, 0xebb00f00, PROBES_T32_TST,
+						 REGS(NOPC, 0, 0, 0, NOSPPC)),
+
+	/* MOV			1110 1010 010x 1111 xxxx xxxx xxxx xxxx */
+	/* MVN			1110 1010 011x 1111 xxxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xffcf0000, 0xea4f0000, PROBES_T32_MOV,
+						 REGS(0, 0, NOSPPC, 0, NOSPPC)),
+
+	/* ???			1110 1010 101x xxxx xxxx xxxx xxxx xxxx */
+	/* ???			1110 1010 111x xxxx xxxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xffa00000, 0xeaa00000),
+	/* ???			1110 1011 001x xxxx xxxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xffe00000, 0xeb200000),
+	/* ???			1110 1011 100x xxxx xxxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xffe00000, 0xeb800000),
+	/* ???			1110 1011 111x xxxx xxxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xffe00000, 0xebe00000),
+
+	/* ADD/SUB SP, SP, Rm, LSL #0..3				*/
+	/*			1110 1011 x0xx 1101 x000 1101 xx00 xxxx */
+	DECODE_EMULATEX	(0xff4f7f30, 0xeb0d0d00, PROBES_T32_ADDSUB,
+						 REGS(SP, 0, SP, 0, NOSPPC)),
+
+	/* ADD/SUB SP, SP, Rm, shift					*/
+	/*			1110 1011 x0xx 1101 xxxx 1101 xxxx xxxx */
+	DECODE_REJECT	(0xff4f0f00, 0xeb0d0d00),
+
+	/* ADD/SUB Rd, SP, Rm, shift					*/
+	/*			1110 1011 x0xx 1101 xxxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xff4f0000, 0xeb0d0000, PROBES_T32_ADDSUB,
+						 REGS(SP, 0, NOPC, 0, NOSPPC)),
+
+	/* AND			1110 1010 000x xxxx xxxx xxxx xxxx xxxx */
+	/* BIC			1110 1010 001x xxxx xxxx xxxx xxxx xxxx */
+	/* ORR			1110 1010 010x xxxx xxxx xxxx xxxx xxxx */
+	/* ORN			1110 1010 011x xxxx xxxx xxxx xxxx xxxx */
+	/* EOR			1110 1010 100x xxxx xxxx xxxx xxxx xxxx */
+	/* PKH			1110 1010 110x xxxx xxxx xxxx xxxx xxxx */
+	/* ADD			1110 1011 000x xxxx xxxx xxxx xxxx xxxx */
+	/* ADC			1110 1011 010x xxxx xxxx xxxx xxxx xxxx */
+	/* SBC			1110 1011 011x xxxx xxxx xxxx xxxx xxxx */
+	/* SUB			1110 1011 101x xxxx xxxx xxxx xxxx xxxx */
+	/* RSB			1110 1011 110x xxxx xxxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xfe000000, 0xea000000, PROBES_T32_LOGICAL,
+						 REGS(NOSPPC, 0, NOSPPC, 0, NOSPPC)),
+
+	DECODE_END
+};
+
+static const union decode_item t32_table_1111_0x0x___0[] = {
+	/* Data-processing (modified immediate)				*/
+
+	/* TST			1111 0x00 0001 xxxx 0xxx 1111 xxxx xxxx */
+	/* TEQ			1111 0x00 1001 xxxx 0xxx 1111 xxxx xxxx */
+	DECODE_EMULATEX	(0xfb708f00, 0xf0100f00, PROBES_T32_TST,
+						 REGS(NOSPPC, 0, 0, 0, 0)),
+
+	/* CMN			1111 0x01 0001 xxxx 0xxx 1111 xxxx xxxx */
+	DECODE_OR	(0xfbf08f00, 0xf1100f00),
+	/* CMP			1111 0x01 1011 xxxx 0xxx 1111 xxxx xxxx */
+	DECODE_EMULATEX	(0xfbf08f00, 0xf1b00f00, PROBES_T32_CMP,
+						 REGS(NOPC, 0, 0, 0, 0)),
+
+	/* MOV			1111 0x00 010x 1111 0xxx xxxx xxxx xxxx */
+	/* MVN			1111 0x00 011x 1111 0xxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xfbcf8000, 0xf04f0000, PROBES_T32_MOV,
+						 REGS(0, 0, NOSPPC, 0, 0)),
+
+	/* ???			1111 0x00 101x xxxx 0xxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xfbe08000, 0xf0a00000),
+	/* ???			1111 0x00 110x xxxx 0xxx xxxx xxxx xxxx */
+	/* ???			1111 0x00 111x xxxx 0xxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xfbc08000, 0xf0c00000),
+	/* ???			1111 0x01 001x xxxx 0xxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xfbe08000, 0xf1200000),
+	/* ???			1111 0x01 100x xxxx 0xxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xfbe08000, 0xf1800000),
+	/* ???			1111 0x01 111x xxxx 0xxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xfbe08000, 0xf1e00000),
+
+	/* ADD Rd, SP, #imm	1111 0x01 000x 1101 0xxx xxxx xxxx xxxx */
+	/* SUB Rd, SP, #imm	1111 0x01 101x 1101 0xxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xfb4f8000, 0xf10d0000, PROBES_T32_ADDSUB,
+						 REGS(SP, 0, NOPC, 0, 0)),
+
+	/* AND			1111 0x00 000x xxxx 0xxx xxxx xxxx xxxx */
+	/* BIC			1111 0x00 001x xxxx 0xxx xxxx xxxx xxxx */
+	/* ORR			1111 0x00 010x xxxx 0xxx xxxx xxxx xxxx */
+	/* ORN			1111 0x00 011x xxxx 0xxx xxxx xxxx xxxx */
+	/* EOR			1111 0x00 100x xxxx 0xxx xxxx xxxx xxxx */
+	/* ADD			1111 0x01 000x xxxx 0xxx xxxx xxxx xxxx */
+	/* ADC			1111 0x01 010x xxxx 0xxx xxxx xxxx xxxx */
+	/* SBC			1111 0x01 011x xxxx 0xxx xxxx xxxx xxxx */
+	/* SUB			1111 0x01 101x xxxx 0xxx xxxx xxxx xxxx */
+	/* RSB			1111 0x01 110x xxxx 0xxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xfa008000, 0xf0000000, PROBES_T32_LOGICAL,
+						 REGS(NOSPPC, 0, NOSPPC, 0, 0)),
+
+	DECODE_END
+};
+
+static const union decode_item t32_table_1111_0x1x___0[] = {
+	/* Data-processing (plain binary immediate)			*/
+
+	/* ADDW Rd, PC, #imm	1111 0x10 0000 1111 0xxx xxxx xxxx xxxx */
+	DECODE_OR	(0xfbff8000, 0xf20f0000),
+	/* SUBW	Rd, PC, #imm	1111 0x10 1010 1111 0xxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xfbff8000, 0xf2af0000, PROBES_T32_ADDWSUBW_PC,
+						 REGS(PC, 0, NOSPPC, 0, 0)),
+
+	/* ADDW SP, SP, #imm	1111 0x10 0000 1101 0xxx 1101 xxxx xxxx */
+	DECODE_OR	(0xfbff8f00, 0xf20d0d00),
+	/* SUBW	SP, SP, #imm	1111 0x10 1010 1101 0xxx 1101 xxxx xxxx */
+	DECODE_EMULATEX	(0xfbff8f00, 0xf2ad0d00, PROBES_T32_ADDWSUBW,
+						 REGS(SP, 0, SP, 0, 0)),
+
+	/* ADDW			1111 0x10 0000 xxxx 0xxx xxxx xxxx xxxx */
+	DECODE_OR	(0xfbf08000, 0xf2000000),
+	/* SUBW			1111 0x10 1010 xxxx 0xxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xfbf08000, 0xf2a00000, PROBES_T32_ADDWSUBW,
+						 REGS(NOPCX, 0, NOSPPC, 0, 0)),
+
+	/* MOVW			1111 0x10 0100 xxxx 0xxx xxxx xxxx xxxx */
+	/* MOVT			1111 0x10 1100 xxxx 0xxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xfb708000, 0xf2400000, PROBES_T32_MOVW,
+						 REGS(0, 0, NOSPPC, 0, 0)),
+
+	/* SSAT16		1111 0x11 0010 xxxx 0000 xxxx 00xx xxxx */
+	/* SSAT			1111 0x11 00x0 xxxx 0xxx xxxx xxxx xxxx */
+	/* USAT16		1111 0x11 1010 xxxx 0000 xxxx 00xx xxxx */
+	/* USAT			1111 0x11 10x0 xxxx 0xxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xfb508000, 0xf3000000, PROBES_T32_SAT,
+						 REGS(NOSPPC, 0, NOSPPC, 0, 0)),
+
+	/* SFBX			1111 0x11 0100 xxxx 0xxx xxxx xxxx xxxx */
+	/* UFBX			1111 0x11 1100 xxxx 0xxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xfb708000, 0xf3400000, PROBES_T32_BITFIELD,
+						 REGS(NOSPPC, 0, NOSPPC, 0, 0)),
+
+	/* BFC			1111 0x11 0110 1111 0xxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xfbff8000, 0xf36f0000, PROBES_T32_BITFIELD,
+						 REGS(0, 0, NOSPPC, 0, 0)),
+
+	/* BFI			1111 0x11 0110 xxxx 0xxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xfbf08000, 0xf3600000, PROBES_T32_BITFIELD,
+						 REGS(NOSPPCX, 0, NOSPPC, 0, 0)),
+
+	DECODE_END
+};
+
+static const union decode_item t32_table_1111_0xxx___1[] = {
+	/* Branches and miscellaneous control				*/
+
+	/* YIELD		1111 0011 1010 xxxx 10x0 x000 0000 0001 */
+	DECODE_OR	(0xfff0d7ff, 0xf3a08001),
+	/* SEV			1111 0011 1010 xxxx 10x0 x000 0000 0100 */
+	DECODE_EMULATE	(0xfff0d7ff, 0xf3a08004, PROBES_T32_SEV),
+	/* NOP			1111 0011 1010 xxxx 10x0 x000 0000 0000 */
+	/* WFE			1111 0011 1010 xxxx 10x0 x000 0000 0010 */
+	/* WFI			1111 0011 1010 xxxx 10x0 x000 0000 0011 */
+	DECODE_SIMULATE	(0xfff0d7fc, 0xf3a08000, PROBES_T32_WFE),
+
+	/* MRS Rd, CPSR		1111 0011 1110 xxxx 10x0 xxxx xxxx xxxx */
+	DECODE_SIMULATEX(0xfff0d000, 0xf3e08000, PROBES_T32_MRS,
+						 REGS(0, 0, NOSPPC, 0, 0)),
+
+	/*
+	 * Unsupported instructions
+	 *			1111 0x11 1xxx xxxx 10x0 xxxx xxxx xxxx
+	 *
+	 * MSR			1111 0011 100x xxxx 10x0 xxxx xxxx xxxx
+	 * DBG hint		1111 0011 1010 xxxx 10x0 x000 1111 xxxx
+	 * Unallocated hints	1111 0011 1010 xxxx 10x0 x000 xxxx xxxx
+	 * CPS			1111 0011 1010 xxxx 10x0 xxxx xxxx xxxx
+	 * CLREX/DSB/DMB/ISB	1111 0011 1011 xxxx 10x0 xxxx xxxx xxxx
+	 * BXJ			1111 0011 1100 xxxx 10x0 xxxx xxxx xxxx
+	 * SUBS PC,LR,#<imm8>	1111 0011 1101 xxxx 10x0 xxxx xxxx xxxx
+	 * MRS Rd, SPSR		1111 0011 1111 xxxx 10x0 xxxx xxxx xxxx
+	 * SMC			1111 0111 1111 xxxx 1000 xxxx xxxx xxxx
+	 * UNDEFINED		1111 0111 1111 xxxx 1010 xxxx xxxx xxxx
+	 * ???			1111 0111 1xxx xxxx 1010 xxxx xxxx xxxx
+	 */
+	DECODE_REJECT	(0xfb80d000, 0xf3808000),
+
+	/* Bcc			1111 0xxx xxxx xxxx 10x0 xxxx xxxx xxxx */
+	DECODE_CUSTOM	(0xf800d000, 0xf0008000, PROBES_T32_BRANCH_COND),
+
+	/* BLX			1111 0xxx xxxx xxxx 11x0 xxxx xxxx xxx0 */
+	DECODE_OR	(0xf800d001, 0xf000c000),
+	/* B			1111 0xxx xxxx xxxx 10x1 xxxx xxxx xxxx */
+	/* BL			1111 0xxx xxxx xxxx 11x1 xxxx xxxx xxxx */
+	DECODE_SIMULATE	(0xf8009000, 0xf0009000, PROBES_T32_BRANCH),
+
+	DECODE_END
+};
+
+static const union decode_item t32_table_1111_100x_x0x1__1111[] = {
+	/* Memory hints							*/
+
+	/* PLD (literal)	1111 1000 x001 1111 1111 xxxx xxxx xxxx */
+	/* PLI (literal)	1111 1001 x001 1111 1111 xxxx xxxx xxxx */
+	DECODE_SIMULATE	(0xfe7ff000, 0xf81ff000, PROBES_T32_PLDI),
+
+	/* PLD{W} (immediate)	1111 1000 10x1 xxxx 1111 xxxx xxxx xxxx */
+	DECODE_OR	(0xffd0f000, 0xf890f000),
+	/* PLD{W} (immediate)	1111 1000 00x1 xxxx 1111 1100 xxxx xxxx */
+	DECODE_OR	(0xffd0ff00, 0xf810fc00),
+	/* PLI (immediate)	1111 1001 1001 xxxx 1111 xxxx xxxx xxxx */
+	DECODE_OR	(0xfff0f000, 0xf990f000),
+	/* PLI (immediate)	1111 1001 0001 xxxx 1111 1100 xxxx xxxx */
+	DECODE_SIMULATEX(0xfff0ff00, 0xf910fc00, PROBES_T32_PLDI,
+						 REGS(NOPCX, 0, 0, 0, 0)),
+
+	/* PLD{W} (register)	1111 1000 00x1 xxxx 1111 0000 00xx xxxx */
+	DECODE_OR	(0xffd0ffc0, 0xf810f000),
+	/* PLI (register)	1111 1001 0001 xxxx 1111 0000 00xx xxxx */
+	DECODE_SIMULATEX(0xfff0ffc0, 0xf910f000, PROBES_T32_PLDI,
+						 REGS(NOPCX, 0, 0, 0, NOSPPC)),
+
+	/* Other unallocated instructions...				*/
+	DECODE_END
+};
+
+static const union decode_item t32_table_1111_100x[] = {
+	/* Store/Load single data item					*/
+
+	/* ???			1111 100x x11x xxxx xxxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xfe600000, 0xf8600000),
+
+	/* ???			1111 1001 0101 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xfff00000, 0xf9500000),
+
+	/* ???			1111 100x 0xxx xxxx xxxx 10x0 xxxx xxxx */
+	DECODE_REJECT	(0xfe800d00, 0xf8000800),
+
+	/* STRBT		1111 1000 0000 xxxx xxxx 1110 xxxx xxxx */
+	/* STRHT		1111 1000 0010 xxxx xxxx 1110 xxxx xxxx */
+	/* STRT			1111 1000 0100 xxxx xxxx 1110 xxxx xxxx */
+	/* LDRBT		1111 1000 0001 xxxx xxxx 1110 xxxx xxxx */
+	/* LDRSBT		1111 1001 0001 xxxx xxxx 1110 xxxx xxxx */
+	/* LDRHT		1111 1000 0011 xxxx xxxx 1110 xxxx xxxx */
+	/* LDRSHT		1111 1001 0011 xxxx xxxx 1110 xxxx xxxx */
+	/* LDRT			1111 1000 0101 xxxx xxxx 1110 xxxx xxxx */
+	DECODE_REJECT	(0xfe800f00, 0xf8000e00),
+
+	/* STR{,B,H} Rn,[PC...]	1111 1000 xxx0 1111 xxxx xxxx xxxx xxxx */
+	DECODE_REJECT	(0xff1f0000, 0xf80f0000),
+
+	/* STR{,B,H} PC,[Rn...]	1111 1000 xxx0 xxxx 1111 xxxx xxxx xxxx */
+	DECODE_REJECT	(0xff10f000, 0xf800f000),
+
+	/* LDR (literal)	1111 1000 x101 1111 xxxx xxxx xxxx xxxx */
+	DECODE_SIMULATEX(0xff7f0000, 0xf85f0000, PROBES_T32_LDR_LIT,
+						 REGS(PC, ANY, 0, 0, 0)),
+
+	/* STR (immediate)	1111 1000 0100 xxxx xxxx 1xxx xxxx xxxx */
+	/* LDR (immediate)	1111 1000 0101 xxxx xxxx 1xxx xxxx xxxx */
+	DECODE_OR	(0xffe00800, 0xf8400800),
+	/* STR (immediate)	1111 1000 1100 xxxx xxxx xxxx xxxx xxxx */
+	/* LDR (immediate)	1111 1000 1101 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xffe00000, 0xf8c00000, PROBES_T32_LDRSTR,
+						 REGS(NOPCX, ANY, 0, 0, 0)),
+
+	/* STR (register)	1111 1000 0100 xxxx xxxx 0000 00xx xxxx */
+	/* LDR (register)	1111 1000 0101 xxxx xxxx 0000 00xx xxxx */
+	DECODE_EMULATEX	(0xffe00fc0, 0xf8400000, PROBES_T32_LDRSTR,
+						 REGS(NOPCX, ANY, 0, 0, NOSPPC)),
+
+	/* LDRB (literal)	1111 1000 x001 1111 xxxx xxxx xxxx xxxx */
+	/* LDRSB (literal)	1111 1001 x001 1111 xxxx xxxx xxxx xxxx */
+	/* LDRH (literal)	1111 1000 x011 1111 xxxx xxxx xxxx xxxx */
+	/* LDRSH (literal)	1111 1001 x011 1111 xxxx xxxx xxxx xxxx */
+	DECODE_SIMULATEX(0xfe5f0000, 0xf81f0000, PROBES_T32_LDR_LIT,
+						 REGS(PC, NOSPPCX, 0, 0, 0)),
+
+	/* STRB (immediate)	1111 1000 0000 xxxx xxxx 1xxx xxxx xxxx */
+	/* STRH (immediate)	1111 1000 0010 xxxx xxxx 1xxx xxxx xxxx */
+	/* LDRB (immediate)	1111 1000 0001 xxxx xxxx 1xxx xxxx xxxx */
+	/* LDRSB (immediate)	1111 1001 0001 xxxx xxxx 1xxx xxxx xxxx */
+	/* LDRH (immediate)	1111 1000 0011 xxxx xxxx 1xxx xxxx xxxx */
+	/* LDRSH (immediate)	1111 1001 0011 xxxx xxxx 1xxx xxxx xxxx */
+	DECODE_OR	(0xfec00800, 0xf8000800),
+	/* STRB (immediate)	1111 1000 1000 xxxx xxxx xxxx xxxx xxxx */
+	/* STRH (immediate)	1111 1000 1010 xxxx xxxx xxxx xxxx xxxx */
+	/* LDRB (immediate)	1111 1000 1001 xxxx xxxx xxxx xxxx xxxx */
+	/* LDRSB (immediate)	1111 1001 1001 xxxx xxxx xxxx xxxx xxxx */
+	/* LDRH (immediate)	1111 1000 1011 xxxx xxxx xxxx xxxx xxxx */
+	/* LDRSH (immediate)	1111 1001 1011 xxxx xxxx xxxx xxxx xxxx */
+	DECODE_EMULATEX	(0xfec00000, 0xf8800000, PROBES_T32_LDRSTR,
+						 REGS(NOPCX, NOSPPCX, 0, 0, 0)),
+
+	/* STRB (register)	1111 1000 0000 xxxx xxxx 0000 00xx xxxx */
+	/* STRH (register)	1111 1000 0010 xxxx xxxx 0000 00xx xxxx */
+	/* LDRB (register)	1111 1000 0001 xxxx xxxx 0000 00xx xxxx */
+	/* LDRSB (register)	1111 1001 0001 xxxx xxxx 0000 00xx xxxx */
+	/* LDRH (register)	1111 1000 0011 xxxx xxxx 0000 00xx xxxx */
+	/* LDRSH (register)	1111 1001 0011 xxxx xxxx 0000 00xx xxxx */
+	DECODE_EMULATEX	(0xfe800fc0, 0xf8000000, PROBES_T32_LDRSTR,
+						 REGS(NOPCX, NOSPPCX, 0, 0, NOSPPC)),
+
+	/* Other unallocated instructions...				*/
+	DECODE_END
+};
+
+static const union decode_item t32_table_1111_1010___1111[] = {
+	/* Data-processing (register)					*/
+
+	/* ???			1111 1010 011x xxxx 1111 xxxx 1xxx xxxx */
+	DECODE_REJECT	(0xffe0f080, 0xfa60f080),
+
+	/* SXTH			1111 1010 0000 1111 1111 xxxx 1xxx xxxx */
+	/* UXTH			1111 1010 0001 1111 1111 xxxx 1xxx xxxx */
+	/* SXTB16		1111 1010 0010 1111 1111 xxxx 1xxx xxxx */
+	/* UXTB16		1111 1010 0011 1111 1111 xxxx 1xxx xxxx */
+	/* SXTB			1111 1010 0100 1111 1111 xxxx 1xxx xxxx */
+	/* UXTB			1111 1010 0101 1111 1111 xxxx 1xxx xxxx */
+	DECODE_EMULATEX	(0xff8ff080, 0xfa0ff080, PROBES_T32_SIGN_EXTEND,
+						 REGS(0, 0, NOSPPC, 0, NOSPPC)),
+
+
+	/* ???			1111 1010 1xxx xxxx 1111 xxxx 0x11 xxxx */
+	DECODE_REJECT	(0xff80f0b0, 0xfa80f030),
+	/* ???			1111 1010 1x11 xxxx 1111 xxxx 0xxx xxxx */
+	DECODE_REJECT	(0xffb0f080, 0xfab0f000),
+
+	/* SADD16		1111 1010 1001 xxxx 1111 xxxx 0000 xxxx */
+	/* SASX			1111 1010 1010 xxxx 1111 xxxx 0000 xxxx */
+	/* SSAX			1111 1010 1110 xxxx 1111 xxxx 0000 xxxx */
+	/* SSUB16		1111 1010 1101 xxxx 1111 xxxx 0000 xxxx */
+	/* SADD8		1111 1010 1000 xxxx 1111 xxxx 0000 xxxx */
+	/* SSUB8		1111 1010 1100 xxxx 1111 xxxx 0000 xxxx */
+
+	/* QADD16		1111 1010 1001 xxxx 1111 xxxx 0001 xxxx */
+	/* QASX			1111 1010 1010 xxxx 1111 xxxx 0001 xxxx */
+	/* QSAX			1111 1010 1110 xxxx 1111 xxxx 0001 xxxx */
+	/* QSUB16		1111 1010 1101 xxxx 1111 xxxx 0001 xxxx */
+	/* QADD8		1111 1010 1000 xxxx 1111 xxxx 0001 xxxx */
+	/* QSUB8		1111 1010 1100 xxxx 1111 xxxx 0001 xxxx */
+
+	/* SHADD16		1111 1010 1001 xxxx 1111 xxxx 0010 xxxx */
+	/* SHASX		1111 1010 1010 xxxx 1111 xxxx 0010 xxxx */
+	/* SHSAX		1111 1010 1110 xxxx 1111 xxxx 0010 xxxx */
+	/* SHSUB16		1111 1010 1101 xxxx 1111 xxxx 0010 xxxx */
+	/* SHADD8		1111 1010 1000 xxxx 1111 xxxx 0010 xxxx */
+	/* SHSUB8		1111 1010 1100 xxxx 1111 xxxx 0010 xxxx */
+
+	/* UADD16		1111 1010 1001 xxxx 1111 xxxx 0100 xxxx */
+	/* UASX			1111 1010 1010 xxxx 1111 xxxx 0100 xxxx */
+	/* USAX			1111 1010 1110 xxxx 1111 xxxx 0100 xxxx */
+	/* USUB16		1111 1010 1101 xxxx 1111 xxxx 0100 xxxx */
+	/* UADD8		1111 1010 1000 xxxx 1111 xxxx 0100 xxxx */
+	/* USUB8		1111 1010 1100 xxxx 1111 xxxx 0100 xxxx */
+
+	/* UQADD16		1111 1010 1001 xxxx 1111 xxxx 0101 xxxx */
+	/* UQASX		1111 1010 1010 xxxx 1111 xxxx 0101 xxxx */
+	/* UQSAX		1111 1010 1110 xxxx 1111 xxxx 0101 xxxx */
+	/* UQSUB16		1111 1010 1101 xxxx 1111 xxxx 0101 xxxx */
+	/* UQADD8		1111 1010 1000 xxxx 1111 xxxx 0101 xxxx */
+	/* UQSUB8		1111 1010 1100 xxxx 1111 xxxx 0101 xxxx */
+
+	/* UHADD16		1111 1010 1001 xxxx 1111 xxxx 0110 xxxx */
+	/* UHASX		1111 1010 1010 xxxx 1111 xxxx 0110 xxxx */
+	/* UHSAX		1111 1010 1110 xxxx 1111 xxxx 0110 xxxx */
+	/* UHSUB16		1111 1010 1101 xxxx 1111 xxxx 0110 xxxx */
+	/* UHADD8		1111 1010 1000 xxxx 1111 xxxx 0110 xxxx */
+	/* UHSUB8		1111 1010 1100 xxxx 1111 xxxx 0110 xxxx */
+	DECODE_OR	(0xff80f080, 0xfa80f000),
+
+	/* SXTAH		1111 1010 0000 xxxx 1111 xxxx 1xxx xxxx */
+	/* UXTAH		1111 1010 0001 xxxx 1111 xxxx 1xxx xxxx */
+	/* SXTAB16		1111 1010 0010 xxxx 1111 xxxx 1xxx xxxx */
+	/* UXTAB16		1111 1010 0011 xxxx 1111 xxxx 1xxx xxxx */
+	/* SXTAB		1111 1010 0100 xxxx 1111 xxxx 1xxx xxxx */
+	/* UXTAB		1111 1010 0101 xxxx 1111 xxxx 1xxx xxxx */
+	DECODE_OR	(0xff80f080, 0xfa00f080),
+
+	/* QADD			1111 1010 1000 xxxx 1111 xxxx 1000 xxxx */
+	/* QDADD		1111 1010 1000 xxxx 1111 xxxx 1001 xxxx */
+	/* QSUB			1111 1010 1000 xxxx 1111 xxxx 1010 xxxx */
+	/* QDSUB		1111 1010 1000 xxxx 1111 xxxx 1011 xxxx */
+	DECODE_OR	(0xfff0f0c0, 0xfa80f080),
+
+	/* SEL			1111 1010 1010 xxxx 1111 xxxx 1000 xxxx */
+	DECODE_OR	(0xfff0f0f0, 0xfaa0f080),
+
+	/* LSL			1111 1010 000x xxxx 1111 xxxx 0000 xxxx */
+	/* LSR			1111 1010 001x xxxx 1111 xxxx 0000 xxxx */
+	/* ASR			1111 1010 010x xxxx 1111 xxxx 0000 xxxx */
+	/* ROR			1111 1010 011x xxxx 1111 xxxx 0000 xxxx */
+	DECODE_EMULATEX	(0xff80f0f0, 0xfa00f000, PROBES_T32_MEDIA,
+						 REGS(NOSPPC, 0, NOSPPC, 0, NOSPPC)),
+
+	/* CLZ			1111 1010 1010 xxxx 1111 xxxx 1000 xxxx */
+	DECODE_OR	(0xfff0f0f0, 0xfab0f080),
+
+	/* REV			1111 1010 1001 xxxx 1111 xxxx 1000 xxxx */
+	/* REV16		1111 1010 1001 xxxx 1111 xxxx 1001 xxxx */
+	/* RBIT			1111 1010 1001 xxxx 1111 xxxx 1010 xxxx */
+	/* REVSH		1111 1010 1001 xxxx 1111 xxxx 1011 xxxx */
+	DECODE_EMULATEX	(0xfff0f0c0, 0xfa90f080, PROBES_T32_REVERSE,
+						 REGS(NOSPPC, 0, NOSPPC, 0, SAMEAS16)),
+
+	/* Other unallocated instructions...				*/
+	DECODE_END
+};
+
+static const union decode_item t32_table_1111_1011_0[] = {
+	/* Multiply, multiply accumulate, and absolute difference	*/
+
+	/* ???			1111 1011 0000 xxxx 1111 xxxx 0001 xxxx */
+	DECODE_REJECT	(0xfff0f0f0, 0xfb00f010),
+	/* ???			1111 1011 0111 xxxx 1111 xxxx 0001 xxxx */
+	DECODE_REJECT	(0xfff0f0f0, 0xfb70f010),
+
+	/* SMULxy		1111 1011 0001 xxxx 1111 xxxx 00xx xxxx */
+	DECODE_OR	(0xfff0f0c0, 0xfb10f000),
+	/* MUL			1111 1011 0000 xxxx 1111 xxxx 0000 xxxx */
+	/* SMUAD{X}		1111 1011 0010 xxxx 1111 xxxx 000x xxxx */
+	/* SMULWy		1111 1011 0011 xxxx 1111 xxxx 000x xxxx */
+	/* SMUSD{X}		1111 1011 0100 xxxx 1111 xxxx 000x xxxx */
+	/* SMMUL{R}		1111 1011 0101 xxxx 1111 xxxx 000x xxxx */
+	/* USAD8		1111 1011 0111 xxxx 1111 xxxx 0000 xxxx */
+	DECODE_EMULATEX	(0xff80f0e0, 0xfb00f000, PROBES_T32_MUL_ADD,
+						 REGS(NOSPPC, 0, NOSPPC, 0, NOSPPC)),
+
+	/* ???			1111 1011 0111 xxxx xxxx xxxx 0001 xxxx */
+	DECODE_REJECT	(0xfff000f0, 0xfb700010),
+
+	/* SMLAxy		1111 1011 0001 xxxx xxxx xxxx 00xx xxxx */
+	DECODE_OR	(0xfff000c0, 0xfb100000),
+	/* MLA			1111 1011 0000 xxxx xxxx xxxx 0000 xxxx */
+	/* MLS			1111 1011 0000 xxxx xxxx xxxx 0001 xxxx */
+	/* SMLAD{X}		1111 1011 0010 xxxx xxxx xxxx 000x xxxx */
+	/* SMLAWy		1111 1011 0011 xxxx xxxx xxxx 000x xxxx */
+	/* SMLSD{X}		1111 1011 0100 xxxx xxxx xxxx 000x xxxx */
+	/* SMMLA{R}		1111 1011 0101 xxxx xxxx xxxx 000x xxxx */
+	/* SMMLS{R}		1111 1011 0110 xxxx xxxx xxxx 000x xxxx */
+	/* USADA8		1111 1011 0111 xxxx xxxx xxxx 0000 xxxx */
+	DECODE_EMULATEX	(0xff8000c0, 0xfb000000,  PROBES_T32_MUL_ADD2,
+						 REGS(NOSPPC, NOSPPCX, NOSPPC, 0, NOSPPC)),
+
+	/* Other unallocated instructions...				*/
+	DECODE_END
+};
+
+static const union decode_item t32_table_1111_1011_1[] = {
+	/* Long multiply, long multiply accumulate, and divide		*/
+
+	/* UMAAL		1111 1011 1110 xxxx xxxx xxxx 0110 xxxx */
+	DECODE_OR	(0xfff000f0, 0xfbe00060),
+	/* SMLALxy		1111 1011 1100 xxxx xxxx xxxx 10xx xxxx */
+	DECODE_OR	(0xfff000c0, 0xfbc00080),
+	/* SMLALD{X}		1111 1011 1100 xxxx xxxx xxxx 110x xxxx */
+	/* SMLSLD{X}		1111 1011 1101 xxxx xxxx xxxx 110x xxxx */
+	DECODE_OR	(0xffe000e0, 0xfbc000c0),
+	/* SMULL		1111 1011 1000 xxxx xxxx xxxx 0000 xxxx */
+	/* UMULL		1111 1011 1010 xxxx xxxx xxxx 0000 xxxx */
+	/* SMLAL		1111 1011 1100 xxxx xxxx xxxx 0000 xxxx */
+	/* UMLAL		1111 1011 1110 xxxx xxxx xxxx 0000 xxxx */
+	DECODE_EMULATEX	(0xff9000f0, 0xfb800000, PROBES_T32_MUL_ADD_LONG,
+						 REGS(NOSPPC, NOSPPC, NOSPPC, 0, NOSPPC)),
+
+	/* SDIV			1111 1011 1001 xxxx xxxx xxxx 1111 xxxx */
+	/* UDIV			1111 1011 1011 xxxx xxxx xxxx 1111 xxxx */
+	/* Other unallocated instructions...				*/
+	DECODE_END
+};
+
+const union decode_item probes_decode_thumb32_table[] = {
+
+	/*
+	 * Load/store multiple instructions
+	 *			1110 100x x0xx xxxx xxxx xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0xfe400000, 0xe8000000, t32_table_1110_100x_x0xx),
+
+	/*
+	 * Load/store dual, load/store exclusive, table branch
+	 *			1110 100x x1xx xxxx xxxx xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0xfe400000, 0xe8400000, t32_table_1110_100x_x1xx),
+
+	/*
+	 * Data-processing (shifted register)
+	 *			1110 101x xxxx xxxx xxxx xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0xfe000000, 0xea000000, t32_table_1110_101x),
+
+	/*
+	 * Coprocessor instructions
+	 *			1110 11xx xxxx xxxx xxxx xxxx xxxx xxxx
+	 */
+	DECODE_REJECT	(0xfc000000, 0xec000000),
+
+	/*
+	 * Data-processing (modified immediate)
+	 *			1111 0x0x xxxx xxxx 0xxx xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0xfa008000, 0xf0000000, t32_table_1111_0x0x___0),
+
+	/*
+	 * Data-processing (plain binary immediate)
+	 *			1111 0x1x xxxx xxxx 0xxx xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0xfa008000, 0xf2000000, t32_table_1111_0x1x___0),
+
+	/*
+	 * Branches and miscellaneous control
+	 *			1111 0xxx xxxx xxxx 1xxx xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0xf8008000, 0xf0008000, t32_table_1111_0xxx___1),
+
+	/*
+	 * Advanced SIMD element or structure load/store instructions
+	 *			1111 1001 xxx0 xxxx xxxx xxxx xxxx xxxx
+	 */
+	DECODE_REJECT	(0xff100000, 0xf9000000),
+
+	/*
+	 * Memory hints
+	 *			1111 100x x0x1 xxxx 1111 xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0xfe50f000, 0xf810f000, t32_table_1111_100x_x0x1__1111),
+
+	/*
+	 * Store single data item
+	 *			1111 1000 xxx0 xxxx xxxx xxxx xxxx xxxx
+	 * Load single data items
+	 *			1111 100x xxx1 xxxx xxxx xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0xfe000000, 0xf8000000, t32_table_1111_100x),
+
+	/*
+	 * Data-processing (register)
+	 *			1111 1010 xxxx xxxx 1111 xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0xff00f000, 0xfa00f000, t32_table_1111_1010___1111),
+
+	/*
+	 * Multiply, multiply accumulate, and absolute difference
+	 *			1111 1011 0xxx xxxx xxxx xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0xff800000, 0xfb000000, t32_table_1111_1011_0),
+
+	/*
+	 * Long multiply, long multiply accumulate, and divide
+	 *			1111 1011 1xxx xxxx xxxx xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0xff800000, 0xfb800000, t32_table_1111_1011_1),
+
+	/*
+	 * Coprocessor instructions
+	 *			1111 11xx xxxx xxxx xxxx xxxx xxxx xxxx
+	 */
+	DECODE_END
+};
+#ifdef CONFIG_ARM_KPROBES_TEST_MODULE
+EXPORT_SYMBOL_GPL(probes_decode_thumb32_table);
+#endif
+
+static const union decode_item t16_table_1011[] = {
+	/* Miscellaneous 16-bit instructions		    */
+
+	/* ADD (SP plus immediate)	1011 0000 0xxx xxxx */
+	/* SUB (SP minus immediate)	1011 0000 1xxx xxxx */
+	DECODE_SIMULATE	(0xff00, 0xb000, PROBES_T16_ADD_SP),
+
+	/* CBZ				1011 00x1 xxxx xxxx */
+	/* CBNZ				1011 10x1 xxxx xxxx */
+	DECODE_SIMULATE	(0xf500, 0xb100, PROBES_T16_CBZ),
+
+	/* SXTH				1011 0010 00xx xxxx */
+	/* SXTB				1011 0010 01xx xxxx */
+	/* UXTH				1011 0010 10xx xxxx */
+	/* UXTB				1011 0010 11xx xxxx */
+	/* REV				1011 1010 00xx xxxx */
+	/* REV16			1011 1010 01xx xxxx */
+	/* ???				1011 1010 10xx xxxx */
+	/* REVSH			1011 1010 11xx xxxx */
+	DECODE_REJECT	(0xffc0, 0xba80),
+	DECODE_EMULATE	(0xf500, 0xb000, PROBES_T16_SIGN_EXTEND),
+
+	/* PUSH				1011 010x xxxx xxxx */
+	DECODE_CUSTOM	(0xfe00, 0xb400, PROBES_T16_PUSH),
+	/* POP				1011 110x xxxx xxxx */
+	DECODE_CUSTOM	(0xfe00, 0xbc00, PROBES_T16_POP),
+
+	/*
+	 * If-Then, and hints
+	 *				1011 1111 xxxx xxxx
+	 */
+
+	/* YIELD			1011 1111 0001 0000 */
+	DECODE_OR	(0xffff, 0xbf10),
+	/* SEV				1011 1111 0100 0000 */
+	DECODE_EMULATE	(0xffff, 0xbf40, PROBES_T16_SEV),
+	/* NOP				1011 1111 0000 0000 */
+	/* WFE				1011 1111 0010 0000 */
+	/* WFI				1011 1111 0011 0000 */
+	DECODE_SIMULATE	(0xffcf, 0xbf00, PROBES_T16_WFE),
+	/* Unassigned hints		1011 1111 xxxx 0000 */
+	DECODE_REJECT	(0xff0f, 0xbf00),
+	/* IT				1011 1111 xxxx xxxx */
+	DECODE_CUSTOM	(0xff00, 0xbf00, PROBES_T16_IT),
+
+	/* SETEND			1011 0110 010x xxxx */
+	/* CPS				1011 0110 011x xxxx */
+	/* BKPT				1011 1110 xxxx xxxx */
+	/* And unallocated instructions...		    */
+	DECODE_END
+};
+
+const union decode_item probes_decode_thumb16_table[] = {
+
+	/*
+	 * Shift (immediate), add, subtract, move, and compare
+	 *				00xx xxxx xxxx xxxx
+	 */
+
+	/* CMP (immediate)		0010 1xxx xxxx xxxx */
+	DECODE_EMULATE	(0xf800, 0x2800, PROBES_T16_CMP),
+
+	/* ADD (register)		0001 100x xxxx xxxx */
+	/* SUB (register)		0001 101x xxxx xxxx */
+	/* LSL (immediate)		0000 0xxx xxxx xxxx */
+	/* LSR (immediate)		0000 1xxx xxxx xxxx */
+	/* ASR (immediate)		0001 0xxx xxxx xxxx */
+	/* ADD (immediate, Thumb)	0001 110x xxxx xxxx */
+	/* SUB (immediate, Thumb)	0001 111x xxxx xxxx */
+	/* MOV (immediate)		0010 0xxx xxxx xxxx */
+	/* ADD (immediate, Thumb)	0011 0xxx xxxx xxxx */
+	/* SUB (immediate, Thumb)	0011 1xxx xxxx xxxx */
+	DECODE_EMULATE	(0xc000, 0x0000, PROBES_T16_ADDSUB),
+
+	/*
+	 * 16-bit Thumb data-processing instructions
+	 *				0100 00xx xxxx xxxx
+	 */
+
+	/* TST (register)		0100 0010 00xx xxxx */
+	DECODE_EMULATE	(0xffc0, 0x4200, PROBES_T16_CMP),
+	/* CMP (register)		0100 0010 10xx xxxx */
+	/* CMN (register)		0100 0010 11xx xxxx */
+	DECODE_EMULATE	(0xff80, 0x4280, PROBES_T16_CMP),
+	/* AND (register)		0100 0000 00xx xxxx */
+	/* EOR (register)		0100 0000 01xx xxxx */
+	/* LSL (register)		0100 0000 10xx xxxx */
+	/* LSR (register)		0100 0000 11xx xxxx */
+	/* ASR (register)		0100 0001 00xx xxxx */
+	/* ADC (register)		0100 0001 01xx xxxx */
+	/* SBC (register)		0100 0001 10xx xxxx */
+	/* ROR (register)		0100 0001 11xx xxxx */
+	/* RSB (immediate)		0100 0010 01xx xxxx */
+	/* ORR (register)		0100 0011 00xx xxxx */
+	/* MUL				0100 0011 00xx xxxx */
+	/* BIC (register)		0100 0011 10xx xxxx */
+	/* MVN (register)		0100 0011 10xx xxxx */
+	DECODE_EMULATE	(0xfc00, 0x4000, PROBES_T16_LOGICAL),
+
+	/*
+	 * Special data instructions and branch and exchange
+	 *				0100 01xx xxxx xxxx
+	 */
+
+	/* BLX pc			0100 0111 1111 1xxx */
+	DECODE_REJECT	(0xfff8, 0x47f8),
+
+	/* BX (register)		0100 0111 0xxx xxxx */
+	/* BLX (register)		0100 0111 1xxx xxxx */
+	DECODE_SIMULATE (0xff00, 0x4700, PROBES_T16_BLX),
+
+	/* ADD pc, pc			0100 0100 1111 1111 */
+	DECODE_REJECT	(0xffff, 0x44ff),
+
+	/* ADD (register)		0100 0100 xxxx xxxx */
+	/* CMP (register)		0100 0101 xxxx xxxx */
+	/* MOV (register)		0100 0110 xxxx xxxx */
+	DECODE_CUSTOM	(0xfc00, 0x4400, PROBES_T16_HIREGOPS),
+
+	/*
+	 * Load from Literal Pool
+	 * LDR (literal)		0100 1xxx xxxx xxxx
+	 */
+	DECODE_SIMULATE	(0xf800, 0x4800, PROBES_T16_LDR_LIT),
+
+	/*
+	 * 16-bit Thumb Load/store instructions
+	 *				0101 xxxx xxxx xxxx
+	 *				011x xxxx xxxx xxxx
+	 *				100x xxxx xxxx xxxx
+	 */
+
+	/* STR (register)		0101 000x xxxx xxxx */
+	/* STRH (register)		0101 001x xxxx xxxx */
+	/* STRB (register)		0101 010x xxxx xxxx */
+	/* LDRSB (register)		0101 011x xxxx xxxx */
+	/* LDR (register)		0101 100x xxxx xxxx */
+	/* LDRH (register)		0101 101x xxxx xxxx */
+	/* LDRB (register)		0101 110x xxxx xxxx */
+	/* LDRSH (register)		0101 111x xxxx xxxx */
+	/* STR (immediate, Thumb)	0110 0xxx xxxx xxxx */
+	/* LDR (immediate, Thumb)	0110 1xxx xxxx xxxx */
+	/* STRB (immediate, Thumb)	0111 0xxx xxxx xxxx */
+	/* LDRB (immediate, Thumb)	0111 1xxx xxxx xxxx */
+	DECODE_EMULATE	(0xc000, 0x4000, PROBES_T16_LDRHSTRH),
+	/* STRH (immediate, Thumb)	1000 0xxx xxxx xxxx */
+	/* LDRH (immediate, Thumb)	1000 1xxx xxxx xxxx */
+	DECODE_EMULATE	(0xf000, 0x8000, PROBES_T16_LDRHSTRH),
+	/* STR (immediate, Thumb)	1001 0xxx xxxx xxxx */
+	/* LDR (immediate, Thumb)	1001 1xxx xxxx xxxx */
+	DECODE_SIMULATE	(0xf000, 0x9000, PROBES_T16_LDRSTR),
+
+	/*
+	 * Generate PC-/SP-relative address
+	 * ADR (literal)		1010 0xxx xxxx xxxx
+	 * ADD (SP plus immediate)	1010 1xxx xxxx xxxx
+	 */
+	DECODE_SIMULATE	(0xf000, 0xa000, PROBES_T16_ADR),
+
+	/*
+	 * Miscellaneous 16-bit instructions
+	 *				1011 xxxx xxxx xxxx
+	 */
+	DECODE_TABLE	(0xf000, 0xb000, t16_table_1011),
+
+	/* STM				1100 0xxx xxxx xxxx */
+	/* LDM				1100 1xxx xxxx xxxx */
+	DECODE_EMULATE	(0xf000, 0xc000, PROBES_T16_LDMSTM),
+
+	/*
+	 * Conditional branch, and Supervisor Call
+	 */
+
+	/* Permanently UNDEFINED	1101 1110 xxxx xxxx */
+	/* SVC				1101 1111 xxxx xxxx */
+	DECODE_REJECT	(0xfe00, 0xde00),
+
+	/* Conditional branch		1101 xxxx xxxx xxxx */
+	DECODE_CUSTOM	(0xf000, 0xd000, PROBES_T16_BRANCH_COND),
+
+	/*
+	 * Unconditional branch
+	 * B				1110 0xxx xxxx xxxx
+	 */
+	DECODE_SIMULATE	(0xf800, 0xe000, PROBES_T16_BRANCH),
+
+	DECODE_END
+};
+#ifdef CONFIG_ARM_KPROBES_TEST_MODULE
+EXPORT_SYMBOL_GPL(probes_decode_thumb16_table);
+#endif
+
+static unsigned long __kprobes thumb_check_cc(unsigned long cpsr)
+{
+	if (unlikely(in_it_block(cpsr)))
+		return probes_condition_checks[current_cond(cpsr)](cpsr);
+	return true;
+}
+
+static void __kprobes thumb16_singlestep(probes_opcode_t opcode,
+		struct arch_probes_insn *asi,
+		struct pt_regs *regs)
+{
+	regs->ARM_pc += 2;
+	asi->insn_handler(opcode, asi, regs);
+	regs->ARM_cpsr = it_advance(regs->ARM_cpsr);
+}
+
+static void __kprobes thumb32_singlestep(probes_opcode_t opcode,
+		struct arch_probes_insn *asi,
+		struct pt_regs *regs)
+{
+	regs->ARM_pc += 4;
+	asi->insn_handler(opcode, asi, regs);
+	regs->ARM_cpsr = it_advance(regs->ARM_cpsr);
+}
+
+enum probes_insn __kprobes
+thumb16_probes_decode_insn(probes_opcode_t insn, struct arch_probes_insn *asi,
+			   bool emulate, const union decode_action *actions,
+			   const struct decode_checker *checkers[])
+{
+	asi->insn_singlestep = thumb16_singlestep;
+	asi->insn_check_cc = thumb_check_cc;
+	return probes_decode_insn(insn, asi, probes_decode_thumb16_table, true,
+				  emulate, actions, checkers);
+}
+
+enum probes_insn __kprobes
+thumb32_probes_decode_insn(probes_opcode_t insn, struct arch_probes_insn *asi,
+			   bool emulate, const union decode_action *actions,
+			   const struct decode_checker *checkers[])
+{
+	asi->insn_singlestep = thumb32_singlestep;
+	asi->insn_check_cc = thumb_check_cc;
+	return probes_decode_insn(insn, asi, probes_decode_thumb32_table, true,
+				  emulate, actions, checkers);
+}

arch/arm/probes/decode-thumb.h

@@ -0,0 +1,101 @@
+/*
+ * arch/arm/probes/decode-thumb.h
+ *
+ * Copyright 2013 Linaro Ltd.
+ * Written by: David A. Long
+ *
+ * The code contained herein is licensed under the GNU General Public
+ * License. You may obtain a copy of the GNU General Public License
+ * Version 2 or later at the following locations:
+ *
+ * http://www.opensource.org/licenses/gpl-license.html
+ * http://www.gnu.org/copyleft/gpl.html
+ */
+
+#ifndef _ARM_KERNEL_PROBES_THUMB_H
+#define  _ARM_KERNEL_PROBES_THUMB_H
+
+#include "decode.h"
+
+/*
+ * True if current instruction is in an IT block.
+ */
+#define in_it_block(cpsr)	((cpsr & 0x06000c00) != 0x00000000)
+
+/*
+ * Return the condition code to check for the currently executing instruction.
+ * This is in ITSTATE<7:4> which is in CPSR<15:12> but is only valid if
+ * in_it_block returns true.
+ */
+#define current_cond(cpsr)	((cpsr >> 12) & 0xf)
+
+enum probes_t32_action {
+	PROBES_T32_EMULATE_NONE,
+	PROBES_T32_SIMULATE_NOP,
+	PROBES_T32_LDMSTM,
+	PROBES_T32_LDRDSTRD,
+	PROBES_T32_TABLE_BRANCH,
+	PROBES_T32_TST,
+	PROBES_T32_CMP,
+	PROBES_T32_MOV,
+	PROBES_T32_ADDSUB,
+	PROBES_T32_LOGICAL,
+	PROBES_T32_ADDWSUBW_PC,
+	PROBES_T32_ADDWSUBW,
+	PROBES_T32_MOVW,
+	PROBES_T32_SAT,
+	PROBES_T32_BITFIELD,
+	PROBES_T32_SEV,
+	PROBES_T32_WFE,
+	PROBES_T32_MRS,
+	PROBES_T32_BRANCH_COND,
+	PROBES_T32_BRANCH,
+	PROBES_T32_PLDI,
+	PROBES_T32_LDR_LIT,
+	PROBES_T32_LDRSTR,
+	PROBES_T32_SIGN_EXTEND,
+	PROBES_T32_MEDIA,
+	PROBES_T32_REVERSE,
+	PROBES_T32_MUL_ADD,
+	PROBES_T32_MUL_ADD2,
+	PROBES_T32_MUL_ADD_LONG,
+	NUM_PROBES_T32_ACTIONS
+};
+
+enum probes_t16_action {
+	PROBES_T16_ADD_SP,
+	PROBES_T16_CBZ,
+	PROBES_T16_SIGN_EXTEND,
+	PROBES_T16_PUSH,
+	PROBES_T16_POP,
+	PROBES_T16_SEV,
+	PROBES_T16_WFE,
+	PROBES_T16_IT,
+	PROBES_T16_CMP,
+	PROBES_T16_ADDSUB,
+	PROBES_T16_LOGICAL,
+	PROBES_T16_BLX,
+	PROBES_T16_HIREGOPS,
+	PROBES_T16_LDR_LIT,
+	PROBES_T16_LDRHSTRH,
+	PROBES_T16_LDRSTR,
+	PROBES_T16_ADR,
+	PROBES_T16_LDMSTM,
+	PROBES_T16_BRANCH_COND,
+	PROBES_T16_BRANCH,
+	NUM_PROBES_T16_ACTIONS
+};
+
+extern const union decode_item probes_decode_thumb32_table[];
+extern const union decode_item probes_decode_thumb16_table[];
+
+enum probes_insn __kprobes
+thumb16_probes_decode_insn(probes_opcode_t insn, struct arch_probes_insn *asi,
+		bool emulate, const union decode_action *actions,
+		const struct decode_checker *checkers[]);
+enum probes_insn __kprobes
+thumb32_probes_decode_insn(probes_opcode_t insn, struct arch_probes_insn *asi,
+		bool emulate, const union decode_action *actions,
+		const struct decode_checker *checkers[]);
+
+#endif

arch/arm/probes/decode.c

@@ -0,0 +1,521 @@
+/*
+ * arch/arm/probes/decode.c
+ *
+ * Copyright (C) 2011 Jon Medhurst <tixy@yxit.co.uk>.
+ *
+ * Some contents moved here from arch/arm/include/asm/kprobes-arm.c which is
+ * Copyright (C) 2006, 2007 Motorola Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <asm/system_info.h>
+#include <asm/ptrace.h>
+#include <linux/bug.h>
+
+#include "decode.h"
+
+
+#ifndef find_str_pc_offset
+
+/*
+ * For STR and STM instructions, an ARM core may choose to use either
+ * a +8 or a +12 displacement from the current instruction's address.
+ * Whichever value is chosen for a given core, it must be the same for
+ * both instructions and may not change.  This function measures it.
+ */
+
+int str_pc_offset;
+
+void __init find_str_pc_offset(void)
+{
+	int addr, scratch, ret;
+
+	__asm__ (
+		"sub	%[ret], pc, #4		\n\t"
+		"str	pc, %[addr]		\n\t"
+		"ldr	%[scr], %[addr]		\n\t"
+		"sub	%[ret], %[scr], %[ret]	\n\t"
+		: [ret] "=r" (ret), [scr] "=r" (scratch), [addr] "+m" (addr));
+
+	str_pc_offset = ret;
+}
+
+#endif /* !find_str_pc_offset */
+
+
+#ifndef test_load_write_pc_interworking
+
+bool load_write_pc_interworks;
+
+void __init test_load_write_pc_interworking(void)
+{
+	int arch = cpu_architecture();
+	BUG_ON(arch == CPU_ARCH_UNKNOWN);
+	load_write_pc_interworks = arch >= CPU_ARCH_ARMv5T;
+}
+
+#endif /* !test_load_write_pc_interworking */
+
+
+#ifndef test_alu_write_pc_interworking
+
+bool alu_write_pc_interworks;
+
+void __init test_alu_write_pc_interworking(void)
+{
+	int arch = cpu_architecture();
+	BUG_ON(arch == CPU_ARCH_UNKNOWN);
+	alu_write_pc_interworks = arch >= CPU_ARCH_ARMv7;
+}
+
+#endif /* !test_alu_write_pc_interworking */
+
+
+void __init arm_probes_decode_init(void)
+{
+	find_str_pc_offset();
+	test_load_write_pc_interworking();
+	test_alu_write_pc_interworking();
+}
+
+
+static unsigned long __kprobes __check_eq(unsigned long cpsr)
+{
+	return cpsr & PSR_Z_BIT;
+}
+
+static unsigned long __kprobes __check_ne(unsigned long cpsr)
+{
+	return (~cpsr) & PSR_Z_BIT;
+}
+
+static unsigned long __kprobes __check_cs(unsigned long cpsr)
+{
+	return cpsr & PSR_C_BIT;
+}
+
+static unsigned long __kprobes __check_cc(unsigned long cpsr)
+{
+	return (~cpsr) & PSR_C_BIT;
+}
+
+static unsigned long __kprobes __check_mi(unsigned long cpsr)
+{
+	return cpsr & PSR_N_BIT;
+}
+
+static unsigned long __kprobes __check_pl(unsigned long cpsr)
+{
+	return (~cpsr) & PSR_N_BIT;
+}
+
+static unsigned long __kprobes __check_vs(unsigned long cpsr)
+{
+	return cpsr & PSR_V_BIT;
+}
+
+static unsigned long __kprobes __check_vc(unsigned long cpsr)
+{
+	return (~cpsr) & PSR_V_BIT;
+}
+
+static unsigned long __kprobes __check_hi(unsigned long cpsr)
+{
+	cpsr &= ~(cpsr >> 1); /* PSR_C_BIT &= ~PSR_Z_BIT */
+	return cpsr & PSR_C_BIT;
+}
+
+static unsigned long __kprobes __check_ls(unsigned long cpsr)
+{
+	cpsr &= ~(cpsr >> 1); /* PSR_C_BIT &= ~PSR_Z_BIT */
+	return (~cpsr) & PSR_C_BIT;
+}
+
+static unsigned long __kprobes __check_ge(unsigned long cpsr)
+{
+	cpsr ^= (cpsr << 3); /* PSR_N_BIT ^= PSR_V_BIT */
+	return (~cpsr) & PSR_N_BIT;
+}
+
+static unsigned long __kprobes __check_lt(unsigned long cpsr)
+{
+	cpsr ^= (cpsr << 3); /* PSR_N_BIT ^= PSR_V_BIT */
+	return cpsr & PSR_N_BIT;
+}
+
+static unsigned long __kprobes __check_gt(unsigned long cpsr)
+{
+	unsigned long temp = cpsr ^ (cpsr << 3); /* PSR_N_BIT ^= PSR_V_BIT */
+	temp |= (cpsr << 1);			 /* PSR_N_BIT |= PSR_Z_BIT */
+	return (~temp) & PSR_N_BIT;
+}
+
+static unsigned long __kprobes __check_le(unsigned long cpsr)
+{
+	unsigned long temp = cpsr ^ (cpsr << 3); /* PSR_N_BIT ^= PSR_V_BIT */
+	temp |= (cpsr << 1);			 /* PSR_N_BIT |= PSR_Z_BIT */
+	return temp & PSR_N_BIT;
+}
+
+static unsigned long __kprobes __check_al(unsigned long cpsr)
+{
+	return true;
+}
+
+probes_check_cc * const probes_condition_checks[16] = {
+	&__check_eq, &__check_ne, &__check_cs, &__check_cc,
+	&__check_mi, &__check_pl, &__check_vs, &__check_vc,
+	&__check_hi, &__check_ls, &__check_ge, &__check_lt,
+	&__check_gt, &__check_le, &__check_al, &__check_al
+};
+
+
+void __kprobes probes_simulate_nop(probes_opcode_t opcode,
+	struct arch_probes_insn *asi,
+	struct pt_regs *regs)
+{
+}
+
+void __kprobes probes_emulate_none(probes_opcode_t opcode,
+	struct arch_probes_insn *asi,
+	struct pt_regs *regs)
+{
+	asi->insn_fn();
+}
+
+/*
+ * Prepare an instruction slot to receive an instruction for emulating.
+ * This is done by placing a subroutine return after the location where the
+ * instruction will be placed. We also modify ARM instructions to be
+ * unconditional as the condition code will already be checked before any
+ * emulation handler is called.
+ */
+static probes_opcode_t __kprobes
+prepare_emulated_insn(probes_opcode_t insn, struct arch_probes_insn *asi,
+		      bool thumb)
+{
+#ifdef CONFIG_THUMB2_KERNEL
+	if (thumb) {
+		u16 *thumb_insn = (u16 *)asi->insn;
+		/* Thumb bx lr */
+		thumb_insn[1] = __opcode_to_mem_thumb16(0x4770);
+		thumb_insn[2] = __opcode_to_mem_thumb16(0x4770);
+		return insn;
+	}
+	asi->insn[1] = __opcode_to_mem_arm(0xe12fff1e); /* ARM bx lr */
+#else
+	asi->insn[1] = __opcode_to_mem_arm(0xe1a0f00e); /* mov pc, lr */
+#endif
+	/* Make an ARM instruction unconditional */
+	if (insn < 0xe0000000)
+		insn = (insn | 0xe0000000) & ~0x10000000;
+	return insn;
+}
+
+/*
+ * Write a (probably modified) instruction into the slot previously prepared by
+ * prepare_emulated_insn
+ */
+static void  __kprobes
+set_emulated_insn(probes_opcode_t insn, struct arch_probes_insn *asi,
+		  bool thumb)
+{
+#ifdef CONFIG_THUMB2_KERNEL
+	if (thumb) {
+		u16 *ip = (u16 *)asi->insn;
+		if (is_wide_instruction(insn))
+			*ip++ = __opcode_to_mem_thumb16(insn >> 16);
+		*ip++ = __opcode_to_mem_thumb16(insn);
+		return;
+	}
+#endif
+	asi->insn[0] = __opcode_to_mem_arm(insn);
+}
+
+/*
+ * When we modify the register numbers encoded in an instruction to be emulated,
+ * the new values come from this define. For ARM and 32-bit Thumb instructions
+ * this gives...
+ *
+ *	bit position	  16  12   8   4   0
+ *	---------------+---+---+---+---+---+
+ *	register	 r2  r0  r1  --  r3
+ */
+#define INSN_NEW_BITS		0x00020103
+
+/* Each nibble has same value as that at INSN_NEW_BITS bit 16 */
+#define INSN_SAMEAS16_BITS	0x22222222
+
+/*
+ * Validate and modify each of the registers encoded in an instruction.
+ *
+ * Each nibble in regs contains a value from enum decode_reg_type. For each
+ * non-zero value, the corresponding nibble in pinsn is validated and modified
+ * according to the type.
+ */
+static bool __kprobes decode_regs(probes_opcode_t *pinsn, u32 regs, bool modify)
+{
+	probes_opcode_t insn = *pinsn;
+	probes_opcode_t mask = 0xf; /* Start at least significant nibble */
+
+	for (; regs != 0; regs >>= 4, mask <<= 4) {
+
+		probes_opcode_t new_bits = INSN_NEW_BITS;
+
+		switch (regs & 0xf) {
+
+		case REG_TYPE_NONE:
+			/* Nibble not a register, skip to next */
+			continue;
+
+		case REG_TYPE_ANY:
+			/* Any register is allowed */
+			break;
+
+		case REG_TYPE_SAMEAS16:
+			/* Replace register with same as at bit position 16 */
+			new_bits = INSN_SAMEAS16_BITS;
+			break;
+
+		case REG_TYPE_SP:
+			/* Only allow SP (R13) */
+			if ((insn ^ 0xdddddddd) & mask)
+				goto reject;
+			break;
+
+		case REG_TYPE_PC:
+			/* Only allow PC (R15) */
+			if ((insn ^ 0xffffffff) & mask)
+				goto reject;
+			break;
+
+		case REG_TYPE_NOSP:
+			/* Reject SP (R13) */
+			if (((insn ^ 0xdddddddd) & mask) == 0)
+				goto reject;
+			break;
+
+		case REG_TYPE_NOSPPC:
+		case REG_TYPE_NOSPPCX:
+			/* Reject SP and PC (R13 and R15) */
+			if (((insn ^ 0xdddddddd) & 0xdddddddd & mask) == 0)
+				goto reject;
+			break;
+
+		case REG_TYPE_NOPCWB:
+			if (!is_writeback(insn))
+				break; /* No writeback, so any register is OK */
+			/* fall through... */
+		case REG_TYPE_NOPC:
+		case REG_TYPE_NOPCX:
+			/* Reject PC (R15) */
+			if (((insn ^ 0xffffffff) & mask) == 0)
+				goto reject;
+			break;
+		}
+
+		/* Replace value of nibble with new register number... */
+		insn &= ~mask;
+		insn |= new_bits & mask;
+	}
+
+	if (modify)
+		*pinsn = insn;
+
+	return true;
+
+reject:
+	return false;
+}
+
+static const int decode_struct_sizes[NUM_DECODE_TYPES] = {
+	[DECODE_TYPE_TABLE]	= sizeof(struct decode_table),
+	[DECODE_TYPE_CUSTOM]	= sizeof(struct decode_custom),
+	[DECODE_TYPE_SIMULATE]	= sizeof(struct decode_simulate),
+	[DECODE_TYPE_EMULATE]	= sizeof(struct decode_emulate),
+	[DECODE_TYPE_OR]	= sizeof(struct decode_or),
+	[DECODE_TYPE_REJECT]	= sizeof(struct decode_reject)
+};
+
+static int run_checkers(const struct decode_checker *checkers[],
+		int action, probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	const struct decode_checker **p;
+
+	if (!checkers)
+		return INSN_GOOD;
+
+	p = checkers;
+	while (*p != NULL) {
+		int retval;
+		probes_check_t *checker_func = (*p)[action].checker;
+
+		retval = INSN_GOOD;
+		if (checker_func)
+			retval = checker_func(insn, asi, h);
+		if (retval == INSN_REJECTED)
+			return retval;
+		p++;
+	}
+	return INSN_GOOD;
+}
+
+/*
+ * probes_decode_insn operates on data tables in order to decode an ARM
+ * architecture instruction onto which a kprobe has been placed.
+ *
+ * These instruction decoding tables are a concatenation of entries each
+ * of which consist of one of the following structs:
+ *
+ *	decode_table
+ *	decode_custom
+ *	decode_simulate
+ *	decode_emulate
+ *	decode_or
+ *	decode_reject
+ *
+ * Each of these starts with a struct decode_header which has the following
+ * fields:
+ *
+ *	type_regs
+ *	mask
+ *	value
+ *
+ * The least significant DECODE_TYPE_BITS of type_regs contains a value
+ * from enum decode_type, this indicates which of the decode_* structs
+ * the entry contains. The value DECODE_TYPE_END indicates the end of the
+ * table.
+ *
+ * When the table is parsed, each entry is checked in turn to see if it
+ * matches the instruction to be decoded using the test:
+ *
+ *	(insn & mask) == value
+ *
+ * If no match is found before the end of the table is reached then decoding
+ * fails with INSN_REJECTED.
+ *
+ * When a match is found, decode_regs() is called to validate and modify each
+ * of the registers encoded in the instruction; the data it uses to do this
+ * is (type_regs >> DECODE_TYPE_BITS). A validation failure will cause decoding
+ * to fail with INSN_REJECTED.
+ *
+ * Once the instruction has passed the above tests, further processing
+ * depends on the type of the table entry's decode struct.
+ *
+ */
+int __kprobes
+probes_decode_insn(probes_opcode_t insn, struct arch_probes_insn *asi,
+		   const union decode_item *table, bool thumb,
+		   bool emulate, const union decode_action *actions,
+		   const struct decode_checker *checkers[])
+{
+	const struct decode_header *h = (struct decode_header *)table;
+	const struct decode_header *next;
+	bool matched = false;
+	/*
+	 * @insn can be modified by decode_regs. Save its original
+	 * value for checkers.
+	 */
+	probes_opcode_t origin_insn = insn;
+
+	/*
+	 * stack_space is initialized to 0 here. Checker functions
+	 * should update is value if they find this is a stack store
+	 * instruction: positive value means bytes of stack usage,
+	 * negitive value means unable to determine stack usage
+	 * statically. For instruction doesn't store to stack, checker
+	 * do nothing with it.
+	 */
+	asi->stack_space = 0;
+
+	/*
+	 * Similarly to stack_space, register_usage_flags is filled by
+	 * checkers. Its default value is set to ~0, which is 'all
+	 * registers are used', to prevent any potential optimization.
+	 */
+	asi->register_usage_flags = ~0UL;
+
+	if (emulate)
+		insn = prepare_emulated_insn(insn, asi, thumb);
+
+	for (;; h = next) {
+		enum decode_type type = h->type_regs.bits & DECODE_TYPE_MASK;
+		u32 regs = h->type_regs.bits >> DECODE_TYPE_BITS;
+
+		if (type == DECODE_TYPE_END)
+			return INSN_REJECTED;
+
+		next = (struct decode_header *)
+				((uintptr_t)h + decode_struct_sizes[type]);
+
+		if (!matched && (insn & h->mask.bits) != h->value.bits)
+			continue;
+
+		if (!decode_regs(&insn, regs, emulate))
+			return INSN_REJECTED;
+
+		switch (type) {
+
+		case DECODE_TYPE_TABLE: {
+			struct decode_table *d = (struct decode_table *)h;
+			next = (struct decode_header *)d->table.table;
+			break;
+		}
+
+		case DECODE_TYPE_CUSTOM: {
+			int err;
+			struct decode_custom *d = (struct decode_custom *)h;
+			int action = d->decoder.action;
+
+			err = run_checkers(checkers, action, origin_insn, asi, h);
+			if (err == INSN_REJECTED)
+				return INSN_REJECTED;
+			return actions[action].decoder(insn, asi, h);
+		}
+
+		case DECODE_TYPE_SIMULATE: {
+			int err;
+			struct decode_simulate *d = (struct decode_simulate *)h;
+			int action = d->handler.action;
+
+			err = run_checkers(checkers, action, origin_insn, asi, h);
+			if (err == INSN_REJECTED)
+				return INSN_REJECTED;
+			asi->insn_handler = actions[action].handler;
+			return INSN_GOOD_NO_SLOT;
+		}
+
+		case DECODE_TYPE_EMULATE: {
+			int err;
+			struct decode_emulate *d = (struct decode_emulate *)h;
+			int action = d->handler.action;
+
+			err = run_checkers(checkers, action, origin_insn, asi, h);
+			if (err == INSN_REJECTED)
+				return INSN_REJECTED;
+
+			if (!emulate)
+				return actions[action].decoder(insn, asi, h);
+
+			asi->insn_handler = actions[action].handler;
+			set_emulated_insn(insn, asi, thumb);
+			return INSN_GOOD;
+		}
+
+		case DECODE_TYPE_OR:
+			matched = true;
+			break;
+
+		case DECODE_TYPE_REJECT:
+		default:
+			return INSN_REJECTED;
+		}
+	}
+}

arch/arm/probes/decode.h

@@ -0,0 +1,416 @@
+/*
+ * arch/arm/probes/decode.h
+ *
+ * Copyright (C) 2011 Jon Medhurst <tixy@yxit.co.uk>.
+ *
+ * Some contents moved here from arch/arm/include/asm/kprobes.h which is
+ * Copyright (C) 2006, 2007 Motorola Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+#ifndef _ARM_KERNEL_PROBES_H
+#define  _ARM_KERNEL_PROBES_H
+
+#include <linux/types.h>
+#include <linux/stddef.h>
+#include <asm/probes.h>
+
+void __init arm_probes_decode_init(void);
+
+extern probes_check_cc * const probes_condition_checks[16];
+
+#if __LINUX_ARM_ARCH__ >= 7
+
+/* str_pc_offset is architecturally defined from ARMv7 onwards */
+#define str_pc_offset 8
+#define find_str_pc_offset()
+
+#else /* __LINUX_ARM_ARCH__ < 7 */
+
+/* We need a run-time check to determine str_pc_offset */
+extern int str_pc_offset;
+void __init find_str_pc_offset(void);
+
+#endif
+
+
+/*
+ * Update ITSTATE after normal execution of an IT block instruction.
+ *
+ * The 8 IT state bits are split into two parts in CPSR:
+ *	ITSTATE<1:0> are in CPSR<26:25>
+ *	ITSTATE<7:2> are in CPSR<15:10>
+ */
+static inline unsigned long it_advance(unsigned long cpsr)
+	{
+	if ((cpsr & 0x06000400) == 0) {
+		/* ITSTATE<2:0> == 0 means end of IT block, so clear IT state */
+		cpsr &= ~PSR_IT_MASK;
+	} else {
+		/* We need to shift left ITSTATE<4:0> */
+		const unsigned long mask = 0x06001c00;  /* Mask ITSTATE<4:0> */
+		unsigned long it = cpsr & mask;
+		it <<= 1;
+		it |= it >> (27 - 10);  /* Carry ITSTATE<2> to correct place */
+		it &= mask;
+		cpsr &= ~mask;
+		cpsr |= it;
+	}
+	return cpsr;
+}
+
+static inline void __kprobes bx_write_pc(long pcv, struct pt_regs *regs)
+{
+	long cpsr = regs->ARM_cpsr;
+	if (pcv & 0x1) {
+		cpsr |= PSR_T_BIT;
+		pcv &= ~0x1;
+	} else {
+		cpsr &= ~PSR_T_BIT;
+		pcv &= ~0x2;	/* Avoid UNPREDICTABLE address allignment */
+	}
+	regs->ARM_cpsr = cpsr;
+	regs->ARM_pc = pcv;
+}
+
+
+#if __LINUX_ARM_ARCH__ >= 6
+
+/* Kernels built for >= ARMv6 should never run on <= ARMv5 hardware, so... */
+#define load_write_pc_interworks true
+#define test_load_write_pc_interworking()
+
+#else /* __LINUX_ARM_ARCH__ < 6 */
+
+/* We need run-time testing to determine if load_write_pc() should interwork. */
+extern bool load_write_pc_interworks;
+void __init test_load_write_pc_interworking(void);
+
+#endif
+
+static inline void __kprobes load_write_pc(long pcv, struct pt_regs *regs)
+{
+	if (load_write_pc_interworks)
+		bx_write_pc(pcv, regs);
+	else
+		regs->ARM_pc = pcv;
+}
+
+
+#if __LINUX_ARM_ARCH__ >= 7
+
+#define alu_write_pc_interworks true
+#define test_alu_write_pc_interworking()
+
+#elif __LINUX_ARM_ARCH__ <= 5
+
+/* Kernels built for <= ARMv5 should never run on >= ARMv6 hardware, so... */
+#define alu_write_pc_interworks false
+#define test_alu_write_pc_interworking()
+
+#else /* __LINUX_ARM_ARCH__ == 6 */
+
+/* We could be an ARMv6 binary on ARMv7 hardware so we need a run-time check. */
+extern bool alu_write_pc_interworks;
+void __init test_alu_write_pc_interworking(void);
+
+#endif /* __LINUX_ARM_ARCH__ == 6 */
+
+static inline void __kprobes alu_write_pc(long pcv, struct pt_regs *regs)
+{
+	if (alu_write_pc_interworks)
+		bx_write_pc(pcv, regs);
+	else
+		regs->ARM_pc = pcv;
+}
+
+
+/*
+ * Test if load/store instructions writeback the address register.
+ * if P (bit 24) == 0 or W (bit 21) == 1
+ */
+#define is_writeback(insn) ((insn ^ 0x01000000) & 0x01200000)
+
+/*
+ * The following definitions and macros are used to build instruction
+ * decoding tables for use by probes_decode_insn.
+ *
+ * These tables are a concatenation of entries each of which consist of one of
+ * the decode_* structs. All of the fields in every type of decode structure
+ * are of the union type decode_item, therefore the entire decode table can be
+ * viewed as an array of these and declared like:
+ *
+ *	static const union decode_item table_name[] = {};
+ *
+ * In order to construct each entry in the table, macros are used to
+ * initialise a number of sequential decode_item values in a layout which
+ * matches the relevant struct. E.g. DECODE_SIMULATE initialise a struct
+ * decode_simulate by initialising four decode_item objects like this...
+ *
+ *	{.bits = _type},
+ *	{.bits = _mask},
+ *	{.bits = _value},
+ *	{.action = _handler},
+ *
+ * Initialising a specified member of the union means that the compiler
+ * will produce a warning if the argument is of an incorrect type.
+ *
+ * Below is a list of each of the macros used to initialise entries and a
+ * description of the action performed when that entry is matched to an
+ * instruction. A match is found when (instruction & mask) == value.
+ *
+ * DECODE_TABLE(mask, value, table)
+ *	Instruction decoding jumps to parsing the new sub-table 'table'.
+ *
+ * DECODE_CUSTOM(mask, value, decoder)
+ *	The value of 'decoder' is used as an index into the array of
+ *	action functions, and the retrieved decoder function is invoked
+ *	to complete decoding of the instruction.
+ *
+ * DECODE_SIMULATE(mask, value, handler)
+ *	The probes instruction handler is set to the value found by
+ *	indexing into the action array using the value of 'handler'. This
+ *	will be used to simulate the instruction when the probe is hit.
+ *	Decoding returns with INSN_GOOD_NO_SLOT.
+ *
+ * DECODE_EMULATE(mask, value, handler)
+ *	The probes instruction handler is set to the value found by
+ *	indexing into the action array using the value of 'handler'. This
+ *	will be used to emulate the instruction when the probe is hit. The
+ *	modified instruction (see below) is placed in the probes instruction
+ *	slot so it may be called by the emulation code. Decoding returns
+ *	with INSN_GOOD.
+ *
+ * DECODE_REJECT(mask, value)
+ *	Instruction decoding fails with INSN_REJECTED
+ *
+ * DECODE_OR(mask, value)
+ *	This allows the mask/value test of multiple table entries to be
+ *	logically ORed. Once an 'or' entry is matched the decoding action to
+ *	be performed is that of the next entry which isn't an 'or'. E.g.
+ *
+ *		DECODE_OR	(mask1, value1)
+ *		DECODE_OR	(mask2, value2)
+ *		DECODE_SIMULATE	(mask3, value3, simulation_handler)
+ *
+ *	This means that if any of the three mask/value pairs match the
+ *	instruction being decoded, then 'simulation_handler' will be used
+ *	for it.
+ *
+ * Both the SIMULATE and EMULATE macros have a second form which take an
+ * additional 'regs' argument.
+ *
+ *	DECODE_SIMULATEX(mask, value, handler, regs)
+ *	DECODE_EMULATEX	(mask, value, handler, regs)
+ *
+ * These are used to specify what kind of CPU register is encoded in each of the
+ * least significant 5 nibbles of the instruction being decoded. The regs value
+ * is specified using the REGS macro, this takes any of the REG_TYPE_* values
+ * from enum decode_reg_type as arguments; only the '*' part of the name is
+ * given. E.g.
+ *
+ *	REGS(0, ANY, NOPC, 0, ANY)
+ *
+ * This indicates an instruction is encoded like:
+ *
+ *	bits 19..16	ignore
+ *	bits 15..12	any register allowed here
+ *	bits 11.. 8	any register except PC allowed here
+ *	bits  7.. 4	ignore
+ *	bits  3.. 0	any register allowed here
+ *
+ * This register specification is checked after a decode table entry is found to
+ * match an instruction (through the mask/value test). Any invalid register then
+ * found in the instruction will cause decoding to fail with INSN_REJECTED. In
+ * the above example this would happen if bits 11..8 of the instruction were
+ * 1111, indicating R15 or PC.
+ *
+ * As well as checking for legal combinations of registers, this data is also
+ * used to modify the registers encoded in the instructions so that an
+ * emulation routines can use it. (See decode_regs() and INSN_NEW_BITS.)
+ *
+ * Here is a real example which matches ARM instructions of the form
+ * "AND <Rd>,<Rn>,<Rm>,<shift> <Rs>"
+ *
+ *	DECODE_EMULATEX	(0x0e000090, 0x00000010, PROBES_DATA_PROCESSING_REG,
+ *						 REGS(ANY, ANY, NOPC, 0, ANY)),
+ *						      ^    ^    ^        ^
+ *						      Rn   Rd   Rs       Rm
+ *
+ * Decoding the instruction "AND R4, R5, R6, ASL R15" will be rejected because
+ * Rs == R15
+ *
+ * Decoding the instruction "AND R4, R5, R6, ASL R7" will be accepted and the
+ * instruction will be modified to "AND R0, R2, R3, ASL R1" and then placed into
+ * the kprobes instruction slot. This can then be called later by the handler
+ * function emulate_rd12rn16rm0rs8_rwflags (a pointer to which is retrieved from
+ * the indicated slot in the action array), in order to simulate the instruction.
+ */
+
+enum decode_type {
+	DECODE_TYPE_END,
+	DECODE_TYPE_TABLE,
+	DECODE_TYPE_CUSTOM,
+	DECODE_TYPE_SIMULATE,
+	DECODE_TYPE_EMULATE,
+	DECODE_TYPE_OR,
+	DECODE_TYPE_REJECT,
+	NUM_DECODE_TYPES /* Must be last enum */
+};
+
+#define DECODE_TYPE_BITS	4
+#define DECODE_TYPE_MASK	((1 << DECODE_TYPE_BITS) - 1)
+
+enum decode_reg_type {
+	REG_TYPE_NONE = 0, /* Not a register, ignore */
+	REG_TYPE_ANY,	   /* Any register allowed */
+	REG_TYPE_SAMEAS16, /* Register should be same as that at bits 19..16 */
+	REG_TYPE_SP,	   /* Register must be SP */
+	REG_TYPE_PC,	   /* Register must be PC */
+	REG_TYPE_NOSP,	   /* Register must not be SP */
+	REG_TYPE_NOSPPC,   /* Register must not be SP or PC */
+	REG_TYPE_NOPC,	   /* Register must not be PC */
+	REG_TYPE_NOPCWB,   /* No PC if load/store write-back flag also set */
+
+	/* The following types are used when the encoding for PC indicates
+	 * another instruction form. This distiction only matters for test
+	 * case coverage checks.
+	 */
+	REG_TYPE_NOPCX,	   /* Register must not be PC */
+	REG_TYPE_NOSPPCX,  /* Register must not be SP or PC */
+
+	/* Alias to allow '0' arg to be used in REGS macro. */
+	REG_TYPE_0 = REG_TYPE_NONE
+};
+
+#define REGS(r16, r12, r8, r4, r0)	\
+	(((REG_TYPE_##r16) << 16) +	\
+	((REG_TYPE_##r12) << 12) +	\
+	((REG_TYPE_##r8) << 8) +	\
+	((REG_TYPE_##r4) << 4) +	\
+	(REG_TYPE_##r0))
+
+union decode_item {
+	u32			bits;
+	const union decode_item	*table;
+	int			action;
+};
+
+struct decode_header;
+typedef enum probes_insn (probes_custom_decode_t)(probes_opcode_t,
+						  struct arch_probes_insn *,
+						  const struct decode_header *);
+
+union decode_action {
+	probes_insn_handler_t	*handler;
+	probes_custom_decode_t	*decoder;
+};
+
+typedef enum probes_insn (probes_check_t)(probes_opcode_t,
+					   struct arch_probes_insn *,
+					   const struct decode_header *);
+
+struct decode_checker {
+	probes_check_t	*checker;
+};
+
+#define DECODE_END			\
+	{.bits = DECODE_TYPE_END}
+
+
+struct decode_header {
+	union decode_item	type_regs;
+	union decode_item	mask;
+	union decode_item	value;
+};
+
+#define DECODE_HEADER(_type, _mask, _value, _regs)		\
+	{.bits = (_type) | ((_regs) << DECODE_TYPE_BITS)},	\
+	{.bits = (_mask)},					\
+	{.bits = (_value)}
+
+
+struct decode_table {
+	struct decode_header	header;
+	union decode_item	table;
+};
+
+#define DECODE_TABLE(_mask, _value, _table)			\
+	DECODE_HEADER(DECODE_TYPE_TABLE, _mask, _value, 0),	\
+	{.table = (_table)}
+
+
+struct decode_custom {
+	struct decode_header	header;
+	union decode_item	decoder;
+};
+
+#define DECODE_CUSTOM(_mask, _value, _decoder)			\
+	DECODE_HEADER(DECODE_TYPE_CUSTOM, _mask, _value, 0),	\
+	{.action = (_decoder)}
+
+
+struct decode_simulate {
+	struct decode_header	header;
+	union decode_item	handler;
+};
+
+#define DECODE_SIMULATEX(_mask, _value, _handler, _regs)		\
+	DECODE_HEADER(DECODE_TYPE_SIMULATE, _mask, _value, _regs),	\
+	{.action = (_handler)}
+
+#define DECODE_SIMULATE(_mask, _value, _handler)	\
+	DECODE_SIMULATEX(_mask, _value, _handler, 0)
+
+
+struct decode_emulate {
+	struct decode_header	header;
+	union decode_item	handler;
+};
+
+#define DECODE_EMULATEX(_mask, _value, _handler, _regs)			\
+	DECODE_HEADER(DECODE_TYPE_EMULATE, _mask, _value, _regs),	\
+	{.action = (_handler)}
+
+#define DECODE_EMULATE(_mask, _value, _handler)		\
+	DECODE_EMULATEX(_mask, _value, _handler, 0)
+
+
+struct decode_or {
+	struct decode_header	header;
+};
+
+#define DECODE_OR(_mask, _value)				\
+	DECODE_HEADER(DECODE_TYPE_OR, _mask, _value, 0)
+
+enum probes_insn {
+	INSN_REJECTED,
+	INSN_GOOD,
+	INSN_GOOD_NO_SLOT
+};
+
+struct decode_reject {
+	struct decode_header	header;
+};
+
+#define DECODE_REJECT(_mask, _value)				\
+	DECODE_HEADER(DECODE_TYPE_REJECT, _mask, _value, 0)
+
+probes_insn_handler_t probes_simulate_nop;
+probes_insn_handler_t probes_emulate_none;
+
+int __kprobes
+probes_decode_insn(probes_opcode_t insn, struct arch_probes_insn *asi,
+		const union decode_item *table, bool thumb, bool emulate,
+		const union decode_action *actions,
+		const struct decode_checker **checkers);
+
+#endif

arch/arm/probes/kprobes/Makefile

@@ -0,0 +1,12 @@
+obj-$(CONFIG_KPROBES)		+= core.o actions-common.o checkers-common.o
+obj-$(CONFIG_ARM_KPROBES_TEST)	+= test-kprobes.o
+test-kprobes-objs		:= test-core.o
+
+ifdef CONFIG_THUMB2_KERNEL
+obj-$(CONFIG_KPROBES)		+= actions-thumb.o checkers-thumb.o
+test-kprobes-objs		+= test-thumb.o
+else
+obj-$(CONFIG_KPROBES)		+= actions-arm.o checkers-arm.o
+obj-$(CONFIG_OPTPROBES)		+= opt-arm.o
+test-kprobes-objs		+= test-arm.o
+endif

arch/arm/probes/kprobes/actions-arm.c

@@ -0,0 +1,344 @@
+/*
+ * arch/arm/probes/kprobes/actions-arm.c
+ *
+ * Copyright (C) 2006, 2007 Motorola Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+/*
+ * We do not have hardware single-stepping on ARM, This
+ * effort is further complicated by the ARM not having a
+ * "next PC" register.  Instructions that change the PC
+ * can't be safely single-stepped in a MP environment, so
+ * we have a lot of work to do:
+ *
+ * In the prepare phase:
+ *   *) If it is an instruction that does anything
+ *      with the CPU mode, we reject it for a kprobe.
+ *      (This is out of laziness rather than need.  The
+ *      instructions could be simulated.)
+ *
+ *   *) Otherwise, decode the instruction rewriting its
+ *      registers to take fixed, ordered registers and
+ *      setting a handler for it to run the instruction.
+ *
+ * In the execution phase by an instruction's handler:
+ *
+ *   *) If the PC is written to by the instruction, the
+ *      instruction must be fully simulated in software.
+ *
+ *   *) Otherwise, a modified form of the instruction is
+ *      directly executed.  Its handler calls the
+ *      instruction in insn[0].  In insn[1] is a
+ *      "mov pc, lr" to return.
+ *
+ *      Before calling, load up the reordered registers
+ *      from the original instruction's registers.  If one
+ *      of the original input registers is the PC, compute
+ *      and adjust the appropriate input register.
+ *
+ *	After call completes, copy the output registers to
+ *      the original instruction's original registers.
+ *
+ * We don't use a real breakpoint instruction since that
+ * would have us in the kernel go from SVC mode to SVC
+ * mode losing the link register.  Instead we use an
+ * undefined instruction.  To simplify processing, the
+ * undefined instruction used for kprobes must be reserved
+ * exclusively for kprobes use.
+ *
+ * TODO: ifdef out some instruction decoding based on architecture.
+ */
+
+#include <linux/kernel.h>
+#include <linux/kprobes.h>
+#include <linux/ptrace.h>
+
+#include "../decode-arm.h"
+#include "core.h"
+#include "checkers.h"
+
+#if  __LINUX_ARM_ARCH__ >= 6
+#define BLX(reg)	"blx	"reg"		\n\t"
+#else
+#define BLX(reg)	"mov	lr, pc		\n\t"	\
+			"mov	pc, "reg"	\n\t"
+#endif
+
+static void __kprobes
+emulate_ldrdstrd(probes_opcode_t insn,
+	struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long pc = regs->ARM_pc + 4;
+	int rt = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	int rm = insn & 0xf;
+
+	register unsigned long rtv asm("r0") = regs->uregs[rt];
+	register unsigned long rt2v asm("r1") = regs->uregs[rt+1];
+	register unsigned long rnv asm("r2") = (rn == 15) ? pc
+							  : regs->uregs[rn];
+	register unsigned long rmv asm("r3") = regs->uregs[rm];
+
+	__asm__ __volatile__ (
+		BLX("%[fn]")
+		: "=r" (rtv), "=r" (rt2v), "=r" (rnv)
+		: "0" (rtv), "1" (rt2v), "2" (rnv), "r" (rmv),
+		  [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	regs->uregs[rt] = rtv;
+	regs->uregs[rt+1] = rt2v;
+	if (is_writeback(insn))
+		regs->uregs[rn] = rnv;
+}
+
+static void __kprobes
+emulate_ldr(probes_opcode_t insn,
+	struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long pc = regs->ARM_pc + 4;
+	int rt = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	int rm = insn & 0xf;
+
+	register unsigned long rtv asm("r0");
+	register unsigned long rnv asm("r2") = (rn == 15) ? pc
+							  : regs->uregs[rn];
+	register unsigned long rmv asm("r3") = regs->uregs[rm];
+
+	__asm__ __volatile__ (
+		BLX("%[fn]")
+		: "=r" (rtv), "=r" (rnv)
+		: "1" (rnv), "r" (rmv), [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	if (rt == 15)
+		load_write_pc(rtv, regs);
+	else
+		regs->uregs[rt] = rtv;
+
+	if (is_writeback(insn))
+		regs->uregs[rn] = rnv;
+}
+
+static void __kprobes
+emulate_str(probes_opcode_t insn,
+	struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long rtpc = regs->ARM_pc - 4 + str_pc_offset;
+	unsigned long rnpc = regs->ARM_pc + 4;
+	int rt = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	int rm = insn & 0xf;
+
+	register unsigned long rtv asm("r0") = (rt == 15) ? rtpc
+							  : regs->uregs[rt];
+	register unsigned long rnv asm("r2") = (rn == 15) ? rnpc
+							  : regs->uregs[rn];
+	register unsigned long rmv asm("r3") = regs->uregs[rm];
+
+	__asm__ __volatile__ (
+		BLX("%[fn]")
+		: "=r" (rnv)
+		: "r" (rtv), "0" (rnv), "r" (rmv), [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	if (is_writeback(insn))
+		regs->uregs[rn] = rnv;
+}
+
+static void __kprobes
+emulate_rd12rn16rm0rs8_rwflags(probes_opcode_t insn,
+	struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long pc = regs->ARM_pc + 4;
+	int rd = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	int rm = insn & 0xf;
+	int rs = (insn >> 8) & 0xf;
+
+	register unsigned long rdv asm("r0") = regs->uregs[rd];
+	register unsigned long rnv asm("r2") = (rn == 15) ? pc
+							  : regs->uregs[rn];
+	register unsigned long rmv asm("r3") = (rm == 15) ? pc
+							  : regs->uregs[rm];
+	register unsigned long rsv asm("r1") = regs->uregs[rs];
+	unsigned long cpsr = regs->ARM_cpsr;
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[cpsr]	\n\t"
+		BLX("%[fn]")
+		"mrs	%[cpsr], cpsr		\n\t"
+		: "=r" (rdv), [cpsr] "=r" (cpsr)
+		: "0" (rdv), "r" (rnv), "r" (rmv), "r" (rsv),
+		  "1" (cpsr), [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	if (rd == 15)
+		alu_write_pc(rdv, regs);
+	else
+		regs->uregs[rd] = rdv;
+	regs->ARM_cpsr = (regs->ARM_cpsr & ~APSR_MASK) | (cpsr & APSR_MASK);
+}
+
+static void __kprobes
+emulate_rd12rn16rm0_rwflags_nopc(probes_opcode_t insn,
+	struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	int rd = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	int rm = insn & 0xf;
+
+	register unsigned long rdv asm("r0") = regs->uregs[rd];
+	register unsigned long rnv asm("r2") = regs->uregs[rn];
+	register unsigned long rmv asm("r3") = regs->uregs[rm];
+	unsigned long cpsr = regs->ARM_cpsr;
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[cpsr]	\n\t"
+		BLX("%[fn]")
+		"mrs	%[cpsr], cpsr		\n\t"
+		: "=r" (rdv), [cpsr] "=r" (cpsr)
+		: "0" (rdv), "r" (rnv), "r" (rmv),
+		  "1" (cpsr), [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	regs->uregs[rd] = rdv;
+	regs->ARM_cpsr = (regs->ARM_cpsr & ~APSR_MASK) | (cpsr & APSR_MASK);
+}
+
+static void __kprobes
+emulate_rd16rn12rm0rs8_rwflags_nopc(probes_opcode_t insn,
+	struct arch_probes_insn *asi,
+	struct pt_regs *regs)
+{
+	int rd = (insn >> 16) & 0xf;
+	int rn = (insn >> 12) & 0xf;
+	int rm = insn & 0xf;
+	int rs = (insn >> 8) & 0xf;
+
+	register unsigned long rdv asm("r2") = regs->uregs[rd];
+	register unsigned long rnv asm("r0") = regs->uregs[rn];
+	register unsigned long rmv asm("r3") = regs->uregs[rm];
+	register unsigned long rsv asm("r1") = regs->uregs[rs];
+	unsigned long cpsr = regs->ARM_cpsr;
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[cpsr]	\n\t"
+		BLX("%[fn]")
+		"mrs	%[cpsr], cpsr		\n\t"
+		: "=r" (rdv), [cpsr] "=r" (cpsr)
+		: "0" (rdv), "r" (rnv), "r" (rmv), "r" (rsv),
+		  "1" (cpsr), [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	regs->uregs[rd] = rdv;
+	regs->ARM_cpsr = (regs->ARM_cpsr & ~APSR_MASK) | (cpsr & APSR_MASK);
+}
+
+static void __kprobes
+emulate_rd12rm0_noflags_nopc(probes_opcode_t insn,
+	struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	int rd = (insn >> 12) & 0xf;
+	int rm = insn & 0xf;
+
+	register unsigned long rdv asm("r0") = regs->uregs[rd];
+	register unsigned long rmv asm("r3") = regs->uregs[rm];
+
+	__asm__ __volatile__ (
+		BLX("%[fn]")
+		: "=r" (rdv)
+		: "0" (rdv), "r" (rmv), [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	regs->uregs[rd] = rdv;
+}
+
+static void __kprobes
+emulate_rdlo12rdhi16rn0rm8_rwflags_nopc(probes_opcode_t insn,
+	struct arch_probes_insn *asi,
+	struct pt_regs *regs)
+{
+	int rdlo = (insn >> 12) & 0xf;
+	int rdhi = (insn >> 16) & 0xf;
+	int rn = insn & 0xf;
+	int rm = (insn >> 8) & 0xf;
+
+	register unsigned long rdlov asm("r0") = regs->uregs[rdlo];
+	register unsigned long rdhiv asm("r2") = regs->uregs[rdhi];
+	register unsigned long rnv asm("r3") = regs->uregs[rn];
+	register unsigned long rmv asm("r1") = regs->uregs[rm];
+	unsigned long cpsr = regs->ARM_cpsr;
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[cpsr]	\n\t"
+		BLX("%[fn]")
+		"mrs	%[cpsr], cpsr		\n\t"
+		: "=r" (rdlov), "=r" (rdhiv), [cpsr] "=r" (cpsr)
+		: "0" (rdlov), "1" (rdhiv), "r" (rnv), "r" (rmv),
+		  "2" (cpsr), [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	regs->uregs[rdlo] = rdlov;
+	regs->uregs[rdhi] = rdhiv;
+	regs->ARM_cpsr = (regs->ARM_cpsr & ~APSR_MASK) | (cpsr & APSR_MASK);
+}
+
+const union decode_action kprobes_arm_actions[NUM_PROBES_ARM_ACTIONS] = {
+	[PROBES_PRELOAD_IMM] = {.handler = probes_simulate_nop},
+	[PROBES_PRELOAD_REG] = {.handler = probes_simulate_nop},
+	[PROBES_BRANCH_IMM] = {.handler = simulate_blx1},
+	[PROBES_MRS] = {.handler = simulate_mrs},
+	[PROBES_BRANCH_REG] = {.handler = simulate_blx2bx},
+	[PROBES_CLZ] = {.handler = emulate_rd12rm0_noflags_nopc},
+	[PROBES_SATURATING_ARITHMETIC] = {
+		.handler = emulate_rd12rn16rm0_rwflags_nopc},
+	[PROBES_MUL1] = {.handler = emulate_rdlo12rdhi16rn0rm8_rwflags_nopc},
+	[PROBES_MUL2] = {.handler = emulate_rd16rn12rm0rs8_rwflags_nopc},
+	[PROBES_SWP] = {.handler = emulate_rd12rn16rm0_rwflags_nopc},
+	[PROBES_LDRSTRD] = {.handler = emulate_ldrdstrd},
+	[PROBES_LOAD_EXTRA] = {.handler = emulate_ldr},
+	[PROBES_LOAD] = {.handler = emulate_ldr},
+	[PROBES_STORE_EXTRA] = {.handler = emulate_str},
+	[PROBES_STORE] = {.handler = emulate_str},
+	[PROBES_MOV_IP_SP] = {.handler = simulate_mov_ipsp},
+	[PROBES_DATA_PROCESSING_REG] = {
+		.handler = emulate_rd12rn16rm0rs8_rwflags},
+	[PROBES_DATA_PROCESSING_IMM] = {
+		.handler = emulate_rd12rn16rm0rs8_rwflags},
+	[PROBES_MOV_HALFWORD] = {.handler = emulate_rd12rm0_noflags_nopc},
+	[PROBES_SEV] = {.handler = probes_emulate_none},
+	[PROBES_WFE] = {.handler = probes_simulate_nop},
+	[PROBES_SATURATE] = {.handler = emulate_rd12rn16rm0_rwflags_nopc},
+	[PROBES_REV] = {.handler = emulate_rd12rm0_noflags_nopc},
+	[PROBES_MMI] = {.handler = emulate_rd12rn16rm0_rwflags_nopc},
+	[PROBES_PACK] = {.handler = emulate_rd12rn16rm0_rwflags_nopc},
+	[PROBES_EXTEND] = {.handler = emulate_rd12rm0_noflags_nopc},
+	[PROBES_EXTEND_ADD] = {.handler = emulate_rd12rn16rm0_rwflags_nopc},
+	[PROBES_MUL_ADD_LONG] = {
+		.handler = emulate_rdlo12rdhi16rn0rm8_rwflags_nopc},
+	[PROBES_MUL_ADD] = {.handler = emulate_rd16rn12rm0rs8_rwflags_nopc},
+	[PROBES_BITFIELD] = {.handler = emulate_rd12rm0_noflags_nopc},
+	[PROBES_BRANCH] = {.handler = simulate_bbl},
+	[PROBES_LDMSTM] = {.decoder = kprobe_decode_ldmstm}
+};
+
+const struct decode_checker *kprobes_arm_checkers[] = {arm_stack_checker, arm_regs_checker, NULL};

arch/arm/probes/kprobes/actions-common.c

@@ -0,0 +1,171 @@
+/*
+ * arch/arm/probes/kprobes/actions-common.c
+ *
+ * Copyright (C) 2011 Jon Medhurst <tixy@yxit.co.uk>.
+ *
+ * Some contents moved here from arch/arm/include/asm/kprobes-arm.c which is
+ * Copyright (C) 2006, 2007 Motorola Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/kprobes.h>
+#include <asm/opcodes.h>
+
+#include "core.h"
+
+
+static void __kprobes simulate_ldm1stm1(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		struct pt_regs *regs)
+{
+	int rn = (insn >> 16) & 0xf;
+	int lbit = insn & (1 << 20);
+	int wbit = insn & (1 << 21);
+	int ubit = insn & (1 << 23);
+	int pbit = insn & (1 << 24);
+	long *addr = (long *)regs->uregs[rn];
+	int reg_bit_vector;
+	int reg_count;
+
+	reg_count = 0;
+	reg_bit_vector = insn & 0xffff;
+	while (reg_bit_vector) {
+		reg_bit_vector &= (reg_bit_vector - 1);
+		++reg_count;
+	}
+
+	if (!ubit)
+		addr -= reg_count;
+	addr += (!pbit == !ubit);
+
+	reg_bit_vector = insn & 0xffff;
+	while (reg_bit_vector) {
+		int reg = __ffs(reg_bit_vector);
+		reg_bit_vector &= (reg_bit_vector - 1);
+		if (lbit)
+			regs->uregs[reg] = *addr++;
+		else
+			*addr++ = regs->uregs[reg];
+	}
+
+	if (wbit) {
+		if (!ubit)
+			addr -= reg_count;
+		addr -= (!pbit == !ubit);
+		regs->uregs[rn] = (long)addr;
+	}
+}
+
+static void __kprobes simulate_stm1_pc(probes_opcode_t insn,
+	struct arch_probes_insn *asi,
+	struct pt_regs *regs)
+{
+	unsigned long addr = regs->ARM_pc - 4;
+
+	regs->ARM_pc = (long)addr + str_pc_offset;
+	simulate_ldm1stm1(insn, asi, regs);
+	regs->ARM_pc = (long)addr + 4;
+}
+
+static void __kprobes simulate_ldm1_pc(probes_opcode_t insn,
+	struct arch_probes_insn *asi,
+	struct pt_regs *regs)
+{
+	simulate_ldm1stm1(insn, asi, regs);
+	load_write_pc(regs->ARM_pc, regs);
+}
+
+static void __kprobes
+emulate_generic_r0_12_noflags(probes_opcode_t insn,
+	struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	register void *rregs asm("r1") = regs;
+	register void *rfn asm("lr") = asi->insn_fn;
+
+	__asm__ __volatile__ (
+		"stmdb	sp!, {%[regs], r11}	\n\t"
+		"ldmia	%[regs], {r0-r12}	\n\t"
+#if __LINUX_ARM_ARCH__ >= 6
+		"blx	%[fn]			\n\t"
+#else
+		"str	%[fn], [sp, #-4]!	\n\t"
+		"adr	lr, 1f			\n\t"
+		"ldr	pc, [sp], #4		\n\t"
+		"1:				\n\t"
+#endif
+		"ldr	lr, [sp], #4		\n\t" /* lr = regs */
+		"stmia	lr, {r0-r12}		\n\t"
+		"ldr	r11, [sp], #4		\n\t"
+		: [regs] "=r" (rregs), [fn] "=r" (rfn)
+		: "0" (rregs), "1" (rfn)
+		: "r0", "r2", "r3", "r4", "r5", "r6", "r7",
+		  "r8", "r9", "r10", "r12", "memory", "cc"
+		);
+}
+
+static void __kprobes
+emulate_generic_r2_14_noflags(probes_opcode_t insn,
+	struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	emulate_generic_r0_12_noflags(insn, asi,
+		(struct pt_regs *)(regs->uregs+2));
+}
+
+static void __kprobes
+emulate_ldm_r3_15(probes_opcode_t insn,
+	struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	emulate_generic_r0_12_noflags(insn, asi,
+		(struct pt_regs *)(regs->uregs+3));
+	load_write_pc(regs->ARM_pc, regs);
+}
+
+enum probes_insn __kprobes
+kprobe_decode_ldmstm(probes_opcode_t insn, struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	probes_insn_handler_t *handler = 0;
+	unsigned reglist = insn & 0xffff;
+	int is_ldm = insn & 0x100000;
+	int rn = (insn >> 16) & 0xf;
+
+	if (rn <= 12 && (reglist & 0xe000) == 0) {
+		/* Instruction only uses registers in the range R0..R12 */
+		handler = emulate_generic_r0_12_noflags;
+
+	} else if (rn >= 2 && (reglist & 0x8003) == 0) {
+		/* Instruction only uses registers in the range R2..R14 */
+		rn -= 2;
+		reglist >>= 2;
+		handler = emulate_generic_r2_14_noflags;
+
+	} else if (rn >= 3 && (reglist & 0x0007) == 0) {
+		/* Instruction only uses registers in the range R3..R15 */
+		if (is_ldm && (reglist & 0x8000)) {
+			rn -= 3;
+			reglist >>= 3;
+			handler = emulate_ldm_r3_15;
+		}
+	}
+
+	if (handler) {
+		/* We can emulate the instruction in (possibly) modified form */
+		asi->insn[0] = __opcode_to_mem_arm((insn & 0xfff00000) |
+						   (rn << 16) | reglist);
+		asi->insn_handler = handler;
+		return INSN_GOOD;
+	}
+
+	/* Fallback to slower simulation... */
+	if (reglist & 0x8000)
+		handler = is_ldm ? simulate_ldm1_pc : simulate_stm1_pc;
+	else
+		handler = simulate_ldm1stm1;
+	asi->insn_handler = handler;
+	return INSN_GOOD_NO_SLOT;
+}
+

arch/arm/probes/kprobes/actions-thumb.c

@@ -0,0 +1,670 @@
+/*
+ * arch/arm/probes/kprobes/actions-thumb.c
+ *
+ * Copyright (C) 2011 Jon Medhurst <tixy@yxit.co.uk>.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/ptrace.h>
+#include <linux/kprobes.h>
+
+#include "../decode-thumb.h"
+#include "core.h"
+#include "checkers.h"
+
+/* These emulation encodings are functionally equivalent... */
+#define t32_emulate_rd8rn16rm0ra12_noflags \
+		t32_emulate_rdlo12rdhi8rn16rm0_noflags
+
+/* t32 thumb actions */
+
+static void __kprobes
+t32_simulate_table_branch(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long pc = regs->ARM_pc;
+	int rn = (insn >> 16) & 0xf;
+	int rm = insn & 0xf;
+
+	unsigned long rnv = (rn == 15) ? pc : regs->uregs[rn];
+	unsigned long rmv = regs->uregs[rm];
+	unsigned int halfwords;
+
+	if (insn & 0x10) /* TBH */
+		halfwords = ((u16 *)rnv)[rmv];
+	else /* TBB */
+		halfwords = ((u8 *)rnv)[rmv];
+
+	regs->ARM_pc = pc + 2 * halfwords;
+}
+
+static void __kprobes
+t32_simulate_mrs(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	int rd = (insn >> 8) & 0xf;
+	unsigned long mask = 0xf8ff03df; /* Mask out execution state */
+	regs->uregs[rd] = regs->ARM_cpsr & mask;
+}
+
+static void __kprobes
+t32_simulate_cond_branch(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long pc = regs->ARM_pc;
+
+	long offset = insn & 0x7ff;		/* imm11 */
+	offset += (insn & 0x003f0000) >> 5;	/* imm6 */
+	offset += (insn & 0x00002000) << 4;	/* J1 */
+	offset += (insn & 0x00000800) << 7;	/* J2 */
+	offset -= (insn & 0x04000000) >> 7;	/* Apply sign bit */
+
+	regs->ARM_pc = pc + (offset * 2);
+}
+
+static enum probes_insn __kprobes
+t32_decode_cond_branch(probes_opcode_t insn, struct arch_probes_insn *asi,
+		const struct decode_header *d)
+{
+	int cc = (insn >> 22) & 0xf;
+	asi->insn_check_cc = probes_condition_checks[cc];
+	asi->insn_handler = t32_simulate_cond_branch;
+	return INSN_GOOD_NO_SLOT;
+}
+
+static void __kprobes
+t32_simulate_branch(probes_opcode_t insn,
+		    struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long pc = regs->ARM_pc;
+
+	long offset = insn & 0x7ff;		/* imm11 */
+	offset += (insn & 0x03ff0000) >> 5;	/* imm10 */
+	offset += (insn & 0x00002000) << 9;	/* J1 */
+	offset += (insn & 0x00000800) << 10;	/* J2 */
+	if (insn & 0x04000000)
+		offset -= 0x00800000; /* Apply sign bit */
+	else
+		offset ^= 0x00600000; /* Invert J1 and J2 */
+
+	if (insn & (1 << 14)) {
+		/* BL or BLX */
+		regs->ARM_lr = regs->ARM_pc | 1;
+		if (!(insn & (1 << 12))) {
+			/* BLX so switch to ARM mode */
+			regs->ARM_cpsr &= ~PSR_T_BIT;
+			pc &= ~3;
+		}
+	}
+
+	regs->ARM_pc = pc + (offset * 2);
+}
+
+static void __kprobes
+t32_simulate_ldr_literal(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long addr = regs->ARM_pc & ~3;
+	int rt = (insn >> 12) & 0xf;
+	unsigned long rtv;
+
+	long offset = insn & 0xfff;
+	if (insn & 0x00800000)
+		addr += offset;
+	else
+		addr -= offset;
+
+	if (insn & 0x00400000) {
+		/* LDR */
+		rtv = *(unsigned long *)addr;
+		if (rt == 15) {
+			bx_write_pc(rtv, regs);
+			return;
+		}
+	} else if (insn & 0x00200000) {
+		/* LDRH */
+		if (insn & 0x01000000)
+			rtv = *(s16 *)addr;
+		else
+			rtv = *(u16 *)addr;
+	} else {
+		/* LDRB */
+		if (insn & 0x01000000)
+			rtv = *(s8 *)addr;
+		else
+			rtv = *(u8 *)addr;
+	}
+
+	regs->uregs[rt] = rtv;
+}
+
+static enum probes_insn __kprobes
+t32_decode_ldmstm(probes_opcode_t insn, struct arch_probes_insn *asi,
+		const struct decode_header *d)
+{
+	enum probes_insn ret = kprobe_decode_ldmstm(insn, asi, d);
+
+	/* Fixup modified instruction to have halfwords in correct order...*/
+	insn = __mem_to_opcode_arm(asi->insn[0]);
+	((u16 *)asi->insn)[0] = __opcode_to_mem_thumb16(insn >> 16);
+	((u16 *)asi->insn)[1] = __opcode_to_mem_thumb16(insn & 0xffff);
+
+	return ret;
+}
+
+static void __kprobes
+t32_emulate_ldrdstrd(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long pc = regs->ARM_pc & ~3;
+	int rt1 = (insn >> 12) & 0xf;
+	int rt2 = (insn >> 8) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+
+	register unsigned long rt1v asm("r0") = regs->uregs[rt1];
+	register unsigned long rt2v asm("r1") = regs->uregs[rt2];
+	register unsigned long rnv asm("r2") = (rn == 15) ? pc
+							  : regs->uregs[rn];
+
+	__asm__ __volatile__ (
+		"blx    %[fn]"
+		: "=r" (rt1v), "=r" (rt2v), "=r" (rnv)
+		: "0" (rt1v), "1" (rt2v), "2" (rnv), [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	if (rn != 15)
+		regs->uregs[rn] = rnv; /* Writeback base register */
+	regs->uregs[rt1] = rt1v;
+	regs->uregs[rt2] = rt2v;
+}
+
+static void __kprobes
+t32_emulate_ldrstr(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	int rt = (insn >> 12) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	int rm = insn & 0xf;
+
+	register unsigned long rtv asm("r0") = regs->uregs[rt];
+	register unsigned long rnv asm("r2") = regs->uregs[rn];
+	register unsigned long rmv asm("r3") = regs->uregs[rm];
+
+	__asm__ __volatile__ (
+		"blx    %[fn]"
+		: "=r" (rtv), "=r" (rnv)
+		: "0" (rtv), "1" (rnv), "r" (rmv), [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	regs->uregs[rn] = rnv; /* Writeback base register */
+	if (rt == 15) /* Can't be true for a STR as they aren't allowed */
+		bx_write_pc(rtv, regs);
+	else
+		regs->uregs[rt] = rtv;
+}
+
+static void __kprobes
+t32_emulate_rd8rn16rm0_rwflags(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	int rd = (insn >> 8) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	int rm = insn & 0xf;
+
+	register unsigned long rdv asm("r1") = regs->uregs[rd];
+	register unsigned long rnv asm("r2") = regs->uregs[rn];
+	register unsigned long rmv asm("r3") = regs->uregs[rm];
+	unsigned long cpsr = regs->ARM_cpsr;
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[cpsr]	\n\t"
+		"blx    %[fn]			\n\t"
+		"mrs	%[cpsr], cpsr		\n\t"
+		: "=r" (rdv), [cpsr] "=r" (cpsr)
+		: "0" (rdv), "r" (rnv), "r" (rmv),
+		  "1" (cpsr), [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	regs->uregs[rd] = rdv;
+	regs->ARM_cpsr = (regs->ARM_cpsr & ~APSR_MASK) | (cpsr & APSR_MASK);
+}
+
+static void __kprobes
+t32_emulate_rd8pc16_noflags(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long pc = regs->ARM_pc;
+	int rd = (insn >> 8) & 0xf;
+
+	register unsigned long rdv asm("r1") = regs->uregs[rd];
+	register unsigned long rnv asm("r2") = pc & ~3;
+
+	__asm__ __volatile__ (
+		"blx    %[fn]"
+		: "=r" (rdv)
+		: "0" (rdv), "r" (rnv), [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	regs->uregs[rd] = rdv;
+}
+
+static void __kprobes
+t32_emulate_rd8rn16_noflags(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	int rd = (insn >> 8) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+
+	register unsigned long rdv asm("r1") = regs->uregs[rd];
+	register unsigned long rnv asm("r2") = regs->uregs[rn];
+
+	__asm__ __volatile__ (
+		"blx    %[fn]"
+		: "=r" (rdv)
+		: "0" (rdv), "r" (rnv), [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	regs->uregs[rd] = rdv;
+}
+
+static void __kprobes
+t32_emulate_rdlo12rdhi8rn16rm0_noflags(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		struct pt_regs *regs)
+{
+	int rdlo = (insn >> 12) & 0xf;
+	int rdhi = (insn >> 8) & 0xf;
+	int rn = (insn >> 16) & 0xf;
+	int rm = insn & 0xf;
+
+	register unsigned long rdlov asm("r0") = regs->uregs[rdlo];
+	register unsigned long rdhiv asm("r1") = regs->uregs[rdhi];
+	register unsigned long rnv asm("r2") = regs->uregs[rn];
+	register unsigned long rmv asm("r3") = regs->uregs[rm];
+
+	__asm__ __volatile__ (
+		"blx    %[fn]"
+		: "=r" (rdlov), "=r" (rdhiv)
+		: "0" (rdlov), "1" (rdhiv), "r" (rnv), "r" (rmv),
+		  [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	regs->uregs[rdlo] = rdlov;
+	regs->uregs[rdhi] = rdhiv;
+}
+/* t16 thumb actions */
+
+static void __kprobes
+t16_simulate_bxblx(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long pc = regs->ARM_pc + 2;
+	int rm = (insn >> 3) & 0xf;
+	unsigned long rmv = (rm == 15) ? pc : regs->uregs[rm];
+
+	if (insn & (1 << 7)) /* BLX ? */
+		regs->ARM_lr = regs->ARM_pc | 1;
+
+	bx_write_pc(rmv, regs);
+}
+
+static void __kprobes
+t16_simulate_ldr_literal(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long *base = (unsigned long *)((regs->ARM_pc + 2) & ~3);
+	long index = insn & 0xff;
+	int rt = (insn >> 8) & 0x7;
+	regs->uregs[rt] = base[index];
+}
+
+static void __kprobes
+t16_simulate_ldrstr_sp_relative(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long* base = (unsigned long *)regs->ARM_sp;
+	long index = insn & 0xff;
+	int rt = (insn >> 8) & 0x7;
+	if (insn & 0x800) /* LDR */
+		regs->uregs[rt] = base[index];
+	else /* STR */
+		base[index] = regs->uregs[rt];
+}
+
+static void __kprobes
+t16_simulate_reladr(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long base = (insn & 0x800) ? regs->ARM_sp
+					    : ((regs->ARM_pc + 2) & ~3);
+	long offset = insn & 0xff;
+	int rt = (insn >> 8) & 0x7;
+	regs->uregs[rt] = base + offset * 4;
+}
+
+static void __kprobes
+t16_simulate_add_sp_imm(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	long imm = insn & 0x7f;
+	if (insn & 0x80) /* SUB */
+		regs->ARM_sp -= imm * 4;
+	else /* ADD */
+		regs->ARM_sp += imm * 4;
+}
+
+static void __kprobes
+t16_simulate_cbz(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	int rn = insn & 0x7;
+	probes_opcode_t nonzero = regs->uregs[rn] ? insn : ~insn;
+	if (nonzero & 0x800) {
+		long i = insn & 0x200;
+		long imm5 = insn & 0xf8;
+		unsigned long pc = regs->ARM_pc + 2;
+		regs->ARM_pc = pc + (i >> 3) + (imm5 >> 2);
+	}
+}
+
+static void __kprobes
+t16_simulate_it(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	/*
+	 * The 8 IT state bits are split into two parts in CPSR:
+	 *	ITSTATE<1:0> are in CPSR<26:25>
+	 *	ITSTATE<7:2> are in CPSR<15:10>
+	 * The new IT state is in the lower byte of insn.
+	 */
+	unsigned long cpsr = regs->ARM_cpsr;
+	cpsr &= ~PSR_IT_MASK;
+	cpsr |= (insn & 0xfc) << 8;
+	cpsr |= (insn & 0x03) << 25;
+	regs->ARM_cpsr = cpsr;
+}
+
+static void __kprobes
+t16_singlestep_it(probes_opcode_t insn,
+		  struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	regs->ARM_pc += 2;
+	t16_simulate_it(insn, asi, regs);
+}
+
+static enum probes_insn __kprobes
+t16_decode_it(probes_opcode_t insn, struct arch_probes_insn *asi,
+		const struct decode_header *d)
+{
+	asi->insn_singlestep = t16_singlestep_it;
+	return INSN_GOOD_NO_SLOT;
+}
+
+static void __kprobes
+t16_simulate_cond_branch(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long pc = regs->ARM_pc + 2;
+	long offset = insn & 0x7f;
+	offset -= insn & 0x80; /* Apply sign bit */
+	regs->ARM_pc = pc + (offset * 2);
+}
+
+static enum probes_insn __kprobes
+t16_decode_cond_branch(probes_opcode_t insn, struct arch_probes_insn *asi,
+		const struct decode_header *d)
+{
+	int cc = (insn >> 8) & 0xf;
+	asi->insn_check_cc = probes_condition_checks[cc];
+	asi->insn_handler = t16_simulate_cond_branch;
+	return INSN_GOOD_NO_SLOT;
+}
+
+static void __kprobes
+t16_simulate_branch(probes_opcode_t insn,
+		   struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long pc = regs->ARM_pc + 2;
+	long offset = insn & 0x3ff;
+	offset -= insn & 0x400; /* Apply sign bit */
+	regs->ARM_pc = pc + (offset * 2);
+}
+
+static unsigned long __kprobes
+t16_emulate_loregs(probes_opcode_t insn,
+		   struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long oldcpsr = regs->ARM_cpsr;
+	unsigned long newcpsr;
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[oldcpsr]	\n\t"
+		"ldmia	%[regs], {r0-r7}	\n\t"
+		"blx	%[fn]			\n\t"
+		"stmia	%[regs], {r0-r7}	\n\t"
+		"mrs	%[newcpsr], cpsr	\n\t"
+		: [newcpsr] "=r" (newcpsr)
+		: [oldcpsr] "r" (oldcpsr), [regs] "r" (regs),
+		  [fn] "r" (asi->insn_fn)
+		: "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7",
+		  "lr", "memory", "cc"
+		);
+
+	return (oldcpsr & ~APSR_MASK) | (newcpsr & APSR_MASK);
+}
+
+static void __kprobes
+t16_emulate_loregs_rwflags(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	regs->ARM_cpsr = t16_emulate_loregs(insn, asi, regs);
+}
+
+static void __kprobes
+t16_emulate_loregs_noitrwflags(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long cpsr = t16_emulate_loregs(insn, asi, regs);
+	if (!in_it_block(cpsr))
+		regs->ARM_cpsr = cpsr;
+}
+
+static void __kprobes
+t16_emulate_hiregs(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	unsigned long pc = regs->ARM_pc + 2;
+	int rdn = (insn & 0x7) | ((insn & 0x80) >> 4);
+	int rm = (insn >> 3) & 0xf;
+
+	register unsigned long rdnv asm("r1");
+	register unsigned long rmv asm("r0");
+	unsigned long cpsr = regs->ARM_cpsr;
+
+	rdnv = (rdn == 15) ? pc : regs->uregs[rdn];
+	rmv = (rm == 15) ? pc : regs->uregs[rm];
+
+	__asm__ __volatile__ (
+		"msr	cpsr_fs, %[cpsr]	\n\t"
+		"blx    %[fn]			\n\t"
+		"mrs	%[cpsr], cpsr		\n\t"
+		: "=r" (rdnv), [cpsr] "=r" (cpsr)
+		: "0" (rdnv), "r" (rmv), "1" (cpsr), [fn] "r" (asi->insn_fn)
+		: "lr", "memory", "cc"
+	);
+
+	if (rdn == 15)
+		rdnv &= ~1;
+
+	regs->uregs[rdn] = rdnv;
+	regs->ARM_cpsr = (regs->ARM_cpsr & ~APSR_MASK) | (cpsr & APSR_MASK);
+}
+
+static enum probes_insn __kprobes
+t16_decode_hiregs(probes_opcode_t insn, struct arch_probes_insn *asi,
+		const struct decode_header *d)
+{
+	insn &= ~0x00ff;
+	insn |= 0x001; /* Set Rdn = R1 and Rm = R0 */
+	((u16 *)asi->insn)[0] = __opcode_to_mem_thumb16(insn);
+	asi->insn_handler = t16_emulate_hiregs;
+	return INSN_GOOD;
+}
+
+static void __kprobes
+t16_emulate_push(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	__asm__ __volatile__ (
+		"ldr	r9, [%[regs], #13*4]	\n\t"
+		"ldr	r8, [%[regs], #14*4]	\n\t"
+		"ldmia	%[regs], {r0-r7}	\n\t"
+		"blx	%[fn]			\n\t"
+		"str	r9, [%[regs], #13*4]	\n\t"
+		:
+		: [regs] "r" (regs), [fn] "r" (asi->insn_fn)
+		: "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9",
+		  "lr", "memory", "cc"
+		);
+}
+
+static enum probes_insn __kprobes
+t16_decode_push(probes_opcode_t insn, struct arch_probes_insn *asi,
+		const struct decode_header *d)
+{
+	/*
+	 * To simulate a PUSH we use a Thumb-2 "STMDB R9!, {registers}"
+	 * and call it with R9=SP and LR in the register list represented
+	 * by R8.
+	 */
+	/* 1st half STMDB R9!,{} */
+	((u16 *)asi->insn)[0] = __opcode_to_mem_thumb16(0xe929);
+	/* 2nd half (register list) */
+	((u16 *)asi->insn)[1] = __opcode_to_mem_thumb16(insn & 0x1ff);
+	asi->insn_handler = t16_emulate_push;
+	return INSN_GOOD;
+}
+
+static void __kprobes
+t16_emulate_pop_nopc(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	__asm__ __volatile__ (
+		"ldr	r9, [%[regs], #13*4]	\n\t"
+		"ldmia	%[regs], {r0-r7}	\n\t"
+		"blx	%[fn]			\n\t"
+		"stmia	%[regs], {r0-r7}	\n\t"
+		"str	r9, [%[regs], #13*4]	\n\t"
+		:
+		: [regs] "r" (regs), [fn] "r" (asi->insn_fn)
+		: "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7", "r9",
+		  "lr", "memory", "cc"
+		);
+}
+
+static void __kprobes
+t16_emulate_pop_pc(probes_opcode_t insn,
+		struct arch_probes_insn *asi, struct pt_regs *regs)
+{
+	register unsigned long pc asm("r8");
+
+	__asm__ __volatile__ (
+		"ldr	r9, [%[regs], #13*4]	\n\t"
+		"ldmia	%[regs], {r0-r7}	\n\t"
+		"blx	%[fn]			\n\t"
+		"stmia	%[regs], {r0-r7}	\n\t"
+		"str	r9, [%[regs], #13*4]	\n\t"
+		: "=r" (pc)
+		: [regs] "r" (regs), [fn] "r" (asi->insn_fn)
+		: "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7", "r9",
+		  "lr", "memory", "cc"
+		);
+
+	bx_write_pc(pc, regs);
+}
+
+static enum probes_insn __kprobes
+t16_decode_pop(probes_opcode_t insn, struct arch_probes_insn *asi,
+		const struct decode_header *d)
+{
+	/*
+	 * To simulate a POP we use a Thumb-2 "LDMDB R9!, {registers}"
+	 * and call it with R9=SP and PC in the register list represented
+	 * by R8.
+	 */
+	/* 1st half LDMIA R9!,{} */
+	((u16 *)asi->insn)[0] = __opcode_to_mem_thumb16(0xe8b9);
+	/* 2nd half (register list) */
+	((u16 *)asi->insn)[1] = __opcode_to_mem_thumb16(insn & 0x1ff);
+	asi->insn_handler = insn & 0x100 ? t16_emulate_pop_pc
+					 : t16_emulate_pop_nopc;
+	return INSN_GOOD;
+}
+
+const union decode_action kprobes_t16_actions[NUM_PROBES_T16_ACTIONS] = {
+	[PROBES_T16_ADD_SP] = {.handler = t16_simulate_add_sp_imm},
+	[PROBES_T16_CBZ] = {.handler = t16_simulate_cbz},
+	[PROBES_T16_SIGN_EXTEND] = {.handler = t16_emulate_loregs_rwflags},
+	[PROBES_T16_PUSH] = {.decoder = t16_decode_push},
+	[PROBES_T16_POP] = {.decoder = t16_decode_pop},
+	[PROBES_T16_SEV] = {.handler = probes_emulate_none},
+	[PROBES_T16_WFE] = {.handler = probes_simulate_nop},
+	[PROBES_T16_IT] = {.decoder = t16_decode_it},
+	[PROBES_T16_CMP] = {.handler = t16_emulate_loregs_rwflags},
+	[PROBES_T16_ADDSUB] = {.handler = t16_emulate_loregs_noitrwflags},
+	[PROBES_T16_LOGICAL] = {.handler = t16_emulate_loregs_noitrwflags},
+	[PROBES_T16_LDR_LIT] = {.handler = t16_simulate_ldr_literal},
+	[PROBES_T16_BLX] = {.handler = t16_simulate_bxblx},
+	[PROBES_T16_HIREGOPS] = {.decoder = t16_decode_hiregs},
+	[PROBES_T16_LDRHSTRH] = {.handler = t16_emulate_loregs_rwflags},
+	[PROBES_T16_LDRSTR] = {.handler = t16_simulate_ldrstr_sp_relative},
+	[PROBES_T16_ADR] = {.handler = t16_simulate_reladr},
+	[PROBES_T16_LDMSTM] = {.handler = t16_emulate_loregs_rwflags},
+	[PROBES_T16_BRANCH_COND] = {.decoder = t16_decode_cond_branch},
+	[PROBES_T16_BRANCH] = {.handler = t16_simulate_branch},
+};
+
+const union decode_action kprobes_t32_actions[NUM_PROBES_T32_ACTIONS] = {
+	[PROBES_T32_LDMSTM] = {.decoder = t32_decode_ldmstm},
+	[PROBES_T32_LDRDSTRD] = {.handler = t32_emulate_ldrdstrd},
+	[PROBES_T32_TABLE_BRANCH] = {.handler = t32_simulate_table_branch},
+	[PROBES_T32_TST] = {.handler = t32_emulate_rd8rn16rm0_rwflags},
+	[PROBES_T32_MOV] = {.handler = t32_emulate_rd8rn16rm0_rwflags},
+	[PROBES_T32_ADDSUB] = {.handler = t32_emulate_rd8rn16rm0_rwflags},
+	[PROBES_T32_LOGICAL] = {.handler = t32_emulate_rd8rn16rm0_rwflags},
+	[PROBES_T32_CMP] = {.handler = t32_emulate_rd8rn16rm0_rwflags},
+	[PROBES_T32_ADDWSUBW_PC] = {.handler = t32_emulate_rd8pc16_noflags,},
+	[PROBES_T32_ADDWSUBW] = {.handler = t32_emulate_rd8rn16_noflags},
+	[PROBES_T32_MOVW] = {.handler = t32_emulate_rd8rn16_noflags},
+	[PROBES_T32_SAT] = {.handler = t32_emulate_rd8rn16rm0_rwflags},
+	[PROBES_T32_BITFIELD] = {.handler = t32_emulate_rd8rn16_noflags},
+	[PROBES_T32_SEV] = {.handler = probes_emulate_none},
+	[PROBES_T32_WFE] = {.handler = probes_simulate_nop},
+	[PROBES_T32_MRS] = {.handler = t32_simulate_mrs},
+	[PROBES_T32_BRANCH_COND] = {.decoder = t32_decode_cond_branch},
+	[PROBES_T32_BRANCH] = {.handler = t32_simulate_branch},
+	[PROBES_T32_PLDI] = {.handler = probes_simulate_nop},
+	[PROBES_T32_LDR_LIT] = {.handler = t32_simulate_ldr_literal},
+	[PROBES_T32_LDRSTR] = {.handler = t32_emulate_ldrstr},
+	[PROBES_T32_SIGN_EXTEND] = {.handler = t32_emulate_rd8rn16rm0_rwflags},
+	[PROBES_T32_MEDIA] = {.handler = t32_emulate_rd8rn16rm0_rwflags},
+	[PROBES_T32_REVERSE] = {.handler = t32_emulate_rd8rn16_noflags},
+	[PROBES_T32_MUL_ADD] = {.handler = t32_emulate_rd8rn16rm0_rwflags},
+	[PROBES_T32_MUL_ADD2] = {.handler = t32_emulate_rd8rn16rm0ra12_noflags},
+	[PROBES_T32_MUL_ADD_LONG] = {
+		.handler = t32_emulate_rdlo12rdhi8rn16rm0_noflags},
+};
+
+const struct decode_checker *kprobes_t32_checkers[] = {t32_stack_checker, NULL};
+const struct decode_checker *kprobes_t16_checkers[] = {t16_stack_checker, NULL};

arch/arm/probes/kprobes/checkers-arm.c

@@ -0,0 +1,192 @@
+/*
+ * arch/arm/probes/kprobes/checkers-arm.c
+ *
+ * Copyright (C) 2014 Huawei Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+#include <linux/kernel.h>
+#include "../decode.h"
+#include "../decode-arm.h"
+#include "checkers.h"
+
+static enum probes_insn __kprobes arm_check_stack(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	/*
+	 * PROBES_LDRSTRD, PROBES_LDMSTM, PROBES_STORE,
+	 * PROBES_STORE_EXTRA may get here. Simply mark all normal
+	 * insns as STACK_USE_NONE.
+	 */
+	static const union decode_item table[] = {
+		/*
+		 * 'STR{,D,B,H}, Rt, [Rn, Rm]' should be marked as UNKNOWN
+		 * if Rn or Rm is SP.
+		 *                                 x
+		 * STR (register)	cccc 011x x0x0 xxxx xxxx xxxx xxxx xxxx
+		 * STRB (register)	cccc 011x x1x0 xxxx xxxx xxxx xxxx xxxx
+		 */
+		DECODE_OR	(0x0e10000f, 0x0600000d),
+		DECODE_OR	(0x0e1f0000, 0x060d0000),
+
+		/*
+		 *                                                     x
+		 * STRD (register)	cccc 000x x0x0 xxxx xxxx xxxx 1111 xxxx
+		 * STRH (register)	cccc 000x x0x0 xxxx xxxx xxxx 1011 xxxx
+		 */
+		DECODE_OR	(0x0e5000bf, 0x000000bd),
+		DECODE_CUSTOM	(0x0e5f00b0, 0x000d00b0, STACK_USE_UNKNOWN),
+
+		/*
+		 * For PROBES_LDMSTM, only stmdx sp, [...] need to examine
+		 *
+		 * Bit B/A (bit 24) encodes arithmetic operation order. 1 means
+		 * before, 0 means after.
+		 * Bit I/D (bit 23) encodes arithmetic operation. 1 means
+		 * increment, 0 means decrement.
+		 *
+		 * So:
+		 *                              B I
+		 *                              / /
+		 *                              A D   | Rn |
+		 * STMDX SP, [...]	cccc 100x 00x0 xxxx xxxx xxxx xxxx xxxx
+		 */
+		DECODE_CUSTOM	(0x0edf0000, 0x080d0000, STACK_USE_STMDX),
+
+		/*                              P U W | Rn | Rt |     imm12    |*/
+		/* STR (immediate)	cccc 010x x0x0 1101 xxxx xxxx xxxx xxxx */
+		/* STRB (immediate)	cccc 010x x1x0 1101 xxxx xxxx xxxx xxxx */
+		/*                              P U W | Rn | Rt |imm4|    |imm4|*/
+		/* STRD (immediate)	cccc 000x x1x0 1101 xxxx xxxx 1111 xxxx */
+		/* STRH (immediate)	cccc 000x x1x0 1101 xxxx xxxx 1011 xxxx */
+		/*
+		 * index = (P == '1'); add = (U == '1').
+		 * Above insns with:
+		 *    index == 0 (str{,d,h} rx, [sp], #+/-imm) or
+		 *    add == 1 (str{,d,h} rx, [sp, #+<imm>])
+		 * should be STACK_USE_NONE.
+		 * Only str{,b,d,h} rx,[sp,#-n] (P == 1 and U == 0) are
+		 * required to be examined.
+		 */
+		/* STR{,B} Rt,[SP,#-n]	cccc 0101 0xx0 1101 xxxx xxxx xxxx xxxx */
+		DECODE_CUSTOM	(0x0f9f0000, 0x050d0000, STACK_USE_FIXED_XXX),
+
+		/* STR{D,H} Rt,[SP,#-n]	cccc 0001 01x0 1101 xxxx xxxx 1x11 xxxx */
+		DECODE_CUSTOM	(0x0fdf00b0, 0x014d00b0, STACK_USE_FIXED_X0X),
+
+		/* fall through */
+		DECODE_CUSTOM	(0, 0, STACK_USE_NONE),
+		DECODE_END
+	};
+
+	return probes_decode_insn(insn, asi, table, false, false, stack_check_actions, NULL);
+}
+
+const struct decode_checker arm_stack_checker[NUM_PROBES_ARM_ACTIONS] = {
+	[PROBES_LDRSTRD] = {.checker = arm_check_stack},
+	[PROBES_STORE_EXTRA] = {.checker = arm_check_stack},
+	[PROBES_STORE] = {.checker = arm_check_stack},
+	[PROBES_LDMSTM] = {.checker = arm_check_stack},
+};
+
+static enum probes_insn __kprobes arm_check_regs_nouse(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	asi->register_usage_flags = 0;
+	return INSN_GOOD;
+}
+
+static enum probes_insn arm_check_regs_normal(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	u32 regs = h->type_regs.bits >> DECODE_TYPE_BITS;
+	int i;
+
+	asi->register_usage_flags = 0;
+	for (i = 0; i < 5; regs >>= 4, insn >>= 4, i++)
+		if (regs & 0xf)
+			asi->register_usage_flags |= 1 << (insn & 0xf);
+
+	return INSN_GOOD;
+}
+
+
+static enum probes_insn arm_check_regs_ldmstm(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	unsigned int reglist = insn & 0xffff;
+	unsigned int rn = (insn >> 16) & 0xf;
+	asi->register_usage_flags = reglist | (1 << rn);
+	return INSN_GOOD;
+}
+
+static enum probes_insn arm_check_regs_mov_ip_sp(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	/* Instruction is 'mov ip, sp' i.e. 'mov r12, r13' */
+	asi->register_usage_flags = (1 << 12) | (1<< 13);
+	return INSN_GOOD;
+}
+
+/*
+ *                                    | Rn |Rt/d|         | Rm |
+ * LDRD (register)      cccc 000x x0x0 xxxx xxxx xxxx 1101 xxxx
+ * STRD (register)      cccc 000x x0x0 xxxx xxxx xxxx 1111 xxxx
+ *                                    | Rn |Rt/d|         |imm4L|
+ * LDRD (immediate)     cccc 000x x1x0 xxxx xxxx xxxx 1101 xxxx
+ * STRD (immediate)     cccc 000x x1x0 xxxx xxxx xxxx 1111 xxxx
+ *
+ * Such instructions access Rt/d and its next register, so different
+ * from others, a specific checker is required to handle this extra
+ * implicit register usage.
+ */
+static enum probes_insn arm_check_regs_ldrdstrd(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	int rdt = (insn >> 12) & 0xf;
+	arm_check_regs_normal(insn, asi, h);
+	asi->register_usage_flags |= 1 << (rdt + 1);
+	return INSN_GOOD;
+}
+
+
+const struct decode_checker arm_regs_checker[NUM_PROBES_ARM_ACTIONS] = {
+	[PROBES_MRS] = {.checker = arm_check_regs_normal},
+	[PROBES_SATURATING_ARITHMETIC] = {.checker = arm_check_regs_normal},
+	[PROBES_MUL1] = {.checker = arm_check_regs_normal},
+	[PROBES_MUL2] = {.checker = arm_check_regs_normal},
+	[PROBES_MUL_ADD_LONG] = {.checker = arm_check_regs_normal},
+	[PROBES_MUL_ADD] = {.checker = arm_check_regs_normal},
+	[PROBES_LOAD] = {.checker = arm_check_regs_normal},
+	[PROBES_LOAD_EXTRA] = {.checker = arm_check_regs_normal},
+	[PROBES_STORE] = {.checker = arm_check_regs_normal},
+	[PROBES_STORE_EXTRA] = {.checker = arm_check_regs_normal},
+	[PROBES_DATA_PROCESSING_REG] = {.checker = arm_check_regs_normal},
+	[PROBES_DATA_PROCESSING_IMM] = {.checker = arm_check_regs_normal},
+	[PROBES_SEV] = {.checker = arm_check_regs_nouse},
+	[PROBES_WFE] = {.checker = arm_check_regs_nouse},
+	[PROBES_SATURATE] = {.checker = arm_check_regs_normal},
+	[PROBES_REV] = {.checker = arm_check_regs_normal},
+	[PROBES_MMI] = {.checker = arm_check_regs_normal},
+	[PROBES_PACK] = {.checker = arm_check_regs_normal},
+	[PROBES_EXTEND] = {.checker = arm_check_regs_normal},
+	[PROBES_EXTEND_ADD] = {.checker = arm_check_regs_normal},
+	[PROBES_BITFIELD] = {.checker = arm_check_regs_normal},
+	[PROBES_LDMSTM] = {.checker = arm_check_regs_ldmstm},
+	[PROBES_MOV_IP_SP] = {.checker = arm_check_regs_mov_ip_sp},
+	[PROBES_LDRSTRD] = {.checker = arm_check_regs_ldrdstrd},
+};

arch/arm/probes/kprobes/checkers-common.c

@@ -0,0 +1,101 @@
+/*
+ * arch/arm/probes/kprobes/checkers-common.c
+ *
+ * Copyright (C) 2014 Huawei Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+#include <linux/kernel.h>
+#include "../decode.h"
+#include "../decode-arm.h"
+#include "checkers.h"
+
+enum probes_insn checker_stack_use_none(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	asi->stack_space = 0;
+	return INSN_GOOD_NO_SLOT;
+}
+
+enum probes_insn checker_stack_use_unknown(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	asi->stack_space = -1;
+	return INSN_GOOD_NO_SLOT;
+}
+
+#ifdef CONFIG_THUMB2_KERNEL
+enum probes_insn checker_stack_use_imm_0xx(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	int imm = insn & 0xff;
+	asi->stack_space = imm;
+	return INSN_GOOD_NO_SLOT;
+}
+
+/*
+ * Different from other insn uses imm8, the real addressing offset of
+ * STRD in T32 encoding should be imm8 * 4. See ARMARM description.
+ */
+enum probes_insn checker_stack_use_t32strd(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	int imm = insn & 0xff;
+	asi->stack_space = imm << 2;
+	return INSN_GOOD_NO_SLOT;
+}
+#else
+enum probes_insn checker_stack_use_imm_x0x(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	int imm = ((insn & 0xf00) >> 4) + (insn & 0xf);
+	asi->stack_space = imm;
+	return INSN_GOOD_NO_SLOT;
+}
+#endif
+
+enum probes_insn checker_stack_use_imm_xxx(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	int imm = insn & 0xfff;
+	asi->stack_space = imm;
+	return INSN_GOOD_NO_SLOT;
+}
+
+enum probes_insn checker_stack_use_stmdx(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	unsigned int reglist = insn & 0xffff;
+	int pbit = insn & (1 << 24);
+	asi->stack_space = (hweight32(reglist) - (!pbit ? 1 : 0)) * 4;
+
+	return INSN_GOOD_NO_SLOT;
+}
+
+const union decode_action stack_check_actions[] = {
+	[STACK_USE_NONE] = {.decoder = checker_stack_use_none},
+	[STACK_USE_UNKNOWN] = {.decoder = checker_stack_use_unknown},
+#ifdef CONFIG_THUMB2_KERNEL
+	[STACK_USE_FIXED_0XX] = {.decoder = checker_stack_use_imm_0xx},
+	[STACK_USE_T32STRD] = {.decoder = checker_stack_use_t32strd},
+#else
+	[STACK_USE_FIXED_X0X] = {.decoder = checker_stack_use_imm_x0x},
+#endif
+	[STACK_USE_FIXED_XXX] = {.decoder = checker_stack_use_imm_xxx},
+	[STACK_USE_STMDX] = {.decoder = checker_stack_use_stmdx},
+};

arch/arm/probes/kprobes/checkers-thumb.c

@@ -0,0 +1,110 @@
+/*
+ * arch/arm/probes/kprobes/checkers-thumb.c
+ *
+ * Copyright (C) 2014 Huawei Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+#include <linux/kernel.h>
+#include "../decode.h"
+#include "../decode-thumb.h"
+#include "checkers.h"
+
+static enum probes_insn __kprobes t32_check_stack(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	/*
+	 * PROBES_T32_LDMSTM, PROBES_T32_LDRDSTRD and PROBES_T32_LDRSTR
+	 * may get here. Simply mark all normal insns as STACK_USE_NONE.
+	 */
+	static const union decode_item table[] = {
+
+		/*
+		 * First, filter out all ldr insns to make our life easier.
+		 * Following load insns may come here:
+		 * LDM, LDRD, LDR.
+		 * In T32 encoding, bit 20 is enough for distinguishing
+		 * load and store. All load insns have this bit set, when
+		 * all store insns have this bit clear.
+		 */
+		DECODE_CUSTOM	(0x00100000, 0x00100000, STACK_USE_NONE),
+
+		/*
+		 * Mark all 'STR{,B,H}, Rt, [Rn, Rm]' as STACK_USE_UNKNOWN
+		 * if Rn or Rm is SP. T32 doesn't encode STRD.
+		 */
+		/*                                 xx | Rn | Rt |         | Rm |*/
+		/* STR (register)	1111 1000 0100 xxxx xxxx 0000 00xx xxxx */
+		/* STRB (register)	1111 1000 0000 xxxx xxxx 0000 00xx xxxx */
+		/* STRH (register)	1111 1000 0010 xxxx xxxx 0000 00xx xxxx */
+		/* INVALID INSN		1111 1000 0110 xxxx xxxx 0000 00xx xxxx */
+		/* By Introducing INVALID INSN, bit 21 and 22 can be ignored. */
+		DECODE_OR	(0xff9f0fc0, 0xf80d0000),
+		DECODE_CUSTOM	(0xff900fcf, 0xf800000d, STACK_USE_UNKNOWN),
+
+
+		/*                                 xx | Rn | Rt | PUW|   imm8  |*/
+		/* STR (imm 8)		1111 1000 0100 1101 xxxx 110x xxxx xxxx */
+		/* STRB (imm 8)		1111 1000 0000 1101 xxxx 110x xxxx xxxx */
+		/* STRH (imm 8)		1111 1000 0010 1101 xxxx 110x xxxx xxxx */
+		/* INVALID INSN		1111 1000 0110 1101 xxxx 110x xxxx xxxx */
+		/* Only consider U == 0 and P == 1: strx rx, [sp, #-<imm>] */
+		DECODE_CUSTOM	(0xff9f0e00, 0xf80d0c00, STACK_USE_FIXED_0XX),
+
+		/* For STR{,B,H} (imm 12), offset is always positive, so ignore them. */
+
+		/*                              P U W | Rn | Rt | Rt2|   imm8  |*/
+		/* STRD (immediate)	1110 1001 01x0 1101 xxxx xxxx xxxx xxxx */
+		/*
+		 * Only consider U == 0 and P == 1.
+		 * Also note that STRD in T32 encoding is special:
+		 * imm = ZeroExtend(imm8:'00', 32)
+		 */
+		DECODE_CUSTOM	(0xffdf0000, 0xe94d0000, STACK_USE_T32STRD),
+
+		/*                                    | Rn | */
+		/* STMDB		1110 1001 00x0 1101 xxxx xxxx xxxx xxxx */
+		DECODE_CUSTOM	(0xffdf0000, 0xe90d0000, STACK_USE_STMDX),
+
+		/* fall through */
+		DECODE_CUSTOM	(0, 0, STACK_USE_NONE),
+		DECODE_END
+	};
+
+	return probes_decode_insn(insn, asi, table, false, false, stack_check_actions, NULL);
+}
+
+const struct decode_checker t32_stack_checker[NUM_PROBES_T32_ACTIONS] = {
+	[PROBES_T32_LDMSTM] = {.checker = t32_check_stack},
+	[PROBES_T32_LDRDSTRD] = {.checker = t32_check_stack},
+	[PROBES_T32_LDRSTR] = {.checker = t32_check_stack},
+};
+
+/*
+ * See following comments. This insn must be 'push'.
+ */
+static enum probes_insn __kprobes t16_check_stack(probes_opcode_t insn,
+		struct arch_probes_insn *asi,
+		const struct decode_header *h)
+{
+	unsigned int reglist = insn & 0x1ff;
+	asi->stack_space = hweight32(reglist) * 4;
+	return INSN_GOOD;
+}
+
+/*
+ * T16 encoding is simple: only the 'push' insn can need extra stack space.
+ * Other insns, like str, can only use r0-r7 as Rn.
+ */
+const struct decode_checker t16_stack_checker[NUM_PROBES_T16_ACTIONS] = {
+	[PROBES_T16_PUSH] = {.checker = t16_check_stack},
+};

arch/arm/probes/kprobes/checkers.h

@@ -0,0 +1,55 @@
+/*
+ * arch/arm/probes/kprobes/checkers.h
+ *
+ * Copyright (C) 2014 Huawei Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+#ifndef _ARM_KERNEL_PROBES_CHECKERS_H
+#define _ARM_KERNEL_PROBES_CHECKERS_H
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include "../decode.h"
+
+extern probes_check_t checker_stack_use_none;
+extern probes_check_t checker_stack_use_unknown;
+#ifdef CONFIG_THUMB2_KERNEL
+extern probes_check_t checker_stack_use_imm_0xx;
+#else
+extern probes_check_t checker_stack_use_imm_x0x;
+#endif
+extern probes_check_t checker_stack_use_imm_xxx;
+extern probes_check_t checker_stack_use_stmdx;
+
+enum {
+	STACK_USE_NONE,
+	STACK_USE_UNKNOWN,
+#ifdef CONFIG_THUMB2_KERNEL
+	STACK_USE_FIXED_0XX,
+	STACK_USE_T32STRD,
+#else
+	STACK_USE_FIXED_X0X,
+#endif
+	STACK_USE_FIXED_XXX,
+	STACK_USE_STMDX,
+	NUM_STACK_USE_TYPES
+};
+
+extern const union decode_action stack_check_actions[];
+
+#ifndef CONFIG_THUMB2_KERNEL
+extern const struct decode_checker arm_stack_checker[];
+extern const struct decode_checker arm_regs_checker[];
+#else
+#endif
+extern const struct decode_checker t32_stack_checker[];
+extern const struct decode_checker t16_stack_checker[];
+#endif

arch/arm/probes/kprobes/core.c

@@ -0,0 +1,673 @@
+/*
+ * arch/arm/kernel/kprobes.c
+ *
+ * Kprobes on ARM
+ *
+ * Abhishek Sagar <sagar.abhishek@gmail.com>
+ * Copyright (C) 2006, 2007 Motorola Inc.
+ *
+ * Nicolas Pitre <nico@marvell.com>
+ * Copyright (C) 2007 Marvell Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/kprobes.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/stop_machine.h>
+#include <linux/stringify.h>
+#include <asm/traps.h>
+#include <asm/opcodes.h>
+#include <asm/cacheflush.h>
+#include <linux/percpu.h>
+#include <linux/bug.h>
+#include <asm/patch.h>
+
+#include "../decode-arm.h"
+#include "../decode-thumb.h"
+#include "core.h"
+
+#define MIN_STACK_SIZE(addr) 				\
+	min((unsigned long)MAX_STACK_SIZE,		\
+	    (unsigned long)current_thread_info() + THREAD_START_SP - (addr))
+
+#define flush_insns(addr, size)				\
+	flush_icache_range((unsigned long)(addr),	\
+			   (unsigned long)(addr) +	\
+			   (size))
+
+/* Used as a marker in ARM_pc to note when we're in a jprobe. */
+#define JPROBE_MAGIC_ADDR		0xffffffff
+
+DEFINE_PER_CPU(struct kprobe *, current_kprobe) = NULL;
+DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
+
+
+int __kprobes arch_prepare_kprobe(struct kprobe *p)
+{
+	kprobe_opcode_t insn;
+	kprobe_opcode_t tmp_insn[MAX_INSN_SIZE];
+	unsigned long addr = (unsigned long)p->addr;
+	bool thumb;
+	kprobe_decode_insn_t *decode_insn;
+	const union decode_action *actions;
+	int is;
+	const struct decode_checker **checkers;
+
+	if (in_exception_text(addr))
+		return -EINVAL;
+
+#ifdef CONFIG_THUMB2_KERNEL
+	thumb = true;
+	addr &= ~1; /* Bit 0 would normally be set to indicate Thumb code */
+	insn = __mem_to_opcode_thumb16(((u16 *)addr)[0]);
+	if (is_wide_instruction(insn)) {
+		u16 inst2 = __mem_to_opcode_thumb16(((u16 *)addr)[1]);
+		insn = __opcode_thumb32_compose(insn, inst2);
+		decode_insn = thumb32_probes_decode_insn;
+		actions = kprobes_t32_actions;
+		checkers = kprobes_t32_checkers;
+	} else {
+		decode_insn = thumb16_probes_decode_insn;
+		actions = kprobes_t16_actions;
+		checkers = kprobes_t16_checkers;
+	}
+#else /* !CONFIG_THUMB2_KERNEL */
+	thumb = false;
+	if (addr & 0x3)
+		return -EINVAL;
+	insn = __mem_to_opcode_arm(*p->addr);
+	decode_insn = arm_probes_decode_insn;
+	actions = kprobes_arm_actions;
+	checkers = kprobes_arm_checkers;
+#endif
+
+	p->opcode = insn;
+	p->ainsn.insn = tmp_insn;
+
+	switch ((*decode_insn)(insn, &p->ainsn, true, actions, checkers)) {
+	case INSN_REJECTED:	/* not supported */
+		return -EINVAL;
+
+	case INSN_GOOD:		/* instruction uses slot */
+		p->ainsn.insn = get_insn_slot();
+		if (!p->ainsn.insn)
+			return -ENOMEM;
+		for (is = 0; is < MAX_INSN_SIZE; ++is)
+			p->ainsn.insn[is] = tmp_insn[is];
+		flush_insns(p->ainsn.insn,
+				sizeof(p->ainsn.insn[0]) * MAX_INSN_SIZE);
+		p->ainsn.insn_fn = (probes_insn_fn_t *)
+					((uintptr_t)p->ainsn.insn | thumb);
+		break;
+
+	case INSN_GOOD_NO_SLOT:	/* instruction doesn't need insn slot */
+		p->ainsn.insn = NULL;
+		break;
+	}
+
+	/*
+	 * Never instrument insn like 'str r0, [sp, +/-r1]'. Also, insn likes
+	 * 'str r0, [sp, #-68]' should also be prohibited.
+	 * See __und_svc.
+	 */
+	if ((p->ainsn.stack_space < 0) ||
+			(p->ainsn.stack_space > MAX_STACK_SIZE))
+		return -EINVAL;
+
+	return 0;
+}
+
+void __kprobes arch_arm_kprobe(struct kprobe *p)
+{
+	unsigned int brkp;
+	void *addr;
+
+	if (IS_ENABLED(CONFIG_THUMB2_KERNEL)) {
+		/* Remove any Thumb flag */
+		addr = (void *)((uintptr_t)p->addr & ~1);
+
+		if (is_wide_instruction(p->opcode))
+			brkp = KPROBE_THUMB32_BREAKPOINT_INSTRUCTION;
+		else
+			brkp = KPROBE_THUMB16_BREAKPOINT_INSTRUCTION;
+	} else {
+		kprobe_opcode_t insn = p->opcode;
+
+		addr = p->addr;
+		brkp = KPROBE_ARM_BREAKPOINT_INSTRUCTION;
+
+		if (insn >= 0xe0000000)
+			brkp |= 0xe0000000;  /* Unconditional instruction */
+		else
+			brkp |= insn & 0xf0000000;  /* Copy condition from insn */
+	}
+
+	patch_text(addr, brkp);
+}
+
+/*
+ * The actual disarming is done here on each CPU and synchronized using
+ * stop_machine. This synchronization is necessary on SMP to avoid removing
+ * a probe between the moment the 'Undefined Instruction' exception is raised
+ * and the moment the exception handler reads the faulting instruction from
+ * memory. It is also needed to atomically set the two half-words of a 32-bit
+ * Thumb breakpoint.
+ */
+struct patch {
+	void *addr;
+	unsigned int insn;
+};
+
+static int __kprobes_remove_breakpoint(void *data)
+{
+	struct patch *p = data;
+	__patch_text(p->addr, p->insn);
+	return 0;
+}
+
+void __kprobes kprobes_remove_breakpoint(void *addr, unsigned int insn)
+{
+	struct patch p = {
+		.addr = addr,
+		.insn = insn,
+	};
+	stop_machine(__kprobes_remove_breakpoint, &p, cpu_online_mask);
+}
+
+void __kprobes arch_disarm_kprobe(struct kprobe *p)
+{
+	kprobes_remove_breakpoint((void *)((uintptr_t)p->addr & ~1),
+			p->opcode);
+}
+
+void __kprobes arch_remove_kprobe(struct kprobe *p)
+{
+	if (p->ainsn.insn) {
+		free_insn_slot(p->ainsn.insn, 0);
+		p->ainsn.insn = NULL;
+	}
+}
+
+static void __kprobes save_previous_kprobe(struct kprobe_ctlblk *kcb)
+{
+	kcb->prev_kprobe.kp = kprobe_running();
+	kcb->prev_kprobe.status = kcb->kprobe_status;
+}
+
+static void __kprobes restore_previous_kprobe(struct kprobe_ctlblk *kcb)
+{
+	__this_cpu_write(current_kprobe, kcb->prev_kprobe.kp);
+	kcb->kprobe_status = kcb->prev_kprobe.status;
+}
+
+static void __kprobes set_current_kprobe(struct kprobe *p)
+{
+	__this_cpu_write(current_kprobe, p);
+}
+
+static void __kprobes
+singlestep_skip(struct kprobe *p, struct pt_regs *regs)
+{
+#ifdef CONFIG_THUMB2_KERNEL
+	regs->ARM_cpsr = it_advance(regs->ARM_cpsr);
+	if (is_wide_instruction(p->opcode))
+		regs->ARM_pc += 4;
+	else
+		regs->ARM_pc += 2;
+#else
+	regs->ARM_pc += 4;
+#endif
+}
+
+static inline void __kprobes
+singlestep(struct kprobe *p, struct pt_regs *regs, struct kprobe_ctlblk *kcb)
+{
+	p->ainsn.insn_singlestep(p->opcode, &p->ainsn, regs);
+}
+
+/*
+ * Called with IRQs disabled. IRQs must remain disabled from that point
+ * all the way until processing this kprobe is complete.  The current
+ * kprobes implementation cannot process more than one nested level of
+ * kprobe, and that level is reserved for user kprobe handlers, so we can't
+ * risk encountering a new kprobe in an interrupt handler.
+ */
+void __kprobes kprobe_handler(struct pt_regs *regs)
+{
+	struct kprobe *p, *cur;
+	struct kprobe_ctlblk *kcb;
+
+	kcb = get_kprobe_ctlblk();
+	cur = kprobe_running();
+
+#ifdef CONFIG_THUMB2_KERNEL
+	/*
+	 * First look for a probe which was registered using an address with
+	 * bit 0 set, this is the usual situation for pointers to Thumb code.
+	 * If not found, fallback to looking for one with bit 0 clear.
+	 */
+	p = get_kprobe((kprobe_opcode_t *)(regs->ARM_pc | 1));
+	if (!p)
+		p = get_kprobe((kprobe_opcode_t *)regs->ARM_pc);
+
+#else /* ! CONFIG_THUMB2_KERNEL */
+	p = get_kprobe((kprobe_opcode_t *)regs->ARM_pc);
+#endif
+
+	if (p) {
+		if (cur) {
+			/* Kprobe is pending, so we're recursing. */
+			switch (kcb->kprobe_status) {
+			case KPROBE_HIT_ACTIVE:
+			case KPROBE_HIT_SSDONE:
+				/* A pre- or post-handler probe got us here. */
+				kprobes_inc_nmissed_count(p);
+				save_previous_kprobe(kcb);
+				set_current_kprobe(p);
+				kcb->kprobe_status = KPROBE_REENTER;
+				singlestep(p, regs, kcb);
+				restore_previous_kprobe(kcb);
+				break;
+			default:
+				/* impossible cases */
+				BUG();
+			}
+		} else if (p->ainsn.insn_check_cc(regs->ARM_cpsr)) {
+			/* Probe hit and conditional execution check ok. */
+			set_current_kprobe(p);
+			kcb->kprobe_status = KPROBE_HIT_ACTIVE;
+
+			/*
+			 * If we have no pre-handler or it returned 0, we
+			 * continue with normal processing.  If we have a
+			 * pre-handler and it returned non-zero, it prepped
+			 * for calling the break_handler below on re-entry,
+			 * so get out doing nothing more here.
+			 */
+			if (!p->pre_handler || !p->pre_handler(p, regs)) {
+				kcb->kprobe_status = KPROBE_HIT_SS;
+				singlestep(p, regs, kcb);
+				if (p->post_handler) {
+					kcb->kprobe_status = KPROBE_HIT_SSDONE;
+					p->post_handler(p, regs, 0);
+				}
+				reset_current_kprobe();
+			}
+		} else {
+			/*
+			 * Probe hit but conditional execution check failed,
+			 * so just skip the instruction and continue as if
+			 * nothing had happened.
+			 */
+			singlestep_skip(p, regs);
+		}
+	} else if (cur) {
+		/* We probably hit a jprobe.  Call its break handler. */
+		if (cur->break_handler && cur->break_handler(cur, regs)) {
+			kcb->kprobe_status = KPROBE_HIT_SS;
+			singlestep(cur, regs, kcb);
+			if (cur->post_handler) {
+				kcb->kprobe_status = KPROBE_HIT_SSDONE;
+				cur->post_handler(cur, regs, 0);
+			}
+		}
+		reset_current_kprobe();
+	} else {
+		/*
+		 * The probe was removed and a race is in progress.
+		 * There is nothing we can do about it.  Let's restart
+		 * the instruction.  By the time we can restart, the
+		 * real instruction will be there.
+		 */
+	}
+}
+
+static int __kprobes kprobe_trap_handler(struct pt_regs *regs, unsigned int instr)
+{
+	unsigned long flags;
+	local_irq_save(flags);
+	kprobe_handler(regs);
+	local_irq_restore(flags);
+	return 0;
+}
+
+int __kprobes kprobe_fault_handler(struct pt_regs *regs, unsigned int fsr)
+{
+	struct kprobe *cur = kprobe_running();
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+	switch (kcb->kprobe_status) {
+	case KPROBE_HIT_SS:
+	case KPROBE_REENTER:
+		/*
+		 * We are here because the instruction being single
+		 * stepped caused a page fault. We reset the current
+		 * kprobe and the PC to point back to the probe address
+		 * and allow the page fault handler to continue as a
+		 * normal page fault.
+		 */
+		regs->ARM_pc = (long)cur->addr;
+		if (kcb->kprobe_status == KPROBE_REENTER) {
+			restore_previous_kprobe(kcb);
+		} else {
+			reset_current_kprobe();
+		}
+		break;
+
+	case KPROBE_HIT_ACTIVE:
+	case KPROBE_HIT_SSDONE:
+		/*
+		 * We increment the nmissed count for accounting,
+		 * we can also use npre/npostfault count for accounting
+		 * these specific fault cases.
+		 */
+		kprobes_inc_nmissed_count(cur);
+
+		/*
+		 * We come here because instructions in the pre/post
+		 * handler caused the page_fault, this could happen
+		 * if handler tries to access user space by
+		 * copy_from_user(), get_user() etc. Let the
+		 * user-specified handler try to fix it.
+		 */
+		if (cur->fault_handler && cur->fault_handler(cur, regs, fsr))
+			return 1;
+		break;
+
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
+				       unsigned long val, void *data)
+{
+	/*
+	 * notify_die() is currently never called on ARM,
+	 * so this callback is currently empty.
+	 */
+	return NOTIFY_DONE;
+}
+
+/*
+ * When a retprobed function returns, trampoline_handler() is called,
+ * calling the kretprobe's handler. We construct a struct pt_regs to
+ * give a view of registers r0-r11 to the user return-handler.  This is
+ * not a complete pt_regs structure, but that should be plenty sufficient
+ * for kretprobe handlers which should normally be interested in r0 only
+ * anyway.
+ */
+void __naked __kprobes kretprobe_trampoline(void)
+{
+	__asm__ __volatile__ (
+		"stmdb	sp!, {r0 - r11}		\n\t"
+		"mov	r0, sp			\n\t"
+		"bl	trampoline_handler	\n\t"
+		"mov	lr, r0			\n\t"
+		"ldmia	sp!, {r0 - r11}		\n\t"
+#ifdef CONFIG_THUMB2_KERNEL
+		"bx	lr			\n\t"
+#else
+		"mov	pc, lr			\n\t"
+#endif
+		: : : "memory");
+}
+
+/* Called from kretprobe_trampoline */
+static __used __kprobes void *trampoline_handler(struct pt_regs *regs)
+{
+	struct kretprobe_instance *ri = NULL;
+	struct hlist_head *head, empty_rp;
+	struct hlist_node *tmp;
+	unsigned long flags, orig_ret_address = 0;
+	unsigned long trampoline_address = (unsigned long)&kretprobe_trampoline;
+	kprobe_opcode_t *correct_ret_addr = NULL;
+
+	INIT_HLIST_HEAD(&empty_rp);
+	kretprobe_hash_lock(current, &head, &flags);
+
+	/*
+	 * It is possible to have multiple instances associated with a given
+	 * task either because multiple functions in the call path have
+	 * a return probe installed on them, and/or more than one return
+	 * probe was registered for a target function.
+	 *
+	 * We can handle this because:
+	 *     - instances are always inserted at the head of the list
+	 *     - when multiple return probes are registered for the same
+	 *       function, the first instance's ret_addr will point to the
+	 *       real return address, and all the rest will point to
+	 *       kretprobe_trampoline
+	 */
+	hlist_for_each_entry_safe(ri, tmp, head, hlist) {
+		if (ri->task != current)
+			/* another task is sharing our hash bucket */
+			continue;
+
+		orig_ret_address = (unsigned long)ri->ret_addr;
+
+		if (orig_ret_address != trampoline_address)
+			/*
+			 * This is the real return address. Any other
+			 * instances associated with this task are for
+			 * other calls deeper on the call stack
+			 */
+			break;
+	}
+
+	kretprobe_assert(ri, orig_ret_address, trampoline_address);
+
+	correct_ret_addr = ri->ret_addr;
+	hlist_for_each_entry_safe(ri, tmp, head, hlist) {
+		if (ri->task != current)
+			/* another task is sharing our hash bucket */
+			continue;
+
+		orig_ret_address = (unsigned long)ri->ret_addr;
+		if (ri->rp && ri->rp->handler) {
+			__this_cpu_write(current_kprobe, &ri->rp->kp);
+			get_kprobe_ctlblk()->kprobe_status = KPROBE_HIT_ACTIVE;
+			ri->ret_addr = correct_ret_addr;
+			ri->rp->handler(ri, regs);
+			__this_cpu_write(current_kprobe, NULL);
+		}
+
+		recycle_rp_inst(ri, &empty_rp);
+
+		if (orig_ret_address != trampoline_address)
+			/*
+			 * This is the real return address. Any other
+			 * instances associated with this task are for
+			 * other calls deeper on the call stack
+			 */
+			break;
+	}
+
+	kretprobe_hash_unlock(current, &flags);
+
+	hlist_for_each_entry_safe(ri, tmp, &empty_rp, hlist) {
+		hlist_del(&ri->hlist);
+		kfree(ri);
+	}
+
+	return (void *)orig_ret_address;
+}
+
+void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
+				      struct pt_regs *regs)
+{
+	ri->ret_addr = (kprobe_opcode_t *)regs->ARM_lr;
+
+	/* Replace the return addr with trampoline addr. */
+	regs->ARM_lr = (unsigned long)&kretprobe_trampoline;
+}
+
+int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct jprobe *jp = container_of(p, struct jprobe, kp);
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+	long sp_addr = regs->ARM_sp;
+	long cpsr;
+
+	kcb->jprobe_saved_regs = *regs;
+	memcpy(kcb->jprobes_stack, (void *)sp_addr, MIN_STACK_SIZE(sp_addr));
+	regs->ARM_pc = (long)jp->entry;
+
+	cpsr = regs->ARM_cpsr | PSR_I_BIT;
+#ifdef CONFIG_THUMB2_KERNEL
+	/* Set correct Thumb state in cpsr */
+	if (regs->ARM_pc & 1)
+		cpsr |= PSR_T_BIT;
+	else
+		cpsr &= ~PSR_T_BIT;
+#endif
+	regs->ARM_cpsr = cpsr;
+
+	preempt_disable();
+	return 1;
+}
+
+void __kprobes jprobe_return(void)
+{
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+	__asm__ __volatile__ (
+		/*
+		 * Setup an empty pt_regs. Fill SP and PC fields as
+		 * they're needed by longjmp_break_handler.
+		 *
+		 * We allocate some slack between the original SP and start of
+		 * our fabricated regs. To be precise we want to have worst case
+		 * covered which is STMFD with all 16 regs so we allocate 2 *
+		 * sizeof(struct_pt_regs)).
+		 *
+		 * This is to prevent any simulated instruction from writing
+		 * over the regs when they are accessing the stack.
+		 */
+#ifdef CONFIG_THUMB2_KERNEL
+		"sub    r0, %0, %1		\n\t"
+		"mov    sp, r0			\n\t"
+#else
+		"sub    sp, %0, %1		\n\t"
+#endif
+		"ldr    r0, ="__stringify(JPROBE_MAGIC_ADDR)"\n\t"
+		"str    %0, [sp, %2]		\n\t"
+		"str    r0, [sp, %3]		\n\t"
+		"mov    r0, sp			\n\t"
+		"bl     kprobe_handler		\n\t"
+
+		/*
+		 * Return to the context saved by setjmp_pre_handler
+		 * and restored by longjmp_break_handler.
+		 */
+#ifdef CONFIG_THUMB2_KERNEL
+		"ldr	lr, [sp, %2]		\n\t" /* lr = saved sp */
+		"ldrd	r0, r1, [sp, %5]	\n\t" /* r0,r1 = saved lr,pc */
+		"ldr	r2, [sp, %4]		\n\t" /* r2 = saved psr */
+		"stmdb	lr!, {r0, r1, r2}	\n\t" /* push saved lr and */
+						      /* rfe context */
+		"ldmia	sp, {r0 - r12}		\n\t"
+		"mov	sp, lr			\n\t"
+		"ldr	lr, [sp], #4		\n\t"
+		"rfeia	sp!			\n\t"
+#else
+		"ldr	r0, [sp, %4]		\n\t"
+		"msr	cpsr_cxsf, r0		\n\t"
+		"ldmia	sp, {r0 - pc}		\n\t"
+#endif
+		:
+		: "r" (kcb->jprobe_saved_regs.ARM_sp),
+		  "I" (sizeof(struct pt_regs) * 2),
+		  "J" (offsetof(struct pt_regs, ARM_sp)),
+		  "J" (offsetof(struct pt_regs, ARM_pc)),
+		  "J" (offsetof(struct pt_regs, ARM_cpsr)),
+		  "J" (offsetof(struct pt_regs, ARM_lr))
+		: "memory", "cc");
+}
+
+int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+	long stack_addr = kcb->jprobe_saved_regs.ARM_sp;
+	long orig_sp = regs->ARM_sp;
+	struct jprobe *jp = container_of(p, struct jprobe, kp);
+
+	if (regs->ARM_pc == JPROBE_MAGIC_ADDR) {
+		if (orig_sp != stack_addr) {
+			struct pt_regs *saved_regs =
+				(struct pt_regs *)kcb->jprobe_saved_regs.ARM_sp;
+			printk("current sp %lx does not match saved sp %lx\n",
+			       orig_sp, stack_addr);
+			printk("Saved registers for jprobe %p\n", jp);
+			show_regs(saved_regs);
+			printk("Current registers\n");
+			show_regs(regs);
+			BUG();
+		}
+		*regs = kcb->jprobe_saved_regs;
+		memcpy((void *)stack_addr, kcb->jprobes_stack,
+		       MIN_STACK_SIZE(stack_addr));
+		preempt_enable_no_resched();
+		return 1;
+	}
+	return 0;
+}
+
+int __kprobes arch_trampoline_kprobe(struct kprobe *p)
+{
+	return 0;
+}
+
+#ifdef CONFIG_THUMB2_KERNEL
+
+static struct undef_hook kprobes_thumb16_break_hook = {
+	.instr_mask	= 0xffff,
+	.instr_val	= KPROBE_THUMB16_BREAKPOINT_INSTRUCTION,
+	.cpsr_mask	= MODE_MASK,
+	.cpsr_val	= SVC_MODE,
+	.fn		= kprobe_trap_handler,
+};
+
+static struct undef_hook kprobes_thumb32_break_hook = {
+	.instr_mask	= 0xffffffff,
+	.instr_val	= KPROBE_THUMB32_BREAKPOINT_INSTRUCTION,
+	.cpsr_mask	= MODE_MASK,
+	.cpsr_val	= SVC_MODE,
+	.fn		= kprobe_trap_handler,
+};
+
+#else  /* !CONFIG_THUMB2_KERNEL */
+
+static struct undef_hook kprobes_arm_break_hook = {
+	.instr_mask	= 0x0fffffff,
+	.instr_val	= KPROBE_ARM_BREAKPOINT_INSTRUCTION,
+	.cpsr_mask	= MODE_MASK,
+	.cpsr_val	= SVC_MODE,
+	.fn		= kprobe_trap_handler,
+};
+
+#endif /* !CONFIG_THUMB2_KERNEL */
+
+int __init arch_init_kprobes()
+{
+	arm_probes_decode_init();
+#ifdef CONFIG_THUMB2_KERNEL
+	register_undef_hook(&kprobes_thumb16_break_hook);
+	register_undef_hook(&kprobes_thumb32_break_hook);
+#else
+	register_undef_hook(&kprobes_arm_break_hook);
+#endif
+	return 0;
+}

arch/arm/probes/kprobes/core.h

@@ -0,0 +1,58 @@
+/*
+ * arch/arm/kernel/kprobes.h
+ *
+ * Copyright (C) 2011 Jon Medhurst <tixy@yxit.co.uk>.
+ *
+ * Some contents moved here from arch/arm/include/asm/kprobes.h which is
+ * Copyright (C) 2006, 2007 Motorola Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+#ifndef _ARM_KERNEL_KPROBES_H
+#define _ARM_KERNEL_KPROBES_H
+
+#include <asm/kprobes.h>
+#include "../decode.h"
+
+/*
+ * These undefined instructions must be unique and
+ * reserved solely for kprobes' use.
+ */
+#define KPROBE_ARM_BREAKPOINT_INSTRUCTION	0x07f001f8
+#define KPROBE_THUMB16_BREAKPOINT_INSTRUCTION	0xde18
+#define KPROBE_THUMB32_BREAKPOINT_INSTRUCTION	0xf7f0a018
+
+extern void kprobes_remove_breakpoint(void *addr, unsigned int insn);
+
+enum probes_insn __kprobes
+kprobe_decode_ldmstm(kprobe_opcode_t insn, struct arch_probes_insn *asi,
+		const struct decode_header *h);
+
+typedef enum probes_insn (kprobe_decode_insn_t)(probes_opcode_t,
+						struct arch_probes_insn *,
+						bool,
+						const union decode_action *,
+						const struct decode_checker *[]);
+
+#ifdef CONFIG_THUMB2_KERNEL
+
+extern const union decode_action kprobes_t32_actions[];
+extern const union decode_action kprobes_t16_actions[];
+extern const struct decode_checker *kprobes_t32_checkers[];
+extern const struct decode_checker *kprobes_t16_checkers[];
+#else /* !CONFIG_THUMB2_KERNEL */
+
+extern const union decode_action kprobes_arm_actions[];
+extern const struct decode_checker *kprobes_arm_checkers[];
+
+#endif
+
+#endif /* _ARM_KERNEL_KPROBES_H */

arch/arm/probes/kprobes/opt-arm.c

@@ -0,0 +1,372 @@
+/*
+ *  Kernel Probes Jump Optimization (Optprobes)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2002, 2004
+ * Copyright (C) Hitachi Ltd., 2012
+ * Copyright (C) Huawei Inc., 2014
+ */
+
+#include <linux/kprobes.h>
+#include <linux/jump_label.h>
+#include <asm/kprobes.h>
+#include <asm/cacheflush.h>
+/* for arm_gen_branch */
+#include <asm/insn.h>
+/* for patch_text */
+#include <asm/patch.h>
+
+#include "core.h"
+
+/*
+ * See register_usage_flags. If the probed instruction doesn't use PC,
+ * we can copy it into template and have it executed directly without
+ * simulation or emulation.
+ */
+#define ARM_REG_PC	15
+#define can_kprobe_direct_exec(m)	(!test_bit(ARM_REG_PC, &(m)))
+
+/*
+ * NOTE: the first sub and add instruction will be modified according
+ * to the stack cost of the instruction.
+ */
+asm (
+			".global optprobe_template_entry\n"
+			"optprobe_template_entry:\n"
+			".global optprobe_template_sub_sp\n"
+			"optprobe_template_sub_sp:"
+			"	sub	sp, sp, #0xff\n"
+			"	stmia	sp, {r0 - r14} \n"
+			".global optprobe_template_add_sp\n"
+			"optprobe_template_add_sp:"
+			"	add	r3, sp, #0xff\n"
+			"	str	r3, [sp, #52]\n"
+			"	mrs	r4, cpsr\n"
+			"	str	r4, [sp, #64]\n"
+			"	mov	r1, sp\n"
+			"	ldr	r0, 1f\n"
+			"	ldr	r2, 2f\n"
+			/*
+			 * AEABI requires an 8-bytes alignment stack. If
+			 * SP % 8 != 0 (SP % 4 == 0 should be ensured),
+			 * alloc more bytes here.
+			 */
+			"	and	r4, sp, #4\n"
+			"	sub	sp, sp, r4\n"
+#if __LINUX_ARM_ARCH__ >= 5
+			"	blx	r2\n"
+#else
+			"	mov     lr, pc\n"
+			"	mov	pc, r2\n"
+#endif
+			"	add	sp, sp, r4\n"
+			"	ldr	r1, [sp, #64]\n"
+			"	tst	r1, #"__stringify(PSR_T_BIT)"\n"
+			"	ldrne	r2, [sp, #60]\n"
+			"	orrne	r2, #1\n"
+			"	strne	r2, [sp, #60] @ set bit0 of PC for thumb\n"
+			"	msr	cpsr_cxsf, r1\n"
+			".global optprobe_template_restore_begin\n"
+			"optprobe_template_restore_begin:\n"
+			"	ldmia	sp, {r0 - r15}\n"
+			".global optprobe_template_restore_orig_insn\n"
+			"optprobe_template_restore_orig_insn:\n"
+			"	nop\n"
+			".global optprobe_template_restore_end\n"
+			"optprobe_template_restore_end:\n"
+			"	nop\n"
+			".global optprobe_template_val\n"
+			"optprobe_template_val:\n"
+			"1:	.long 0\n"
+			".global optprobe_template_call\n"
+			"optprobe_template_call:\n"
+			"2:	.long 0\n"
+			".global optprobe_template_end\n"
+			"optprobe_template_end:\n");
+
+#define TMPL_VAL_IDX \
+	((unsigned long *)&optprobe_template_val - (unsigned long *)&optprobe_template_entry)
+#define TMPL_CALL_IDX \
+	((unsigned long *)&optprobe_template_call - (unsigned long *)&optprobe_template_entry)
+#define TMPL_END_IDX \
+	((unsigned long *)&optprobe_template_end - (unsigned long *)&optprobe_template_entry)
+#define TMPL_ADD_SP \
+	((unsigned long *)&optprobe_template_add_sp - (unsigned long *)&optprobe_template_entry)
+#define TMPL_SUB_SP \
+	((unsigned long *)&optprobe_template_sub_sp - (unsigned long *)&optprobe_template_entry)
+#define TMPL_RESTORE_BEGIN \
+	((unsigned long *)&optprobe_template_restore_begin - (unsigned long *)&optprobe_template_entry)
+#define TMPL_RESTORE_ORIGN_INSN \
+	((unsigned long *)&optprobe_template_restore_orig_insn - (unsigned long *)&optprobe_template_entry)
+#define TMPL_RESTORE_END \
+	((unsigned long *)&optprobe_template_restore_end - (unsigned long *)&optprobe_template_entry)
+
+/*
+ * ARM can always optimize an instruction when using ARM ISA, except
+ * instructions like 'str r0, [sp, r1]' which store to stack and unable
+ * to determine stack space consumption statically.
+ */
+int arch_prepared_optinsn(struct arch_optimized_insn *optinsn)
+{
+	return optinsn->insn != NULL;
+}
+
+/*
+ * In ARM ISA, kprobe opt always replace one instruction (4 bytes
+ * aligned and 4 bytes long). It is impossible to encounter another
+ * kprobe in the address range. So always return 0.
+ */
+int arch_check_optimized_kprobe(struct optimized_kprobe *op)
+{
+	return 0;
+}
+
+/* Caller must ensure addr & 3 == 0 */
+static int can_optimize(struct kprobe *kp)
+{
+	if (kp->ainsn.stack_space < 0)
+		return 0;
+	/*
+	 * 255 is the biggest imm can be used in 'sub r0, r0, #<imm>'.
+	 * Number larger than 255 needs special encoding.
+	 */
+	if (kp->ainsn.stack_space > 255 - sizeof(struct pt_regs))
+		return 0;
+	return 1;
+}
+
+/* Free optimized instruction slot */
+static void
+__arch_remove_optimized_kprobe(struct optimized_kprobe *op, int dirty)
+{
+	if (op->optinsn.insn) {
+		free_optinsn_slot(op->optinsn.insn, dirty);
+		op->optinsn.insn = NULL;
+	}
+}
+
+extern void kprobe_handler(struct pt_regs *regs);
+
+static void
+optimized_callback(struct optimized_kprobe *op, struct pt_regs *regs)
+{
+	unsigned long flags;
+	struct kprobe *p = &op->kp;
+	struct kprobe_ctlblk *kcb;
+
+	/* Save skipped registers */
+	regs->ARM_pc = (unsigned long)op->kp.addr;
+	regs->ARM_ORIG_r0 = ~0UL;
+
+	local_irq_save(flags);
+	kcb = get_kprobe_ctlblk();
+
+	if (kprobe_running()) {
+		kprobes_inc_nmissed_count(&op->kp);
+	} else {
+		__this_cpu_write(current_kprobe, &op->kp);
+		kcb->kprobe_status = KPROBE_HIT_ACTIVE;
+		opt_pre_handler(&op->kp, regs);
+		__this_cpu_write(current_kprobe, NULL);
+	}
+
+	/*
+	 * We singlestep the replaced instruction only when it can't be
+	 * executed directly during restore.
+	 */
+	if (!p->ainsn.kprobe_direct_exec)
+		op->kp.ainsn.insn_singlestep(p->opcode, &p->ainsn, regs);
+
+	local_irq_restore(flags);
+}
+NOKPROBE_SYMBOL(optimized_callback)
+
+int arch_prepare_optimized_kprobe(struct optimized_kprobe *op, struct kprobe *orig)
+{
+	kprobe_opcode_t *code;
+	unsigned long rel_chk;
+	unsigned long val;
+	unsigned long stack_protect = sizeof(struct pt_regs);
+
+	if (!can_optimize(orig))
+		return -EILSEQ;
+
+	code = get_optinsn_slot();
+	if (!code)
+		return -ENOMEM;
+
+	/*
+	 * Verify if the address gap is in 32MiB range, because this uses
+	 * a relative jump.
+	 *
+	 * kprobe opt use a 'b' instruction to branch to optinsn.insn.
+	 * According to ARM manual, branch instruction is:
+	 *
+	 *   31  28 27           24 23             0
+	 *  +------+---+---+---+---+----------------+
+	 *  | cond | 1 | 0 | 1 | 0 |      imm24     |
+	 *  +------+---+---+---+---+----------------+
+	 *
+	 * imm24 is a signed 24 bits integer. The real branch offset is computed
+	 * by: imm32 = SignExtend(imm24:'00', 32);
+	 *
+	 * So the maximum forward branch should be:
+	 *   (0x007fffff << 2) = 0x01fffffc =  0x1fffffc
+	 * The maximum backword branch should be:
+	 *   (0xff800000 << 2) = 0xfe000000 = -0x2000000
+	 *
+	 * We can simply check (rel & 0xfe000003):
+	 *  if rel is positive, (rel & 0xfe000000) shoule be 0
+	 *  if rel is negitive, (rel & 0xfe000000) should be 0xfe000000
+	 *  the last '3' is used for alignment checking.
+	 */
+	rel_chk = (unsigned long)((long)code -
+			(long)orig->addr + 8) & 0xfe000003;
+
+	if ((rel_chk != 0) && (rel_chk != 0xfe000000)) {
+		/*
+		 * Different from x86, we free code buf directly instead of
+		 * calling __arch_remove_optimized_kprobe() because
+		 * we have not fill any field in op.
+		 */
+		free_optinsn_slot(code, 0);
+		return -ERANGE;
+	}
+
+	/* Copy arch-dep-instance from template. */
+	memcpy(code, &optprobe_template_entry,
+			TMPL_END_IDX * sizeof(kprobe_opcode_t));
+
+	/* Adjust buffer according to instruction. */
+	BUG_ON(orig->ainsn.stack_space < 0);
+
+	stack_protect += orig->ainsn.stack_space;
+
+	/* Should have been filtered by can_optimize(). */
+	BUG_ON(stack_protect > 255);
+
+	/* Create a 'sub sp, sp, #<stack_protect>' */
+	code[TMPL_SUB_SP] = __opcode_to_mem_arm(0xe24dd000 | stack_protect);
+	/* Create a 'add r3, sp, #<stack_protect>' */
+	code[TMPL_ADD_SP] = __opcode_to_mem_arm(0xe28d3000 | stack_protect);
+
+	/* Set probe information */
+	val = (unsigned long)op;
+	code[TMPL_VAL_IDX] = val;
+
+	/* Set probe function call */
+	val = (unsigned long)optimized_callback;
+	code[TMPL_CALL_IDX] = val;
+
+	/* If possible, copy insn and have it executed during restore */
+	orig->ainsn.kprobe_direct_exec = false;
+	if (can_kprobe_direct_exec(orig->ainsn.register_usage_flags)) {
+		kprobe_opcode_t final_branch = arm_gen_branch(
+				(unsigned long)(&code[TMPL_RESTORE_END]),
+				(unsigned long)(op->kp.addr) + 4);
+		if (final_branch != 0) {
+			/*
+			 * Replace original 'ldmia sp, {r0 - r15}' with
+			 * 'ldmia {r0 - r14}', restore all registers except pc.
+			 */
+			code[TMPL_RESTORE_BEGIN] = __opcode_to_mem_arm(0xe89d7fff);
+
+			/* The original probed instruction */
+			code[TMPL_RESTORE_ORIGN_INSN] = __opcode_to_mem_arm(orig->opcode);
+
+			/* Jump back to next instruction */
+			code[TMPL_RESTORE_END] = __opcode_to_mem_arm(final_branch);
+			orig->ainsn.kprobe_direct_exec = true;
+		}
+	}
+
+	flush_icache_range((unsigned long)code,
+			   (unsigned long)(&code[TMPL_END_IDX]));
+
+	/* Set op->optinsn.insn means prepared. */
+	op->optinsn.insn = code;
+	return 0;
+}
+
+void __kprobes arch_optimize_kprobes(struct list_head *oplist)
+{
+	struct optimized_kprobe *op, *tmp;
+
+	list_for_each_entry_safe(op, tmp, oplist, list) {
+		unsigned long insn;
+		WARN_ON(kprobe_disabled(&op->kp));
+
+		/*
+		 * Backup instructions which will be replaced
+		 * by jump address
+		 */
+		memcpy(op->optinsn.copied_insn, op->kp.addr,
+				RELATIVEJUMP_SIZE);
+
+		insn = arm_gen_branch((unsigned long)op->kp.addr,
+				(unsigned long)op->optinsn.insn);
+		BUG_ON(insn == 0);
+
+		/*
+		 * Make it a conditional branch if replaced insn
+		 * is consitional
+		 */
+		insn = (__mem_to_opcode_arm(
+			  op->optinsn.copied_insn[0]) & 0xf0000000) |
+			(insn & 0x0fffffff);
+
+		/*
+		 * Similar to __arch_disarm_kprobe, operations which
+		 * removing breakpoints must be wrapped by stop_machine
+		 * to avoid racing.
+		 */
+		kprobes_remove_breakpoint(op->kp.addr, insn);
+
+		list_del_init(&op->list);
+	}
+}
+
+void arch_unoptimize_kprobe(struct optimized_kprobe *op)
+{
+	arch_arm_kprobe(&op->kp);
+}
+
+/*
+ * Recover original instructions and breakpoints from relative jumps.
+ * Caller must call with locking kprobe_mutex.
+ */
+void arch_unoptimize_kprobes(struct list_head *oplist,
+			    struct list_head *done_list)
+{
+	struct optimized_kprobe *op, *tmp;
+
+	list_for_each_entry_safe(op, tmp, oplist, list) {
+		arch_unoptimize_kprobe(op);
+		list_move(&op->list, done_list);
+	}
+}
+
+int arch_within_optimized_kprobe(struct optimized_kprobe *op,
+				unsigned long addr)
+{
+	return ((unsigned long)op->kp.addr <= addr &&
+		(unsigned long)op->kp.addr + RELATIVEJUMP_SIZE > addr);
+}
+
+void arch_remove_optimized_kprobe(struct optimized_kprobe *op)
+{
+	__arch_remove_optimized_kprobe(op, 1);
+}

arch/arm/probes/kprobes/test-arm.c

@@ -0,0 +1,1366 @@
+/*
+ * arch/arm/kernel/kprobes-test-arm.c
+ *
+ * Copyright (C) 2011 Jon Medhurst <tixy@yxit.co.uk>.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <asm/system_info.h>
+#include <asm/opcodes.h>
+#include <asm/probes.h>
+
+#include "test-core.h"
+
+
+#define TEST_ISA "32"
+
+#define TEST_ARM_TO_THUMB_INTERWORK_R(code1, reg, val, code2)	\
+	TESTCASE_START(code1 #reg code2)			\
+	TEST_ARG_REG(reg, val)					\
+	TEST_ARG_REG(14, 99f)					\
+	TEST_ARG_END("")					\
+	"50:	nop			\n\t"			\
+	"1:	"code1 #reg code2"	\n\t"			\
+	"	bx	lr		\n\t"			\
+	".thumb				\n\t"			\
+	"3:	adr	lr, 2f		\n\t"			\
+	"	bx	lr		\n\t"			\
+	".arm				\n\t"			\
+	"2:	nop			\n\t"			\
+	TESTCASE_END
+
+#define TEST_ARM_TO_THUMB_INTERWORK_P(code1, reg, val, code2)	\
+	TESTCASE_START(code1 #reg code2)			\
+	TEST_ARG_PTR(reg, val)					\
+	TEST_ARG_REG(14, 99f)					\
+	TEST_ARG_MEM(15, 3f+1)					\
+	TEST_ARG_END("")					\
+	"50:	nop			\n\t"			\
+	"1:	"code1 #reg code2"	\n\t"			\
+	"	bx	lr		\n\t"			\
+	".thumb				\n\t"			\
+	"3:	adr	lr, 2f		\n\t"			\
+	"	bx	lr		\n\t"			\
+	".arm				\n\t"			\
+	"2:	nop			\n\t"			\
+	TESTCASE_END
+
+
+void kprobe_arm_test_cases(void)
+{
+	kprobe_test_flags = 0;
+
+	TEST_GROUP("Data-processing (register), (register-shifted register), (immediate)")
+
+#define _DATA_PROCESSING_DNM(op,s,val)						\
+	TEST_RR(  op "eq" s "	r0,  r",1, VAL1,", r",2, val, "")		\
+	TEST_RR(  op "ne" s "	r1,  r",1, VAL1,", r",2, val, ", lsl #3")	\
+	TEST_RR(  op "cs" s "	r2,  r",3, VAL1,", r",2, val, ", lsr #4")	\
+	TEST_RR(  op "cc" s "	r3,  r",3, VAL1,", r",2, val, ", asr #5")	\
+	TEST_RR(  op "mi" s "	r4,  r",5, VAL1,", r",2, N(val),", asr #6")	\
+	TEST_RR(  op "pl" s "	r5,  r",5, VAL1,", r",2, val, ", ror #7")	\
+	TEST_RR(  op "vs" s "	r6,  r",7, VAL1,", r",2, val, ", rrx")		\
+	TEST_R(   op "vc" s "	r6,  r",7, VAL1,", pc, lsl #3")			\
+	TEST_R(   op "vc" s "	r6,  r",7, VAL1,", sp, lsr #4")			\
+	TEST_R(   op "vc" s "	r6,  pc, r",7, VAL1,", asr #5")			\
+	TEST_R(   op "vc" s "	r6,  sp, r",7, VAL1,", ror #6")			\
+	TEST_RRR( op "hi" s "	r8,  r",9, VAL1,", r",14,val, ", lsl r",0, 3,"")\
+	TEST_RRR( op "ls" s "	r9,  r",9, VAL1,", r",14,val, ", lsr r",7, 4,"")\
+	TEST_RRR( op "ge" s "	r10, r",11,VAL1,", r",14,val, ", asr r",7, 5,"")\
+	TEST_RRR( op "lt" s "	r11, r",11,VAL1,", r",14,N(val),", asr r",7, 6,"")\
+	TEST_RR(  op "gt" s "	r12, r13"       ", r",14,val, ", ror r",14,7,"")\
+	TEST_RR(  op "le" s "	r14, r",0, val, ", r13"       ", lsl r",14,8,"")\
+	TEST_R(   op "eq" s "	r0,  r",11,VAL1,", #0xf5")			\
+	TEST_R(   op "ne" s "	r11, r",0, VAL1,", #0xf5000000")		\
+	TEST_R(   op s "	r7,  r",8, VAL2,", #0x000af000")		\
+	TEST(     op s "	r4,  pc"        ", #0x00005a00")
+
+#define DATA_PROCESSING_DNM(op,val)		\
+	_DATA_PROCESSING_DNM(op,"",val)		\
+	_DATA_PROCESSING_DNM(op,"s",val)
+
+#define DATA_PROCESSING_NM(op,val)						\
+	TEST_RR(  op "ne	r",1, VAL1,", r",2, val, "")			\
+	TEST_RR(  op "eq	r",1, VAL1,", r",2, val, ", lsl #3")		\
+	TEST_RR(  op "cc	r",3, VAL1,", r",2, val, ", lsr #4")		\
+	TEST_RR(  op "cs	r",3, VAL1,", r",2, val, ", asr #5")		\
+	TEST_RR(  op "pl	r",5, VAL1,", r",2, N(val),", asr #6")		\
+	TEST_RR(  op "mi	r",5, VAL1,", r",2, val, ", ror #7")		\
+	TEST_RR(  op "vc	r",7, VAL1,", r",2, val, ", rrx")		\
+	TEST_R (  op "vs	r",7, VAL1,", pc, lsl #3")			\
+	TEST_R (  op "vs	r",7, VAL1,", sp, lsr #4")			\
+	TEST_R(   op "vs	pc, r",7, VAL1,", asr #5")			\
+	TEST_R(   op "vs	sp, r",7, VAL1,", ror #6")			\
+	TEST_RRR( op "ls	r",9, VAL1,", r",14,val, ", lsl r",0, 3,"")	\
+	TEST_RRR( op "hi	r",9, VAL1,", r",14,val, ", lsr r",7, 4,"")	\
+	TEST_RRR( op "lt	r",11,VAL1,", r",14,val, ", asr r",7, 5,"")	\
+	TEST_RRR( op "ge	r",11,VAL1,", r",14,N(val),", asr r",7, 6,"")	\
+	TEST_RR(  op "le	r13"       ", r",14,val, ", ror r",14,7,"")	\
+	TEST_RR(  op "gt	r",0, val, ", r13"       ", lsl r",14,8,"")	\
+	TEST_R(   op "eq	r",11,VAL1,", #0xf5")				\
+	TEST_R(   op "ne	r",0, VAL1,", #0xf5000000")			\
+	TEST_R(   op "	r",8, VAL2,", #0x000af000")
+
+#define _DATA_PROCESSING_DM(op,s,val)					\
+	TEST_R(   op "eq" s "	r0,  r",1, val, "")			\
+	TEST_R(   op "ne" s "	r1,  r",1, val, ", lsl #3")		\
+	TEST_R(   op "cs" s "	r2,  r",3, val, ", lsr #4")		\
+	TEST_R(   op "cc" s "	r3,  r",3, val, ", asr #5")		\
+	TEST_R(   op "mi" s "	r4,  r",5, N(val),", asr #6")		\
+	TEST_R(   op "pl" s "	r5,  r",5, val, ", ror #7")		\
+	TEST_R(   op "vs" s "	r6,  r",10,val, ", rrx")		\
+	TEST(     op "vs" s "	r7,  pc, lsl #3")			\
+	TEST(     op "vs" s "	r7,  sp, lsr #4")			\
+	TEST_RR(  op "vc" s "	r8,  r",7, val, ", lsl r",0, 3,"")	\
+	TEST_RR(  op "hi" s "	r9,  r",9, val, ", lsr r",7, 4,"")	\
+	TEST_RR(  op "ls" s "	r10, r",9, val, ", asr r",7, 5,"")	\
+	TEST_RR(  op "ge" s "	r11, r",11,N(val),", asr r",7, 6,"")	\
+	TEST_RR(  op "lt" s "	r12, r",11,val, ", ror r",14,7,"")	\
+	TEST_R(   op "gt" s "	r14, r13"       ", lsl r",14,8,"")	\
+	TEST(     op "eq" s "	r0,  #0xf5")				\
+	TEST(     op "ne" s "	r11, #0xf5000000")			\
+	TEST(     op s "	r7,  #0x000af000")			\
+	TEST(     op s "	r4,  #0x00005a00")
+
+#define DATA_PROCESSING_DM(op,val)		\
+	_DATA_PROCESSING_DM(op,"",val)		\
+	_DATA_PROCESSING_DM(op,"s",val)
+
+	DATA_PROCESSING_DNM("and",0xf00f00ff)
+	DATA_PROCESSING_DNM("eor",0xf00f00ff)
+	DATA_PROCESSING_DNM("sub",VAL2)
+	DATA_PROCESSING_DNM("rsb",VAL2)
+	DATA_PROCESSING_DNM("add",VAL2)
+	DATA_PROCESSING_DNM("adc",VAL2)
+	DATA_PROCESSING_DNM("sbc",VAL2)
+	DATA_PROCESSING_DNM("rsc",VAL2)
+	DATA_PROCESSING_NM("tst",0xf00f00ff)
+	DATA_PROCESSING_NM("teq",0xf00f00ff)
+	DATA_PROCESSING_NM("cmp",VAL2)
+	DATA_PROCESSING_NM("cmn",VAL2)
+	DATA_PROCESSING_DNM("orr",0xf00f00ff)
+	DATA_PROCESSING_DM("mov",VAL2)
+	DATA_PROCESSING_DNM("bic",0xf00f00ff)
+	DATA_PROCESSING_DM("mvn",VAL2)
+
+	TEST("mov	ip, sp") /* This has special case emulation code */
+
+	TEST_SUPPORTED("mov	pc, #0x1000");
+	TEST_SUPPORTED("mov	sp, #0x1000");
+	TEST_SUPPORTED("cmp	pc, #0x1000");
+	TEST_SUPPORTED("cmp	sp, #0x1000");
+
+	/* Data-processing with PC and a shift count in a register */
+	TEST_UNSUPPORTED(__inst_arm(0xe15c0f1e) "	@ cmp	r12, r14, asl pc")
+	TEST_UNSUPPORTED(__inst_arm(0xe1a0cf1e) "	@ mov	r12, r14, asl pc")
+	TEST_UNSUPPORTED(__inst_arm(0xe08caf1e) "	@ add	r10, r12, r14, asl pc")
+	TEST_UNSUPPORTED(__inst_arm(0xe151021f) "	@ cmp	r1, pc, lsl r2")
+	TEST_UNSUPPORTED(__inst_arm(0xe17f0211) "	@ cmn	pc, r1, lsl r2")
+	TEST_UNSUPPORTED(__inst_arm(0xe1a0121f) "	@ mov	r1, pc, lsl r2")
+	TEST_UNSUPPORTED(__inst_arm(0xe1a0f211) "	@ mov	pc, r1, lsl r2")
+	TEST_UNSUPPORTED(__inst_arm(0xe042131f) "	@ sub	r1, r2, pc, lsl r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe1cf1312) "	@ bic	r1, pc, r2, lsl r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe081f312) "	@ add	pc, r1, r2, lsl r3")
+
+	/* Data-processing with PC as a target and status registers updated */
+	TEST_UNSUPPORTED("movs	pc, r1")
+	TEST_UNSUPPORTED("movs	pc, r1, lsl r2")
+	TEST_UNSUPPORTED("movs	pc, #0x10000")
+	TEST_UNSUPPORTED("adds	pc, lr, r1")
+	TEST_UNSUPPORTED("adds	pc, lr, r1, lsl r2")
+	TEST_UNSUPPORTED("adds	pc, lr, #4")
+
+	/* Data-processing with SP as target */
+	TEST("add	sp, sp, #16")
+	TEST("sub	sp, sp, #8")
+	TEST("bic	sp, sp, #0x20")
+	TEST("orr	sp, sp, #0x20")
+	TEST_PR( "add	sp, r",10,0,", r",11,4,"")
+	TEST_PRR("add	sp, r",10,0,", r",11,4,", asl r",12,1,"")
+	TEST_P(  "mov	sp, r",10,0,"")
+	TEST_PR( "mov	sp, r",10,0,", asl r",12,0,"")
+
+	/* Data-processing with PC as target */
+	TEST_BF(   "add	pc, pc, #2f-1b-8")
+	TEST_BF_R ("add	pc, pc, r",14,2f-1f-8,"")
+	TEST_BF_R ("add	pc, r",14,2f-1f-8,", pc")
+	TEST_BF_R ("mov	pc, r",0,2f,"")
+	TEST_BF_R ("add	pc, pc, r",14,(2f-1f-8)*2,", asr #1")
+	TEST_BB(   "sub	pc, pc, #1b-2b+8")
+#if __LINUX_ARM_ARCH__ == 6 && !defined(CONFIG_CPU_V7)
+	TEST_BB(   "sub	pc, pc, #1b-2b+8-2") /* UNPREDICTABLE before and after ARMv6 */
+#endif
+	TEST_BB_R( "sub	pc, pc, r",14, 1f-2f+8,"")
+	TEST_BB_R( "rsb	pc, r",14,1f-2f+8,", pc")
+	TEST_R(    "add	pc, pc, r",10,-2,", asl #1")
+#ifdef CONFIG_THUMB2_KERNEL
+	TEST_ARM_TO_THUMB_INTERWORK_R("add	pc, pc, r",0,3f-1f-8+1,"")
+	TEST_ARM_TO_THUMB_INTERWORK_R("sub	pc, r",0,3f+8+1,", #8")
+#endif
+	TEST_GROUP("Miscellaneous instructions")
+
+	TEST_RMASKED("mrs	r",0,~PSR_IGNORE_BITS,", cpsr")
+	TEST_RMASKED("mrspl	r",7,~PSR_IGNORE_BITS,", cpsr")
+	TEST_RMASKED("mrs	r",14,~PSR_IGNORE_BITS,", cpsr")
+	TEST_UNSUPPORTED(__inst_arm(0xe10ff000) "	@ mrs r15, cpsr")
+	TEST_UNSUPPORTED("mrs	r0, spsr")
+	TEST_UNSUPPORTED("mrs	lr, spsr")
+
+	TEST_UNSUPPORTED("msr	cpsr, r0")
+	TEST_UNSUPPORTED("msr	cpsr_f, lr")
+	TEST_UNSUPPORTED("msr	spsr, r0")
+
+#if __LINUX_ARM_ARCH__ >= 5 || \
+    (__LINUX_ARM_ARCH__ == 4 && !defined(CONFIG_CPU_32v4))
+	TEST_BF_R("bx	r",0,2f,"")
+	TEST_BB_R("bx	r",7,2f,"")
+	TEST_BF_R("bxeq	r",14,2f,"")
+#endif
+
+#if __LINUX_ARM_ARCH__ >= 5
+	TEST_R("clz	r0, r",0, 0x0,"")
+	TEST_R("clzeq	r7, r",14,0x1,"")
+	TEST_R("clz	lr, r",7, 0xffffffff,"")
+	TEST(  "clz	r4, sp")
+	TEST_UNSUPPORTED(__inst_arm(0x016fff10) "	@ clz pc, r0")
+	TEST_UNSUPPORTED(__inst_arm(0x016f0f1f) "	@ clz r0, pc")
+
+#if __LINUX_ARM_ARCH__ >= 6
+	TEST_UNSUPPORTED("bxj	r0")
+#endif
+
+	TEST_BF_R("blx	r",0,2f,"")
+	TEST_BB_R("blx	r",7,2f,"")
+	TEST_BF_R("blxeq	r",14,2f,"")
+	TEST_UNSUPPORTED(__inst_arm(0x0120003f) "	@ blx pc")
+
+	TEST_RR(   "qadd	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(   "qaddvs	lr, r",9, VAL2,", r",8, VAL1,"")
+	TEST_R(    "qadd	lr, r",9, VAL2,", r13")
+	TEST_RR(   "qsub	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(   "qsubvs	lr, r",9, VAL2,", r",8, VAL1,"")
+	TEST_R(    "qsub	lr, r",9, VAL2,", r13")
+	TEST_RR(   "qdadd	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(   "qdaddvs	lr, r",9, VAL2,", r",8, VAL1,"")
+	TEST_R(    "qdadd	lr, r",9, VAL2,", r13")
+	TEST_RR(   "qdsub	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(   "qdsubvs	lr, r",9, VAL2,", r",8, VAL1,"")
+	TEST_R(    "qdsub	lr, r",9, VAL2,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe101f050) "	@ qadd pc, r0, r1")
+	TEST_UNSUPPORTED(__inst_arm(0xe121f050) "	@ qsub pc, r0, r1")
+	TEST_UNSUPPORTED(__inst_arm(0xe141f050) "	@ qdadd pc, r0, r1")
+	TEST_UNSUPPORTED(__inst_arm(0xe161f050) "	@ qdsub pc, r0, r1")
+	TEST_UNSUPPORTED(__inst_arm(0xe16f2050) "	@ qdsub r2, r0, pc")
+	TEST_UNSUPPORTED(__inst_arm(0xe161205f) "	@ qdsub r2, pc, r1")
+
+	TEST_UNSUPPORTED("bkpt	0xffff")
+	TEST_UNSUPPORTED("bkpt	0x0000")
+
+	TEST_UNSUPPORTED(__inst_arm(0xe1600070) " @ smc #0")
+
+	TEST_GROUP("Halfword multiply and multiply-accumulate")
+
+	TEST_RRR(    "smlabb	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(    "smlabbge	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RR(     "smlabb	lr, r",1, VAL2,", r",2, VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe10f3281) " @ smlabb pc, r1, r2, r3")
+	TEST_RRR(    "smlatb	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(    "smlatbge	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RR(     "smlatb	lr, r",1, VAL2,", r",2, VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe10f32a1) " @ smlatb pc, r1, r2, r3")
+	TEST_RRR(    "smlabt	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(    "smlabtge	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RR(     "smlabt	lr, r",1, VAL2,", r",2, VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe10f32c1) " @ smlabt pc, r1, r2, r3")
+	TEST_RRR(    "smlatt	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(    "smlattge	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RR(     "smlatt	lr, r",1, VAL2,", r",2, VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe10f32e1) " @ smlatt pc, r1, r2, r3")
+
+	TEST_RRR(    "smlawb	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(    "smlawbge	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RR(     "smlawb	lr, r",1, VAL2,", r",2, VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe12f3281) " @ smlawb pc, r1, r2, r3")
+	TEST_RRR(    "smlawt	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(    "smlawtge	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RR(     "smlawt	lr, r",1, VAL2,", r",2, VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe12f32c1) " @ smlawt pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe12032cf) " @ smlawt r0, pc, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe1203fc1) " @ smlawt r0, r1, pc, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe120f2c1) " @ smlawt r0, r1, r2, pc")
+
+	TEST_RR(    "smulwb	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "smulwbge	r7, r",8, VAL3,", r",9, VAL1,"")
+	TEST_R(     "smulwb	lr, r",1, VAL2,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe12f02a1) " @ smulwb pc, r1, r2")
+	TEST_RR(    "smulwt	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "smulwtge	r7, r",8, VAL3,", r",9, VAL1,"")
+	TEST_R(     "smulwt	lr, r",1, VAL2,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe12f02e1) " @ smulwt pc, r1, r2")
+
+	TEST_RRRR(  "smlalbb	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR(  "smlalbble	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+	TEST_RRR(   "smlalbb	r",14,VAL3,", r",7, VAL4,", r",5, VAL1,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe14f1382) " @ smlalbb pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe141f382) " @ smlalbb r1, pc, r2, r3")
+	TEST_RRRR(  "smlaltb	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR(  "smlaltble	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+	TEST_RRR(   "smlaltb	r",14,VAL3,", r",7, VAL4,", r",5, VAL1,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe14f13a2) " @ smlaltb pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe141f3a2) " @ smlaltb r1, pc, r2, r3")
+	TEST_RRRR(  "smlalbt	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR(  "smlalbtle	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+	TEST_RRR(   "smlalbt	r",14,VAL3,", r",7, VAL4,", r",5, VAL1,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe14f13c2) " @ smlalbt pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe141f3c2) " @ smlalbt r1, pc, r2, r3")
+	TEST_RRRR(  "smlaltt	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR(  "smlalttle	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+	TEST_RRR(   "smlaltt	r",14,VAL3,", r",7, VAL4,", r",5, VAL1,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe14f13e2) " @ smlalbb pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe140f3e2) " @ smlalbb r0, pc, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe14013ef) " @ smlalbb r0, r1, pc, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe1401fe2) " @ smlalbb r0, r1, r2, pc")
+
+	TEST_RR(    "smulbb	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "smulbbge	r7, r",8, VAL3,", r",9, VAL1,"")
+	TEST_R(     "smulbb	lr, r",1, VAL2,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe16f0281) " @ smulbb pc, r1, r2")
+	TEST_RR(    "smultb	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "smultbge	r7, r",8, VAL3,", r",9, VAL1,"")
+	TEST_R(     "smultb	lr, r",1, VAL2,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe16f02a1) " @ smultb pc, r1, r2")
+	TEST_RR(    "smulbt	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "smulbtge	r7, r",8, VAL3,", r",9, VAL1,"")
+	TEST_R(     "smulbt	lr, r",1, VAL2,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe16f02c1) " @ smultb pc, r1, r2")
+	TEST_RR(    "smultt	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "smulttge	r7, r",8, VAL3,", r",9, VAL1,"")
+	TEST_R(     "smultt	lr, r",1, VAL2,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe16f02e1) " @ smultt pc, r1, r2")
+	TEST_UNSUPPORTED(__inst_arm(0xe16002ef) " @ smultt r0, pc, r2")
+	TEST_UNSUPPORTED(__inst_arm(0xe1600fe1) " @ smultt r0, r1, pc")
+#endif
+
+	TEST_GROUP("Multiply and multiply-accumulate")
+
+	TEST_RR(    "mul	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "mulls	r7, r",8, VAL2,", r",9, VAL2,"")
+	TEST_R(     "mul	lr, r",4, VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe00f0291) " @ mul pc, r1, r2")
+	TEST_UNSUPPORTED(__inst_arm(0xe000029f) " @ mul r0, pc, r2")
+	TEST_UNSUPPORTED(__inst_arm(0xe0000f91) " @ mul r0, r1, pc")
+	TEST_RR(    "muls	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "mullss	r7, r",8, VAL2,", r",9, VAL2,"")
+	TEST_R(     "muls	lr, r",4, VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe01f0291) " @ muls pc, r1, r2")
+
+	TEST_RRR(    "mla	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(    "mlahi	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RR(     "mla	lr, r",1, VAL2,", r",2, VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe02f3291) " @ mla pc, r1, r2, r3")
+	TEST_RRR(    "mlas	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(    "mlahis	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RR(     "mlas	lr, r",1, VAL2,", r",2, VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe03f3291) " @ mlas pc, r1, r2, r3")
+
+#if __LINUX_ARM_ARCH__ >= 6
+	TEST_RR(  "umaal	r0, r1, r",2, VAL1,", r",3, VAL2,"")
+	TEST_RR(  "umaalls	r7, r8, r",9, VAL2,", r",10, VAL1,"")
+	TEST_R(   "umaal	lr, r12, r",11,VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe041f392) " @ umaal pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe04f0392) " @ umaal r0, pc, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe0500090) " @ undef")
+	TEST_UNSUPPORTED(__inst_arm(0xe05fff9f) " @ undef")
+#endif
+
+#if __LINUX_ARM_ARCH__ >= 7
+	TEST_RRR(  "mls		r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(  "mlshi	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RR(   "mls		lr, r",1, VAL2,", r",2, VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe06f3291) " @ mls pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe060329f) " @ mls r0, pc, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe0603f91) " @ mls r0, r1, pc, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe060f291) " @ mls r0, r1, r2, pc")
+#endif
+
+	TEST_UNSUPPORTED(__inst_arm(0xe0700090) " @ undef")
+	TEST_UNSUPPORTED(__inst_arm(0xe07fff9f) " @ undef")
+
+	TEST_RR(  "umull	r0, r1, r",2, VAL1,", r",3, VAL2,"")
+	TEST_RR(  "umullls	r7, r8, r",9, VAL2,", r",10, VAL1,"")
+	TEST_R(   "umull	lr, r12, r",11,VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe081f392) " @ umull pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe08f1392) " @ umull r1, pc, r2, r3")
+	TEST_RR(  "umulls	r0, r1, r",2, VAL1,", r",3, VAL2,"")
+	TEST_RR(  "umulllss	r7, r8, r",9, VAL2,", r",10, VAL1,"")
+	TEST_R(   "umulls	lr, r12, r",11,VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe091f392) " @ umulls pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe09f1392) " @ umulls r1, pc, r2, r3")
+
+	TEST_RRRR(  "umlal	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR(  "umlalle	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+	TEST_RRR(   "umlal	r",14,VAL3,", r",7, VAL4,", r",5, VAL1,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe0af1392) " @ umlal pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe0a1f392) " @ umlal r1, pc, r2, r3")
+	TEST_RRRR(  "umlals	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR(  "umlalles	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+	TEST_RRR(   "umlals	r",14,VAL3,", r",7, VAL4,", r",5, VAL1,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe0bf1392) " @ umlals pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe0b1f392) " @ umlals r1, pc, r2, r3")
+
+	TEST_RR(  "smull	r0, r1, r",2, VAL1,", r",3, VAL2,"")
+	TEST_RR(  "smullls	r7, r8, r",9, VAL2,", r",10, VAL1,"")
+	TEST_R(   "smull	lr, r12, r",11,VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe0c1f392) " @ smull pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe0cf1392) " @ smull r1, pc, r2, r3")
+	TEST_RR(  "smulls	r0, r1, r",2, VAL1,", r",3, VAL2,"")
+	TEST_RR(  "smulllss	r7, r8, r",9, VAL2,", r",10, VAL1,"")
+	TEST_R(   "smulls	lr, r12, r",11,VAL3,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe0d1f392) " @ smulls pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe0df1392) " @ smulls r1, pc, r2, r3")
+
+	TEST_RRRR(  "smlal	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR(  "smlalle	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+	TEST_RRR(   "smlal	r",14,VAL3,", r",7, VAL4,", r",5, VAL1,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe0ef1392) " @ smlal pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe0e1f392) " @ smlal r1, pc, r2, r3")
+	TEST_RRRR(  "smlals	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR(  "smlalles	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+	TEST_RRR(   "smlals	r",14,VAL3,", r",7, VAL4,", r",5, VAL1,", r13")
+	TEST_UNSUPPORTED(__inst_arm(0xe0ff1392) " @ smlals pc, r1, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe0f0f392) " @ smlals r0, pc, r2, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe0f0139f) " @ smlals r0, r1, pc, r3")
+	TEST_UNSUPPORTED(__inst_arm(0xe0f01f92) " @ smlals r0, r1, r2, pc")
+
+	TEST_GROUP("Synchronization primitives")
+
+#if __LINUX_ARM_ARCH__ < 6
+	TEST_RP("swp	lr, r",7,VAL2,", [r",8,0,"]")
+	TEST_R( "swpvs	r0, r",1,VAL1,", [sp]")
+	TEST_RP("swp	sp, r",14,VAL2,", [r",12,13*4,"]")
+#else
+	TEST_UNSUPPORTED(__inst_arm(0xe108e097) " @ swp	lr, r7, [r8]")
+	TEST_UNSUPPORTED(__inst_arm(0x610d0091) " @ swpvs	r0, r1, [sp]")
+	TEST_UNSUPPORTED(__inst_arm(0xe10cd09e) " @ swp	sp, r14 [r12]")
+#endif
+	TEST_UNSUPPORTED(__inst_arm(0xe102f091) " @ swp pc, r1, [r2]")
+	TEST_UNSUPPORTED(__inst_arm(0xe102009f) " @ swp r0, pc, [r2]")
+	TEST_UNSUPPORTED(__inst_arm(0xe10f0091) " @ swp r0, r1, [pc]")
+#if __LINUX_ARM_ARCH__ < 6
+	TEST_RP("swpb	lr, r",7,VAL2,", [r",8,0,"]")
+	TEST_R( "swpvsb	r0, r",1,VAL1,", [sp]")
+#else
+	TEST_UNSUPPORTED(__inst_arm(0xe148e097) " @ swpb	lr, r7, [r8]")
+	TEST_UNSUPPORTED(__inst_arm(0x614d0091) " @ swpvsb	r0, r1, [sp]")
+#endif
+	TEST_UNSUPPORTED(__inst_arm(0xe142f091) " @ swpb pc, r1, [r2]")
+
+	TEST_UNSUPPORTED(__inst_arm(0xe1100090)) /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe1200090)) /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe1300090)) /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe1500090)) /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe1600090)) /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe1700090)) /* Unallocated space */
+#if __LINUX_ARM_ARCH__ >= 6
+	TEST_UNSUPPORTED("ldrex	r2, [sp]")
+#endif
+#if (__LINUX_ARM_ARCH__ >= 7) || defined(CONFIG_CPU_32v6K)
+	TEST_UNSUPPORTED("strexd	r0, r2, r3, [sp]")
+	TEST_UNSUPPORTED("ldrexd	r2, r3, [sp]")
+	TEST_UNSUPPORTED("strexb	r0, r2, [sp]")
+	TEST_UNSUPPORTED("ldrexb	r2, [sp]")
+	TEST_UNSUPPORTED("strexh	r0, r2, [sp]")
+	TEST_UNSUPPORTED("ldrexh	r2, [sp]")
+#endif
+	TEST_GROUP("Extra load/store instructions")
+
+	TEST_RPR(  "strh	r",0, VAL1,", [r",1, 48,", -r",2, 24,"]")
+	TEST_RPR(  "streqh	r",14,VAL2,", [r",11,0, ", r",12, 48,"]")
+	TEST_UNSUPPORTED(  "streqh	r14, [r13, r12]")
+	TEST_UNSUPPORTED(  "streqh	r14, [r12, r13]")
+	TEST_RPR(  "strh	r",1, VAL1,", [r",2, 24,", r",3,  48,"]!")
+	TEST_RPR(  "strneh	r",12,VAL2,", [r",11,48,", -r",10,24,"]!")
+	TEST_RPR(  "strh	r",2, VAL1,", [r",3, 24,"], r",4, 48,"")
+	TEST_RPR(  "strh	r",10,VAL2,", [r",9, 48,"], -r",11,24,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe1afc0ba) "	@ strh r12, [pc, r10]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe089f0bb) "	@ strh pc, [r9], r11")
+	TEST_UNSUPPORTED(__inst_arm(0xe089a0bf) "	@ strh r10, [r9], pc")
+
+	TEST_PR(   "ldrh	r0, [r",0,  48,", -r",2, 24,"]")
+	TEST_PR(   "ldrcsh	r14, [r",13,0, ", r",12, 48,"]")
+	TEST_PR(   "ldrh	r1, [r",2,  24,", r",3,  48,"]!")
+	TEST_PR(   "ldrcch	r12, [r",11,48,", -r",10,24,"]!")
+	TEST_PR(   "ldrh	r2, [r",3,  24,"], r",4, 48,"")
+	TEST_PR(   "ldrh	r10, [r",9, 48,"], -r",11,24,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe1bfc0ba) "	@ ldrh r12, [pc, r10]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe099f0bb) "	@ ldrh pc, [r9], r11")
+	TEST_UNSUPPORTED(__inst_arm(0xe099a0bf) "	@ ldrh r10, [r9], pc")
+
+	TEST_RP(   "strh	r",0, VAL1,", [r",1, 24,", #-2]")
+	TEST_RP(   "strmih	r",14,VAL2,", [r",13,0, ", #2]")
+	TEST_RP(   "strh	r",1, VAL1,", [r",2, 24,", #4]!")
+	TEST_RP(   "strplh	r",12,VAL2,", [r",11,24,", #-4]!")
+	TEST_RP(   "strh	r",2, VAL1,", [r",3, 24,"], #48")
+	TEST_RP(   "strh	r",10,VAL2,", [r",9, 64,"], #-48")
+	TEST_RP(   "strh	r",3, VAL1,", [r",13,TEST_MEMORY_SIZE,", #-"__stringify(MAX_STACK_SIZE)"]!")
+	TEST_UNSUPPORTED("strh r3, [r13, #-"__stringify(MAX_STACK_SIZE)"-8]!")
+	TEST_RP(   "strh	r",4, VAL1,", [r",14,TEST_MEMORY_SIZE,", #-"__stringify(MAX_STACK_SIZE)"-8]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe1efc3b0) "	@ strh r12, [pc, #48]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe0c9f3b0) "	@ strh pc, [r9], #48")
+
+	TEST_P(	   "ldrh	r0, [r",0,  24,", #-2]")
+	TEST_P(	   "ldrvsh	r14, [r",13,0, ", #2]")
+	TEST_P(	   "ldrh	r1, [r",2,  24,", #4]!")
+	TEST_P(	   "ldrvch	r12, [r",11,24,", #-4]!")
+	TEST_P(	   "ldrh	r2, [r",3,  24,"], #48")
+	TEST_P(	   "ldrh	r10, [r",9, 64,"], #-48")
+	TEST(      "ldrh	r0, [pc, #0]")
+	TEST_UNSUPPORTED(__inst_arm(0xe1ffc3b0) "	@ ldrh r12, [pc, #48]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe0d9f3b0) "	@ ldrh pc, [r9], #48")
+
+	TEST_PR(   "ldrsb	r0, [r",0,  48,", -r",2, 24,"]")
+	TEST_PR(   "ldrhisb	r14, [r",13,0,", r",12,  48,"]")
+	TEST_PR(   "ldrsb	r1, [r",2,  24,", r",3,  48,"]!")
+	TEST_PR(   "ldrlssb	r12, [r",11,48,", -r",10,24,"]!")
+	TEST_PR(   "ldrsb	r2, [r",3,  24,"], r",4, 48,"")
+	TEST_PR(   "ldrsb	r10, [r",9, 48,"], -r",11,24,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe1bfc0da) "	@ ldrsb r12, [pc, r10]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe099f0db) "	@ ldrsb pc, [r9], r11")
+
+	TEST_P(	   "ldrsb	r0, [r",0,  24,", #-1]")
+	TEST_P(	   "ldrgesb	r14, [r",13,0, ", #1]")
+	TEST_P(	   "ldrsb	r1, [r",2,  24,", #4]!")
+	TEST_P(	   "ldrltsb	r12, [r",11,24,", #-4]!")
+	TEST_P(	   "ldrsb	r2, [r",3,  24,"], #48")
+	TEST_P(	   "ldrsb	r10, [r",9, 64,"], #-48")
+	TEST(      "ldrsb	r0, [pc, #0]")
+	TEST_UNSUPPORTED(__inst_arm(0xe1ffc3d0) "	@ ldrsb r12, [pc, #48]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe0d9f3d0) "	@ ldrsb pc, [r9], #48")
+
+	TEST_PR(   "ldrsh	r0, [r",0,  48,", -r",2, 24,"]")
+	TEST_PR(   "ldrgtsh	r14, [r",13,0, ", r",12, 48,"]")
+	TEST_PR(   "ldrsh	r1, [r",2,  24,", r",3,  48,"]!")
+	TEST_PR(   "ldrlesh	r12, [r",11,48,", -r",10,24,"]!")
+	TEST_PR(   "ldrsh	r2, [r",3,  24,"], r",4, 48,"")
+	TEST_PR(   "ldrsh	r10, [r",9, 48,"], -r",11,24,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe1bfc0fa) "	@ ldrsh r12, [pc, r10]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe099f0fb) "	@ ldrsh pc, [r9], r11")
+
+	TEST_P(	   "ldrsh	r0, [r",0,  24,", #-1]")
+	TEST_P(	   "ldreqsh	r14, [r",13,0 ,", #1]")
+	TEST_P(	   "ldrsh	r1, [r",2,  24,", #4]!")
+	TEST_P(	   "ldrnesh	r12, [r",11,24,", #-4]!")
+	TEST_P(	   "ldrsh	r2, [r",3,  24,"], #48")
+	TEST_P(	   "ldrsh	r10, [r",9, 64,"], #-48")
+	TEST(      "ldrsh	r0, [pc, #0]")
+	TEST_UNSUPPORTED(__inst_arm(0xe1ffc3f0) "	@ ldrsh r12, [pc, #48]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe0d9f3f0) "	@ ldrsh pc, [r9], #48")
+
+#if __LINUX_ARM_ARCH__ >= 7
+	TEST_UNSUPPORTED("strht	r1, [r2], r3")
+	TEST_UNSUPPORTED("ldrht	r1, [r2], r3")
+	TEST_UNSUPPORTED("strht	r1, [r2], #48")
+	TEST_UNSUPPORTED("ldrht	r1, [r2], #48")
+	TEST_UNSUPPORTED("ldrsbt	r1, [r2], r3")
+	TEST_UNSUPPORTED("ldrsbt	r1, [r2], #48")
+	TEST_UNSUPPORTED("ldrsht	r1, [r2], r3")
+	TEST_UNSUPPORTED("ldrsht	r1, [r2], #48")
+#endif
+
+#if __LINUX_ARM_ARCH__ >= 5
+	TEST_RPR(  "strd	r",0, VAL1,", [r",1, 48,", -r",2,24,"]")
+	TEST_RPR(  "strccd	r",8, VAL2,", [r",11,0, ", r",12,48,"]")
+	TEST_UNSUPPORTED(  "strccd r8, [r13, r12]")
+	TEST_UNSUPPORTED(  "strccd r8, [r12, r13]")
+	TEST_RPR(  "strd	r",4, VAL1,", [r",2, 24,", r",3, 48,"]!")
+	TEST_RPR(  "strcsd	r",12,VAL2,", [r",11,48,", -r",10,24,"]!")
+	TEST_RPR(  "strd	r",2, VAL1,", [r",5, 24,"], r",4,48,"")
+	TEST_RPR(  "strd	r",10,VAL2,", [r",9, 48,"], -r",7,24,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe1afc0fa) "	@ strd r12, [pc, r10]!")
+
+	TEST_PR(   "ldrd	r0, [r",0, 48,", -r",2,24,"]")
+	TEST_PR(   "ldrmid	r8, [r",13,0, ", r",12,48,"]")
+	TEST_PR(   "ldrd	r4, [r",2, 24,", r",3, 48,"]!")
+	TEST_PR(   "ldrpld	r6, [r",11,48,", -r",10,24,"]!")
+	TEST_PR(   "ldrd	r2, [r",5, 24,"], r",4,48,"")
+	TEST_PR(   "ldrd	r10, [r",9,48,"], -r",7,24,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe1afc0da) "	@ ldrd r12, [pc, r10]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe089f0db) "	@ ldrd pc, [r9], r11")
+	TEST_UNSUPPORTED(__inst_arm(0xe089e0db) "	@ ldrd lr, [r9], r11")
+	TEST_UNSUPPORTED(__inst_arm(0xe089c0df) "	@ ldrd r12, [r9], pc")
+
+	TEST_RP(   "strd	r",0, VAL1,", [r",1, 24,", #-8]")
+	TEST_RP(   "strvsd	r",8, VAL2,", [r",13,0, ", #8]")
+	TEST_RP(   "strd	r",4, VAL1,", [r",2, 24,", #16]!")
+	TEST_RP(   "strvcd	r",12,VAL2,", [r",11,24,", #-16]!")
+	TEST_RP(   "strd	r",2, VAL1,", [r",4, 24,"], #48")
+	TEST_RP(   "strd	r",10,VAL2,", [r",9, 64,"], #-48")
+	TEST_RP(   "strd	r",6, VAL1,", [r",13,TEST_MEMORY_SIZE,", #-"__stringify(MAX_STACK_SIZE)"]!")
+	TEST_UNSUPPORTED("strd r6, [r13, #-"__stringify(MAX_STACK_SIZE)"-8]!")
+	TEST_RP(   "strd	r",4, VAL1,", [r",12,TEST_MEMORY_SIZE,", #-"__stringify(MAX_STACK_SIZE)"-8]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe1efc3f0) "	@ strd r12, [pc, #48]!")
+
+	TEST_P(	   "ldrd	r0, [r",0, 24,", #-8]")
+	TEST_P(	   "ldrhid	r8, [r",13,0, ", #8]")
+	TEST_P(	   "ldrd	r4, [r",2, 24,", #16]!")
+	TEST_P(	   "ldrlsd	r6, [r",11,24,", #-16]!")
+	TEST_P(	   "ldrd	r2, [r",5, 24,"], #48")
+	TEST_P(	   "ldrd	r10, [r",9,6,"], #-48")
+	TEST_UNSUPPORTED(__inst_arm(0xe1efc3d0) "	@ ldrd r12, [pc, #48]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe0c9f3d0) "	@ ldrd pc, [r9], #48")
+	TEST_UNSUPPORTED(__inst_arm(0xe0c9e3d0) "	@ ldrd lr, [r9], #48")
+#endif
+
+	TEST_GROUP("Miscellaneous")
+
+#if __LINUX_ARM_ARCH__ >= 7
+	TEST("movw	r0, #0")
+	TEST("movw	r0, #0xffff")
+	TEST("movw	lr, #0xffff")
+	TEST_UNSUPPORTED(__inst_arm(0xe300f000) "	@ movw pc, #0")
+	TEST_R("movt	r",0, VAL1,", #0")
+	TEST_R("movt	r",0, VAL2,", #0xffff")
+	TEST_R("movt	r",14,VAL1,", #0xffff")
+	TEST_UNSUPPORTED(__inst_arm(0xe340f000) "	@ movt pc, #0")
+#endif
+
+	TEST_UNSUPPORTED("msr	cpsr, 0x13")
+	TEST_UNSUPPORTED("msr	cpsr_f, 0xf0000000")
+	TEST_UNSUPPORTED("msr	spsr, 0x13")
+
+#if __LINUX_ARM_ARCH__ >= 7
+	TEST_SUPPORTED("yield")
+	TEST("sev")
+	TEST("nop")
+	TEST("wfi")
+	TEST_SUPPORTED("wfe")
+	TEST_UNSUPPORTED("dbg #0")
+#endif
+
+	TEST_GROUP("Load/store word and unsigned byte")
+
+#define LOAD_STORE(byte)							\
+	TEST_RP( "str"byte"	r",0, VAL1,", [r",1, 24,", #-2]")		\
+	TEST_RP( "str"byte"	r",14,VAL2,", [r",13,0, ", #2]")		\
+	TEST_RP( "str"byte"	r",1, VAL1,", [r",2, 24,", #4]!")		\
+	TEST_RP( "str"byte"	r",12,VAL2,", [r",11,24,", #-4]!")		\
+	TEST_RP( "str"byte"	r",2, VAL1,", [r",3, 24,"], #48")		\
+	TEST_RP( "str"byte"	r",10,VAL2,", [r",9, 64,"], #-48")		\
+	TEST_RP( "str"byte"	r",3, VAL1,", [r",13,TEST_MEMORY_SIZE,", #-"__stringify(MAX_STACK_SIZE)"]!") \
+	TEST_UNSUPPORTED("str"byte" r3, [r13, #-"__stringify(MAX_STACK_SIZE)"-8]!")				\
+	TEST_RP( "str"byte"	r",4, VAL1,", [r",10,TEST_MEMORY_SIZE,", #-"__stringify(MAX_STACK_SIZE)"-8]!") \
+	TEST_RPR("str"byte"	r",0, VAL1,", [r",1, 48,", -r",2, 24,"]")	\
+	TEST_RPR("str"byte"	r",14,VAL2,", [r",11,0, ", r",12, 48,"]")	\
+	TEST_UNSUPPORTED("str"byte" r14, [r13, r12]")				\
+	TEST_UNSUPPORTED("str"byte" r14, [r12, r13]")				\
+	TEST_RPR("str"byte"	r",1, VAL1,", [r",2, 24,", r",3,  48,"]!")	\
+	TEST_RPR("str"byte"	r",12,VAL2,", [r",11,48,", -r",10,24,"]!")	\
+	TEST_RPR("str"byte"	r",2, VAL1,", [r",3, 24,"], r",4, 48,"")	\
+	TEST_RPR("str"byte"	r",10,VAL2,", [r",9, 48,"], -r",11,24,"")	\
+	TEST_RPR("str"byte"	r",0, VAL1,", [r",1, 24,", r",2,  32,", asl #1]")\
+	TEST_RPR("str"byte"	r",14,VAL2,", [r",11,0, ", r",12, 32,", lsr #2]")\
+	TEST_UNSUPPORTED("str"byte"	r14, [r13, r12, lsr #2]")		\
+	TEST_RPR("str"byte"	r",1, VAL1,", [r",2, 24,", r",3,  32,", asr #3]!")\
+	TEST_RPR("str"byte"	r",12,VAL2,", [r",11,24,", r",10, 4,", ror #31]!")\
+	TEST_P(  "ldr"byte"	r0, [r",0,  24,", #-2]")			\
+	TEST_P(  "ldr"byte"	r14, [r",13,0, ", #2]")				\
+	TEST_P(  "ldr"byte"	r1, [r",2,  24,", #4]!")			\
+	TEST_P(  "ldr"byte"	r12, [r",11,24,", #-4]!")			\
+	TEST_P(  "ldr"byte"	r2, [r",3,  24,"], #48")			\
+	TEST_P(  "ldr"byte"	r10, [r",9, 64,"], #-48")			\
+	TEST_PR( "ldr"byte"	r0, [r",0,  48,", -r",2, 24,"]")		\
+	TEST_PR( "ldr"byte"	r14, [r",13,0, ", r",12, 48,"]")		\
+	TEST_PR( "ldr"byte"	r1, [r",2,  24,", r",3, 48,"]!")		\
+	TEST_PR( "ldr"byte"	r12, [r",11,48,", -r",10,24,"]!")		\
+	TEST_PR( "ldr"byte"	r2, [r",3,  24,"], r",4, 48,"")			\
+	TEST_PR( "ldr"byte"	r10, [r",9, 48,"], -r",11,24,"")		\
+	TEST_PR( "ldr"byte"	r0, [r",0,  24,", r",2,  32,", asl #1]")	\
+	TEST_PR( "ldr"byte"	r14, [r",13,0, ", r",12, 32,", lsr #2]")	\
+	TEST_PR( "ldr"byte"	r1, [r",2,  24,", r",3,  32,", asr #3]!")	\
+	TEST_PR( "ldr"byte"	r12, [r",11,24,", r",10, 4,", ror #31]!")	\
+	TEST(    "ldr"byte"	r0, [pc, #0]")					\
+	TEST_R(  "ldr"byte"	r12, [pc, r",14,0,"]")
+
+	LOAD_STORE("")
+	TEST_P(   "str	pc, [r",0,0,", #15*4]")
+	TEST_UNSUPPORTED(   "str	pc, [sp, r2]")
+	TEST_BF(  "ldr	pc, [sp, #15*4]")
+	TEST_BF_R("ldr	pc, [sp, r",2,15*4,"]")
+
+	TEST_P(   "str	sp, [r",0,0,", #13*4]")
+	TEST_UNSUPPORTED(   "str	sp, [sp, r2]")
+	TEST_BF(  "ldr	sp, [sp, #13*4]")
+	TEST_BF_R("ldr	sp, [sp, r",2,13*4,"]")
+
+#ifdef CONFIG_THUMB2_KERNEL
+	TEST_ARM_TO_THUMB_INTERWORK_P("ldr	pc, [r",0,0,", #15*4]")
+#endif
+	TEST_UNSUPPORTED(__inst_arm(0xe5af6008) "	@ str r6, [pc, #8]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe7af6008) "	@ str r6, [pc, r8]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe5bf6008) "	@ ldr r6, [pc, #8]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe7bf6008) "	@ ldr r6, [pc, r8]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe788600f) "	@ str r6, [r8, pc]")
+	TEST_UNSUPPORTED(__inst_arm(0xe798600f) "	@ ldr r6, [r8, pc]")
+
+	LOAD_STORE("b")
+	TEST_UNSUPPORTED(__inst_arm(0xe5f7f008) "	@ ldrb pc, [r7, #8]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe7f7f008) "	@ ldrb pc, [r7, r8]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe5ef6008) "	@ strb r6, [pc, #8]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe7ef6008) "	@ strb r6, [pc, r3]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe5ff6008) "	@ ldrb r6, [pc, #8]!")
+	TEST_UNSUPPORTED(__inst_arm(0xe7ff6008) "	@ ldrb r6, [pc, r3]!")
+
+	TEST_UNSUPPORTED("ldrt	r0, [r1], #4")
+	TEST_UNSUPPORTED("ldrt	r1, [r2], r3")
+	TEST_UNSUPPORTED("strt	r2, [r3], #4")
+	TEST_UNSUPPORTED("strt	r3, [r4], r5")
+	TEST_UNSUPPORTED("ldrbt	r4, [r5], #4")
+	TEST_UNSUPPORTED("ldrbt	r5, [r6], r7")
+	TEST_UNSUPPORTED("strbt	r6, [r7], #4")
+	TEST_UNSUPPORTED("strbt	r7, [r8], r9")
+
+#if __LINUX_ARM_ARCH__ >= 7
+	TEST_GROUP("Parallel addition and subtraction, signed")
+
+	TEST_UNSUPPORTED(__inst_arm(0xe6000010) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe60fffff) "") /* Unallocated space */
+
+	TEST_RR(    "sadd16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "sadd16	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe61cff1a) "	@ sadd16	pc, r12, r10")
+	TEST_RR(    "sasx	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "sasx	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe61cff3a) "	@ sasx	pc, r12, r10")
+	TEST_RR(    "ssax	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "ssax	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe61cff5a) "	@ ssax	pc, r12, r10")
+	TEST_RR(    "ssub16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "ssub16	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe61cff7a) "	@ ssub16	pc, r12, r10")
+	TEST_RR(    "sadd8	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "sadd8	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe61cff9a) "	@ sadd8	pc, r12, r10")
+	TEST_UNSUPPORTED(__inst_arm(0xe61000b0) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe61fffbf) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe61000d0) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe61fffdf) "") /* Unallocated space */
+	TEST_RR(    "ssub8	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "ssub8	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe61cfffa) "	@ ssub8	pc, r12, r10")
+
+	TEST_RR(    "qadd16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "qadd16	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe62cff1a) "	@ qadd16	pc, r12, r10")
+	TEST_RR(    "qasx	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "qasx	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe62cff3a) "	@ qasx	pc, r12, r10")
+	TEST_RR(    "qsax	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "qsax	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe62cff5a) "	@ qsax	pc, r12, r10")
+	TEST_RR(    "qsub16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "qsub16	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe62cff7a) "	@ qsub16	pc, r12, r10")
+	TEST_RR(    "qadd8	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "qadd8	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe62cff9a) "	@ qadd8	pc, r12, r10")
+	TEST_UNSUPPORTED(__inst_arm(0xe62000b0) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe62fffbf) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe62000d0) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe62fffdf) "") /* Unallocated space */
+	TEST_RR(    "qsub8	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "qsub8	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe62cfffa) "	@ qsub8	pc, r12, r10")
+
+	TEST_RR(    "shadd16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "shadd16	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe63cff1a) "	@ shadd16	pc, r12, r10")
+	TEST_RR(    "shasx	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "shasx	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe63cff3a) "	@ shasx	pc, r12, r10")
+	TEST_RR(    "shsax	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "shsax	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe63cff5a) "	@ shsax	pc, r12, r10")
+	TEST_RR(    "shsub16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "shsub16	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe63cff7a) "	@ shsub16	pc, r12, r10")
+	TEST_RR(    "shadd8	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "shadd8	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe63cff9a) "	@ shadd8	pc, r12, r10")
+	TEST_UNSUPPORTED(__inst_arm(0xe63000b0) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe63fffbf) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe63000d0) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe63fffdf) "") /* Unallocated space */
+	TEST_RR(    "shsub8	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "shsub8	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe63cfffa) "	@ shsub8	pc, r12, r10")
+
+	TEST_GROUP("Parallel addition and subtraction, unsigned")
+
+	TEST_UNSUPPORTED(__inst_arm(0xe6400010) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe64fffff) "") /* Unallocated space */
+
+	TEST_RR(    "uadd16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uadd16	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe65cff1a) "	@ uadd16	pc, r12, r10")
+	TEST_RR(    "uasx	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uasx	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe65cff3a) "	@ uasx	pc, r12, r10")
+	TEST_RR(    "usax	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "usax	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe65cff5a) "	@ usax	pc, r12, r10")
+	TEST_RR(    "usub16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "usub16	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe65cff7a) "	@ usub16	pc, r12, r10")
+	TEST_RR(    "uadd8	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uadd8	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe65cff9a) "	@ uadd8	pc, r12, r10")
+	TEST_UNSUPPORTED(__inst_arm(0xe65000b0) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe65fffbf) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe65000d0) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe65fffdf) "") /* Unallocated space */
+	TEST_RR(    "usub8	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "usub8	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe65cfffa) "	@ usub8	pc, r12, r10")
+
+	TEST_RR(    "uqadd16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uqadd16	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe66cff1a) "	@ uqadd16	pc, r12, r10")
+	TEST_RR(    "uqasx	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uqasx	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe66cff3a) "	@ uqasx	pc, r12, r10")
+	TEST_RR(    "uqsax	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uqsax	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe66cff5a) "	@ uqsax	pc, r12, r10")
+	TEST_RR(    "uqsub16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uqsub16	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe66cff7a) "	@ uqsub16	pc, r12, r10")
+	TEST_RR(    "uqadd8	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uqadd8	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe66cff9a) "	@ uqadd8	pc, r12, r10")
+	TEST_UNSUPPORTED(__inst_arm(0xe66000b0) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe66fffbf) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe66000d0) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe66fffdf) "") /* Unallocated space */
+	TEST_RR(    "uqsub8	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uqsub8	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe66cfffa) "	@ uqsub8	pc, r12, r10")
+
+	TEST_RR(    "uhadd16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uhadd16	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe67cff1a) "	@ uhadd16	pc, r12, r10")
+	TEST_RR(    "uhasx	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uhasx	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe67cff3a) "	@ uhasx	pc, r12, r10")
+	TEST_RR(    "uhsax	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uhsax	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe67cff5a) "	@ uhsax	pc, r12, r10")
+	TEST_RR(    "uhsub16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uhsub16	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe67cff7a) "	@ uhsub16	pc, r12, r10")
+	TEST_RR(    "uhadd8	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uhadd8	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe67cff9a) "	@ uhadd8	pc, r12, r10")
+	TEST_UNSUPPORTED(__inst_arm(0xe67000b0) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe67fffbf) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe67000d0) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe67fffdf) "") /* Unallocated space */
+	TEST_RR(    "uhsub8	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uhsub8	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe67cfffa) "	@ uhsub8	pc, r12, r10")
+	TEST_UNSUPPORTED(__inst_arm(0xe67feffa) "	@ uhsub8	r14, pc, r10")
+	TEST_UNSUPPORTED(__inst_arm(0xe67cefff) "	@ uhsub8	r14, r12, pc")
+#endif /* __LINUX_ARM_ARCH__ >= 7 */
+
+#if __LINUX_ARM_ARCH__ >= 6
+	TEST_GROUP("Packing, unpacking, saturation, and reversal")
+
+	TEST_RR(    "pkhbt	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "pkhbt	r14,r",12, HH1,", r",10,HH2,", lsl #2")
+	TEST_UNSUPPORTED(__inst_arm(0xe68cf11a) "	@ pkhbt	pc, r12, r10, lsl #2")
+	TEST_RR(    "pkhtb	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "pkhtb	r14,r",12, HH1,", r",10,HH2,", asr #2")
+	TEST_UNSUPPORTED(__inst_arm(0xe68cf15a) "	@ pkhtb	pc, r12, r10, asr #2")
+	TEST_UNSUPPORTED(__inst_arm(0xe68fe15a) "	@ pkhtb	r14, pc, r10, asr #2")
+	TEST_UNSUPPORTED(__inst_arm(0xe68ce15f) "	@ pkhtb	r14, r12, pc, asr #2")
+	TEST_UNSUPPORTED(__inst_arm(0xe6900010) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe69fffdf) "") /* Unallocated space */
+
+	TEST_R(     "ssat	r0, #24, r",0,   VAL1,"")
+	TEST_R(     "ssat	r14, #24, r",12, VAL2,"")
+	TEST_R(     "ssat	r0, #24, r",0,   VAL1,", lsl #8")
+	TEST_R(     "ssat	r14, #24, r",12, VAL2,", asr #8")
+	TEST_UNSUPPORTED(__inst_arm(0xe6b7f01c) "	@ ssat	pc, #24, r12")
+
+	TEST_R(     "usat	r0, #24, r",0,   VAL1,"")
+	TEST_R(     "usat	r14, #24, r",12, VAL2,"")
+	TEST_R(     "usat	r0, #24, r",0,   VAL1,", lsl #8")
+	TEST_R(     "usat	r14, #24, r",12, VAL2,", asr #8")
+	TEST_UNSUPPORTED(__inst_arm(0xe6f7f01c) "	@ usat	pc, #24, r12")
+
+	TEST_RR(    "sxtab16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "sxtab16	r14,r",12, HH2,", r",10,HH1,", ror #8")
+	TEST_R(     "sxtb16	r8, r",7,  HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe68cf47a) "	@ sxtab16	pc,r12, r10, ror #8")
+
+	TEST_RR(    "sel	r0, r",0,  VAL1,", r",1, VAL2,"")
+	TEST_RR(    "sel	r14, r",12,VAL1,", r",10, VAL2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe68cffba) "	@ sel	pc, r12, r10")
+	TEST_UNSUPPORTED(__inst_arm(0xe68fefba) "	@ sel	r14, pc, r10")
+	TEST_UNSUPPORTED(__inst_arm(0xe68cefbf) "	@ sel	r14, r12, pc")
+
+	TEST_R(     "ssat16	r0, #12, r",0,   HH1,"")
+	TEST_R(     "ssat16	r14, #12, r",12, HH2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe6abff3c) "	@ ssat16	pc, #12, r12")
+
+	TEST_RR(    "sxtab	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "sxtab	r14,r",12, HH2,", r",10,HH1,", ror #8")
+	TEST_R(     "sxtb	r8, r",7,  HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe6acf47a) "	@ sxtab	pc,r12, r10, ror #8")
+
+	TEST_R(     "rev	r0, r",0,   VAL1,"")
+	TEST_R(     "rev	r14, r",12, VAL2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe6bfff3c) "	@ rev	pc, r12")
+
+	TEST_RR(    "sxtah	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "sxtah	r14,r",12, HH2,", r",10,HH1,", ror #8")
+	TEST_R(     "sxth	r8, r",7,  HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe6bcf47a) "	@ sxtah	pc,r12, r10, ror #8")
+
+	TEST_R(     "rev16	r0, r",0,   VAL1,"")
+	TEST_R(     "rev16	r14, r",12, VAL2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe6bfffbc) "	@ rev16	pc, r12")
+
+	TEST_RR(    "uxtab16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uxtab16	r14,r",12, HH2,", r",10,HH1,", ror #8")
+	TEST_R(     "uxtb16	r8, r",7,  HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe6ccf47a) "	@ uxtab16	pc,r12, r10, ror #8")
+
+	TEST_R(     "usat16	r0, #12, r",0,   HH1,"")
+	TEST_R(     "usat16	r14, #12, r",12, HH2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe6ecff3c) "	@ usat16	pc, #12, r12")
+	TEST_UNSUPPORTED(__inst_arm(0xe6ecef3f) "	@ usat16	r14, #12, pc")
+
+	TEST_RR(    "uxtab	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uxtab	r14,r",12, HH2,", r",10,HH1,", ror #8")
+	TEST_R(     "uxtb	r8, r",7,  HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe6ecf47a) "	@ uxtab	pc,r12, r10, ror #8")
+
+#if __LINUX_ARM_ARCH__ >= 7
+	TEST_R(     "rbit	r0, r",0,   VAL1,"")
+	TEST_R(     "rbit	r14, r",12, VAL2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe6ffff3c) "	@ rbit	pc, r12")
+#endif
+
+	TEST_RR(    "uxtah	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uxtah	r14,r",12, HH2,", r",10,HH1,", ror #8")
+	TEST_R(     "uxth	r8, r",7,  HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe6fff077) "	@ uxth	pc, r7")
+	TEST_UNSUPPORTED(__inst_arm(0xe6ff807f) "	@ uxth	r8, pc")
+	TEST_UNSUPPORTED(__inst_arm(0xe6fcf47a) "	@ uxtah	pc, r12, r10, ror #8")
+	TEST_UNSUPPORTED(__inst_arm(0xe6fce47f) "	@ uxtah	r14, r12, pc, ror #8")
+
+	TEST_R(     "revsh	r0, r",0,   VAL1,"")
+	TEST_R(     "revsh	r14, r",12, VAL2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe6ffff3c) "	@ revsh	pc, r12")
+	TEST_UNSUPPORTED(__inst_arm(0xe6ffef3f) "	@ revsh	r14, pc")
+
+	TEST_UNSUPPORTED(__inst_arm(0xe6900070) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe69fff7f) "") /* Unallocated space */
+
+	TEST_UNSUPPORTED(__inst_arm(0xe6d00070) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_arm(0xe6dfff7f) "") /* Unallocated space */
+#endif /* __LINUX_ARM_ARCH__ >= 6 */
+
+#if __LINUX_ARM_ARCH__ >= 6
+	TEST_GROUP("Signed multiplies")
+
+	TEST_RRR(   "smlad	r0, r",0,  HH1,", r",1, HH2,", r",2, VAL1,"")
+	TEST_RRR(   "smlad	r14, r",12,HH2,", r",10,HH1,", r",8, VAL2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe70f8a1c) "	@ smlad	pc, r12, r10, r8")
+	TEST_RRR(   "smladx	r0, r",0,  HH1,", r",1, HH2,", r",2, VAL1,"")
+	TEST_RRR(   "smladx	r14, r",12,HH2,", r",10,HH1,", r",8, VAL2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe70f8a3c) "	@ smladx	pc, r12, r10, r8")
+
+	TEST_RR(   "smuad	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(   "smuad	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe70ffa1c) "	@ smuad	pc, r12, r10")
+	TEST_RR(   "smuadx	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(   "smuadx	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe70ffa3c) "	@ smuadx	pc, r12, r10")
+
+	TEST_RRR(   "smlsd	r0, r",0,  HH1,", r",1, HH2,", r",2, VAL1,"")
+	TEST_RRR(   "smlsd	r14, r",12,HH2,", r",10,HH1,", r",8, VAL2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe70f8a5c) "	@ smlsd	pc, r12, r10, r8")
+	TEST_RRR(   "smlsdx	r0, r",0,  HH1,", r",1, HH2,", r",2, VAL1,"")
+	TEST_RRR(   "smlsdx	r14, r",12,HH2,", r",10,HH1,", r",8, VAL2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe70f8a7c) "	@ smlsdx	pc, r12, r10, r8")
+
+	TEST_RR(   "smusd	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(   "smusd	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe70ffa5c) "	@ smusd	pc, r12, r10")
+	TEST_RR(   "smusdx	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(   "smusdx	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe70ffa7c) "	@ smusdx	pc, r12, r10")
+
+	TEST_RRRR( "smlald	r",0, VAL1,", r",1, VAL2, ", r",0, HH1,", r",1, HH2)
+	TEST_RRRR( "smlald	r",11,VAL2,", r",10,VAL1, ", r",9, HH2,", r",8, HH1)
+	TEST_UNSUPPORTED(__inst_arm(0xe74af819) "	@ smlald	pc, r10, r9, r8")
+	TEST_UNSUPPORTED(__inst_arm(0xe74fb819) "	@ smlald	r11, pc, r9, r8")
+	TEST_UNSUPPORTED(__inst_arm(0xe74ab81f) "	@ smlald	r11, r10, pc, r8")
+	TEST_UNSUPPORTED(__inst_arm(0xe74abf19) "	@ smlald	r11, r10, r9, pc")
+
+	TEST_RRRR( "smlaldx	r",0, VAL1,", r",1, VAL2, ", r",0, HH1,", r",1, HH2)
+	TEST_RRRR( "smlaldx	r",11,VAL2,", r",10,VAL1, ", r",9, HH2,", r",8, HH1)
+	TEST_UNSUPPORTED(__inst_arm(0xe74af839) "	@ smlaldx	pc, r10, r9, r8")
+	TEST_UNSUPPORTED(__inst_arm(0xe74fb839) "	@ smlaldx	r11, pc, r9, r8")
+
+	TEST_RRR(  "smmla	r0, r",0,  VAL1,", r",1, VAL2,", r",2, VAL1,"")
+	TEST_RRR(  "smmla	r14, r",12,VAL2,", r",10,VAL1,", r",8, VAL2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe75f8a1c) "	@ smmla	pc, r12, r10, r8")
+	TEST_RRR(  "smmlar	r0, r",0,  VAL1,", r",1, VAL2,", r",2, VAL1,"")
+	TEST_RRR(  "smmlar	r14, r",12,VAL2,", r",10,VAL1,", r",8, VAL2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe75f8a3c) "	@ smmlar	pc, r12, r10, r8")
+
+	TEST_RR(   "smmul	r0, r",0,  VAL1,", r",1, VAL2,"")
+	TEST_RR(   "smmul	r14, r",12,VAL2,", r",10,VAL1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe75ffa1c) "	@ smmul	pc, r12, r10")
+	TEST_RR(   "smmulr	r0, r",0,  VAL1,", r",1, VAL2,"")
+	TEST_RR(   "smmulr	r14, r",12,VAL2,", r",10,VAL1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe75ffa3c) "	@ smmulr	pc, r12, r10")
+
+	TEST_RRR(  "smmls	r0, r",0,  VAL1,", r",1, VAL2,", r",2, VAL1,"")
+	TEST_RRR(  "smmls	r14, r",12,VAL2,", r",10,VAL1,", r",8, VAL2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe75f8adc) "	@ smmls	pc, r12, r10, r8")
+	TEST_RRR(  "smmlsr	r0, r",0,  VAL1,", r",1, VAL2,", r",2, VAL1,"")
+	TEST_RRR(  "smmlsr	r14, r",12,VAL2,", r",10,VAL1,", r",8, VAL2,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe75f8afc) "	@ smmlsr	pc, r12, r10, r8")
+	TEST_UNSUPPORTED(__inst_arm(0xe75e8aff) "	@ smmlsr	r14, pc, r10, r8")
+	TEST_UNSUPPORTED(__inst_arm(0xe75e8ffc) "	@ smmlsr	r14, r12, pc, r8")
+	TEST_UNSUPPORTED(__inst_arm(0xe75efafc) "	@ smmlsr	r14, r12, r10, pc")
+
+	TEST_RR(   "usad8	r0, r",0,  VAL1,", r",1, VAL2,"")
+	TEST_RR(   "usad8	r14, r",12,VAL2,", r",10,VAL1,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe75ffa1c) "	@ usad8	pc, r12, r10")
+	TEST_UNSUPPORTED(__inst_arm(0xe75efa1f) "	@ usad8	r14, pc, r10")
+	TEST_UNSUPPORTED(__inst_arm(0xe75eff1c) "	@ usad8	r14, r12, pc")
+
+	TEST_RRR(  "usada8	r0, r",0,  VAL1,", r",1, VAL2,", r",2, VAL3,"")
+	TEST_RRR(  "usada8	r14, r",12,VAL2,", r",10,VAL1,", r",8, VAL3,"")
+	TEST_UNSUPPORTED(__inst_arm(0xe78f8a1c) "	@ usada8	pc, r12, r10, r8")
+	TEST_UNSUPPORTED(__inst_arm(0xe78e8a1f) "	@ usada8	r14, pc, r10, r8")
+	TEST_UNSUPPORTED(__inst_arm(0xe78e8f1c) "	@ usada8	r14, r12, pc, r8")
+#endif /* __LINUX_ARM_ARCH__ >= 6 */
+
+#if __LINUX_ARM_ARCH__ >= 7
+	TEST_GROUP("Bit Field")
+
+	TEST_R(     "sbfx	r0, r",0  , VAL1,", #0, #31")
+	TEST_R(     "sbfxeq	r14, r",12, VAL2,", #8, #16")
+	TEST_R(     "sbfx	r4, r",10,  VAL1,", #16, #15")
+	TEST_UNSUPPORTED(__inst_arm(0xe7aff45c) "	@ sbfx	pc, r12, #8, #16")
+
+	TEST_R(     "ubfx	r0, r",0  , VAL1,", #0, #31")
+	TEST_R(     "ubfxcs	r14, r",12, VAL2,", #8, #16")
+	TEST_R(     "ubfx	r4, r",10,  VAL1,", #16, #15")
+	TEST_UNSUPPORTED(__inst_arm(0xe7eff45c) "	@ ubfx	pc, r12, #8, #16")
+	TEST_UNSUPPORTED(__inst_arm(0xe7efc45f) "	@ ubfx	r12, pc, #8, #16")
+
+	TEST_R(     "bfc	r",0, VAL1,", #4, #20")
+	TEST_R(     "bfcvs	r",14,VAL2,", #4, #20")
+	TEST_R(     "bfc	r",7, VAL1,", #0, #31")
+	TEST_R(     "bfc	r",8, VAL2,", #0, #31")
+	TEST_UNSUPPORTED(__inst_arm(0xe7def01f) "	@ bfc	pc, #0, #31");
+
+	TEST_RR(    "bfi	r",0, VAL1,", r",0  , VAL2,", #0, #31")
+	TEST_RR(    "bfipl	r",12,VAL1,", r",14 , VAL2,", #4, #20")
+	TEST_UNSUPPORTED(__inst_arm(0xe7d7f21e) "	@ bfi	pc, r14, #4, #20")
+
+	TEST_UNSUPPORTED(__inst_arm(0x07f000f0) "")  /* Permanently UNDEFINED */
+	TEST_UNSUPPORTED(__inst_arm(0x07ffffff) "")  /* Permanently UNDEFINED */
+#endif /* __LINUX_ARM_ARCH__ >= 6 */
+
+	TEST_GROUP("Branch, branch with link, and block data transfer")
+
+	TEST_P(   "stmda	r",0, 16*4,", {r0}")
+	TEST_P(   "stmeqda	r",4, 16*4,", {r0-r15}")
+	TEST_P(   "stmneda	r",8, 16*4,"!, {r8-r15}")
+	TEST_P(   "stmda	r",12,16*4,"!, {r1,r3,r5,r7,r8-r11,r14}")
+	TEST_P(   "stmda	r",13,0,   "!, {pc}")
+
+	TEST_P(   "ldmda	r",0, 16*4,", {r0}")
+	TEST_BF_P("ldmcsda	r",4, 15*4,", {r0-r15}")
+	TEST_BF_P("ldmccda	r",7, 15*4,"!, {r8-r15}")
+	TEST_P(   "ldmda	r",12,16*4,"!, {r1,r3,r5,r7,r8-r11,r14}")
+	TEST_BF_P("ldmda	r",14,15*4,"!, {pc}")
+
+	TEST_P(   "stmia	r",0, 16*4,", {r0}")
+	TEST_P(   "stmmiia	r",4, 16*4,", {r0-r15}")
+	TEST_P(   "stmplia	r",8, 16*4,"!, {r8-r15}")
+	TEST_P(   "stmia	r",12,16*4,"!, {r1,r3,r5,r7,r8-r11,r14}")
+	TEST_P(   "stmia	r",14,0,   "!, {pc}")
+
+	TEST_P(   "ldmia	r",0, 16*4,", {r0}")
+	TEST_BF_P("ldmvsia	r",4, 0,   ", {r0-r15}")
+	TEST_BF_P("ldmvcia	r",7, 8*4, "!, {r8-r15}")
+	TEST_P(   "ldmia	r",12,16*4,"!, {r1,r3,r5,r7,r8-r11,r14}")
+	TEST_BF_P("ldmia	r",14,15*4,"!, {pc}")
+
+	TEST_P(   "stmdb	r",0, 16*4,", {r0}")
+	TEST_P(   "stmhidb	r",4, 16*4,", {r0-r15}")
+	TEST_P(   "stmlsdb	r",8, 16*4,"!, {r8-r15}")
+	TEST_P(   "stmdb	r",12,16*4,"!, {r1,r3,r5,r7,r8-r11,r14}")
+	TEST_P(   "stmdb	r",13,4,   "!, {pc}")
+
+	TEST_P(   "ldmdb	r",0, 16*4,", {r0}")
+	TEST_BF_P("ldmgedb	r",4, 16*4,", {r0-r15}")
+	TEST_BF_P("ldmltdb	r",7, 16*4,"!, {r8-r15}")
+	TEST_P(   "ldmdb	r",12,16*4,"!, {r1,r3,r5,r7,r8-r11,r14}")
+	TEST_BF_P("ldmdb	r",14,16*4,"!, {pc}")
+
+	TEST_P(   "stmib	r",0, 16*4,", {r0}")
+	TEST_P(   "stmgtib	r",4, 16*4,", {r0-r15}")
+	TEST_P(   "stmleib	r",8, 16*4,"!, {r8-r15}")
+	TEST_P(   "stmib	r",12,16*4,"!, {r1,r3,r5,r7,r8-r11,r14}")
+	TEST_P(   "stmib	r",13,-4,  "!, {pc}")
+
+	TEST_P(   "ldmib	r",0, 16*4,", {r0}")
+	TEST_BF_P("ldmeqib	r",4, -4,", {r0-r15}")
+	TEST_BF_P("ldmneib	r",7, 7*4,"!, {r8-r15}")
+	TEST_P(   "ldmib	r",12,16*4,"!, {r1,r3,r5,r7,r8-r11,r14}")
+	TEST_BF_P("ldmib	r",14,14*4,"!, {pc}")
+
+	TEST_P(   "stmdb	r",13,16*4,"!, {r3-r12,lr}")
+	TEST_P(	  "stmeqdb	r",13,16*4,"!, {r3-r12}")
+	TEST_P(   "stmnedb	r",2, 16*4,", {r3-r12,lr}")
+	TEST_P(   "stmdb	r",13,16*4,"!, {r2-r12,lr}")
+	TEST_P(   "stmdb	r",0, 16*4,", {r0-r12}")
+	TEST_P(   "stmdb	r",0, 16*4,", {r0-r12,lr}")
+
+	TEST_BF_P("ldmia	r",13,5*4, "!, {r3-r12,pc}")
+	TEST_P(	  "ldmccia	r",13,5*4, "!, {r3-r12}")
+	TEST_BF_P("ldmcsia	r",2, 5*4, "!, {r3-r12,pc}")
+	TEST_BF_P("ldmia	r",13,4*4, "!, {r2-r12,pc}")
+	TEST_P(   "ldmia	r",0, 16*4,", {r0-r12}")
+	TEST_P(   "ldmia	r",0, 16*4,", {r0-r12,lr}")
+
+#ifdef CONFIG_THUMB2_KERNEL
+	TEST_ARM_TO_THUMB_INTERWORK_P("ldmplia	r",0,15*4,", {pc}")
+	TEST_ARM_TO_THUMB_INTERWORK_P("ldmmiia	r",13,0,", {r0-r15}")
+#endif
+	TEST_BF("b	2f")
+	TEST_BF("bl	2f")
+	TEST_BB("b	2b")
+	TEST_BB("bl	2b")
+
+	TEST_BF("beq	2f")
+	TEST_BF("bleq	2f")
+	TEST_BB("bne	2b")
+	TEST_BB("blne	2b")
+
+	TEST_BF("bgt	2f")
+	TEST_BF("blgt	2f")
+	TEST_BB("blt	2b")
+	TEST_BB("bllt	2b")
+
+	TEST_GROUP("Supervisor Call, and coprocessor instructions")
+
+	/*
+	 * We can't really test these by executing them, so all
+	 * we can do is check that probes are, or are not allowed.
+	 * At the moment none are allowed...
+	 */
+#define TEST_COPROCESSOR(code) TEST_UNSUPPORTED(code)
+
+#define COPROCESSOR_INSTRUCTIONS_ST_LD(two,cc)					\
+	TEST_COPROCESSOR("stc"two"	0, cr0, [r13, #4]")			\
+	TEST_COPROCESSOR("stc"two"	0, cr0, [r13, #-4]")			\
+	TEST_COPROCESSOR("stc"two"	0, cr0, [r13, #4]!")			\
+	TEST_COPROCESSOR("stc"two"	0, cr0, [r13, #-4]!")			\
+	TEST_COPROCESSOR("stc"two"	0, cr0, [r13], #4")			\
+	TEST_COPROCESSOR("stc"two"	0, cr0, [r13], #-4")			\
+	TEST_COPROCESSOR("stc"two"	0, cr0, [r13], {1}")			\
+	TEST_COPROCESSOR("stc"two"l	0, cr0, [r13, #4]")			\
+	TEST_COPROCESSOR("stc"two"l	0, cr0, [r13, #-4]")			\
+	TEST_COPROCESSOR("stc"two"l	0, cr0, [r13, #4]!")			\
+	TEST_COPROCESSOR("stc"two"l	0, cr0, [r13, #-4]!")			\
+	TEST_COPROCESSOR("stc"two"l	0, cr0, [r13], #4")			\
+	TEST_COPROCESSOR("stc"two"l	0, cr0, [r13], #-4")			\
+	TEST_COPROCESSOR("stc"two"l	0, cr0, [r13], {1}")			\
+	TEST_COPROCESSOR("ldc"two"	0, cr0, [r13, #4]")			\
+	TEST_COPROCESSOR("ldc"two"	0, cr0, [r13, #-4]")			\
+	TEST_COPROCESSOR("ldc"two"	0, cr0, [r13, #4]!")			\
+	TEST_COPROCESSOR("ldc"two"	0, cr0, [r13, #-4]!")			\
+	TEST_COPROCESSOR("ldc"two"	0, cr0, [r13], #4")			\
+	TEST_COPROCESSOR("ldc"two"	0, cr0, [r13], #-4")			\
+	TEST_COPROCESSOR("ldc"two"	0, cr0, [r13], {1}")			\
+	TEST_COPROCESSOR("ldc"two"l	0, cr0, [r13, #4]")			\
+	TEST_COPROCESSOR("ldc"two"l	0, cr0, [r13, #-4]")			\
+	TEST_COPROCESSOR("ldc"two"l	0, cr0, [r13, #4]!")			\
+	TEST_COPROCESSOR("ldc"two"l	0, cr0, [r13, #-4]!")			\
+	TEST_COPROCESSOR("ldc"two"l	0, cr0, [r13], #4")			\
+	TEST_COPROCESSOR("ldc"two"l	0, cr0, [r13], #-4")			\
+	TEST_COPROCESSOR("ldc"two"l	0, cr0, [r13], {1}")			\
+										\
+	TEST_COPROCESSOR( "stc"two"	0, cr0, [r15, #4]")			\
+	TEST_COPROCESSOR( "stc"two"	0, cr0, [r15, #-4]")			\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##daf0001) "	@ stc"two"	0, cr0, [r15, #4]!")	\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##d2f0001) "	@ stc"two"	0, cr0, [r15, #-4]!")	\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##caf0001) "	@ stc"two"	0, cr0, [r15], #4")	\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##c2f0001) "	@ stc"two"	0, cr0, [r15], #-4")	\
+	TEST_COPROCESSOR( "stc"two"	0, cr0, [r15], {1}")			\
+	TEST_COPROCESSOR( "stc"two"l	0, cr0, [r15, #4]")			\
+	TEST_COPROCESSOR( "stc"two"l	0, cr0, [r15, #-4]")			\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##def0001) "	@ stc"two"l	0, cr0, [r15, #4]!")	\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##d6f0001) "	@ stc"two"l	0, cr0, [r15, #-4]!")	\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##cef0001) "	@ stc"two"l	0, cr0, [r15], #4")	\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##c6f0001) "	@ stc"two"l	0, cr0, [r15], #-4")	\
+	TEST_COPROCESSOR( "stc"two"l	0, cr0, [r15], {1}")			\
+	TEST_COPROCESSOR( "ldc"two"	0, cr0, [r15, #4]")			\
+	TEST_COPROCESSOR( "ldc"two"	0, cr0, [r15, #-4]")			\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##dbf0001) "	@ ldc"two"	0, cr0, [r15, #4]!")	\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##d3f0001) "	@ ldc"two"	0, cr0, [r15, #-4]!")	\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##cbf0001) "	@ ldc"two"	0, cr0, [r15], #4")	\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##c3f0001) "	@ ldc"two"	0, cr0, [r15], #-4")	\
+	TEST_COPROCESSOR( "ldc"two"	0, cr0, [r15], {1}")			\
+	TEST_COPROCESSOR( "ldc"two"l	0, cr0, [r15, #4]")			\
+	TEST_COPROCESSOR( "ldc"two"l	0, cr0, [r15, #-4]")			\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##dff0001) "	@ ldc"two"l	0, cr0, [r15, #4]!")	\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##d7f0001) "	@ ldc"two"l	0, cr0, [r15, #-4]!")	\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##cff0001) "	@ ldc"two"l	0, cr0, [r15], #4")	\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##c7f0001) "	@ ldc"two"l	0, cr0, [r15], #-4")	\
+	TEST_COPROCESSOR( "ldc"two"l	0, cr0, [r15], {1}")
+
+#define COPROCESSOR_INSTRUCTIONS_MC_MR(two,cc)					\
+										\
+	TEST_COPROCESSOR( "mcrr"two"	0, 15, r0, r14, cr0")			\
+	TEST_COPROCESSOR( "mcrr"two"	15, 0, r14, r0, cr15")			\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##c4f00f0) "	@ mcrr"two"	0, 15, r0, r15, cr0")	\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##c40ff0f) "	@ mcrr"two"	15, 0, r15, r0, cr15")	\
+	TEST_COPROCESSOR( "mrrc"two"	0, 15, r0, r14, cr0")			\
+	TEST_COPROCESSOR( "mrrc"two"	15, 0, r14, r0, cr15")			\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##c5f00f0) "	@ mrrc"two"	0, 15, r0, r15, cr0")	\
+	TEST_UNSUPPORTED(__inst_arm(0x##cc##c50ff0f) "	@ mrrc"two"	15, 0, r15, r0, cr15")	\
+	TEST_COPROCESSOR( "cdp"two"	15, 15, cr15, cr15, cr15, 7")		\
+	TEST_COPROCESSOR( "cdp"two"	0, 0, cr0, cr0, cr0, 0")		\
+	TEST_COPROCESSOR( "mcr"two"	15, 7, r15, cr15, cr15, 7")		\
+	TEST_COPROCESSOR( "mcr"two"	0, 0, r0, cr0, cr0, 0")			\
+	TEST_COPROCESSOR( "mrc"two"	15, 7, r15, cr15, cr15, 7")		\
+	TEST_COPROCESSOR( "mrc"two"	0, 0, r0, cr0, cr0, 0")
+
+	COPROCESSOR_INSTRUCTIONS_ST_LD("",e)
+#if __LINUX_ARM_ARCH__ >= 5
+	COPROCESSOR_INSTRUCTIONS_MC_MR("",e)
+#endif
+	TEST_UNSUPPORTED("svc	0")
+	TEST_UNSUPPORTED("svc	0xffffff")
+
+	TEST_UNSUPPORTED("svc	0")
+
+	TEST_GROUP("Unconditional instruction")
+
+#if __LINUX_ARM_ARCH__ >= 6
+	TEST_UNSUPPORTED("srsda	sp, 0x13")
+	TEST_UNSUPPORTED("srsdb	sp, 0x13")
+	TEST_UNSUPPORTED("srsia	sp, 0x13")
+	TEST_UNSUPPORTED("srsib	sp, 0x13")
+	TEST_UNSUPPORTED("srsda	sp!, 0x13")
+	TEST_UNSUPPORTED("srsdb	sp!, 0x13")
+	TEST_UNSUPPORTED("srsia	sp!, 0x13")
+	TEST_UNSUPPORTED("srsib	sp!, 0x13")
+
+	TEST_UNSUPPORTED("rfeda	sp")
+	TEST_UNSUPPORTED("rfedb	sp")
+	TEST_UNSUPPORTED("rfeia	sp")
+	TEST_UNSUPPORTED("rfeib	sp")
+	TEST_UNSUPPORTED("rfeda	sp!")
+	TEST_UNSUPPORTED("rfedb	sp!")
+	TEST_UNSUPPORTED("rfeia	sp!")
+	TEST_UNSUPPORTED("rfeib	sp!")
+	TEST_UNSUPPORTED(__inst_arm(0xf81d0a00) "	@ rfeda	pc")
+	TEST_UNSUPPORTED(__inst_arm(0xf91d0a00) "	@ rfedb	pc")
+	TEST_UNSUPPORTED(__inst_arm(0xf89d0a00) "	@ rfeia	pc")
+	TEST_UNSUPPORTED(__inst_arm(0xf99d0a00) "	@ rfeib	pc")
+	TEST_UNSUPPORTED(__inst_arm(0xf83d0a00) "	@ rfeda	pc!")
+	TEST_UNSUPPORTED(__inst_arm(0xf93d0a00) "	@ rfedb	pc!")
+	TEST_UNSUPPORTED(__inst_arm(0xf8bd0a00) "	@ rfeia	pc!")
+	TEST_UNSUPPORTED(__inst_arm(0xf9bd0a00) "	@ rfeib	pc!")
+#endif /* __LINUX_ARM_ARCH__ >= 6 */
+
+#if __LINUX_ARM_ARCH__ >= 6
+	TEST_X(	"blx	__dummy_thumb_subroutine_even",
+		".thumb				\n\t"
+		".space 4			\n\t"
+		".type __dummy_thumb_subroutine_even, %%function \n\t"
+		"__dummy_thumb_subroutine_even:	\n\t"
+		"mov	r0, pc			\n\t"
+		"bx	lr			\n\t"
+		".arm				\n\t"
+	)
+	TEST(	"blx	__dummy_thumb_subroutine_even")
+
+	TEST_X(	"blx	__dummy_thumb_subroutine_odd",
+		".thumb				\n\t"
+		".space 2			\n\t"
+		".type __dummy_thumb_subroutine_odd, %%function	\n\t"
+		"__dummy_thumb_subroutine_odd:	\n\t"
+		"mov	r0, pc			\n\t"
+		"bx	lr			\n\t"
+		".arm				\n\t"
+	)
+	TEST(	"blx	__dummy_thumb_subroutine_odd")
+#endif /* __LINUX_ARM_ARCH__ >= 6 */
+
+#if __LINUX_ARM_ARCH__ >= 5
+	COPROCESSOR_INSTRUCTIONS_ST_LD("2",f)
+#endif
+#if __LINUX_ARM_ARCH__ >= 6
+	COPROCESSOR_INSTRUCTIONS_MC_MR("2",f)
+#endif
+
+	TEST_GROUP("Miscellaneous instructions, memory hints, and Advanced SIMD instructions")
+
+#if __LINUX_ARM_ARCH__ >= 6
+	TEST_UNSUPPORTED("cps	0x13")
+	TEST_UNSUPPORTED("cpsie	i")
+	TEST_UNSUPPORTED("cpsid	i")
+	TEST_UNSUPPORTED("cpsie	i,0x13")
+	TEST_UNSUPPORTED("cpsid	i,0x13")
+	TEST_UNSUPPORTED("setend	le")
+	TEST_UNSUPPORTED("setend	be")
+#endif
+
+#if __LINUX_ARM_ARCH__ >= 7
+	TEST_P("pli	[r",0,0b,", #16]")
+	TEST(  "pli	[pc, #0]")
+	TEST_RR("pli	[r",12,0b,", r",0, 16,"]")
+	TEST_RR("pli	[r",0, 0b,", -r",12,16,", lsl #4]")
+#endif
+
+#if __LINUX_ARM_ARCH__ >= 5
+	TEST_P("pld	[r",0,32,", #-16]")
+	TEST(  "pld	[pc, #0]")
+	TEST_PR("pld	[r",7, 24, ", r",0, 16,"]")
+	TEST_PR("pld	[r",8, 24, ", -r",12,16,", lsl #4]")
+#endif
+
+#if __LINUX_ARM_ARCH__ >= 7
+	TEST_SUPPORTED(  __inst_arm(0xf590f000) "	@ pldw [r0, #0]")
+	TEST_SUPPORTED(  __inst_arm(0xf797f000) "	@ pldw	[r7, r0]")
+	TEST_SUPPORTED(  __inst_arm(0xf798f18c) "	@ pldw	[r8, r12, lsl #3]");
+#endif
+
+#if __LINUX_ARM_ARCH__ >= 7
+	TEST_UNSUPPORTED("clrex")
+	TEST_UNSUPPORTED("dsb")
+	TEST_UNSUPPORTED("dmb")
+	TEST_UNSUPPORTED("isb")
+#endif
+
+	verbose("\n");
+}
+

arch/arm/probes/kprobes/test-core.c

@@ -0,0 +1,1732 @@
+/*
+ * arch/arm/kernel/kprobes-test.c
+ *
+ * Copyright (C) 2011 Jon Medhurst <tixy@yxit.co.uk>.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/*
+ * This file contains test code for ARM kprobes.
+ *
+ * The top level function run_all_tests() executes tests for all of the
+ * supported instruction sets: ARM, 16-bit Thumb, and 32-bit Thumb. These tests
+ * fall into two categories; run_api_tests() checks basic functionality of the
+ * kprobes API, and run_test_cases() is a comprehensive test for kprobes
+ * instruction decoding and simulation.
+ *
+ * run_test_cases() first checks the kprobes decoding table for self consistency
+ * (using table_test()) then executes a series of test cases for each of the CPU
+ * instruction forms. coverage_start() and coverage_end() are used to verify
+ * that these test cases cover all of the possible combinations of instructions
+ * described by the kprobes decoding tables.
+ *
+ * The individual test cases are in kprobes-test-arm.c and kprobes-test-thumb.c
+ * which use the macros defined in kprobes-test.h. The rest of this
+ * documentation will describe the operation of the framework used by these
+ * test cases.
+ */
+
+/*
+ * TESTING METHODOLOGY
+ * -------------------
+ *
+ * The methodology used to test an ARM instruction 'test_insn' is to use
+ * inline assembler like:
+ *
+ * test_before: nop
+ * test_case:	test_insn
+ * test_after:	nop
+ *
+ * When the test case is run a kprobe is placed of each nop. The
+ * post-handler of the test_before probe is used to modify the saved CPU
+ * register context to that which we require for the test case. The
+ * pre-handler of the of the test_after probe saves a copy of the CPU
+ * register context. In this way we can execute test_insn with a specific
+ * register context and see the results afterwards.
+ *
+ * To actually test the kprobes instruction emulation we perform the above
+ * step a second time but with an additional kprobe on the test_case
+ * instruction itself. If the emulation is accurate then the results seen
+ * by the test_after probe will be identical to the first run which didn't
+ * have a probe on test_case.
+ *
+ * Each test case is run several times with a variety of variations in the
+ * flags value of stored in CPSR, and for Thumb code, different ITState.
+ *
+ * For instructions which can modify PC, a second test_after probe is used
+ * like this:
+ *
+ * test_before: nop
+ * test_case:	test_insn
+ * test_after:	nop
+ *		b test_done
+ * test_after2: nop
+ * test_done:
+ *
+ * The test case is constructed such that test_insn branches to
+ * test_after2, or, if testing a conditional instruction, it may just
+ * continue to test_after. The probes inserted at both locations let us
+ * determine which happened. A similar approach is used for testing
+ * backwards branches...
+ *
+ *		b test_before
+ *		b test_done  @ helps to cope with off by 1 branches
+ * test_after2: nop
+ *		b test_done
+ * test_before: nop
+ * test_case:	test_insn
+ * test_after:	nop
+ * test_done:
+ *
+ * The macros used to generate the assembler instructions describe above
+ * are TEST_INSTRUCTION, TEST_BRANCH_F (branch forwards) and TEST_BRANCH_B
+ * (branch backwards). In these, the local variables numbered 1, 50, 2 and
+ * 99 represent: test_before, test_case, test_after2 and test_done.
+ *
+ * FRAMEWORK
+ * ---------
+ *
+ * Each test case is wrapped between the pair of macros TESTCASE_START and
+ * TESTCASE_END. As well as performing the inline assembler boilerplate,
+ * these call out to the kprobes_test_case_start() and
+ * kprobes_test_case_end() functions which drive the execution of the test
+ * case. The specific arguments to use for each test case are stored as
+ * inline data constructed using the various TEST_ARG_* macros. Putting
+ * this all together, a simple test case may look like:
+ *
+ *	TESTCASE_START("Testing mov r0, r7")
+ *	TEST_ARG_REG(7, 0x12345678) // Set r7=0x12345678
+ *	TEST_ARG_END("")
+ *	TEST_INSTRUCTION("mov r0, r7")
+ *	TESTCASE_END
+ *
+ * Note, in practice the single convenience macro TEST_R would be used for this
+ * instead.
+ *
+ * The above would expand to assembler looking something like:
+ *
+ *	@ TESTCASE_START
+ *	bl	__kprobes_test_case_start
+ *	.pushsection .rodata
+ *	"10:
+ *	.ascii "mov r0, r7"	@ text title for test case
+ *	.byte	0
+ *	.popsection
+ *	@ start of inline data...
+ *	.word	10b		@ pointer to title in .rodata section
+ *
+ *	@ TEST_ARG_REG
+ *	.byte	ARG_TYPE_REG
+ *	.byte	7
+ *	.short	0
+ *	.word	0x1234567
+ *
+ *	@ TEST_ARG_END
+ *	.byte	ARG_TYPE_END
+ *	.byte	TEST_ISA	@ flags, including ISA being tested
+ *	.short	50f-0f		@ offset of 'test_before'
+ *	.short	2f-0f		@ offset of 'test_after2' (if relevent)
+ *	.short	99f-0f		@ offset of 'test_done'
+ *	@ start of test case code...
+ *	0:
+ *	.code	TEST_ISA	@ switch to ISA being tested
+ *
+ *	@ TEST_INSTRUCTION
+ *	50:	nop		@ location for 'test_before' probe
+ *	1:	mov r0, r7	@ the test case instruction 'test_insn'
+ *		nop		@ location for 'test_after' probe
+ *
+ *	// TESTCASE_END
+ *	2:
+ *	99:	bl __kprobes_test_case_end_##TEST_ISA
+ *	.code	NONMAL_ISA
+ *
+ * When the above is execute the following happens...
+ *
+ * __kprobes_test_case_start() is an assembler wrapper which sets up space
+ * for a stack buffer and calls the C function kprobes_test_case_start().
+ * This C function will do some initial processing of the inline data and
+ * setup some global state. It then inserts the test_before and test_after
+ * kprobes and returns a value which causes the assembler wrapper to jump
+ * to the start of the test case code, (local label '0').
+ *
+ * When the test case code executes, the test_before probe will be hit and
+ * test_before_post_handler will call setup_test_context(). This fills the
+ * stack buffer and CPU registers with a test pattern and then processes
+ * the test case arguments. In our example there is one TEST_ARG_REG which
+ * indicates that R7 should be loaded with the value 0x12345678.
+ *
+ * When the test_before probe ends, the test case continues and executes
+ * the "mov r0, r7" instruction. It then hits the test_after probe and the
+ * pre-handler for this (test_after_pre_handler) will save a copy of the
+ * CPU register context. This should now have R0 holding the same value as
+ * R7.
+ *
+ * Finally we get to the call to __kprobes_test_case_end_{32,16}. This is
+ * an assembler wrapper which switches back to the ISA used by the test
+ * code and calls the C function kprobes_test_case_end().
+ *
+ * For each run through the test case, test_case_run_count is incremented
+ * by one. For even runs, kprobes_test_case_end() saves a copy of the
+ * register and stack buffer contents from the test case just run. It then
+ * inserts a kprobe on the test case instruction 'test_insn' and returns a
+ * value to cause the test case code to be re-run.
+ *
+ * For odd numbered runs, kprobes_test_case_end() compares the register and
+ * stack buffer contents to those that were saved on the previous even
+ * numbered run (the one without the kprobe on test_insn). These should be
+ * the same if the kprobe instruction simulation routine is correct.
+ *
+ * The pair of test case runs is repeated with different combinations of
+ * flag values in CPSR and, for Thumb, different ITState. This is
+ * controlled by test_context_cpsr().
+ *
+ * BUILDING TEST CASES
+ * -------------------
+ *
+ *
+ * As an aid to building test cases, the stack buffer is initialised with
+ * some special values:
+ *
+ *   [SP+13*4]	Contains SP+120. This can be used to test instructions
+ *		which load a value into SP.
+ *
+ *   [SP+15*4]	When testing branching instructions using TEST_BRANCH_{F,B},
+ *		this holds the target address of the branch, 'test_after2'.
+ *		This can be used to test instructions which load a PC value
+ *		from memory.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/kprobes.h>
+#include <linux/errno.h>
+#include <linux/stddef.h>
+#include <linux/bug.h>
+#include <asm/opcodes.h>
+
+#include "core.h"
+#include "test-core.h"
+#include "../decode-arm.h"
+#include "../decode-thumb.h"
+
+
+#define BENCHMARKING	1
+
+
+/*
+ * Test basic API
+ */
+
+static bool test_regs_ok;
+static int test_func_instance;
+static int pre_handler_called;
+static int post_handler_called;
+static int jprobe_func_called;
+static int kretprobe_handler_called;
+static int tests_failed;
+
+#define FUNC_ARG1 0x12345678
+#define FUNC_ARG2 0xabcdef
+
+
+#ifndef CONFIG_THUMB2_KERNEL
+
+#define RET(reg)	"mov	pc, "#reg
+
+long arm_func(long r0, long r1);
+
+static void __used __naked __arm_kprobes_test_func(void)
+{
+	__asm__ __volatile__ (
+		".arm					\n\t"
+		".type arm_func, %%function		\n\t"
+		"arm_func:				\n\t"
+		"adds	r0, r0, r1			\n\t"
+		"mov	pc, lr				\n\t"
+		".code "NORMAL_ISA	 /* Back to Thumb if necessary */
+		: : : "r0", "r1", "cc"
+	);
+}
+
+#else /* CONFIG_THUMB2_KERNEL */
+
+#define RET(reg)	"bx	"#reg
+
+long thumb16_func(long r0, long r1);
+long thumb32even_func(long r0, long r1);
+long thumb32odd_func(long r0, long r1);
+
+static void __used __naked __thumb_kprobes_test_funcs(void)
+{
+	__asm__ __volatile__ (
+		".type thumb16_func, %%function		\n\t"
+		"thumb16_func:				\n\t"
+		"adds.n	r0, r0, r1			\n\t"
+		"bx	lr				\n\t"
+
+		".align					\n\t"
+		".type thumb32even_func, %%function	\n\t"
+		"thumb32even_func:			\n\t"
+		"adds.w	r0, r0, r1			\n\t"
+		"bx	lr				\n\t"
+
+		".align					\n\t"
+		"nop.n					\n\t"
+		".type thumb32odd_func, %%function	\n\t"
+		"thumb32odd_func:			\n\t"
+		"adds.w	r0, r0, r1			\n\t"
+		"bx	lr				\n\t"
+
+		: : : "r0", "r1", "cc"
+	);
+}
+
+#endif /* CONFIG_THUMB2_KERNEL */
+
+
+static int call_test_func(long (*func)(long, long), bool check_test_regs)
+{
+	long ret;
+
+	++test_func_instance;
+	test_regs_ok = false;
+
+	ret = (*func)(FUNC_ARG1, FUNC_ARG2);
+	if (ret != FUNC_ARG1 + FUNC_ARG2) {
+		pr_err("FAIL: call_test_func: func returned %lx\n", ret);
+		return false;
+	}
+
+	if (check_test_regs && !test_regs_ok) {
+		pr_err("FAIL: test regs not OK\n");
+		return false;
+	}
+
+	return true;
+}
+
+static int __kprobes pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	pre_handler_called = test_func_instance;
+	if (regs->ARM_r0 == FUNC_ARG1 && regs->ARM_r1 == FUNC_ARG2)
+		test_regs_ok = true;
+	return 0;
+}
+
+static void __kprobes post_handler(struct kprobe *p, struct pt_regs *regs,
+				unsigned long flags)
+{
+	post_handler_called = test_func_instance;
+	if (regs->ARM_r0 != FUNC_ARG1 + FUNC_ARG2 || regs->ARM_r1 != FUNC_ARG2)
+		test_regs_ok = false;
+}
+
+static struct kprobe the_kprobe = {
+	.addr		= 0,
+	.pre_handler	= pre_handler,
+	.post_handler	= post_handler
+};
+
+static int test_kprobe(long (*func)(long, long))
+{
+	int ret;
+
+	the_kprobe.addr = (kprobe_opcode_t *)func;
+	ret = register_kprobe(&the_kprobe);
+	if (ret < 0) {
+		pr_err("FAIL: register_kprobe failed with %d\n", ret);
+		return ret;
+	}
+
+	ret = call_test_func(func, true);
+
+	unregister_kprobe(&the_kprobe);
+	the_kprobe.flags = 0; /* Clear disable flag to allow reuse */
+
+	if (!ret)
+		return -EINVAL;
+	if (pre_handler_called != test_func_instance) {
+		pr_err("FAIL: kprobe pre_handler not called\n");
+		return -EINVAL;
+	}
+	if (post_handler_called != test_func_instance) {
+		pr_err("FAIL: kprobe post_handler not called\n");
+		return -EINVAL;
+	}
+	if (!call_test_func(func, false))
+		return -EINVAL;
+	if (pre_handler_called == test_func_instance ||
+				post_handler_called == test_func_instance) {
+		pr_err("FAIL: probe called after unregistering\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void __kprobes jprobe_func(long r0, long r1)
+{
+	jprobe_func_called = test_func_instance;
+	if (r0 == FUNC_ARG1 && r1 == FUNC_ARG2)
+		test_regs_ok = true;
+	jprobe_return();
+}
+
+static struct jprobe the_jprobe = {
+	.entry		= jprobe_func,
+};
+
+static int test_jprobe(long (*func)(long, long))
+{
+	int ret;
+
+	the_jprobe.kp.addr = (kprobe_opcode_t *)func;
+	ret = register_jprobe(&the_jprobe);
+	if (ret < 0) {
+		pr_err("FAIL: register_jprobe failed with %d\n", ret);
+		return ret;
+	}
+
+	ret = call_test_func(func, true);
+
+	unregister_jprobe(&the_jprobe);
+	the_jprobe.kp.flags = 0; /* Clear disable flag to allow reuse */
+
+	if (!ret)
+		return -EINVAL;
+	if (jprobe_func_called != test_func_instance) {
+		pr_err("FAIL: jprobe handler function not called\n");
+		return -EINVAL;
+	}
+	if (!call_test_func(func, false))
+		return -EINVAL;
+	if (jprobe_func_called == test_func_instance) {
+		pr_err("FAIL: probe called after unregistering\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int __kprobes
+kretprobe_handler(struct kretprobe_instance *ri, struct pt_regs *regs)
+{
+	kretprobe_handler_called = test_func_instance;
+	if (regs_return_value(regs) == FUNC_ARG1 + FUNC_ARG2)
+		test_regs_ok = true;
+	return 0;
+}
+
+static struct kretprobe the_kretprobe = {
+	.handler	= kretprobe_handler,
+};
+
+static int test_kretprobe(long (*func)(long, long))
+{
+	int ret;
+
+	the_kretprobe.kp.addr = (kprobe_opcode_t *)func;
+	ret = register_kretprobe(&the_kretprobe);
+	if (ret < 0) {
+		pr_err("FAIL: register_kretprobe failed with %d\n", ret);
+		return ret;
+	}
+
+	ret = call_test_func(func, true);
+
+	unregister_kretprobe(&the_kretprobe);
+	the_kretprobe.kp.flags = 0; /* Clear disable flag to allow reuse */
+
+	if (!ret)
+		return -EINVAL;
+	if (kretprobe_handler_called != test_func_instance) {
+		pr_err("FAIL: kretprobe handler not called\n");
+		return -EINVAL;
+	}
+	if (!call_test_func(func, false))
+		return -EINVAL;
+	if (jprobe_func_called == test_func_instance) {
+		pr_err("FAIL: kretprobe called after unregistering\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int run_api_tests(long (*func)(long, long))
+{
+	int ret;
+
+	pr_info("    kprobe\n");
+	ret = test_kprobe(func);
+	if (ret < 0)
+		return ret;
+
+	pr_info("    jprobe\n");
+	ret = test_jprobe(func);
+#if defined(CONFIG_THUMB2_KERNEL) && !defined(MODULE)
+	if (ret == -EINVAL) {
+		pr_err("FAIL: Known longtime bug with jprobe on Thumb kernels\n");
+		tests_failed = ret;
+		ret = 0;
+	}
+#endif
+	if (ret < 0)
+		return ret;
+
+	pr_info("    kretprobe\n");
+	ret = test_kretprobe(func);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+
+/*
+ * Benchmarking
+ */
+
+#if BENCHMARKING
+
+static void __naked benchmark_nop(void)
+{
+	__asm__ __volatile__ (
+		"nop		\n\t"
+		RET(lr)"	\n\t"
+	);
+}
+
+#ifdef CONFIG_THUMB2_KERNEL
+#define wide ".w"
+#else
+#define wide
+#endif
+
+static void __naked benchmark_pushpop1(void)
+{
+	__asm__ __volatile__ (
+		"stmdb"wide"	sp!, {r3-r11,lr}  \n\t"
+		"ldmia"wide"	sp!, {r3-r11,pc}"
+	);
+}
+
+static void __naked benchmark_pushpop2(void)
+{
+	__asm__ __volatile__ (
+		"stmdb"wide"	sp!, {r0-r8,lr}  \n\t"
+		"ldmia"wide"	sp!, {r0-r8,pc}"
+	);
+}
+
+static void __naked benchmark_pushpop3(void)
+{
+	__asm__ __volatile__ (
+		"stmdb"wide"	sp!, {r4,lr}  \n\t"
+		"ldmia"wide"	sp!, {r4,pc}"
+	);
+}
+
+static void __naked benchmark_pushpop4(void)
+{
+	__asm__ __volatile__ (
+		"stmdb"wide"	sp!, {r0,lr}  \n\t"
+		"ldmia"wide"	sp!, {r0,pc}"
+	);
+}
+
+
+#ifdef CONFIG_THUMB2_KERNEL
+
+static void __naked benchmark_pushpop_thumb(void)
+{
+	__asm__ __volatile__ (
+		"push.n	{r0-r7,lr}  \n\t"
+		"pop.n	{r0-r7,pc}"
+	);
+}
+
+#endif
+
+static int __kprobes
+benchmark_pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	return 0;
+}
+
+static int benchmark(void(*fn)(void))
+{
+	unsigned n, i, t, t0;
+
+	for (n = 1000; ; n *= 2) {
+		t0 = sched_clock();
+		for (i = n; i > 0; --i)
+			fn();
+		t = sched_clock() - t0;
+		if (t >= 250000000)
+			break; /* Stop once we took more than 0.25 seconds */
+	}
+	return t / n; /* Time for one iteration in nanoseconds */
+};
+
+static int kprobe_benchmark(void(*fn)(void), unsigned offset)
+{
+	struct kprobe k = {
+		.addr		= (kprobe_opcode_t *)((uintptr_t)fn + offset),
+		.pre_handler	= benchmark_pre_handler,
+	};
+
+	int ret = register_kprobe(&k);
+	if (ret < 0) {
+		pr_err("FAIL: register_kprobe failed with %d\n", ret);
+		return ret;
+	}
+
+	ret = benchmark(fn);
+
+	unregister_kprobe(&k);
+	return ret;
+};
+
+struct benchmarks {
+	void		(*fn)(void);
+	unsigned	offset;
+	const char	*title;
+};
+
+static int run_benchmarks(void)
+{
+	int ret;
+	struct benchmarks list[] = {
+		{&benchmark_nop, 0, "nop"},
+		/*
+		 * benchmark_pushpop{1,3} will have the optimised
+		 * instruction emulation, whilst benchmark_pushpop{2,4} will
+		 * be the equivalent unoptimised instructions.
+		 */
+		{&benchmark_pushpop1, 0, "stmdb	sp!, {r3-r11,lr}"},
+		{&benchmark_pushpop1, 4, "ldmia	sp!, {r3-r11,pc}"},
+		{&benchmark_pushpop2, 0, "stmdb	sp!, {r0-r8,lr}"},
+		{&benchmark_pushpop2, 4, "ldmia	sp!, {r0-r8,pc}"},
+		{&benchmark_pushpop3, 0, "stmdb	sp!, {r4,lr}"},
+		{&benchmark_pushpop3, 4, "ldmia	sp!, {r4,pc}"},
+		{&benchmark_pushpop4, 0, "stmdb	sp!, {r0,lr}"},
+		{&benchmark_pushpop4, 4, "ldmia	sp!, {r0,pc}"},
+#ifdef CONFIG_THUMB2_KERNEL
+		{&benchmark_pushpop_thumb, 0, "push.n	{r0-r7,lr}"},
+		{&benchmark_pushpop_thumb, 2, "pop.n	{r0-r7,pc}"},
+#endif
+		{0}
+	};
+
+	struct benchmarks *b;
+	for (b = list; b->fn; ++b) {
+		ret = kprobe_benchmark(b->fn, b->offset);
+		if (ret < 0)
+			return ret;
+		pr_info("    %dns for kprobe %s\n", ret, b->title);
+	}
+
+	pr_info("\n");
+	return 0;
+}
+
+#endif /* BENCHMARKING */
+
+
+/*
+ * Decoding table self-consistency tests
+ */
+
+static const int decode_struct_sizes[NUM_DECODE_TYPES] = {
+	[DECODE_TYPE_TABLE]	= sizeof(struct decode_table),
+	[DECODE_TYPE_CUSTOM]	= sizeof(struct decode_custom),
+	[DECODE_TYPE_SIMULATE]	= sizeof(struct decode_simulate),
+	[DECODE_TYPE_EMULATE]	= sizeof(struct decode_emulate),
+	[DECODE_TYPE_OR]	= sizeof(struct decode_or),
+	[DECODE_TYPE_REJECT]	= sizeof(struct decode_reject)
+};
+
+static int table_iter(const union decode_item *table,
+			int (*fn)(const struct decode_header *, void *),
+			void *args)
+{
+	const struct decode_header *h = (struct decode_header *)table;
+	int result;
+
+	for (;;) {
+		enum decode_type type = h->type_regs.bits & DECODE_TYPE_MASK;
+
+		if (type == DECODE_TYPE_END)
+			return 0;
+
+		result = fn(h, args);
+		if (result)
+			return result;
+
+		h = (struct decode_header *)
+			((uintptr_t)h + decode_struct_sizes[type]);
+
+	}
+}
+
+static int table_test_fail(const struct decode_header *h, const char* message)
+{
+
+	pr_err("FAIL: kprobes test failure \"%s\" (mask %08x, value %08x)\n",
+					message, h->mask.bits, h->value.bits);
+	return -EINVAL;
+}
+
+struct table_test_args {
+	const union decode_item *root_table;
+	u32			parent_mask;
+	u32			parent_value;
+};
+
+static int table_test_fn(const struct decode_header *h, void *args)
+{
+	struct table_test_args *a = (struct table_test_args *)args;
+	enum decode_type type = h->type_regs.bits & DECODE_TYPE_MASK;
+
+	if (h->value.bits & ~h->mask.bits)
+		return table_test_fail(h, "Match value has bits not in mask");
+
+	if ((h->mask.bits & a->parent_mask) != a->parent_mask)
+		return table_test_fail(h, "Mask has bits not in parent mask");
+
+	if ((h->value.bits ^ a->parent_value) & a->parent_mask)
+		return table_test_fail(h, "Value is inconsistent with parent");
+
+	if (type == DECODE_TYPE_TABLE) {
+		struct decode_table *d = (struct decode_table *)h;
+		struct table_test_args args2 = *a;
+		args2.parent_mask = h->mask.bits;
+		args2.parent_value = h->value.bits;
+		return table_iter(d->table.table, table_test_fn, &args2);
+	}
+
+	return 0;
+}
+
+static int table_test(const union decode_item *table)
+{
+	struct table_test_args args = {
+		.root_table	= table,
+		.parent_mask	= 0,
+		.parent_value	= 0
+	};
+	return table_iter(args.root_table, table_test_fn, &args);
+}
+
+
+/*
+ * Decoding table test coverage analysis
+ *
+ * coverage_start() builds a coverage_table which contains a list of
+ * coverage_entry's to match each entry in the specified kprobes instruction
+ * decoding table.
+ *
+ * When test cases are run, coverage_add() is called to process each case.
+ * This looks up the corresponding entry in the coverage_table and sets it as
+ * being matched, as well as clearing the regs flag appropriate for the test.
+ *
+ * After all test cases have been run, coverage_end() is called to check that
+ * all entries in coverage_table have been matched and that all regs flags are
+ * cleared. I.e. that all possible combinations of instructions described by
+ * the kprobes decoding tables have had a test case executed for them.
+ */
+
+bool coverage_fail;
+
+#define MAX_COVERAGE_ENTRIES 256
+
+struct coverage_entry {
+	const struct decode_header	*header;
+	unsigned			regs;
+	unsigned			nesting;
+	char				matched;
+};
+
+struct coverage_table {
+	struct coverage_entry	*base;
+	unsigned		num_entries;
+	unsigned		nesting;
+};
+
+struct coverage_table coverage;
+
+#define COVERAGE_ANY_REG	(1<<0)
+#define COVERAGE_SP		(1<<1)
+#define COVERAGE_PC		(1<<2)
+#define COVERAGE_PCWB		(1<<3)
+
+static const char coverage_register_lookup[16] = {
+	[REG_TYPE_ANY]		= COVERAGE_ANY_REG | COVERAGE_SP | COVERAGE_PC,
+	[REG_TYPE_SAMEAS16]	= COVERAGE_ANY_REG,
+	[REG_TYPE_SP]		= COVERAGE_SP,
+	[REG_TYPE_PC]		= COVERAGE_PC,
+	[REG_TYPE_NOSP]		= COVERAGE_ANY_REG | COVERAGE_SP,
+	[REG_TYPE_NOSPPC]	= COVERAGE_ANY_REG | COVERAGE_SP | COVERAGE_PC,
+	[REG_TYPE_NOPC]		= COVERAGE_ANY_REG | COVERAGE_PC,
+	[REG_TYPE_NOPCWB]	= COVERAGE_ANY_REG | COVERAGE_PC | COVERAGE_PCWB,
+	[REG_TYPE_NOPCX]	= COVERAGE_ANY_REG,
+	[REG_TYPE_NOSPPCX]	= COVERAGE_ANY_REG | COVERAGE_SP,
+};
+
+unsigned coverage_start_registers(const struct decode_header *h)
+{
+	unsigned regs = 0;
+	int i;
+	for (i = 0; i < 20; i += 4) {
+		int r = (h->type_regs.bits >> (DECODE_TYPE_BITS + i)) & 0xf;
+		regs |= coverage_register_lookup[r] << i;
+	}
+	return regs;
+}
+
+static int coverage_start_fn(const struct decode_header *h, void *args)
+{
+	struct coverage_table *coverage = (struct coverage_table *)args;
+	enum decode_type type = h->type_regs.bits & DECODE_TYPE_MASK;
+	struct coverage_entry *entry = coverage->base + coverage->num_entries;
+
+	if (coverage->num_entries == MAX_COVERAGE_ENTRIES - 1) {
+		pr_err("FAIL: Out of space for test coverage data");
+		return -ENOMEM;
+	}
+
+	++coverage->num_entries;
+
+	entry->header = h;
+	entry->regs = coverage_start_registers(h);
+	entry->nesting = coverage->nesting;
+	entry->matched = false;
+
+	if (type == DECODE_TYPE_TABLE) {
+		struct decode_table *d = (struct decode_table *)h;
+		int ret;
+		++coverage->nesting;
+		ret = table_iter(d->table.table, coverage_start_fn, coverage);
+		--coverage->nesting;
+		return ret;
+	}
+
+	return 0;
+}
+
+static int coverage_start(const union decode_item *table)
+{
+	coverage.base = kmalloc(MAX_COVERAGE_ENTRIES *
+				sizeof(struct coverage_entry), GFP_KERNEL);
+	coverage.num_entries = 0;
+	coverage.nesting = 0;
+	return table_iter(table, coverage_start_fn, &coverage);
+}
+
+static void
+coverage_add_registers(struct coverage_entry *entry, kprobe_opcode_t insn)
+{
+	int regs = entry->header->type_regs.bits >> DECODE_TYPE_BITS;
+	int i;
+	for (i = 0; i < 20; i += 4) {
+		enum decode_reg_type reg_type = (regs >> i) & 0xf;
+		int reg = (insn >> i) & 0xf;
+		int flag;
+
+		if (!reg_type)
+			continue;
+
+		if (reg == 13)
+			flag = COVERAGE_SP;
+		else if (reg == 15)
+			flag = COVERAGE_PC;
+		else
+			flag = COVERAGE_ANY_REG;
+		entry->regs &= ~(flag << i);
+
+		switch (reg_type) {
+
+		case REG_TYPE_NONE:
+		case REG_TYPE_ANY:
+		case REG_TYPE_SAMEAS16:
+			break;
+
+		case REG_TYPE_SP:
+			if (reg != 13)
+				return;
+			break;
+
+		case REG_TYPE_PC:
+			if (reg != 15)
+				return;
+			break;
+
+		case REG_TYPE_NOSP:
+			if (reg == 13)
+				return;
+			break;
+
+		case REG_TYPE_NOSPPC:
+		case REG_TYPE_NOSPPCX:
+			if (reg == 13 || reg == 15)
+				return;
+			break;
+
+		case REG_TYPE_NOPCWB:
+			if (!is_writeback(insn))
+				break;
+			if (reg == 15) {
+				entry->regs &= ~(COVERAGE_PCWB << i);
+				return;
+			}
+			break;
+
+		case REG_TYPE_NOPC:
+		case REG_TYPE_NOPCX:
+			if (reg == 15)
+				return;
+			break;
+		}
+
+	}
+}
+
+static void coverage_add(kprobe_opcode_t insn)
+{
+	struct coverage_entry *entry = coverage.base;
+	struct coverage_entry *end = coverage.base + coverage.num_entries;
+	bool matched = false;
+	unsigned nesting = 0;
+
+	for (; entry < end; ++entry) {
+		const struct decode_header *h = entry->header;
+		enum decode_type type = h->type_regs.bits & DECODE_TYPE_MASK;
+
+		if (entry->nesting > nesting)
+			continue; /* Skip sub-table we didn't match */
+
+		if (entry->nesting < nesting)
+			break; /* End of sub-table we were scanning */
+
+		if (!matched) {
+			if ((insn & h->mask.bits) != h->value.bits)
+				continue;
+			entry->matched = true;
+		}
+
+		switch (type) {
+
+		case DECODE_TYPE_TABLE:
+			++nesting;
+			break;
+
+		case DECODE_TYPE_CUSTOM:
+		case DECODE_TYPE_SIMULATE:
+		case DECODE_TYPE_EMULATE:
+			coverage_add_registers(entry, insn);
+			return;
+
+		case DECODE_TYPE_OR:
+			matched = true;
+			break;
+
+		case DECODE_TYPE_REJECT:
+		default:
+			return;
+		}
+
+	}
+}
+
+static void coverage_end(void)
+{
+	struct coverage_entry *entry = coverage.base;
+	struct coverage_entry *end = coverage.base + coverage.num_entries;
+
+	for (; entry < end; ++entry) {
+		u32 mask = entry->header->mask.bits;
+		u32 value = entry->header->value.bits;
+
+		if (entry->regs) {
+			pr_err("FAIL: Register test coverage missing for %08x %08x (%05x)\n",
+				mask, value, entry->regs);
+			coverage_fail = true;
+		}
+		if (!entry->matched) {
+			pr_err("FAIL: Test coverage entry missing for %08x %08x\n",
+				mask, value);
+			coverage_fail = true;
+		}
+	}
+
+	kfree(coverage.base);
+}
+
+
+/*
+ * Framework for instruction set test cases
+ */
+
+void __naked __kprobes_test_case_start(void)
+{
+	__asm__ __volatile__ (
+		"mov	r2, sp					\n\t"
+		"bic	r3, r2, #7				\n\t"
+		"mov	sp, r3					\n\t"
+		"stmdb	sp!, {r2-r11}				\n\t"
+		"sub	sp, sp, #"__stringify(TEST_MEMORY_SIZE)"\n\t"
+		"bic	r0, lr, #1  @ r0 = inline data		\n\t"
+		"mov	r1, sp					\n\t"
+		"bl	kprobes_test_case_start			\n\t"
+		RET(r0)"					\n\t"
+	);
+}
+
+#ifndef CONFIG_THUMB2_KERNEL
+
+void __naked __kprobes_test_case_end_32(void)
+{
+	__asm__ __volatile__ (
+		"mov	r4, lr					\n\t"
+		"bl	kprobes_test_case_end			\n\t"
+		"cmp	r0, #0					\n\t"
+		"movne	pc, r0					\n\t"
+		"mov	r0, r4					\n\t"
+		"add	sp, sp, #"__stringify(TEST_MEMORY_SIZE)"\n\t"
+		"ldmia	sp!, {r2-r11}				\n\t"
+		"mov	sp, r2					\n\t"
+		"mov	pc, r0					\n\t"
+	);
+}
+
+#else /* CONFIG_THUMB2_KERNEL */
+
+void __naked __kprobes_test_case_end_16(void)
+{
+	__asm__ __volatile__ (
+		"mov	r4, lr					\n\t"
+		"bl	kprobes_test_case_end			\n\t"
+		"cmp	r0, #0					\n\t"
+		"bxne	r0					\n\t"
+		"mov	r0, r4					\n\t"
+		"add	sp, sp, #"__stringify(TEST_MEMORY_SIZE)"\n\t"
+		"ldmia	sp!, {r2-r11}				\n\t"
+		"mov	sp, r2					\n\t"
+		"bx	r0					\n\t"
+	);
+}
+
+void __naked __kprobes_test_case_end_32(void)
+{
+	__asm__ __volatile__ (
+		".arm						\n\t"
+		"orr	lr, lr, #1  @ will return to Thumb code	\n\t"
+		"ldr	pc, 1f					\n\t"
+		"1:						\n\t"
+		".word	__kprobes_test_case_end_16		\n\t"
+	);
+}
+
+#endif
+
+
+int kprobe_test_flags;
+int kprobe_test_cc_position;
+
+static int test_try_count;
+static int test_pass_count;
+static int test_fail_count;
+
+static struct pt_regs initial_regs;
+static struct pt_regs expected_regs;
+static struct pt_regs result_regs;
+
+static u32 expected_memory[TEST_MEMORY_SIZE/sizeof(u32)];
+
+static const char *current_title;
+static struct test_arg *current_args;
+static u32 *current_stack;
+static uintptr_t current_branch_target;
+
+static uintptr_t current_code_start;
+static kprobe_opcode_t current_instruction;
+
+
+#define TEST_CASE_PASSED -1
+#define TEST_CASE_FAILED -2
+
+static int test_case_run_count;
+static bool test_case_is_thumb;
+static int test_instance;
+
+static unsigned long test_check_cc(int cc, unsigned long cpsr)
+{
+	int ret = arm_check_condition(cc << 28, cpsr);
+
+	return (ret != ARM_OPCODE_CONDTEST_FAIL);
+}
+
+static int is_last_scenario;
+static int probe_should_run; /* 0 = no, 1 = yes, -1 = unknown */
+static int memory_needs_checking;
+
+static unsigned long test_context_cpsr(int scenario)
+{
+	unsigned long cpsr;
+
+	probe_should_run = 1;
+
+	/* Default case is that we cycle through 16 combinations of flags */
+	cpsr  = (scenario & 0xf) << 28; /* N,Z,C,V flags */
+	cpsr |= (scenario & 0xf) << 16; /* GE flags */
+	cpsr |= (scenario & 0x1) << 27; /* Toggle Q flag */
+
+	if (!test_case_is_thumb) {
+		/* Testing ARM code */
+		int cc = current_instruction >> 28;
+
+		probe_should_run = test_check_cc(cc, cpsr) != 0;
+		if (scenario == 15)
+			is_last_scenario = true;
+
+	} else if (kprobe_test_flags & TEST_FLAG_NO_ITBLOCK) {
+		/* Testing Thumb code without setting ITSTATE */
+		if (kprobe_test_cc_position) {
+			int cc = (current_instruction >> kprobe_test_cc_position) & 0xf;
+			probe_should_run = test_check_cc(cc, cpsr) != 0;
+		}
+
+		if (scenario == 15)
+			is_last_scenario = true;
+
+	} else if (kprobe_test_flags & TEST_FLAG_FULL_ITBLOCK) {
+		/* Testing Thumb code with all combinations of ITSTATE */
+		unsigned x = (scenario >> 4);
+		unsigned cond_base = x % 7; /* ITSTATE<7:5> */
+		unsigned mask = x / 7 + 2;  /* ITSTATE<4:0>, bits reversed */
+
+		if (mask > 0x1f) {
+			/* Finish by testing state from instruction 'itt al' */
+			cond_base = 7;
+			mask = 0x4;
+			if ((scenario & 0xf) == 0xf)
+				is_last_scenario = true;
+		}
+
+		cpsr |= cond_base << 13;	/* ITSTATE<7:5> */
+		cpsr |= (mask & 0x1) << 12;	/* ITSTATE<4> */
+		cpsr |= (mask & 0x2) << 10;	/* ITSTATE<3> */
+		cpsr |= (mask & 0x4) << 8;	/* ITSTATE<2> */
+		cpsr |= (mask & 0x8) << 23;	/* ITSTATE<1> */
+		cpsr |= (mask & 0x10) << 21;	/* ITSTATE<0> */
+
+		probe_should_run = test_check_cc((cpsr >> 12) & 0xf, cpsr) != 0;
+
+	} else {
+		/* Testing Thumb code with several combinations of ITSTATE */
+		switch (scenario) {
+		case 16: /* Clear NZCV flags and 'it eq' state (false as Z=0) */
+			cpsr = 0x00000800;
+			probe_should_run = 0;
+			break;
+		case 17: /* Set NZCV flags and 'it vc' state (false as V=1) */
+			cpsr = 0xf0007800;
+			probe_should_run = 0;
+			break;
+		case 18: /* Clear NZCV flags and 'it ls' state (true as C=0) */
+			cpsr = 0x00009800;
+			break;
+		case 19: /* Set NZCV flags and 'it cs' state (true as C=1) */
+			cpsr = 0xf0002800;
+			is_last_scenario = true;
+			break;
+		}
+	}
+
+	return cpsr;
+}
+
+static void setup_test_context(struct pt_regs *regs)
+{
+	int scenario = test_case_run_count>>1;
+	unsigned long val;
+	struct test_arg *args;
+	int i;
+
+	is_last_scenario = false;
+	memory_needs_checking = false;
+
+	/* Initialise test memory on stack */
+	val = (scenario & 1) ? VALM : ~VALM;
+	for (i = 0; i < TEST_MEMORY_SIZE / sizeof(current_stack[0]); ++i)
+		current_stack[i] = val + (i << 8);
+	/* Put target of branch on stack for tests which load PC from memory */
+	if (current_branch_target)
+		current_stack[15] = current_branch_target;
+	/* Put a value for SP on stack for tests which load SP from memory */
+	current_stack[13] = (u32)current_stack + 120;
+
+	/* Initialise register values to their default state */
+	val = (scenario & 2) ? VALR : ~VALR;
+	for (i = 0; i < 13; ++i)
+		regs->uregs[i] = val ^ (i << 8);
+	regs->ARM_lr = val ^ (14 << 8);
+	regs->ARM_cpsr &= ~(APSR_MASK | PSR_IT_MASK);
+	regs->ARM_cpsr |= test_context_cpsr(scenario);
+
+	/* Perform testcase specific register setup  */
+	args = current_args;
+	for (; args[0].type != ARG_TYPE_END; ++args)
+		switch (args[0].type) {
+		case ARG_TYPE_REG: {
+			struct test_arg_regptr *arg =
+				(struct test_arg_regptr *)args;
+			regs->uregs[arg->reg] = arg->val;
+			break;
+		}
+		case ARG_TYPE_PTR: {
+			struct test_arg_regptr *arg =
+				(struct test_arg_regptr *)args;
+			regs->uregs[arg->reg] =
+				(unsigned long)current_stack + arg->val;
+			memory_needs_checking = true;
+			/*
+			 * Test memory at an address below SP is in danger of
+			 * being altered by an interrupt occurring and pushing
+			 * data onto the stack. Disable interrupts to stop this.
+			 */
+			if (arg->reg == 13)
+				regs->ARM_cpsr |= PSR_I_BIT;
+			break;
+		}
+		case ARG_TYPE_MEM: {
+			struct test_arg_mem *arg = (struct test_arg_mem *)args;
+			current_stack[arg->index] = arg->val;
+			break;
+		}
+		default:
+			break;
+		}
+}
+
+struct test_probe {
+	struct kprobe	kprobe;
+	bool		registered;
+	int		hit;
+};
+
+static void unregister_test_probe(struct test_probe *probe)
+{
+	if (probe->registered) {
+		unregister_kprobe(&probe->kprobe);
+		probe->kprobe.flags = 0; /* Clear disable flag to allow reuse */
+	}
+	probe->registered = false;
+}
+
+static int register_test_probe(struct test_probe *probe)
+{
+	int ret;
+
+	if (probe->registered)
+		BUG();
+
+	ret = register_kprobe(&probe->kprobe);
+	if (ret >= 0) {
+		probe->registered = true;
+		probe->hit = -1;
+	}
+	return ret;
+}
+
+static int __kprobes
+test_before_pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	container_of(p, struct test_probe, kprobe)->hit = test_instance;
+	return 0;
+}
+
+static void __kprobes
+test_before_post_handler(struct kprobe *p, struct pt_regs *regs,
+							unsigned long flags)
+{
+	setup_test_context(regs);
+	initial_regs = *regs;
+	initial_regs.ARM_cpsr &= ~PSR_IGNORE_BITS;
+}
+
+static int __kprobes
+test_case_pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	container_of(p, struct test_probe, kprobe)->hit = test_instance;
+	return 0;
+}
+
+static int __kprobes
+test_after_pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct test_arg *args;
+
+	if (container_of(p, struct test_probe, kprobe)->hit == test_instance)
+		return 0; /* Already run for this test instance */
+
+	result_regs = *regs;
+
+	/* Mask out results which are indeterminate */
+	result_regs.ARM_cpsr &= ~PSR_IGNORE_BITS;
+	for (args = current_args; args[0].type != ARG_TYPE_END; ++args)
+		if (args[0].type == ARG_TYPE_REG_MASKED) {
+			struct test_arg_regptr *arg =
+				(struct test_arg_regptr *)args;
+			result_regs.uregs[arg->reg] &= arg->val;
+		}
+
+	/* Undo any changes done to SP by the test case */
+	regs->ARM_sp = (unsigned long)current_stack;
+	/* Enable interrupts in case setup_test_context disabled them */
+	regs->ARM_cpsr &= ~PSR_I_BIT;
+
+	container_of(p, struct test_probe, kprobe)->hit = test_instance;
+	return 0;
+}
+
+static struct test_probe test_before_probe = {
+	.kprobe.pre_handler	= test_before_pre_handler,
+	.kprobe.post_handler	= test_before_post_handler,
+};
+
+static struct test_probe test_case_probe = {
+	.kprobe.pre_handler	= test_case_pre_handler,
+};
+
+static struct test_probe test_after_probe = {
+	.kprobe.pre_handler	= test_after_pre_handler,
+};
+
+static struct test_probe test_after2_probe = {
+	.kprobe.pre_handler	= test_after_pre_handler,
+};
+
+static void test_case_cleanup(void)
+{
+	unregister_test_probe(&test_before_probe);
+	unregister_test_probe(&test_case_probe);
+	unregister_test_probe(&test_after_probe);
+	unregister_test_probe(&test_after2_probe);
+}
+
+static void print_registers(struct pt_regs *regs)
+{
+	pr_err("r0  %08lx | r1  %08lx | r2  %08lx | r3  %08lx\n",
+		regs->ARM_r0, regs->ARM_r1, regs->ARM_r2, regs->ARM_r3);
+	pr_err("r4  %08lx | r5  %08lx | r6  %08lx | r7  %08lx\n",
+		regs->ARM_r4, regs->ARM_r5, regs->ARM_r6, regs->ARM_r7);
+	pr_err("r8  %08lx | r9  %08lx | r10 %08lx | r11 %08lx\n",
+		regs->ARM_r8, regs->ARM_r9, regs->ARM_r10, regs->ARM_fp);
+	pr_err("r12 %08lx | sp  %08lx | lr  %08lx | pc  %08lx\n",
+		regs->ARM_ip, regs->ARM_sp, regs->ARM_lr, regs->ARM_pc);
+	pr_err("cpsr %08lx\n", regs->ARM_cpsr);
+}
+
+static void print_memory(u32 *mem, size_t size)
+{
+	int i;
+	for (i = 0; i < size / sizeof(u32); i += 4)
+		pr_err("%08x %08x %08x %08x\n", mem[i], mem[i+1],
+						mem[i+2], mem[i+3]);
+}
+
+static size_t expected_memory_size(u32 *sp)
+{
+	size_t size = sizeof(expected_memory);
+	int offset = (uintptr_t)sp - (uintptr_t)current_stack;
+	if (offset > 0)
+		size -= offset;
+	return size;
+}
+
+static void test_case_failed(const char *message)
+{
+	test_case_cleanup();
+
+	pr_err("FAIL: %s\n", message);
+	pr_err("FAIL: Test %s\n", current_title);
+	pr_err("FAIL: Scenario %d\n", test_case_run_count >> 1);
+}
+
+static unsigned long next_instruction(unsigned long pc)
+{
+#ifdef CONFIG_THUMB2_KERNEL
+	if ((pc & 1) &&
+	    !is_wide_instruction(__mem_to_opcode_thumb16(*(u16 *)(pc - 1))))
+		return pc + 2;
+	else
+#endif
+	return pc + 4;
+}
+
+static uintptr_t __used kprobes_test_case_start(const char **title, void *stack)
+{
+	struct test_arg *args;
+	struct test_arg_end *end_arg;
+	unsigned long test_code;
+
+	current_title = *title++;
+	args = (struct test_arg *)title;
+	current_args = args;
+	current_stack = stack;
+
+	++test_try_count;
+
+	while (args->type != ARG_TYPE_END)
+		++args;
+	end_arg = (struct test_arg_end *)args;
+
+	test_code = (unsigned long)(args + 1); /* Code starts after args */
+
+	test_case_is_thumb = end_arg->flags & ARG_FLAG_THUMB;
+	if (test_case_is_thumb)
+		test_code |= 1;
+
+	current_code_start = test_code;
+
+	current_branch_target = 0;
+	if (end_arg->branch_offset != end_arg->end_offset)
+		current_branch_target = test_code + end_arg->branch_offset;
+
+	test_code += end_arg->code_offset;
+	test_before_probe.kprobe.addr = (kprobe_opcode_t *)test_code;
+
+	test_code = next_instruction(test_code);
+	test_case_probe.kprobe.addr = (kprobe_opcode_t *)test_code;
+
+	if (test_case_is_thumb) {
+		u16 *p = (u16 *)(test_code & ~1);
+		current_instruction = __mem_to_opcode_thumb16(p[0]);
+		if (is_wide_instruction(current_instruction)) {
+			u16 instr2 = __mem_to_opcode_thumb16(p[1]);
+			current_instruction = __opcode_thumb32_compose(current_instruction, instr2);
+		}
+	} else {
+		current_instruction = __mem_to_opcode_arm(*(u32 *)test_code);
+	}
+
+	if (current_title[0] == '.')
+		verbose("%s\n", current_title);
+	else
+		verbose("%s\t@ %0*x\n", current_title,
+					test_case_is_thumb ? 4 : 8,
+					current_instruction);
+
+	test_code = next_instruction(test_code);
+	test_after_probe.kprobe.addr = (kprobe_opcode_t *)test_code;
+
+	if (kprobe_test_flags & TEST_FLAG_NARROW_INSTR) {
+		if (!test_case_is_thumb ||
+			is_wide_instruction(current_instruction)) {
+				test_case_failed("expected 16-bit instruction");
+				goto fail;
+		}
+	} else {
+		if (test_case_is_thumb &&
+			!is_wide_instruction(current_instruction)) {
+				test_case_failed("expected 32-bit instruction");
+				goto fail;
+		}
+	}
+
+	coverage_add(current_instruction);
+
+	if (end_arg->flags & ARG_FLAG_UNSUPPORTED) {
+		if (register_test_probe(&test_case_probe) < 0)
+			goto pass;
+		test_case_failed("registered probe for unsupported instruction");
+		goto fail;
+	}
+
+	if (end_arg->flags & ARG_FLAG_SUPPORTED) {
+		if (register_test_probe(&test_case_probe) >= 0)
+			goto pass;
+		test_case_failed("couldn't register probe for supported instruction");
+		goto fail;
+	}
+
+	if (register_test_probe(&test_before_probe) < 0) {
+		test_case_failed("register test_before_probe failed");
+		goto fail;
+	}
+	if (register_test_probe(&test_after_probe) < 0) {
+		test_case_failed("register test_after_probe failed");
+		goto fail;
+	}
+	if (current_branch_target) {
+		test_after2_probe.kprobe.addr =
+				(kprobe_opcode_t *)current_branch_target;
+		if (register_test_probe(&test_after2_probe) < 0) {
+			test_case_failed("register test_after2_probe failed");
+			goto fail;
+		}
+	}
+
+	/* Start first run of test case */
+	test_case_run_count = 0;
+	++test_instance;
+	return current_code_start;
+pass:
+	test_case_run_count = TEST_CASE_PASSED;
+	return (uintptr_t)test_after_probe.kprobe.addr;
+fail:
+	test_case_run_count = TEST_CASE_FAILED;
+	return (uintptr_t)test_after_probe.kprobe.addr;
+}
+
+static bool check_test_results(void)
+{
+	size_t mem_size = 0;
+	u32 *mem = 0;
+
+	if (memcmp(&expected_regs, &result_regs, sizeof(expected_regs))) {
+		test_case_failed("registers differ");
+		goto fail;
+	}
+
+	if (memory_needs_checking) {
+		mem = (u32 *)result_regs.ARM_sp;
+		mem_size = expected_memory_size(mem);
+		if (memcmp(expected_memory, mem, mem_size)) {
+			test_case_failed("test memory differs");
+			goto fail;
+		}
+	}
+
+	return true;
+
+fail:
+	pr_err("initial_regs:\n");
+	print_registers(&initial_regs);
+	pr_err("expected_regs:\n");
+	print_registers(&expected_regs);
+	pr_err("result_regs:\n");
+	print_registers(&result_regs);
+
+	if (mem) {
+		pr_err("current_stack=%p\n", current_stack);
+		pr_err("expected_memory:\n");
+		print_memory(expected_memory, mem_size);
+		pr_err("result_memory:\n");
+		print_memory(mem, mem_size);
+	}
+
+	return false;
+}
+
+static uintptr_t __used kprobes_test_case_end(void)
+{
+	if (test_case_run_count < 0) {
+		if (test_case_run_count == TEST_CASE_PASSED)
+			/* kprobes_test_case_start did all the needed testing */
+			goto pass;
+		else
+			/* kprobes_test_case_start failed */
+			goto fail;
+	}
+
+	if (test_before_probe.hit != test_instance) {
+		test_case_failed("test_before_handler not run");
+		goto fail;
+	}
+
+	if (test_after_probe.hit != test_instance &&
+				test_after2_probe.hit != test_instance) {
+		test_case_failed("test_after_handler not run");
+		goto fail;
+	}
+
+	/*
+	 * Even numbered test runs ran without a probe on the test case so
+	 * we can gather reference results. The subsequent odd numbered run
+	 * will have the probe inserted.
+	*/
+	if ((test_case_run_count & 1) == 0) {
+		/* Save results from run without probe */
+		u32 *mem = (u32 *)result_regs.ARM_sp;
+		expected_regs = result_regs;
+		memcpy(expected_memory, mem, expected_memory_size(mem));
+
+		/* Insert probe onto test case instruction */
+		if (register_test_probe(&test_case_probe) < 0) {
+			test_case_failed("register test_case_probe failed");
+			goto fail;
+		}
+	} else {
+		/* Check probe ran as expected */
+		if (probe_should_run == 1) {
+			if (test_case_probe.hit != test_instance) {
+				test_case_failed("test_case_handler not run");
+				goto fail;
+			}
+		} else if (probe_should_run == 0) {
+			if (test_case_probe.hit == test_instance) {
+				test_case_failed("test_case_handler ran");
+				goto fail;
+			}
+		}
+
+		/* Remove probe for any subsequent reference run */
+		unregister_test_probe(&test_case_probe);
+
+		if (!check_test_results())
+			goto fail;
+
+		if (is_last_scenario)
+			goto pass;
+	}
+
+	/* Do next test run */
+	++test_case_run_count;
+	++test_instance;
+	return current_code_start;
+fail:
+	++test_fail_count;
+	goto end;
+pass:
+	++test_pass_count;
+end:
+	test_case_cleanup();
+	return 0;
+}
+
+
+/*
+ * Top level test functions
+ */
+
+static int run_test_cases(void (*tests)(void), const union decode_item *table)
+{
+	int ret;
+
+	pr_info("    Check decoding tables\n");
+	ret = table_test(table);
+	if (ret)
+		return ret;
+
+	pr_info("    Run test cases\n");
+	ret = coverage_start(table);
+	if (ret)
+		return ret;
+
+	tests();
+
+	coverage_end();
+	return 0;
+}
+
+
+static int __init run_all_tests(void)
+{
+	int ret = 0;
+
+	pr_info("Beginning kprobe tests...\n");
+
+#ifndef CONFIG_THUMB2_KERNEL
+
+	pr_info("Probe ARM code\n");
+	ret = run_api_tests(arm_func);
+	if (ret)
+		goto out;
+
+	pr_info("ARM instruction simulation\n");
+	ret = run_test_cases(kprobe_arm_test_cases, probes_decode_arm_table);
+	if (ret)
+		goto out;
+
+#else /* CONFIG_THUMB2_KERNEL */
+
+	pr_info("Probe 16-bit Thumb code\n");
+	ret = run_api_tests(thumb16_func);
+	if (ret)
+		goto out;
+
+	pr_info("Probe 32-bit Thumb code, even halfword\n");
+	ret = run_api_tests(thumb32even_func);
+	if (ret)
+		goto out;
+
+	pr_info("Probe 32-bit Thumb code, odd halfword\n");
+	ret = run_api_tests(thumb32odd_func);
+	if (ret)
+		goto out;
+
+	pr_info("16-bit Thumb instruction simulation\n");
+	ret = run_test_cases(kprobe_thumb16_test_cases,
+				probes_decode_thumb16_table);
+	if (ret)
+		goto out;
+
+	pr_info("32-bit Thumb instruction simulation\n");
+	ret = run_test_cases(kprobe_thumb32_test_cases,
+				probes_decode_thumb32_table);
+	if (ret)
+		goto out;
+#endif
+
+	pr_info("Total instruction simulation tests=%d, pass=%d fail=%d\n",
+		test_try_count, test_pass_count, test_fail_count);
+	if (test_fail_count) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+#if BENCHMARKING
+	pr_info("Benchmarks\n");
+	ret = run_benchmarks();
+	if (ret)
+		goto out;
+#endif
+
+#if __LINUX_ARM_ARCH__ >= 7
+	/* We are able to run all test cases so coverage should be complete */
+	if (coverage_fail) {
+		pr_err("FAIL: Test coverage checks failed\n");
+		ret = -EINVAL;
+		goto out;
+	}
+#endif
+
+out:
+	if (ret == 0)
+		ret = tests_failed;
+	if (ret == 0)
+		pr_info("Finished kprobe tests OK\n");
+	else
+		pr_err("kprobe tests failed\n");
+
+	return ret;
+}
+
+
+/*
+ * Module setup
+ */
+
+#ifdef MODULE
+
+static void __exit kprobe_test_exit(void)
+{
+}
+
+module_init(run_all_tests)
+module_exit(kprobe_test_exit)
+MODULE_LICENSE("GPL");
+
+#else /* !MODULE */
+
+late_initcall(run_all_tests);
+
+#endif

arch/arm/probes/kprobes/test-core.h

@@ -0,0 +1,458 @@
+/*
+ * arch/arm/probes/kprobes/test-core.h
+ *
+ * Copyright (C) 2011 Jon Medhurst <tixy@yxit.co.uk>.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#define VERBOSE 0 /* Set to '1' for more logging of test cases */
+
+#ifdef CONFIG_THUMB2_KERNEL
+#define NORMAL_ISA "16"
+#else
+#define NORMAL_ISA "32"
+#endif
+
+
+/* Flags used in kprobe_test_flags */
+#define TEST_FLAG_NO_ITBLOCK	(1<<0)
+#define TEST_FLAG_FULL_ITBLOCK	(1<<1)
+#define TEST_FLAG_NARROW_INSTR	(1<<2)
+
+extern int kprobe_test_flags;
+extern int kprobe_test_cc_position;
+
+
+#define TEST_MEMORY_SIZE 256
+
+
+/*
+ * Test case structures.
+ *
+ * The arguments given to test cases can be one of three types.
+ *
+ *   ARG_TYPE_REG
+ *	Load a register with the given value.
+ *
+ *   ARG_TYPE_PTR
+ *	Load a register with a pointer into the stack buffer (SP + given value).
+ *
+ *   ARG_TYPE_MEM
+ *	Store the given value into the stack buffer at [SP+index].
+ *
+ */
+
+#define	ARG_TYPE_END		0
+#define	ARG_TYPE_REG		1
+#define	ARG_TYPE_PTR		2
+#define	ARG_TYPE_MEM		3
+#define	ARG_TYPE_REG_MASKED	4
+
+#define ARG_FLAG_UNSUPPORTED	0x01
+#define ARG_FLAG_SUPPORTED	0x02
+#define ARG_FLAG_THUMB		0x10	/* Must be 16 so TEST_ISA can be used */
+#define ARG_FLAG_ARM		0x20	/* Must be 32 so TEST_ISA can be used */
+
+struct test_arg {
+	u8	type;		/* ARG_TYPE_x */
+	u8	_padding[7];
+};
+
+struct test_arg_regptr {
+	u8	type;		/* ARG_TYPE_REG or ARG_TYPE_PTR or ARG_TYPE_REG_MASKED */
+	u8	reg;
+	u8	_padding[2];
+	u32	val;
+};
+
+struct test_arg_mem {
+	u8	type;		/* ARG_TYPE_MEM */
+	u8	index;
+	u8	_padding[2];
+	u32	val;
+};
+
+struct test_arg_end {
+	u8	type;		/* ARG_TYPE_END */
+	u8	flags;		/* ARG_FLAG_x */
+	u16	code_offset;
+	u16	branch_offset;
+	u16	end_offset;
+};
+
+
+/*
+ * Building blocks for test cases.
+ *
+ * Each test case is wrapped between TESTCASE_START and TESTCASE_END.
+ *
+ * To specify arguments for a test case the TEST_ARG_{REG,PTR,MEM} macros are
+ * used followed by a terminating TEST_ARG_END.
+ *
+ * After this, the instruction to be tested is defined with TEST_INSTRUCTION.
+ * Or for branches, TEST_BRANCH_B and TEST_BRANCH_F (branch forwards/backwards).
+ *
+ * Some specific test cases may make use of other custom constructs.
+ */
+
+#if VERBOSE
+#define verbose(fmt, ...) pr_info(fmt, ##__VA_ARGS__)
+#else
+#define verbose(fmt, ...)
+#endif
+
+#define TEST_GROUP(title)					\
+	verbose("\n");						\
+	verbose(title"\n");					\
+	verbose("---------------------------------------------------------\n");
+
+#define TESTCASE_START(title)					\
+	__asm__ __volatile__ (					\
+	"bl	__kprobes_test_case_start		\n\t"	\
+	".pushsection .rodata				\n\t"	\
+	"10:						\n\t"	\
+	/* don't use .asciz here as 'title' may be */		\
+	/* multiple strings to be concatenated.  */		\
+	".ascii "#title"				\n\t"	\
+	".byte	0					\n\t"	\
+	".popsection					\n\t"	\
+	".word	10b					\n\t"
+
+#define	TEST_ARG_REG(reg, val)					\
+	".byte	"__stringify(ARG_TYPE_REG)"		\n\t"	\
+	".byte	"#reg"					\n\t"	\
+	".short	0					\n\t"	\
+	".word	"#val"					\n\t"
+
+#define	TEST_ARG_PTR(reg, val)					\
+	".byte	"__stringify(ARG_TYPE_PTR)"		\n\t"	\
+	".byte	"#reg"					\n\t"	\
+	".short	0					\n\t"	\
+	".word	"#val"					\n\t"
+
+#define	TEST_ARG_MEM(index, val)				\
+	".byte	"__stringify(ARG_TYPE_MEM)"		\n\t"	\
+	".byte	"#index"				\n\t"	\
+	".short	0					\n\t"	\
+	".word	"#val"					\n\t"
+
+#define	TEST_ARG_REG_MASKED(reg, val)				\
+	".byte	"__stringify(ARG_TYPE_REG_MASKED)"	\n\t"	\
+	".byte	"#reg"					\n\t"	\
+	".short	0					\n\t"	\
+	".word	"#val"					\n\t"
+
+#define	TEST_ARG_END(flags)					\
+	".byte	"__stringify(ARG_TYPE_END)"		\n\t"	\
+	".byte	"TEST_ISA flags"			\n\t"	\
+	".short	50f-0f					\n\t"	\
+	".short	2f-0f					\n\t"	\
+	".short	99f-0f					\n\t"	\
+	".code "TEST_ISA"				\n\t"	\
+	"0:						\n\t"
+
+#define TEST_INSTRUCTION(instruction)				\
+	"50:	nop					\n\t"	\
+	"1:	"instruction"				\n\t"	\
+	"	nop					\n\t"
+
+#define TEST_BRANCH_F(instruction)				\
+	TEST_INSTRUCTION(instruction)				\
+	"	b	99f				\n\t"	\
+	"2:	nop					\n\t"
+
+#define TEST_BRANCH_B(instruction)				\
+	"	b	50f				\n\t"	\
+	"	b	99f				\n\t"	\
+	"2:	nop					\n\t"	\
+	"	b	99f				\n\t"	\
+	TEST_INSTRUCTION(instruction)
+
+#define TEST_BRANCH_FX(instruction, codex)			\
+	TEST_INSTRUCTION(instruction)				\
+	"	b	99f				\n\t"	\
+	codex"						\n\t"	\
+	"	b	99f				\n\t"	\
+	"2:	nop					\n\t"
+
+#define TEST_BRANCH_BX(instruction, codex)			\
+	"	b	50f				\n\t"	\
+	"	b	99f				\n\t"	\
+	"2:	nop					\n\t"	\
+	"	b	99f				\n\t"	\
+	codex"						\n\t"	\
+	TEST_INSTRUCTION(instruction)
+
+#define TESTCASE_END						\
+	"2:						\n\t"	\
+	"99:						\n\t"	\
+	"	bl __kprobes_test_case_end_"TEST_ISA"	\n\t"	\
+	".code "NORMAL_ISA"				\n\t"	\
+	: :							\
+	: "r0", "r1", "r2", "r3", "ip", "lr", "memory", "cc"	\
+	);
+
+
+/*
+ * Macros to define test cases.
+ *
+ * Those of the form TEST_{R,P,M}* can be used to define test cases
+ * which take combinations of the three basic types of arguments. E.g.
+ *
+ *   TEST_R	One register argument
+ *   TEST_RR	Two register arguments
+ *   TEST_RPR	A register, a pointer, then a register argument
+ *
+ * For testing instructions which may branch, there are macros TEST_BF_*
+ * and TEST_BB_* for branching forwards and backwards.
+ *
+ * TEST_SUPPORTED and TEST_UNSUPPORTED don't cause the code to be executed,
+ * the just verify that a kprobe is or is not allowed on the given instruction.
+ */
+
+#define TEST(code)				\
+	TESTCASE_START(code)			\
+	TEST_ARG_END("")			\
+	TEST_INSTRUCTION(code)			\
+	TESTCASE_END
+
+#define TEST_UNSUPPORTED(code)					\
+	TESTCASE_START(code)					\
+	TEST_ARG_END("|"__stringify(ARG_FLAG_UNSUPPORTED))	\
+	TEST_INSTRUCTION(code)					\
+	TESTCASE_END
+
+#define TEST_SUPPORTED(code)					\
+	TESTCASE_START(code)					\
+	TEST_ARG_END("|"__stringify(ARG_FLAG_SUPPORTED))	\
+	TEST_INSTRUCTION(code)					\
+	TESTCASE_END
+
+#define TEST_R(code1, reg, val, code2)			\
+	TESTCASE_START(code1 #reg code2)		\
+	TEST_ARG_REG(reg, val)				\
+	TEST_ARG_END("")				\
+	TEST_INSTRUCTION(code1 #reg code2)		\
+	TESTCASE_END
+
+#define TEST_RR(code1, reg1, val1, code2, reg2, val2, code3)	\
+	TESTCASE_START(code1 #reg1 code2 #reg2 code3)		\
+	TEST_ARG_REG(reg1, val1)				\
+	TEST_ARG_REG(reg2, val2)				\
+	TEST_ARG_END("")					\
+	TEST_INSTRUCTION(code1 #reg1 code2 #reg2 code3)		\
+	TESTCASE_END
+
+#define TEST_RRR(code1, reg1, val1, code2, reg2, val2, code3, reg3, val3, code4)\
+	TESTCASE_START(code1 #reg1 code2 #reg2 code3 #reg3 code4)		\
+	TEST_ARG_REG(reg1, val1)						\
+	TEST_ARG_REG(reg2, val2)						\
+	TEST_ARG_REG(reg3, val3)						\
+	TEST_ARG_END("")							\
+	TEST_INSTRUCTION(code1 #reg1 code2 #reg2 code3 #reg3 code4)		\
+	TESTCASE_END
+
+#define TEST_RRRR(code1, reg1, val1, code2, reg2, val2, code3, reg3, val3, code4, reg4, val4)	\
+	TESTCASE_START(code1 #reg1 code2 #reg2 code3 #reg3 code4 #reg4)		\
+	TEST_ARG_REG(reg1, val1)						\
+	TEST_ARG_REG(reg2, val2)						\
+	TEST_ARG_REG(reg3, val3)						\
+	TEST_ARG_REG(reg4, val4)						\
+	TEST_ARG_END("")							\
+	TEST_INSTRUCTION(code1 #reg1 code2 #reg2 code3 #reg3 code4 #reg4)	\
+	TESTCASE_END
+
+#define TEST_P(code1, reg1, val1, code2)	\
+	TESTCASE_START(code1 #reg1 code2)	\
+	TEST_ARG_PTR(reg1, val1)		\
+	TEST_ARG_END("")			\
+	TEST_INSTRUCTION(code1 #reg1 code2)	\
+	TESTCASE_END
+
+#define TEST_PR(code1, reg1, val1, code2, reg2, val2, code3)	\
+	TESTCASE_START(code1 #reg1 code2 #reg2 code3)		\
+	TEST_ARG_PTR(reg1, val1)				\
+	TEST_ARG_REG(reg2, val2)				\
+	TEST_ARG_END("")					\
+	TEST_INSTRUCTION(code1 #reg1 code2 #reg2 code3)		\
+	TESTCASE_END
+
+#define TEST_RP(code1, reg1, val1, code2, reg2, val2, code3)	\
+	TESTCASE_START(code1 #reg1 code2 #reg2 code3)		\
+	TEST_ARG_REG(reg1, val1)				\
+	TEST_ARG_PTR(reg2, val2)				\
+	TEST_ARG_END("")					\
+	TEST_INSTRUCTION(code1 #reg1 code2 #reg2 code3)		\
+	TESTCASE_END
+
+#define TEST_PRR(code1, reg1, val1, code2, reg2, val2, code3, reg3, val3, code4)\
+	TESTCASE_START(code1 #reg1 code2 #reg2 code3 #reg3 code4)		\
+	TEST_ARG_PTR(reg1, val1)						\
+	TEST_ARG_REG(reg2, val2)						\
+	TEST_ARG_REG(reg3, val3)						\
+	TEST_ARG_END("")							\
+	TEST_INSTRUCTION(code1 #reg1 code2 #reg2 code3 #reg3 code4)		\
+	TESTCASE_END
+
+#define TEST_RPR(code1, reg1, val1, code2, reg2, val2, code3, reg3, val3, code4)\
+	TESTCASE_START(code1 #reg1 code2 #reg2 code3 #reg3 code4)		\
+	TEST_ARG_REG(reg1, val1)						\
+	TEST_ARG_PTR(reg2, val2)						\
+	TEST_ARG_REG(reg3, val3)						\
+	TEST_ARG_END("")							\
+	TEST_INSTRUCTION(code1 #reg1 code2 #reg2 code3 #reg3 code4)		\
+	TESTCASE_END
+
+#define TEST_RRP(code1, reg1, val1, code2, reg2, val2, code3, reg3, val3, code4)\
+	TESTCASE_START(code1 #reg1 code2 #reg2 code3 #reg3 code4)		\
+	TEST_ARG_REG(reg1, val1)						\
+	TEST_ARG_REG(reg2, val2)						\
+	TEST_ARG_PTR(reg3, val3)						\
+	TEST_ARG_END("")							\
+	TEST_INSTRUCTION(code1 #reg1 code2 #reg2 code3 #reg3 code4)		\
+	TESTCASE_END
+
+#define TEST_BF_P(code1, reg1, val1, code2)	\
+	TESTCASE_START(code1 #reg1 code2)	\
+	TEST_ARG_PTR(reg1, val1)		\
+	TEST_ARG_END("")			\
+	TEST_BRANCH_F(code1 #reg1 code2)	\
+	TESTCASE_END
+
+#define TEST_BF(code)				\
+	TESTCASE_START(code)			\
+	TEST_ARG_END("")			\
+	TEST_BRANCH_F(code)			\
+	TESTCASE_END
+
+#define TEST_BB(code)				\
+	TESTCASE_START(code)			\
+	TEST_ARG_END("")			\
+	TEST_BRANCH_B(code)			\
+	TESTCASE_END
+
+#define TEST_BF_R(code1, reg, val, code2)	\
+	TESTCASE_START(code1 #reg code2)	\
+	TEST_ARG_REG(reg, val)			\
+	TEST_ARG_END("")			\
+	TEST_BRANCH_F(code1 #reg code2)		\
+	TESTCASE_END
+
+#define TEST_BB_R(code1, reg, val, code2)	\
+	TESTCASE_START(code1 #reg code2)	\
+	TEST_ARG_REG(reg, val)			\
+	TEST_ARG_END("")			\
+	TEST_BRANCH_B(code1 #reg code2)		\
+	TESTCASE_END
+
+#define TEST_BF_RR(code1, reg1, val1, code2, reg2, val2, code3)	\
+	TESTCASE_START(code1 #reg1 code2 #reg2 code3)		\
+	TEST_ARG_REG(reg1, val1)				\
+	TEST_ARG_REG(reg2, val2)				\
+	TEST_ARG_END("")					\
+	TEST_BRANCH_F(code1 #reg1 code2 #reg2 code3)		\
+	TESTCASE_END
+
+#define TEST_BF_X(code, codex)			\
+	TESTCASE_START(code)			\
+	TEST_ARG_END("")			\
+	TEST_BRANCH_FX(code, codex)		\
+	TESTCASE_END
+
+#define TEST_BB_X(code, codex)			\
+	TESTCASE_START(code)			\
+	TEST_ARG_END("")			\
+	TEST_BRANCH_BX(code, codex)		\
+	TESTCASE_END
+
+#define TEST_BF_RX(code1, reg, val, code2, codex)	\
+	TESTCASE_START(code1 #reg code2)		\
+	TEST_ARG_REG(reg, val)				\
+	TEST_ARG_END("")				\
+	TEST_BRANCH_FX(code1 #reg code2, codex)		\
+	TESTCASE_END
+
+#define TEST_X(code, codex)			\
+	TESTCASE_START(code)			\
+	TEST_ARG_END("")			\
+	TEST_INSTRUCTION(code)			\
+	"	b	99f		\n\t"	\
+	"	"codex"			\n\t"	\
+	TESTCASE_END
+
+#define TEST_RX(code1, reg, val, code2, codex)		\
+	TESTCASE_START(code1 #reg code2)		\
+	TEST_ARG_REG(reg, val)				\
+	TEST_ARG_END("")				\
+	TEST_INSTRUCTION(code1 __stringify(reg) code2)	\
+	"	b	99f		\n\t"		\
+	"	"codex"			\n\t"		\
+	TESTCASE_END
+
+#define TEST_RRX(code1, reg1, val1, code2, reg2, val2, code3, codex)		\
+	TESTCASE_START(code1 #reg1 code2 #reg2 code3)				\
+	TEST_ARG_REG(reg1, val1)						\
+	TEST_ARG_REG(reg2, val2)						\
+	TEST_ARG_END("")							\
+	TEST_INSTRUCTION(code1 __stringify(reg1) code2 __stringify(reg2) code3)	\
+	"	b	99f		\n\t"					\
+	"	"codex"			\n\t"					\
+	TESTCASE_END
+
+#define TEST_RMASKED(code1, reg, mask, code2)		\
+	TESTCASE_START(code1 #reg code2)		\
+	TEST_ARG_REG_MASKED(reg, mask)			\
+	TEST_ARG_END("")				\
+	TEST_INSTRUCTION(code1 #reg code2)		\
+	TESTCASE_END
+
+/*
+ * We ignore the state of the imprecise abort disable flag (CPSR.A) because this
+ * can change randomly as the kernel doesn't take care to preserve or initialise
+ * this across context switches. Also, with Security Extensions, the flag may
+ * not be under control of the kernel; for this reason we ignore the state of
+ * the FIQ disable flag CPSR.F as well.
+ */
+#define PSR_IGNORE_BITS (PSR_A_BIT | PSR_F_BIT)
+
+
+/*
+ * Macros for defining space directives spread over multiple lines.
+ * These are required so the compiler guesses better the length of inline asm
+ * code and will spill the literal pool early enough to avoid generating PC
+ * relative loads with out of range offsets.
+ */
+#define TWICE(x)	x x
+#define SPACE_0x8	TWICE(".space 4\n\t")
+#define SPACE_0x10	TWICE(SPACE_0x8)
+#define SPACE_0x20	TWICE(SPACE_0x10)
+#define SPACE_0x40	TWICE(SPACE_0x20)
+#define SPACE_0x80	TWICE(SPACE_0x40)
+#define SPACE_0x100	TWICE(SPACE_0x80)
+#define SPACE_0x200	TWICE(SPACE_0x100)
+#define SPACE_0x400	TWICE(SPACE_0x200)
+#define SPACE_0x800	TWICE(SPACE_0x400)
+#define SPACE_0x1000	TWICE(SPACE_0x800)
+
+
+/* Various values used in test cases... */
+#define N(val)	(val ^ 0xffffffff)
+#define VAL1	0x12345678
+#define VAL2	N(VAL1)
+#define VAL3	0xa5f801
+#define VAL4	N(VAL3)
+#define VALM	0x456789ab
+#define VALR	0xdeaddead
+#define HH1	0x0123fecb
+#define HH2	0xa9874567
+
+
+#ifdef CONFIG_THUMB2_KERNEL
+void kprobe_thumb16_test_cases(void);
+void kprobe_thumb32_test_cases(void);
+#else
+void kprobe_arm_test_cases(void);
+#endif

arch/arm/probes/kprobes/test-thumb.c

@@ -0,0 +1,1200 @@
+/*
+ * arch/arm/probes/kprobes/test-thumb.c
+ *
+ * Copyright (C) 2011 Jon Medhurst <tixy@yxit.co.uk>.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <asm/opcodes.h>
+#include <asm/probes.h>
+
+#include "test-core.h"
+
+
+#define TEST_ISA "16"
+
+#define DONT_TEST_IN_ITBLOCK(tests)			\
+	kprobe_test_flags |= TEST_FLAG_NO_ITBLOCK;	\
+	tests						\
+	kprobe_test_flags &= ~TEST_FLAG_NO_ITBLOCK;
+
+#define CONDITION_INSTRUCTIONS(cc_pos, tests)		\
+	kprobe_test_cc_position = cc_pos;		\
+	DONT_TEST_IN_ITBLOCK(tests)			\
+	kprobe_test_cc_position = 0;
+
+#define TEST_ITBLOCK(code)				\
+	kprobe_test_flags |= TEST_FLAG_FULL_ITBLOCK;	\
+	TESTCASE_START(code)				\
+	TEST_ARG_END("")				\
+	"50:	nop			\n\t"		\
+	"1:	"code"			\n\t"		\
+	"	mov r1, #0x11		\n\t"		\
+	"	mov r2, #0x22		\n\t"		\
+	"	mov r3, #0x33		\n\t"		\
+	"2:	nop			\n\t"		\
+	TESTCASE_END					\
+	kprobe_test_flags &= ~TEST_FLAG_FULL_ITBLOCK;
+
+#define TEST_THUMB_TO_ARM_INTERWORK_P(code1, reg, val, code2)	\
+	TESTCASE_START(code1 #reg code2)			\
+	TEST_ARG_PTR(reg, val)					\
+	TEST_ARG_REG(14, 99f+1)					\
+	TEST_ARG_MEM(15, 3f)					\
+	TEST_ARG_END("")					\
+	"	nop			\n\t" /* To align 1f */	\
+	"50:	nop			\n\t"			\
+	"1:	"code1 #reg code2"	\n\t"			\
+	"	bx	lr		\n\t"			\
+	".arm				\n\t"			\
+	"3:	adr	lr, 2f+1	\n\t"			\
+	"	bx	lr		\n\t"			\
+	".thumb				\n\t"			\
+	"2:	nop			\n\t"			\
+	TESTCASE_END
+
+
+void kprobe_thumb16_test_cases(void)
+{
+	kprobe_test_flags = TEST_FLAG_NARROW_INSTR;
+
+	TEST_GROUP("Shift (immediate), add, subtract, move, and compare")
+
+	TEST_R(    "lsls	r7, r",0,VAL1,", #5")
+	TEST_R(    "lsls	r0, r",7,VAL2,", #11")
+	TEST_R(    "lsrs	r7, r",0,VAL1,", #5")
+	TEST_R(    "lsrs	r0, r",7,VAL2,", #11")
+	TEST_R(    "asrs	r7, r",0,VAL1,", #5")
+	TEST_R(    "asrs	r0, r",7,VAL2,", #11")
+	TEST_RR(   "adds	r2, r",0,VAL1,", r",7,VAL2,"")
+	TEST_RR(   "adds	r5, r",7,VAL2,", r",0,VAL2,"")
+	TEST_RR(   "subs	r2, r",0,VAL1,", r",7,VAL2,"")
+	TEST_RR(   "subs	r5, r",7,VAL2,", r",0,VAL2,"")
+	TEST_R(    "adds	r7, r",0,VAL1,", #5")
+	TEST_R(    "adds	r0, r",7,VAL2,", #2")
+	TEST_R(    "subs	r7, r",0,VAL1,", #5")
+	TEST_R(    "subs	r0, r",7,VAL2,", #2")
+	TEST(      "movs.n	r0, #0x5f")
+	TEST(      "movs.n	r7, #0xa0")
+	TEST_R(    "cmp.n	r",0,0x5e, ", #0x5f")
+	TEST_R(    "cmp.n	r",5,0x15f,", #0x5f")
+	TEST_R(    "cmp.n	r",7,0xa0, ", #0xa0")
+	TEST_R(    "adds.n	r",0,VAL1,", #0x5f")
+	TEST_R(    "adds.n	r",7,VAL2,", #0xa0")
+	TEST_R(    "subs.n	r",0,VAL1,", #0x5f")
+	TEST_R(    "subs.n	r",7,VAL2,", #0xa0")
+
+	TEST_GROUP("16-bit Thumb data-processing instructions")
+
+#define DATA_PROCESSING16(op,val)			\
+	TEST_RR(   op"	r",0,VAL1,", r",7,val,"")	\
+	TEST_RR(   op"	r",7,VAL2,", r",0,val,"")
+
+	DATA_PROCESSING16("ands",0xf00f00ff)
+	DATA_PROCESSING16("eors",0xf00f00ff)
+	DATA_PROCESSING16("lsls",11)
+	DATA_PROCESSING16("lsrs",11)
+	DATA_PROCESSING16("asrs",11)
+	DATA_PROCESSING16("adcs",VAL2)
+	DATA_PROCESSING16("sbcs",VAL2)
+	DATA_PROCESSING16("rors",11)
+	DATA_PROCESSING16("tst",0xf00f00ff)
+	TEST_R("rsbs	r",0,VAL1,", #0")
+	TEST_R("rsbs	r",7,VAL2,", #0")
+	DATA_PROCESSING16("cmp",0xf00f00ff)
+	DATA_PROCESSING16("cmn",0xf00f00ff)
+	DATA_PROCESSING16("orrs",0xf00f00ff)
+	DATA_PROCESSING16("muls",VAL2)
+	DATA_PROCESSING16("bics",0xf00f00ff)
+	DATA_PROCESSING16("mvns",VAL2)
+
+	TEST_GROUP("Special data instructions and branch and exchange")
+
+	TEST_RR(  "add	r",0, VAL1,", r",7,VAL2,"")
+	TEST_RR(  "add	r",3, VAL2,", r",8,VAL3,"")
+	TEST_RR(  "add	r",8, VAL3,", r",0,VAL1,"")
+	TEST_R(   "add	sp"        ", r",8,-8,  "")
+	TEST_R(   "add	r",14,VAL1,", pc")
+	TEST_BF_R("add	pc"        ", r",0,2f-1f-8,"")
+	TEST_UNSUPPORTED(__inst_thumb16(0x44ff) "	@ add pc, pc")
+
+	TEST_RR(  "cmp	r",3,VAL1,", r",8,VAL2,"")
+	TEST_RR(  "cmp	r",8,VAL2,", r",0,VAL1,"")
+	TEST_R(   "cmp	sp"       ", r",8,-8,  "")
+
+	TEST_R(   "mov	r0, r",7,VAL2,"")
+	TEST_R(   "mov	r3, r",8,VAL3,"")
+	TEST_R(   "mov	r8, r",0,VAL1,"")
+	TEST_P(   "mov	sp, r",8,-8,  "")
+	TEST(     "mov	lr, pc")
+	TEST_BF_R("mov	pc, r",0,2f,  "")
+
+	TEST_BF_R("bx	r",0, 2f+1,"")
+	TEST_BF_R("bx	r",14,2f+1,"")
+	TESTCASE_START("bx	pc")
+		TEST_ARG_REG(14, 99f+1)
+		TEST_ARG_END("")
+		"	nop			\n\t" /* To align the bx pc*/
+		"50:	nop			\n\t"
+		"1:	bx	pc		\n\t"
+		"	bx	lr		\n\t"
+		".arm				\n\t"
+		"	adr	lr, 2f+1	\n\t"
+		"	bx	lr		\n\t"
+		".thumb				\n\t"
+		"2:	nop			\n\t"
+	TESTCASE_END
+
+	TEST_BF_R("blx	r",0, 2f+1,"")
+	TEST_BB_R("blx	r",14,2f+1,"")
+	TEST_UNSUPPORTED(__inst_thumb16(0x47f8) "	@ blx pc")
+
+	TEST_GROUP("Load from Literal Pool")
+
+	TEST_X( "ldr	r0, 3f",
+		".align					\n\t"
+		"3:	.word	"__stringify(VAL1))
+	TEST_X( "ldr	r7, 3f",
+		".space 128				\n\t"
+		".align					\n\t"
+		"3:	.word	"__stringify(VAL2))
+
+	TEST_GROUP("16-bit Thumb Load/store instructions")
+
+	TEST_RPR("str	r",0, VAL1,", [r",1, 24,", r",2,  48,"]")
+	TEST_RPR("str	r",7, VAL2,", [r",6, 24,", r",5,  48,"]")
+	TEST_RPR("strh	r",0, VAL1,", [r",1, 24,", r",2,  48,"]")
+	TEST_RPR("strh	r",7, VAL2,", [r",6, 24,", r",5,  48,"]")
+	TEST_RPR("strb	r",0, VAL1,", [r",1, 24,", r",2,  48,"]")
+	TEST_RPR("strb	r",7, VAL2,", [r",6, 24,", r",5,  48,"]")
+	TEST_PR( "ldrsb	r0, [r",1, 24,", r",2,  48,"]")
+	TEST_PR( "ldrsb	r7, [r",6, 24,", r",5,  50,"]")
+	TEST_PR( "ldr	r0, [r",1, 24,", r",2,  48,"]")
+	TEST_PR( "ldr	r7, [r",6, 24,", r",5,  48,"]")
+	TEST_PR( "ldrh	r0, [r",1, 24,", r",2,  48,"]")
+	TEST_PR( "ldrh	r7, [r",6, 24,", r",5,  50,"]")
+	TEST_PR( "ldrb	r0, [r",1, 24,", r",2,  48,"]")
+	TEST_PR( "ldrb	r7, [r",6, 24,", r",5,  50,"]")
+	TEST_PR( "ldrsh	r0, [r",1, 24,", r",2,  48,"]")
+	TEST_PR( "ldrsh	r7, [r",6, 24,", r",5,  50,"]")
+
+	TEST_RP("str	r",0, VAL1,", [r",1, 24,", #120]")
+	TEST_RP("str	r",7, VAL2,", [r",6, 24,", #120]")
+	TEST_P( "ldr	r0, [r",1, 24,", #120]")
+	TEST_P( "ldr	r7, [r",6, 24,", #120]")
+	TEST_RP("strb	r",0, VAL1,", [r",1, 24,", #30]")
+	TEST_RP("strb	r",7, VAL2,", [r",6, 24,", #30]")
+	TEST_P( "ldrb	r0, [r",1, 24,", #30]")
+	TEST_P( "ldrb	r7, [r",6, 24,", #30]")
+	TEST_RP("strh	r",0, VAL1,", [r",1, 24,", #60]")
+	TEST_RP("strh	r",7, VAL2,", [r",6, 24,", #60]")
+	TEST_P( "ldrh	r0, [r",1, 24,", #60]")
+	TEST_P( "ldrh	r7, [r",6, 24,", #60]")
+
+	TEST_R( "str	r",0, VAL1,", [sp, #0]")
+	TEST_R( "str	r",7, VAL2,", [sp, #160]")
+	TEST(   "ldr	r0, [sp, #0]")
+	TEST(   "ldr	r7, [sp, #160]")
+
+	TEST_RP("str	r",0, VAL1,", [r",0, 24,"]")
+	TEST_P( "ldr	r0, [r",0, 24,"]")
+
+	TEST_GROUP("Generate PC-/SP-relative address")
+
+	TEST("add	r0, pc, #4")
+	TEST("add	r7, pc, #1020")
+	TEST("add	r0, sp, #4")
+	TEST("add	r7, sp, #1020")
+
+	TEST_GROUP("Miscellaneous 16-bit instructions")
+
+	TEST_UNSUPPORTED( "cpsie	i")
+	TEST_UNSUPPORTED( "cpsid	i")
+	TEST_UNSUPPORTED( "setend	le")
+	TEST_UNSUPPORTED( "setend	be")
+
+	TEST("add	sp, #"__stringify(TEST_MEMORY_SIZE)) /* Assumes TEST_MEMORY_SIZE < 0x400 */
+	TEST("sub	sp, #0x7f*4")
+
+DONT_TEST_IN_ITBLOCK(
+	TEST_BF_R(  "cbnz	r",0,0, ", 2f")
+	TEST_BF_R(  "cbz	r",2,-1,", 2f")
+	TEST_BF_RX( "cbnz	r",4,1, ", 2f", SPACE_0x20)
+	TEST_BF_RX( "cbz	r",7,0, ", 2f", SPACE_0x40)
+)
+	TEST_R("sxth	r0, r",7, HH1,"")
+	TEST_R("sxth	r7, r",0, HH2,"")
+	TEST_R("sxtb	r0, r",7, HH1,"")
+	TEST_R("sxtb	r7, r",0, HH2,"")
+	TEST_R("uxth	r0, r",7, HH1,"")
+	TEST_R("uxth	r7, r",0, HH2,"")
+	TEST_R("uxtb	r0, r",7, HH1,"")
+	TEST_R("uxtb	r7, r",0, HH2,"")
+	TEST_R("rev	r0, r",7, VAL1,"")
+	TEST_R("rev	r7, r",0, VAL2,"")
+	TEST_R("rev16	r0, r",7, VAL1,"")
+	TEST_R("rev16	r7, r",0, VAL2,"")
+	TEST_UNSUPPORTED(__inst_thumb16(0xba80) "")
+	TEST_UNSUPPORTED(__inst_thumb16(0xbabf) "")
+	TEST_R("revsh	r0, r",7, VAL1,"")
+	TEST_R("revsh	r7, r",0, VAL2,"")
+
+#define TEST_POPPC(code, offset)	\
+	TESTCASE_START(code)		\
+	TEST_ARG_PTR(13, offset)	\
+	TEST_ARG_END("")		\
+	TEST_BRANCH_F(code)		\
+	TESTCASE_END
+
+	TEST("push	{r0}")
+	TEST("push	{r7}")
+	TEST("push	{r14}")
+	TEST("push	{r0-r7,r14}")
+	TEST("push	{r0,r2,r4,r6,r14}")
+	TEST("push	{r1,r3,r5,r7}")
+	TEST("pop	{r0}")
+	TEST("pop	{r7}")
+	TEST("pop	{r0,r2,r4,r6}")
+	TEST_POPPC("pop	{pc}",15*4)
+	TEST_POPPC("pop	{r0-r7,pc}",7*4)
+	TEST_POPPC("pop	{r1,r3,r5,r7,pc}",11*4)
+	TEST_THUMB_TO_ARM_INTERWORK_P("pop	{pc}	@ ",13,15*4,"")
+	TEST_THUMB_TO_ARM_INTERWORK_P("pop	{r0-r7,pc}	@ ",13,7*4,"")
+
+	TEST_UNSUPPORTED("bkpt.n	0")
+	TEST_UNSUPPORTED("bkpt.n	255")
+
+	TEST_SUPPORTED("yield")
+	TEST("sev")
+	TEST("nop")
+	TEST("wfi")
+	TEST_SUPPORTED("wfe")
+	TEST_UNSUPPORTED(__inst_thumb16(0xbf50) "") /* Unassigned hints */
+	TEST_UNSUPPORTED(__inst_thumb16(0xbff0) "") /* Unassigned hints */
+
+#define TEST_IT(code, code2)			\
+	TESTCASE_START(code)			\
+	TEST_ARG_END("")			\
+	"50:	nop			\n\t"	\
+	"1:	"code"			\n\t"	\
+	"	"code2"			\n\t"	\
+	"2:	nop			\n\t"	\
+	TESTCASE_END
+
+DONT_TEST_IN_ITBLOCK(
+	TEST_IT("it	eq","moveq r0,#0")
+	TEST_IT("it	vc","movvc r0,#0")
+	TEST_IT("it	le","movle r0,#0")
+	TEST_IT("ite	eq","moveq r0,#0\n\t  movne r1,#1")
+	TEST_IT("itet	vc","movvc r0,#0\n\t  movvs r1,#1\n\t  movvc r2,#2")
+	TEST_IT("itete	le","movle r0,#0\n\t  movgt r1,#1\n\t  movle r2,#2\n\t  movgt r3,#3")
+	TEST_IT("itttt	le","movle r0,#0\n\t  movle r1,#1\n\t  movle r2,#2\n\t  movle r3,#3")
+	TEST_IT("iteee	le","movle r0,#0\n\t  movgt r1,#1\n\t  movgt r2,#2\n\t  movgt r3,#3")
+)
+
+	TEST_GROUP("Load and store multiple")
+
+	TEST_P("ldmia	r",4, 16*4,"!, {r0,r7}")
+	TEST_P("ldmia	r",7, 16*4,"!, {r0-r6}")
+	TEST_P("stmia	r",4, 16*4,"!, {r0,r7}")
+	TEST_P("stmia	r",0, 16*4,"!, {r0-r7}")
+
+	TEST_GROUP("Conditional branch and Supervisor Call instructions")
+
+CONDITION_INSTRUCTIONS(8,
+	TEST_BF("beq	2f")
+	TEST_BB("bne	2b")
+	TEST_BF("bgt	2f")
+	TEST_BB("blt	2b")
+)
+	TEST_UNSUPPORTED(__inst_thumb16(0xde00) "")
+	TEST_UNSUPPORTED(__inst_thumb16(0xdeff) "")
+	TEST_UNSUPPORTED("svc	#0x00")
+	TEST_UNSUPPORTED("svc	#0xff")
+
+	TEST_GROUP("Unconditional branch")
+
+	TEST_BF(  "b	2f")
+	TEST_BB(  "b	2b")
+	TEST_BF_X("b	2f", SPACE_0x400)
+	TEST_BB_X("b	2b", SPACE_0x400)
+
+	TEST_GROUP("Testing instructions in IT blocks")
+
+	TEST_ITBLOCK("subs.n r0, r0")
+
+	verbose("\n");
+}
+
+
+void kprobe_thumb32_test_cases(void)
+{
+	kprobe_test_flags = 0;
+
+	TEST_GROUP("Load/store multiple")
+
+	TEST_UNSUPPORTED("rfedb	sp")
+	TEST_UNSUPPORTED("rfeia	sp")
+	TEST_UNSUPPORTED("rfedb	sp!")
+	TEST_UNSUPPORTED("rfeia	sp!")
+
+	TEST_P(   "stmia	r",0, 16*4,", {r0,r8}")
+	TEST_P(   "stmia	r",4, 16*4,", {r0-r12,r14}")
+	TEST_P(   "stmia	r",7, 16*4,"!, {r8-r12,r14}")
+	TEST_P(   "stmia	r",12,16*4,"!, {r1,r3,r5,r7,r8-r11,r14}")
+
+	TEST_P(   "ldmia	r",0, 16*4,", {r0,r8}")
+	TEST_P(   "ldmia	r",4, 0,   ", {r0-r12,r14}")
+	TEST_BF_P("ldmia	r",5, 8*4, "!, {r6-r12,r15}")
+	TEST_P(   "ldmia	r",12,16*4,"!, {r1,r3,r5,r7,r8-r11,r14}")
+	TEST_BF_P("ldmia	r",14,14*4,"!, {r4,pc}")
+
+	TEST_P(   "stmdb	r",0, 16*4,", {r0,r8}")
+	TEST_P(   "stmdb	r",4, 16*4,", {r0-r12,r14}")
+	TEST_P(   "stmdb	r",5, 16*4,"!, {r8-r12,r14}")
+	TEST_P(   "stmdb	r",12,16*4,"!, {r1,r3,r5,r7,r8-r11,r14}")
+
+	TEST_P(   "ldmdb	r",0, 16*4,", {r0,r8}")
+	TEST_P(   "ldmdb	r",4, 16*4,", {r0-r12,r14}")
+	TEST_BF_P("ldmdb	r",5, 16*4,"!, {r6-r12,r15}")
+	TEST_P(   "ldmdb	r",12,16*4,"!, {r1,r3,r5,r7,r8-r11,r14}")
+	TEST_BF_P("ldmdb	r",14,16*4,"!, {r4,pc}")
+
+	TEST_P(   "stmdb	r",13,16*4,"!, {r3-r12,lr}")
+	TEST_P(	  "stmdb	r",13,16*4,"!, {r3-r12}")
+	TEST_P(   "stmdb	r",2, 16*4,", {r3-r12,lr}")
+	TEST_P(   "stmdb	r",13,16*4,"!, {r2-r12,lr}")
+	TEST_P(   "stmdb	r",0, 16*4,", {r0-r12}")
+	TEST_P(   "stmdb	r",0, 16*4,", {r0-r12,lr}")
+
+	TEST_BF_P("ldmia	r",13,5*4, "!, {r3-r12,pc}")
+	TEST_P(	  "ldmia	r",13,5*4, "!, {r3-r12}")
+	TEST_BF_P("ldmia	r",2, 5*4, "!, {r3-r12,pc}")
+	TEST_BF_P("ldmia	r",13,4*4, "!, {r2-r12,pc}")
+	TEST_P(   "ldmia	r",0, 16*4,", {r0-r12}")
+	TEST_P(   "ldmia	r",0, 16*4,", {r0-r12,lr}")
+
+	TEST_THUMB_TO_ARM_INTERWORK_P("ldmia	r",0,14*4,", {r12,pc}")
+	TEST_THUMB_TO_ARM_INTERWORK_P("ldmia	r",13,2*4,", {r0-r12,pc}")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xe88f0101) "	@ stmia	pc, {r0,r8}")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe92f5f00) "	@ stmdb	pc!, {r8-r12,r14}")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe8bdc000) "	@ ldmia	r13!, {r14,pc}")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe93ec000) "	@ ldmdb	r14!, {r14,pc}")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe8a73f00) "	@ stmia	r7!, {r8-r12,sp}")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe8a79f00) "	@ stmia	r7!, {r8-r12,pc}")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe93e2010) "	@ ldmdb	r14!, {r4,sp}")
+
+	TEST_GROUP("Load/store double or exclusive, table branch")
+
+	TEST_P(  "ldrd	r0, r1, [r",1, 24,", #-16]")
+	TEST(    "ldrd	r12, r14, [sp, #16]")
+	TEST_P(  "ldrd	r1, r0, [r",7, 24,", #-16]!")
+	TEST(    "ldrd	r14, r12, [sp, #16]!")
+	TEST_P(  "ldrd	r1, r0, [r",7, 24,"], #16")
+	TEST(    "ldrd	r7, r8, [sp], #-16")
+
+	TEST_X( "ldrd	r12, r14, 3f",
+		".align 3				\n\t"
+		"3:	.word	"__stringify(VAL1)"	\n\t"
+		"	.word	"__stringify(VAL2))
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xe9ffec04) "	@ ldrd	r14, r12, [pc, #16]!")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe8ffec04) "	@ ldrd	r14, r12, [pc], #16")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe9d4d800) "	@ ldrd	sp, r8, [r4]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe9d4f800) "	@ ldrd	pc, r8, [r4]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe9d47d00) "	@ ldrd	r7, sp, [r4]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe9d47f00) "	@ ldrd	r7, pc, [r4]")
+
+	TEST_RRP("strd	r",0, VAL1,", r",1, VAL2,", [r",1, 24,", #-16]")
+	TEST_RR( "strd	r",12,VAL2,", r",14,VAL1,", [sp, #16]")
+	TEST_RRP("strd	r",1, VAL1,", r",0, VAL2,", [r",7, 24,", #-16]!")
+	TEST_RR( "strd	r",14,VAL2,", r",12,VAL1,", [sp, #16]!")
+	TEST_RRP("strd	r",1, VAL1,", r",0, VAL2,", [r",7, 24,"], #16")
+	TEST_RR( "strd	r",7, VAL2,", r",8, VAL1,", [sp], #-16")
+	TEST_RRP("strd	r",6, VAL1,", r",7, VAL2,", [r",13, TEST_MEMORY_SIZE,", #-"__stringify(MAX_STACK_SIZE)"]!")
+	TEST_UNSUPPORTED("strd r6, r7, [r13, #-"__stringify(MAX_STACK_SIZE)"-8]!")
+	TEST_RRP("strd	r",4, VAL1,", r",5, VAL2,", [r",14, TEST_MEMORY_SIZE,", #-"__stringify(MAX_STACK_SIZE)"-8]!")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe9efec04) "	@ strd	r14, r12, [pc, #16]!")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe8efec04) "	@ strd	r14, r12, [pc], #16")
+
+	TEST_RX("tbb	[pc, r",0, (9f-(1f+4)),"]",
+		"9:			\n\t"
+		".byte	(2f-1b-4)>>1	\n\t"
+		".byte	(3f-1b-4)>>1	\n\t"
+		"3:	mvn	r0, r0	\n\t"
+		"2:	nop		\n\t")
+
+	TEST_RX("tbb	[pc, r",4, (9f-(1f+4)+1),"]",
+		"9:			\n\t"
+		".byte	(2f-1b-4)>>1	\n\t"
+		".byte	(3f-1b-4)>>1	\n\t"
+		"3:	mvn	r0, r0	\n\t"
+		"2:	nop		\n\t")
+
+	TEST_RRX("tbb	[r",1,9f,", r",2,0,"]",
+		"9:			\n\t"
+		".byte	(2f-1b-4)>>1	\n\t"
+		".byte	(3f-1b-4)>>1	\n\t"
+		"3:	mvn	r0, r0	\n\t"
+		"2:	nop		\n\t")
+
+	TEST_RX("tbh	[pc, r",7, (9f-(1f+4))>>1,"]",
+		"9:			\n\t"
+		".short	(2f-1b-4)>>1	\n\t"
+		".short	(3f-1b-4)>>1	\n\t"
+		"3:	mvn	r0, r0	\n\t"
+		"2:	nop		\n\t")
+
+	TEST_RX("tbh	[pc, r",12, ((9f-(1f+4))>>1)+1,"]",
+		"9:			\n\t"
+		".short	(2f-1b-4)>>1	\n\t"
+		".short	(3f-1b-4)>>1	\n\t"
+		"3:	mvn	r0, r0	\n\t"
+		"2:	nop		\n\t")
+
+	TEST_RRX("tbh	[r",1,9f, ", r",14,1,"]",
+		"9:			\n\t"
+		".short	(2f-1b-4)>>1	\n\t"
+		".short	(3f-1b-4)>>1	\n\t"
+		"3:	mvn	r0, r0	\n\t"
+		"2:	nop		\n\t")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xe8d1f01f) "	@ tbh [r1, pc]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe8d1f01d) "	@ tbh [r1, sp]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xe8ddf012) "	@ tbh [sp, r2]")
+
+	TEST_UNSUPPORTED("strexb	r0, r1, [r2]")
+	TEST_UNSUPPORTED("strexh	r0, r1, [r2]")
+	TEST_UNSUPPORTED("strexd	r0, r1, [r2]")
+	TEST_UNSUPPORTED("ldrexb	r0, [r1]")
+	TEST_UNSUPPORTED("ldrexh	r0, [r1]")
+	TEST_UNSUPPORTED("ldrexd	r0, [r1]")
+
+	TEST_GROUP("Data-processing (shifted register) and (modified immediate)")
+
+#define _DATA_PROCESSING32_DNM(op,s,val)					\
+	TEST_RR(op s".w	r0,  r",1, VAL1,", r",2, val, "")			\
+	TEST_RR(op s"	r1,  r",1, VAL1,", r",2, val, ", lsl #3")		\
+	TEST_RR(op s"	r2,  r",3, VAL1,", r",2, val, ", lsr #4")		\
+	TEST_RR(op s"	r3,  r",3, VAL1,", r",2, val, ", asr #5")		\
+	TEST_RR(op s"	r4,  r",5, VAL1,", r",2, N(val),", asr #6")		\
+	TEST_RR(op s"	r5,  r",5, VAL1,", r",2, val, ", ror #7")		\
+	TEST_RR(op s"	r8,  r",9, VAL1,", r",10,val, ", rrx")			\
+	TEST_R( op s"	r0,  r",11,VAL1,", #0x00010001")			\
+	TEST_R( op s"	r11, r",0, VAL1,", #0xf5000000")			\
+	TEST_R( op s"	r7,  r",8, VAL2,", #0x000af000")
+
+#define DATA_PROCESSING32_DNM(op,val)		\
+	_DATA_PROCESSING32_DNM(op,"",val)	\
+	_DATA_PROCESSING32_DNM(op,"s",val)
+
+#define DATA_PROCESSING32_NM(op,val)					\
+	TEST_RR(op".w	r",1, VAL1,", r",2, val, "")			\
+	TEST_RR(op"	r",1, VAL1,", r",2, val, ", lsl #3")		\
+	TEST_RR(op"	r",3, VAL1,", r",2, val, ", lsr #4")		\
+	TEST_RR(op"	r",3, VAL1,", r",2, val, ", asr #5")		\
+	TEST_RR(op"	r",5, VAL1,", r",2, N(val),", asr #6")		\
+	TEST_RR(op"	r",5, VAL1,", r",2, val, ", ror #7")		\
+	TEST_RR(op"	r",9, VAL1,", r",10,val, ", rrx")		\
+	TEST_R( op"	r",11,VAL1,", #0x00010001")			\
+	TEST_R( op"	r",0, VAL1,", #0xf5000000")			\
+	TEST_R( op"	r",8, VAL2,", #0x000af000")
+
+#define _DATA_PROCESSING32_DM(op,s,val)				\
+	TEST_R( op s".w	r0,  r",14, val, "")			\
+	TEST_R( op s"	r1,  r",12, val, ", lsl #3")		\
+	TEST_R( op s"	r2,  r",11, val, ", lsr #4")		\
+	TEST_R( op s"	r3,  r",10, val, ", asr #5")		\
+	TEST_R( op s"	r4,  r",9, N(val),", asr #6")		\
+	TEST_R( op s"	r5,  r",8, val, ", ror #7")		\
+	TEST_R( op s"	r8,  r",7,val, ", rrx")			\
+	TEST(   op s"	r0,  #0x00010001")			\
+	TEST(   op s"	r11, #0xf5000000")			\
+	TEST(   op s"	r7,  #0x000af000")			\
+	TEST(   op s"	r4,  #0x00005a00")
+
+#define DATA_PROCESSING32_DM(op,val)		\
+	_DATA_PROCESSING32_DM(op,"",val)	\
+	_DATA_PROCESSING32_DM(op,"s",val)
+
+	DATA_PROCESSING32_DNM("and",0xf00f00ff)
+	DATA_PROCESSING32_NM("tst",0xf00f00ff)
+	DATA_PROCESSING32_DNM("bic",0xf00f00ff)
+	DATA_PROCESSING32_DNM("orr",0xf00f00ff)
+	DATA_PROCESSING32_DM("mov",VAL2)
+	DATA_PROCESSING32_DNM("orn",0xf00f00ff)
+	DATA_PROCESSING32_DM("mvn",VAL2)
+	DATA_PROCESSING32_DNM("eor",0xf00f00ff)
+	DATA_PROCESSING32_NM("teq",0xf00f00ff)
+	DATA_PROCESSING32_DNM("add",VAL2)
+	DATA_PROCESSING32_NM("cmn",VAL2)
+	DATA_PROCESSING32_DNM("adc",VAL2)
+	DATA_PROCESSING32_DNM("sbc",VAL2)
+	DATA_PROCESSING32_DNM("sub",VAL2)
+	DATA_PROCESSING32_NM("cmp",VAL2)
+	DATA_PROCESSING32_DNM("rsb",VAL2)
+
+	TEST_RR("pkhbt	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR("pkhbt	r14,r",12, HH1,", r",10,HH2,", lsl #2")
+	TEST_RR("pkhtb	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR("pkhtb	r14,r",12, HH1,", r",10,HH2,", asr #2")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xea170f0d) "	@ tst.w r7, sp")
+	TEST_UNSUPPORTED(__inst_thumb32(0xea170f0f) "	@ tst.w r7, pc")
+	TEST_UNSUPPORTED(__inst_thumb32(0xea1d0f07) "	@ tst.w sp, r7")
+	TEST_UNSUPPORTED(__inst_thumb32(0xea1f0f07) "	@ tst.w pc, r7")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf01d1f08) "	@ tst sp, #0x00080008")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf01f1f08) "	@ tst pc, #0x00080008")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xea970f0d) "	@ teq.w r7, sp")
+	TEST_UNSUPPORTED(__inst_thumb32(0xea970f0f) "	@ teq.w r7, pc")
+	TEST_UNSUPPORTED(__inst_thumb32(0xea9d0f07) "	@ teq.w sp, r7")
+	TEST_UNSUPPORTED(__inst_thumb32(0xea9f0f07) "	@ teq.w pc, r7")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf09d1f08) "	@ tst sp, #0x00080008")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf09f1f08) "	@ tst pc, #0x00080008")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb170f0d) "	@ cmn.w r7, sp")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb170f0f) "	@ cmn.w r7, pc")
+	TEST_P("cmn.w	sp, r",7,0,"")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb1f0f07) "	@ cmn.w pc, r7")
+	TEST(  "cmn	sp, #0x00080008")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf11f1f08) "	@ cmn pc, #0x00080008")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xebb70f0d) "	@ cmp.w r7, sp")
+	TEST_UNSUPPORTED(__inst_thumb32(0xebb70f0f) "	@ cmp.w r7, pc")
+	TEST_P("cmp.w	sp, r",7,0,"")
+	TEST_UNSUPPORTED(__inst_thumb32(0xebbf0f07) "	@ cmp.w pc, r7")
+	TEST(  "cmp	sp, #0x00080008")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf1bf1f08) "	@ cmp pc, #0x00080008")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xea5f070d) "	@ movs.w r7, sp")
+	TEST_UNSUPPORTED(__inst_thumb32(0xea5f070f) "	@ movs.w r7, pc")
+	TEST_UNSUPPORTED(__inst_thumb32(0xea5f0d07) "	@ movs.w sp, r7")
+	TEST_UNSUPPORTED(__inst_thumb32(0xea4f0f07) "	@ mov.w  pc, r7")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf04f1d08) "	@ mov sp, #0x00080008")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf04f1f08) "	@ mov pc, #0x00080008")
+
+	TEST_R("add.w	r0, sp, r",1, 4,"")
+	TEST_R("adds	r0, sp, r",1, 4,", asl #3")
+	TEST_R("add	r0, sp, r",1, 4,", asl #4")
+	TEST_R("add	r0, sp, r",1, 16,", ror #1")
+	TEST_R("add.w	sp, sp, r",1, 4,"")
+	TEST_R("add	sp, sp, r",1, 4,", asl #3")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb0d1d01) "	@ add sp, sp, r1, asl #4")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb0d0d71) "	@ add sp, sp, r1, ror #1")
+	TEST(  "add.w	r0, sp, #24")
+	TEST(  "add.w	sp, sp, #24")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb0d0f01) "	@ add pc, sp, r1")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb0d000f) "	@ add r0, sp, pc")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb0d000d) "	@ add r0, sp, sp")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb0d0d0f) "	@ add sp, sp, pc")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb0d0d0d) "	@ add sp, sp, sp")
+
+	TEST_R("sub.w	r0, sp, r",1, 4,"")
+	TEST_R("subs	r0, sp, r",1, 4,", asl #3")
+	TEST_R("sub	r0, sp, r",1, 4,", asl #4")
+	TEST_R("sub	r0, sp, r",1, 16,", ror #1")
+	TEST_R("sub.w	sp, sp, r",1, 4,"")
+	TEST_R("sub	sp, sp, r",1, 4,", asl #3")
+	TEST_UNSUPPORTED(__inst_thumb32(0xebad1d01) "	@ sub sp, sp, r1, asl #4")
+	TEST_UNSUPPORTED(__inst_thumb32(0xebad0d71) "	@ sub sp, sp, r1, ror #1")
+	TEST_UNSUPPORTED(__inst_thumb32(0xebad0f01) "	@ sub pc, sp, r1")
+	TEST(  "sub.w	r0, sp, #24")
+	TEST(  "sub.w	sp, sp, #24")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xea02010f) "	@ and r1, r2, pc")
+	TEST_UNSUPPORTED(__inst_thumb32(0xea0f0103) "	@ and r1, pc, r3")
+	TEST_UNSUPPORTED(__inst_thumb32(0xea020f03) "	@ and pc, r2, r3")
+	TEST_UNSUPPORTED(__inst_thumb32(0xea02010d) "	@ and r1, r2, sp")
+	TEST_UNSUPPORTED(__inst_thumb32(0xea0d0103) "	@ and r1, sp, r3")
+	TEST_UNSUPPORTED(__inst_thumb32(0xea020d03) "	@ and sp, r2, r3")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf00d1108) "	@ and r1, sp, #0x00080008")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf00f1108) "	@ and r1, pc, #0x00080008")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf0021d08) "	@ and sp, r8, #0x00080008")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf0021f08) "	@ and pc, r8, #0x00080008")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb02010f) "	@ add r1, r2, pc")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb0f0103) "	@ add r1, pc, r3")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb020f03) "	@ add pc, r2, r3")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb02010d) "	@ add r1, r2, sp")
+	TEST_SUPPORTED(  __inst_thumb32(0xeb0d0103) "	@ add r1, sp, r3")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb020d03) "	@ add sp, r2, r3")
+	TEST_SUPPORTED(  __inst_thumb32(0xf10d1108) "	@ add r1, sp, #0x00080008")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf10d1f08) "	@ add pc, sp, #0x00080008")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf10f1108) "	@ add r1, pc, #0x00080008")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf1021d08) "	@ add sp, r8, #0x00080008")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf1021f08) "	@ add pc, r8, #0x00080008")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xeaa00000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeaf00000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb200000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeb800000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xebe00000) "")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xf0a00000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf0c00000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf0f00000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf1200000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf1800000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf1e00000) "")
+
+	TEST_GROUP("Coprocessor instructions")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xec000000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xeff00000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfc000000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfff00000) "")
+
+	TEST_GROUP("Data-processing (plain binary immediate)")
+
+	TEST_R("addw	r0,  r",1, VAL1,", #0x123")
+	TEST(  "addw	r14, sp, #0xf5a")
+	TEST(  "addw	sp, sp, #0x20")
+	TEST(  "addw	r7,  pc, #0x888")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf20f1f20) "	@ addw pc, pc, #0x120")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf20d1f20) "	@ addw pc, sp, #0x120")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf20f1d20) "	@ addw sp, pc, #0x120")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf2001d20) "	@ addw sp, r0, #0x120")
+
+	TEST_R("subw	r0,  r",1, VAL1,", #0x123")
+	TEST(  "subw	r14, sp, #0xf5a")
+	TEST(  "subw	sp, sp, #0x20")
+	TEST(  "subw	r7,  pc, #0x888")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf2af1f20) "	@ subw pc, pc, #0x120")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf2ad1f20) "	@ subw pc, sp, #0x120")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf2af1d20) "	@ subw sp, pc, #0x120")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf2a01d20) "	@ subw sp, r0, #0x120")
+
+	TEST("movw	r0, #0")
+	TEST("movw	r0, #0xffff")
+	TEST("movw	lr, #0xffff")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf2400d00) "	@ movw sp, #0")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf2400f00) "	@ movw pc, #0")
+
+	TEST_R("movt	r",0, VAL1,", #0")
+	TEST_R("movt	r",0, VAL2,", #0xffff")
+	TEST_R("movt	r",14,VAL1,", #0xffff")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf2c00d00) "	@ movt sp, #0")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf2c00f00) "	@ movt pc, #0")
+
+	TEST_R(     "ssat	r0, #24, r",0,   VAL1,"")
+	TEST_R(     "ssat	r14, #24, r",12, VAL2,"")
+	TEST_R(     "ssat	r0, #24, r",0,   VAL1,", lsl #8")
+	TEST_R(     "ssat	r14, #24, r",12, VAL2,", asr #8")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf30c0d17) "	@ ssat	sp, #24, r12")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf30c0f17) "	@ ssat	pc, #24, r12")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf30d0c17) "	@ ssat	r12, #24, sp")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf30f0c17) "	@ ssat	r12, #24, pc")
+
+	TEST_R(     "usat	r0, #24, r",0,   VAL1,"")
+	TEST_R(     "usat	r14, #24, r",12, VAL2,"")
+	TEST_R(     "usat	r0, #24, r",0,   VAL1,", lsl #8")
+	TEST_R(     "usat	r14, #24, r",12, VAL2,", asr #8")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf38c0d17) "	@ usat	sp, #24, r12")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf38c0f17) "	@ usat	pc, #24, r12")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf38d0c17) "	@ usat	r12, #24, sp")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf38f0c17) "	@ usat	r12, #24, pc")
+
+	TEST_R(     "ssat16	r0, #12, r",0,   HH1,"")
+	TEST_R(     "ssat16	r14, #12, r",12, HH2,"")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf32c0d0b) "	@ ssat16	sp, #12, r12")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf32c0f0b) "	@ ssat16	pc, #12, r12")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf32d0c0b) "	@ ssat16	r12, #12, sp")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf32f0c0b) "	@ ssat16	r12, #12, pc")
+
+	TEST_R(     "usat16	r0, #12, r",0,   HH1,"")
+	TEST_R(     "usat16	r14, #12, r",12, HH2,"")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf3ac0d0b) "	@ usat16	sp, #12, r12")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf3ac0f0b) "	@ usat16	pc, #12, r12")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf3ad0c0b) "	@ usat16	r12, #12, sp")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf3af0c0b) "	@ usat16	r12, #12, pc")
+
+	TEST_R(     "sbfx	r0, r",0  , VAL1,", #0, #31")
+	TEST_R(     "sbfx	r14, r",12, VAL2,", #8, #16")
+	TEST_R(     "sbfx	r4, r",10,  VAL1,", #16, #15")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf34c2d0f) "	@ sbfx	sp, r12, #8, #16")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf34c2f0f) "	@ sbfx	pc, r12, #8, #16")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf34d2c0f) "	@ sbfx	r12, sp, #8, #16")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf34f2c0f) "	@ sbfx	r12, pc, #8, #16")
+
+	TEST_R(     "ubfx	r0, r",0  , VAL1,", #0, #31")
+	TEST_R(     "ubfx	r14, r",12, VAL2,", #8, #16")
+	TEST_R(     "ubfx	r4, r",10,  VAL1,", #16, #15")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf3cc2d0f) "	@ ubfx	sp, r12, #8, #16")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf3cc2f0f) "	@ ubfx	pc, r12, #8, #16")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf3cd2c0f) "	@ ubfx	r12, sp, #8, #16")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf3cf2c0f) "	@ ubfx	r12, pc, #8, #16")
+
+	TEST_R(     "bfc	r",0, VAL1,", #4, #20")
+	TEST_R(     "bfc	r",14,VAL2,", #4, #20")
+	TEST_R(     "bfc	r",7, VAL1,", #0, #31")
+	TEST_R(     "bfc	r",8, VAL2,", #0, #31")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf36f0d1e) "	@ bfc	sp, #0, #31")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf36f0f1e) "	@ bfc	pc, #0, #31")
+
+	TEST_RR(    "bfi	r",0, VAL1,", r",0  , VAL2,", #0, #31")
+	TEST_RR(    "bfi	r",12,VAL1,", r",14 , VAL2,", #4, #20")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf36e1d17) "	@ bfi	sp, r14, #4, #20")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf36e1f17) "	@ bfi	pc, r14, #4, #20")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf36d1e17) "	@ bfi	r14, sp, #4, #20")
+
+	TEST_GROUP("Branches and miscellaneous control")
+
+CONDITION_INSTRUCTIONS(22,
+	TEST_BF("beq.w	2f")
+	TEST_BB("bne.w	2b")
+	TEST_BF("bgt.w	2f")
+	TEST_BB("blt.w	2b")
+	TEST_BF_X("bpl.w	2f", SPACE_0x1000)
+)
+
+	TEST_UNSUPPORTED("msr	cpsr, r0")
+	TEST_UNSUPPORTED("msr	cpsr_f, r1")
+	TEST_UNSUPPORTED("msr	spsr, r2")
+
+	TEST_UNSUPPORTED("cpsie.w	i")
+	TEST_UNSUPPORTED("cpsid.w	i")
+	TEST_UNSUPPORTED("cps	0x13")
+
+	TEST_SUPPORTED("yield.w")
+	TEST("sev.w")
+	TEST("nop.w")
+	TEST("wfi.w")
+	TEST_SUPPORTED("wfe.w")
+	TEST_UNSUPPORTED("dbg.w	#0")
+
+	TEST_UNSUPPORTED("clrex")
+	TEST_UNSUPPORTED("dsb")
+	TEST_UNSUPPORTED("dmb")
+	TEST_UNSUPPORTED("isb")
+
+	TEST_UNSUPPORTED("bxj	r0")
+
+	TEST_UNSUPPORTED("subs	pc, lr, #4")
+
+	TEST_RMASKED("mrs	r",0,~PSR_IGNORE_BITS,", cpsr")
+	TEST_RMASKED("mrs	r",14,~PSR_IGNORE_BITS,", cpsr")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf3ef8d00) "	@ mrs	sp, spsr")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf3ef8f00) "	@ mrs	pc, spsr")
+	TEST_UNSUPPORTED("mrs	r0, spsr")
+	TEST_UNSUPPORTED("mrs	lr, spsr")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xf7f08000) " @ smc #0")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xf7f0a000) " @ undefeined")
+
+	TEST_BF(  "b.w	2f")
+	TEST_BB(  "b.w	2b")
+	TEST_BF_X("b.w	2f", SPACE_0x1000)
+
+	TEST_BF(  "bl.w	2f")
+	TEST_BB(  "bl.w	2b")
+	TEST_BB_X("bl.w	2b", SPACE_0x1000)
+
+	TEST_X(	"blx	__dummy_arm_subroutine",
+		".arm				\n\t"
+		".align				\n\t"
+		".type __dummy_arm_subroutine, %%function \n\t"
+		"__dummy_arm_subroutine:	\n\t"
+		"mov	r0, pc			\n\t"
+		"bx	lr			\n\t"
+		".thumb				\n\t"
+	)
+	TEST(	"blx	__dummy_arm_subroutine")
+
+	TEST_GROUP("Store single data item")
+
+#define SINGLE_STORE(size)							\
+	TEST_RP( "str"size"	r",0, VAL1,", [r",11,-1024,", #1024]")		\
+	TEST_RP( "str"size"	r",14,VAL2,", [r",1, -1024,", #1080]")		\
+	TEST_RP( "str"size"	r",0, VAL1,", [r",11,256,  ", #-120]")		\
+	TEST_RP( "str"size"	r",14,VAL2,", [r",1, 256,  ", #-128]")		\
+	TEST_RP( "str"size"	r",0, VAL1,", [r",11,24,  "], #120")		\
+	TEST_RP( "str"size"	r",14,VAL2,", [r",1, 24,  "], #128")		\
+	TEST_RP( "str"size"	r",0, VAL1,", [r",11,24,  "], #-120")		\
+	TEST_RP( "str"size"	r",14,VAL2,", [r",1, 24,  "], #-128")		\
+	TEST_RP( "str"size"	r",0, VAL1,", [r",11,24,   ", #120]!")		\
+	TEST_RP( "str"size"	r",14,VAL2,", [r",1, 24,   ", #128]!")		\
+	TEST_RP( "str"size"	r",0, VAL1,", [r",11,256,  ", #-120]!")		\
+	TEST_RP( "str"size"	r",14,VAL2,", [r",1, 256,  ", #-128]!")		\
+	TEST_RPR("str"size".w	r",0, VAL1,", [r",1, 0,", r",2, 4,"]")		\
+	TEST_RPR("str"size"	r",14,VAL2,", [r",10,0,", r",11,4,", lsl #1]")	\
+	TEST_UNSUPPORTED("str"size"	r0, [r13, r1]")				\
+	TEST_R(  "str"size".w	r",7, VAL1,", [sp, #24]")			\
+	TEST_RP( "str"size".w	r",0, VAL2,", [r",0,0, "]")			\
+	TEST_RP( "str"size"	r",6, VAL1,", [r",13, TEST_MEMORY_SIZE,", #-"__stringify(MAX_STACK_SIZE)"]!") \
+	TEST_UNSUPPORTED("str"size"	r6, [r13, #-"__stringify(MAX_STACK_SIZE)"-8]!")			\
+	TEST_RP( "str"size"	r",4, VAL2,", [r",12, TEST_MEMORY_SIZE,", #-"__stringify(MAX_STACK_SIZE)"-8]!") \
+	TEST_UNSUPPORTED("str"size"t	r0, [r1, #4]")
+
+	SINGLE_STORE("b")
+	SINGLE_STORE("h")
+	SINGLE_STORE("")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xf801000d) "	@ strb	r0, [r1, r13]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf821000d) "	@ strh	r0, [r1, r13]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf841000d) "	@ str	r0, [r1, r13]")
+
+	TEST("str	sp, [sp]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf8cfe000) "	@ str	r14, [pc]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf8cef000) "	@ str	pc, [r14]")
+
+	TEST_GROUP("Advanced SIMD element or structure load/store instructions")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xf9000000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf92fffff) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf9800000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf9efffff) "")
+
+	TEST_GROUP("Load single data item and memory hints")
+
+#define SINGLE_LOAD(size)						\
+	TEST_P( "ldr"size"	r0, [r",11,-1024, ", #1024]")		\
+	TEST_P( "ldr"size"	r14, [r",1, -1024,", #1080]")		\
+	TEST_P( "ldr"size"	r0, [r",11,256,   ", #-120]")		\
+	TEST_P( "ldr"size"	r14, [r",1, 256,  ", #-128]")		\
+	TEST_P( "ldr"size"	r0, [r",11,24,   "], #120")		\
+	TEST_P( "ldr"size"	r14, [r",1, 24,  "], #128")		\
+	TEST_P( "ldr"size"	r0, [r",11,24,   "], #-120")		\
+	TEST_P( "ldr"size"	r14, [r",1,24,   "], #-128")		\
+	TEST_P( "ldr"size"	r0, [r",11,24,    ", #120]!")		\
+	TEST_P( "ldr"size"	r14, [r",1, 24,   ", #128]!")		\
+	TEST_P( "ldr"size"	r0, [r",11,256,   ", #-120]!")		\
+	TEST_P( "ldr"size"	r14, [r",1, 256,  ", #-128]!")		\
+	TEST_PR("ldr"size".w	r0, [r",1, 0,", r",2, 4,"]")		\
+	TEST_PR("ldr"size"	r14, [r",10,0,", r",11,4,", lsl #1]")	\
+	TEST_X( "ldr"size".w	r0, 3f",				\
+		".align 3				\n\t"		\
+		"3:	.word	"__stringify(VAL1))			\
+	TEST_X( "ldr"size".w	r14, 3f",				\
+		".align 3				\n\t"		\
+		"3:	.word	"__stringify(VAL2))			\
+	TEST(   "ldr"size".w	r7, 3b")				\
+	TEST(   "ldr"size".w	r7, [sp, #24]")				\
+	TEST_P( "ldr"size".w	r0, [r",0,0, "]")			\
+	TEST_UNSUPPORTED("ldr"size"t	r0, [r1, #4]")
+
+	SINGLE_LOAD("b")
+	SINGLE_LOAD("sb")
+	SINGLE_LOAD("h")
+	SINGLE_LOAD("sh")
+	SINGLE_LOAD("")
+
+	TEST_BF_P("ldr	pc, [r",14, 15*4,"]")
+	TEST_P(   "ldr	sp, [r",14, 13*4,"]")
+	TEST_BF_R("ldr	pc, [sp, r",14, 15*4,"]")
+	TEST_R(   "ldr	sp, [sp, r",14, 13*4,"]")
+	TEST_THUMB_TO_ARM_INTERWORK_P("ldr	pc, [r",0,0,", #15*4]")
+	TEST_SUPPORTED("ldr	sp, 99f")
+	TEST_SUPPORTED("ldr	pc, 99f")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xf854700d) "	@ ldr	r7, [r4, sp]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf854700f) "	@ ldr	r7, [r4, pc]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf814700d) "	@ ldrb	r7, [r4, sp]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf814700f) "	@ ldrb	r7, [r4, pc]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf89fd004) "	@ ldrb	sp, 99f")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf814d008) "	@ ldrb	sp, [r4, r8]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf894d000) "	@ ldrb	sp, [r4]")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xf8600000) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_thumb32(0xf9ffffff) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_thumb32(0xf9500000) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_thumb32(0xf95fffff) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_thumb32(0xf8000800) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_thumb32(0xf97ffaff) "") /* Unallocated space */
+
+	TEST(   "pli	[pc, #4]")
+	TEST(   "pli	[pc, #-4]")
+	TEST(   "pld	[pc, #4]")
+	TEST(   "pld	[pc, #-4]")
+
+	TEST_P( "pld	[r",0,-1024,", #1024]")
+	TEST(   __inst_thumb32(0xf8b0f400) "	@ pldw	[r0, #1024]")
+	TEST_P( "pli	[r",4, 0b,", #1024]")
+	TEST_P( "pld	[r",7, 120,", #-120]")
+	TEST(   __inst_thumb32(0xf837fc78) "	@ pldw	[r7, #-120]")
+	TEST_P( "pli	[r",11,120,", #-120]")
+	TEST(   "pld	[sp, #0]")
+
+	TEST_PR("pld	[r",7, 24, ", r",0, 16,"]")
+	TEST_PR("pld	[r",8, 24, ", r",12,16,", lsl #3]")
+	TEST_SUPPORTED(__inst_thumb32(0xf837f000) "	@ pldw	[r7, r0]")
+	TEST_SUPPORTED(__inst_thumb32(0xf838f03c) "	@ pldw	[r8, r12, lsl #3]");
+	TEST_RR("pli	[r",12,0b,", r",0, 16,"]")
+	TEST_RR("pli	[r",0, 0b,", r",12,16,", lsl #3]")
+	TEST_R( "pld	[sp, r",1, 16,"]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf817f00d) "  @pld	[r7, sp]")
+	TEST_UNSUPPORTED(__inst_thumb32(0xf817f00f) "  @pld	[r7, pc]")
+
+	TEST_GROUP("Data-processing (register)")
+
+#define SHIFTS32(op)					\
+	TEST_RR(op"	r0,  r",1, VAL1,", r",2, 3, "")	\
+	TEST_RR(op"	r14, r",12,VAL2,", r",11,10,"")
+
+	SHIFTS32("lsl")
+	SHIFTS32("lsls")
+	SHIFTS32("lsr")
+	SHIFTS32("lsrs")
+	SHIFTS32("asr")
+	SHIFTS32("asrs")
+	SHIFTS32("ror")
+	SHIFTS32("rors")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa01ff02) "	@ lsl	pc, r1, r2")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa01fd02) "	@ lsl	sp, r1, r2")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa0ff002) "	@ lsl	r0, pc, r2")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa0df002) "	@ lsl	r0, sp, r2")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa01f00f) "	@ lsl	r0, r1, pc")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa01f00d) "	@ lsl	r0, r1, sp")
+
+	TEST_RR(    "sxtah	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "sxtah	r14,r",12, HH2,", r",10,HH1,", ror #8")
+	TEST_R(     "sxth	r8, r",7,  HH1,"")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa0fff87) "	@ sxth	pc, r7");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa0ffd87) "	@ sxth	sp, r7");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa0ff88f) "	@ sxth	r8, pc");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa0ff88d) "	@ sxth	r8, sp");
+
+	TEST_RR(    "uxtah	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uxtah	r14,r",12, HH2,", r",10,HH1,", ror #8")
+	TEST_R(     "uxth	r8, r",7,  HH1,"")
+
+	TEST_RR(    "sxtab16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "sxtab16	r14,r",12, HH2,", r",10,HH1,", ror #8")
+	TEST_R(     "sxtb16	r8, r",7,  HH1,"")
+
+	TEST_RR(    "uxtab16	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uxtab16	r14,r",12, HH2,", r",10,HH1,", ror #8")
+	TEST_R(     "uxtb16	r8, r",7,  HH1,"")
+
+	TEST_RR(    "sxtab	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "sxtab	r14,r",12, HH2,", r",10,HH1,", ror #8")
+	TEST_R(     "sxtb	r8, r",7,  HH1,"")
+
+	TEST_RR(    "uxtab	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "uxtab	r14,r",12, HH2,", r",10,HH1,", ror #8")
+	TEST_R(     "uxtb	r8, r",7,  HH1,"")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa6000f0) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa7fffff) "")
+
+#define PARALLEL_ADD_SUB(op)					\
+	TEST_RR(  op"add16	r0, r",0,  HH1,", r",1, HH2,"")	\
+	TEST_RR(  op"add16	r14, r",12,HH2,", r",10,HH1,"")	\
+	TEST_RR(  op"asx	r0, r",0,  HH1,", r",1, HH2,"")	\
+	TEST_RR(  op"asx	r14, r",12,HH2,", r",10,HH1,"")	\
+	TEST_RR(  op"sax	r0, r",0,  HH1,", r",1, HH2,"")	\
+	TEST_RR(  op"sax	r14, r",12,HH2,", r",10,HH1,"")	\
+	TEST_RR(  op"sub16	r0, r",0,  HH1,", r",1, HH2,"")	\
+	TEST_RR(  op"sub16	r14, r",12,HH2,", r",10,HH1,"")	\
+	TEST_RR(  op"add8	r0, r",0,  HH1,", r",1, HH2,"")	\
+	TEST_RR(  op"add8	r14, r",12,HH2,", r",10,HH1,"")	\
+	TEST_RR(  op"sub8	r0, r",0,  HH1,", r",1, HH2,"")	\
+	TEST_RR(  op"sub8	r14, r",12,HH2,", r",10,HH1,"")
+
+	TEST_GROUP("Parallel addition and subtraction, signed")
+
+	PARALLEL_ADD_SUB("s")
+	PARALLEL_ADD_SUB("q")
+	PARALLEL_ADD_SUB("sh")
+
+	TEST_GROUP("Parallel addition and subtraction, unsigned")
+
+	PARALLEL_ADD_SUB("u")
+	PARALLEL_ADD_SUB("uq")
+	PARALLEL_ADD_SUB("uh")
+
+	TEST_GROUP("Miscellaneous operations")
+
+	TEST_RR("qadd	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR("qadd	lr, r",9, VAL2,", r",8, VAL1,"")
+	TEST_RR("qsub	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR("qsub	lr, r",9, VAL2,", r",8, VAL1,"")
+	TEST_RR("qdadd	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR("qdadd	lr, r",9, VAL2,", r",8, VAL1,"")
+	TEST_RR("qdsub	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR("qdsub	lr, r",9, VAL2,", r",8, VAL1,"")
+
+	TEST_R("rev.w	r0, r",0,   VAL1,"")
+	TEST_R("rev	r14, r",12, VAL2,"")
+	TEST_R("rev16.w	r0, r",0,   VAL1,"")
+	TEST_R("rev16	r14, r",12, VAL2,"")
+	TEST_R("rbit	r0, r",0,   VAL1,"")
+	TEST_R("rbit	r14, r",12, VAL2,"")
+	TEST_R("revsh.w	r0, r",0,   VAL1,"")
+	TEST_R("revsh	r14, r",12, VAL2,"")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa9cff8c) "	@ rev	pc, r12");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa9cfd8c) "	@ rev	sp, r12");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa9ffe8f) "	@ rev	r14, pc");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa9dfe8d) "	@ rev	r14, sp");
+
+	TEST_RR("sel	r0, r",0,  VAL1,", r",1, VAL2,"")
+	TEST_RR("sel	r14, r",12,VAL1,", r",10, VAL2,"")
+
+	TEST_R("clz	r0, r",0, 0x0,"")
+	TEST_R("clz	r7, r",14,0x1,"")
+	TEST_R("clz	lr, r",7, 0xffffffff,"")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xfa80f030) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_thumb32(0xfaffff7f) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_thumb32(0xfab0f000) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_thumb32(0xfaffff7f) "") /* Unallocated space */
+
+	TEST_GROUP("Multiply, multiply accumulate, and absolute difference operations")
+
+	TEST_RR(    "mul	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "mul	r7, r",8, VAL2,", r",9, VAL2,"")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb08ff09) "	@ mul	pc, r8, r9")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb08fd09) "	@ mul	sp, r8, r9")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb0ff709) "	@ mul	r7, pc, r9")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb0df709) "	@ mul	r7, sp, r9")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb08f70f) "	@ mul	r7, r8, pc")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb08f70d) "	@ mul	r7, r8, sp")
+
+	TEST_RRR(   "mla	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(   "mla	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb08af09) "	@ mla	pc, r8, r9, r10");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb08ad09) "	@ mla	sp, r8, r9, r10");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb0fa709) "	@ mla	r7, pc, r9, r10");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb0da709) "	@ mla	r7, sp, r9, r10");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb08a70f) "	@ mla	r7, r8, pc, r10");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb08a70d) "	@ mla	r7, r8, sp, r10");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb08d709) "	@ mla	r7, r8, r9, sp");
+
+	TEST_RRR(   "mls	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(   "mls	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+
+	TEST_RRR(   "smlabb	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(   "smlabb	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RRR(   "smlatb	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(   "smlatb	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RRR(   "smlabt	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(   "smlabt	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RRR(   "smlatt	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(   "smlatt	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RR(    "smulbb	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "smulbb	r7, r",8, VAL3,", r",9, VAL1,"")
+	TEST_RR(    "smultb	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "smultb	r7, r",8, VAL3,", r",9, VAL1,"")
+	TEST_RR(    "smulbt	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "smulbt	r7, r",8, VAL3,", r",9, VAL1,"")
+	TEST_RR(    "smultt	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "smultt	r7, r",8, VAL3,", r",9, VAL1,"")
+
+	TEST_RRR(   "smlad	r0, r",0,  HH1,", r",1, HH2,", r",2, VAL1,"")
+	TEST_RRR(   "smlad	r14, r",12,HH2,", r",10,HH1,", r",8, VAL2,"")
+	TEST_RRR(   "smladx	r0, r",0,  HH1,", r",1, HH2,", r",2, VAL1,"")
+	TEST_RRR(   "smladx	r14, r",12,HH2,", r",10,HH1,", r",8, VAL2,"")
+	TEST_RR(    "smuad	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "smuad	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_RR(    "smuadx	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "smuadx	r14, r",12,HH2,", r",10,HH1,"")
+
+	TEST_RRR(   "smlawb	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(   "smlawb	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RRR(   "smlawt	r0, r",1, VAL1,", r",2, VAL2,", r",3,  VAL3,"")
+	TEST_RRR(   "smlawt	r7, r",8, VAL3,", r",9, VAL1,", r",10, VAL2,"")
+	TEST_RR(    "smulwb	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "smulwb	r7, r",8, VAL3,", r",9, VAL1,"")
+	TEST_RR(    "smulwt	r0, r",1, VAL1,", r",2, VAL2,"")
+	TEST_RR(    "smulwt	r7, r",8, VAL3,", r",9, VAL1,"")
+
+	TEST_RRR(   "smlsd	r0, r",0,  HH1,", r",1, HH2,", r",2, VAL1,"")
+	TEST_RRR(   "smlsd	r14, r",12,HH2,", r",10,HH1,", r",8, VAL2,"")
+	TEST_RRR(   "smlsdx	r0, r",0,  HH1,", r",1, HH2,", r",2, VAL1,"")
+	TEST_RRR(   "smlsdx	r14, r",12,HH2,", r",10,HH1,", r",8, VAL2,"")
+	TEST_RR(    "smusd	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "smusd	r14, r",12,HH2,", r",10,HH1,"")
+	TEST_RR(    "smusdx	r0, r",0,  HH1,", r",1, HH2,"")
+	TEST_RR(    "smusdx	r14, r",12,HH2,", r",10,HH1,"")
+
+	TEST_RRR(   "smmla	r0, r",0,  VAL1,", r",1, VAL2,", r",2, VAL1,"")
+	TEST_RRR(   "smmla	r14, r",12,VAL2,", r",10,VAL1,", r",8, VAL2,"")
+	TEST_RRR(   "smmlar	r0, r",0,  VAL1,", r",1, VAL2,", r",2, VAL1,"")
+	TEST_RRR(   "smmlar	r14, r",12,VAL2,", r",10,VAL1,", r",8, VAL2,"")
+	TEST_RR(    "smmul	r0, r",0,  VAL1,", r",1, VAL2,"")
+	TEST_RR(    "smmul	r14, r",12,VAL2,", r",10,VAL1,"")
+	TEST_RR(    "smmulr	r0, r",0,  VAL1,", r",1, VAL2,"")
+	TEST_RR(    "smmulr	r14, r",12,VAL2,", r",10,VAL1,"")
+
+	TEST_RRR(   "smmls	r0, r",0,  VAL1,", r",1, VAL2,", r",2, VAL1,"")
+	TEST_RRR(   "smmls	r14, r",12,VAL2,", r",10,VAL1,", r",8, VAL2,"")
+	TEST_RRR(   "smmlsr	r0, r",0,  VAL1,", r",1, VAL2,", r",2, VAL1,"")
+	TEST_RRR(   "smmlsr	r14, r",12,VAL2,", r",10,VAL1,", r",8, VAL2,"")
+
+	TEST_RRR(   "usada8	r0, r",0,  VAL1,", r",1, VAL2,", r",2, VAL3,"")
+	TEST_RRR(   "usada8	r14, r",12,VAL2,", r",10,VAL1,", r",8, VAL3,"")
+	TEST_RR(    "usad8	r0, r",0,  VAL1,", r",1, VAL2,"")
+	TEST_RR(    "usad8	r14, r",12,VAL2,", r",10,VAL1,"")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb00f010) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb0fff1f) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb70f010) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb7fff1f) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb700010) "") /* Unallocated space */
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb7fff1f) "") /* Unallocated space */
+
+	TEST_GROUP("Long multiply, long multiply accumulate, and divide")
+
+	TEST_RR(   "smull	r0, r1, r",2, VAL1,", r",3, VAL2,"")
+	TEST_RR(   "smull	r7, r8, r",9, VAL2,", r",10, VAL1,"")
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb89f80a) "	@ smull	pc, r8, r9, r10");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb89d80a) "	@ smull	sp, r8, r9, r10");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb897f0a) "	@ smull	r7, pc, r9, r10");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb897d0a) "	@ smull	r7, sp, r9, r10");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb8f780a) "	@ smull	r7, r8, pc, r10");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb8d780a) "	@ smull	r7, r8, sp, r10");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb89780f) "	@ smull	r7, r8, r9, pc");
+	TEST_UNSUPPORTED(__inst_thumb32(0xfb89780d) "	@ smull	r7, r8, r9, sp");
+
+	TEST_RR(   "umull	r0, r1, r",2, VAL1,", r",3, VAL2,"")
+	TEST_RR(   "umull	r7, r8, r",9, VAL2,", r",10, VAL1,"")
+
+	TEST_RRRR( "smlal	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR( "smlal	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+
+	TEST_RRRR( "smlalbb	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR( "smlalbb	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+	TEST_RRRR( "smlalbt	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR( "smlalbt	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+	TEST_RRRR( "smlaltb	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR( "smlaltb	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+	TEST_RRRR( "smlaltt	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR( "smlaltt	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+
+	TEST_RRRR( "smlald	r",0, VAL1,", r",1, VAL2, ", r",0, HH1,", r",1, HH2)
+	TEST_RRRR( "smlald	r",11,VAL2,", r",10,VAL1, ", r",9, HH2,", r",8, HH1)
+	TEST_RRRR( "smlaldx	r",0, VAL1,", r",1, VAL2, ", r",0, HH1,", r",1, HH2)
+	TEST_RRRR( "smlaldx	r",11,VAL2,", r",10,VAL1, ", r",9, HH2,", r",8, HH1)
+
+	TEST_RRRR( "smlsld	r",0, VAL1,", r",1, VAL2, ", r",0, HH1,", r",1, HH2)
+	TEST_RRRR( "smlsld	r",11,VAL2,", r",10,VAL1, ", r",9, HH2,", r",8, HH1)
+	TEST_RRRR( "smlsldx	r",0, VAL1,", r",1, VAL2, ", r",0, HH1,", r",1, HH2)
+	TEST_RRRR( "smlsldx	r",11,VAL2,", r",10,VAL1, ", r",9, HH2,", r",8, HH1)
+
+	TEST_RRRR( "umlal	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR( "umlal	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+	TEST_RRRR( "umaal	r",0, VAL1,", r",1, VAL2,", r",2, VAL3,", r",3, VAL4)
+	TEST_RRRR( "umaal	r",8, VAL4,", r",9, VAL1,", r",10,VAL2,", r",11,VAL3)
+
+	TEST_GROUP("Coprocessor instructions")
+
+	TEST_UNSUPPORTED(__inst_thumb32(0xfc000000) "")
+	TEST_UNSUPPORTED(__inst_thumb32(0xffffffff) "")
+
+	TEST_GROUP("Testing instructions in IT blocks")
+
+	TEST_ITBLOCK("sub.w	r0, r0")
+
+	verbose("\n");
+}
+

arch/arm/probes/uprobes/Makefile

@@ -0,0 +1 @@
+obj-$(CONFIG_UPROBES)		+= core.o actions-arm.o

arch/arm/probes/uprobes/actions-arm.c

@@ -0,0 +1,232 @@
+/*
+ * Copyright (C) 2012 Rabin Vincent <rabin at rab.in>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/stddef.h>
+#include <linux/wait.h>
+#include <linux/uprobes.h>
+#include <linux/module.h>
+
+#include "../decode.h"
+#include "../decode-arm.h"
+#include "core.h"
+
+static int uprobes_substitute_pc(unsigned long *pinsn, u32 oregs)
+{
+	probes_opcode_t insn = __mem_to_opcode_arm(*pinsn);
+	probes_opcode_t temp;
+	probes_opcode_t mask;
+	int freereg;
+	u32 free = 0xffff;
+	u32 regs;
+
+	for (regs = oregs; regs; regs >>= 4, insn >>= 4) {
+		if ((regs & 0xf) == REG_TYPE_NONE)
+			continue;
+
+		free &= ~(1 << (insn & 0xf));
+	}
+
+	/* No PC, no problem */
+	if (free & (1 << 15))
+		return 15;
+
+	if (!free)
+		return -1;
+
+	/*
+	 * fls instead of ffs ensures that for "ldrd r0, r1, [pc]" we would
+	 * pick LR instead of R1.
+	 */
+	freereg = free = fls(free) - 1;
+
+	temp = __mem_to_opcode_arm(*pinsn);
+	insn = temp;
+	regs = oregs;
+	mask = 0xf;
+
+	for (; regs; regs >>= 4, mask <<= 4, free <<= 4, temp >>= 4) {
+		if ((regs & 0xf) == REG_TYPE_NONE)
+			continue;
+
+		if ((temp & 0xf) != 15)
+			continue;
+
+		insn &= ~mask;
+		insn |= free & mask;
+	}
+
+	*pinsn = __opcode_to_mem_arm(insn);
+	return freereg;
+}
+
+static void uprobe_set_pc(struct arch_uprobe *auprobe,
+			  struct arch_uprobe_task *autask,
+			  struct pt_regs *regs)
+{
+	u32 pcreg = auprobe->pcreg;
+
+	autask->backup = regs->uregs[pcreg];
+	regs->uregs[pcreg] = regs->ARM_pc + 8;
+}
+
+static void uprobe_unset_pc(struct arch_uprobe *auprobe,
+			    struct arch_uprobe_task *autask,
+			    struct pt_regs *regs)
+{
+	/* PC will be taken care of by common code */
+	regs->uregs[auprobe->pcreg] = autask->backup;
+}
+
+static void uprobe_aluwrite_pc(struct arch_uprobe *auprobe,
+			       struct arch_uprobe_task *autask,
+			       struct pt_regs *regs)
+{
+	u32 pcreg = auprobe->pcreg;
+
+	alu_write_pc(regs->uregs[pcreg], regs);
+	regs->uregs[pcreg] = autask->backup;
+}
+
+static void uprobe_write_pc(struct arch_uprobe *auprobe,
+			    struct arch_uprobe_task *autask,
+			    struct pt_regs *regs)
+{
+	u32 pcreg = auprobe->pcreg;
+
+	load_write_pc(regs->uregs[pcreg], regs);
+	regs->uregs[pcreg] = autask->backup;
+}
+
+enum probes_insn
+decode_pc_ro(probes_opcode_t insn, struct arch_probes_insn *asi,
+	     const struct decode_header *d)
+{
+	struct arch_uprobe *auprobe = container_of(asi, struct arch_uprobe,
+						   asi);
+	struct decode_emulate *decode = (struct decode_emulate *) d;
+	u32 regs = decode->header.type_regs.bits >> DECODE_TYPE_BITS;
+	int reg;
+
+	reg = uprobes_substitute_pc(&auprobe->ixol[0], regs);
+	if (reg == 15)
+		return INSN_GOOD;
+
+	if (reg == -1)
+		return INSN_REJECTED;
+
+	auprobe->pcreg = reg;
+	auprobe->prehandler = uprobe_set_pc;
+	auprobe->posthandler = uprobe_unset_pc;
+
+	return INSN_GOOD;
+}
+
+enum probes_insn
+decode_wb_pc(probes_opcode_t insn, struct arch_probes_insn *asi,
+	     const struct decode_header *d, bool alu)
+{
+	struct arch_uprobe *auprobe = container_of(asi, struct arch_uprobe,
+						   asi);
+	enum probes_insn ret = decode_pc_ro(insn, asi, d);
+
+	if (((insn >> 12) & 0xf) == 15)
+		auprobe->posthandler = alu ? uprobe_aluwrite_pc
+					   : uprobe_write_pc;
+
+	return ret;
+}
+
+enum probes_insn
+decode_rd12rn16rm0rs8_rwflags(probes_opcode_t insn,
+			      struct arch_probes_insn *asi,
+			      const struct decode_header *d)
+{
+	return decode_wb_pc(insn, asi, d, true);
+}
+
+enum probes_insn
+decode_ldr(probes_opcode_t insn, struct arch_probes_insn *asi,
+	   const struct decode_header *d)
+{
+	return decode_wb_pc(insn, asi, d, false);
+}
+
+enum probes_insn
+uprobe_decode_ldmstm(probes_opcode_t insn,
+		     struct arch_probes_insn *asi,
+		     const struct decode_header *d)
+{
+	struct arch_uprobe *auprobe = container_of(asi, struct arch_uprobe,
+						   asi);
+	unsigned reglist = insn & 0xffff;
+	int rn = (insn >> 16) & 0xf;
+	int lbit = insn & (1 << 20);
+	unsigned used = reglist | (1 << rn);
+
+	if (rn == 15)
+		return INSN_REJECTED;
+
+	if (!(used & (1 << 15)))
+		return INSN_GOOD;
+
+	if (used & (1 << 14))
+		return INSN_REJECTED;
+
+	/* Use LR instead of PC */
+	insn ^= 0xc000;
+
+	auprobe->pcreg = 14;
+	auprobe->ixol[0] = __opcode_to_mem_arm(insn);
+
+	auprobe->prehandler = uprobe_set_pc;
+	if (lbit)
+		auprobe->posthandler = uprobe_write_pc;
+	else
+		auprobe->posthandler = uprobe_unset_pc;
+
+	return INSN_GOOD;
+}
+
+const union decode_action uprobes_probes_actions[] = {
+	[PROBES_PRELOAD_IMM] = {.handler = probes_simulate_nop},
+	[PROBES_PRELOAD_REG] = {.handler = probes_simulate_nop},
+	[PROBES_BRANCH_IMM] = {.handler = simulate_blx1},
+	[PROBES_MRS] = {.handler = simulate_mrs},
+	[PROBES_BRANCH_REG] = {.handler = simulate_blx2bx},
+	[PROBES_CLZ] = {.handler = probes_simulate_nop},
+	[PROBES_SATURATING_ARITHMETIC] = {.handler = probes_simulate_nop},
+	[PROBES_MUL1] = {.handler = probes_simulate_nop},
+	[PROBES_MUL2] = {.handler = probes_simulate_nop},
+	[PROBES_SWP] = {.handler = probes_simulate_nop},
+	[PROBES_LDRSTRD] = {.decoder = decode_pc_ro},
+	[PROBES_LOAD_EXTRA] = {.decoder = decode_pc_ro},
+	[PROBES_LOAD] = {.decoder = decode_ldr},
+	[PROBES_STORE_EXTRA] = {.decoder = decode_pc_ro},
+	[PROBES_STORE] = {.decoder = decode_pc_ro},
+	[PROBES_MOV_IP_SP] = {.handler = simulate_mov_ipsp},
+	[PROBES_DATA_PROCESSING_REG] = {
+		.decoder = decode_rd12rn16rm0rs8_rwflags},
+	[PROBES_DATA_PROCESSING_IMM] = {
+		.decoder = decode_rd12rn16rm0rs8_rwflags},
+	[PROBES_MOV_HALFWORD] = {.handler = probes_simulate_nop},
+	[PROBES_SEV] = {.handler = probes_simulate_nop},
+	[PROBES_WFE] = {.handler = probes_simulate_nop},
+	[PROBES_SATURATE] = {.handler = probes_simulate_nop},
+	[PROBES_REV] = {.handler = probes_simulate_nop},
+	[PROBES_MMI] = {.handler = probes_simulate_nop},
+	[PROBES_PACK] = {.handler = probes_simulate_nop},
+	[PROBES_EXTEND] = {.handler = probes_simulate_nop},
+	[PROBES_EXTEND_ADD] = {.handler = probes_simulate_nop},
+	[PROBES_MUL_ADD_LONG] = {.handler = probes_simulate_nop},
+	[PROBES_MUL_ADD] = {.handler = probes_simulate_nop},
+	[PROBES_BITFIELD] = {.handler = probes_simulate_nop},
+	[PROBES_BRANCH] = {.handler = simulate_bbl},
+	[PROBES_LDMSTM] = {.decoder = uprobe_decode_ldmstm}
+};

arch/arm/probes/uprobes/core.c

@@ -0,0 +1,230 @@
+/*
+ * Copyright (C) 2012 Rabin Vincent <rabin at rab.in>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/stddef.h>
+#include <linux/errno.h>
+#include <linux/highmem.h>
+#include <linux/sched.h>
+#include <linux/uprobes.h>
+#include <linux/notifier.h>
+
+#include <asm/opcodes.h>
+#include <asm/traps.h>
+
+#include "../decode.h"
+#include "../decode-arm.h"
+#include "core.h"
+
+#define UPROBE_TRAP_NR	UINT_MAX
+
+bool is_swbp_insn(uprobe_opcode_t *insn)
+{
+	return (__mem_to_opcode_arm(*insn) & 0x0fffffff) ==
+		(UPROBE_SWBP_ARM_INSN & 0x0fffffff);
+}
+
+int set_swbp(struct arch_uprobe *auprobe, struct mm_struct *mm,
+	     unsigned long vaddr)
+{
+	return uprobe_write_opcode(mm, vaddr,
+		   __opcode_to_mem_arm(auprobe->bpinsn));
+}
+
+bool arch_uprobe_ignore(struct arch_uprobe *auprobe, struct pt_regs *regs)
+{
+	if (!auprobe->asi.insn_check_cc(regs->ARM_cpsr)) {
+		regs->ARM_pc += 4;
+		return true;
+	}
+
+	return false;
+}
+
+bool arch_uprobe_skip_sstep(struct arch_uprobe *auprobe, struct pt_regs *regs)
+{
+	probes_opcode_t opcode;
+
+	if (!auprobe->simulate)
+		return false;
+
+	opcode = __mem_to_opcode_arm(*(unsigned int *) auprobe->insn);
+
+	auprobe->asi.insn_singlestep(opcode, &auprobe->asi, regs);
+
+	return true;
+}
+
+unsigned long
+arch_uretprobe_hijack_return_addr(unsigned long trampoline_vaddr,
+				  struct pt_regs *regs)
+{
+	unsigned long orig_ret_vaddr;
+
+	orig_ret_vaddr = regs->ARM_lr;
+	/* Replace the return addr with trampoline addr */
+	regs->ARM_lr = trampoline_vaddr;
+	return orig_ret_vaddr;
+}
+
+int arch_uprobe_analyze_insn(struct arch_uprobe *auprobe, struct mm_struct *mm,
+			     unsigned long addr)
+{
+	unsigned int insn;
+	unsigned int bpinsn;
+	enum probes_insn ret;
+
+	/* Thumb not yet support */
+	if (addr & 0x3)
+		return -EINVAL;
+
+	insn = __mem_to_opcode_arm(*(unsigned int *)auprobe->insn);
+	auprobe->ixol[0] = __opcode_to_mem_arm(insn);
+	auprobe->ixol[1] = __opcode_to_mem_arm(UPROBE_SS_ARM_INSN);
+
+	ret = arm_probes_decode_insn(insn, &auprobe->asi, false,
+				     uprobes_probes_actions, NULL);
+	switch (ret) {
+	case INSN_REJECTED:
+		return -EINVAL;
+
+	case INSN_GOOD_NO_SLOT:
+		auprobe->simulate = true;
+		break;
+
+	case INSN_GOOD:
+	default:
+		break;
+	}
+
+	bpinsn = UPROBE_SWBP_ARM_INSN & 0x0fffffff;
+	if (insn >= 0xe0000000)
+		bpinsn |= 0xe0000000;  /* Unconditional instruction */
+	else
+		bpinsn |= insn & 0xf0000000;  /* Copy condition from insn */
+
+	auprobe->bpinsn = bpinsn;
+
+	return 0;
+}
+
+void arch_uprobe_copy_ixol(struct page *page, unsigned long vaddr,
+			   void *src, unsigned long len)
+{
+	void *xol_page_kaddr = kmap_atomic(page);
+	void *dst = xol_page_kaddr + (vaddr & ~PAGE_MASK);
+
+	preempt_disable();
+
+	/* Initialize the slot */
+	memcpy(dst, src, len);
+
+	/* flush caches (dcache/icache) */
+	flush_uprobe_xol_access(page, vaddr, dst, len);
+
+	preempt_enable();
+
+	kunmap_atomic(xol_page_kaddr);
+}
+
+
+int arch_uprobe_pre_xol(struct arch_uprobe *auprobe, struct pt_regs *regs)
+{
+	struct uprobe_task *utask = current->utask;
+
+	if (auprobe->prehandler)
+		auprobe->prehandler(auprobe, &utask->autask, regs);
+
+	utask->autask.saved_trap_no = current->thread.trap_no;
+	current->thread.trap_no = UPROBE_TRAP_NR;
+	regs->ARM_pc = utask->xol_vaddr;
+
+	return 0;
+}
+
+int arch_uprobe_post_xol(struct arch_uprobe *auprobe, struct pt_regs *regs)
+{
+	struct uprobe_task *utask = current->utask;
+
+	WARN_ON_ONCE(current->thread.trap_no != UPROBE_TRAP_NR);
+
+	current->thread.trap_no = utask->autask.saved_trap_no;
+	regs->ARM_pc = utask->vaddr + 4;
+
+	if (auprobe->posthandler)
+		auprobe->posthandler(auprobe, &utask->autask, regs);
+
+	return 0;
+}
+
+bool arch_uprobe_xol_was_trapped(struct task_struct *t)
+{
+	if (t->thread.trap_no != UPROBE_TRAP_NR)
+		return true;
+
+	return false;
+}
+
+void arch_uprobe_abort_xol(struct arch_uprobe *auprobe, struct pt_regs *regs)
+{
+	struct uprobe_task *utask = current->utask;
+
+	current->thread.trap_no = utask->autask.saved_trap_no;
+	instruction_pointer_set(regs, utask->vaddr);
+}
+
+int arch_uprobe_exception_notify(struct notifier_block *self,
+				 unsigned long val, void *data)
+{
+	return NOTIFY_DONE;
+}
+
+static int uprobe_trap_handler(struct pt_regs *regs, unsigned int instr)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	instr &= 0x0fffffff;
+	if (instr == (UPROBE_SWBP_ARM_INSN & 0x0fffffff))
+		uprobe_pre_sstep_notifier(regs);
+	else if (instr == (UPROBE_SS_ARM_INSN & 0x0fffffff))
+		uprobe_post_sstep_notifier(regs);
+	local_irq_restore(flags);
+
+	return 0;
+}
+
+unsigned long uprobe_get_swbp_addr(struct pt_regs *regs)
+{
+	return instruction_pointer(regs);
+}
+
+static struct undef_hook uprobes_arm_break_hook = {
+	.instr_mask	= 0x0fffffff,
+	.instr_val	= (UPROBE_SWBP_ARM_INSN & 0x0fffffff),
+	.cpsr_mask	= MODE_MASK,
+	.cpsr_val	= USR_MODE,
+	.fn		= uprobe_trap_handler,
+};
+
+static struct undef_hook uprobes_arm_ss_hook = {
+	.instr_mask	= 0x0fffffff,
+	.instr_val	= (UPROBE_SS_ARM_INSN & 0x0fffffff),
+	.cpsr_mask	= MODE_MASK,
+	.cpsr_val	= USR_MODE,
+	.fn		= uprobe_trap_handler,
+};
+
+static int arch_uprobes_init(void)
+{
+	register_undef_hook(&uprobes_arm_break_hook);
+	register_undef_hook(&uprobes_arm_ss_hook);
+
+	return 0;
+}
+device_initcall(arch_uprobes_init);

arch/arm/probes/uprobes/core.h

@@ -0,0 +1,35 @@
+/*
+ * Copyright (C) 2012 Rabin Vincent <rabin at rab.in>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef __ARM_KERNEL_UPROBES_H
+#define __ARM_KERNEL_UPROBES_H
+
+enum probes_insn uprobe_decode_ldmstm(probes_opcode_t insn,
+				      struct arch_probes_insn *asi,
+				      const struct decode_header *d);
+
+enum probes_insn decode_ldr(probes_opcode_t insn,
+			    struct arch_probes_insn *asi,
+			    const struct decode_header *d);
+
+enum probes_insn
+decode_rd12rn16rm0rs8_rwflags(probes_opcode_t insn,
+			      struct arch_probes_insn *asi,
+			      const struct decode_header *d);
+
+enum probes_insn
+decode_wb_pc(probes_opcode_t insn, struct arch_probes_insn *asi,
+	     const struct decode_header *d, bool alu);
+
+enum probes_insn
+decode_pc_ro(probes_opcode_t insn, struct arch_probes_insn *asi,
+	     const struct decode_header *d);
+
+extern const union decode_action uprobes_probes_actions[];
+
+#endif

arch/arm/tools/Makefile

@@ -0,0 +1,13 @@
+#
+# linux/arch/arm/tools/Makefile
+#
+# Copyright (C) 2001 Russell King
+#
+
+quiet_cmd_gen_mach = GEN     $@
+      cmd_gen_mach = mkdir -p $(dir $@) && \
+		     $(AWK) -f $(filter-out $(PHONY),$^) > $@ || \
+		     { rm -f $@; /bin/false; }
+
+include/generated/mach-types.h: $(src)/gen-mach-types $(src)/mach-types FORCE
+	$(call if_changed,gen_mach)

arch/arm/tools/gen-mach-types

@@ -0,0 +1,72 @@
+#!/bin/awk
+#
+# Awk script to generate include/generated/mach-types.h
+#
+BEGIN	{ nr = 0 }
+/^#/	{ next }
+/^[ 	]*$/ { next }
+
+NF == 4 {
+	  machine_is[nr] = "machine_is_"$1;
+	  config[nr] = "CONFIG_"$2;
+	  mach_type[nr] = "MACH_TYPE_"$3;
+	  num[nr] = $4; nr++
+	}
+
+NF == 3 {
+	  machine_is[nr] = "machine_is_"$1;
+	  config[nr] = "CONFIG_"$2;
+	  mach_type[nr] = "MACH_TYPE_"$3;
+	  num[nr] = ""; nr++
+	}
+
+
+END	{
+	  printf("/*\n");
+	  printf(" * This was automagically generated from %s!\n", FILENAME);
+	  printf(" * Do NOT edit\n");
+	  printf(" */\n\n");
+	  printf("#ifndef __ASM_ARM_MACH_TYPE_H\n");
+	  printf("#define __ASM_ARM_MACH_TYPE_H\n\n");
+	  printf("#ifndef __ASSEMBLY__\n");
+	  printf("/* The type of machine we're running on */\n");
+	  printf("extern unsigned int __machine_arch_type;\n");
+	  printf("#endif\n\n");
+
+	  printf("/* see arch/arm/kernel/arch.c for a description of these */\n");
+	  for (i = 0; i < nr; i++)
+	    if (num[i] ~ /..*/)
+	      printf("#define %-30s %d\n", mach_type[i], num[i]);
+
+	  printf("\n");
+
+	  for (i = 0; i < nr; i++)
+	    if (num[i] ~ /..*/) {
+	      printf("#ifdef %s\n", config[i]);
+	      printf("# ifdef machine_arch_type\n");
+	      printf("#  undef machine_arch_type\n");
+	      printf("#  define machine_arch_type\t__machine_arch_type\n");
+	      printf("# else\n");
+	      printf("#  define machine_arch_type\t%s\n", mach_type[i]);
+	      printf("# endif\n");
+	      printf("# define %s()\t(machine_arch_type == %s)\n", machine_is[i], mach_type[i]);
+	      printf("#else\n");
+	      printf("# define %s()\t(0)\n", machine_is[i]);
+	      printf("#endif\n\n");
+	    }
+
+	  printf("/*\n * These have not yet been registered\n */\n");
+	  for (i = 0; i < nr; i++)
+	    if (num[i] !~ /..*/)
+	      printf("/* #define %-30s <<not registered>> */\n", mach_type[i]);
+
+	  for (i = 0; i < nr; i++)
+	    if (num[i] !~ /..*/) {
+	      printf("#define %s()\t(0)\n", machine_is[i]);
+	    }
+
+	  printf("\n#ifndef machine_arch_type\n");
+	  printf("#define machine_arch_type\t__machine_arch_type\n");
+	  printf("#endif\n\n");
+	  printf("#endif\n");
+	}

arch/arm/tools/mach-types

@@ -0,0 +1,1008 @@
+# Database of machine macros and numbers
+#
+# This file is linux/arch/arm/tools/mach-types
+#
+# Up to date versions of this file can be obtained from:
+#
+#   http://www.arm.linux.org.uk/developer/machines/download.php
+#
+# Please do not send patches to this file; it is automatically generated!
+# To add an entry into this database, please see Documentation/arm/README,
+# or visit:
+#
+#   http://www.arm.linux.org.uk/developer/machines/?action=new
+#
+# This is a cut-down version of the file; it contains only machines that
+# are merged into mainline or have been edited in the machine database
+# within the last 12 months.  References to machine_is_NAME() do not count!
+#
+# Last update: Fri Mar 22 17:24:50 2013
+#
+# machine_is_xxx	CONFIG_xxxx		MACH_TYPE_xxx		number
+#
+ebsa110			ARCH_EBSA110		EBSA110			0
+riscpc			ARCH_RPC		RISCPC			1
+ebsa285			ARCH_EBSA285		EBSA285			4
+netwinder		ARCH_NETWINDER		NETWINDER		5
+cats			ARCH_CATS		CATS			6
+shark			ARCH_SHARK		SHARK			15
+brutus			SA1100_BRUTUS		BRUTUS			16
+personal_server		ARCH_PERSONAL_SERVER	PERSONAL_SERVER		17
+l7200			ARCH_L7200		L7200			19
+pleb			SA1100_PLEB		PLEB			20
+integrator		ARCH_INTEGRATOR		INTEGRATOR		21
+h3600			SA1100_H3600		H3600			22
+p720t			ARCH_P720T		P720T			24
+assabet			SA1100_ASSABET		ASSABET			25
+lart			SA1100_LART		LART			27
+graphicsclient		SA1100_GRAPHICSCLIENT	GRAPHICSCLIENT		29
+xp860			SA1100_XP860		XP860			30
+cerf			SA1100_CERF		CERF			31
+nanoengine		SA1100_NANOENGINE	NANOENGINE		32
+jornada720		SA1100_JORNADA720	JORNADA720		48
+edb7211			ARCH_EDB7211		EDB7211			50
+pfs168			SA1100_PFS168		PFS168			52
+flexanet		SA1100_FLEXANET		FLEXANET		54
+simpad			SA1100_SIMPAD		SIMPAD			87
+lubbock			ARCH_LUBBOCK		LUBBOCK			89
+clep7212		ARCH_CLEP7212		CLEP7212		91
+shannon			SA1100_SHANNON		SHANNON			97
+consus			SA1100_CONSUS		CONSUS			105
+aaed2000		ARCH_AAED2000		AAED2000		106
+cdb89712		ARCH_CDB89712		CDB89712		107
+graphicsmaster		SA1100_GRAPHICSMASTER	GRAPHICSMASTER		108
+adsbitsy		SA1100_ADSBITSY		ADSBITSY		109
+pxa_idp			ARCH_PXA_IDP		PXA_IDP			110
+pt_system3		SA1100_PT_SYSTEM3	PT_SYSTEM3		112
+autcpu12		ARCH_AUTCPU12		AUTCPU12		118
+h3100			SA1100_H3100		H3100			136
+collie			SA1100_COLLIE		COLLIE			146
+badge4			SA1100_BADGE4		BADGE4			148
+fortunet		ARCH_FORTUNET		FORTUNET		152
+mx1ads			ARCH_MX1ADS		MX1ADS			160
+h7201			ARCH_H7201		H7201			161
+h7202			ARCH_H7202		H7202			162
+iq80321			ARCH_IQ80321		IQ80321			169
+ks8695			ARCH_KS8695		KS8695			180
+smdk2410		ARCH_SMDK2410		SMDK2410		193
+ceiva			ARCH_CEIVA		CEIVA			200
+voiceblue		MACH_VOICEBLUE		VOICEBLUE		218
+h5400			ARCH_H5400		H5400			220
+omap_innovator		MACH_OMAP_INNOVATOR	OMAP_INNOVATOR		234
+ixdp2400		ARCH_IXDP2400		IXDP2400		242
+ixdp2800		ARCH_IXDP2800		IXDP2800		243
+ixdp425			ARCH_IXDP425		IXDP425			245
+hackkit			SA1100_HACKKIT		HACKKIT			254
+ixcdp1100		ARCH_IXCDP1100		IXCDP1100		260
+at91rm9200dk		ARCH_AT91RM9200DK	AT91RM9200DK		262
+cintegrator		ARCH_CINTEGRATOR	CINTEGRATOR		275
+viper			ARCH_VIPER		VIPER			283
+adi_coyote		ARCH_ADI_COYOTE		ADI_COYOTE		290
+ixdp2401		ARCH_IXDP2401		IXDP2401		299
+ixdp2801		ARCH_IXDP2801		IXDP2801		300
+iq31244			ARCH_IQ31244		IQ31244			327
+bast			ARCH_BAST		BAST			331
+h1940			ARCH_H1940		H1940			347
+enp2611			ARCH_ENP2611		ENP2611			356
+s3c2440			ARCH_S3C2440		S3C2440			362
+gumstix			ARCH_GUMSTIX		GUMSTIX			373
+omap_h2			MACH_OMAP_H2		OMAP_H2			382
+e740			MACH_E740		E740			384
+iq80331			ARCH_IQ80331		IQ80331			385
+versatile_pb		ARCH_VERSATILE_PB	VERSATILE_PB		387
+kev7a400		MACH_KEV7A400		KEV7A400		388
+lpd7a400		MACH_LPD7A400		LPD7A400		389
+lpd7a404		MACH_LPD7A404		LPD7A404		390
+csb337			MACH_CSB337		CSB337			399
+mainstone		MACH_MAINSTONE		MAINSTONE		406
+lite300			MACH_LITE300		LITE300			408
+xcep			MACH_XCEP		XCEP			413
+arcom_vulcan		MACH_ARCOM_VULCAN	ARCOM_VULCAN		414
+nomadik			MACH_NOMADIK		NOMADIK			420
+corgi			MACH_CORGI		CORGI			423
+poodle			MACH_POODLE		POODLE			424
+armcore			MACH_ARMCORE		ARMCORE			438
+mx31ads			MACH_MX31ADS		MX31ADS			447
+himalaya		MACH_HIMALAYA		HIMALAYA		448
+edb9312			MACH_EDB9312		EDB9312			451
+omap_generic		MACH_OMAP_GENERIC	OMAP_GENERIC		452
+edb9301			MACH_EDB9301		EDB9301			462
+edb9315			MACH_EDB9315		EDB9315			463
+vr1000			MACH_VR1000		VR1000			475
+omap_perseus2		MACH_OMAP_PERSEUS2	OMAP_PERSEUS2		491
+e800			MACH_E800		E800			496
+e750			MACH_E750		E750			497
+scb9328			MACH_SCB9328		SCB9328			508
+omap_h3			MACH_OMAP_H3		OMAP_H3			509
+omap_h4			MACH_OMAP_H4		OMAP_H4			510
+omap_osk		MACH_OMAP_OSK		OMAP_OSK		515
+tosa			MACH_TOSA		TOSA			520
+avila			MACH_AVILA		AVILA			526
+edb9302			MACH_EDB9302		EDB9302			538
+husky			MACH_HUSKY		HUSKY			543
+shepherd		MACH_SHEPHERD		SHEPHERD		545
+h4700			MACH_H4700		H4700			562
+rx3715			MACH_RX3715		RX3715			592
+nslu2			MACH_NSLU2		NSLU2			597
+e400			MACH_E400		E400			598
+ixdpg425		MACH_IXDPG425		IXDPG425		604
+versatile_ab		MACH_VERSATILE_AB	VERSATILE_AB		606
+edb9307			MACH_EDB9307		EDB9307			607
+kb9200			MACH_KB9200		KB9200			612
+sx1			MACH_SX1		SX1			613
+ixdp465			MACH_IXDP465		IXDP465			618
+ixdp2351		MACH_IXDP2351		IXDP2351		619
+cm4008			MACH_CM4008		CM4008			624
+iq80332			MACH_IQ80332		IQ80332			629
+gtwx5715		MACH_GTWX5715		GTWX5715		641
+csb637			MACH_CSB637		CSB637			648
+n30			MACH_N30		N30			656
+nec_mp900		MACH_NEC_MP900		NEC_MP900		659
+kafa			MACH_KAFA		KAFA			662
+cm41xx			MACH_CM41XX		CM41XX			672
+ts72xx			MACH_TS72XX		TS72XX			673
+otom			MACH_OTOM		OTOM			680
+nexcoder_2440		MACH_NEXCODER_2440	NEXCODER_2440		681
+eco920			MACH_ECO920		ECO920			702
+roadrunner		MACH_ROADRUNNER		ROADRUNNER		704
+at91rm9200ek		MACH_AT91RM9200EK	AT91RM9200EK		705
+spitz			MACH_SPITZ		SPITZ			713
+adssphere		MACH_ADSSPHERE		ADSSPHERE		723
+colibri			MACH_COLIBRI		COLIBRI			729
+gateway7001		MACH_GATEWAY7001	GATEWAY7001		731
+pcm027			MACH_PCM027		PCM027			732
+anubis			MACH_ANUBIS		ANUBIS			734
+xboardgp8		MACH_XBOARDGP8		XBOARDGP8		742
+akita			MACH_AKITA		AKITA			744
+e330			MACH_E330		E330			753
+nokia770		MACH_NOKIA770		NOKIA770		755
+carmeva			MACH_CARMEVA		CARMEVA			769
+edb9315a		MACH_EDB9315A		EDB9315A		772
+stargate2		MACH_STARGATE2		STARGATE2		774
+intelmote2		MACH_INTELMOTE2		INTELMOTE2		775
+trizeps4		MACH_TRIZEPS4		TRIZEPS4		776
+pnx4008			MACH_PNX4008		PNX4008			782
+cpuat91			MACH_CPUAT91		CPUAT91			787
+iq81340sc		MACH_IQ81340SC		IQ81340SC		799
+iq81340mc		MACH_IQ81340MC		IQ81340MC		801
+se4200			MACH_SE4200		SE4200			809
+micro9			MACH_MICRO9		MICRO9			811
+micro9l			MACH_MICRO9L		MICRO9L			812
+omap_palmte		MACH_OMAP_PALMTE	OMAP_PALMTE		817
+realview_eb		MACH_REALVIEW_EB	REALVIEW_EB		827
+borzoi			MACH_BORZOI		BORZOI			831
+palmld			MACH_PALMLD		PALMLD			835
+ixdp28x5		MACH_IXDP28X5		IXDP28X5		838
+omap_palmtt		MACH_OMAP_PALMTT	OMAP_PALMTT		839
+arcom_zeus		MACH_ARCOM_ZEUS		ARCOM_ZEUS		841
+osiris			MACH_OSIRIS		OSIRIS			842
+palmte2			MACH_PALMTE2		PALMTE2			844
+mx27ads			MACH_MX27ADS		MX27ADS			846
+at91sam9261ek		MACH_AT91SAM9261EK	AT91SAM9261EK		848
+loft			MACH_LOFT		LOFT			849
+mx21ads			MACH_MX21ADS		MX21ADS			851
+ams_delta		MACH_AMS_DELTA		AMS_DELTA		862
+nas100d			MACH_NAS100D		NAS100D			865
+magician		MACH_MAGICIAN		MAGICIAN		875
+cm4002			MACH_CM4002		CM4002			876
+nxdkn			MACH_NXDKN		NXDKN			880
+palmtx			MACH_PALMTX		PALMTX			885
+s3c2413			MACH_S3C2413		S3C2413			887
+wg302v2			MACH_WG302V2		WG302V2			890
+omap_2430sdp		MACH_OMAP_2430SDP	OMAP_2430SDP		900
+davinci_evm		MACH_DAVINCI_EVM	DAVINCI_EVM		901
+palmz72			MACH_PALMZ72		PALMZ72			904
+nxdb500			MACH_NXDB500		NXDB500			905
+apf9328			MACH_APF9328		APF9328			906
+palmt5			MACH_PALMT5		PALMT5			917
+palmtc			MACH_PALMTC		PALMTC			918
+omap_apollon		MACH_OMAP_APOLLON	OMAP_APOLLON		919
+ateb9200		MACH_ATEB9200		ATEB9200		923
+n35			MACH_N35		N35			927
+logicpd_pxa270		MACH_LOGICPD_PXA270	LOGICPD_PXA270		930
+nxeb500hmi		MACH_NXEB500HMI		NXEB500HMI		941
+espresso		MACH_ESPRESSO		ESPRESSO		949
+rx1950			MACH_RX1950		RX1950			952
+gesbc9312		MACH_GESBC9312		GESBC9312		958
+picotux2xx		MACH_PICOTUX2XX		PICOTUX2XX		963
+dsmg600			MACH_DSMG600		DSMG600			964
+omap_fsample		MACH_OMAP_FSAMPLE	OMAP_FSAMPLE		970
+snapper_cl15		MACH_SNAPPER_CL15	SNAPPER_CL15		986
+omap_palmz71		MACH_OMAP_PALMZ71	OMAP_PALMZ71		993
+smdk2412		MACH_SMDK2412		SMDK2412		1009
+smdk2413		MACH_SMDK2413		SMDK2413		1022
+aml_m5900		MACH_AML_M5900		AML_M5900		1024
+balloon3		MACH_BALLOON3		BALLOON3		1029
+ecbat91			MACH_ECBAT91		ECBAT91			1072
+onearm			MACH_ONEARM		ONEARM			1075
+smdk2443		MACH_SMDK2443		SMDK2443		1084
+fsg			MACH_FSG		FSG			1091
+at91sam9260ek		MACH_AT91SAM9260EK	AT91SAM9260EK		1099
+glantank		MACH_GLANTANK		GLANTANK		1100
+n2100			MACH_N2100		N2100			1101
+im42xx			MACH_IM42XX		IM42XX			1105
+qt2410			MACH_QT2410		QT2410			1108
+kixrp435		MACH_KIXRP435		KIXRP435		1109
+cc9p9360dev		MACH_CC9P9360DEV	CC9P9360DEV		1114
+edb9302a		MACH_EDB9302A		EDB9302A		1127
+edb9307a		MACH_EDB9307A		EDB9307A		1128
+omap_3430sdp		MACH_OMAP_3430SDP	OMAP_3430SDP		1138
+vstms			MACH_VSTMS		VSTMS			1140
+micro9m			MACH_MICRO9M		MICRO9M			1169
+bug			MACH_BUG		BUG			1179
+at91sam9263ek		MACH_AT91SAM9263EK	AT91SAM9263EK		1202
+em7210			MACH_EM7210		EM7210			1212
+vpac270			MACH_VPAC270		VPAC270			1227
+treo680			MACH_TREO680		TREO680			1230
+zylonite		MACH_ZYLONITE		ZYLONITE		1233
+mx31lite		MACH_MX31LITE		MX31LITE		1236
+mioa701			MACH_MIOA701		MIOA701			1257
+armadillo5x0		MACH_ARMADILLO5X0	ARMADILLO5X0		1260
+cc9p9360js		MACH_CC9P9360JS		CC9P9360JS		1264
+smdk6400		MACH_SMDK6400		SMDK6400		1270
+nokia_n800		MACH_NOKIA_N800		NOKIA_N800		1271
+ep80219			MACH_EP80219		EP80219			1281
+goramo_mlr		MACH_GORAMO_MLR		GORAMO_MLR		1292
+em_x270			MACH_EM_X270		EM_X270			1297
+neo1973_gta02		MACH_NEO1973_GTA02	NEO1973_GTA02		1304
+at91sam9rlek		MACH_AT91SAM9RLEK	AT91SAM9RLEK		1326
+colibri320		MACH_COLIBRI320		COLIBRI320		1340
+cam60			MACH_CAM60		CAM60			1351
+at91eb01		MACH_AT91EB01		AT91EB01		1354
+db88f5281		MACH_DB88F5281		DB88F5281		1358
+csb726			MACH_CSB726		CSB726			1359
+davinci_dm6467_evm	MACH_DAVINCI_DM6467_EVM	DAVINCI_DM6467_EVM	1380
+davinci_dm355_evm	MACH_DAVINCI_DM355_EVM	DAVINCI_DM355_EVM	1381
+littleton		MACH_LITTLETON		LITTLETON		1388
+im4004			MACH_IM4004		IM4004			1400
+realview_pb11mp		MACH_REALVIEW_PB11MP	REALVIEW_PB11MP		1407
+mx27_3ds		MACH_MX27_3DS		MX27_3DS		1430
+halibut			MACH_HALIBUT		HALIBUT			1439
+trout			MACH_TROUT		TROUT			1440
+tct_hammer		MACH_TCT_HAMMER		TCT_HAMMER		1460
+herald			MACH_HERALD		HERALD			1461
+sim_one			MACH_SIM_ONE		SIM_ONE			1476
+jive			MACH_JIVE		JIVE			1490
+sam9_l9260		MACH_SAM9_L9260		SAM9_L9260		1501
+realview_pb1176		MACH_REALVIEW_PB1176	REALVIEW_PB1176		1504
+yl9200			MACH_YL9200		YL9200			1507
+rd88f5182		MACH_RD88F5182		RD88F5182		1508
+kurobox_pro		MACH_KUROBOX_PRO	KUROBOX_PRO		1509
+mx31_3ds		MACH_MX31_3DS		MX31_3DS		1511
+qong			MACH_QONG		QONG			1524
+omap2evm		MACH_OMAP2EVM		OMAP2EVM		1534
+omap3evm		MACH_OMAP3EVM		OMAP3EVM		1535
+dns323			MACH_DNS323		DNS323			1542
+omap3_beagle		MACH_OMAP3_BEAGLE	OMAP3_BEAGLE		1546
+nokia_n810		MACH_NOKIA_N810		NOKIA_N810		1548
+pcm038			MACH_PCM038		PCM038			1551
+sg310			MACH_SG310		SG310			1564
+ts209			MACH_TS209		TS209			1565
+at91cap9adk		MACH_AT91CAP9ADK	AT91CAP9ADK		1566
+mx31moboard		MACH_MX31MOBOARD	MX31MOBOARD		1574
+vision_ep9307		MACH_VISION_EP9307	VISION_EP9307		1578
+terastation_pro2	MACH_TERASTATION_PRO2	TERASTATION_PRO2	1584
+linkstation_pro		MACH_LINKSTATION_PRO	LINKSTATION_PRO		1585
+e350			MACH_E350		E350			1596
+ts409			MACH_TS409		TS409			1601
+rsi_ews			MACH_RSI_EWS		RSI_EWS			1609
+cm_x300			MACH_CM_X300		CM_X300			1616
+at91sam9g20ek		MACH_AT91SAM9G20EK	AT91SAM9G20EK		1624
+smdk6410		MACH_SMDK6410		SMDK6410		1626
+u300			MACH_U300		U300			1627
+wrt350n_v2		MACH_WRT350N_V2		WRT350N_V2		1633
+omap_ldp		MACH_OMAP_LDP		OMAP_LDP		1639
+mx35_3ds		MACH_MX35_3DS		MX35_3DS		1645
+neuros_osd2		MACH_NEUROS_OSD2	NEUROS_OSD2		1647
+trizeps4wl		MACH_TRIZEPS4WL		TRIZEPS4WL		1649
+ts78xx			MACH_TS78XX		TS78XX			1652
+sffsdr			MACH_SFFSDR		SFFSDR			1657
+pcm037			MACH_PCM037		PCM037			1673
+db88f6281_bp		MACH_DB88F6281_BP	DB88F6281_BP		1680
+rd88f6192_nas		MACH_RD88F6192_NAS	RD88F6192_NAS		1681
+rd88f6281		MACH_RD88F6281		RD88F6281		1682
+db78x00_bp		MACH_DB78X00_BP		DB78X00_BP		1683
+smdk2416		MACH_SMDK2416		SMDK2416		1685
+wbd111			MACH_WBD111		WBD111			1690
+mv2120			MACH_MV2120		MV2120			1693
+mx51_3ds		MACH_MX51_3DS		MX51_3DS		1696
+imx27lite		MACH_IMX27LITE		IMX27LITE		1701
+usb_a9260		MACH_USB_A9260		USB_A9260		1709
+usb_a9263		MACH_USB_A9263		USB_A9263		1710
+qil_a9260		MACH_QIL_A9260		QIL_A9260		1711
+kzm_arm11_01		MACH_KZM_ARM11_01	KZM_ARM11_01		1722
+nokia_n810_wimax	MACH_NOKIA_N810_WIMAX	NOKIA_N810_WIMAX	1727
+sapphire		MACH_SAPPHIRE		SAPPHIRE		1729
+stmp37xx		MACH_STMP37XX		STMP37XX		1732
+stmp378x		MACH_STMP378X		STMP378X		1733
+ezx_a780		MACH_EZX_A780		EZX_A780		1740
+ezx_e680		MACH_EZX_E680		EZX_E680		1741
+ezx_a1200		MACH_EZX_A1200		EZX_A1200		1742
+ezx_e6			MACH_EZX_E6		EZX_E6			1743
+ezx_e2			MACH_EZX_E2		EZX_E2			1744
+ezx_a910		MACH_EZX_A910		EZX_A910		1745
+edmini_v2		MACH_EDMINI_V2		EDMINI_V2		1756
+zipit2			MACH_ZIPIT2		ZIPIT2			1757
+omap3_pandora		MACH_OMAP3_PANDORA	OMAP3_PANDORA		1761
+mss2			MACH_MSS2		MSS2			1766
+lb88rc8480		MACH_LB88RC8480		LB88RC8480		1769
+mx25_3ds		MACH_MX25_3DS		MX25_3DS		1771
+omap3530_lv_som		MACH_OMAP3530_LV_SOM	OMAP3530_LV_SOM		1773
+davinci_da830_evm	MACH_DAVINCI_DA830_EVM	DAVINCI_DA830_EVM	1781
+dove_db			MACH_DOVE_DB		DOVE_DB			1788
+overo			MACH_OVERO		OVERO			1798
+at2440evb		MACH_AT2440EVB		AT2440EVB		1799
+neocore926		MACH_NEOCORE926		NEOCORE926		1800
+wnr854t			MACH_WNR854T		WNR854T			1801
+rd88f5181l_ge		MACH_RD88F5181L_GE	RD88F5181L_GE		1812
+rd88f5181l_fxo		MACH_RD88F5181L_FXO	RD88F5181L_FXO		1818
+stamp9g20		MACH_STAMP9G20		STAMP9G20		1824
+smdkc100		MACH_SMDKC100		SMDKC100		1826
+tavorevb		MACH_TAVOREVB		TAVOREVB		1827
+saar			MACH_SAAR		SAAR			1828
+at91sam9m10g45ek	MACH_AT91SAM9M10G45EK	AT91SAM9M10G45EK	1830
+usb_a9g20		MACH_USB_A9G20		USB_A9G20		1841
+mxlads			MACH_MXLADS		MXLADS			1851
+linkstation_mini	MACH_LINKSTATION_MINI	LINKSTATION_MINI	1858
+afeb9260		MACH_AFEB9260		AFEB9260		1859
+imx27ipcam		MACH_IMX27IPCAM		IMX27IPCAM		1871
+rd88f6183ap_ge		MACH_RD88F6183AP_GE	RD88F6183AP_GE		1894
+realview_pba8		MACH_REALVIEW_PBA8	REALVIEW_PBA8		1897
+realview_pbx		MACH_REALVIEW_PBX	REALVIEW_PBX		1901
+micro9s			MACH_MICRO9S		MICRO9S			1902
+rut100			MACH_RUT100		RUT100			1908
+g3evm			MACH_G3EVM		G3EVM			1919
+w90p910evb		MACH_W90P910EVB		W90P910EVB		1921
+w90p950evb		MACH_W90P950EVB		W90P950EVB		1923
+w90n960evb		MACH_W90N960EVB		W90N960EVB		1924
+mv88f6281gtw_ge		MACH_MV88F6281GTW_GE	MV88F6281GTW_GE		1932
+ncp			MACH_NCP		NCP			1933
+davinci_dm365_evm	MACH_DAVINCI_DM365_EVM	DAVINCI_DM365_EVM	1939
+centro			MACH_CENTRO		CENTRO			1944
+nokia_rx51		MACH_NOKIA_RX51		NOKIA_RX51		1955
+omap_zoom2		MACH_OMAP_ZOOM2		OMAP_ZOOM2		1967
+cpuat9260		MACH_CPUAT9260		CPUAT9260		1973
+eukrea_cpuimx27		MACH_EUKREA_CPUIMX27	EUKREA_CPUIMX27		1975
+acs5k			MACH_ACS5K		ACS5K			1982
+snapper_9260		MACH_SNAPPER_9260	SNAPPER_9260		1987
+dsm320			MACH_DSM320		DSM320			1988
+exeda			MACH_EXEDA		EXEDA			1994
+mini2440		MACH_MINI2440		MINI2440		1999
+colibri300		MACH_COLIBRI300		COLIBRI300		2000
+linkstation_ls_hgl	MACH_LINKSTATION_LS_HGL	LINKSTATION_LS_HGL	2005
+cpuat9g20		MACH_CPUAT9G20		CPUAT9G20		2031
+smdk6440		MACH_SMDK6440		SMDK6440		2032
+nas4220b		MACH_NAS4220B		NAS4220B		2038
+zylonite2		MACH_ZYLONITE2		ZYLONITE2		2042
+aspenite		MACH_ASPENITE		ASPENITE		2043
+ttc_dkb			MACH_TTC_DKB		TTC_DKB			2045
+pcm043			MACH_PCM043		PCM043			2072
+sheevaplug		MACH_SHEEVAPLUG		SHEEVAPLUG		2097
+avengers_lite		MACH_AVENGERS_LITE	AVENGERS_LITE		2104
+mx51_babbage		MACH_MX51_BABBAGE	MX51_BABBAGE		2125
+rd78x00_masa		MACH_RD78X00_MASA	RD78X00_MASA		2135
+dm355_leopard		MACH_DM355_LEOPARD	DM355_LEOPARD		2138
+ts219			MACH_TS219		TS219			2139
+pca100			MACH_PCA100		PCA100			2149
+davinci_da850_evm	MACH_DAVINCI_DA850_EVM	DAVINCI_DA850_EVM	2157
+at91sam9g10ek		MACH_AT91SAM9G10EK	AT91SAM9G10EK		2159
+omap_4430sdp		MACH_OMAP_4430SDP	OMAP_4430SDP		2160
+magx_zn5		MACH_MAGX_ZN5		MAGX_ZN5		2162
+omap3_torpedo		MACH_OMAP3_TORPEDO	OMAP3_TORPEDO		2178
+anw6410			MACH_ANW6410		ANW6410			2183
+imx27_visstrim_m10	MACH_IMX27_VISSTRIM_M10	IMX27_VISSTRIM_M10	2187
+portuxg20		MACH_PORTUXG20		PORTUXG20		2191
+smdkc110		MACH_SMDKC110		SMDKC110		2193
+cabespresso		MACH_CABESPRESSO	CABESPRESSO		2194
+omap3517evm		MACH_OMAP3517EVM	OMAP3517EVM		2200
+netspace_v2		MACH_NETSPACE_V2	NETSPACE_V2		2201
+netspace_max_v2		MACH_NETSPACE_MAX_V2	NETSPACE_MAX_V2		2202
+d2net_v2		MACH_D2NET_V2		D2NET_V2		2203
+net2big_v2		MACH_NET2BIG_V2		NET2BIG_V2		2204
+net5big_v2		MACH_NET5BIG_V2		NET5BIG_V2		2206
+inetspace_v2		MACH_INETSPACE_V2	INETSPACE_V2		2208
+at91sam9g45ekes		MACH_AT91SAM9G45EKES	AT91SAM9G45EKES		2212
+spear600		MACH_SPEAR600		SPEAR600		2236
+spear300		MACH_SPEAR300		SPEAR300		2237
+lilly1131		MACH_LILLY1131		LILLY1131		2239
+hmt			MACH_HMT		HMT			2254
+vexpress		MACH_VEXPRESS		VEXPRESS		2272
+d2net			MACH_D2NET		D2NET			2282
+bigdisk			MACH_BIGDISK		BIGDISK			2283
+at91sam9g20ek_2mmc	MACH_AT91SAM9G20EK_2MMC	AT91SAM9G20EK_2MMC	2288
+bcmring			MACH_BCMRING		BCMRING			2289
+mahimahi		MACH_MAHIMAHI		MAHIMAHI		2304
+cerebric		MACH_CEREBRIC		CEREBRIC		2311
+smdk6442		MACH_SMDK6442		SMDK6442		2324
+openrd_base		MACH_OPENRD_BASE	OPENRD_BASE		2325
+devkit8000		MACH_DEVKIT8000		DEVKIT8000		2330
+mx51_efikamx		MACH_MX51_EFIKAMX	MX51_EFIKAMX		2336
+cm_t35			MACH_CM_T35		CM_T35			2341
+net2big			MACH_NET2BIG		NET2BIG			2342
+igep0020		MACH_IGEP0020		IGEP0020		2344
+nuc932evb		MACH_NUC932EVB		NUC932EVB		2356
+openrd_client		MACH_OPENRD_CLIENT	OPENRD_CLIENT		2361
+u8500			MACH_U8500		U8500			2368
+mx51_efikasb		MACH_MX51_EFIKASB	MX51_EFIKASB		2370
+marvell_jasper		MACH_MARVELL_JASPER	MARVELL_JASPER		2382
+flint			MACH_FLINT		FLINT			2383
+tavorevb3		MACH_TAVOREVB3		TAVOREVB3		2384
+touchbook		MACH_TOUCHBOOK		TOUCHBOOK		2393
+raumfeld_rc		MACH_RAUMFELD_RC	RAUMFELD_RC		2413
+raumfeld_connector	MACH_RAUMFELD_CONNECTOR	RAUMFELD_CONNECTOR	2414
+raumfeld_speaker	MACH_RAUMFELD_SPEAKER	RAUMFELD_SPEAKER	2415
+tnetv107x		MACH_TNETV107X		TNETV107X		2418
+smdkv210		MACH_SMDKV210		SMDKV210		2456
+omap_zoom3		MACH_OMAP_ZOOM3		OMAP_ZOOM3		2464
+omap_3630sdp		MACH_OMAP_3630SDP	OMAP_3630SDP		2465
+cybook2440		MACH_CYBOOK2440		CYBOOK2440		2466
+smartq7			MACH_SMARTQ7		SMARTQ7			2479
+watson_efm_plugin	MACH_WATSON_EFM_PLUGIN	WATSON_EFM_PLUGIN	2491
+g4evm			MACH_G4EVM		G4EVM			2493
+omapl138_hawkboard	MACH_OMAPL138_HAWKBOARD	OMAPL138_HAWKBOARD	2495
+ts41x			MACH_TS41X		TS41X			2502
+phy3250			MACH_PHY3250		PHY3250			2511
+mini6410		MACH_MINI6410		MINI6410		2520
+mx28evk			MACH_MX28EVK		MX28EVK			2531
+smartq5			MACH_SMARTQ5		SMARTQ5			2534
+davinci_dm6467tevm	MACH_DAVINCI_DM6467TEVM	DAVINCI_DM6467TEVM	2548
+mxt_td60		MACH_MXT_TD60		MXT_TD60		2550
+capc7117		MACH_CAPC7117		CAPC7117		2612
+icontrol		MACH_ICONTROL		ICONTROL		2624
+gplugd			MACH_GPLUGD		GPLUGD			2625
+qsd8x50a_st1_5		MACH_QSD8X50A_ST1_5	QSD8X50A_ST1_5		2627
+mx23evk			MACH_MX23EVK		MX23EVK			2629
+ap4evb			MACH_AP4EVB		AP4EVB			2630
+mityomapl138		MACH_MITYOMAPL138	MITYOMAPL138		2650
+guruplug		MACH_GURUPLUG		GURUPLUG		2659
+spear310		MACH_SPEAR310		SPEAR310		2660
+spear320		MACH_SPEAR320		SPEAR320		2661
+aquila			MACH_AQUILA		AQUILA			2676
+esata_sheevaplug	MACH_ESATA_SHEEVAPLUG	ESATA_SHEEVAPLUG	2678
+msm7x30_surf		MACH_MSM7X30_SURF	MSM7X30_SURF		2679
+terastation_wxl		MACH_TERASTATION_WXL	TERASTATION_WXL		2697
+msm7x25_surf		MACH_MSM7X25_SURF	MSM7X25_SURF		2703
+msm7x25_ffa		MACH_MSM7X25_FFA	MSM7X25_FFA		2704
+msm7x27_surf		MACH_MSM7X27_SURF	MSM7X27_SURF		2705
+msm7x27_ffa		MACH_MSM7X27_FFA	MSM7X27_FFA		2706
+msm7x30_ffa		MACH_MSM7X30_FFA	MSM7X30_FFA		2707
+qsd8x50_surf		MACH_QSD8X50_SURF	QSD8X50_SURF		2708
+mx53_evk		MACH_MX53_EVK		MX53_EVK		2716
+igep0030		MACH_IGEP0030		IGEP0030		2717
+sbc3530			MACH_SBC3530		SBC3530			2722
+saarb			MACH_SAARB		SAARB			2727
+harmony			MACH_HARMONY		HARMONY			2731
+cybook_orizon		MACH_CYBOOK_ORIZON	CYBOOK_ORIZON		2733
+msm7x30_fluid		MACH_MSM7X30_FLUID	MSM7X30_FLUID		2741
+cm_t3517		MACH_CM_T3517		CM_T3517		2750
+wbd222			MACH_WBD222		WBD222			2753
+msm8x60_surf		MACH_MSM8X60_SURF	MSM8X60_SURF		2755
+msm8x60_sim		MACH_MSM8X60_SIM	MSM8X60_SIM		2756
+tcc8000_sdk		MACH_TCC8000_SDK	TCC8000_SDK		2758
+cns3420vb		MACH_CNS3420VB		CNS3420VB		2776
+omap4_panda		MACH_OMAP4_PANDA	OMAP4_PANDA		2791
+ti8168evm		MACH_TI8168EVM		TI8168EVM		2800
+teton_bga		MACH_TETON_BGA		TETON_BGA		2816
+eukrea_cpuimx25sd	MACH_EUKREA_CPUIMX25SD	EUKREA_CPUIMX25SD	2820
+eukrea_cpuimx35sd	MACH_EUKREA_CPUIMX35SD	EUKREA_CPUIMX35SD	2821
+eukrea_cpuimx51sd	MACH_EUKREA_CPUIMX51SD	EUKREA_CPUIMX51SD	2822
+eukrea_cpuimx51		MACH_EUKREA_CPUIMX51	EUKREA_CPUIMX51		2823
+smdkc210		MACH_SMDKC210		SMDKC210		2838
+t5325			MACH_T5325		T5325			2846
+income			MACH_INCOME		INCOME			2849
+goni			MACH_GONI		GONI			2862
+bv07			MACH_BV07		BV07			2882
+openrd_ultimate		MACH_OPENRD_ULTIMATE	OPENRD_ULTIMATE		2884
+devixp			MACH_DEVIXP		DEVIXP			2885
+miccpt			MACH_MICCPT		MICCPT			2886
+mic256			MACH_MIC256		MIC256			2887
+u5500			MACH_U5500		U5500			2890
+linkstation_lschl	MACH_LINKSTATION_LSCHL	LINKSTATION_LSCHL	2913
+smdkv310		MACH_SMDKV310		SMDKV310		2925
+wm8505_7in_netbook	MACH_WM8505_7IN_NETBOOK	WM8505_7IN_NETBOOK	2928
+craneboard		MACH_CRANEBOARD		CRANEBOARD		2932
+smdk6450		MACH_SMDK6450		SMDK6450		2938
+brownstone		MACH_BROWNSTONE		BROWNSTONE		2957
+flexibity		MACH_FLEXIBITY		FLEXIBITY		2965
+mx50_rdp		MACH_MX50_RDP		MX50_RDP		2988
+universal_c210		MACH_UNIVERSAL_C210	UNIVERSAL_C210		2989
+real6410		MACH_REAL6410		REAL6410		2990
+dockstar		MACH_DOCKSTAR		DOCKSTAR		2998
+ti8148evm		MACH_TI8148EVM		TI8148EVM		3004
+seaboard		MACH_SEABOARD		SEABOARD		3005
+mx53_ard		MACH_MX53_ARD		MX53_ARD		3010
+mx53_smd		MACH_MX53_SMD		MX53_SMD		3011
+msm8x60_rumi3		MACH_MSM8X60_RUMI3	MSM8X60_RUMI3		3016
+msm8x60_ffa		MACH_MSM8X60_FFA	MSM8X60_FFA		3017
+cm_a510			MACH_CM_A510		CM_A510			3020
+tx28			MACH_TX28		TX28			3043
+pcontrol_g20		MACH_PCONTROL_G20	PCONTROL_G20		3062
+vpr200			MACH_VPR200		VPR200			3087
+torbreck		MACH_TORBRECK		TORBRECK		3090
+prima2_evb		MACH_PRIMA2_EVB		PRIMA2_EVB		3103
+paz00			MACH_PAZ00		PAZ00			3128
+acmenetusfoxg20		MACH_ACMENETUSFOXG20	ACMENETUSFOXG20		3129
+ag5evm			MACH_AG5EVM		AG5EVM			3189
+ics_if_voip		MACH_ICS_IF_VOIP	ICS_IF_VOIP		3206
+wlf_cragg_6410		MACH_WLF_CRAGG_6410	WLF_CRAGG_6410		3207
+trimslice		MACH_TRIMSLICE		TRIMSLICE		3209
+kaen			MACH_KAEN		KAEN			3217
+nokia_rm680		MACH_NOKIA_RM680	NOKIA_RM680		3220
+msm8960_sim		MACH_MSM8960_SIM	MSM8960_SIM		3230
+msm8960_rumi3		MACH_MSM8960_RUMI3	MSM8960_RUMI3		3231
+gsia18s			MACH_GSIA18S		GSIA18S			3234
+mx53_loco		MACH_MX53_LOCO		MX53_LOCO		3273
+wario			MACH_WARIO		WARIO			3288
+cm_t3730		MACH_CM_T3730		CM_T3730		3290
+hrefv60			MACH_HREFV60		HREFV60			3293
+armlex4210		MACH_ARMLEX4210		ARMLEX4210		3361
+snowball		MACH_SNOWBALL		SNOWBALL		3363
+xilinx_ep107		MACH_XILINX_EP107	XILINX_EP107		3378
+nuri			MACH_NURI		NURI			3379
+origen			MACH_ORIGEN		ORIGEN			3455
+nspire			MACH_NSPIRE		NSPIRE			3503
+nokia_rm696		MACH_NOKIA_RM696	NOKIA_RM696		3522
+mikrap_x168		MACH_MIKRAP_X168	MIKRAP_X168		3543
+deto_macarm9		MACH_DETO_MACARM9	DETO_MACARM9		3568
+m28evk			MACH_M28EVK		M28EVK			3613
+kota2			MACH_KOTA2		KOTA2			3616
+bonito			MACH_BONITO		BONITO			3623
+omap3_egf		MACH_OMAP3_EGF		OMAP3_EGF		3637
+smdk4212		MACH_SMDK4212		SMDK4212		3638
+apx4devkit		MACH_APX4DEVKIT		APX4DEVKIT		3712
+smdk4412		MACH_SMDK4412		SMDK4412		3765
+marzen			MACH_MARZEN		MARZEN			3790
+krome			MACH_KROME		KROME			3797
+armadillo800eva		MACH_ARMADILLO800EVA	ARMADILLO800EVA		3863
+mx53_umobo		MACH_MX53_UMOBO		MX53_UMOBO		3927
+mt4			MACH_MT4		MT4			3981
+u8520			MACH_U8520		U8520			3990
+chupacabra		MACH_CHUPACABRA		CHUPACABRA		4098
+scorpion		MACH_SCORPION		SCORPION		4099
+davinci_he_hmi10	MACH_DAVINCI_HE_HMI10	DAVINCI_HE_HMI10	4100
+topkick			MACH_TOPKICK		TOPKICK			4101
+m3_auguestrush		MACH_M3_AUGUESTRUSH	M3_AUGUESTRUSH		4102
+ipc335x			MACH_IPC335X		IPC335X			4103
+sun4i			MACH_SUN4I		SUN4I			4104
+imx233_olinuxino	MACH_IMX233_OLINUXINO	IMX233_OLINUXINO	4105
+k2_wl			MACH_K2_WL		K2_WL			4106
+k2_ul			MACH_K2_UL		K2_UL			4107
+k2_cl			MACH_K2_CL		K2_CL			4108
+minbari_w		MACH_MINBARI_W		MINBARI_W		4109
+minbari_m		MACH_MINBARI_M		MINBARI_M		4110
+k035			MACH_K035		K035			4111
+ariel			MACH_ARIEL		ARIEL			4112
+arielsaarc		MACH_ARIELSAARC		ARIELSAARC		4113
+arieldkb		MACH_ARIELDKB		ARIELDKB		4114
+armadillo810		MACH_ARMADILLO810	ARMADILLO810		4115
+tam335x			MACH_TAM335X		TAM335X			4116
+grouper			MACH_GROUPER		GROUPER			4117
+mpcsa21_9g20		MACH_MPCSA21_9G20	MPCSA21_9G20		4118
+m6u_cpu			MACH_M6U_CPU		M6U_CPU			4119
+ginkgo			MACH_GINKGO		GINKGO			4121
+cgt_qmx6		MACH_CGT_QMX6		CGT_QMX6		4122
+profpga			MACH_PROFPGA		PROFPGA			4123
+acfx100oc		MACH_ACFX100OC		ACFX100OC		4124
+acfx100nb		MACH_ACFX100NB		ACFX100NB		4125
+capricorn		MACH_CAPRICORN		CAPRICORN		4126
+pisces			MACH_PISCES		PISCES			4127
+aries			MACH_ARIES		ARIES			4128
+cancer			MACH_CANCER		CANCER			4129
+leo			MACH_LEO		LEO			4130
+virgo			MACH_VIRGO		VIRGO			4131
+sagittarius		MACH_SAGITTARIUS	SAGITTARIUS		4132
+devil			MACH_DEVIL		DEVIL			4133
+ballantines		MACH_BALLANTINES	BALLANTINES		4134
+omap3_procerusvpu	MACH_OMAP3_PROCERUSVPU	OMAP3_PROCERUSVPU	4135
+my27			MACH_MY27		MY27			4136
+sun6i			MACH_SUN6I		SUN6I			4137
+sun5i			MACH_SUN5I		SUN5I			4138
+mx512_mx		MACH_MX512_MX		MX512_MX		4139
+kzm9g			MACH_KZM9G		KZM9G			4140
+vdstbn			MACH_VDSTBN		VDSTBN			4141
+cfa10036		MACH_CFA10036		CFA10036		4142
+cfa10049		MACH_CFA10049		CFA10049		4143
+pcm051			MACH_PCM051		PCM051			4144
+vybrid_vf7xx		MACH_VYBRID_VF7XX	VYBRID_VF7XX		4145
+vybrid_vf6xx		MACH_VYBRID_VF6XX	VYBRID_VF6XX		4146
+vybrid_vf5xx		MACH_VYBRID_VF5XX	VYBRID_VF5XX		4147
+vybrid_vf4xx		MACH_VYBRID_VF4XX	VYBRID_VF4XX		4148
+aria_g25		MACH_ARIA_G25		ARIA_G25		4149
+bcm21553		MACH_BCM21553		BCM21553		4150
+smdk5410		MACH_SMDK5410		SMDK5410		4151
+lpc18xx			MACH_LPC18XX		LPC18XX			4152
+oratisparty		MACH_ORATISPARTY	ORATISPARTY		4153
+qseven			MACH_QSEVEN		QSEVEN			4154
+gmv_generic		MACH_GMV_GENERIC	GMV_GENERIC		4155
+th_link_eth		MACH_TH_LINK_ETH	TH_LINK_ETH		4156
+tn_muninn		MACH_TN_MUNINN		TN_MUNINN		4157
+rampage			MACH_RAMPAGE		RAMPAGE			4158
+visstrim_mv10		MACH_VISSTRIM_MV10	VISSTRIM_MV10		4159
+mx28_wilma		MACH_MX28_WILMA		MX28_WILMA		4164
+msm8625_ffa		MACH_MSM8625_FFA	MSM8625_FFA		4166
+vpu101			MACH_VPU101		VPU101			4167
+baileys			MACH_BAILEYS		BAILEYS			4169
+familybox		MACH_FAMILYBOX		FAMILYBOX		4170
+ensemble_mx35		MACH_ENSEMBLE_MX35	ENSEMBLE_MX35		4171
+sc_sps_1		MACH_SC_SPS_1		SC_SPS_1		4172
+ucsimply_sam9260	MACH_UCSIMPLY_SAM9260	UCSIMPLY_SAM9260	4173
+unicorn			MACH_UNICORN		UNICORN			4174
+m9g45a			MACH_M9G45A		M9G45A			4175
+mtwebif			MACH_MTWEBIF		MTWEBIF			4176
+playstone		MACH_PLAYSTONE		PLAYSTONE		4177
+chelsea			MACH_CHELSEA		CHELSEA			4178
+bayern			MACH_BAYERN		BAYERN			4179
+mitwo			MACH_MITWO		MITWO			4180
+mx25_noah		MACH_MX25_NOAH		MX25_NOAH		4181
+stm_b2020		MACH_STM_B2020		STM_B2020		4182
+annax_src		MACH_ANNAX_SRC		ANNAX_SRC		4183
+ionics_stratus		MACH_IONICS_STRATUS	IONICS_STRATUS		4184
+hugo			MACH_HUGO		HUGO			4185
+em300			MACH_EM300		EM300			4186
+mmp3_qseven		MACH_MMP3_QSEVEN	MMP3_QSEVEN		4187
+bosphorus2		MACH_BOSPHORUS2		BOSPHORUS2		4188
+tt2200			MACH_TT2200		TT2200			4189
+ocelot3			MACH_OCELOT3		OCELOT3			4190
+tek_cobra		MACH_TEK_COBRA		TEK_COBRA		4191
+protou			MACH_PROTOU		PROTOU			4192
+msm8625_evt		MACH_MSM8625_EVT	MSM8625_EVT		4193
+mx53_sellwood		MACH_MX53_SELLWOOD	MX53_SELLWOOD		4194
+somiq_am35		MACH_SOMIQ_AM35		SOMIQ_AM35		4195
+somiq_am37		MACH_SOMIQ_AM37		SOMIQ_AM37		4196
+k2_plc_cl		MACH_K2_PLC_CL		K2_PLC_CL		4197
+tc2			MACH_TC2		TC2			4198
+dulex_j			MACH_DULEX_J		DULEX_J			4199
+stm_b2044		MACH_STM_B2044		STM_B2044		4200
+deluxe_j		MACH_DELUXE_J		DELUXE_J		4201
+mango2443		MACH_MANGO2443		MANGO2443		4202
+cp2dcg			MACH_CP2DCG		CP2DCG			4203
+cp2dtg			MACH_CP2DTG		CP2DTG			4204
+cp2dug			MACH_CP2DUG		CP2DUG			4205
+var_som_am33		MACH_VAR_SOM_AM33	VAR_SOM_AM33		4206
+pepper			MACH_PEPPER		PEPPER			4207
+mango2450		MACH_MANGO2450		MANGO2450		4208
+valente_wx_c9		MACH_VALENTE_WX_C9	VALENTE_WX_C9		4209
+minitv			MACH_MINITV		MINITV			4210
+u8540			MACH_U8540		U8540			4211
+iv_atlas_i_z7e		MACH_IV_ATLAS_I_Z7E	IV_ATLAS_I_Z7E		4212
+mach_type_sky		MACH_MACH_TYPE_SKY	MACH_TYPE_SKY		4214
+bluesky			MACH_BLUESKY		BLUESKY			4215
+ngrouter		MACH_NGROUTER		NGROUTER		4216
+mx53_denetim		MACH_MX53_DENETIM	MX53_DENETIM		4217
+opal			MACH_OPAL		OPAL			4218
+gnet_us3gref		MACH_GNET_US3GREF	GNET_US3GREF		4219
+gnet_nc3g		MACH_GNET_NC3G		GNET_NC3G		4220
+gnet_ge3g		MACH_GNET_GE3G		GNET_GE3G		4221
+adp2			MACH_ADP2		ADP2			4222
+tqma28			MACH_TQMA28		TQMA28			4223
+kacom3			MACH_KACOM3		KACOM3			4224
+rrhdemo			MACH_RRHDEMO		RRHDEMO			4225
+protodug		MACH_PROTODUG		PROTODUG		4226
+lago			MACH_LAGO		LAGO			4227
+ktt30			MACH_KTT30		KTT30			4228
+ts43xx			MACH_TS43XX		TS43XX			4229
+mx6q_denso		MACH_MX6Q_DENSO		MX6Q_DENSO		4230
+comsat_gsmumts8		MACH_COMSAT_GSMUMTS8	COMSAT_GSMUMTS8		4231
+dreamx			MACH_DREAMX		DREAMX			4232
+thunderstonem		MACH_THUNDERSTONEM	THUNDERSTONEM		4233
+yoyopad			MACH_YOYOPAD		YOYOPAD			4234
+yoyopatient		MACH_YOYOPATIENT	YOYOPATIENT		4235
+a10l			MACH_A10L		A10L			4236
+mq60			MACH_MQ60		MQ60			4237
+linkstation_lsql	MACH_LINKSTATION_LSQL	LINKSTATION_LSQL	4238
+am3703gateway		MACH_AM3703GATEWAY	AM3703GATEWAY		4239
+accipiter		MACH_ACCIPITER		ACCIPITER		4240
+magnidug		MACH_MAGNIDUG		MAGNIDUG		4242
+hydra			MACH_HYDRA		HYDRA			4243
+sun3i			MACH_SUN3I		SUN3I			4244
+stm_b2078		MACH_STM_B2078		STM_B2078		4245
+at91sam9263deskv2	MACH_AT91SAM9263DESKV2	AT91SAM9263DESKV2	4246
+deluxe_r		MACH_DELUXE_R		DELUXE_R		4247
+p_98_v			MACH_P_98_V		P_98_V			4248
+p_98_c			MACH_P_98_C		P_98_C			4249
+davinci_am18xx_omn	MACH_DAVINCI_AM18XX_OMN	DAVINCI_AM18XX_OMN	4250
+socfpga_cyclone5	MACH_SOCFPGA_CYCLONE5	SOCFPGA_CYCLONE5	4251
+cabatuin		MACH_CABATUIN		CABATUIN		4252
+yoyopad_ft		MACH_YOYOPAD_FT		YOYOPAD_FT		4253
+dan2400evb		MACH_DAN2400EVB		DAN2400EVB		4254
+dan3400evb		MACH_DAN3400EVB		DAN3400EVB		4255
+edm_sf_imx6		MACH_EDM_SF_IMX6	EDM_SF_IMX6		4256
+edm_cf_imx6		MACH_EDM_CF_IMX6	EDM_CF_IMX6		4257
+vpos3xx			MACH_VPOS3XX		VPOS3XX			4258
+vulcano_9x5		MACH_VULCANO_9X5	VULCANO_9X5		4259
+spmp8000		MACH_SPMP8000		SPMP8000		4260
+catalina		MACH_CATALINA		CATALINA		4261
+rd88f5181l_fe		MACH_RD88F5181L_FE	RD88F5181L_FE		4262
+mx535_mx		MACH_MX535_MX		MX535_MX		4263
+armadillo840		MACH_ARMADILLO840	ARMADILLO840		4264
+spc9000baseboard	MACH_SPC9000BASEBOARD	SPC9000BASEBOARD	4265
+iris			MACH_IRIS		IRIS			4266
+protodcg		MACH_PROTODCG		PROTODCG		4267
+palmtree		MACH_PALMTREE		PALMTREE		4268
+novena			MACH_NOVENA		NOVENA			4269
+ma_um			MACH_MA_UM		MA_UM			4270
+ma_am			MACH_MA_AM		MA_AM			4271
+ems348			MACH_EMS348		EMS348			4272
+cm_fx6			MACH_CM_FX6		CM_FX6			4273
+arndale			MACH_ARNDALE		ARNDALE			4274
+q5xr5			MACH_Q5XR5		Q5XR5			4275
+willow			MACH_WILLOW		WILLOW			4276
+omap3621_odyv3		MACH_OMAP3621_ODYV3	OMAP3621_ODYV3		4277
+omapl138_presonus	MACH_OMAPL138_PRESONUS	OMAPL138_PRESONUS	4278
+dvf99			MACH_DVF99		DVF99			4279
+impression_j		MACH_IMPRESSION_J	IMPRESSION_J		4280
+qblissa9		MACH_QBLISSA9		QBLISSA9		4281
+robin_heliview10	MACH_ROBIN_HELIVIEW10	ROBIN_HELIVIEW10	4282
+sun7i			MACH_SUN7I		SUN7I			4283
+mx6q_hdmidongle		MACH_MX6Q_HDMIDONGLE	MX6Q_HDMIDONGLE		4284
+mx6_sid2		MACH_MX6_SID2		MX6_SID2		4285
+helios_v3		MACH_HELIOS_V3		HELIOS_V3		4286
+helios_v4		MACH_HELIOS_V4		HELIOS_V4		4287
+q7_imx6			MACH_Q7_IMX6		Q7_IMX6			4288
+odroidx			MACH_ODROIDX		ODROIDX			4289
+robpro			MACH_ROBPRO		ROBPRO			4290
+research59if_mk1	MACH_RESEARCH59IF_MK1	RESEARCH59IF_MK1	4291
+bobsleigh		MACH_BOBSLEIGH		BOBSLEIGH		4292
+dcshgwt3		MACH_DCSHGWT3		DCSHGWT3		4293
+gld1018			MACH_GLD1018		GLD1018			4294
+ev10			MACH_EV10		EV10			4295
+nitrogen6x		MACH_NITROGEN6X		NITROGEN6X		4296
+p_107_bb		MACH_P_107_BB		P_107_BB		4297
+evita_utl		MACH_EVITA_UTL		EVITA_UTL		4298
+falconwing		MACH_FALCONWING		FALCONWING		4299
+dct3			MACH_DCT3		DCT3			4300
+cpx2e_cell		MACH_CPX2E_CELL		CPX2E_CELL		4301
+amiro			MACH_AMIRO		AMIRO			4302
+mx6q_brassboard		MACH_MX6Q_BRASSBOARD	MX6Q_BRASSBOARD		4303
+dalmore			MACH_DALMORE		DALMORE			4304
+omap3_portal7cp		MACH_OMAP3_PORTAL7CP	OMAP3_PORTAL7CP		4305
+tegra_pluto		MACH_TEGRA_PLUTO	TEGRA_PLUTO		4306
+mx6sl_evk		MACH_MX6SL_EVK		MX6SL_EVK		4307
+m7			MACH_M7			M7			4308
+pxm2			MACH_PXM2		PXM2			4309
+haba_knx_lite		MACH_HABA_KNX_LITE	HABA_KNX_LITE		4310
+tai			MACH_TAI		TAI			4311
+prototd			MACH_PROTOTD		PROTOTD			4312
+dst_tonto		MACH_DST_TONTO		DST_TONTO		4313
+draco			MACH_DRACO		DRACO			4314
+dxr2			MACH_DXR2		DXR2			4315
+rut			MACH_RUT		RUT			4316
+am180x_wsc		MACH_AM180X_WSC		AM180X_WSC		4317
+deluxe_u		MACH_DELUXE_U		DELUXE_U		4318
+deluxe_ul		MACH_DELUXE_UL		DELUXE_UL		4319
+at91sam9260medths	MACH_AT91SAM9260MEDTHS	AT91SAM9260MEDTHS	4320
+matrix516		MACH_MATRIX516		MATRIX516		4321
+vid401x			MACH_VID401X		VID401X			4322
+helios_v5		MACH_HELIOS_V5		HELIOS_V5		4323
+playpaq2		MACH_PLAYPAQ2		PLAYPAQ2		4324
+igam			MACH_IGAM		IGAM			4325
+amico_i			MACH_AMICO_I		AMICO_I			4326
+amico_e			MACH_AMICO_E		AMICO_E			4327
+sentient_mm3_ck		MACH_SENTIENT_MM3_CK	SENTIENT_MM3_CK		4328
+smx6			MACH_SMX6		SMX6			4329
+pango			MACH_PANGO		PANGO			4330
+ns115_stick		MACH_NS115_STICK	NS115_STICK		4331
+bctrm3			MACH_BCTRM3		BCTRM3			4332
+doctorws		MACH_DOCTORWS		DOCTORWS		4333
+m2601			MACH_M2601		M2601			4334
+vgg1111			MACH_VGG1111		VGG1111			4337
+countach		MACH_COUNTACH		COUNTACH		4338
+visstrim_sm20		MACH_VISSTRIM_SM20	VISSTRIM_SM20		4339
+a639			MACH_A639		A639			4340
+spacemonkey		MACH_SPACEMONKEY	SPACEMONKEY		4341
+zpdu_stamp		MACH_ZPDU_STAMP		ZPDU_STAMP		4342
+htc_g7_clone		MACH_HTC_G7_CLONE	HTC_G7_CLONE		4343
+ft2080_corvus		MACH_FT2080_CORVUS	FT2080_CORVUS		4344
+fisland			MACH_FISLAND		FISLAND			4345
+zpdu			MACH_ZPDU		ZPDU			4346
+urt			MACH_URT		URT			4347
+conti_ovip		MACH_CONTI_OVIP		CONTI_OVIP		4348
+omapl138_nagra		MACH_OMAPL138_NAGRA	OMAPL138_NAGRA		4349
+da850_at3kp1		MACH_DA850_AT3KP1	DA850_AT3KP1		4350
+da850_at3kp2		MACH_DA850_AT3KP2	DA850_AT3KP2		4351
+surma			MACH_SURMA		SURMA			4352
+stm_b2092		MACH_STM_B2092		STM_B2092		4353
+mx535_ycr		MACH_MX535_YCR		MX535_YCR		4354
+m7_wl			MACH_M7_WL		M7_WL			4355
+m7_u			MACH_M7_U		M7_U			4356
+omap3_stndt_evm		MACH_OMAP3_STNDT_EVM	OMAP3_STNDT_EVM		4357
+m7_wlv			MACH_M7_WLV		M7_WLV			4358
+xam3517			MACH_XAM3517		XAM3517			4359
+a220			MACH_A220		A220			4360
+aclima_odie		MACH_ACLIMA_ODIE	ACLIMA_ODIE		4361
+vibble			MACH_VIBBLE		VIBBLE			4362
+k2_u			MACH_K2_U		K2_U			4363
+mx53_egf		MACH_MX53_EGF		MX53_EGF		4364
+novpek_imx53		MACH_NOVPEK_IMX53	NOVPEK_IMX53		4365
+novpek_imx6x		MACH_NOVPEK_IMX6X	NOVPEK_IMX6X		4366
+mx25_smartbox		MACH_MX25_SMARTBOX	MX25_SMARTBOX		4367
+eicg6410		MACH_EICG6410		EICG6410		4368
+picasso_e3		MACH_PICASSO_E3		PICASSO_E3		4369
+motonavigator		MACH_MOTONAVIGATOR	MOTONAVIGATOR		4370
+varioconnect2		MACH_VARIOCONNECT2	VARIOCONNECT2		4371
+deluxe_tw		MACH_DELUXE_TW		DELUXE_TW		4372
+kore3			MACH_KORE3		KORE3			4374
+mx6s_drs		MACH_MX6S_DRS		MX6S_DRS		4375
+cmimx6			MACH_CMIMX6		CMIMX6			4376
+roth			MACH_ROTH		ROTH			4377
+eq4ux			MACH_EQ4UX		EQ4UX			4378
+x1plus			MACH_X1PLUS		X1PLUS			4379
+modimx27		MACH_MODIMX27		MODIMX27		4380
+videon_hduac		MACH_VIDEON_HDUAC	VIDEON_HDUAC		4381
+blackbird		MACH_BLACKBIRD		BLACKBIRD		4382
+runmaster		MACH_RUNMASTER		RUNMASTER		4383
+ceres			MACH_CERES		CERES			4384
+nad435			MACH_NAD435		NAD435			4385
+ns115_proto_type	MACH_NS115_PROTO_TYPE	NS115_PROTO_TYPE	4386
+fs20_vcc		MACH_FS20_VCC		FS20_VCC		4387
+meson6tv_skt		MACH_MESON6TV_SKT	MESON6TV_SKT		4389
+keystone		MACH_KEYSTONE		KEYSTONE		4390
+pcm052			MACH_PCM052		PCM052			4391
+qrd_skud_prime		MACH_QRD_SKUD_PRIME	QRD_SKUD_PRIME		4393
+guf_santaro		MACH_GUF_SANTARO	GUF_SANTARO		4395
+sheepshead		MACH_SHEEPSHEAD		SHEEPSHEAD		4396
+mx6_iwg15m_mxm		MACH_MX6_IWG15M_MXM	MX6_IWG15M_MXM		4397
+mx6_iwg15m_q7		MACH_MX6_IWG15M_Q7	MX6_IWG15M_Q7		4398
+at91sam9263if8mic	MACH_AT91SAM9263IF8MIC	AT91SAM9263IF8MIC	4399
+marcopolo		MACH_MARCOPOLO		MARCOPOLO		4401
+mx535_sdcr		MACH_MX535_SDCR		MX535_SDCR		4402
+mx53_csb2733		MACH_MX53_CSB2733	MX53_CSB2733		4403
+diva			MACH_DIVA		DIVA			4404
+ncr_7744		MACH_NCR_7744		NCR_7744		4405
+macallan		MACH_MACALLAN		MACALLAN		4406
+wnr3500			MACH_WNR3500		WNR3500			4407
+pgavrf			MACH_PGAVRF		PGAVRF			4408
+helios_v6		MACH_HELIOS_V6		HELIOS_V6		4409
+lcct			MACH_LCCT		LCCT			4410
+csndug			MACH_CSNDUG		CSNDUG			4411
+wandboard_imx6		MACH_WANDBOARD_IMX6	WANDBOARD_IMX6		4412
+omap4_jet		MACH_OMAP4_JET		OMAP4_JET		4413
+tegra_roth		MACH_TEGRA_ROTH		TEGRA_ROTH		4414
+m7dcg			MACH_M7DCG		M7DCG			4415
+m7dug			MACH_M7DUG		M7DUG			4416
+m7dtg			MACH_M7DTG		M7DTG			4417
+ap42x			MACH_AP42X		AP42X			4418
+var_som_mx6		MACH_VAR_SOM_MX6	VAR_SOM_MX6		4419
+pdlu			MACH_PDLU		PDLU			4420
+hydrogen		MACH_HYDROGEN		HYDROGEN		4421
+npa211e			MACH_NPA211E		NPA211E			4422
+arcadia			MACH_ARCADIA		ARCADIA			4423
+arcadia_l		MACH_ARCADIA_L		ARCADIA_L		4424
+msm8930dt		MACH_MSM8930DT		MSM8930DT		4425
+ktam3874		MACH_KTAM3874		KTAM3874		4426
+cec4			MACH_CEC4		CEC4			4427
+ape6evm			MACH_APE6EVM		APE6EVM			4428
+tx6			MACH_TX6		TX6			4429
+cfa10037		MACH_CFA10037		CFA10037		4431
+ezp1000			MACH_EZP1000		EZP1000			4433
+wgr826v			MACH_WGR826V		WGR826V			4434
+exuma			MACH_EXUMA		EXUMA			4435
+fregate			MACH_FREGATE		FREGATE			4436
+osirisimx508		MACH_OSIRISIMX508	OSIRISIMX508		4437
+st_exigo		MACH_ST_EXIGO		ST_EXIGO		4438
+pismo			MACH_PISMO		PISMO			4439
+atc7			MACH_ATC7		ATC7			4440
+nspireclp		MACH_NSPIRECLP		NSPIRECLP		4441
+nspiretp		MACH_NSPIRETP		NSPIRETP		4442
+nspirecx		MACH_NSPIRECX		NSPIRECX		4443
+maya			MACH_MAYA		MAYA			4444
+wecct			MACH_WECCT		WECCT			4445
+m2s			MACH_M2S		M2S			4446
+msm8625q_evbd		MACH_MSM8625Q_EVBD	MSM8625Q_EVBD		4447
+tiny210			MACH_TINY210		TINY210			4448
+g3			MACH_G3			G3			4449
+hurricane		MACH_HURRICANE		HURRICANE		4450
+mx6_pod			MACH_MX6_POD		MX6_POD			4451
+elondcn			MACH_ELONDCN		ELONDCN			4452
+cwmx535			MACH_CWMX535		CWMX535			4453
+m7_wlj			MACH_M7_WLJ		M7_WLJ			4454
+qsp_arm			MACH_QSP_ARM		QSP_ARM			4455
+msm8625q_skud		MACH_MSM8625Q_SKUD	MSM8625Q_SKUD		4456
+htcmondrian		MACH_HTCMONDRIAN	HTCMONDRIAN		4457
+watson_ead		MACH_WATSON_EAD		WATSON_EAD		4458
+mitwoa			MACH_MITWOA		MITWOA			4459
+omap3_wolverine		MACH_OMAP3_WOLVERINE	OMAP3_WOLVERINE		4460
+mapletree		MACH_MAPLETREE		MAPLETREE		4461
+msm8625_fih_sae		MACH_MSM8625_FIH_SAE	MSM8625_FIH_SAE		4462
+epc35			MACH_EPC35		EPC35			4463
+smartrtu		MACH_SMARTRTU		SMARTRTU		4464
+rcm101			MACH_RCM101		RCM101			4465
+amx_imx53_mxx		MACH_AMX_IMX53_MXX	AMX_IMX53_MXX		4466
+acer_a12		MACH_ACER_A12		ACER_A12		4470
+sbc6x			MACH_SBC6X		SBC6X			4471
+u2			MACH_U2			U2			4472
+smdk4270		MACH_SMDK4270		SMDK4270		4473
+priscillag		MACH_PRISCILLAG		PRISCILLAG		4474
+priscillac		MACH_PRISCILLAC		PRISCILLAC		4475
+priscilla		MACH_PRISCILLA		PRISCILLA		4476
+innova_shpu_v2		MACH_INNOVA_SHPU_V2	INNOVA_SHPU_V2		4477
+mach_type_dep2410	MACH_MACH_TYPE_DEP2410	MACH_TYPE_DEP2410	4479
+bctre3			MACH_BCTRE3		BCTRE3			4480
+omap_m100		MACH_OMAP_M100		OMAP_M100		4481
+flo			MACH_FLO		FLO			4482
+nanobone		MACH_NANOBONE		NANOBONE		4483
+stm_b2105		MACH_STM_B2105		STM_B2105		4484
+omap4_bsc_bap_v3	MACH_OMAP4_BSC_BAP_V3	OMAP4_BSC_BAP_V3	4485
+ss1pam			MACH_SS1PAM		SS1PAM			4486
+primominiu		MACH_PRIMOMINIU		PRIMOMINIU		4488
+mrt_35hd_dualnas_e	MACH_MRT_35HD_DUALNAS_E	MRT_35HD_DUALNAS_E	4489
+kiwi			MACH_KIWI		KIWI			4490
+hw90496			MACH_HW90496		HW90496			4491
+mep2440			MACH_MEP2440		MEP2440			4492
+colibri_t30		MACH_COLIBRI_T30	COLIBRI_T30		4493
+cwv1			MACH_CWV1		CWV1			4494
+nsa325			MACH_NSA325		NSA325			4495
+dpxmtc			MACH_DPXMTC		DPXMTC			4497
+tt_stuttgart		MACH_TT_STUTTGART	TT_STUTTGART		4498
+miranda_apcii		MACH_MIRANDA_APCII	MIRANDA_APCII		4499
+mx6q_moderox		MACH_MX6Q_MODEROX	MX6Q_MODEROX		4500
+mudskipper		MACH_MUDSKIPPER		MUDSKIPPER		4501
+urania			MACH_URANIA		URANIA			4502
+stm_b2112		MACH_STM_B2112		STM_B2112		4503
+mx6q_ats_phoenix	MACH_MX6Q_ATS_PHOENIX	MX6Q_ATS_PHOENIX	4505
+stm_b2116		MACH_STM_B2116		STM_B2116		4506
+mythology		MACH_MYTHOLOGY		MYTHOLOGY		4507
+fc360v1			MACH_FC360V1		FC360V1			4508
+gps_sensor		MACH_GPS_SENSOR		GPS_SENSOR		4509
+gazelle			MACH_GAZELLE		GAZELLE			4510
+mpq8064_dma		MACH_MPQ8064_DMA	MPQ8064_DMA		4511
+wems_asd01		MACH_WEMS_ASD01		WEMS_ASD01		4512
+apalis_t30		MACH_APALIS_T30		APALIS_T30		4513
+armstonea9		MACH_ARMSTONEA9		ARMSTONEA9		4515
+omap_blazetablet	MACH_OMAP_BLAZETABLET	OMAP_BLAZETABLET	4516
+ar6mxq			MACH_AR6MXQ		AR6MXQ			4517
+ar6mxs			MACH_AR6MXS		AR6MXS			4518
+gwventana		MACH_GWVENTANA		GWVENTANA		4520
+igep0033		MACH_IGEP0033		IGEP0033		4521
+h52c1_concerto		MACH_H52C1_CONCERTO	H52C1_CONCERTO		4524
+fcmbrd			MACH_FCMBRD		FCMBRD			4525
+pcaaxs1			MACH_PCAAXS1		PCAAXS1			4526
+ls_orca			MACH_LS_ORCA		LS_ORCA			4527
+pcm051lb		MACH_PCM051LB		PCM051LB		4528
+mx6s_lp507_gvci		MACH_MX6S_LP507_GVCI	MX6S_LP507_GVCI		4529
+dido			MACH_DIDO		DIDO			4530
+swarco_itc3_9g20	MACH_SWARCO_ITC3_9G20	SWARCO_ITC3_9G20	4531
+robo_roady		MACH_ROBO_ROADY		ROBO_ROADY		4532
+rskrza1			MACH_RSKRZA1		RSKRZA1			4533
+swarco_sid		MACH_SWARCO_SID		SWARCO_SID		4534
+mx6_iwg15s_sbc		MACH_MX6_IWG15S_SBC	MX6_IWG15S_SBC		4535
+mx6q_camaro		MACH_MX6Q_CAMARO	MX6Q_CAMARO		4536
+hb6mxs			MACH_HB6MXS		HB6MXS			4537
+lager			MACH_LAGER		LAGER			4538
+lp8x4x			MACH_LP8X4X		LP8X4X			4539
+tegratab7		MACH_TEGRATAB7		TEGRATAB7		4540
+andromeda		MACH_ANDROMEDA		ANDROMEDA		4541
+bootes			MACH_BOOTES		BOOTES			4542
+nethmi			MACH_NETHMI		NETHMI			4543
+tegratab		MACH_TEGRATAB		TEGRATAB		4544
+som5_evb		MACH_SOM5_EVB		SOM5_EVB		4545
+venaticorum		MACH_VENATICORUM	VENATICORUM		4546
+stm_b2110		MACH_STM_B2110		STM_B2110		4547
+elux_hathor		MACH_ELUX_HATHOR	ELUX_HATHOR		4548
+helios_v7		MACH_HELIOS_V7		HELIOS_V7		4549
+xc10v1			MACH_XC10V1		XC10V1			4550
+cp2u			MACH_CP2U		CP2U			4551
+iap_f			MACH_IAP_F		IAP_F			4552
+iap_g			MACH_IAP_G		IAP_G			4553
+aae			MACH_AAE		AAE			4554
+pegasus			MACH_PEGASUS		PEGASUS			4555
+cygnus			MACH_CYGNUS		CYGNUS			4556
+centaurus		MACH_CENTAURUS		CENTAURUS		4557
+msm8930_qrd8930		MACH_MSM8930_QRD8930	MSM8930_QRD8930		4558
+quby_tim		MACH_QUBY_TIM		QUBY_TIM		4559
+zedi3250a		MACH_ZEDI3250A		ZEDI3250A		4560
+grus			MACH_GRUS		GRUS			4561
+apollo3			MACH_APOLLO3		APOLLO3			4562
+cowon_r7		MACH_COWON_R7		COWON_R7		4563
+tonga3			MACH_TONGA3		TONGA3			4564
+p535			MACH_P535		P535			4565
+sa3874i			MACH_SA3874I		SA3874I			4566
+mx6_navico_com		MACH_MX6_NAVICO_COM	MX6_NAVICO_COM		4567
+proxmobil2		MACH_PROXMOBIL2		PROXMOBIL2		4568
+ubinux1			MACH_UBINUX1		UBINUX1			4569
+istos			MACH_ISTOS		ISTOS			4570
+benvolio4		MACH_BENVOLIO4		BENVOLIO4		4571
+eco5_bx2		MACH_ECO5_BX2		ECO5_BX2		4572
+eukrea_cpuimx28sd	MACH_EUKREA_CPUIMX28SD	EUKREA_CPUIMX28SD	4573
+domotab			MACH_DOMOTAB		DOMOTAB			4574
+pfla03			MACH_PFLA03		PFLA03			4575

arch/arm/vdso/Makefile

@@ -0,0 +1,78 @@
+hostprogs-y := vdsomunge
+
+obj-vdso := vgettimeofday.o datapage.o
+
+# Build rules
+targets := $(obj-vdso) vdso.so vdso.so.dbg vdso.so.raw vdso.lds
+obj-vdso := $(addprefix $(obj)/, $(obj-vdso))
+
+ccflags-y := -fPIC -fno-common -fno-builtin -fno-stack-protector
+ccflags-y += -DDISABLE_BRANCH_PROFILING
+
+VDSO_LDFLAGS := -Wl,-Bsymbolic -Wl,--no-undefined -Wl,-soname=linux-vdso.so.1
+VDSO_LDFLAGS += -Wl,-z,max-page-size=4096 -Wl,-z,common-page-size=4096
+VDSO_LDFLAGS += -nostdlib -shared
+VDSO_LDFLAGS += $(call cc-ldoption, -Wl$(comma)--hash-style=sysv)
+VDSO_LDFLAGS += $(call cc-ldoption, -Wl$(comma)--build-id)
+VDSO_LDFLAGS += $(call cc-ldoption, -fuse-ld=bfd)
+
+obj-$(CONFIG_VDSO) += vdso.o
+extra-$(CONFIG_VDSO) += vdso.lds
+CPPFLAGS_vdso.lds += -P -C -U$(ARCH)
+
+CFLAGS_REMOVE_vdso.o = -pg
+
+# Force -O2 to avoid libgcc dependencies
+CFLAGS_REMOVE_vgettimeofday.o = -pg -Os
+CFLAGS_vgettimeofday.o = -O2
+
+# Disable gcov profiling for VDSO code
+GCOV_PROFILE := n
+
+# Force dependency
+$(obj)/vdso.o : $(obj)/vdso.so
+
+# Link rule for the .so file
+$(obj)/vdso.so.raw: $(src)/vdso.lds $(obj-vdso) FORCE
+	$(call if_changed,vdsold)
+
+$(obj)/vdso.so.dbg: $(obj)/vdso.so.raw $(obj)/vdsomunge FORCE
+	$(call if_changed,vdsomunge)
+
+# Strip rule for the .so file
+$(obj)/%.so: OBJCOPYFLAGS := -S
+$(obj)/%.so: $(obj)/%.so.dbg FORCE
+	$(call if_changed,objcopy)
+
+# Actual build commands
+quiet_cmd_vdsold = VDSO    $@
+      cmd_vdsold = $(CC) $(c_flags) $(VDSO_LDFLAGS) \
+                   -Wl,-T $(filter %.lds,$^) $(filter %.o,$^) -o $@
+
+quiet_cmd_vdsomunge = MUNGE   $@
+      cmd_vdsomunge = $(objtree)/$(obj)/vdsomunge $< $@
+
+#
+# Install the unstripped copy of vdso.so.dbg.  If our toolchain
+# supports build-id, install .build-id links as well.
+#
+# Cribbed from arch/x86/vdso/Makefile.
+#
+quiet_cmd_vdso_install = INSTALL $<
+define cmd_vdso_install
+	cp $< "$(MODLIB)/vdso/vdso.so"; \
+	if readelf -n $< | grep -q 'Build ID'; then \
+	  buildid=`readelf -n $< |grep 'Build ID' |sed -e 's/^.*Build ID: \(.*\)$$/\1/'`; \
+	  first=`echo $$buildid | cut -b-2`; \
+	  last=`echo $$buildid | cut -b3-`; \
+	  mkdir -p "$(MODLIB)/vdso/.build-id/$$first"; \
+	  ln -sf "../../vdso.so" "$(MODLIB)/vdso/.build-id/$$first/$$last.debug"; \
+	fi
+endef
+
+$(MODLIB)/vdso: FORCE
+	@mkdir -p $(MODLIB)/vdso
+
+PHONY += vdso_install
+vdso_install: $(obj)/vdso.so.dbg $(MODLIB)/vdso
+	$(call cmd,vdso_install)

arch/arm/vdso/datapage.S

@@ -0,0 +1,15 @@
+#include <linux/linkage.h>
+#include <asm/asm-offsets.h>
+
+	.align 2
+.L_vdso_data_ptr:
+	.long	_start - . - VDSO_DATA_SIZE
+
+ENTRY(__get_datapage)
+	.fnstart
+	adr	r0, .L_vdso_data_ptr
+	ldr	r1, [r0]
+	add	r0, r0, r1
+	bx	lr
+	.fnend
+ENDPROC(__get_datapage)

arch/arm/vdso/vdso.S

@@ -0,0 +1,34 @@
+/*
+ * Adapted from arm64 version.
+ *
+ * Copyright (C) 2012 ARM Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Author: Will Deacon <will.deacon@arm.com>
+ */
+
+#include <linux/init.h>
+#include <linux/linkage.h>
+#include <linux/const.h>
+#include <asm/page.h>
+
+	.globl vdso_start, vdso_end
+	.section .data..ro_after_init
+	.balign PAGE_SIZE
+vdso_start:
+	.incbin "arch/arm/vdso/vdso.so"
+	.balign PAGE_SIZE
+vdso_end:
+
+	.previous

arch/arm/vdso/vdso.lds.S

@@ -0,0 +1,87 @@
+/*
+ * Adapted from arm64 version.
+ *
+ * GNU linker script for the VDSO library.
+ *
+ * Copyright (C) 2012 ARM Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Author: Will Deacon <will.deacon@arm.com>
+ * Heavily based on the vDSO linker scripts for other archs.
+ */
+
+#include <linux/const.h>
+#include <asm/page.h>
+#include <asm/vdso.h>
+
+OUTPUT_FORMAT("elf32-littlearm", "elf32-bigarm", "elf32-littlearm")
+OUTPUT_ARCH(arm)
+
+SECTIONS
+{
+	PROVIDE(_start = .);
+
+	. = SIZEOF_HEADERS;
+
+	.hash		: { *(.hash) }			:text
+	.gnu.hash	: { *(.gnu.hash) }
+	.dynsym		: { *(.dynsym) }
+	.dynstr		: { *(.dynstr) }
+	.gnu.version	: { *(.gnu.version) }
+	.gnu.version_d	: { *(.gnu.version_d) }
+	.gnu.version_r	: { *(.gnu.version_r) }
+
+	.note		: { *(.note.*) }		:text	:note
+
+
+	.eh_frame_hdr	: { *(.eh_frame_hdr) }		:text	:eh_frame_hdr
+	.eh_frame	: { KEEP (*(.eh_frame)) }	:text
+
+	.dynamic	: { *(.dynamic) }		:text	:dynamic
+
+	.rodata		: { *(.rodata*) }		:text
+
+	.text		: { *(.text*) }			:text	=0xe7f001f2
+
+	.got		: { *(.got) }
+	.rel.plt	: { *(.rel.plt) }
+
+	/DISCARD/	: {
+		*(.note.GNU-stack)
+		*(.data .data.* .gnu.linkonce.d.* .sdata*)
+		*(.bss .sbss .dynbss .dynsbss)
+	}
+}
+
+/*
+ * We must supply the ELF program headers explicitly to get just one
+ * PT_LOAD segment, and set the flags explicitly to make segments read-only.
+ */
+PHDRS
+{
+	text		PT_LOAD		FLAGS(5) FILEHDR PHDRS; /* PF_R|PF_X */
+	dynamic		PT_DYNAMIC	FLAGS(4);		/* PF_R */
+	note		PT_NOTE		FLAGS(4);		/* PF_R */
+	eh_frame_hdr	PT_GNU_EH_FRAME;
+}
+
+VERSION
+{
+	LINUX_2.6 {
+	global:
+		__vdso_clock_gettime;
+		__vdso_gettimeofday;
+	local: *;
+	};
+}

arch/arm/vdso/vdsomunge.c

@@ -0,0 +1,220 @@
+/*
+ * Copyright 2015 Mentor Graphics Corporation.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ *
+ * vdsomunge - Host program which produces a shared object
+ * architecturally specified to be usable by both soft- and hard-float
+ * programs.
+ *
+ * The Procedure Call Standard for the ARM Architecture (ARM IHI
+ * 0042E) says:
+ *
+ *	6.4.1 VFP and Base Standard Compatibility
+ *
+ *	Code compiled for the VFP calling standard is compatible with
+ *	the base standard (and vice-versa) if no floating-point or
+ *	containerized vector arguments or results are used.
+ *
+ * And ELF for the ARM Architecture (ARM IHI 0044E) (Table 4-2) says:
+ *
+ *	If both EF_ARM_ABI_FLOAT_XXXX bits are clear, conformance to the
+ *	base procedure-call standard is implied.
+ *
+ * The VDSO is built with -msoft-float, as with the rest of the ARM
+ * kernel, and uses no floating point arguments or results.  The build
+ * process will produce a shared object that may or may not have the
+ * EF_ARM_ABI_FLOAT_SOFT flag set (it seems to depend on the binutils
+ * version; binutils starting with 2.24 appears to set it).  The
+ * EF_ARM_ABI_FLOAT_HARD flag should definitely not be set, and this
+ * program will error out if it is.
+ *
+ * If the soft-float flag is set, this program clears it.  That's all
+ * it does.
+ */
+
+#include <elf.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <stdarg.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/mman.h>
+#include <sys/stat.h>
+#include <sys/types.h>
+#include <unistd.h>
+
+#define swab16(x) \
+	((((x) & 0x00ff) << 8) | \
+	 (((x) & 0xff00) >> 8))
+
+#define swab32(x) \
+	((((x) & 0x000000ff) << 24) | \
+	 (((x) & 0x0000ff00) <<  8) | \
+	 (((x) & 0x00ff0000) >>  8) | \
+	 (((x) & 0xff000000) >> 24))
+
+#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
+#define HOST_ORDER ELFDATA2LSB
+#elif __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
+#define HOST_ORDER ELFDATA2MSB
+#endif
+
+/* Some of the ELF constants we'd like to use were added to <elf.h>
+ * relatively recently.
+ */
+#ifndef EF_ARM_EABI_VER5
+#define EF_ARM_EABI_VER5 0x05000000
+#endif
+
+#ifndef EF_ARM_ABI_FLOAT_SOFT
+#define EF_ARM_ABI_FLOAT_SOFT 0x200
+#endif
+
+#ifndef EF_ARM_ABI_FLOAT_HARD
+#define EF_ARM_ABI_FLOAT_HARD 0x400
+#endif
+
+static int failed;
+static const char *argv0;
+static const char *outfile;
+
+static void fail(const char *fmt, ...)
+{
+	va_list ap;
+
+	failed = 1;
+	fprintf(stderr, "%s: ", argv0);
+	va_start(ap, fmt);
+	vfprintf(stderr, fmt, ap);
+	va_end(ap);
+	exit(EXIT_FAILURE);
+}
+
+static void cleanup(void)
+{
+	if (failed && outfile != NULL)
+		unlink(outfile);
+}
+
+static Elf32_Word read_elf_word(Elf32_Word word, bool swap)
+{
+	return swap ? swab32(word) : word;
+}
+
+static Elf32_Half read_elf_half(Elf32_Half half, bool swap)
+{
+	return swap ? swab16(half) : half;
+}
+
+static void write_elf_word(Elf32_Word val, Elf32_Word *dst, bool swap)
+{
+	*dst = swap ? swab32(val) : val;
+}
+
+int main(int argc, char **argv)
+{
+	const Elf32_Ehdr *inhdr;
+	bool clear_soft_float;
+	const char *infile;
+	Elf32_Word e_flags;
+	const void *inbuf;
+	struct stat stat;
+	void *outbuf;
+	bool swap;
+	int outfd;
+	int infd;
+
+	atexit(cleanup);
+	argv0 = argv[0];
+
+	if (argc != 3)
+		fail("Usage: %s [infile] [outfile]\n", argv[0]);
+
+	infile = argv[1];
+	outfile = argv[2];
+
+	infd = open(infile, O_RDONLY);
+	if (infd < 0)
+		fail("Cannot open %s: %s\n", infile, strerror(errno));
+
+	if (fstat(infd, &stat) != 0)
+		fail("Failed stat for %s: %s\n", infile, strerror(errno));
+
+	inbuf = mmap(NULL, stat.st_size, PROT_READ, MAP_PRIVATE, infd, 0);
+	if (inbuf == MAP_FAILED)
+		fail("Failed to map %s: %s\n", infile, strerror(errno));
+
+	close(infd);
+
+	inhdr = inbuf;
+
+	if (memcmp(&inhdr->e_ident, ELFMAG, SELFMAG) != 0)
+		fail("Not an ELF file\n");
+
+	if (inhdr->e_ident[EI_CLASS] != ELFCLASS32)
+		fail("Unsupported ELF class\n");
+
+	swap = inhdr->e_ident[EI_DATA] != HOST_ORDER;
+
+	if (read_elf_half(inhdr->e_type, swap) != ET_DYN)
+		fail("Not a shared object\n");
+
+	if (read_elf_half(inhdr->e_machine, swap) != EM_ARM)
+		fail("Unsupported architecture %#x\n", inhdr->e_machine);
+
+	e_flags = read_elf_word(inhdr->e_flags, swap);
+
+	if (EF_ARM_EABI_VERSION(e_flags) != EF_ARM_EABI_VER5) {
+		fail("Unsupported EABI version %#x\n",
+		     EF_ARM_EABI_VERSION(e_flags));
+	}
+
+	if (e_flags & EF_ARM_ABI_FLOAT_HARD)
+		fail("Unexpected hard-float flag set in e_flags\n");
+
+	clear_soft_float = !!(e_flags & EF_ARM_ABI_FLOAT_SOFT);
+
+	outfd = open(outfile, O_RDWR | O_CREAT | O_TRUNC, S_IRUSR | S_IWUSR);
+	if (outfd < 0)
+		fail("Cannot open %s: %s\n", outfile, strerror(errno));
+
+	if (ftruncate(outfd, stat.st_size) != 0)
+		fail("Cannot truncate %s: %s\n", outfile, strerror(errno));
+
+	outbuf = mmap(NULL, stat.st_size, PROT_READ | PROT_WRITE, MAP_SHARED,
+		      outfd, 0);
+	if (outbuf == MAP_FAILED)
+		fail("Failed to map %s: %s\n", outfile, strerror(errno));
+
+	close(outfd);
+
+	memcpy(outbuf, inbuf, stat.st_size);
+
+	if (clear_soft_float) {
+		Elf32_Ehdr *outhdr;
+
+		outhdr = outbuf;
+		e_flags &= ~EF_ARM_ABI_FLOAT_SOFT;
+		write_elf_word(e_flags, &outhdr->e_flags, swap);
+	}
+
+	if (msync(outbuf, stat.st_size, MS_SYNC) != 0)
+		fail("Failed to sync %s: %s\n", outfile, strerror(errno));
+
+	return EXIT_SUCCESS;
+}

arch/arm/vdso/vgettimeofday.c

@@ -0,0 +1,282 @@
+/*
+ * Copyright 2015 Mentor Graphics Corporation.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2 of the
+ * License.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/compiler.h>
+#include <linux/hrtimer.h>
+#include <linux/time.h>
+#include <asm/arch_timer.h>
+#include <asm/barrier.h>
+#include <asm/bug.h>
+#include <asm/page.h>
+#include <asm/unistd.h>
+#include <asm/vdso_datapage.h>
+
+#ifndef CONFIG_AEABI
+#error This code depends on AEABI system call conventions
+#endif
+
+extern struct vdso_data *__get_datapage(void);
+
+static notrace u32 __vdso_read_begin(const struct vdso_data *vdata)
+{
+	u32 seq;
+repeat:
+	seq = ACCESS_ONCE(vdata->seq_count);
+	if (seq & 1) {
+		cpu_relax();
+		goto repeat;
+	}
+	return seq;
+}
+
+static notrace u32 vdso_read_begin(const struct vdso_data *vdata)
+{
+	u32 seq;
+
+	seq = __vdso_read_begin(vdata);
+
+	smp_rmb(); /* Pairs with smp_wmb in vdso_write_end */
+	return seq;
+}
+
+static notrace int vdso_read_retry(const struct vdso_data *vdata, u32 start)
+{
+	smp_rmb(); /* Pairs with smp_wmb in vdso_write_begin */
+	return vdata->seq_count != start;
+}
+
+static notrace long clock_gettime_fallback(clockid_t _clkid,
+					   struct timespec *_ts)
+{
+	register struct timespec *ts asm("r1") = _ts;
+	register clockid_t clkid asm("r0") = _clkid;
+	register long ret asm ("r0");
+	register long nr asm("r7") = __NR_clock_gettime;
+
+	asm volatile(
+	"	swi #0\n"
+	: "=r" (ret)
+	: "r" (clkid), "r" (ts), "r" (nr)
+	: "memory");
+
+	return ret;
+}
+
+static notrace int do_realtime_coarse(struct timespec *ts,
+				      struct vdso_data *vdata)
+{
+	u32 seq;
+
+	do {
+		seq = vdso_read_begin(vdata);
+
+		ts->tv_sec = vdata->xtime_coarse_sec;
+		ts->tv_nsec = vdata->xtime_coarse_nsec;
+
+	} while (vdso_read_retry(vdata, seq));
+
+	return 0;
+}
+
+static notrace int do_monotonic_coarse(struct timespec *ts,
+				       struct vdso_data *vdata)
+{
+	struct timespec tomono;
+	u32 seq;
+
+	do {
+		seq = vdso_read_begin(vdata);
+
+		ts->tv_sec = vdata->xtime_coarse_sec;
+		ts->tv_nsec = vdata->xtime_coarse_nsec;
+
+		tomono.tv_sec = vdata->wtm_clock_sec;
+		tomono.tv_nsec = vdata->wtm_clock_nsec;
+
+	} while (vdso_read_retry(vdata, seq));
+
+	ts->tv_sec += tomono.tv_sec;
+	timespec_add_ns(ts, tomono.tv_nsec);
+
+	return 0;
+}
+
+#ifdef CONFIG_ARM_ARCH_TIMER
+
+static notrace u64 get_ns(struct vdso_data *vdata)
+{
+	u64 cycle_delta;
+	u64 cycle_now;
+	u64 nsec;
+
+	cycle_now = arch_counter_get_cntvct();
+
+	cycle_delta = (cycle_now - vdata->cs_cycle_last) & vdata->cs_mask;
+
+	nsec = (cycle_delta * vdata->cs_mult) + vdata->xtime_clock_snsec;
+	nsec >>= vdata->cs_shift;
+
+	return nsec;
+}
+
+static notrace int do_realtime(struct timespec *ts, struct vdso_data *vdata)
+{
+	u64 nsecs;
+	u32 seq;
+
+	do {
+		seq = vdso_read_begin(vdata);
+
+		if (!vdata->tk_is_cntvct)
+			return -1;
+
+		ts->tv_sec = vdata->xtime_clock_sec;
+		nsecs = get_ns(vdata);
+
+	} while (vdso_read_retry(vdata, seq));
+
+	ts->tv_nsec = 0;
+	timespec_add_ns(ts, nsecs);
+
+	return 0;
+}
+
+static notrace int do_monotonic(struct timespec *ts, struct vdso_data *vdata)
+{
+	struct timespec tomono;
+	u64 nsecs;
+	u32 seq;
+
+	do {
+		seq = vdso_read_begin(vdata);
+
+		if (!vdata->tk_is_cntvct)
+			return -1;
+
+		ts->tv_sec = vdata->xtime_clock_sec;
+		nsecs = get_ns(vdata);
+
+		tomono.tv_sec = vdata->wtm_clock_sec;
+		tomono.tv_nsec = vdata->wtm_clock_nsec;
+
+	} while (vdso_read_retry(vdata, seq));
+
+	ts->tv_sec += tomono.tv_sec;
+	ts->tv_nsec = 0;
+	timespec_add_ns(ts, nsecs + tomono.tv_nsec);
+
+	return 0;
+}
+
+#else /* CONFIG_ARM_ARCH_TIMER */
+
+static notrace int do_realtime(struct timespec *ts, struct vdso_data *vdata)
+{
+	return -1;
+}
+
+static notrace int do_monotonic(struct timespec *ts, struct vdso_data *vdata)
+{
+	return -1;
+}
+
+#endif /* CONFIG_ARM_ARCH_TIMER */
+
+notrace int __vdso_clock_gettime(clockid_t clkid, struct timespec *ts)
+{
+	struct vdso_data *vdata;
+	int ret = -1;
+
+	vdata = __get_datapage();
+
+	switch (clkid) {
+	case CLOCK_REALTIME_COARSE:
+		ret = do_realtime_coarse(ts, vdata);
+		break;
+	case CLOCK_MONOTONIC_COARSE:
+		ret = do_monotonic_coarse(ts, vdata);
+		break;
+	case CLOCK_REALTIME:
+		ret = do_realtime(ts, vdata);
+		break;
+	case CLOCK_MONOTONIC:
+		ret = do_monotonic(ts, vdata);
+		break;
+	default:
+		break;
+	}
+
+	if (ret)
+		ret = clock_gettime_fallback(clkid, ts);
+
+	return ret;
+}
+
+static notrace long gettimeofday_fallback(struct timeval *_tv,
+					  struct timezone *_tz)
+{
+	register struct timezone *tz asm("r1") = _tz;
+	register struct timeval *tv asm("r0") = _tv;
+	register long ret asm ("r0");
+	register long nr asm("r7") = __NR_gettimeofday;
+
+	asm volatile(
+	"	swi #0\n"
+	: "=r" (ret)
+	: "r" (tv), "r" (tz), "r" (nr)
+	: "memory");
+
+	return ret;
+}
+
+notrace int __vdso_gettimeofday(struct timeval *tv, struct timezone *tz)
+{
+	struct timespec ts;
+	struct vdso_data *vdata;
+	int ret;
+
+	vdata = __get_datapage();
+
+	ret = do_realtime(&ts, vdata);
+	if (ret)
+		return gettimeofday_fallback(tv, tz);
+
+	if (tv) {
+		tv->tv_sec = ts.tv_sec;
+		tv->tv_usec = ts.tv_nsec / 1000;
+	}
+	if (tz) {
+		tz->tz_minuteswest = vdata->tz_minuteswest;
+		tz->tz_dsttime = vdata->tz_dsttime;
+	}
+
+	return ret;
+}
+
+/* Avoid unresolved references emitted by GCC */
+
+void __aeabi_unwind_cpp_pr0(void)
+{
+}
+
+void __aeabi_unwind_cpp_pr1(void)
+{
+}
+
+void __aeabi_unwind_cpp_pr2(void)
+{
+}

arch/arm/vfp/Makefile

@@ -0,0 +1,15 @@
+#
+# linux/arch/arm/vfp/Makefile
+#
+# Copyright (C) 2001 ARM Limited
+#
+
+# ccflags-y := -DDEBUG
+# asflags-y := -DDEBUG
+
+KBUILD_AFLAGS	:=$(KBUILD_AFLAGS:-msoft-float=-Wa,-mfpu=softvfp+vfp -mfloat-abi=soft)
+LDFLAGS		+=--no-warn-mismatch
+
+obj-y			+= vfp.o
+
+vfp-$(CONFIG_VFP)	+= vfpmodule.o entry.o vfphw.o vfpsingle.o vfpdouble.o

arch/arm/vfp/entry.S

@@ -0,0 +1,59 @@
+/*
+ *  linux/arch/arm/vfp/entry.S
+ *
+ *  Copyright (C) 2004 ARM Limited.
+ *  Written by Deep Blue Solutions Limited.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/init.h>
+#include <linux/linkage.h>
+#include <asm/thread_info.h>
+#include <asm/vfpmacros.h>
+#include <asm/assembler.h>
+#include <asm/asm-offsets.h>
+
+@ VFP entry point.
+@
+@  r0  = instruction opcode (32-bit ARM or two 16-bit Thumb)
+@  r2  = PC value to resume execution after successful emulation
+@  r9  = normal "successful" return address
+@  r10 = this threads thread_info structure
+@  lr  = unrecognised instruction return address
+@  IRQs enabled.
+@
+ENTRY(do_vfp)
+	inc_preempt_count r10, r4
+ 	ldr	r4, .LCvfp
+	ldr	r11, [r10, #TI_CPU]	@ CPU number
+	add	r10, r10, #TI_VFPSTATE	@ r10 = workspace
+	ldr	pc, [r4]		@ call VFP entry point
+ENDPROC(do_vfp)
+
+ENTRY(vfp_null_entry)
+	dec_preempt_count_ti r10, r4
+	ret	lr
+ENDPROC(vfp_null_entry)
+
+	.align	2
+.LCvfp:
+	.word	vfp_vector
+
+@ This code is called if the VFP does not exist. It needs to flag the
+@ failure to the VFP initialisation code.
+
+	__INIT
+ENTRY(vfp_testing_entry)
+	dec_preempt_count_ti r10, r4
+	ldr	r0, VFP_arch_address
+	str	r0, [r0]		@ set to non-zero value
+	ret	r9			@ we have handled the fault
+ENDPROC(vfp_testing_entry)
+
+	.align	2
+VFP_arch_address:
+	.word	VFP_arch
+
+	__FINIT

arch/arm/vfp/vfp.h

@@ -0,0 +1,380 @@
+/*
+ *  linux/arch/arm/vfp/vfp.h
+ *
+ *  Copyright (C) 2004 ARM Limited.
+ *  Written by Deep Blue Solutions Limited.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+static inline u32 vfp_shiftright32jamming(u32 val, unsigned int shift)
+{
+	if (shift) {
+		if (shift < 32)
+			val = val >> shift | ((val << (32 - shift)) != 0);
+		else
+			val = val != 0;
+	}
+	return val;
+}
+
+static inline u64 vfp_shiftright64jamming(u64 val, unsigned int shift)
+{
+	if (shift) {
+		if (shift < 64)
+			val = val >> shift | ((val << (64 - shift)) != 0);
+		else
+			val = val != 0;
+	}
+	return val;
+}
+
+static inline u32 vfp_hi64to32jamming(u64 val)
+{
+	u32 v;
+
+	asm(
+	"cmp	%Q1, #1		@ vfp_hi64to32jamming\n\t"
+	"movcc	%0, %R1\n\t"
+	"orrcs	%0, %R1, #1"
+	: "=r" (v) : "r" (val) : "cc");
+
+	return v;
+}
+
+static inline void add128(u64 *resh, u64 *resl, u64 nh, u64 nl, u64 mh, u64 ml)
+{
+	asm(	"adds	%Q0, %Q2, %Q4\n\t"
+		"adcs	%R0, %R2, %R4\n\t"
+		"adcs	%Q1, %Q3, %Q5\n\t"
+		"adc	%R1, %R3, %R5"
+	    : "=r" (nl), "=r" (nh)
+	    : "0" (nl), "1" (nh), "r" (ml), "r" (mh)
+	    : "cc");
+	*resh = nh;
+	*resl = nl;
+}
+
+static inline void sub128(u64 *resh, u64 *resl, u64 nh, u64 nl, u64 mh, u64 ml)
+{
+	asm(	"subs	%Q0, %Q2, %Q4\n\t"
+		"sbcs	%R0, %R2, %R4\n\t"
+		"sbcs	%Q1, %Q3, %Q5\n\t"
+		"sbc	%R1, %R3, %R5\n\t"
+	    : "=r" (nl), "=r" (nh)
+	    : "0" (nl), "1" (nh), "r" (ml), "r" (mh)
+	    : "cc");
+	*resh = nh;
+	*resl = nl;
+}
+
+static inline void mul64to128(u64 *resh, u64 *resl, u64 n, u64 m)
+{
+	u32 nh, nl, mh, ml;
+	u64 rh, rma, rmb, rl;
+
+	nl = n;
+	ml = m;
+	rl = (u64)nl * ml;
+
+	nh = n >> 32;
+	rma = (u64)nh * ml;
+
+	mh = m >> 32;
+	rmb = (u64)nl * mh;
+	rma += rmb;
+
+	rh = (u64)nh * mh;
+	rh += ((u64)(rma < rmb) << 32) + (rma >> 32);
+
+	rma <<= 32;
+	rl += rma;
+	rh += (rl < rma);
+
+	*resl = rl;
+	*resh = rh;
+}
+
+static inline void shift64left(u64 *resh, u64 *resl, u64 n)
+{
+	*resh = n >> 63;
+	*resl = n << 1;
+}
+
+static inline u64 vfp_hi64multiply64(u64 n, u64 m)
+{
+	u64 rh, rl;
+	mul64to128(&rh, &rl, n, m);
+	return rh | (rl != 0);
+}
+
+static inline u64 vfp_estimate_div128to64(u64 nh, u64 nl, u64 m)
+{
+	u64 mh, ml, remh, reml, termh, terml, z;
+
+	if (nh >= m)
+		return ~0ULL;
+	mh = m >> 32;
+	if (mh << 32 <= nh) {
+		z = 0xffffffff00000000ULL;
+	} else {
+		z = nh;
+		do_div(z, mh);
+		z <<= 32;
+	}
+	mul64to128(&termh, &terml, m, z);
+	sub128(&remh, &reml, nh, nl, termh, terml);
+	ml = m << 32;
+	while ((s64)remh < 0) {
+		z -= 0x100000000ULL;
+		add128(&remh, &reml, remh, reml, mh, ml);
+	}
+	remh = (remh << 32) | (reml >> 32);
+	if (mh << 32 <= remh) {
+		z |= 0xffffffff;
+	} else {
+		do_div(remh, mh);
+		z |= remh;
+	}
+	return z;
+}
+
+/*
+ * Operations on unpacked elements
+ */
+#define vfp_sign_negate(sign)	(sign ^ 0x8000)
+
+/*
+ * Single-precision
+ */
+struct vfp_single {
+	s16	exponent;
+	u16	sign;
+	u32	significand;
+};
+
+extern s32 vfp_get_float(unsigned int reg);
+extern void vfp_put_float(s32 val, unsigned int reg);
+
+/*
+ * VFP_SINGLE_MANTISSA_BITS - number of bits in the mantissa
+ * VFP_SINGLE_EXPONENT_BITS - number of bits in the exponent
+ * VFP_SINGLE_LOW_BITS - number of low bits in the unpacked significand
+ *  which are not propagated to the float upon packing.
+ */
+#define VFP_SINGLE_MANTISSA_BITS	(23)
+#define VFP_SINGLE_EXPONENT_BITS	(8)
+#define VFP_SINGLE_LOW_BITS		(32 - VFP_SINGLE_MANTISSA_BITS - 2)
+#define VFP_SINGLE_LOW_BITS_MASK	((1 << VFP_SINGLE_LOW_BITS) - 1)
+
+/*
+ * The bit in an unpacked float which indicates that it is a quiet NaN
+ */
+#define VFP_SINGLE_SIGNIFICAND_QNAN	(1 << (VFP_SINGLE_MANTISSA_BITS - 1 + VFP_SINGLE_LOW_BITS))
+
+/*
+ * Operations on packed single-precision numbers
+ */
+#define vfp_single_packed_sign(v)	((v) & 0x80000000)
+#define vfp_single_packed_negate(v)	((v) ^ 0x80000000)
+#define vfp_single_packed_abs(v)	((v) & ~0x80000000)
+#define vfp_single_packed_exponent(v)	(((v) >> VFP_SINGLE_MANTISSA_BITS) & ((1 << VFP_SINGLE_EXPONENT_BITS) - 1))
+#define vfp_single_packed_mantissa(v)	((v) & ((1 << VFP_SINGLE_MANTISSA_BITS) - 1))
+
+/*
+ * Unpack a single-precision float.  Note that this returns the magnitude
+ * of the single-precision float mantissa with the 1. if necessary,
+ * aligned to bit 30.
+ */
+static inline void vfp_single_unpack(struct vfp_single *s, s32 val)
+{
+	u32 significand;
+
+	s->sign = vfp_single_packed_sign(val) >> 16,
+	s->exponent = vfp_single_packed_exponent(val);
+
+	significand = (u32) val;
+	significand = (significand << (32 - VFP_SINGLE_MANTISSA_BITS)) >> 2;
+	if (s->exponent && s->exponent != 255)
+		significand |= 0x40000000;
+	s->significand = significand;
+}
+
+/*
+ * Re-pack a single-precision float.  This assumes that the float is
+ * already normalised such that the MSB is bit 30, _not_ bit 31.
+ */
+static inline s32 vfp_single_pack(struct vfp_single *s)
+{
+	u32 val;
+	val = (s->sign << 16) +
+	      (s->exponent << VFP_SINGLE_MANTISSA_BITS) +
+	      (s->significand >> VFP_SINGLE_LOW_BITS);
+	return (s32)val;
+}
+
+#define VFP_NUMBER		(1<<0)
+#define VFP_ZERO		(1<<1)
+#define VFP_DENORMAL		(1<<2)
+#define VFP_INFINITY		(1<<3)
+#define VFP_NAN			(1<<4)
+#define VFP_NAN_SIGNAL		(1<<5)
+
+#define VFP_QNAN		(VFP_NAN)
+#define VFP_SNAN		(VFP_NAN|VFP_NAN_SIGNAL)
+
+static inline int vfp_single_type(struct vfp_single *s)
+{
+	int type = VFP_NUMBER;
+	if (s->exponent == 255) {
+		if (s->significand == 0)
+			type = VFP_INFINITY;
+		else if (s->significand & VFP_SINGLE_SIGNIFICAND_QNAN)
+			type = VFP_QNAN;
+		else
+			type = VFP_SNAN;
+	} else if (s->exponent == 0) {
+		if (s->significand == 0)
+			type |= VFP_ZERO;
+		else
+			type |= VFP_DENORMAL;
+	}
+	return type;
+}
+
+#ifndef DEBUG
+#define vfp_single_normaliseround(sd,vsd,fpscr,except,func) __vfp_single_normaliseround(sd,vsd,fpscr,except)
+u32 __vfp_single_normaliseround(int sd, struct vfp_single *vs, u32 fpscr, u32 exceptions);
+#else
+u32 vfp_single_normaliseround(int sd, struct vfp_single *vs, u32 fpscr, u32 exceptions, const char *func);
+#endif
+
+/*
+ * Double-precision
+ */
+struct vfp_double {
+	s16	exponent;
+	u16	sign;
+	u64	significand;
+};
+
+/*
+ * VFP_REG_ZERO is a special register number for vfp_get_double
+ * which returns (double)0.0.  This is useful for the compare with
+ * zero instructions.
+ */
+#ifdef CONFIG_VFPv3
+#define VFP_REG_ZERO	32
+#else
+#define VFP_REG_ZERO	16
+#endif
+extern u64 vfp_get_double(unsigned int reg);
+extern void vfp_put_double(u64 val, unsigned int reg);
+
+#define VFP_DOUBLE_MANTISSA_BITS	(52)
+#define VFP_DOUBLE_EXPONENT_BITS	(11)
+#define VFP_DOUBLE_LOW_BITS		(64 - VFP_DOUBLE_MANTISSA_BITS - 2)
+#define VFP_DOUBLE_LOW_BITS_MASK	((1 << VFP_DOUBLE_LOW_BITS) - 1)
+
+/*
+ * The bit in an unpacked double which indicates that it is a quiet NaN
+ */
+#define VFP_DOUBLE_SIGNIFICAND_QNAN	(1ULL << (VFP_DOUBLE_MANTISSA_BITS - 1 + VFP_DOUBLE_LOW_BITS))
+
+/*
+ * Operations on packed single-precision numbers
+ */
+#define vfp_double_packed_sign(v)	((v) & (1ULL << 63))
+#define vfp_double_packed_negate(v)	((v) ^ (1ULL << 63))
+#define vfp_double_packed_abs(v)	((v) & ~(1ULL << 63))
+#define vfp_double_packed_exponent(v)	(((v) >> VFP_DOUBLE_MANTISSA_BITS) & ((1 << VFP_DOUBLE_EXPONENT_BITS) - 1))
+#define vfp_double_packed_mantissa(v)	((v) & ((1ULL << VFP_DOUBLE_MANTISSA_BITS) - 1))
+
+/*
+ * Unpack a double-precision float.  Note that this returns the magnitude
+ * of the double-precision float mantissa with the 1. if necessary,
+ * aligned to bit 62.
+ */
+static inline void vfp_double_unpack(struct vfp_double *s, s64 val)
+{
+	u64 significand;
+
+	s->sign = vfp_double_packed_sign(val) >> 48;
+	s->exponent = vfp_double_packed_exponent(val);
+
+	significand = (u64) val;
+	significand = (significand << (64 - VFP_DOUBLE_MANTISSA_BITS)) >> 2;
+	if (s->exponent && s->exponent != 2047)
+		significand |= (1ULL << 62);
+	s->significand = significand;
+}
+
+/*
+ * Re-pack a double-precision float.  This assumes that the float is
+ * already normalised such that the MSB is bit 30, _not_ bit 31.
+ */
+static inline s64 vfp_double_pack(struct vfp_double *s)
+{
+	u64 val;
+	val = ((u64)s->sign << 48) +
+	      ((u64)s->exponent << VFP_DOUBLE_MANTISSA_BITS) +
+	      (s->significand >> VFP_DOUBLE_LOW_BITS);
+	return (s64)val;
+}
+
+static inline int vfp_double_type(struct vfp_double *s)
+{
+	int type = VFP_NUMBER;
+	if (s->exponent == 2047) {
+		if (s->significand == 0)
+			type = VFP_INFINITY;
+		else if (s->significand & VFP_DOUBLE_SIGNIFICAND_QNAN)
+			type = VFP_QNAN;
+		else
+			type = VFP_SNAN;
+	} else if (s->exponent == 0) {
+		if (s->significand == 0)
+			type |= VFP_ZERO;
+		else
+			type |= VFP_DENORMAL;
+	}
+	return type;
+}
+
+u32 vfp_double_normaliseround(int dd, struct vfp_double *vd, u32 fpscr, u32 exceptions, const char *func);
+
+u32 vfp_estimate_sqrt_significand(u32 exponent, u32 significand);
+
+/*
+ * A special flag to tell the normalisation code not to normalise.
+ */
+#define VFP_NAN_FLAG	0x100
+
+/*
+ * A bit pattern used to indicate the initial (unset) value of the
+ * exception mask, in case nothing handles an instruction.  This
+ * doesn't include the NAN flag, which get masked out before
+ * we check for an error.
+ */
+#define VFP_EXCEPTION_ERROR	((u32)-1 & ~VFP_NAN_FLAG)
+
+/*
+ * A flag to tell vfp instruction type.
+ *  OP_SCALAR - this operation always operates in scalar mode
+ *  OP_SD - the instruction exceptionally writes to a single precision result.
+ *  OP_DD - the instruction exceptionally writes to a double precision result.
+ *  OP_SM - the instruction exceptionally reads from a single precision operand.
+ */
+#define OP_SCALAR	(1 << 0)
+#define OP_SD		(1 << 1)
+#define OP_DD		(1 << 1)
+#define OP_SM		(1 << 2)
+
+struct op {
+	u32 (* const fn)(int dd, int dn, int dm, u32 fpscr);
+	u32 flags;
+};
+
+extern void vfp_save_state(void *location, u32 fpexc);

arch/arm/vfp/vfpdouble.c

@@ -0,0 +1,1206 @@
+/*
+ *  linux/arch/arm/vfp/vfpdouble.c
+ *
+ * This code is derived in part from John R. Housers softfloat library, which
+ * carries the following notice:
+ *
+ * ===========================================================================
+ * This C source file is part of the SoftFloat IEC/IEEE Floating-point
+ * Arithmetic Package, Release 2.
+ *
+ * Written by John R. Hauser.  This work was made possible in part by the
+ * International Computer Science Institute, located at Suite 600, 1947 Center
+ * Street, Berkeley, California 94704.  Funding was partially provided by the
+ * National Science Foundation under grant MIP-9311980.  The original version
+ * of this code was written as part of a project to build a fixed-point vector
+ * processor in collaboration with the University of California at Berkeley,
+ * overseen by Profs. Nelson Morgan and John Wawrzynek.  More information
+ * is available through the web page `http://HTTP.CS.Berkeley.EDU/~jhauser/
+ * arithmetic/softfloat.html'.
+ *
+ * THIS SOFTWARE IS DISTRIBUTED AS IS, FOR FREE.  Although reasonable effort
+ * has been made to avoid it, THIS SOFTWARE MAY CONTAIN FAULTS THAT WILL AT
+ * TIMES RESULT IN INCORRECT BEHAVIOR.  USE OF THIS SOFTWARE IS RESTRICTED TO
+ * PERSONS AND ORGANIZATIONS WHO CAN AND WILL TAKE FULL RESPONSIBILITY FOR ANY
+ * AND ALL LOSSES, COSTS, OR OTHER PROBLEMS ARISING FROM ITS USE.
+ *
+ * Derivative works are acceptable, even for commercial purposes, so long as
+ * (1) they include prominent notice that the work is derivative, and (2) they
+ * include prominent notice akin to these three paragraphs for those parts of
+ * this code that are retained.
+ * ===========================================================================
+ */
+#include <linux/kernel.h>
+#include <linux/bitops.h>
+
+#include <asm/div64.h>
+#include <asm/vfp.h>
+
+#include "vfpinstr.h"
+#include "vfp.h"
+
+static struct vfp_double vfp_double_default_qnan = {
+	.exponent	= 2047,
+	.sign		= 0,
+	.significand	= VFP_DOUBLE_SIGNIFICAND_QNAN,
+};
+
+static void vfp_double_dump(const char *str, struct vfp_double *d)
+{
+	pr_debug("VFP: %s: sign=%d exponent=%d significand=%016llx\n",
+		 str, d->sign != 0, d->exponent, d->significand);
+}
+
+static void vfp_double_normalise_denormal(struct vfp_double *vd)
+{
+	int bits = 31 - fls(vd->significand >> 32);
+	if (bits == 31)
+		bits = 63 - fls(vd->significand);
+
+	vfp_double_dump("normalise_denormal: in", vd);
+
+	if (bits) {
+		vd->exponent -= bits - 1;
+		vd->significand <<= bits;
+	}
+
+	vfp_double_dump("normalise_denormal: out", vd);
+}
+
+u32 vfp_double_normaliseround(int dd, struct vfp_double *vd, u32 fpscr, u32 exceptions, const char *func)
+{
+	u64 significand, incr;
+	int exponent, shift, underflow;
+	u32 rmode;
+
+	vfp_double_dump("pack: in", vd);
+
+	/*
+	 * Infinities and NaNs are a special case.
+	 */
+	if (vd->exponent == 2047 && (vd->significand == 0 || exceptions))
+		goto pack;
+
+	/*
+	 * Special-case zero.
+	 */
+	if (vd->significand == 0) {
+		vd->exponent = 0;
+		goto pack;
+	}
+
+	exponent = vd->exponent;
+	significand = vd->significand;
+
+	shift = 32 - fls(significand >> 32);
+	if (shift == 32)
+		shift = 64 - fls(significand);
+	if (shift) {
+		exponent -= shift;
+		significand <<= shift;
+	}
+
+#ifdef DEBUG
+	vd->exponent = exponent;
+	vd->significand = significand;
+	vfp_double_dump("pack: normalised", vd);
+#endif
+
+	/*
+	 * Tiny number?
+	 */
+	underflow = exponent < 0;
+	if (underflow) {
+		significand = vfp_shiftright64jamming(significand, -exponent);
+		exponent = 0;
+#ifdef DEBUG
+		vd->exponent = exponent;
+		vd->significand = significand;
+		vfp_double_dump("pack: tiny number", vd);
+#endif
+		if (!(significand & ((1ULL << (VFP_DOUBLE_LOW_BITS + 1)) - 1)))
+			underflow = 0;
+	}
+
+	/*
+	 * Select rounding increment.
+	 */
+	incr = 0;
+	rmode = fpscr & FPSCR_RMODE_MASK;
+
+	if (rmode == FPSCR_ROUND_NEAREST) {
+		incr = 1ULL << VFP_DOUBLE_LOW_BITS;
+		if ((significand & (1ULL << (VFP_DOUBLE_LOW_BITS + 1))) == 0)
+			incr -= 1;
+	} else if (rmode == FPSCR_ROUND_TOZERO) {
+		incr = 0;
+	} else if ((rmode == FPSCR_ROUND_PLUSINF) ^ (vd->sign != 0))
+		incr = (1ULL << (VFP_DOUBLE_LOW_BITS + 1)) - 1;
+
+	pr_debug("VFP: rounding increment = 0x%08llx\n", incr);
+
+	/*
+	 * Is our rounding going to overflow?
+	 */
+	if ((significand + incr) < significand) {
+		exponent += 1;
+		significand = (significand >> 1) | (significand & 1);
+		incr >>= 1;
+#ifdef DEBUG
+		vd->exponent = exponent;
+		vd->significand = significand;
+		vfp_double_dump("pack: overflow", vd);
+#endif
+	}
+
+	/*
+	 * If any of the low bits (which will be shifted out of the
+	 * number) are non-zero, the result is inexact.
+	 */
+	if (significand & ((1 << (VFP_DOUBLE_LOW_BITS + 1)) - 1))
+		exceptions |= FPSCR_IXC;
+
+	/*
+	 * Do our rounding.
+	 */
+	significand += incr;
+
+	/*
+	 * Infinity?
+	 */
+	if (exponent >= 2046) {
+		exceptions |= FPSCR_OFC | FPSCR_IXC;
+		if (incr == 0) {
+			vd->exponent = 2045;
+			vd->significand = 0x7fffffffffffffffULL;
+		} else {
+			vd->exponent = 2047;		/* infinity */
+			vd->significand = 0;
+		}
+	} else {
+		if (significand >> (VFP_DOUBLE_LOW_BITS + 1) == 0)
+			exponent = 0;
+		if (exponent || significand > 0x8000000000000000ULL)
+			underflow = 0;
+		if (underflow)
+			exceptions |= FPSCR_UFC;
+		vd->exponent = exponent;
+		vd->significand = significand >> 1;
+	}
+
+ pack:
+	vfp_double_dump("pack: final", vd);
+	{
+		s64 d = vfp_double_pack(vd);
+		pr_debug("VFP: %s: d(d%d)=%016llx exceptions=%08x\n", func,
+			 dd, d, exceptions);
+		vfp_put_double(d, dd);
+	}
+	return exceptions;
+}
+
+/*
+ * Propagate the NaN, setting exceptions if it is signalling.
+ * 'n' is always a NaN.  'm' may be a number, NaN or infinity.
+ */
+static u32
+vfp_propagate_nan(struct vfp_double *vdd, struct vfp_double *vdn,
+		  struct vfp_double *vdm, u32 fpscr)
+{
+	struct vfp_double *nan;
+	int tn, tm = 0;
+
+	tn = vfp_double_type(vdn);
+
+	if (vdm)
+		tm = vfp_double_type(vdm);
+
+	if (fpscr & FPSCR_DEFAULT_NAN)
+		/*
+		 * Default NaN mode - always returns a quiet NaN
+		 */
+		nan = &vfp_double_default_qnan;
+	else {
+		/*
+		 * Contemporary mode - select the first signalling
+		 * NAN, or if neither are signalling, the first
+		 * quiet NAN.
+		 */
+		if (tn == VFP_SNAN || (tm != VFP_SNAN && tn == VFP_QNAN))
+			nan = vdn;
+		else
+			nan = vdm;
+		/*
+		 * Make the NaN quiet.
+		 */
+		nan->significand |= VFP_DOUBLE_SIGNIFICAND_QNAN;
+	}
+
+	*vdd = *nan;
+
+	/*
+	 * If one was a signalling NAN, raise invalid operation.
+	 */
+	return tn == VFP_SNAN || tm == VFP_SNAN ? FPSCR_IOC : VFP_NAN_FLAG;
+}
+
+/*
+ * Extended operations
+ */
+static u32 vfp_double_fabs(int dd, int unused, int dm, u32 fpscr)
+{
+	vfp_put_double(vfp_double_packed_abs(vfp_get_double(dm)), dd);
+	return 0;
+}
+
+static u32 vfp_double_fcpy(int dd, int unused, int dm, u32 fpscr)
+{
+	vfp_put_double(vfp_get_double(dm), dd);
+	return 0;
+}
+
+static u32 vfp_double_fneg(int dd, int unused, int dm, u32 fpscr)
+{
+	vfp_put_double(vfp_double_packed_negate(vfp_get_double(dm)), dd);
+	return 0;
+}
+
+static u32 vfp_double_fsqrt(int dd, int unused, int dm, u32 fpscr)
+{
+	struct vfp_double vdm, vdd;
+	int ret, tm;
+
+	vfp_double_unpack(&vdm, vfp_get_double(dm));
+	tm = vfp_double_type(&vdm);
+	if (tm & (VFP_NAN|VFP_INFINITY)) {
+		struct vfp_double *vdp = &vdd;
+
+		if (tm & VFP_NAN)
+			ret = vfp_propagate_nan(vdp, &vdm, NULL, fpscr);
+		else if (vdm.sign == 0) {
+ sqrt_copy:
+			vdp = &vdm;
+			ret = 0;
+		} else {
+ sqrt_invalid:
+			vdp = &vfp_double_default_qnan;
+			ret = FPSCR_IOC;
+		}
+		vfp_put_double(vfp_double_pack(vdp), dd);
+		return ret;
+	}
+
+	/*
+	 * sqrt(+/- 0) == +/- 0
+	 */
+	if (tm & VFP_ZERO)
+		goto sqrt_copy;
+
+	/*
+	 * Normalise a denormalised number
+	 */
+	if (tm & VFP_DENORMAL)
+		vfp_double_normalise_denormal(&vdm);
+
+	/*
+	 * sqrt(<0) = invalid
+	 */
+	if (vdm.sign)
+		goto sqrt_invalid;
+
+	vfp_double_dump("sqrt", &vdm);
+
+	/*
+	 * Estimate the square root.
+	 */
+	vdd.sign = 0;
+	vdd.exponent = ((vdm.exponent - 1023) >> 1) + 1023;
+	vdd.significand = (u64)vfp_estimate_sqrt_significand(vdm.exponent, vdm.significand >> 32) << 31;
+
+	vfp_double_dump("sqrt estimate1", &vdd);
+
+	vdm.significand >>= 1 + (vdm.exponent & 1);
+	vdd.significand += 2 + vfp_estimate_div128to64(vdm.significand, 0, vdd.significand);
+
+	vfp_double_dump("sqrt estimate2", &vdd);
+
+	/*
+	 * And now adjust.
+	 */
+	if ((vdd.significand & VFP_DOUBLE_LOW_BITS_MASK) <= 5) {
+		if (vdd.significand < 2) {
+			vdd.significand = ~0ULL;
+		} else {
+			u64 termh, terml, remh, reml;
+			vdm.significand <<= 2;
+			mul64to128(&termh, &terml, vdd.significand, vdd.significand);
+			sub128(&remh, &reml, vdm.significand, 0, termh, terml);
+			while ((s64)remh < 0) {
+				vdd.significand -= 1;
+				shift64left(&termh, &terml, vdd.significand);
+				terml |= 1;
+				add128(&remh, &reml, remh, reml, termh, terml);
+			}
+			vdd.significand |= (remh | reml) != 0;
+		}
+	}
+	vdd.significand = vfp_shiftright64jamming(vdd.significand, 1);
+
+	return vfp_double_normaliseround(dd, &vdd, fpscr, 0, "fsqrt");
+}
+
+/*
+ * Equal	:= ZC
+ * Less than	:= N
+ * Greater than	:= C
+ * Unordered	:= CV
+ */
+static u32 vfp_compare(int dd, int signal_on_qnan, int dm, u32 fpscr)
+{
+	s64 d, m;
+	u32 ret = 0;
+
+	m = vfp_get_double(dm);
+	if (vfp_double_packed_exponent(m) == 2047 && vfp_double_packed_mantissa(m)) {
+		ret |= FPSCR_C | FPSCR_V;
+		if (signal_on_qnan || !(vfp_double_packed_mantissa(m) & (1ULL << (VFP_DOUBLE_MANTISSA_BITS - 1))))
+			/*
+			 * Signalling NaN, or signalling on quiet NaN
+			 */
+			ret |= FPSCR_IOC;
+	}
+
+	d = vfp_get_double(dd);
+	if (vfp_double_packed_exponent(d) == 2047 && vfp_double_packed_mantissa(d)) {
+		ret |= FPSCR_C | FPSCR_V;
+		if (signal_on_qnan || !(vfp_double_packed_mantissa(d) & (1ULL << (VFP_DOUBLE_MANTISSA_BITS - 1))))
+			/*
+			 * Signalling NaN, or signalling on quiet NaN
+			 */
+			ret |= FPSCR_IOC;
+	}
+
+	if (ret == 0) {
+		if (d == m || vfp_double_packed_abs(d | m) == 0) {
+			/*
+			 * equal
+			 */
+			ret |= FPSCR_Z | FPSCR_C;
+		} else if (vfp_double_packed_sign(d ^ m)) {
+			/*
+			 * different signs
+			 */
+			if (vfp_double_packed_sign(d))
+				/*
+				 * d is negative, so d < m
+				 */
+				ret |= FPSCR_N;
+			else
+				/*
+				 * d is positive, so d > m
+				 */
+				ret |= FPSCR_C;
+		} else if ((vfp_double_packed_sign(d) != 0) ^ (d < m)) {
+			/*
+			 * d < m
+			 */
+			ret |= FPSCR_N;
+		} else if ((vfp_double_packed_sign(d) != 0) ^ (d > m)) {
+			/*
+			 * d > m
+			 */
+			ret |= FPSCR_C;
+		}
+	}
+
+	return ret;
+}
+
+static u32 vfp_double_fcmp(int dd, int unused, int dm, u32 fpscr)
+{
+	return vfp_compare(dd, 0, dm, fpscr);
+}
+
+static u32 vfp_double_fcmpe(int dd, int unused, int dm, u32 fpscr)
+{
+	return vfp_compare(dd, 1, dm, fpscr);
+}
+
+static u32 vfp_double_fcmpz(int dd, int unused, int dm, u32 fpscr)
+{
+	return vfp_compare(dd, 0, VFP_REG_ZERO, fpscr);
+}
+
+static u32 vfp_double_fcmpez(int dd, int unused, int dm, u32 fpscr)
+{
+	return vfp_compare(dd, 1, VFP_REG_ZERO, fpscr);
+}
+
+static u32 vfp_double_fcvts(int sd, int unused, int dm, u32 fpscr)
+{
+	struct vfp_double vdm;
+	struct vfp_single vsd;
+	int tm;
+	u32 exceptions = 0;
+
+	vfp_double_unpack(&vdm, vfp_get_double(dm));
+
+	tm = vfp_double_type(&vdm);
+
+	/*
+	 * If we have a signalling NaN, signal invalid operation.
+	 */
+	if (tm == VFP_SNAN)
+		exceptions = FPSCR_IOC;
+
+	if (tm & VFP_DENORMAL)
+		vfp_double_normalise_denormal(&vdm);
+
+	vsd.sign = vdm.sign;
+	vsd.significand = vfp_hi64to32jamming(vdm.significand);
+
+	/*
+	 * If we have an infinity or a NaN, the exponent must be 255
+	 */
+	if (tm & (VFP_INFINITY|VFP_NAN)) {
+		vsd.exponent = 255;
+		if (tm == VFP_QNAN)
+			vsd.significand |= VFP_SINGLE_SIGNIFICAND_QNAN;
+		goto pack_nan;
+	} else if (tm & VFP_ZERO)
+		vsd.exponent = 0;
+	else
+		vsd.exponent = vdm.exponent - (1023 - 127);
+
+	return vfp_single_normaliseround(sd, &vsd, fpscr, exceptions, "fcvts");
+
+ pack_nan:
+	vfp_put_float(vfp_single_pack(&vsd), sd);
+	return exceptions;
+}
+
+static u32 vfp_double_fuito(int dd, int unused, int dm, u32 fpscr)
+{
+	struct vfp_double vdm;
+	u32 m = vfp_get_float(dm);
+
+	vdm.sign = 0;
+	vdm.exponent = 1023 + 63 - 1;
+	vdm.significand = (u64)m;
+
+	return vfp_double_normaliseround(dd, &vdm, fpscr, 0, "fuito");
+}
+
+static u32 vfp_double_fsito(int dd, int unused, int dm, u32 fpscr)
+{
+	struct vfp_double vdm;
+	u32 m = vfp_get_float(dm);
+
+	vdm.sign = (m & 0x80000000) >> 16;
+	vdm.exponent = 1023 + 63 - 1;
+	vdm.significand = vdm.sign ? -m : m;
+
+	return vfp_double_normaliseround(dd, &vdm, fpscr, 0, "fsito");
+}
+
+static u32 vfp_double_ftoui(int sd, int unused, int dm, u32 fpscr)
+{
+	struct vfp_double vdm;
+	u32 d, exceptions = 0;
+	int rmode = fpscr & FPSCR_RMODE_MASK;
+	int tm;
+
+	vfp_double_unpack(&vdm, vfp_get_double(dm));
+
+	/*
+	 * Do we have a denormalised number?
+	 */
+	tm = vfp_double_type(&vdm);
+	if (tm & VFP_DENORMAL)
+		exceptions |= FPSCR_IDC;
+
+	if (tm & VFP_NAN)
+		vdm.sign = 0;
+
+	if (vdm.exponent >= 1023 + 32) {
+		d = vdm.sign ? 0 : 0xffffffff;
+		exceptions = FPSCR_IOC;
+	} else if (vdm.exponent >= 1023 - 1) {
+		int shift = 1023 + 63 - vdm.exponent;
+		u64 rem, incr = 0;
+
+		/*
+		 * 2^0 <= m < 2^32-2^8
+		 */
+		d = (vdm.significand << 1) >> shift;
+		rem = vdm.significand << (65 - shift);
+
+		if (rmode == FPSCR_ROUND_NEAREST) {
+			incr = 0x8000000000000000ULL;
+			if ((d & 1) == 0)
+				incr -= 1;
+		} else if (rmode == FPSCR_ROUND_TOZERO) {
+			incr = 0;
+		} else if ((rmode == FPSCR_ROUND_PLUSINF) ^ (vdm.sign != 0)) {
+			incr = ~0ULL;
+		}
+
+		if ((rem + incr) < rem) {
+			if (d < 0xffffffff)
+				d += 1;
+			else
+				exceptions |= FPSCR_IOC;
+		}
+
+		if (d && vdm.sign) {
+			d = 0;
+			exceptions |= FPSCR_IOC;
+		} else if (rem)
+			exceptions |= FPSCR_IXC;
+	} else {
+		d = 0;
+		if (vdm.exponent | vdm.significand) {
+			exceptions |= FPSCR_IXC;
+			if (rmode == FPSCR_ROUND_PLUSINF && vdm.sign == 0)
+				d = 1;
+			else if (rmode == FPSCR_ROUND_MINUSINF && vdm.sign) {
+				d = 0;
+				exceptions |= FPSCR_IOC;
+			}
+		}
+	}
+
+	pr_debug("VFP: ftoui: d(s%d)=%08x exceptions=%08x\n", sd, d, exceptions);
+
+	vfp_put_float(d, sd);
+
+	return exceptions;
+}
+
+static u32 vfp_double_ftouiz(int sd, int unused, int dm, u32 fpscr)
+{
+	return vfp_double_ftoui(sd, unused, dm, FPSCR_ROUND_TOZERO);
+}
+
+static u32 vfp_double_ftosi(int sd, int unused, int dm, u32 fpscr)
+{
+	struct vfp_double vdm;
+	u32 d, exceptions = 0;
+	int rmode = fpscr & FPSCR_RMODE_MASK;
+	int tm;
+
+	vfp_double_unpack(&vdm, vfp_get_double(dm));
+	vfp_double_dump("VDM", &vdm);
+
+	/*
+	 * Do we have denormalised number?
+	 */
+	tm = vfp_double_type(&vdm);
+	if (tm & VFP_DENORMAL)
+		exceptions |= FPSCR_IDC;
+
+	if (tm & VFP_NAN) {
+		d = 0;
+		exceptions |= FPSCR_IOC;
+	} else if (vdm.exponent >= 1023 + 32) {
+		d = 0x7fffffff;
+		if (vdm.sign)
+			d = ~d;
+		exceptions |= FPSCR_IOC;
+	} else if (vdm.exponent >= 1023 - 1) {
+		int shift = 1023 + 63 - vdm.exponent;	/* 58 */
+		u64 rem, incr = 0;
+
+		d = (vdm.significand << 1) >> shift;
+		rem = vdm.significand << (65 - shift);
+
+		if (rmode == FPSCR_ROUND_NEAREST) {
+			incr = 0x8000000000000000ULL;
+			if ((d & 1) == 0)
+				incr -= 1;
+		} else if (rmode == FPSCR_ROUND_TOZERO) {
+			incr = 0;
+		} else if ((rmode == FPSCR_ROUND_PLUSINF) ^ (vdm.sign != 0)) {
+			incr = ~0ULL;
+		}
+
+		if ((rem + incr) < rem && d < 0xffffffff)
+			d += 1;
+		if (d > 0x7fffffff + (vdm.sign != 0)) {
+			d = 0x7fffffff + (vdm.sign != 0);
+			exceptions |= FPSCR_IOC;
+		} else if (rem)
+			exceptions |= FPSCR_IXC;
+
+		if (vdm.sign)
+			d = -d;
+	} else {
+		d = 0;
+		if (vdm.exponent | vdm.significand) {
+			exceptions |= FPSCR_IXC;
+			if (rmode == FPSCR_ROUND_PLUSINF && vdm.sign == 0)
+				d = 1;
+			else if (rmode == FPSCR_ROUND_MINUSINF && vdm.sign)
+				d = -1;
+		}
+	}
+
+	pr_debug("VFP: ftosi: d(s%d)=%08x exceptions=%08x\n", sd, d, exceptions);
+
+	vfp_put_float((s32)d, sd);
+
+	return exceptions;
+}
+
+static u32 vfp_double_ftosiz(int dd, int unused, int dm, u32 fpscr)
+{
+	return vfp_double_ftosi(dd, unused, dm, FPSCR_ROUND_TOZERO);
+}
+
+
+static struct op fops_ext[32] = {
+	[FEXT_TO_IDX(FEXT_FCPY)]	= { vfp_double_fcpy,   0 },
+	[FEXT_TO_IDX(FEXT_FABS)]	= { vfp_double_fabs,   0 },
+	[FEXT_TO_IDX(FEXT_FNEG)]	= { vfp_double_fneg,   0 },
+	[FEXT_TO_IDX(FEXT_FSQRT)]	= { vfp_double_fsqrt,  0 },
+	[FEXT_TO_IDX(FEXT_FCMP)]	= { vfp_double_fcmp,   OP_SCALAR },
+	[FEXT_TO_IDX(FEXT_FCMPE)]	= { vfp_double_fcmpe,  OP_SCALAR },
+	[FEXT_TO_IDX(FEXT_FCMPZ)]	= { vfp_double_fcmpz,  OP_SCALAR },
+	[FEXT_TO_IDX(FEXT_FCMPEZ)]	= { vfp_double_fcmpez, OP_SCALAR },
+	[FEXT_TO_IDX(FEXT_FCVT)]	= { vfp_double_fcvts,  OP_SCALAR|OP_SD },
+	[FEXT_TO_IDX(FEXT_FUITO)]	= { vfp_double_fuito,  OP_SCALAR|OP_SM },
+	[FEXT_TO_IDX(FEXT_FSITO)]	= { vfp_double_fsito,  OP_SCALAR|OP_SM },
+	[FEXT_TO_IDX(FEXT_FTOUI)]	= { vfp_double_ftoui,  OP_SCALAR|OP_SD },
+	[FEXT_TO_IDX(FEXT_FTOUIZ)]	= { vfp_double_ftouiz, OP_SCALAR|OP_SD },
+	[FEXT_TO_IDX(FEXT_FTOSI)]	= { vfp_double_ftosi,  OP_SCALAR|OP_SD },
+	[FEXT_TO_IDX(FEXT_FTOSIZ)]	= { vfp_double_ftosiz, OP_SCALAR|OP_SD },
+};
+
+
+
+
+static u32
+vfp_double_fadd_nonnumber(struct vfp_double *vdd, struct vfp_double *vdn,
+			  struct vfp_double *vdm, u32 fpscr)
+{
+	struct vfp_double *vdp;
+	u32 exceptions = 0;
+	int tn, tm;
+
+	tn = vfp_double_type(vdn);
+	tm = vfp_double_type(vdm);
+
+	if (tn & tm & VFP_INFINITY) {
+		/*
+		 * Two infinities.  Are they different signs?
+		 */
+		if (vdn->sign ^ vdm->sign) {
+			/*
+			 * different signs -> invalid
+			 */
+			exceptions = FPSCR_IOC;
+			vdp = &vfp_double_default_qnan;
+		} else {
+			/*
+			 * same signs -> valid
+			 */
+			vdp = vdn;
+		}
+	} else if (tn & VFP_INFINITY && tm & VFP_NUMBER) {
+		/*
+		 * One infinity and one number -> infinity
+		 */
+		vdp = vdn;
+	} else {
+		/*
+		 * 'n' is a NaN of some type
+		 */
+		return vfp_propagate_nan(vdd, vdn, vdm, fpscr);
+	}
+	*vdd = *vdp;
+	return exceptions;
+}
+
+static u32
+vfp_double_add(struct vfp_double *vdd, struct vfp_double *vdn,
+	       struct vfp_double *vdm, u32 fpscr)
+{
+	u32 exp_diff;
+	u64 m_sig;
+
+	if (vdn->significand & (1ULL << 63) ||
+	    vdm->significand & (1ULL << 63)) {
+		pr_info("VFP: bad FP values in %s\n", __func__);
+		vfp_double_dump("VDN", vdn);
+		vfp_double_dump("VDM", vdm);
+	}
+
+	/*
+	 * Ensure that 'n' is the largest magnitude number.  Note that
+	 * if 'n' and 'm' have equal exponents, we do not swap them.
+	 * This ensures that NaN propagation works correctly.
+	 */
+	if (vdn->exponent < vdm->exponent) {
+		struct vfp_double *t = vdn;
+		vdn = vdm;
+		vdm = t;
+	}
+
+	/*
+	 * Is 'n' an infinity or a NaN?  Note that 'm' may be a number,
+	 * infinity or a NaN here.
+	 */
+	if (vdn->exponent == 2047)
+		return vfp_double_fadd_nonnumber(vdd, vdn, vdm, fpscr);
+
+	/*
+	 * We have two proper numbers, where 'vdn' is the larger magnitude.
+	 *
+	 * Copy 'n' to 'd' before doing the arithmetic.
+	 */
+	*vdd = *vdn;
+
+	/*
+	 * Align 'm' with the result.
+	 */
+	exp_diff = vdn->exponent - vdm->exponent;
+	m_sig = vfp_shiftright64jamming(vdm->significand, exp_diff);
+
+	/*
+	 * If the signs are different, we are really subtracting.
+	 */
+	if (vdn->sign ^ vdm->sign) {
+		m_sig = vdn->significand - m_sig;
+		if ((s64)m_sig < 0) {
+			vdd->sign = vfp_sign_negate(vdd->sign);
+			m_sig = -m_sig;
+		} else if (m_sig == 0) {
+			vdd->sign = (fpscr & FPSCR_RMODE_MASK) ==
+				      FPSCR_ROUND_MINUSINF ? 0x8000 : 0;
+		}
+	} else {
+		m_sig += vdn->significand;
+	}
+	vdd->significand = m_sig;
+
+	return 0;
+}
+
+static u32
+vfp_double_multiply(struct vfp_double *vdd, struct vfp_double *vdn,
+		    struct vfp_double *vdm, u32 fpscr)
+{
+	vfp_double_dump("VDN", vdn);
+	vfp_double_dump("VDM", vdm);
+
+	/*
+	 * Ensure that 'n' is the largest magnitude number.  Note that
+	 * if 'n' and 'm' have equal exponents, we do not swap them.
+	 * This ensures that NaN propagation works correctly.
+	 */
+	if (vdn->exponent < vdm->exponent) {
+		struct vfp_double *t = vdn;
+		vdn = vdm;
+		vdm = t;
+		pr_debug("VFP: swapping M <-> N\n");
+	}
+
+	vdd->sign = vdn->sign ^ vdm->sign;
+
+	/*
+	 * If 'n' is an infinity or NaN, handle it.  'm' may be anything.
+	 */
+	if (vdn->exponent == 2047) {
+		if (vdn->significand || (vdm->exponent == 2047 && vdm->significand))
+			return vfp_propagate_nan(vdd, vdn, vdm, fpscr);
+		if ((vdm->exponent | vdm->significand) == 0) {
+			*vdd = vfp_double_default_qnan;
+			return FPSCR_IOC;
+		}
+		vdd->exponent = vdn->exponent;
+		vdd->significand = 0;
+		return 0;
+	}
+
+	/*
+	 * If 'm' is zero, the result is always zero.  In this case,
+	 * 'n' may be zero or a number, but it doesn't matter which.
+	 */
+	if ((vdm->exponent | vdm->significand) == 0) {
+		vdd->exponent = 0;
+		vdd->significand = 0;
+		return 0;
+	}
+
+	/*
+	 * We add 2 to the destination exponent for the same reason
+	 * as the addition case - though this time we have +1 from
+	 * each input operand.
+	 */
+	vdd->exponent = vdn->exponent + vdm->exponent - 1023 + 2;
+	vdd->significand = vfp_hi64multiply64(vdn->significand, vdm->significand);
+
+	vfp_double_dump("VDD", vdd);
+	return 0;
+}
+
+#define NEG_MULTIPLY	(1 << 0)
+#define NEG_SUBTRACT	(1 << 1)
+
+static u32
+vfp_double_multiply_accumulate(int dd, int dn, int dm, u32 fpscr, u32 negate, char *func)
+{
+	struct vfp_double vdd, vdp, vdn, vdm;
+	u32 exceptions;
+
+	vfp_double_unpack(&vdn, vfp_get_double(dn));
+	if (vdn.exponent == 0 && vdn.significand)
+		vfp_double_normalise_denormal(&vdn);
+
+	vfp_double_unpack(&vdm, vfp_get_double(dm));
+	if (vdm.exponent == 0 && vdm.significand)
+		vfp_double_normalise_denormal(&vdm);
+
+	exceptions = vfp_double_multiply(&vdp, &vdn, &vdm, fpscr);
+	if (negate & NEG_MULTIPLY)
+		vdp.sign = vfp_sign_negate(vdp.sign);
+
+	vfp_double_unpack(&vdn, vfp_get_double(dd));
+	if (vdn.exponent == 0 && vdn.significand)
+		vfp_double_normalise_denormal(&vdn);
+	if (negate & NEG_SUBTRACT)
+		vdn.sign = vfp_sign_negate(vdn.sign);
+
+	exceptions |= vfp_double_add(&vdd, &vdn, &vdp, fpscr);
+
+	return vfp_double_normaliseround(dd, &vdd, fpscr, exceptions, func);
+}
+
+/*
+ * Standard operations
+ */
+
+/*
+ * sd = sd + (sn * sm)
+ */
+static u32 vfp_double_fmac(int dd, int dn, int dm, u32 fpscr)
+{
+	return vfp_double_multiply_accumulate(dd, dn, dm, fpscr, 0, "fmac");
+}
+
+/*
+ * sd = sd - (sn * sm)
+ */
+static u32 vfp_double_fnmac(int dd, int dn, int dm, u32 fpscr)
+{
+	return vfp_double_multiply_accumulate(dd, dn, dm, fpscr, NEG_MULTIPLY, "fnmac");
+}
+
+/*
+ * sd = -sd + (sn * sm)
+ */
+static u32 vfp_double_fmsc(int dd, int dn, int dm, u32 fpscr)
+{
+	return vfp_double_multiply_accumulate(dd, dn, dm, fpscr, NEG_SUBTRACT, "fmsc");
+}
+
+/*
+ * sd = -sd - (sn * sm)
+ */
+static u32 vfp_double_fnmsc(int dd, int dn, int dm, u32 fpscr)
+{
+	return vfp_double_multiply_accumulate(dd, dn, dm, fpscr, NEG_SUBTRACT | NEG_MULTIPLY, "fnmsc");
+}
+
+/*
+ * sd = sn * sm
+ */
+static u32 vfp_double_fmul(int dd, int dn, int dm, u32 fpscr)
+{
+	struct vfp_double vdd, vdn, vdm;
+	u32 exceptions;
+
+	vfp_double_unpack(&vdn, vfp_get_double(dn));
+	if (vdn.exponent == 0 && vdn.significand)
+		vfp_double_normalise_denormal(&vdn);
+
+	vfp_double_unpack(&vdm, vfp_get_double(dm));
+	if (vdm.exponent == 0 && vdm.significand)
+		vfp_double_normalise_denormal(&vdm);
+
+	exceptions = vfp_double_multiply(&vdd, &vdn, &vdm, fpscr);
+	return vfp_double_normaliseround(dd, &vdd, fpscr, exceptions, "fmul");
+}
+
+/*
+ * sd = -(sn * sm)
+ */
+static u32 vfp_double_fnmul(int dd, int dn, int dm, u32 fpscr)
+{
+	struct vfp_double vdd, vdn, vdm;
+	u32 exceptions;
+
+	vfp_double_unpack(&vdn, vfp_get_double(dn));
+	if (vdn.exponent == 0 && vdn.significand)
+		vfp_double_normalise_denormal(&vdn);
+
+	vfp_double_unpack(&vdm, vfp_get_double(dm));
+	if (vdm.exponent == 0 && vdm.significand)
+		vfp_double_normalise_denormal(&vdm);
+
+	exceptions = vfp_double_multiply(&vdd, &vdn, &vdm, fpscr);
+	vdd.sign = vfp_sign_negate(vdd.sign);
+
+	return vfp_double_normaliseround(dd, &vdd, fpscr, exceptions, "fnmul");
+}
+
+/*
+ * sd = sn + sm
+ */
+static u32 vfp_double_fadd(int dd, int dn, int dm, u32 fpscr)
+{
+	struct vfp_double vdd, vdn, vdm;
+	u32 exceptions;
+
+	vfp_double_unpack(&vdn, vfp_get_double(dn));
+	if (vdn.exponent == 0 && vdn.significand)
+		vfp_double_normalise_denormal(&vdn);
+
+	vfp_double_unpack(&vdm, vfp_get_double(dm));
+	if (vdm.exponent == 0 && vdm.significand)
+		vfp_double_normalise_denormal(&vdm);
+
+	exceptions = vfp_double_add(&vdd, &vdn, &vdm, fpscr);
+
+	return vfp_double_normaliseround(dd, &vdd, fpscr, exceptions, "fadd");
+}
+
+/*
+ * sd = sn - sm
+ */
+static u32 vfp_double_fsub(int dd, int dn, int dm, u32 fpscr)
+{
+	struct vfp_double vdd, vdn, vdm;
+	u32 exceptions;
+
+	vfp_double_unpack(&vdn, vfp_get_double(dn));
+	if (vdn.exponent == 0 && vdn.significand)
+		vfp_double_normalise_denormal(&vdn);
+
+	vfp_double_unpack(&vdm, vfp_get_double(dm));
+	if (vdm.exponent == 0 && vdm.significand)
+		vfp_double_normalise_denormal(&vdm);
+
+	/*
+	 * Subtraction is like addition, but with a negated operand.
+	 */
+	vdm.sign = vfp_sign_negate(vdm.sign);
+
+	exceptions = vfp_double_add(&vdd, &vdn, &vdm, fpscr);
+
+	return vfp_double_normaliseround(dd, &vdd, fpscr, exceptions, "fsub");
+}
+
+/*
+ * sd = sn / sm
+ */
+static u32 vfp_double_fdiv(int dd, int dn, int dm, u32 fpscr)
+{
+	struct vfp_double vdd, vdn, vdm;
+	u32 exceptions = 0;
+	int tm, tn;
+
+	vfp_double_unpack(&vdn, vfp_get_double(dn));
+	vfp_double_unpack(&vdm, vfp_get_double(dm));
+
+	vdd.sign = vdn.sign ^ vdm.sign;
+
+	tn = vfp_double_type(&vdn);
+	tm = vfp_double_type(&vdm);
+
+	/*
+	 * Is n a NAN?
+	 */
+	if (tn & VFP_NAN)
+		goto vdn_nan;
+
+	/*
+	 * Is m a NAN?
+	 */
+	if (tm & VFP_NAN)
+		goto vdm_nan;
+
+	/*
+	 * If n and m are infinity, the result is invalid
+	 * If n and m are zero, the result is invalid
+	 */
+	if (tm & tn & (VFP_INFINITY|VFP_ZERO))
+		goto invalid;
+
+	/*
+	 * If n is infinity, the result is infinity
+	 */
+	if (tn & VFP_INFINITY)
+		goto infinity;
+
+	/*
+	 * If m is zero, raise div0 exceptions
+	 */
+	if (tm & VFP_ZERO)
+		goto divzero;
+
+	/*
+	 * If m is infinity, or n is zero, the result is zero
+	 */
+	if (tm & VFP_INFINITY || tn & VFP_ZERO)
+		goto zero;
+
+	if (tn & VFP_DENORMAL)
+		vfp_double_normalise_denormal(&vdn);
+	if (tm & VFP_DENORMAL)
+		vfp_double_normalise_denormal(&vdm);
+
+	/*
+	 * Ok, we have two numbers, we can perform division.
+	 */
+	vdd.exponent = vdn.exponent - vdm.exponent + 1023 - 1;
+	vdm.significand <<= 1;
+	if (vdm.significand <= (2 * vdn.significand)) {
+		vdn.significand >>= 1;
+		vdd.exponent++;
+	}
+	vdd.significand = vfp_estimate_div128to64(vdn.significand, 0, vdm.significand);
+	if ((vdd.significand & 0x1ff) <= 2) {
+		u64 termh, terml, remh, reml;
+		mul64to128(&termh, &terml, vdm.significand, vdd.significand);
+		sub128(&remh, &reml, vdn.significand, 0, termh, terml);
+		while ((s64)remh < 0) {
+			vdd.significand -= 1;
+			add128(&remh, &reml, remh, reml, 0, vdm.significand);
+		}
+		vdd.significand |= (reml != 0);
+	}
+	return vfp_double_normaliseround(dd, &vdd, fpscr, 0, "fdiv");
+
+ vdn_nan:
+	exceptions = vfp_propagate_nan(&vdd, &vdn, &vdm, fpscr);
+ pack:
+	vfp_put_double(vfp_double_pack(&vdd), dd);
+	return exceptions;
+
+ vdm_nan:
+	exceptions = vfp_propagate_nan(&vdd, &vdm, &vdn, fpscr);
+	goto pack;
+
+ zero:
+	vdd.exponent = 0;
+	vdd.significand = 0;
+	goto pack;
+
+ divzero:
+	exceptions = FPSCR_DZC;
+ infinity:
+	vdd.exponent = 2047;
+	vdd.significand = 0;
+	goto pack;
+
+ invalid:
+	vfp_put_double(vfp_double_pack(&vfp_double_default_qnan), dd);
+	return FPSCR_IOC;
+}
+
+static struct op fops[16] = {
+	[FOP_TO_IDX(FOP_FMAC)]	= { vfp_double_fmac,  0 },
+	[FOP_TO_IDX(FOP_FNMAC)]	= { vfp_double_fnmac, 0 },
+	[FOP_TO_IDX(FOP_FMSC)]	= { vfp_double_fmsc,  0 },
+	[FOP_TO_IDX(FOP_FNMSC)]	= { vfp_double_fnmsc, 0 },
+	[FOP_TO_IDX(FOP_FMUL)]	= { vfp_double_fmul,  0 },
+	[FOP_TO_IDX(FOP_FNMUL)]	= { vfp_double_fnmul, 0 },
+	[FOP_TO_IDX(FOP_FADD)]	= { vfp_double_fadd,  0 },
+	[FOP_TO_IDX(FOP_FSUB)]	= { vfp_double_fsub,  0 },
+	[FOP_TO_IDX(FOP_FDIV)]	= { vfp_double_fdiv,  0 },
+};
+
+#define FREG_BANK(x)	((x) & 0x0c)
+#define FREG_IDX(x)	((x) & 3)
+
+u32 vfp_double_cpdo(u32 inst, u32 fpscr)
+{
+	u32 op = inst & FOP_MASK;
+	u32 exceptions = 0;
+	unsigned int dest;
+	unsigned int dn = vfp_get_dn(inst);
+	unsigned int dm;
+	unsigned int vecitr, veclen, vecstride;
+	struct op *fop;
+
+	vecstride = (1 + ((fpscr & FPSCR_STRIDE_MASK) == FPSCR_STRIDE_MASK));
+
+	fop = (op == FOP_EXT) ? &fops_ext[FEXT_TO_IDX(inst)] : &fops[FOP_TO_IDX(op)];
+
+	/*
+	 * fcvtds takes an sN register number as destination, not dN.
+	 * It also always operates on scalars.
+	 */
+	if (fop->flags & OP_SD)
+		dest = vfp_get_sd(inst);
+	else
+		dest = vfp_get_dd(inst);
+
+	/*
+	 * f[us]ito takes a sN operand, not a dN operand.
+	 */
+	if (fop->flags & OP_SM)
+		dm = vfp_get_sm(inst);
+	else
+		dm = vfp_get_dm(inst);
+
+	/*
+	 * If destination bank is zero, vector length is always '1'.
+	 * ARM DDI0100F C5.1.3, C5.3.2.
+	 */
+	if ((fop->flags & OP_SCALAR) || (FREG_BANK(dest) == 0))
+		veclen = 0;
+	else
+		veclen = fpscr & FPSCR_LENGTH_MASK;
+
+	pr_debug("VFP: vecstride=%u veclen=%u\n", vecstride,
+		 (veclen >> FPSCR_LENGTH_BIT) + 1);
+
+	if (!fop->fn)
+		goto invalid;
+
+	for (vecitr = 0; vecitr <= veclen; vecitr += 1 << FPSCR_LENGTH_BIT) {
+		u32 except;
+		char type;
+
+		type = fop->flags & OP_SD ? 's' : 'd';
+		if (op == FOP_EXT)
+			pr_debug("VFP: itr%d (%c%u) = op[%u] (d%u)\n",
+				 vecitr >> FPSCR_LENGTH_BIT,
+				 type, dest, dn, dm);
+		else
+			pr_debug("VFP: itr%d (%c%u) = (d%u) op[%u] (d%u)\n",
+				 vecitr >> FPSCR_LENGTH_BIT,
+				 type, dest, dn, FOP_TO_IDX(op), dm);
+
+		except = fop->fn(dest, dn, dm, fpscr);
+		pr_debug("VFP: itr%d: exceptions=%08x\n",
+			 vecitr >> FPSCR_LENGTH_BIT, except);
+
+		exceptions |= except;
+
+		/*
+		 * CHECK: It appears to be undefined whether we stop when
+		 * we encounter an exception.  We continue.
+		 */
+		dest = FREG_BANK(dest) + ((FREG_IDX(dest) + vecstride) & 3);
+		dn = FREG_BANK(dn) + ((FREG_IDX(dn) + vecstride) & 3);
+		if (FREG_BANK(dm) != 0)
+			dm = FREG_BANK(dm) + ((FREG_IDX(dm) + vecstride) & 3);
+	}
+	return exceptions;
+
+ invalid:
+	return ~0;
+}

arch/arm/vfp/vfphw.S

@@ -0,0 +1,323 @@
+/*
+ *  linux/arch/arm/vfp/vfphw.S
+ *
+ *  Copyright (C) 2004 ARM Limited.
+ *  Written by Deep Blue Solutions Limited.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This code is called from the kernel's undefined instruction trap.
+ * r9 holds the return address for successful handling.
+ * lr holds the return address for unrecognised instructions.
+ * r10 points at the start of the private FP workspace in the thread structure
+ * sp points to a struct pt_regs (as defined in include/asm/proc/ptrace.h)
+ */
+#include <linux/init.h>
+#include <linux/linkage.h>
+#include <asm/thread_info.h>
+#include <asm/vfpmacros.h>
+#include <linux/kern_levels.h>
+#include <asm/assembler.h>
+#include <asm/asm-offsets.h>
+
+	.macro	DBGSTR, str
+#ifdef DEBUG
+	stmfd	sp!, {r0-r3, ip, lr}
+	ldr	r0, =1f
+	bl	printk
+	ldmfd	sp!, {r0-r3, ip, lr}
+
+	.pushsection .rodata, "a"
+1:	.ascii	KERN_DEBUG "VFP: \str\n"
+	.byte	0
+	.previous
+#endif
+	.endm
+
+	.macro  DBGSTR1, str, arg
+#ifdef DEBUG
+	stmfd	sp!, {r0-r3, ip, lr}
+	mov	r1, \arg
+	ldr	r0, =1f
+	bl	printk
+	ldmfd	sp!, {r0-r3, ip, lr}
+
+	.pushsection .rodata, "a"
+1:	.ascii	KERN_DEBUG "VFP: \str\n"
+	.byte	0
+	.previous
+#endif
+	.endm
+
+	.macro  DBGSTR3, str, arg1, arg2, arg3
+#ifdef DEBUG
+	stmfd	sp!, {r0-r3, ip, lr}
+	mov	r3, \arg3
+	mov	r2, \arg2
+	mov	r1, \arg1
+	ldr	r0, =1f
+	bl	printk
+	ldmfd	sp!, {r0-r3, ip, lr}
+
+	.pushsection .rodata, "a"
+1:	.ascii	KERN_DEBUG "VFP: \str\n"
+	.byte	0
+	.previous
+#endif
+	.endm
+
+
+@ VFP hardware support entry point.
+@
+@  r0  = instruction opcode (32-bit ARM or two 16-bit Thumb)
+@  r2  = PC value to resume execution after successful emulation
+@  r9  = normal "successful" return address
+@  r10 = vfp_state union
+@  r11 = CPU number
+@  lr  = unrecognised instruction return address
+@  IRQs enabled.
+ENTRY(vfp_support_entry)
+	DBGSTR3	"instr %08x pc %08x state %p", r0, r2, r10
+
+	ldr	r3, [sp, #S_PSR]	@ Neither lazy restore nor FP exceptions
+	and	r3, r3, #MODE_MASK	@ are supported in kernel mode
+	teq	r3, #USR_MODE
+	bne	vfp_kmode_exception	@ Returns through lr
+
+	VFPFMRX	r1, FPEXC		@ Is the VFP enabled?
+	DBGSTR1	"fpexc %08x", r1
+	tst	r1, #FPEXC_EN
+	bne	look_for_VFP_exceptions	@ VFP is already enabled
+
+	DBGSTR1 "enable %x", r10
+	ldr	r3, vfp_current_hw_state_address
+	orr	r1, r1, #FPEXC_EN	@ user FPEXC has the enable bit set
+	ldr	r4, [r3, r11, lsl #2]	@ vfp_current_hw_state pointer
+	bic	r5, r1, #FPEXC_EX	@ make sure exceptions are disabled
+	cmp	r4, r10			@ this thread owns the hw context?
+#ifndef CONFIG_SMP
+	@ For UP, checking that this thread owns the hw context is
+	@ sufficient to determine that the hardware state is valid.
+	beq	vfp_hw_state_valid
+
+	@ On UP, we lazily save the VFP context.  As a different
+	@ thread wants ownership of the VFP hardware, save the old
+	@ state if there was a previous (valid) owner.
+
+	VFPFMXR	FPEXC, r5		@ enable VFP, disable any pending
+					@ exceptions, so we can get at the
+					@ rest of it
+
+	DBGSTR1	"save old state %p", r4
+	cmp	r4, #0			@ if the vfp_current_hw_state is NULL
+	beq	vfp_reload_hw		@ then the hw state needs reloading
+	VFPFSTMIA r4, r5		@ save the working registers
+	VFPFMRX	r5, FPSCR		@ current status
+#ifndef CONFIG_CPU_FEROCEON
+	tst	r1, #FPEXC_EX		@ is there additional state to save?
+	beq	1f
+	VFPFMRX	r6, FPINST		@ FPINST (only if FPEXC.EX is set)
+	tst	r1, #FPEXC_FP2V		@ is there an FPINST2 to read?
+	beq	1f
+	VFPFMRX	r8, FPINST2		@ FPINST2 if needed (and present)
+1:
+#endif
+	stmia	r4, {r1, r5, r6, r8}	@ save FPEXC, FPSCR, FPINST, FPINST2
+vfp_reload_hw:
+
+#else
+	@ For SMP, if this thread does not own the hw context, then we
+	@ need to reload it.  No need to save the old state as on SMP,
+	@ we always save the state when we switch away from a thread.
+	bne	vfp_reload_hw
+
+	@ This thread has ownership of the current hardware context.
+	@ However, it may have been migrated to another CPU, in which
+	@ case the saved state is newer than the hardware context.
+	@ Check this by looking at the CPU number which the state was
+	@ last loaded onto.
+	ldr	ip, [r10, #VFP_CPU]
+	teq	ip, r11
+	beq	vfp_hw_state_valid
+
+vfp_reload_hw:
+	@ We're loading this threads state into the VFP hardware. Update
+	@ the CPU number which contains the most up to date VFP context.
+	str	r11, [r10, #VFP_CPU]
+
+	VFPFMXR	FPEXC, r5		@ enable VFP, disable any pending
+					@ exceptions, so we can get at the
+					@ rest of it
+#endif
+
+	DBGSTR1	"load state %p", r10
+	str	r10, [r3, r11, lsl #2]	@ update the vfp_current_hw_state pointer
+					@ Load the saved state back into the VFP
+	VFPFLDMIA r10, r5		@ reload the working registers while
+					@ FPEXC is in a safe state
+	ldmia	r10, {r1, r5, r6, r8}	@ load FPEXC, FPSCR, FPINST, FPINST2
+#ifndef CONFIG_CPU_FEROCEON
+	tst	r1, #FPEXC_EX		@ is there additional state to restore?
+	beq	1f
+	VFPFMXR	FPINST, r6		@ restore FPINST (only if FPEXC.EX is set)
+	tst	r1, #FPEXC_FP2V		@ is there an FPINST2 to write?
+	beq	1f
+	VFPFMXR	FPINST2, r8		@ FPINST2 if needed (and present)
+1:
+#endif
+	VFPFMXR	FPSCR, r5		@ restore status
+
+@ The context stored in the VFP hardware is up to date with this thread
+vfp_hw_state_valid:
+	tst	r1, #FPEXC_EX
+	bne	process_exception	@ might as well handle the pending
+					@ exception before retrying branch
+					@ out before setting an FPEXC that
+					@ stops us reading stuff
+	VFPFMXR	FPEXC, r1		@ Restore FPEXC last
+	sub	r2, r2, #4		@ Retry current instruction - if Thumb
+	str	r2, [sp, #S_PC]		@ mode it's two 16-bit instructions,
+					@ else it's one 32-bit instruction, so
+					@ always subtract 4 from the following
+					@ instruction address.
+	dec_preempt_count_ti r10, r4
+	ret	r9			@ we think we have handled things
+
+
+look_for_VFP_exceptions:
+	@ Check for synchronous or asynchronous exception
+	tst	r1, #FPEXC_EX | FPEXC_DEX
+	bne	process_exception
+	@ On some implementations of the VFP subarch 1, setting FPSCR.IXE
+	@ causes all the CDP instructions to be bounced synchronously without
+	@ setting the FPEXC.EX bit
+	VFPFMRX	r5, FPSCR
+	tst	r5, #FPSCR_IXE
+	bne	process_exception
+
+	tst	r5, #FPSCR_LENGTH_MASK
+	beq	skip
+	orr	r1, r1, #FPEXC_DEX
+	b	process_exception
+skip:
+
+	@ Fall into hand on to next handler - appropriate coproc instr
+	@ not recognised by VFP
+
+	DBGSTR	"not VFP"
+	dec_preempt_count_ti r10, r4
+	ret	lr
+
+process_exception:
+	DBGSTR	"bounce"
+	mov	r2, sp			@ nothing stacked - regdump is at TOS
+	mov	lr, r9			@ setup for a return to the user code.
+
+	@ Now call the C code to package up the bounce to the support code
+	@   r0 holds the trigger instruction
+	@   r1 holds the FPEXC value
+	@   r2 pointer to register dump
+	b	VFP_bounce		@ we have handled this - the support
+					@ code will raise an exception if
+					@ required. If not, the user code will
+					@ retry the faulted instruction
+ENDPROC(vfp_support_entry)
+
+ENTRY(vfp_save_state)
+	@ Save the current VFP state
+	@ r0 - save location
+	@ r1 - FPEXC
+	DBGSTR1	"save VFP state %p", r0
+	VFPFSTMIA r0, r2		@ save the working registers
+	VFPFMRX	r2, FPSCR		@ current status
+	tst	r1, #FPEXC_EX		@ is there additional state to save?
+	beq	1f
+	VFPFMRX	r3, FPINST		@ FPINST (only if FPEXC.EX is set)
+	tst	r1, #FPEXC_FP2V		@ is there an FPINST2 to read?
+	beq	1f
+	VFPFMRX	r12, FPINST2		@ FPINST2 if needed (and present)
+1:
+	stmia	r0, {r1, r2, r3, r12}	@ save FPEXC, FPSCR, FPINST, FPINST2
+	ret	lr
+ENDPROC(vfp_save_state)
+
+	.align
+vfp_current_hw_state_address:
+	.word	vfp_current_hw_state
+
+	.macro	tbl_branch, base, tmp, shift
+#ifdef CONFIG_THUMB2_KERNEL
+	adr	\tmp, 1f
+	add	\tmp, \tmp, \base, lsl \shift
+	ret	\tmp
+#else
+	add	pc, pc, \base, lsl \shift
+	mov	r0, r0
+#endif
+1:
+	.endm
+
+ENTRY(vfp_get_float)
+	tbl_branch r0, r3, #3
+	.irp	dr,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+1:	mrc	p10, 0, r0, c\dr, c0, 0	@ fmrs	r0, s0
+	ret	lr
+	.org	1b + 8
+1:	mrc	p10, 0, r0, c\dr, c0, 4	@ fmrs	r0, s1
+	ret	lr
+	.org	1b + 8
+	.endr
+ENDPROC(vfp_get_float)
+
+ENTRY(vfp_put_float)
+	tbl_branch r1, r3, #3
+	.irp	dr,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+1:	mcr	p10, 0, r0, c\dr, c0, 0	@ fmsr	r0, s0
+	ret	lr
+	.org	1b + 8
+1:	mcr	p10, 0, r0, c\dr, c0, 4	@ fmsr	r0, s1
+	ret	lr
+	.org	1b + 8
+	.endr
+ENDPROC(vfp_put_float)
+
+ENTRY(vfp_get_double)
+	tbl_branch r0, r3, #3
+	.irp	dr,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+1:	fmrrd	r0, r1, d\dr
+	ret	lr
+	.org	1b + 8
+	.endr
+#ifdef CONFIG_VFPv3
+	@ d16 - d31 registers
+	.irp	dr,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+1:	mrrc	p11, 3, r0, r1, c\dr	@ fmrrd	r0, r1, d\dr
+	ret	lr
+	.org	1b + 8
+	.endr
+#endif
+
+	@ virtual register 16 (or 32 if VFPv3) for compare with zero
+	mov	r0, #0
+	mov	r1, #0
+	ret	lr
+ENDPROC(vfp_get_double)
+
+ENTRY(vfp_put_double)
+	tbl_branch r2, r3, #3
+	.irp	dr,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+1:	fmdrr	d\dr, r0, r1
+	ret	lr
+	.org	1b + 8
+	.endr
+#ifdef CONFIG_VFPv3
+	@ d16 - d31 registers
+	.irp	dr,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+1:	mcrr	p11, 3, r0, r1, c\dr	@ fmdrr	r0, r1, d\dr
+	ret	lr
+	.org	1b + 8
+	.endr
+#endif
+ENDPROC(vfp_put_double)

arch/arm/vfp/vfpinstr.h

@@ -0,0 +1,88 @@
+/*
+ *  linux/arch/arm/vfp/vfpinstr.h
+ *
+ *  Copyright (C) 2004 ARM Limited.
+ *  Written by Deep Blue Solutions Limited.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * VFP instruction masks.
+ */
+#define INST_CPRTDO(inst)	(((inst) & 0x0f000000) == 0x0e000000)
+#define INST_CPRT(inst)		((inst) & (1 << 4))
+#define INST_CPRT_L(inst)	((inst) & (1 << 20))
+#define INST_CPRT_Rd(inst)	(((inst) & (15 << 12)) >> 12)
+#define INST_CPRT_OP(inst)	(((inst) >> 21) & 7)
+#define INST_CPNUM(inst)	((inst) & 0xf00)
+#define CPNUM(cp)		((cp) << 8)
+
+#define FOP_MASK	(0x00b00040)
+#define FOP_FMAC	(0x00000000)
+#define FOP_FNMAC	(0x00000040)
+#define FOP_FMSC	(0x00100000)
+#define FOP_FNMSC	(0x00100040)
+#define FOP_FMUL	(0x00200000)
+#define FOP_FNMUL	(0x00200040)
+#define FOP_FADD	(0x00300000)
+#define FOP_FSUB	(0x00300040)
+#define FOP_FDIV	(0x00800000)
+#define FOP_EXT		(0x00b00040)
+
+#define FOP_TO_IDX(inst)	((inst & 0x00b00000) >> 20 | (inst & (1 << 6)) >> 4)
+
+#define FEXT_MASK	(0x000f0080)
+#define FEXT_FCPY	(0x00000000)
+#define FEXT_FABS	(0x00000080)
+#define FEXT_FNEG	(0x00010000)
+#define FEXT_FSQRT	(0x00010080)
+#define FEXT_FCMP	(0x00040000)
+#define FEXT_FCMPE	(0x00040080)
+#define FEXT_FCMPZ	(0x00050000)
+#define FEXT_FCMPEZ	(0x00050080)
+#define FEXT_FCVT	(0x00070080)
+#define FEXT_FUITO	(0x00080000)
+#define FEXT_FSITO	(0x00080080)
+#define FEXT_FTOUI	(0x000c0000)
+#define FEXT_FTOUIZ	(0x000c0080)
+#define FEXT_FTOSI	(0x000d0000)
+#define FEXT_FTOSIZ	(0x000d0080)
+
+#define FEXT_TO_IDX(inst)	((inst & 0x000f0000) >> 15 | (inst & (1 << 7)) >> 7)
+
+#define vfp_get_sd(inst)	((inst & 0x0000f000) >> 11 | (inst & (1 << 22)) >> 22)
+#define vfp_get_dd(inst)	((inst & 0x0000f000) >> 12 | (inst & (1 << 22)) >> 18)
+#define vfp_get_sm(inst)	((inst & 0x0000000f) << 1 | (inst & (1 << 5)) >> 5)
+#define vfp_get_dm(inst)	((inst & 0x0000000f) | (inst & (1 << 5)) >> 1)
+#define vfp_get_sn(inst)	((inst & 0x000f0000) >> 15 | (inst & (1 << 7)) >> 7)
+#define vfp_get_dn(inst)	((inst & 0x000f0000) >> 16 | (inst & (1 << 7)) >> 3)
+
+#define vfp_single(inst)	(((inst) & 0x0000f00) == 0xa00)
+
+#define FPSCR_N	(1 << 31)
+#define FPSCR_Z	(1 << 30)
+#define FPSCR_C (1 << 29)
+#define FPSCR_V	(1 << 28)
+
+/*
+ * Since we aren't building with -mfpu=vfp, we need to code
+ * these instructions using their MRC/MCR equivalents.
+ */
+#define vfpreg(_vfp_) #_vfp_
+
+#define fmrx(_vfp_) ({			\
+	u32 __v;			\
+	asm("mrc p10, 7, %0, " vfpreg(_vfp_) ", cr0, 0 @ fmrx	%0, " #_vfp_	\
+	    : "=r" (__v) : : "cc");	\
+	__v;				\
+ })
+
+#define fmxr(_vfp_,_var_)		\
+	asm("mcr p10, 7, %0, " vfpreg(_vfp_) ", cr0, 0 @ fmxr	" #_vfp_ ", %0"	\
+	   : : "r" (_var_) : "cc")
+
+u32 vfp_single_cpdo(u32 inst, u32 fpscr);
+u32 vfp_single_cprt(u32 inst, u32 fpscr, struct pt_regs *regs);
+
+u32 vfp_double_cpdo(u32 inst, u32 fpscr);

arch/arm/vfp/vfpmodule.c

@@ -0,0 +1,826 @@
+/*
+ *  linux/arch/arm/vfp/vfpmodule.c
+ *
+ *  Copyright (C) 2004 ARM Limited.
+ *  Written by Deep Blue Solutions Limited.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/types.h>
+#include <linux/cpu.h>
+#include <linux/cpu_pm.h>
+#include <linux/hardirq.h>
+#include <linux/kernel.h>
+#include <linux/notifier.h>
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/smp.h>
+#include <linux/init.h>
+#include <linux/uaccess.h>
+#include <linux/user.h>
+#include <linux/export.h>
+
+#include <asm/cp15.h>
+#include <asm/cputype.h>
+#include <asm/system_info.h>
+#include <asm/thread_notify.h>
+#include <asm/vfp.h>
+
+#include "vfpinstr.h"
+#include "vfp.h"
+
+/*
+ * Our undef handlers (in entry.S)
+ */
+void vfp_testing_entry(void);
+void vfp_support_entry(void);
+void vfp_null_entry(void);
+
+void (*vfp_vector)(void) = vfp_null_entry;
+
+/*
+ * Dual-use variable.
+ * Used in startup: set to non-zero if VFP checks fail
+ * After startup, holds VFP architecture
+ */
+unsigned int VFP_arch;
+
+/*
+ * The pointer to the vfpstate structure of the thread which currently
+ * owns the context held in the VFP hardware, or NULL if the hardware
+ * context is invalid.
+ *
+ * For UP, this is sufficient to tell which thread owns the VFP context.
+ * However, for SMP, we also need to check the CPU number stored in the
+ * saved state too to catch migrations.
+ */
+union vfp_state *vfp_current_hw_state[NR_CPUS];
+
+/*
+ * Is 'thread's most up to date state stored in this CPUs hardware?
+ * Must be called from non-preemptible context.
+ */
+static bool vfp_state_in_hw(unsigned int cpu, struct thread_info *thread)
+{
+#ifdef CONFIG_SMP
+	if (thread->vfpstate.hard.cpu != cpu)
+		return false;
+#endif
+	return vfp_current_hw_state[cpu] == &thread->vfpstate;
+}
+
+/*
+ * Force a reload of the VFP context from the thread structure.  We do
+ * this by ensuring that access to the VFP hardware is disabled, and
+ * clear vfp_current_hw_state.  Must be called from non-preemptible context.
+ */
+static void vfp_force_reload(unsigned int cpu, struct thread_info *thread)
+{
+	if (vfp_state_in_hw(cpu, thread)) {
+		fmxr(FPEXC, fmrx(FPEXC) & ~FPEXC_EN);
+		vfp_current_hw_state[cpu] = NULL;
+	}
+#ifdef CONFIG_SMP
+	thread->vfpstate.hard.cpu = NR_CPUS;
+#endif
+}
+
+/*
+ * Per-thread VFP initialization.
+ */
+static void vfp_thread_flush(struct thread_info *thread)
+{
+	union vfp_state *vfp = &thread->vfpstate;
+	unsigned int cpu;
+
+	/*
+	 * Disable VFP to ensure we initialize it first.  We must ensure
+	 * that the modification of vfp_current_hw_state[] and hardware
+	 * disable are done for the same CPU and without preemption.
+	 *
+	 * Do this first to ensure that preemption won't overwrite our
+	 * state saving should access to the VFP be enabled at this point.
+	 */
+	cpu = get_cpu();
+	if (vfp_current_hw_state[cpu] == vfp)
+		vfp_current_hw_state[cpu] = NULL;
+	fmxr(FPEXC, fmrx(FPEXC) & ~FPEXC_EN);
+	put_cpu();
+
+	memset(vfp, 0, sizeof(union vfp_state));
+
+	vfp->hard.fpexc = FPEXC_EN;
+	vfp->hard.fpscr = FPSCR_ROUND_NEAREST;
+#ifdef CONFIG_SMP
+	vfp->hard.cpu = NR_CPUS;
+#endif
+}
+
+static void vfp_thread_exit(struct thread_info *thread)
+{
+	/* release case: Per-thread VFP cleanup. */
+	union vfp_state *vfp = &thread->vfpstate;
+	unsigned int cpu = get_cpu();
+
+	if (vfp_current_hw_state[cpu] == vfp)
+		vfp_current_hw_state[cpu] = NULL;
+	put_cpu();
+}
+
+static void vfp_thread_copy(struct thread_info *thread)
+{
+	struct thread_info *parent = current_thread_info();
+
+	vfp_sync_hwstate(parent);
+	thread->vfpstate = parent->vfpstate;
+#ifdef CONFIG_SMP
+	thread->vfpstate.hard.cpu = NR_CPUS;
+#endif
+}
+
+/*
+ * When this function is called with the following 'cmd's, the following
+ * is true while this function is being run:
+ *  THREAD_NOFTIFY_SWTICH:
+ *   - the previously running thread will not be scheduled onto another CPU.
+ *   - the next thread to be run (v) will not be running on another CPU.
+ *   - thread->cpu is the local CPU number
+ *   - not preemptible as we're called in the middle of a thread switch
+ *  THREAD_NOTIFY_FLUSH:
+ *   - the thread (v) will be running on the local CPU, so
+ *	v === current_thread_info()
+ *   - thread->cpu is the local CPU number at the time it is accessed,
+ *	but may change at any time.
+ *   - we could be preempted if tree preempt rcu is enabled, so
+ *	it is unsafe to use thread->cpu.
+ *  THREAD_NOTIFY_EXIT
+ *   - we could be preempted if tree preempt rcu is enabled, so
+ *	it is unsafe to use thread->cpu.
+ */
+static int vfp_notifier(struct notifier_block *self, unsigned long cmd, void *v)
+{
+	struct thread_info *thread = v;
+	u32 fpexc;
+#ifdef CONFIG_SMP
+	unsigned int cpu;
+#endif
+
+	switch (cmd) {
+	case THREAD_NOTIFY_SWITCH:
+		fpexc = fmrx(FPEXC);
+
+#ifdef CONFIG_SMP
+		cpu = thread->cpu;
+
+		/*
+		 * On SMP, if VFP is enabled, save the old state in
+		 * case the thread migrates to a different CPU. The
+		 * restoring is done lazily.
+		 */
+		if ((fpexc & FPEXC_EN) && vfp_current_hw_state[cpu])
+			vfp_save_state(vfp_current_hw_state[cpu], fpexc);
+#endif
+
+		/*
+		 * Always disable VFP so we can lazily save/restore the
+		 * old state.
+		 */
+		fmxr(FPEXC, fpexc & ~FPEXC_EN);
+		break;
+
+	case THREAD_NOTIFY_FLUSH:
+		vfp_thread_flush(thread);
+		break;
+
+	case THREAD_NOTIFY_EXIT:
+		vfp_thread_exit(thread);
+		break;
+
+	case THREAD_NOTIFY_COPY:
+		vfp_thread_copy(thread);
+		break;
+	}
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block vfp_notifier_block = {
+	.notifier_call	= vfp_notifier,
+};
+
+/*
+ * Raise a SIGFPE for the current process.
+ * sicode describes the signal being raised.
+ */
+static void vfp_raise_sigfpe(unsigned int sicode, struct pt_regs *regs)
+{
+	siginfo_t info;
+
+	memset(&info, 0, sizeof(info));
+
+	info.si_signo = SIGFPE;
+	info.si_code = sicode;
+	info.si_addr = (void __user *)(instruction_pointer(regs) - 4);
+
+	/*
+	 * This is the same as NWFPE, because it's not clear what
+	 * this is used for
+	 */
+	current->thread.error_code = 0;
+	current->thread.trap_no = 6;
+
+	send_sig_info(SIGFPE, &info, current);
+}
+
+static void vfp_panic(char *reason, u32 inst)
+{
+	int i;
+
+	pr_err("VFP: Error: %s\n", reason);
+	pr_err("VFP: EXC 0x%08x SCR 0x%08x INST 0x%08x\n",
+		fmrx(FPEXC), fmrx(FPSCR), inst);
+	for (i = 0; i < 32; i += 2)
+		pr_err("VFP: s%2u: 0x%08x s%2u: 0x%08x\n",
+		       i, vfp_get_float(i), i+1, vfp_get_float(i+1));
+}
+
+/*
+ * Process bitmask of exception conditions.
+ */
+static void vfp_raise_exceptions(u32 exceptions, u32 inst, u32 fpscr, struct pt_regs *regs)
+{
+	int si_code = 0;
+
+	pr_debug("VFP: raising exceptions %08x\n", exceptions);
+
+	if (exceptions == VFP_EXCEPTION_ERROR) {
+		vfp_panic("unhandled bounce", inst);
+		vfp_raise_sigfpe(0, regs);
+		return;
+	}
+
+	/*
+	 * If any of the status flags are set, update the FPSCR.
+	 * Comparison instructions always return at least one of
+	 * these flags set.
+	 */
+	if (exceptions & (FPSCR_N|FPSCR_Z|FPSCR_C|FPSCR_V))
+		fpscr &= ~(FPSCR_N|FPSCR_Z|FPSCR_C|FPSCR_V);
+
+	fpscr |= exceptions;
+
+	fmxr(FPSCR, fpscr);
+
+#define RAISE(stat,en,sig)				\
+	if (exceptions & stat && fpscr & en)		\
+		si_code = sig;
+
+	/*
+	 * These are arranged in priority order, least to highest.
+	 */
+	RAISE(FPSCR_DZC, FPSCR_DZE, FPE_FLTDIV);
+	RAISE(FPSCR_IXC, FPSCR_IXE, FPE_FLTRES);
+	RAISE(FPSCR_UFC, FPSCR_UFE, FPE_FLTUND);
+	RAISE(FPSCR_OFC, FPSCR_OFE, FPE_FLTOVF);
+	RAISE(FPSCR_IOC, FPSCR_IOE, FPE_FLTINV);
+
+	if (si_code)
+		vfp_raise_sigfpe(si_code, regs);
+}
+
+/*
+ * Emulate a VFP instruction.
+ */
+static u32 vfp_emulate_instruction(u32 inst, u32 fpscr, struct pt_regs *regs)
+{
+	u32 exceptions = VFP_EXCEPTION_ERROR;
+
+	pr_debug("VFP: emulate: INST=0x%08x SCR=0x%08x\n", inst, fpscr);
+
+	if (INST_CPRTDO(inst)) {
+		if (!INST_CPRT(inst)) {
+			/*
+			 * CPDO
+			 */
+			if (vfp_single(inst)) {
+				exceptions = vfp_single_cpdo(inst, fpscr);
+			} else {
+				exceptions = vfp_double_cpdo(inst, fpscr);
+			}
+		} else {
+			/*
+			 * A CPRT instruction can not appear in FPINST2, nor
+			 * can it cause an exception.  Therefore, we do not
+			 * have to emulate it.
+			 */
+		}
+	} else {
+		/*
+		 * A CPDT instruction can not appear in FPINST2, nor can
+		 * it cause an exception.  Therefore, we do not have to
+		 * emulate it.
+		 */
+	}
+	return exceptions & ~VFP_NAN_FLAG;
+}
+
+/*
+ * Package up a bounce condition.
+ */
+void VFP_bounce(u32 trigger, u32 fpexc, struct pt_regs *regs)
+{
+	u32 fpscr, orig_fpscr, fpsid, exceptions;
+
+	pr_debug("VFP: bounce: trigger %08x fpexc %08x\n", trigger, fpexc);
+
+	/*
+	 * At this point, FPEXC can have the following configuration:
+	 *
+	 *  EX DEX IXE
+	 *  0   1   x   - synchronous exception
+	 *  1   x   0   - asynchronous exception
+	 *  1   x   1   - sychronous on VFP subarch 1 and asynchronous on later
+	 *  0   0   1   - synchronous on VFP9 (non-standard subarch 1
+	 *                implementation), undefined otherwise
+	 *
+	 * Clear various bits and enable access to the VFP so we can
+	 * handle the bounce.
+	 */
+	fmxr(FPEXC, fpexc & ~(FPEXC_EX|FPEXC_DEX|FPEXC_FP2V|FPEXC_VV|FPEXC_TRAP_MASK));
+
+	fpsid = fmrx(FPSID);
+	orig_fpscr = fpscr = fmrx(FPSCR);
+
+	/*
+	 * Check for the special VFP subarch 1 and FPSCR.IXE bit case
+	 */
+	if ((fpsid & FPSID_ARCH_MASK) == (1 << FPSID_ARCH_BIT)
+	    && (fpscr & FPSCR_IXE)) {
+		/*
+		 * Synchronous exception, emulate the trigger instruction
+		 */
+		goto emulate;
+	}
+
+	if (fpexc & FPEXC_EX) {
+#ifndef CONFIG_CPU_FEROCEON
+		/*
+		 * Asynchronous exception. The instruction is read from FPINST
+		 * and the interrupted instruction has to be restarted.
+		 */
+		trigger = fmrx(FPINST);
+		regs->ARM_pc -= 4;
+#endif
+	} else if (!(fpexc & FPEXC_DEX)) {
+		/*
+		 * Illegal combination of bits. It can be caused by an
+		 * unallocated VFP instruction but with FPSCR.IXE set and not
+		 * on VFP subarch 1.
+		 */
+		 vfp_raise_exceptions(VFP_EXCEPTION_ERROR, trigger, fpscr, regs);
+		goto exit;
+	}
+
+	/*
+	 * Modify fpscr to indicate the number of iterations remaining.
+	 * If FPEXC.EX is 0, FPEXC.DEX is 1 and the FPEXC.VV bit indicates
+	 * whether FPEXC.VECITR or FPSCR.LEN is used.
+	 */
+	if (fpexc & (FPEXC_EX | FPEXC_VV)) {
+		u32 len;
+
+		len = fpexc + (1 << FPEXC_LENGTH_BIT);
+
+		fpscr &= ~FPSCR_LENGTH_MASK;
+		fpscr |= (len & FPEXC_LENGTH_MASK) << (FPSCR_LENGTH_BIT - FPEXC_LENGTH_BIT);
+	}
+
+	/*
+	 * Handle the first FP instruction.  We used to take note of the
+	 * FPEXC bounce reason, but this appears to be unreliable.
+	 * Emulate the bounced instruction instead.
+	 */
+	exceptions = vfp_emulate_instruction(trigger, fpscr, regs);
+	if (exceptions)
+		vfp_raise_exceptions(exceptions, trigger, orig_fpscr, regs);
+
+	/*
+	 * If there isn't a second FP instruction, exit now. Note that
+	 * the FPEXC.FP2V bit is valid only if FPEXC.EX is 1.
+	 */
+	if ((fpexc & (FPEXC_EX | FPEXC_FP2V)) != (FPEXC_EX | FPEXC_FP2V))
+		goto exit;
+
+	/*
+	 * The barrier() here prevents fpinst2 being read
+	 * before the condition above.
+	 */
+	barrier();
+	trigger = fmrx(FPINST2);
+
+ emulate:
+	exceptions = vfp_emulate_instruction(trigger, orig_fpscr, regs);
+	if (exceptions)
+		vfp_raise_exceptions(exceptions, trigger, orig_fpscr, regs);
+ exit:
+	preempt_enable();
+}
+
+static void vfp_enable(void *unused)
+{
+	u32 access;
+
+	BUG_ON(preemptible());
+	access = get_copro_access();
+
+	/*
+	 * Enable full access to VFP (cp10 and cp11)
+	 */
+	set_copro_access(access | CPACC_FULL(10) | CPACC_FULL(11));
+}
+
+/* Called by platforms on which we want to disable VFP because it may not be
+ * present on all CPUs within a SMP complex. Needs to be called prior to
+ * vfp_init().
+ */
+void vfp_disable(void)
+{
+	if (VFP_arch) {
+		pr_debug("%s: should be called prior to vfp_init\n", __func__);
+		return;
+	}
+	VFP_arch = 1;
+}
+
+#ifdef CONFIG_CPU_PM
+static int vfp_pm_suspend(void)
+{
+	struct thread_info *ti = current_thread_info();
+	u32 fpexc = fmrx(FPEXC);
+
+	/* if vfp is on, then save state for resumption */
+	if (fpexc & FPEXC_EN) {
+		pr_debug("%s: saving vfp state\n", __func__);
+		vfp_save_state(&ti->vfpstate, fpexc);
+
+		/* disable, just in case */
+		fmxr(FPEXC, fmrx(FPEXC) & ~FPEXC_EN);
+	} else if (vfp_current_hw_state[ti->cpu]) {
+#ifndef CONFIG_SMP
+		fmxr(FPEXC, fpexc | FPEXC_EN);
+		vfp_save_state(vfp_current_hw_state[ti->cpu], fpexc);
+		fmxr(FPEXC, fpexc);
+#endif
+	}
+
+	/* clear any information we had about last context state */
+	vfp_current_hw_state[ti->cpu] = NULL;
+
+	return 0;
+}
+
+static void vfp_pm_resume(void)
+{
+	/* ensure we have access to the vfp */
+	vfp_enable(NULL);
+
+	/* and disable it to ensure the next usage restores the state */
+	fmxr(FPEXC, fmrx(FPEXC) & ~FPEXC_EN);
+}
+
+static int vfp_cpu_pm_notifier(struct notifier_block *self, unsigned long cmd,
+	void *v)
+{
+	switch (cmd) {
+	case CPU_PM_ENTER:
+		vfp_pm_suspend();
+		break;
+	case CPU_PM_ENTER_FAILED:
+	case CPU_PM_EXIT:
+		vfp_pm_resume();
+		break;
+	}
+	return NOTIFY_OK;
+}
+
+static struct notifier_block vfp_cpu_pm_notifier_block = {
+	.notifier_call = vfp_cpu_pm_notifier,
+};
+
+static void vfp_pm_init(void)
+{
+	cpu_pm_register_notifier(&vfp_cpu_pm_notifier_block);
+}
+
+#else
+static inline void vfp_pm_init(void) { }
+#endif /* CONFIG_CPU_PM */
+
+/*
+ * Ensure that the VFP state stored in 'thread->vfpstate' is up to date
+ * with the hardware state.
+ */
+void vfp_sync_hwstate(struct thread_info *thread)
+{
+	unsigned int cpu = get_cpu();
+
+	if (vfp_state_in_hw(cpu, thread)) {
+		u32 fpexc = fmrx(FPEXC);
+
+		/*
+		 * Save the last VFP state on this CPU.
+		 */
+		fmxr(FPEXC, fpexc | FPEXC_EN);
+		vfp_save_state(&thread->vfpstate, fpexc | FPEXC_EN);
+		fmxr(FPEXC, fpexc);
+	}
+
+	put_cpu();
+}
+
+/* Ensure that the thread reloads the hardware VFP state on the next use. */
+void vfp_flush_hwstate(struct thread_info *thread)
+{
+	unsigned int cpu = get_cpu();
+
+	vfp_force_reload(cpu, thread);
+
+	put_cpu();
+}
+
+/*
+ * Save the current VFP state into the provided structures and prepare
+ * for entry into a new function (signal handler).
+ */
+int vfp_preserve_user_clear_hwstate(struct user_vfp __user *ufp,
+				    struct user_vfp_exc __user *ufp_exc)
+{
+	struct thread_info *thread = current_thread_info();
+	struct vfp_hard_struct *hwstate = &thread->vfpstate.hard;
+	int err = 0;
+
+	/* Ensure that the saved hwstate is up-to-date. */
+	vfp_sync_hwstate(thread);
+
+	/*
+	 * Copy the floating point registers. There can be unused
+	 * registers see asm/hwcap.h for details.
+	 */
+	err |= __copy_to_user(&ufp->fpregs, &hwstate->fpregs,
+			      sizeof(hwstate->fpregs));
+	/*
+	 * Copy the status and control register.
+	 */
+	__put_user_error(hwstate->fpscr, &ufp->fpscr, err);
+
+	/*
+	 * Copy the exception registers.
+	 */
+	__put_user_error(hwstate->fpexc, &ufp_exc->fpexc, err);
+	__put_user_error(hwstate->fpinst, &ufp_exc->fpinst, err);
+	__put_user_error(hwstate->fpinst2, &ufp_exc->fpinst2, err);
+
+	if (err)
+		return -EFAULT;
+
+	/* Ensure that VFP is disabled. */
+	vfp_flush_hwstate(thread);
+
+	/*
+	 * As per the PCS, clear the length and stride bits for function
+	 * entry.
+	 */
+	hwstate->fpscr &= ~(FPSCR_LENGTH_MASK | FPSCR_STRIDE_MASK);
+	return 0;
+}
+
+/* Sanitise and restore the current VFP state from the provided structures. */
+int vfp_restore_user_hwstate(struct user_vfp __user *ufp,
+			     struct user_vfp_exc __user *ufp_exc)
+{
+	struct thread_info *thread = current_thread_info();
+	struct vfp_hard_struct *hwstate = &thread->vfpstate.hard;
+	unsigned long fpexc;
+	int err = 0;
+
+	/* Disable VFP to avoid corrupting the new thread state. */
+	vfp_flush_hwstate(thread);
+
+	/*
+	 * Copy the floating point registers. There can be unused
+	 * registers see asm/hwcap.h for details.
+	 */
+	err |= __copy_from_user(&hwstate->fpregs, &ufp->fpregs,
+				sizeof(hwstate->fpregs));
+	/*
+	 * Copy the status and control register.
+	 */
+	__get_user_error(hwstate->fpscr, &ufp->fpscr, err);
+
+	/*
+	 * Sanitise and restore the exception registers.
+	 */
+	__get_user_error(fpexc, &ufp_exc->fpexc, err);
+
+	/* Ensure the VFP is enabled. */
+	fpexc |= FPEXC_EN;
+
+	/* Ensure FPINST2 is invalid and the exception flag is cleared. */
+	fpexc &= ~(FPEXC_EX | FPEXC_FP2V);
+	hwstate->fpexc = fpexc;
+
+	__get_user_error(hwstate->fpinst, &ufp_exc->fpinst, err);
+	__get_user_error(hwstate->fpinst2, &ufp_exc->fpinst2, err);
+
+	return err ? -EFAULT : 0;
+}
+
+/*
+ * VFP hardware can lose all context when a CPU goes offline.
+ * As we will be running in SMP mode with CPU hotplug, we will save the
+ * hardware state at every thread switch.  We clear our held state when
+ * a CPU has been killed, indicating that the VFP hardware doesn't contain
+ * a threads VFP state.  When a CPU starts up, we re-enable access to the
+ * VFP hardware. The callbacks below are called on the CPU which
+ * is being offlined/onlined.
+ */
+static int vfp_dying_cpu(unsigned int cpu)
+{
+	vfp_current_hw_state[cpu] = NULL;
+	return 0;
+}
+
+static int vfp_starting_cpu(unsigned int unused)
+{
+	vfp_enable(NULL);
+	return 0;
+}
+
+void vfp_kmode_exception(void)
+{
+	/*
+	 * If we reach this point, a floating point exception has been raised
+	 * while running in kernel mode. If the NEON/VFP unit was enabled at the
+	 * time, it means a VFP instruction has been issued that requires
+	 * software assistance to complete, something which is not currently
+	 * supported in kernel mode.
+	 * If the NEON/VFP unit was disabled, and the location pointed to below
+	 * is properly preceded by a call to kernel_neon_begin(), something has
+	 * caused the task to be scheduled out and back in again. In this case,
+	 * rebuilding and running with CONFIG_DEBUG_ATOMIC_SLEEP enabled should
+	 * be helpful in localizing the problem.
+	 */
+	if (fmrx(FPEXC) & FPEXC_EN)
+		pr_crit("BUG: unsupported FP instruction in kernel mode\n");
+	else
+		pr_crit("BUG: FP instruction issued in kernel mode with FP unit disabled\n");
+}
+
+#ifdef CONFIG_KERNEL_MODE_NEON
+
+/*
+ * Kernel-side NEON support functions
+ */
+void kernel_neon_begin(void)
+{
+	struct thread_info *thread = current_thread_info();
+	unsigned int cpu;
+	u32 fpexc;
+
+	/*
+	 * Kernel mode NEON is only allowed outside of interrupt context
+	 * with preemption disabled. This will make sure that the kernel
+	 * mode NEON register contents never need to be preserved.
+	 */
+	BUG_ON(in_interrupt());
+	cpu = get_cpu();
+
+	fpexc = fmrx(FPEXC) | FPEXC_EN;
+	fmxr(FPEXC, fpexc);
+
+	/*
+	 * Save the userland NEON/VFP state. Under UP,
+	 * the owner could be a task other than 'current'
+	 */
+	if (vfp_state_in_hw(cpu, thread))
+		vfp_save_state(&thread->vfpstate, fpexc);
+#ifndef CONFIG_SMP
+	else if (vfp_current_hw_state[cpu] != NULL)
+		vfp_save_state(vfp_current_hw_state[cpu], fpexc);
+#endif
+	vfp_current_hw_state[cpu] = NULL;
+}
+EXPORT_SYMBOL(kernel_neon_begin);
+
+void kernel_neon_end(void)
+{
+	/* Disable the NEON/VFP unit. */
+	fmxr(FPEXC, fmrx(FPEXC) & ~FPEXC_EN);
+	put_cpu();
+}
+EXPORT_SYMBOL(kernel_neon_end);
+
+#endif /* CONFIG_KERNEL_MODE_NEON */
+
+/*
+ * VFP support code initialisation.
+ */
+static int __init vfp_init(void)
+{
+	unsigned int vfpsid;
+	unsigned int cpu_arch = cpu_architecture();
+
+	/*
+	 * Enable the access to the VFP on all online CPUs so the
+	 * following test on FPSID will succeed.
+	 */
+	if (cpu_arch >= CPU_ARCH_ARMv6)
+		on_each_cpu(vfp_enable, NULL, 1);
+
+	/*
+	 * First check that there is a VFP that we can use.
+	 * The handler is already setup to just log calls, so
+	 * we just need to read the VFPSID register.
+	 */
+	vfp_vector = vfp_testing_entry;
+	barrier();
+	vfpsid = fmrx(FPSID);
+	barrier();
+	vfp_vector = vfp_null_entry;
+
+	pr_info("VFP support v0.3: ");
+	if (VFP_arch) {
+		pr_cont("not present\n");
+		return 0;
+	/* Extract the architecture on CPUID scheme */
+	} else if ((read_cpuid_id() & 0x000f0000) == 0x000f0000) {
+		VFP_arch = vfpsid & FPSID_CPUID_ARCH_MASK;
+		VFP_arch >>= FPSID_ARCH_BIT;
+		/*
+		 * Check for the presence of the Advanced SIMD
+		 * load/store instructions, integer and single
+		 * precision floating point operations. Only check
+		 * for NEON if the hardware has the MVFR registers.
+		 */
+		if (IS_ENABLED(CONFIG_NEON) &&
+		   (fmrx(MVFR1) & 0x000fff00) == 0x00011100)
+			elf_hwcap |= HWCAP_NEON;
+
+		if (IS_ENABLED(CONFIG_VFPv3)) {
+			u32 mvfr0 = fmrx(MVFR0);
+			if (((mvfr0 & MVFR0_DP_MASK) >> MVFR0_DP_BIT) == 0x2 ||
+			    ((mvfr0 & MVFR0_SP_MASK) >> MVFR0_SP_BIT) == 0x2) {
+				elf_hwcap |= HWCAP_VFPv3;
+				/*
+				 * Check for VFPv3 D16 and VFPv4 D16.  CPUs in
+				 * this configuration only have 16 x 64bit
+				 * registers.
+				 */
+				if ((mvfr0 & MVFR0_A_SIMD_MASK) == 1)
+					/* also v4-D16 */
+					elf_hwcap |= HWCAP_VFPv3D16;
+				else
+					elf_hwcap |= HWCAP_VFPD32;
+			}
+
+			if ((fmrx(MVFR1) & 0xf0000000) == 0x10000000)
+				elf_hwcap |= HWCAP_VFPv4;
+		}
+	/* Extract the architecture version on pre-cpuid scheme */
+	} else {
+		if (vfpsid & FPSID_NODOUBLE) {
+			pr_cont("no double precision support\n");
+			return 0;
+		}
+
+		VFP_arch = (vfpsid & FPSID_ARCH_MASK) >> FPSID_ARCH_BIT;
+	}
+
+	cpuhp_setup_state_nocalls(CPUHP_AP_ARM_VFP_STARTING,
+				  "AP_ARM_VFP_STARTING", vfp_starting_cpu,
+				  vfp_dying_cpu);
+
+	vfp_vector = vfp_support_entry;
+
+	thread_register_notifier(&vfp_notifier_block);
+	vfp_pm_init();
+
+	/*
+	 * We detected VFP, and the support code is
+	 * in place; report VFP support to userspace.
+	 */
+	elf_hwcap |= HWCAP_VFP;
+
+	pr_cont("implementor %02x architecture %d part %02x variant %x rev %x\n",
+		(vfpsid & FPSID_IMPLEMENTER_MASK) >> FPSID_IMPLEMENTER_BIT,
+		VFP_arch,
+		(vfpsid & FPSID_PART_MASK) >> FPSID_PART_BIT,
+		(vfpsid & FPSID_VARIANT_MASK) >> FPSID_VARIANT_BIT,
+		(vfpsid & FPSID_REV_MASK) >> FPSID_REV_BIT);
+
+	return 0;
+}
+
+core_initcall(vfp_init);

arch/arm/vfp/vfpsingle.c

@@ -0,0 +1,1246 @@
+/*
+ *  linux/arch/arm/vfp/vfpsingle.c
+ *
+ * This code is derived in part from John R. Housers softfloat library, which
+ * carries the following notice:
+ *
+ * ===========================================================================
+ * This C source file is part of the SoftFloat IEC/IEEE Floating-point
+ * Arithmetic Package, Release 2.
+ *
+ * Written by John R. Hauser.  This work was made possible in part by the
+ * International Computer Science Institute, located at Suite 600, 1947 Center
+ * Street, Berkeley, California 94704.  Funding was partially provided by the
+ * National Science Foundation under grant MIP-9311980.  The original version
+ * of this code was written as part of a project to build a fixed-point vector
+ * processor in collaboration with the University of California at Berkeley,
+ * overseen by Profs. Nelson Morgan and John Wawrzynek.  More information
+ * is available through the web page `http://HTTP.CS.Berkeley.EDU/~jhauser/
+ * arithmetic/softfloat.html'.
+ *
+ * THIS SOFTWARE IS DISTRIBUTED AS IS, FOR FREE.  Although reasonable effort
+ * has been made to avoid it, THIS SOFTWARE MAY CONTAIN FAULTS THAT WILL AT
+ * TIMES RESULT IN INCORRECT BEHAVIOR.  USE OF THIS SOFTWARE IS RESTRICTED TO
+ * PERSONS AND ORGANIZATIONS WHO CAN AND WILL TAKE FULL RESPONSIBILITY FOR ANY
+ * AND ALL LOSSES, COSTS, OR OTHER PROBLEMS ARISING FROM ITS USE.
+ *
+ * Derivative works are acceptable, even for commercial purposes, so long as
+ * (1) they include prominent notice that the work is derivative, and (2) they
+ * include prominent notice akin to these three paragraphs for those parts of
+ * this code that are retained.
+ * ===========================================================================
+ */
+#include <linux/kernel.h>
+#include <linux/bitops.h>
+
+#include <asm/div64.h>
+#include <asm/vfp.h>
+
+#include "vfpinstr.h"
+#include "vfp.h"
+
+static struct vfp_single vfp_single_default_qnan = {
+	.exponent	= 255,
+	.sign		= 0,
+	.significand	= VFP_SINGLE_SIGNIFICAND_QNAN,
+};
+
+static void vfp_single_dump(const char *str, struct vfp_single *s)
+{
+	pr_debug("VFP: %s: sign=%d exponent=%d significand=%08x\n",
+		 str, s->sign != 0, s->exponent, s->significand);
+}
+
+static void vfp_single_normalise_denormal(struct vfp_single *vs)
+{
+	int bits = 31 - fls(vs->significand);
+
+	vfp_single_dump("normalise_denormal: in", vs);
+
+	if (bits) {
+		vs->exponent -= bits - 1;
+		vs->significand <<= bits;
+	}
+
+	vfp_single_dump("normalise_denormal: out", vs);
+}
+
+#ifndef DEBUG
+#define vfp_single_normaliseround(sd,vsd,fpscr,except,func) __vfp_single_normaliseround(sd,vsd,fpscr,except)
+u32 __vfp_single_normaliseround(int sd, struct vfp_single *vs, u32 fpscr, u32 exceptions)
+#else
+u32 vfp_single_normaliseround(int sd, struct vfp_single *vs, u32 fpscr, u32 exceptions, const char *func)
+#endif
+{
+	u32 significand, incr, rmode;
+	int exponent, shift, underflow;
+
+	vfp_single_dump("pack: in", vs);
+
+	/*
+	 * Infinities and NaNs are a special case.
+	 */
+	if (vs->exponent == 255 && (vs->significand == 0 || exceptions))
+		goto pack;
+
+	/*
+	 * Special-case zero.
+	 */
+	if (vs->significand == 0) {
+		vs->exponent = 0;
+		goto pack;
+	}
+
+	exponent = vs->exponent;
+	significand = vs->significand;
+
+	/*
+	 * Normalise first.  Note that we shift the significand up to
+	 * bit 31, so we have VFP_SINGLE_LOW_BITS + 1 below the least
+	 * significant bit.
+	 */
+	shift = 32 - fls(significand);
+	if (shift < 32 && shift) {
+		exponent -= shift;
+		significand <<= shift;
+	}
+
+#ifdef DEBUG
+	vs->exponent = exponent;
+	vs->significand = significand;
+	vfp_single_dump("pack: normalised", vs);
+#endif
+
+	/*
+	 * Tiny number?
+	 */
+	underflow = exponent < 0;
+	if (underflow) {
+		significand = vfp_shiftright32jamming(significand, -exponent);
+		exponent = 0;
+#ifdef DEBUG
+		vs->exponent = exponent;
+		vs->significand = significand;
+		vfp_single_dump("pack: tiny number", vs);
+#endif
+		if (!(significand & ((1 << (VFP_SINGLE_LOW_BITS + 1)) - 1)))
+			underflow = 0;
+	}
+
+	/*
+	 * Select rounding increment.
+	 */
+	incr = 0;
+	rmode = fpscr & FPSCR_RMODE_MASK;
+
+	if (rmode == FPSCR_ROUND_NEAREST) {
+		incr = 1 << VFP_SINGLE_LOW_BITS;
+		if ((significand & (1 << (VFP_SINGLE_LOW_BITS + 1))) == 0)
+			incr -= 1;
+	} else if (rmode == FPSCR_ROUND_TOZERO) {
+		incr = 0;
+	} else if ((rmode == FPSCR_ROUND_PLUSINF) ^ (vs->sign != 0))
+		incr = (1 << (VFP_SINGLE_LOW_BITS + 1)) - 1;
+
+	pr_debug("VFP: rounding increment = 0x%08x\n", incr);
+
+	/*
+	 * Is our rounding going to overflow?
+	 */
+	if ((significand + incr) < significand) {
+		exponent += 1;
+		significand = (significand >> 1) | (significand & 1);
+		incr >>= 1;
+#ifdef DEBUG
+		vs->exponent = exponent;
+		vs->significand = significand;
+		vfp_single_dump("pack: overflow", vs);
+#endif
+	}
+
+	/*
+	 * If any of the low bits (which will be shifted out of the
+	 * number) are non-zero, the result is inexact.
+	 */
+	if (significand & ((1 << (VFP_SINGLE_LOW_BITS + 1)) - 1))
+		exceptions |= FPSCR_IXC;
+
+	/*
+	 * Do our rounding.
+	 */
+	significand += incr;
+
+	/*
+	 * Infinity?
+	 */
+	if (exponent >= 254) {
+		exceptions |= FPSCR_OFC | FPSCR_IXC;
+		if (incr == 0) {
+			vs->exponent = 253;
+			vs->significand = 0x7fffffff;
+		} else {
+			vs->exponent = 255;		/* infinity */
+			vs->significand = 0;
+		}
+	} else {
+		if (significand >> (VFP_SINGLE_LOW_BITS + 1) == 0)
+			exponent = 0;
+		if (exponent || significand > 0x80000000)
+			underflow = 0;
+		if (underflow)
+			exceptions |= FPSCR_UFC;
+		vs->exponent = exponent;
+		vs->significand = significand >> 1;
+	}
+
+ pack:
+	vfp_single_dump("pack: final", vs);
+	{
+		s32 d = vfp_single_pack(vs);
+#ifdef DEBUG
+		pr_debug("VFP: %s: d(s%d)=%08x exceptions=%08x\n", func,
+			 sd, d, exceptions);
+#endif
+		vfp_put_float(d, sd);
+	}
+
+	return exceptions;
+}
+
+/*
+ * Propagate the NaN, setting exceptions if it is signalling.
+ * 'n' is always a NaN.  'm' may be a number, NaN or infinity.
+ */
+static u32
+vfp_propagate_nan(struct vfp_single *vsd, struct vfp_single *vsn,
+		  struct vfp_single *vsm, u32 fpscr)
+{
+	struct vfp_single *nan;
+	int tn, tm = 0;
+
+	tn = vfp_single_type(vsn);
+
+	if (vsm)
+		tm = vfp_single_type(vsm);
+
+	if (fpscr & FPSCR_DEFAULT_NAN)
+		/*
+		 * Default NaN mode - always returns a quiet NaN
+		 */
+		nan = &vfp_single_default_qnan;
+	else {
+		/*
+		 * Contemporary mode - select the first signalling
+		 * NAN, or if neither are signalling, the first
+		 * quiet NAN.
+		 */
+		if (tn == VFP_SNAN || (tm != VFP_SNAN && tn == VFP_QNAN))
+			nan = vsn;
+		else
+			nan = vsm;
+		/*
+		 * Make the NaN quiet.
+		 */
+		nan->significand |= VFP_SINGLE_SIGNIFICAND_QNAN;
+	}
+
+	*vsd = *nan;
+
+	/*
+	 * If one was a signalling NAN, raise invalid operation.
+	 */
+	return tn == VFP_SNAN || tm == VFP_SNAN ? FPSCR_IOC : VFP_NAN_FLAG;
+}
+
+
+/*
+ * Extended operations
+ */
+static u32 vfp_single_fabs(int sd, int unused, s32 m, u32 fpscr)
+{
+	vfp_put_float(vfp_single_packed_abs(m), sd);
+	return 0;
+}
+
+static u32 vfp_single_fcpy(int sd, int unused, s32 m, u32 fpscr)
+{
+	vfp_put_float(m, sd);
+	return 0;
+}
+
+static u32 vfp_single_fneg(int sd, int unused, s32 m, u32 fpscr)
+{
+	vfp_put_float(vfp_single_packed_negate(m), sd);
+	return 0;
+}
+
+static const u16 sqrt_oddadjust[] = {
+	0x0004, 0x0022, 0x005d, 0x00b1, 0x011d, 0x019f, 0x0236, 0x02e0,
+	0x039c, 0x0468, 0x0545, 0x0631, 0x072b, 0x0832, 0x0946, 0x0a67
+};
+
+static const u16 sqrt_evenadjust[] = {
+	0x0a2d, 0x08af, 0x075a, 0x0629, 0x051a, 0x0429, 0x0356, 0x029e,
+	0x0200, 0x0179, 0x0109, 0x00af, 0x0068, 0x0034, 0x0012, 0x0002
+};
+
+u32 vfp_estimate_sqrt_significand(u32 exponent, u32 significand)
+{
+	int index;
+	u32 z, a;
+
+	if ((significand & 0xc0000000) != 0x40000000) {
+		pr_warn("VFP: estimate_sqrt: invalid significand\n");
+	}
+
+	a = significand << 1;
+	index = (a >> 27) & 15;
+	if (exponent & 1) {
+		z = 0x4000 + (a >> 17) - sqrt_oddadjust[index];
+		z = ((a / z) << 14) + (z << 15);
+		a >>= 1;
+	} else {
+		z = 0x8000 + (a >> 17) - sqrt_evenadjust[index];
+		z = a / z + z;
+		z = (z >= 0x20000) ? 0xffff8000 : (z << 15);
+		if (z <= a)
+			return (s32)a >> 1;
+	}
+	{
+		u64 v = (u64)a << 31;
+		do_div(v, z);
+		return v + (z >> 1);
+	}
+}
+
+static u32 vfp_single_fsqrt(int sd, int unused, s32 m, u32 fpscr)
+{
+	struct vfp_single vsm, vsd;
+	int ret, tm;
+
+	vfp_single_unpack(&vsm, m);
+	tm = vfp_single_type(&vsm);
+	if (tm & (VFP_NAN|VFP_INFINITY)) {
+		struct vfp_single *vsp = &vsd;
+
+		if (tm & VFP_NAN)
+			ret = vfp_propagate_nan(vsp, &vsm, NULL, fpscr);
+		else if (vsm.sign == 0) {
+ sqrt_copy:
+			vsp = &vsm;
+			ret = 0;
+		} else {
+ sqrt_invalid:
+			vsp = &vfp_single_default_qnan;
+			ret = FPSCR_IOC;
+		}
+		vfp_put_float(vfp_single_pack(vsp), sd);
+		return ret;
+	}
+
+	/*
+	 * sqrt(+/- 0) == +/- 0
+	 */
+	if (tm & VFP_ZERO)
+		goto sqrt_copy;
+
+	/*
+	 * Normalise a denormalised number
+	 */
+	if (tm & VFP_DENORMAL)
+		vfp_single_normalise_denormal(&vsm);
+
+	/*
+	 * sqrt(<0) = invalid
+	 */
+	if (vsm.sign)
+		goto sqrt_invalid;
+
+	vfp_single_dump("sqrt", &vsm);
+
+	/*
+	 * Estimate the square root.
+	 */
+	vsd.sign = 0;
+	vsd.exponent = ((vsm.exponent - 127) >> 1) + 127;
+	vsd.significand = vfp_estimate_sqrt_significand(vsm.exponent, vsm.significand) + 2;
+
+	vfp_single_dump("sqrt estimate", &vsd);
+
+	/*
+	 * And now adjust.
+	 */
+	if ((vsd.significand & VFP_SINGLE_LOW_BITS_MASK) <= 5) {
+		if (vsd.significand < 2) {
+			vsd.significand = 0xffffffff;
+		} else {
+			u64 term;
+			s64 rem;
+			vsm.significand <<= !(vsm.exponent & 1);
+			term = (u64)vsd.significand * vsd.significand;
+			rem = ((u64)vsm.significand << 32) - term;
+
+			pr_debug("VFP: term=%016llx rem=%016llx\n", term, rem);
+
+			while (rem < 0) {
+				vsd.significand -= 1;
+				rem += ((u64)vsd.significand << 1) | 1;
+			}
+			vsd.significand |= rem != 0;
+		}
+	}
+	vsd.significand = vfp_shiftright32jamming(vsd.significand, 1);
+
+	return vfp_single_normaliseround(sd, &vsd, fpscr, 0, "fsqrt");
+}
+
+/*
+ * Equal	:= ZC
+ * Less than	:= N
+ * Greater than	:= C
+ * Unordered	:= CV
+ */
+static u32 vfp_compare(int sd, int signal_on_qnan, s32 m, u32 fpscr)
+{
+	s32 d;
+	u32 ret = 0;
+
+	d = vfp_get_float(sd);
+	if (vfp_single_packed_exponent(m) == 255 && vfp_single_packed_mantissa(m)) {
+		ret |= FPSCR_C | FPSCR_V;
+		if (signal_on_qnan || !(vfp_single_packed_mantissa(m) & (1 << (VFP_SINGLE_MANTISSA_BITS - 1))))
+			/*
+			 * Signalling NaN, or signalling on quiet NaN
+			 */
+			ret |= FPSCR_IOC;
+	}
+
+	if (vfp_single_packed_exponent(d) == 255 && vfp_single_packed_mantissa(d)) {
+		ret |= FPSCR_C | FPSCR_V;
+		if (signal_on_qnan || !(vfp_single_packed_mantissa(d) & (1 << (VFP_SINGLE_MANTISSA_BITS - 1))))
+			/*
+			 * Signalling NaN, or signalling on quiet NaN
+			 */
+			ret |= FPSCR_IOC;
+	}
+
+	if (ret == 0) {
+		if (d == m || vfp_single_packed_abs(d | m) == 0) {
+			/*
+			 * equal
+			 */
+			ret |= FPSCR_Z | FPSCR_C;
+		} else if (vfp_single_packed_sign(d ^ m)) {
+			/*
+			 * different signs
+			 */
+			if (vfp_single_packed_sign(d))
+				/*
+				 * d is negative, so d < m
+				 */
+				ret |= FPSCR_N;
+			else
+				/*
+				 * d is positive, so d > m
+				 */
+				ret |= FPSCR_C;
+		} else if ((vfp_single_packed_sign(d) != 0) ^ (d < m)) {
+			/*
+			 * d < m
+			 */
+			ret |= FPSCR_N;
+		} else if ((vfp_single_packed_sign(d) != 0) ^ (d > m)) {
+			/*
+			 * d > m
+			 */
+			ret |= FPSCR_C;
+		}
+	}
+	return ret;
+}
+
+static u32 vfp_single_fcmp(int sd, int unused, s32 m, u32 fpscr)
+{
+	return vfp_compare(sd, 0, m, fpscr);
+}
+
+static u32 vfp_single_fcmpe(int sd, int unused, s32 m, u32 fpscr)
+{
+	return vfp_compare(sd, 1, m, fpscr);
+}
+
+static u32 vfp_single_fcmpz(int sd, int unused, s32 m, u32 fpscr)
+{
+	return vfp_compare(sd, 0, 0, fpscr);
+}
+
+static u32 vfp_single_fcmpez(int sd, int unused, s32 m, u32 fpscr)
+{
+	return vfp_compare(sd, 1, 0, fpscr);
+}
+
+static u32 vfp_single_fcvtd(int dd, int unused, s32 m, u32 fpscr)
+{
+	struct vfp_single vsm;
+	struct vfp_double vdd;
+	int tm;
+	u32 exceptions = 0;
+
+	vfp_single_unpack(&vsm, m);
+
+	tm = vfp_single_type(&vsm);
+
+	/*
+	 * If we have a signalling NaN, signal invalid operation.
+	 */
+	if (tm == VFP_SNAN)
+		exceptions = FPSCR_IOC;
+
+	if (tm & VFP_DENORMAL)
+		vfp_single_normalise_denormal(&vsm);
+
+	vdd.sign = vsm.sign;
+	vdd.significand = (u64)vsm.significand << 32;
+
+	/*
+	 * If we have an infinity or NaN, the exponent must be 2047.
+	 */
+	if (tm & (VFP_INFINITY|VFP_NAN)) {
+		vdd.exponent = 2047;
+		if (tm == VFP_QNAN)
+			vdd.significand |= VFP_DOUBLE_SIGNIFICAND_QNAN;
+		goto pack_nan;
+	} else if (tm & VFP_ZERO)
+		vdd.exponent = 0;
+	else
+		vdd.exponent = vsm.exponent + (1023 - 127);
+
+	return vfp_double_normaliseround(dd, &vdd, fpscr, exceptions, "fcvtd");
+
+ pack_nan:
+	vfp_put_double(vfp_double_pack(&vdd), dd);
+	return exceptions;
+}
+
+static u32 vfp_single_fuito(int sd, int unused, s32 m, u32 fpscr)
+{
+	struct vfp_single vs;
+
+	vs.sign = 0;
+	vs.exponent = 127 + 31 - 1;
+	vs.significand = (u32)m;
+
+	return vfp_single_normaliseround(sd, &vs, fpscr, 0, "fuito");
+}
+
+static u32 vfp_single_fsito(int sd, int unused, s32 m, u32 fpscr)
+{
+	struct vfp_single vs;
+
+	vs.sign = (m & 0x80000000) >> 16;
+	vs.exponent = 127 + 31 - 1;
+	vs.significand = vs.sign ? -m : m;
+
+	return vfp_single_normaliseround(sd, &vs, fpscr, 0, "fsito");
+}
+
+static u32 vfp_single_ftoui(int sd, int unused, s32 m, u32 fpscr)
+{
+	struct vfp_single vsm;
+	u32 d, exceptions = 0;
+	int rmode = fpscr & FPSCR_RMODE_MASK;
+	int tm;
+
+	vfp_single_unpack(&vsm, m);
+	vfp_single_dump("VSM", &vsm);
+
+	/*
+	 * Do we have a denormalised number?
+	 */
+	tm = vfp_single_type(&vsm);
+	if (tm & VFP_DENORMAL)
+		exceptions |= FPSCR_IDC;
+
+	if (tm & VFP_NAN)
+		vsm.sign = 0;
+
+	if (vsm.exponent >= 127 + 32) {
+		d = vsm.sign ? 0 : 0xffffffff;
+		exceptions = FPSCR_IOC;
+	} else if (vsm.exponent >= 127 - 1) {
+		int shift = 127 + 31 - vsm.exponent;
+		u32 rem, incr = 0;
+
+		/*
+		 * 2^0 <= m < 2^32-2^8
+		 */
+		d = (vsm.significand << 1) >> shift;
+		rem = vsm.significand << (33 - shift);
+
+		if (rmode == FPSCR_ROUND_NEAREST) {
+			incr = 0x80000000;
+			if ((d & 1) == 0)
+				incr -= 1;
+		} else if (rmode == FPSCR_ROUND_TOZERO) {
+			incr = 0;
+		} else if ((rmode == FPSCR_ROUND_PLUSINF) ^ (vsm.sign != 0)) {
+			incr = ~0;
+		}
+
+		if ((rem + incr) < rem) {
+			if (d < 0xffffffff)
+				d += 1;
+			else
+				exceptions |= FPSCR_IOC;
+		}
+
+		if (d && vsm.sign) {
+			d = 0;
+			exceptions |= FPSCR_IOC;
+		} else if (rem)
+			exceptions |= FPSCR_IXC;
+	} else {
+		d = 0;
+		if (vsm.exponent | vsm.significand) {
+			exceptions |= FPSCR_IXC;
+			if (rmode == FPSCR_ROUND_PLUSINF && vsm.sign == 0)
+				d = 1;
+			else if (rmode == FPSCR_ROUND_MINUSINF && vsm.sign) {
+				d = 0;
+				exceptions |= FPSCR_IOC;
+			}
+		}
+	}
+
+	pr_debug("VFP: ftoui: d(s%d)=%08x exceptions=%08x\n", sd, d, exceptions);
+
+	vfp_put_float(d, sd);
+
+	return exceptions;
+}
+
+static u32 vfp_single_ftouiz(int sd, int unused, s32 m, u32 fpscr)
+{
+	return vfp_single_ftoui(sd, unused, m, FPSCR_ROUND_TOZERO);
+}
+
+static u32 vfp_single_ftosi(int sd, int unused, s32 m, u32 fpscr)
+{
+	struct vfp_single vsm;
+	u32 d, exceptions = 0;
+	int rmode = fpscr & FPSCR_RMODE_MASK;
+	int tm;
+
+	vfp_single_unpack(&vsm, m);
+	vfp_single_dump("VSM", &vsm);
+
+	/*
+	 * Do we have a denormalised number?
+	 */
+	tm = vfp_single_type(&vsm);
+	if (vfp_single_type(&vsm) & VFP_DENORMAL)
+		exceptions |= FPSCR_IDC;
+
+	if (tm & VFP_NAN) {
+		d = 0;
+		exceptions |= FPSCR_IOC;
+	} else if (vsm.exponent >= 127 + 32) {
+		/*
+		 * m >= 2^31-2^7: invalid
+		 */
+		d = 0x7fffffff;
+		if (vsm.sign)
+			d = ~d;
+		exceptions |= FPSCR_IOC;
+	} else if (vsm.exponent >= 127 - 1) {
+		int shift = 127 + 31 - vsm.exponent;
+		u32 rem, incr = 0;
+
+		/* 2^0 <= m <= 2^31-2^7 */
+		d = (vsm.significand << 1) >> shift;
+		rem = vsm.significand << (33 - shift);
+
+		if (rmode == FPSCR_ROUND_NEAREST) {
+			incr = 0x80000000;
+			if ((d & 1) == 0)
+				incr -= 1;
+		} else if (rmode == FPSCR_ROUND_TOZERO) {
+			incr = 0;
+		} else if ((rmode == FPSCR_ROUND_PLUSINF) ^ (vsm.sign != 0)) {
+			incr = ~0;
+		}
+
+		if ((rem + incr) < rem && d < 0xffffffff)
+			d += 1;
+		if (d > 0x7fffffff + (vsm.sign != 0)) {
+			d = 0x7fffffff + (vsm.sign != 0);
+			exceptions |= FPSCR_IOC;
+		} else if (rem)
+			exceptions |= FPSCR_IXC;
+
+		if (vsm.sign)
+			d = -d;
+	} else {
+		d = 0;
+		if (vsm.exponent | vsm.significand) {
+			exceptions |= FPSCR_IXC;
+			if (rmode == FPSCR_ROUND_PLUSINF && vsm.sign == 0)
+				d = 1;
+			else if (rmode == FPSCR_ROUND_MINUSINF && vsm.sign)
+				d = -1;
+		}
+	}
+
+	pr_debug("VFP: ftosi: d(s%d)=%08x exceptions=%08x\n", sd, d, exceptions);
+
+	vfp_put_float((s32)d, sd);
+
+	return exceptions;
+}
+
+static u32 vfp_single_ftosiz(int sd, int unused, s32 m, u32 fpscr)
+{
+	return vfp_single_ftosi(sd, unused, m, FPSCR_ROUND_TOZERO);
+}
+
+static struct op fops_ext[32] = {
+	[FEXT_TO_IDX(FEXT_FCPY)]	= { vfp_single_fcpy,   0 },
+	[FEXT_TO_IDX(FEXT_FABS)]	= { vfp_single_fabs,   0 },
+	[FEXT_TO_IDX(FEXT_FNEG)]	= { vfp_single_fneg,   0 },
+	[FEXT_TO_IDX(FEXT_FSQRT)]	= { vfp_single_fsqrt,  0 },
+	[FEXT_TO_IDX(FEXT_FCMP)]	= { vfp_single_fcmp,   OP_SCALAR },
+	[FEXT_TO_IDX(FEXT_FCMPE)]	= { vfp_single_fcmpe,  OP_SCALAR },
+	[FEXT_TO_IDX(FEXT_FCMPZ)]	= { vfp_single_fcmpz,  OP_SCALAR },
+	[FEXT_TO_IDX(FEXT_FCMPEZ)]	= { vfp_single_fcmpez, OP_SCALAR },
+	[FEXT_TO_IDX(FEXT_FCVT)]	= { vfp_single_fcvtd,  OP_SCALAR|OP_DD },
+	[FEXT_TO_IDX(FEXT_FUITO)]	= { vfp_single_fuito,  OP_SCALAR },
+	[FEXT_TO_IDX(FEXT_FSITO)]	= { vfp_single_fsito,  OP_SCALAR },
+	[FEXT_TO_IDX(FEXT_FTOUI)]	= { vfp_single_ftoui,  OP_SCALAR },
+	[FEXT_TO_IDX(FEXT_FTOUIZ)]	= { vfp_single_ftouiz, OP_SCALAR },
+	[FEXT_TO_IDX(FEXT_FTOSI)]	= { vfp_single_ftosi,  OP_SCALAR },
+	[FEXT_TO_IDX(FEXT_FTOSIZ)]	= { vfp_single_ftosiz, OP_SCALAR },
+};
+
+
+
+
+
+static u32
+vfp_single_fadd_nonnumber(struct vfp_single *vsd, struct vfp_single *vsn,
+			  struct vfp_single *vsm, u32 fpscr)
+{
+	struct vfp_single *vsp;
+	u32 exceptions = 0;
+	int tn, tm;
+
+	tn = vfp_single_type(vsn);
+	tm = vfp_single_type(vsm);
+
+	if (tn & tm & VFP_INFINITY) {
+		/*
+		 * Two infinities.  Are they different signs?
+		 */
+		if (vsn->sign ^ vsm->sign) {
+			/*
+			 * different signs -> invalid
+			 */
+			exceptions = FPSCR_IOC;
+			vsp = &vfp_single_default_qnan;
+		} else {
+			/*
+			 * same signs -> valid
+			 */
+			vsp = vsn;
+		}
+	} else if (tn & VFP_INFINITY && tm & VFP_NUMBER) {
+		/*
+		 * One infinity and one number -> infinity
+		 */
+		vsp = vsn;
+	} else {
+		/*
+		 * 'n' is a NaN of some type
+		 */
+		return vfp_propagate_nan(vsd, vsn, vsm, fpscr);
+	}
+	*vsd = *vsp;
+	return exceptions;
+}
+
+static u32
+vfp_single_add(struct vfp_single *vsd, struct vfp_single *vsn,
+	       struct vfp_single *vsm, u32 fpscr)
+{
+	u32 exp_diff, m_sig;
+
+	if (vsn->significand & 0x80000000 ||
+	    vsm->significand & 0x80000000) {
+		pr_info("VFP: bad FP values in %s\n", __func__);
+		vfp_single_dump("VSN", vsn);
+		vfp_single_dump("VSM", vsm);
+	}
+
+	/*
+	 * Ensure that 'n' is the largest magnitude number.  Note that
+	 * if 'n' and 'm' have equal exponents, we do not swap them.
+	 * This ensures that NaN propagation works correctly.
+	 */
+	if (vsn->exponent < vsm->exponent) {
+		struct vfp_single *t = vsn;
+		vsn = vsm;
+		vsm = t;
+	}
+
+	/*
+	 * Is 'n' an infinity or a NaN?  Note that 'm' may be a number,
+	 * infinity or a NaN here.
+	 */
+	if (vsn->exponent == 255)
+		return vfp_single_fadd_nonnumber(vsd, vsn, vsm, fpscr);
+
+	/*
+	 * We have two proper numbers, where 'vsn' is the larger magnitude.
+	 *
+	 * Copy 'n' to 'd' before doing the arithmetic.
+	 */
+	*vsd = *vsn;
+
+	/*
+	 * Align both numbers.
+	 */
+	exp_diff = vsn->exponent - vsm->exponent;
+	m_sig = vfp_shiftright32jamming(vsm->significand, exp_diff);
+
+	/*
+	 * If the signs are different, we are really subtracting.
+	 */
+	if (vsn->sign ^ vsm->sign) {
+		m_sig = vsn->significand - m_sig;
+		if ((s32)m_sig < 0) {
+			vsd->sign = vfp_sign_negate(vsd->sign);
+			m_sig = -m_sig;
+		} else if (m_sig == 0) {
+			vsd->sign = (fpscr & FPSCR_RMODE_MASK) ==
+				      FPSCR_ROUND_MINUSINF ? 0x8000 : 0;
+		}
+	} else {
+		m_sig = vsn->significand + m_sig;
+	}
+	vsd->significand = m_sig;
+
+	return 0;
+}
+
+static u32
+vfp_single_multiply(struct vfp_single *vsd, struct vfp_single *vsn, struct vfp_single *vsm, u32 fpscr)
+{
+	vfp_single_dump("VSN", vsn);
+	vfp_single_dump("VSM", vsm);
+
+	/*
+	 * Ensure that 'n' is the largest magnitude number.  Note that
+	 * if 'n' and 'm' have equal exponents, we do not swap them.
+	 * This ensures that NaN propagation works correctly.
+	 */
+	if (vsn->exponent < vsm->exponent) {
+		struct vfp_single *t = vsn;
+		vsn = vsm;
+		vsm = t;
+		pr_debug("VFP: swapping M <-> N\n");
+	}
+
+	vsd->sign = vsn->sign ^ vsm->sign;
+
+	/*
+	 * If 'n' is an infinity or NaN, handle it.  'm' may be anything.
+	 */
+	if (vsn->exponent == 255) {
+		if (vsn->significand || (vsm->exponent == 255 && vsm->significand))
+			return vfp_propagate_nan(vsd, vsn, vsm, fpscr);
+		if ((vsm->exponent | vsm->significand) == 0) {
+			*vsd = vfp_single_default_qnan;
+			return FPSCR_IOC;
+		}
+		vsd->exponent = vsn->exponent;
+		vsd->significand = 0;
+		return 0;
+	}
+
+	/*
+	 * If 'm' is zero, the result is always zero.  In this case,
+	 * 'n' may be zero or a number, but it doesn't matter which.
+	 */
+	if ((vsm->exponent | vsm->significand) == 0) {
+		vsd->exponent = 0;
+		vsd->significand = 0;
+		return 0;
+	}
+
+	/*
+	 * We add 2 to the destination exponent for the same reason as
+	 * the addition case - though this time we have +1 from each
+	 * input operand.
+	 */
+	vsd->exponent = vsn->exponent + vsm->exponent - 127 + 2;
+	vsd->significand = vfp_hi64to32jamming((u64)vsn->significand * vsm->significand);
+
+	vfp_single_dump("VSD", vsd);
+	return 0;
+}
+
+#define NEG_MULTIPLY	(1 << 0)
+#define NEG_SUBTRACT	(1 << 1)
+
+static u32
+vfp_single_multiply_accumulate(int sd, int sn, s32 m, u32 fpscr, u32 negate, char *func)
+{
+	struct vfp_single vsd, vsp, vsn, vsm;
+	u32 exceptions;
+	s32 v;
+
+	v = vfp_get_float(sn);
+	pr_debug("VFP: s%u = %08x\n", sn, v);
+	vfp_single_unpack(&vsn, v);
+	if (vsn.exponent == 0 && vsn.significand)
+		vfp_single_normalise_denormal(&vsn);
+
+	vfp_single_unpack(&vsm, m);
+	if (vsm.exponent == 0 && vsm.significand)
+		vfp_single_normalise_denormal(&vsm);
+
+	exceptions = vfp_single_multiply(&vsp, &vsn, &vsm, fpscr);
+	if (negate & NEG_MULTIPLY)
+		vsp.sign = vfp_sign_negate(vsp.sign);
+
+	v = vfp_get_float(sd);
+	pr_debug("VFP: s%u = %08x\n", sd, v);
+	vfp_single_unpack(&vsn, v);
+	if (vsn.exponent == 0 && vsn.significand)
+		vfp_single_normalise_denormal(&vsn);
+	if (negate & NEG_SUBTRACT)
+		vsn.sign = vfp_sign_negate(vsn.sign);
+
+	exceptions |= vfp_single_add(&vsd, &vsn, &vsp, fpscr);
+
+	return vfp_single_normaliseround(sd, &vsd, fpscr, exceptions, func);
+}
+
+/*
+ * Standard operations
+ */
+
+/*
+ * sd = sd + (sn * sm)
+ */
+static u32 vfp_single_fmac(int sd, int sn, s32 m, u32 fpscr)
+{
+	return vfp_single_multiply_accumulate(sd, sn, m, fpscr, 0, "fmac");
+}
+
+/*
+ * sd = sd - (sn * sm)
+ */
+static u32 vfp_single_fnmac(int sd, int sn, s32 m, u32 fpscr)
+{
+	return vfp_single_multiply_accumulate(sd, sn, m, fpscr, NEG_MULTIPLY, "fnmac");
+}
+
+/*
+ * sd = -sd + (sn * sm)
+ */
+static u32 vfp_single_fmsc(int sd, int sn, s32 m, u32 fpscr)
+{
+	return vfp_single_multiply_accumulate(sd, sn, m, fpscr, NEG_SUBTRACT, "fmsc");
+}
+
+/*
+ * sd = -sd - (sn * sm)
+ */
+static u32 vfp_single_fnmsc(int sd, int sn, s32 m, u32 fpscr)
+{
+	return vfp_single_multiply_accumulate(sd, sn, m, fpscr, NEG_SUBTRACT | NEG_MULTIPLY, "fnmsc");
+}
+
+/*
+ * sd = sn * sm
+ */
+static u32 vfp_single_fmul(int sd, int sn, s32 m, u32 fpscr)
+{
+	struct vfp_single vsd, vsn, vsm;
+	u32 exceptions;
+	s32 n = vfp_get_float(sn);
+
+	pr_debug("VFP: s%u = %08x\n", sn, n);
+
+	vfp_single_unpack(&vsn, n);
+	if (vsn.exponent == 0 && vsn.significand)
+		vfp_single_normalise_denormal(&vsn);
+
+	vfp_single_unpack(&vsm, m);
+	if (vsm.exponent == 0 && vsm.significand)
+		vfp_single_normalise_denormal(&vsm);
+
+	exceptions = vfp_single_multiply(&vsd, &vsn, &vsm, fpscr);
+	return vfp_single_normaliseround(sd, &vsd, fpscr, exceptions, "fmul");
+}
+
+/*
+ * sd = -(sn * sm)
+ */
+static u32 vfp_single_fnmul(int sd, int sn, s32 m, u32 fpscr)
+{
+	struct vfp_single vsd, vsn, vsm;
+	u32 exceptions;
+	s32 n = vfp_get_float(sn);
+
+	pr_debug("VFP: s%u = %08x\n", sn, n);
+
+	vfp_single_unpack(&vsn, n);
+	if (vsn.exponent == 0 && vsn.significand)
+		vfp_single_normalise_denormal(&vsn);
+
+	vfp_single_unpack(&vsm, m);
+	if (vsm.exponent == 0 && vsm.significand)
+		vfp_single_normalise_denormal(&vsm);
+
+	exceptions = vfp_single_multiply(&vsd, &vsn, &vsm, fpscr);
+	vsd.sign = vfp_sign_negate(vsd.sign);
+	return vfp_single_normaliseround(sd, &vsd, fpscr, exceptions, "fnmul");
+}
+
+/*
+ * sd = sn + sm
+ */
+static u32 vfp_single_fadd(int sd, int sn, s32 m, u32 fpscr)
+{
+	struct vfp_single vsd, vsn, vsm;
+	u32 exceptions;
+	s32 n = vfp_get_float(sn);
+
+	pr_debug("VFP: s%u = %08x\n", sn, n);
+
+	/*
+	 * Unpack and normalise denormals.
+	 */
+	vfp_single_unpack(&vsn, n);
+	if (vsn.exponent == 0 && vsn.significand)
+		vfp_single_normalise_denormal(&vsn);
+
+	vfp_single_unpack(&vsm, m);
+	if (vsm.exponent == 0 && vsm.significand)
+		vfp_single_normalise_denormal(&vsm);
+
+	exceptions = vfp_single_add(&vsd, &vsn, &vsm, fpscr);
+
+	return vfp_single_normaliseround(sd, &vsd, fpscr, exceptions, "fadd");
+}
+
+/*
+ * sd = sn - sm
+ */
+static u32 vfp_single_fsub(int sd, int sn, s32 m, u32 fpscr)
+{
+	/*
+	 * Subtraction is addition with one sign inverted.
+	 */
+	return vfp_single_fadd(sd, sn, vfp_single_packed_negate(m), fpscr);
+}
+
+/*
+ * sd = sn / sm
+ */
+static u32 vfp_single_fdiv(int sd, int sn, s32 m, u32 fpscr)
+{
+	struct vfp_single vsd, vsn, vsm;
+	u32 exceptions = 0;
+	s32 n = vfp_get_float(sn);
+	int tm, tn;
+
+	pr_debug("VFP: s%u = %08x\n", sn, n);
+
+	vfp_single_unpack(&vsn, n);
+	vfp_single_unpack(&vsm, m);
+
+	vsd.sign = vsn.sign ^ vsm.sign;
+
+	tn = vfp_single_type(&vsn);
+	tm = vfp_single_type(&vsm);
+
+	/*
+	 * Is n a NAN?
+	 */
+	if (tn & VFP_NAN)
+		goto vsn_nan;
+
+	/*
+	 * Is m a NAN?
+	 */
+	if (tm & VFP_NAN)
+		goto vsm_nan;
+
+	/*
+	 * If n and m are infinity, the result is invalid
+	 * If n and m are zero, the result is invalid
+	 */
+	if (tm & tn & (VFP_INFINITY|VFP_ZERO))
+		goto invalid;
+
+	/*
+	 * If n is infinity, the result is infinity
+	 */
+	if (tn & VFP_INFINITY)
+		goto infinity;
+
+	/*
+	 * If m is zero, raise div0 exception
+	 */
+	if (tm & VFP_ZERO)
+		goto divzero;
+
+	/*
+	 * If m is infinity, or n is zero, the result is zero
+	 */
+	if (tm & VFP_INFINITY || tn & VFP_ZERO)
+		goto zero;
+
+	if (tn & VFP_DENORMAL)
+		vfp_single_normalise_denormal(&vsn);
+	if (tm & VFP_DENORMAL)
+		vfp_single_normalise_denormal(&vsm);
+
+	/*
+	 * Ok, we have two numbers, we can perform division.
+	 */
+	vsd.exponent = vsn.exponent - vsm.exponent + 127 - 1;
+	vsm.significand <<= 1;
+	if (vsm.significand <= (2 * vsn.significand)) {
+		vsn.significand >>= 1;
+		vsd.exponent++;
+	}
+	{
+		u64 significand = (u64)vsn.significand << 32;
+		do_div(significand, vsm.significand);
+		vsd.significand = significand;
+	}
+	if ((vsd.significand & 0x3f) == 0)
+		vsd.significand |= ((u64)vsm.significand * vsd.significand != (u64)vsn.significand << 32);
+
+	return vfp_single_normaliseround(sd, &vsd, fpscr, 0, "fdiv");
+
+ vsn_nan:
+	exceptions = vfp_propagate_nan(&vsd, &vsn, &vsm, fpscr);
+ pack:
+	vfp_put_float(vfp_single_pack(&vsd), sd);
+	return exceptions;
+
+ vsm_nan:
+	exceptions = vfp_propagate_nan(&vsd, &vsm, &vsn, fpscr);
+	goto pack;
+
+ zero:
+	vsd.exponent = 0;
+	vsd.significand = 0;
+	goto pack;
+
+ divzero:
+	exceptions = FPSCR_DZC;
+ infinity:
+	vsd.exponent = 255;
+	vsd.significand = 0;
+	goto pack;
+
+ invalid:
+	vfp_put_float(vfp_single_pack(&vfp_single_default_qnan), sd);
+	return FPSCR_IOC;
+}
+
+static struct op fops[16] = {
+	[FOP_TO_IDX(FOP_FMAC)]	= { vfp_single_fmac,  0 },
+	[FOP_TO_IDX(FOP_FNMAC)]	= { vfp_single_fnmac, 0 },
+	[FOP_TO_IDX(FOP_FMSC)]	= { vfp_single_fmsc,  0 },
+	[FOP_TO_IDX(FOP_FNMSC)]	= { vfp_single_fnmsc, 0 },
+	[FOP_TO_IDX(FOP_FMUL)]	= { vfp_single_fmul,  0 },
+	[FOP_TO_IDX(FOP_FNMUL)]	= { vfp_single_fnmul, 0 },
+	[FOP_TO_IDX(FOP_FADD)]	= { vfp_single_fadd,  0 },
+	[FOP_TO_IDX(FOP_FSUB)]	= { vfp_single_fsub,  0 },
+	[FOP_TO_IDX(FOP_FDIV)]	= { vfp_single_fdiv,  0 },
+};
+
+#define FREG_BANK(x)	((x) & 0x18)
+#define FREG_IDX(x)	((x) & 7)
+
+u32 vfp_single_cpdo(u32 inst, u32 fpscr)
+{
+	u32 op = inst & FOP_MASK;
+	u32 exceptions = 0;
+	unsigned int dest;
+	unsigned int sn = vfp_get_sn(inst);
+	unsigned int sm = vfp_get_sm(inst);
+	unsigned int vecitr, veclen, vecstride;
+	struct op *fop;
+
+	vecstride = 1 + ((fpscr & FPSCR_STRIDE_MASK) == FPSCR_STRIDE_MASK);
+
+	fop = (op == FOP_EXT) ? &fops_ext[FEXT_TO_IDX(inst)] : &fops[FOP_TO_IDX(op)];
+
+	/*
+	 * fcvtsd takes a dN register number as destination, not sN.
+	 * Technically, if bit 0 of dd is set, this is an invalid
+	 * instruction.  However, we ignore this for efficiency.
+	 * It also only operates on scalars.
+	 */
+	if (fop->flags & OP_DD)
+		dest = vfp_get_dd(inst);
+	else
+		dest = vfp_get_sd(inst);
+
+	/*
+	 * If destination bank is zero, vector length is always '1'.
+	 * ARM DDI0100F C5.1.3, C5.3.2.
+	 */
+	if ((fop->flags & OP_SCALAR) || FREG_BANK(dest) == 0)
+		veclen = 0;
+	else
+		veclen = fpscr & FPSCR_LENGTH_MASK;
+
+	pr_debug("VFP: vecstride=%u veclen=%u\n", vecstride,
+		 (veclen >> FPSCR_LENGTH_BIT) + 1);
+
+	if (!fop->fn)
+		goto invalid;
+
+	for (vecitr = 0; vecitr <= veclen; vecitr += 1 << FPSCR_LENGTH_BIT) {
+		s32 m = vfp_get_float(sm);
+		u32 except;
+		char type;
+
+		type = fop->flags & OP_DD ? 'd' : 's';
+		if (op == FOP_EXT)
+			pr_debug("VFP: itr%d (%c%u) = op[%u] (s%u=%08x)\n",
+				 vecitr >> FPSCR_LENGTH_BIT, type, dest, sn,
+				 sm, m);
+		else
+			pr_debug("VFP: itr%d (%c%u) = (s%u) op[%u] (s%u=%08x)\n",
+				 vecitr >> FPSCR_LENGTH_BIT, type, dest, sn,
+				 FOP_TO_IDX(op), sm, m);
+
+		except = fop->fn(dest, sn, m, fpscr);
+		pr_debug("VFP: itr%d: exceptions=%08x\n",
+			 vecitr >> FPSCR_LENGTH_BIT, except);
+
+		exceptions |= except;
+
+		/*
+		 * CHECK: It appears to be undefined whether we stop when
+		 * we encounter an exception.  We continue.
+		 */
+		dest = FREG_BANK(dest) + ((FREG_IDX(dest) + vecstride) & 7);
+		sn = FREG_BANK(sn) + ((FREG_IDX(sn) + vecstride) & 7);
+		if (FREG_BANK(sm) != 0)
+			sm = FREG_BANK(sm) + ((FREG_IDX(sm) + vecstride) & 7);
+	}
+	return exceptions;
+
+ invalid:
+	return (u32)-1;
+}

arch/arm/xen/Makefile

@@ -0,0 +1,2 @@
+obj-y		:= enlighten.o hypercall.o grant-table.o p2m.o mm.o
+obj-$(CONFIG_XEN_EFI) += efi.o

arch/arm/xen/efi.c

@@ -0,0 +1,40 @@
+/*
+ * Copyright (c) 2015, Linaro Limited, Shannon Zhao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/efi.h>
+#include <xen/xen-ops.h>
+#include <asm/xen/xen-ops.h>
+
+/* Set XEN EFI runtime services function pointers. Other fields of struct efi,
+ * e.g. efi.systab, will be set like normal EFI.
+ */
+void __init xen_efi_runtime_setup(void)
+{
+	efi.get_time                 = xen_efi_get_time;
+	efi.set_time                 = xen_efi_set_time;
+	efi.get_wakeup_time          = xen_efi_get_wakeup_time;
+	efi.set_wakeup_time          = xen_efi_set_wakeup_time;
+	efi.get_variable             = xen_efi_get_variable;
+	efi.get_next_variable        = xen_efi_get_next_variable;
+	efi.set_variable             = xen_efi_set_variable;
+	efi.query_variable_info      = xen_efi_query_variable_info;
+	efi.update_capsule           = xen_efi_update_capsule;
+	efi.query_capsule_caps       = xen_efi_query_capsule_caps;
+	efi.get_next_high_mono_count = xen_efi_get_next_high_mono_count;
+	efi.reset_system             = NULL; /* Functionality provided by Xen. */
+}
+EXPORT_SYMBOL_GPL(xen_efi_runtime_setup);

arch/arm/xen/enlighten.c

@@ -0,0 +1,460 @@
+#include <xen/xen.h>
+#include <xen/events.h>
+#include <xen/grant_table.h>
+#include <xen/hvm.h>
+#include <xen/interface/vcpu.h>
+#include <xen/interface/xen.h>
+#include <xen/interface/memory.h>
+#include <xen/interface/hvm/params.h>
+#include <xen/features.h>
+#include <xen/platform_pci.h>
+#include <xen/xenbus.h>
+#include <xen/page.h>
+#include <xen/interface/sched.h>
+#include <xen/xen-ops.h>
+#include <asm/xen/hypervisor.h>
+#include <asm/xen/hypercall.h>
+#include <asm/xen/xen-ops.h>
+#include <asm/system_misc.h>
+#include <asm/efi.h>
+#include <linux/interrupt.h>
+#include <linux/irqreturn.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_fdt.h>
+#include <linux/of_irq.h>
+#include <linux/of_address.h>
+#include <linux/cpuidle.h>
+#include <linux/cpufreq.h>
+#include <linux/cpu.h>
+#include <linux/console.h>
+#include <linux/pvclock_gtod.h>
+#include <linux/time64.h>
+#include <linux/timekeeping.h>
+#include <linux/timekeeper_internal.h>
+#include <linux/acpi.h>
+
+#include <linux/mm.h>
+
+struct start_info _xen_start_info;
+struct start_info *xen_start_info = &_xen_start_info;
+EXPORT_SYMBOL(xen_start_info);
+
+enum xen_domain_type xen_domain_type = XEN_NATIVE;
+EXPORT_SYMBOL(xen_domain_type);
+
+struct shared_info xen_dummy_shared_info;
+struct shared_info *HYPERVISOR_shared_info = (void *)&xen_dummy_shared_info;
+
+DEFINE_PER_CPU(struct vcpu_info *, xen_vcpu);
+static struct vcpu_info __percpu *xen_vcpu_info;
+
+/* Linux <-> Xen vCPU id mapping */
+DEFINE_PER_CPU(uint32_t, xen_vcpu_id);
+EXPORT_PER_CPU_SYMBOL(xen_vcpu_id);
+
+/* These are unused until we support booting "pre-ballooned" */
+unsigned long xen_released_pages;
+struct xen_memory_region xen_extra_mem[XEN_EXTRA_MEM_MAX_REGIONS] __initdata;
+
+static __read_mostly unsigned int xen_events_irq;
+
+int xen_remap_domain_gfn_array(struct vm_area_struct *vma,
+			       unsigned long addr,
+			       xen_pfn_t *gfn, int nr,
+			       int *err_ptr, pgprot_t prot,
+			       unsigned domid,
+			       struct page **pages)
+{
+	return xen_xlate_remap_gfn_array(vma, addr, gfn, nr, err_ptr,
+					 prot, domid, pages);
+}
+EXPORT_SYMBOL_GPL(xen_remap_domain_gfn_array);
+
+/* Not used by XENFEAT_auto_translated guests. */
+int xen_remap_domain_gfn_range(struct vm_area_struct *vma,
+                              unsigned long addr,
+                              xen_pfn_t gfn, int nr,
+                              pgprot_t prot, unsigned domid,
+                              struct page **pages)
+{
+	return -ENOSYS;
+}
+EXPORT_SYMBOL_GPL(xen_remap_domain_gfn_range);
+
+int xen_unmap_domain_gfn_range(struct vm_area_struct *vma,
+			       int nr, struct page **pages)
+{
+	return xen_xlate_unmap_gfn_range(vma, nr, pages);
+}
+EXPORT_SYMBOL_GPL(xen_unmap_domain_gfn_range);
+
+static void xen_read_wallclock(struct timespec64 *ts)
+{
+	u32 version;
+	struct timespec64 now, ts_monotonic;
+	struct shared_info *s = HYPERVISOR_shared_info;
+	struct pvclock_wall_clock *wall_clock = &(s->wc);
+
+	/* get wallclock at system boot */
+	do {
+		version = wall_clock->version;
+		rmb();		/* fetch version before time */
+		now.tv_sec  = ((uint64_t)wall_clock->sec_hi << 32) | wall_clock->sec;
+		now.tv_nsec = wall_clock->nsec;
+		rmb();		/* fetch time before checking version */
+	} while ((wall_clock->version & 1) || (version != wall_clock->version));
+
+	/* time since system boot */
+	ktime_get_ts64(&ts_monotonic);
+	*ts = timespec64_add(now, ts_monotonic);
+}
+
+static int xen_pvclock_gtod_notify(struct notifier_block *nb,
+				   unsigned long was_set, void *priv)
+{
+	/* Protected by the calling core code serialization */
+	static struct timespec64 next_sync;
+
+	struct xen_platform_op op;
+	struct timespec64 now, system_time;
+	struct timekeeper *tk = priv;
+
+	now.tv_sec = tk->xtime_sec;
+	now.tv_nsec = (long)(tk->tkr_mono.xtime_nsec >> tk->tkr_mono.shift);
+	system_time = timespec64_add(now, tk->wall_to_monotonic);
+
+	/*
+	 * We only take the expensive HV call when the clock was set
+	 * or when the 11 minutes RTC synchronization time elapsed.
+	 */
+	if (!was_set && timespec64_compare(&now, &next_sync) < 0)
+		return NOTIFY_OK;
+
+	op.cmd = XENPF_settime64;
+	op.u.settime64.mbz = 0;
+	op.u.settime64.secs = now.tv_sec;
+	op.u.settime64.nsecs = now.tv_nsec;
+	op.u.settime64.system_time = timespec64_to_ns(&system_time);
+	(void)HYPERVISOR_platform_op(&op);
+
+	/*
+	 * Move the next drift compensation time 11 minutes
+	 * ahead. That's emulating the sync_cmos_clock() update for
+	 * the hardware RTC.
+	 */
+	next_sync = now;
+	next_sync.tv_sec += 11 * 60;
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block xen_pvclock_gtod_notifier = {
+	.notifier_call = xen_pvclock_gtod_notify,
+};
+
+static int xen_starting_cpu(unsigned int cpu)
+{
+	struct vcpu_register_vcpu_info info;
+	struct vcpu_info *vcpup;
+	int err;
+
+	/* 
+	 * VCPUOP_register_vcpu_info cannot be called twice for the same
+	 * vcpu, so if vcpu_info is already registered, just get out. This
+	 * can happen with cpu-hotplug.
+	 */
+	if (per_cpu(xen_vcpu, cpu) != NULL)
+		goto after_register_vcpu_info;
+
+	pr_info("Xen: initializing cpu%d\n", cpu);
+	vcpup = per_cpu_ptr(xen_vcpu_info, cpu);
+
+	info.mfn = virt_to_gfn(vcpup);
+	info.offset = xen_offset_in_page(vcpup);
+
+	err = HYPERVISOR_vcpu_op(VCPUOP_register_vcpu_info, xen_vcpu_nr(cpu),
+				 &info);
+	BUG_ON(err);
+	per_cpu(xen_vcpu, cpu) = vcpup;
+
+	xen_setup_runstate_info(cpu);
+
+after_register_vcpu_info:
+	enable_percpu_irq(xen_events_irq, 0);
+	return 0;
+}
+
+static int xen_dying_cpu(unsigned int cpu)
+{
+	disable_percpu_irq(xen_events_irq);
+	return 0;
+}
+
+static void xen_restart(enum reboot_mode reboot_mode, const char *cmd)
+{
+	struct sched_shutdown r = { .reason = SHUTDOWN_reboot };
+	int rc;
+	rc = HYPERVISOR_sched_op(SCHEDOP_shutdown, &r);
+	BUG_ON(rc);
+}
+
+static void xen_power_off(void)
+{
+	struct sched_shutdown r = { .reason = SHUTDOWN_poweroff };
+	int rc;
+	rc = HYPERVISOR_sched_op(SCHEDOP_shutdown, &r);
+	BUG_ON(rc);
+}
+
+static irqreturn_t xen_arm_callback(int irq, void *arg)
+{
+	xen_hvm_evtchn_do_upcall();
+	return IRQ_HANDLED;
+}
+
+static __initdata struct {
+	const char *compat;
+	const char *prefix;
+	const char *version;
+	bool found;
+} hyper_node = {"xen,xen", "xen,xen-", NULL, false};
+
+static int __init fdt_find_hyper_node(unsigned long node, const char *uname,
+				      int depth, void *data)
+{
+	const void *s = NULL;
+	int len;
+
+	if (depth != 1 || strcmp(uname, "hypervisor") != 0)
+		return 0;
+
+	if (of_flat_dt_is_compatible(node, hyper_node.compat))
+		hyper_node.found = true;
+
+	s = of_get_flat_dt_prop(node, "compatible", &len);
+	if (strlen(hyper_node.prefix) + 3  < len &&
+	    !strncmp(hyper_node.prefix, s, strlen(hyper_node.prefix)))
+		hyper_node.version = s + strlen(hyper_node.prefix);
+
+	/*
+	 * Check if Xen supports EFI by checking whether there is the
+	 * "/hypervisor/uefi" node in DT. If so, runtime services are available
+	 * through proxy functions (e.g. in case of Xen dom0 EFI implementation
+	 * they call special hypercall which executes relevant EFI functions)
+	 * and that is why they are always enabled.
+	 */
+	if (IS_ENABLED(CONFIG_XEN_EFI)) {
+		if ((of_get_flat_dt_subnode_by_name(node, "uefi") > 0) &&
+		    !efi_runtime_disabled())
+			set_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+	}
+
+	return 0;
+}
+
+/*
+ * see Documentation/devicetree/bindings/arm/xen.txt for the
+ * documentation of the Xen Device Tree format.
+ */
+#define GRANT_TABLE_PHYSADDR 0
+void __init xen_early_init(void)
+{
+	of_scan_flat_dt(fdt_find_hyper_node, NULL);
+	if (!hyper_node.found) {
+		pr_debug("No Xen support\n");
+		return;
+	}
+
+	if (hyper_node.version == NULL) {
+		pr_debug("Xen version not found\n");
+		return;
+	}
+
+	pr_info("Xen %s support found\n", hyper_node.version);
+
+	xen_domain_type = XEN_HVM_DOMAIN;
+
+	xen_setup_features();
+
+	if (xen_feature(XENFEAT_dom0))
+		xen_start_info->flags |= SIF_INITDOMAIN|SIF_PRIVILEGED;
+	else
+		xen_start_info->flags &= ~(SIF_INITDOMAIN|SIF_PRIVILEGED);
+
+	if (!console_set_on_cmdline && !xen_initial_domain())
+		add_preferred_console("hvc", 0, NULL);
+}
+
+static void __init xen_acpi_guest_init(void)
+{
+#ifdef CONFIG_ACPI
+	struct xen_hvm_param a;
+	int interrupt, trigger, polarity;
+
+	a.domid = DOMID_SELF;
+	a.index = HVM_PARAM_CALLBACK_IRQ;
+
+	if (HYPERVISOR_hvm_op(HVMOP_get_param, &a)
+	    || (a.value >> 56) != HVM_PARAM_CALLBACK_TYPE_PPI) {
+		xen_events_irq = 0;
+		return;
+	}
+
+	interrupt = a.value & 0xff;
+	trigger = ((a.value >> 8) & 0x1) ? ACPI_EDGE_SENSITIVE
+					 : ACPI_LEVEL_SENSITIVE;
+	polarity = ((a.value >> 8) & 0x2) ? ACPI_ACTIVE_LOW
+					  : ACPI_ACTIVE_HIGH;
+	xen_events_irq = acpi_register_gsi(NULL, interrupt, trigger, polarity);
+#endif
+}
+
+static void __init xen_dt_guest_init(void)
+{
+	struct device_node *xen_node;
+
+	xen_node = of_find_compatible_node(NULL, NULL, "xen,xen");
+	if (!xen_node) {
+		pr_err("Xen support was detected before, but it has disappeared\n");
+		return;
+	}
+
+	xen_events_irq = irq_of_parse_and_map(xen_node, 0);
+}
+
+static int __init xen_guest_init(void)
+{
+	struct xen_add_to_physmap xatp;
+	struct shared_info *shared_info_page = NULL;
+	int cpu;
+
+	if (!xen_domain())
+		return 0;
+
+	if (!acpi_disabled)
+		xen_acpi_guest_init();
+	else
+		xen_dt_guest_init();
+
+	if (!xen_events_irq) {
+		pr_err("Xen event channel interrupt not found\n");
+		return -ENODEV;
+	}
+
+	/*
+	 * The fdt parsing codes have set EFI_RUNTIME_SERVICES if Xen EFI
+	 * parameters are found. Force enable runtime services.
+	 */
+	if (efi_enabled(EFI_RUNTIME_SERVICES))
+		xen_efi_runtime_setup();
+
+	shared_info_page = (struct shared_info *)get_zeroed_page(GFP_KERNEL);
+
+	if (!shared_info_page) {
+		pr_err("not enough memory\n");
+		return -ENOMEM;
+	}
+	xatp.domid = DOMID_SELF;
+	xatp.idx = 0;
+	xatp.space = XENMAPSPACE_shared_info;
+	xatp.gpfn = virt_to_gfn(shared_info_page);
+	if (HYPERVISOR_memory_op(XENMEM_add_to_physmap, &xatp))
+		BUG();
+
+	HYPERVISOR_shared_info = (struct shared_info *)shared_info_page;
+
+	/* xen_vcpu is a pointer to the vcpu_info struct in the shared_info
+	 * page, we use it in the event channel upcall and in some pvclock
+	 * related functions. 
+	 * The shared info contains exactly 1 CPU (the boot CPU). The guest
+	 * is required to use VCPUOP_register_vcpu_info to place vcpu info
+	 * for secondary CPUs as they are brought up.
+	 * For uniformity we use VCPUOP_register_vcpu_info even on cpu0.
+	 */
+	xen_vcpu_info = alloc_percpu(struct vcpu_info);
+	if (xen_vcpu_info == NULL)
+		return -ENOMEM;
+
+	/* Direct vCPU id mapping for ARM guests. */
+	for_each_possible_cpu(cpu)
+		per_cpu(xen_vcpu_id, cpu) = cpu;
+
+	xen_auto_xlat_grant_frames.count = gnttab_max_grant_frames();
+	if (xen_xlate_map_ballooned_pages(&xen_auto_xlat_grant_frames.pfn,
+					  &xen_auto_xlat_grant_frames.vaddr,
+					  xen_auto_xlat_grant_frames.count)) {
+		free_percpu(xen_vcpu_info);
+		return -ENOMEM;
+	}
+	gnttab_init();
+	if (!xen_initial_domain())
+		xenbus_probe(NULL);
+
+	/*
+	 * Making sure board specific code will not set up ops for
+	 * cpu idle and cpu freq.
+	 */
+	disable_cpuidle();
+	disable_cpufreq();
+
+	xen_init_IRQ();
+
+	if (request_percpu_irq(xen_events_irq, xen_arm_callback,
+			       "events", &xen_vcpu)) {
+		pr_err("Error request IRQ %d\n", xen_events_irq);
+		return -EINVAL;
+	}
+
+	xen_time_setup_guest();
+
+	if (xen_initial_domain())
+		pvclock_gtod_register_notifier(&xen_pvclock_gtod_notifier);
+
+	return cpuhp_setup_state(CPUHP_AP_ARM_XEN_STARTING,
+				 "AP_ARM_XEN_STARTING", xen_starting_cpu,
+				 xen_dying_cpu);
+}
+early_initcall(xen_guest_init);
+
+static int __init xen_pm_init(void)
+{
+	if (!xen_domain())
+		return -ENODEV;
+
+	pm_power_off = xen_power_off;
+	arm_pm_restart = xen_restart;
+	if (!xen_initial_domain()) {
+		struct timespec64 ts;
+		xen_read_wallclock(&ts);
+		do_settimeofday64(&ts);
+	}
+
+	return 0;
+}
+late_initcall(xen_pm_init);
+
+
+/* empty stubs */
+void xen_arch_pre_suspend(void) { }
+void xen_arch_post_suspend(int suspend_cancelled) { }
+void xen_timer_resume(void) { }
+void xen_arch_resume(void) { }
+void xen_arch_suspend(void) { }
+
+
+/* In the hypercall.S file. */
+EXPORT_SYMBOL_GPL(HYPERVISOR_event_channel_op);
+EXPORT_SYMBOL_GPL(HYPERVISOR_grant_table_op);
+EXPORT_SYMBOL_GPL(HYPERVISOR_xen_version);
+EXPORT_SYMBOL_GPL(HYPERVISOR_console_io);
+EXPORT_SYMBOL_GPL(HYPERVISOR_sched_op);
+EXPORT_SYMBOL_GPL(HYPERVISOR_hvm_op);
+EXPORT_SYMBOL_GPL(HYPERVISOR_memory_op);
+EXPORT_SYMBOL_GPL(HYPERVISOR_physdev_op);
+EXPORT_SYMBOL_GPL(HYPERVISOR_vcpu_op);
+EXPORT_SYMBOL_GPL(HYPERVISOR_tmem_op);
+EXPORT_SYMBOL_GPL(HYPERVISOR_platform_op);
+EXPORT_SYMBOL_GPL(HYPERVISOR_multicall);
+EXPORT_SYMBOL_GPL(HYPERVISOR_vm_assist);
+EXPORT_SYMBOL_GPL(privcmd_call);

arch/arm/xen/grant-table.c

@@ -0,0 +1,51 @@
+/******************************************************************************
+ * grant_table.c
+ * ARM specific part
+ *
+ * Granting foreign access to our memory reservation.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#include <xen/interface/xen.h>
+#include <xen/page.h>
+#include <xen/grant_table.h>
+
+int arch_gnttab_map_shared(xen_pfn_t *frames, unsigned long nr_gframes,
+			   unsigned long max_nr_gframes,
+			   void **__shared)
+{
+	return -ENOSYS;
+}
+
+void arch_gnttab_unmap(void *shared, unsigned long nr_gframes)
+{
+	return;
+}
+
+int arch_gnttab_init(unsigned long nr_shared)
+{
+	return 0;
+}

arch/arm/xen/hypercall.S

@@ -0,0 +1,122 @@
+/******************************************************************************
+ * hypercall.S
+ *
+ * Xen hypercall wrappers
+ *
+ * Stefano Stabellini <stefano.stabellini@eu.citrix.com>, Citrix, 2012
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+/*
+ * The Xen hypercall calling convention is very similar to the ARM
+ * procedure calling convention: the first paramter is passed in r0, the
+ * second in r1, the third in r2 and the fourth in r3. Considering that
+ * Xen hypercalls have 5 arguments at most, the fifth paramter is passed
+ * in r4, differently from the procedure calling convention of using the
+ * stack for that case.
+ *
+ * The hypercall number is passed in r12.
+ *
+ * The return value is in r0.
+ *
+ * The hvc ISS is required to be 0xEA1, that is the Xen specific ARM
+ * hypercall tag.
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/opcodes-virt.h>
+#include <xen/interface/xen.h>
+
+
+#define XEN_IMM 0xEA1
+
+#define HYPERCALL_SIMPLE(hypercall)		\
+ENTRY(HYPERVISOR_##hypercall)			\
+	mov r12, #__HYPERVISOR_##hypercall;	\
+	__HVC(XEN_IMM);						\
+	ret lr;					\
+ENDPROC(HYPERVISOR_##hypercall)
+
+#define HYPERCALL0 HYPERCALL_SIMPLE
+#define HYPERCALL1 HYPERCALL_SIMPLE
+#define HYPERCALL2 HYPERCALL_SIMPLE
+#define HYPERCALL3 HYPERCALL_SIMPLE
+#define HYPERCALL4 HYPERCALL_SIMPLE
+
+#define HYPERCALL5(hypercall)			\
+ENTRY(HYPERVISOR_##hypercall)			\
+	stmdb sp!, {r4}						\
+	ldr r4, [sp, #4]					\
+	mov r12, #__HYPERVISOR_##hypercall;	\
+	__HVC(XEN_IMM);						\
+	ldm sp!, {r4}						\
+	ret lr					\
+ENDPROC(HYPERVISOR_##hypercall)
+
+                .text
+
+HYPERCALL2(xen_version);
+HYPERCALL3(console_io);
+HYPERCALL3(grant_table_op);
+HYPERCALL2(sched_op);
+HYPERCALL2(event_channel_op);
+HYPERCALL2(hvm_op);
+HYPERCALL2(memory_op);
+HYPERCALL2(physdev_op);
+HYPERCALL3(vcpu_op);
+HYPERCALL1(tmem_op);
+HYPERCALL1(platform_op_raw);
+HYPERCALL2(multicall);
+HYPERCALL2(vm_assist);
+
+ENTRY(privcmd_call)
+	stmdb sp!, {r4}
+	mov r12, r0
+	mov r0, r1
+	mov r1, r2
+	mov r2, r3
+	ldr r3, [sp, #8]
+	/*
+	 * Privcmd calls are issued by the userspace. We need to allow the
+	 * kernel to access the userspace memory before issuing the hypercall.
+	 */
+	uaccess_enable r4
+
+	/* r4 is loaded now as we use it as scratch register before */
+	ldr r4, [sp, #4]
+	__HVC(XEN_IMM)
+
+	/*
+	 * Disable userspace access from kernel. This is fine to do it
+	 * unconditionally as no set_fs(KERNEL_DS)/set_fs(get_ds()) is
+	 * called before.
+	 */
+	uaccess_disable r4
+
+	ldm sp!, {r4}
+	ret lr
+ENDPROC(privcmd_call);

arch/arm/xen/mm.c

@@ -0,0 +1,221 @@
+#include <linux/cpu.h>
+#include <linux/dma-mapping.h>
+#include <linux/bootmem.h>
+#include <linux/gfp.h>
+#include <linux/highmem.h>
+#include <linux/export.h>
+#include <linux/memblock.h>
+#include <linux/of_address.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+#include <linux/dma-mapping.h>
+#include <linux/vmalloc.h>
+#include <linux/swiotlb.h>
+
+#include <xen/xen.h>
+#include <xen/interface/grant_table.h>
+#include <xen/interface/memory.h>
+#include <xen/page.h>
+#include <xen/swiotlb-xen.h>
+
+#include <asm/cacheflush.h>
+#include <asm/xen/hypercall.h>
+#include <asm/xen/interface.h>
+
+unsigned long xen_get_swiotlb_free_pages(unsigned int order)
+{
+	struct memblock_region *reg;
+	gfp_t flags = __GFP_NOWARN|__GFP_KSWAPD_RECLAIM;
+
+	for_each_memblock(memory, reg) {
+		if (reg->base < (phys_addr_t)0xffffffff) {
+			flags |= __GFP_DMA;
+			break;
+		}
+	}
+	return __get_free_pages(flags, order);
+}
+
+enum dma_cache_op {
+       DMA_UNMAP,
+       DMA_MAP,
+};
+static bool hypercall_cflush = false;
+
+/* functions called by SWIOTLB */
+
+static void dma_cache_maint(dma_addr_t handle, unsigned long offset,
+	size_t size, enum dma_data_direction dir, enum dma_cache_op op)
+{
+	struct gnttab_cache_flush cflush;
+	unsigned long xen_pfn;
+	size_t left = size;
+
+	xen_pfn = (handle >> XEN_PAGE_SHIFT) + offset / XEN_PAGE_SIZE;
+	offset %= XEN_PAGE_SIZE;
+
+	do {
+		size_t len = left;
+	
+		/* buffers in highmem or foreign pages cannot cross page
+		 * boundaries */
+		if (len + offset > XEN_PAGE_SIZE)
+			len = XEN_PAGE_SIZE - offset;
+
+		cflush.op = 0;
+		cflush.a.dev_bus_addr = xen_pfn << XEN_PAGE_SHIFT;
+		cflush.offset = offset;
+		cflush.length = len;
+
+		if (op == DMA_UNMAP && dir != DMA_TO_DEVICE)
+			cflush.op = GNTTAB_CACHE_INVAL;
+		if (op == DMA_MAP) {
+			if (dir == DMA_FROM_DEVICE)
+				cflush.op = GNTTAB_CACHE_INVAL;
+			else
+				cflush.op = GNTTAB_CACHE_CLEAN;
+		}
+		if (cflush.op)
+			HYPERVISOR_grant_table_op(GNTTABOP_cache_flush, &cflush, 1);
+
+		offset = 0;
+		xen_pfn++;
+		left -= len;
+	} while (left);
+}
+
+static void __xen_dma_page_dev_to_cpu(struct device *hwdev, dma_addr_t handle,
+		size_t size, enum dma_data_direction dir)
+{
+	dma_cache_maint(handle & PAGE_MASK, handle & ~PAGE_MASK, size, dir, DMA_UNMAP);
+}
+
+static void __xen_dma_page_cpu_to_dev(struct device *hwdev, dma_addr_t handle,
+		size_t size, enum dma_data_direction dir)
+{
+	dma_cache_maint(handle & PAGE_MASK, handle & ~PAGE_MASK, size, dir, DMA_MAP);
+}
+
+void __xen_dma_map_page(struct device *hwdev, struct page *page,
+	     dma_addr_t dev_addr, unsigned long offset, size_t size,
+	     enum dma_data_direction dir, unsigned long attrs)
+{
+	if (is_device_dma_coherent(hwdev))
+		return;
+	if (attrs & DMA_ATTR_SKIP_CPU_SYNC)
+		return;
+
+	__xen_dma_page_cpu_to_dev(hwdev, dev_addr, size, dir);
+}
+
+void __xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle,
+		size_t size, enum dma_data_direction dir,
+		unsigned long attrs)
+
+{
+	if (is_device_dma_coherent(hwdev))
+		return;
+	if (attrs & DMA_ATTR_SKIP_CPU_SYNC)
+		return;
+
+	__xen_dma_page_dev_to_cpu(hwdev, handle, size, dir);
+}
+
+void __xen_dma_sync_single_for_cpu(struct device *hwdev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir)
+{
+	if (is_device_dma_coherent(hwdev))
+		return;
+	__xen_dma_page_dev_to_cpu(hwdev, handle, size, dir);
+}
+
+void __xen_dma_sync_single_for_device(struct device *hwdev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir)
+{
+	if (is_device_dma_coherent(hwdev))
+		return;
+	__xen_dma_page_cpu_to_dev(hwdev, handle, size, dir);
+}
+
+bool xen_arch_need_swiotlb(struct device *dev,
+			   phys_addr_t phys,
+			   dma_addr_t dev_addr)
+{
+	unsigned int xen_pfn = XEN_PFN_DOWN(phys);
+	unsigned int bfn = XEN_PFN_DOWN(dev_addr);
+
+	/*
+	 * The swiotlb buffer should be used if
+	 *	- Xen doesn't have the cache flush hypercall
+	 *	- The Linux page refers to foreign memory
+	 *	- The device doesn't support coherent DMA request
+	 *
+	 * The Linux page may be spanned acrros multiple Xen page, although
+	 * it's not possible to have a mix of local and foreign Xen page.
+	 * Furthermore, range_straddles_page_boundary is already checking
+	 * if buffer is physically contiguous in the host RAM.
+	 *
+	 * Therefore we only need to check the first Xen page to know if we
+	 * require a bounce buffer because the device doesn't support coherent
+	 * memory and we are not able to flush the cache.
+	 */
+	return (!hypercall_cflush && (xen_pfn != bfn) &&
+		!is_device_dma_coherent(dev));
+}
+
+int xen_create_contiguous_region(phys_addr_t pstart, unsigned int order,
+				 unsigned int address_bits,
+				 dma_addr_t *dma_handle)
+{
+	if (!xen_initial_domain())
+		return -EINVAL;
+
+	/* we assume that dom0 is mapped 1:1 for now */
+	*dma_handle = pstart;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(xen_create_contiguous_region);
+
+void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order)
+{
+	return;
+}
+EXPORT_SYMBOL_GPL(xen_destroy_contiguous_region);
+
+struct dma_map_ops *xen_dma_ops;
+EXPORT_SYMBOL(xen_dma_ops);
+
+static struct dma_map_ops xen_swiotlb_dma_ops = {
+	.mapping_error = xen_swiotlb_dma_mapping_error,
+	.alloc = xen_swiotlb_alloc_coherent,
+	.free = xen_swiotlb_free_coherent,
+	.sync_single_for_cpu = xen_swiotlb_sync_single_for_cpu,
+	.sync_single_for_device = xen_swiotlb_sync_single_for_device,
+	.sync_sg_for_cpu = xen_swiotlb_sync_sg_for_cpu,
+	.sync_sg_for_device = xen_swiotlb_sync_sg_for_device,
+	.map_sg = xen_swiotlb_map_sg_attrs,
+	.unmap_sg = xen_swiotlb_unmap_sg_attrs,
+	.map_page = xen_swiotlb_map_page,
+	.unmap_page = xen_swiotlb_unmap_page,
+	.dma_supported = xen_swiotlb_dma_supported,
+	.set_dma_mask = xen_swiotlb_set_dma_mask,
+	.mmap = xen_swiotlb_dma_mmap,
+};
+
+int __init xen_mm_init(void)
+{
+	struct gnttab_cache_flush cflush;
+	if (!xen_initial_domain())
+		return 0;
+	xen_swiotlb_init(1, false);
+	xen_dma_ops = &xen_swiotlb_dma_ops;
+
+	cflush.op = 0;
+	cflush.a.dev_bus_addr = 0;
+	cflush.offset = 0;
+	cflush.length = 0;
+	if (HYPERVISOR_grant_table_op(GNTTABOP_cache_flush, &cflush, 1) != -ENOSYS)
+		hypercall_cflush = true;
+	return 0;
+}
+arch_initcall(xen_mm_init);

arch/arm/xen/p2m.c

@@ -0,0 +1,177 @@
+#include <linux/bootmem.h>
+#include <linux/gfp.h>
+#include <linux/export.h>
+#include <linux/rwlock.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+#include <linux/dma-mapping.h>
+#include <linux/vmalloc.h>
+#include <linux/swiotlb.h>
+
+#include <xen/xen.h>
+#include <xen/interface/memory.h>
+#include <xen/page.h>
+#include <xen/swiotlb-xen.h>
+
+#include <asm/cacheflush.h>
+#include <asm/xen/hypercall.h>
+#include <asm/xen/interface.h>
+
+struct xen_p2m_entry {
+	unsigned long pfn;
+	unsigned long mfn;
+	unsigned long nr_pages;
+	struct rb_node rbnode_phys;
+};
+
+static rwlock_t p2m_lock;
+struct rb_root phys_to_mach = RB_ROOT;
+EXPORT_SYMBOL_GPL(phys_to_mach);
+
+static int xen_add_phys_to_mach_entry(struct xen_p2m_entry *new)
+{
+	struct rb_node **link = &phys_to_mach.rb_node;
+	struct rb_node *parent = NULL;
+	struct xen_p2m_entry *entry;
+	int rc = 0;
+
+	while (*link) {
+		parent = *link;
+		entry = rb_entry(parent, struct xen_p2m_entry, rbnode_phys);
+
+		if (new->pfn == entry->pfn)
+			goto err_out;
+
+		if (new->pfn < entry->pfn)
+			link = &(*link)->rb_left;
+		else
+			link = &(*link)->rb_right;
+	}
+	rb_link_node(&new->rbnode_phys, parent, link);
+	rb_insert_color(&new->rbnode_phys, &phys_to_mach);
+	goto out;
+
+err_out:
+	rc = -EINVAL;
+	pr_warn("%s: cannot add pfn=%pa -> mfn=%pa: pfn=%pa -> mfn=%pa already exists\n",
+			__func__, &new->pfn, &new->mfn, &entry->pfn, &entry->mfn);
+out:
+	return rc;
+}
+
+unsigned long __pfn_to_mfn(unsigned long pfn)
+{
+	struct rb_node *n = phys_to_mach.rb_node;
+	struct xen_p2m_entry *entry;
+	unsigned long irqflags;
+
+	read_lock_irqsave(&p2m_lock, irqflags);
+	while (n) {
+		entry = rb_entry(n, struct xen_p2m_entry, rbnode_phys);
+		if (entry->pfn <= pfn &&
+				entry->pfn + entry->nr_pages > pfn) {
+			read_unlock_irqrestore(&p2m_lock, irqflags);
+			return entry->mfn + (pfn - entry->pfn);
+		}
+		if (pfn < entry->pfn)
+			n = n->rb_left;
+		else
+			n = n->rb_right;
+	}
+	read_unlock_irqrestore(&p2m_lock, irqflags);
+
+	return INVALID_P2M_ENTRY;
+}
+EXPORT_SYMBOL_GPL(__pfn_to_mfn);
+
+int set_foreign_p2m_mapping(struct gnttab_map_grant_ref *map_ops,
+			    struct gnttab_map_grant_ref *kmap_ops,
+			    struct page **pages, unsigned int count)
+{
+	int i;
+
+	for (i = 0; i < count; i++) {
+		if (map_ops[i].status)
+			continue;
+		set_phys_to_machine(map_ops[i].host_addr >> XEN_PAGE_SHIFT,
+				    map_ops[i].dev_bus_addr >> XEN_PAGE_SHIFT);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(set_foreign_p2m_mapping);
+
+int clear_foreign_p2m_mapping(struct gnttab_unmap_grant_ref *unmap_ops,
+			      struct gnttab_unmap_grant_ref *kunmap_ops,
+			      struct page **pages, unsigned int count)
+{
+	int i;
+
+	for (i = 0; i < count; i++) {
+		set_phys_to_machine(unmap_ops[i].host_addr >> XEN_PAGE_SHIFT,
+				    INVALID_P2M_ENTRY);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(clear_foreign_p2m_mapping);
+
+bool __set_phys_to_machine_multi(unsigned long pfn,
+		unsigned long mfn, unsigned long nr_pages)
+{
+	int rc;
+	unsigned long irqflags;
+	struct xen_p2m_entry *p2m_entry;
+	struct rb_node *n = phys_to_mach.rb_node;
+
+	if (mfn == INVALID_P2M_ENTRY) {
+		write_lock_irqsave(&p2m_lock, irqflags);
+		while (n) {
+			p2m_entry = rb_entry(n, struct xen_p2m_entry, rbnode_phys);
+			if (p2m_entry->pfn <= pfn &&
+					p2m_entry->pfn + p2m_entry->nr_pages > pfn) {
+				rb_erase(&p2m_entry->rbnode_phys, &phys_to_mach);
+				write_unlock_irqrestore(&p2m_lock, irqflags);
+				kfree(p2m_entry);
+				return true;
+			}
+			if (pfn < p2m_entry->pfn)
+				n = n->rb_left;
+			else
+				n = n->rb_right;
+		}
+		write_unlock_irqrestore(&p2m_lock, irqflags);
+		return true;
+	}
+
+	p2m_entry = kzalloc(sizeof(struct xen_p2m_entry), GFP_NOWAIT);
+	if (!p2m_entry) {
+		pr_warn("cannot allocate xen_p2m_entry\n");
+		return false;
+	}
+	p2m_entry->pfn = pfn;
+	p2m_entry->nr_pages = nr_pages;
+	p2m_entry->mfn = mfn;
+
+	write_lock_irqsave(&p2m_lock, irqflags);
+	if ((rc = xen_add_phys_to_mach_entry(p2m_entry)) < 0) {
+		write_unlock_irqrestore(&p2m_lock, irqflags);
+		return false;
+	}
+	write_unlock_irqrestore(&p2m_lock, irqflags);
+	return true;
+}
+EXPORT_SYMBOL_GPL(__set_phys_to_machine_multi);
+
+bool __set_phys_to_machine(unsigned long pfn, unsigned long mfn)
+{
+	return __set_phys_to_machine_multi(pfn, mfn, 1);
+}
+EXPORT_SYMBOL_GPL(__set_phys_to_machine);
+
+static int p2m_init(void)
+{
+	rwlock_init(&p2m_lock);
+	return 0;
+}
+arch_initcall(p2m_init);