Commit Graph

115 Commits

Author SHA1 Message Date
Leif Lindholm
86c8b27a01 arm64: ignore DT memreserve entries when booting in UEFI mode
UEFI provides its own method for marking regions to reserve, via the
memory map which is also used to initialise memblock. So when using the
UEFI memory map, ignore any memreserve entries present in the DT.

Reported-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: Mark Rutland <mark.rutland@arm.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
Signed-off-by: Will Deacon <will.deacon@arm.com>
2014-08-19 20:22:03 +01:00
Linus Torvalds
5167d09ffa arm64 updates for 3.17
Changes include:
  - Context tracking support (NO_HZ_FULL) which narrowly missed 3.16
  - vDSO layout rework following Andy's work on x86
  - TEXT_OFFSET fuzzing for bootloader testing
  - /proc/cpuinfo tidy-up
  - Preliminary work to support 48-bit virtual addresses, but this is
    currently disabled until KVM has been ported to use it (the patches
    do, however, bring some nice clean-up)
  - Boot-time CPU sanity checks (especially useful on heterogenous
    systems)
  - Support for syscall auditing
  - Support for CC_STACKPROTECTOR
  - defconfig updates
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1
 
 iQIcBAABCgAGBQJT3qkzAAoJEC379FI+VC/ZxwEP/3uYs9glDLTd1hmVFr1cRutg
 j4m1Kc7RCO+zpbYCXJLAQLPjwjOaUWPZUeZPQZib6bO+4sTqFYe9vsaqRyvn/bxM
 BaQhytpyxymfG8m3rmXaI97TzBwnRB2oQ0k36rsjMwG/VQMLf9kVuEwURoAHF07l
 RyMK2sAwE0/8XIJZQFNo5SAbkO52EiHlehdlTzCXGWWOWdHDyVfks/k6YhIS991r
 0W9Y0ghHaMz+mAumTSq7jzPQa3aF3GjTp0W7gJjk/PRBDHfPisphEO36zsA0yHtE
 3uvEH0kUQK/ve4ZUQiNvuEZCSqalPFag6j5Z8BnFtafa66J5h414CGPAfER6Kz7+
 KGpoEve+7Rpvvb1S4T0tTMg7HoGrvqc5wKS3uFxfoGooGUcUOchSkYiVTBMDJSKn
 QlJbb1QSvuNFGhcKntTOe1QMT+x0w9urq/e+QfnQrZ/m5Er7J3qCZzeOfA2JFTjQ
 sB24yjzAz5a5VwbKbuB2b4gDILY9oYNe94HFP08o/rJfANnL0dpP1Oyl0b12ILsI
 a69EMdpaeEQo8703KLIlzfW6u92PqYs6UkYvya8o27FAvmNvDfB/PffjgVsOAHFi
 Qc+dpYbnzNfwJgG9w0qhJ+MR8g5fiBYHqNpfGOY+g5M50j0hZUX9comoWw1xkl0X
 HlvG7xzrTF7/VbWEtZ2o
 =6XMc
 -----END PGP SIGNATURE-----

Merge tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux

Pull arm64 updates from Will Deacon:
 "Once again, Catalin's off on holiday and I'm looking after the arm64
  tree.  Please can you pull the following arm64 updates for 3.17?

  Note that this branch also includes the new GICv3 driver (merged via a
  stable tag from Jason's irqchip tree), since there is a fix for older
  binutils on top.

  Changes include:
   - context tracking support (NO_HZ_FULL) which narrowly missed 3.16
   - vDSO layout rework following Andy's work on x86
   - TEXT_OFFSET fuzzing for bootloader testing
   - /proc/cpuinfo tidy-up
   - preliminary work to support 48-bit virtual addresses, but this is
     currently disabled until KVM has been ported to use it (the patches
     do, however, bring some nice clean-up)
   - boot-time CPU sanity checks (especially useful on heterogenous
     systems)
   - support for syscall auditing
   - support for CC_STACKPROTECTOR
   - defconfig updates"

* tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (55 commits)
  arm64: add newline to I-cache policy string
  Revert "arm64: dmi: Add SMBIOS/DMI support"
  arm64: fpsimd: fix a typo in fpsimd_save_partial_state ENDPROC
  arm64: don't call break hooks for BRK exceptions from EL0
  arm64: defconfig: enable devtmpfs mount option
  arm64: vdso: fix build error when switching from LE to BE
  arm64: defconfig: add virtio support for running as a kvm guest
  arm64: gicv3: Allow GICv3 compilation with older binutils
  arm64: fix soft lockup due to large tlb flush range
  arm64/crypto: fix makefile rule for aes-glue-%.o
  arm64: Do not invoke audit_syscall_* functions if !CONFIG_AUDIT_SYSCALL
  arm64: Fix barriers used for page table modifications
  arm64: Add support for 48-bit VA space with 64KB page configuration
  arm64: asm/pgtable.h pmd/pud definitions clean-up
  arm64: Determine the vmalloc/vmemmap space at build time based on VA_BITS
  arm64: Clean up the initial page table creation in head.S
  arm64: Remove asm/pgtable-*level-types.h files
  arm64: Remove asm/pgtable-*level-hwdef.h files
  arm64: Convert bool ARM64_x_LEVELS to int ARM64_PGTABLE_LEVELS
  arm64: mm: Implement 4 levels of translation tables
  ...
2014-08-04 12:31:53 -07:00
Catalin Marinas
08375198b0 arm64: Determine the vmalloc/vmemmap space at build time based on VA_BITS
Rather than guessing what the maximum vmmemap space should be, this
patch allows the calculation based on the VA_BITS and sizeof(struct
page). The vmalloc space extends to the beginning of the vmemmap space.

Since the virtual kernel memory layout now depends on the build
configuration, this patch removes the detailed description in
Documentation/arm64/memory.txt in favour of information printed during
kernel booting.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Tested-by: Jungseok Lee <jungseoklee85@gmail.com>
2014-07-23 15:28:05 +01:00
Catalin Marinas
abe669d7e1 arm64: Convert bool ARM64_x_LEVELS to int ARM64_PGTABLE_LEVELS
Rather than having several Kconfig options, define int
ARM64_PGTABLE_LEVELS which will be also useful in converting some of the
pgtable macros.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Tested-by: Jungseok Lee <jungseoklee85@gmail.com>
2014-07-23 15:27:46 +01:00
Jungseok Lee
c79b954bf6 arm64: mm: Implement 4 levels of translation tables
This patch implements 4 levels of translation tables since 3 levels
of page tables with 4KB pages cannot support 40-bit physical address
space described in [1] due to the following issue.

It is a restriction that kernel logical memory map with 4KB + 3 levels
(0xffffffc000000000-0xffffffffffffffff) cannot cover RAM region from
544GB to 1024GB in [1]. Specifically, ARM64 kernel fails to create
mapping for this region in map_mem function since __phys_to_virt for
this region reaches to address overflow.

If SoC design follows the document, [1], over 32GB RAM would be placed
from 544GB. Even 64GB system is supposed to use the region from 544GB
to 576GB for only 32GB RAM. Naturally, it would reach to enable 4 levels
of page tables to avoid hacking __virt_to_phys and __phys_to_virt.

However, it is recommended 4 levels of page table should be only enabled
if memory map is too sparse or there is about 512GB RAM.

References
----------
[1]: Principles of ARM Memory Maps, White Paper, Issue C

Signed-off-by: Jungseok Lee <jays.lee@samsung.com>
Reviewed-by: Sungjinn Chung <sungjinn.chung@samsung.com>
Acked-by: Kukjin Kim <kgene.kim@samsung.com>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Reviewed-by: Steve Capper <steve.capper@linaro.org>
[catalin.marinas@arm.com: MEMBLOCK_INITIAL_LIMIT removed, same as PUD_SIZE]
[catalin.marinas@arm.com: early_ioremap_init() updated for 4 levels]
[catalin.marinas@arm.com: 48-bit VA depends on BROKEN until KVM is fixed]
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Tested-by: Jungseok Lee <jungseoklee85@gmail.com>
2014-07-23 15:27:40 +01:00
Catalin Marinas
7edd88ad7e arm64: Do not initialise the fixmap page tables in head.S
The early_ioremap_init() function already handles fixmap pte
initialisation, so upgrade this to cover all of pud/pmd/pte and remove
one page from swapper_pg_dir.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Tested-by: Jungseok Lee <jungseoklee85@gmail.com>
2014-07-23 15:27:00 +01:00
Catalin Marinas
d50314a6b0 arm64: Create non-empty ZONE_DMA when DRAM starts above 4GB
ZONE_DMA is created to allow 32-bit only devices to access memory in the
absence of an IOMMU. On systems where the memory starts above 4GB, it is
expected that some devices have a DMA offset hardwired to be able to
access the bottom of the memory. Linux currently supports DT bindings
for the DMA offsets but they are not (easily) available early during
boot.

This patch tries to guess a DMA offset and assumes that ZONE_DMA
corresponds to the 32-bit mask above the start of DRAM.

Fixes: 2d5a5612bc (arm64: Limit the CMA buffer to 32-bit if ZONE_DMA)
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Reported-by: Mark Salter <msalter@redhat.com>
Tested-by: Mark Salter <msalter@redhat.com>
Tested-by: Anup Patel <anup.patel@linaro.org>
2014-07-23 11:23:58 +01:00
Mark Rutland
bd00cd5f8c arm64: place initial page tables above the kernel
Currently we place swapper_pg_dir and idmap_pg_dir below the kernel
image, between PHYS_OFFSET and (PHYS_OFFSET + TEXT_OFFSET). However,
bootloaders may use portions of this memory below the kernel and we do
not parse the memory reservation list until after the MMU has been
enabled. As such we may clobber some memory a bootloader wishes to have
preserved.

To enable the use of all of this memory by bootloaders (when the
required memory reservations are communicated to the kernel) it is
necessary to move our initial page tables elsewhere. As we currently
have an effectively unbound requirement for memory at the end of the
kernel image for .bss, we can place the page tables here.

This patch moves the initial page table to the end of the kernel image,
after the BSS. As they do not consist of any initialised data they will
be stripped from the kernel Image as with the BSS. The BSS clearing
routine is updated to stop at __bss_stop rather than _end so as to not
clobber the page tables, and memory reservations made redundant by the
new organisation are removed.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Tested-by: Laura Abbott <lauraa@codeaurora.org>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-07-10 12:36:12 +01:00
Mark Salter
bec7cedc8a arm64: export __cpu_{clear,copy}_user_page functions
The __cpu_clear_user_page() and __cpu_copy_user_page() functions
are not currently exported. This prevents modules from using
clear_user_page() and copy_user_page().

Signed-off-by: Mark Salter <msalter@redhat.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-07-08 17:30:51 +01:00
Steve Capper
923b8f5044 arm64: mm: Make icache synchronisation logic huge page aware
The __sync_icache_dcache routine will only flush the dcache for the
first page of a compound page, potentially leading to stale icache
data residing further on in a hugetlb page.

This patch addresses this issue by taking into consideration the
order of the page when flushing the dcache.

Reported-by: Mark Brown <broonie@linaro.org>
Tested-by: Mark Brown <broonie@linaro.org>
Signed-off-by: Steve Capper <steve.capper@linaro.org>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Cc: <stable@vger.kernel.org> # v3.11+
2014-07-04 14:26:01 +01:00
Catalin Marinas
2d5a5612bc arm64: Limit the CMA buffer to 32-bit if ZONE_DMA
When the CMA buffer is allocated, it is too early to know whether
devices will require ZONE_DMA memory. This patch limits the CMA buffer
to (DMA_BIT_MASK(32) + 1) if CONFIG_ZONE_DMA is enabled.

In addition, it computes the dma_to_phys(DMA_BIT_MASK(32)) before the
increment (no current functional change).

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-06-18 12:40:56 +01:00
Linus Torvalds
cc07aabc53 - Optimised assembly string/memory routines (based on the AArch64 Cortex
Strings library contributed to glibc but re-licensed under GPLv2)
 - Optimised crypto algorithms making use of the ARMv8 crypto extensions
   (together with kernel API for using FPSIMD instructions in interrupt
   context)
 - Ftrace support
 - CPU topology parsing from DT
 - ESR_EL1 (Exception Syndrome Register) exposed to user space signal
   handlers for SIGSEGV/SIGBUS (useful to emulation tools like Qemu)
 - 1GB section linear mapping if applicable
 - Barriers usage clean-up
 - Default pgprot clean-up
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1.4.9 (GNU/Linux)
 
 iQIcBAABAgAGBQJTkb+CAAoJEGvWsS0AyF7xLyEQAJgL8s2SdDyd+R8aukNDu3n9
 tCK7yVHO9Kg96dfeXVuSOVEo2jszo6R3nxzUL05FMovr230WBcmoeHvHz8ETGnw1
 g0yO8Ltkckjevog4UleCa3wGtYISjvwwrTalzbqoEWzsF2AV8oiqv/yuIn/EdkUr
 jaOqfNsnAQa8TIz4vMhi/AVdJWTTU/F6WP80oqCbxqXu/WL2InuBlHtOJMbk1HDI
 u1DJUGDQ1B9OgSVRkAOjCjSsEtz8sDY3lXsg3V1qT5+NbZTyomYM2IiBLdgQcX4P
 t/rqX9nX4VmRQtzefeP5WhKFks2x80C0BKibWC4teeL++tJHbgbFkyjoZZGcP27o
 zued3cYABrjrcAEU6ko/LUiL2Q4ozBOzosClpjpWulCxNPzsOps82UZWo3F3XbAt
 xjE3k7WF9WeNBOJdDGrarEaSLdnjjgCLoWVs8cOUYLpOOrtdSw16D29jJ68U0Y5g
 31wdwKxoueC8SFt8M9fP9J9Jyau08g+kvW1xQXrRmroppweFxjSpSy90imARyux/
 wUFz79HxkQB79ZHpJ0I5TNrw/w+7pBnfVSKGPOzrk+ZUsaH76caNRBoffUCzFMzz
 T3Sc8A36TZtOIcGR/Q4DMZNFXlIUXDSzCHP2Iu0QoIjTd5Ex96cqNvy3nswCYWwv
 yGe3ZEqUq9+WL7snNW4v
 =Jj8U
 -----END PGP SIGNATURE-----

Merge tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux into next

Pull arm64 updates from Catalin Marinas:
 - Optimised assembly string/memory routines (based on the AArch64
   Cortex Strings library contributed to glibc but re-licensed under
   GPLv2)
 - Optimised crypto algorithms making use of the ARMv8 crypto extensions
   (together with kernel API for using FPSIMD instructions in interrupt
   context)
 - Ftrace support
 - CPU topology parsing from DT
 - ESR_EL1 (Exception Syndrome Register) exposed to user space signal
   handlers for SIGSEGV/SIGBUS (useful to emulation tools like Qemu)
 - 1GB section linear mapping if applicable
 - Barriers usage clean-up
 - Default pgprot clean-up

Conflicts as per Catalin.

* tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (57 commits)
  arm64: kernel: initialize broadcast hrtimer based clock event device
  arm64: ftrace: Add system call tracepoint
  arm64: ftrace: Add CALLER_ADDRx macros
  arm64: ftrace: Add dynamic ftrace support
  arm64: Add ftrace support
  ftrace: Add arm64 support to recordmcount
  arm64: Add 'notrace' attribute to unwind_frame() for ftrace
  arm64: add __ASSEMBLY__ in asm/insn.h
  arm64: Fix linker script entry point
  arm64: lib: Implement optimized string length routines
  arm64: lib: Implement optimized string compare routines
  arm64: lib: Implement optimized memcmp routine
  arm64: lib: Implement optimized memset routine
  arm64: lib: Implement optimized memmove routine
  arm64: lib: Implement optimized memcpy routine
  arm64: defconfig: enable a few more common/useful options in defconfig
  ftrace: Make CALLER_ADDRx macros more generic
  arm64: Fix deadlock scenario with smp_send_stop()
  arm64: Fix machine_shutdown() definition
  arm64: Support arch_irq_work_raise() via self IPIs
  ...
2014-06-06 10:43:28 -07:00
Linus Torvalds
c3c55a0720 Merge branch 'arm64-efi-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip into next
Pull ARM64 EFI update from Peter Anvin:
 "By agreement with the ARM64 EFI maintainers, we have agreed to make
  -tip the upstream for all EFI patches.  That is why this patchset
  comes from me :)

  This patchset enables EFI stub support for ARM64, like we already have
  on x86"

* 'arm64-efi-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  arm64: efi: only attempt efi map setup if booting via EFI
  efi/arm64: ignore dtb= when UEFI SecureBoot is enabled
  doc: arm64: add description of EFI stub support
  arm64: efi: add EFI stub
  doc: arm: add UEFI support documentation
  arm64: add EFI runtime services
  efi: Add shared FDT related functions for ARM/ARM64
  arm64: Add function to create identity mappings
  efi: add helper function to get UEFI params from FDT
  doc: efi-stub.txt updates for ARM
  lib: add fdt_empty_tree.c
2014-06-05 13:15:32 -07:00
Naoya Horiguchi
c177c81e09 hugetlb: restrict hugepage_migration_support() to x86_64
Currently hugepage migration is available for all archs which support
pmd-level hugepage, but testing is done only for x86_64 and there're
bugs for other archs.  So to avoid breaking such archs, this patch
limits the availability strictly to x86_64 until developers of other
archs get interested in enabling this feature.

Simply disabling hugepage migration on non-x86_64 archs is not enough to
fix the reported problem where sys_move_pages() hits the BUG_ON() in
follow_page(FOLL_GET), so let's fix this by checking if hugepage
migration is supported in vma_migratable().

Signed-off-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Reported-by: Michael Ellerman <mpe@ellerman.id.au>
Tested-by: Michael Ellerman <mpe@ellerman.id.au>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: James Hogan <james.hogan@imgtec.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: David Miller <davem@davemloft.net>
Cc: <stable@vger.kernel.org>	[3.12+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-06-04 16:53:51 -07:00
Linus Torvalds
d27050641e DeviceTree for 3.16:
- Another round of clean-up of FDT related code in architecture code.
   This removes knowledge of internal FDT details from most architectures
   except powerpc.
 - Conversion of kernel's custom FDT parsing code to use libfdt.
 - DT based initialization for generic serial earlycon. The introduction
   of generic serial earlycon support went in thru tty tree.
 - Improve the platform device naming for DT probed devices to ensure
   unique naming and use parent names instead of a global index.
 - Fix a race condition in of_update_property.
 - Unify the various linker section OF match tables and fix several
   function prototype errors.
 - Update platform_get_irq_byname to work in deferred probe cases.
 - 2 binding doc updates
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1
 
 iQEcBAABAgAGBQJTjzgyAAoJEMhvYp4jgsXiFsUH/1PMTGo8CyD62VQD5ZKdAoW+
 Fq6vCiRQ8assF5i5ZLcW1DqhjtoRaCKYhVbRKa5lj7cZdjlSpacI/qQPrF5Br2Ii
 bTE3Ff/AQwipQaz/Bj7HqJCgGwfWK8xdfgW0abKsyXMWDN86Bov/zzeu8apmws0x
 H1XjJRgnc/rzM4m9ny6+lss0iq6YL54SuTYNzHR33+Ywxls69SfHXIhCW0KpZcBl
 5U3YUOomt40GfO46sxFA4xApAhypEK4oVq7asyiA2ArTZ/c2Pkc9p5CBqzhDLmlq
 yioWTwHIISv0q+yMLCuQrVGIsbUDkQyy7RQ15z6U+/e/iGO/M+j3A5yxMc3qOi4=
 =Onff
 -----END PGP SIGNATURE-----

Merge tag 'devicetree-for-3.16' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux into next

Pull DeviceTree updates from Rob Herring:
 - Another round of clean-up of FDT related code in architecture code.
   This removes knowledge of internal FDT details from most
   architectures except powerpc.
 - Conversion of kernel's custom FDT parsing code to use libfdt.
 - DT based initialization for generic serial earlycon.  The
   introduction of generic serial earlycon support went in through the
   tty tree.
 - Improve the platform device naming for DT probed devices to ensure
   unique naming and use parent names instead of a global index.
 - Fix a race condition in of_update_property.
 - Unify the various linker section OF match tables and fix several
   function prototype errors.
 - Update platform_get_irq_byname to work in deferred probe cases.
 - 2 binding doc updates

* tag 'devicetree-for-3.16' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux: (58 commits)
  of: handle NULL node in next_child iterators
  of/irq: provide more wrappers for !CONFIG_OF
  devicetree: bindings: Document micrel vendor prefix
  dt: bindings: dwc2: fix required value for the phy-names property
  of_pci_irq: kill useless variable in of_irq_parse_pci()
  of/irq: do irq resolution in platform_get_irq_byname()
  of: Add a testcase for of_find_node_by_path()
  of: Make of_find_node_by_path() handle /aliases
  of: Create unlocked version of for_each_child_of_node()
  lib: add glibc style strchrnul() variant
  of: Handle memory@0 node on PPC32 only
  pci/of: Remove dead code
  of: fix race between search and remove in of_update_property()
  of: Use NULL for pointers
  of: Stop naming platform_device using dcr address
  of: Ensure unique names without sacrificing determinism
  tty/serial: pl011: add DT based earlycon support
  of/fdt: add FDT serial scanning for earlycon
  of/fdt: add FDT address translation support
  serial: earlycon: add DT support
  ...
2014-06-04 10:02:38 -07:00
Mark Salter
4797ec2dc8 arm64: fix pud_huge() for 2-level pagetables
The following happens when trying to run a kvm guest on a kernel
configured for 64k pages. This doesn't happen with 4k pages:

  BUG: failure at include/linux/mm.h:297/put_page_testzero()!
  Kernel panic - not syncing: BUG!
  CPU: 2 PID: 4228 Comm: qemu-system-aar Tainted: GF            3.13.0-0.rc7.31.sa2.k32v1.aarch64.debug #1
  Call trace:
  [<fffffe0000096034>] dump_backtrace+0x0/0x16c
  [<fffffe00000961b4>] show_stack+0x14/0x1c
  [<fffffe000066e648>] dump_stack+0x84/0xb0
  [<fffffe0000668678>] panic+0xf4/0x220
  [<fffffe000018ec78>] free_reserved_area+0x0/0x110
  [<fffffe000018edd8>] free_pages+0x50/0x88
  [<fffffe00000a759c>] kvm_free_stage2_pgd+0x30/0x40
  [<fffffe00000a5354>] kvm_arch_destroy_vm+0x18/0x44
  [<fffffe00000a1854>] kvm_put_kvm+0xf0/0x184
  [<fffffe00000a1938>] kvm_vm_release+0x10/0x1c
  [<fffffe00001edc1c>] __fput+0xb0/0x288
  [<fffffe00001ede4c>] ____fput+0xc/0x14
  [<fffffe00000d5a2c>] task_work_run+0xa8/0x11c
  [<fffffe0000095c14>] do_notify_resume+0x54/0x58

In arch/arm/kvm/mmu.c:unmap_range(), we end up doing an extra put_page()
on the stage2 pgd which leads to the BUG in put_page_testzero(). This
happens because a pud_huge() test in unmap_range() returns true when it
should always be false with 2-level pages tables used by 64k pages.
This patch removes support for huge puds if 2-level pagetables are
being used.

Signed-off-by: Mark Salter <msalter@redhat.com>
[catalin.marinas@arm.com: removed #ifndef around PUD_SIZE check]
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Cc: <stable@vger.kernel.org> # v3.11+
2014-05-16 17:34:40 +01:00
Catalin Marinas
5a0fdfada3 Revert "arm64: Introduce execute-only page access permissions"
This reverts commit bc07c2c6e9.

While the aim is increased security for --x memory maps, it does not
protect against kernel level reads. Until SECCOMP is implemented for
arm64, revert this patch to avoid giving a false idea of execute-only
mappings.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-05-16 16:44:32 +01:00
Rob Herring
eafd370dfe Merge branch 'dt-bus-name' into for-next 2014-05-13 18:34:35 -05:00
Will Deacon
dc60b777fc arm64: mm: use inner-shareable barriers for inner-shareable maintenance
In order to ensure ordering and completion of inner-shareable maintenance
instructions (cache and TLB) on AArch64, we can use the -ish suffix to
the dmb and dsb instructions respectively.

This patch updates our low-level cache and tlb maintenance routines to
use the inner-shareable barrier variants where appropriate.

Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-05-09 17:21:24 +01:00
Steve Capper
fa48e6f780 arm64: mm: Optimise tlb flush logic where we have >4K granule
The tlb maintainence functions: __cpu_flush_user_tlb_range and
__cpu_flush_kern_tlb_range do not take into consideration the page
granule when looping through the address range, and repeatedly flush
tlb entries for the same page when operating with 64K pages.

This patch re-works the logic s.t. we instead advance the loop by
 1 << (PAGE_SHIFT - 12), so avoid repeating ourselves.

Also the routines have been converted from assembler to static inline
functions to aid with legibility and potential compiler optimisations.

The isb() has been removed from flush_tlb_kernel_range(.) as it is
only needed when changing the execute permission of a mapping. If one
needs to set an area of the kernel as execute/non-execute an isb()
must be inserted after the call to flush_tlb_kernel_range.

Cc: Laura Abbott <lauraa@codeaurora.org>
Signed-off-by: Steve Capper <steve.capper@linaro.org>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-05-09 17:00:48 +01:00
Steve Capper
206a2a73a6 arm64: mm: Create gigabyte kernel logical mappings where possible
We have the capability to map 1GB level 1 blocks when using a 4K
granule.

This patch adjusts the create_mapping logic s.t. when mapping physical
memory on boot, we attempt to use a 1GB block if both the VA and PA
start and end are 1GB aligned. This both reduces the levels of lookup
required to resolve a kernel logical address, as well as reduces TLB
pressure on cores that support 1GB TLB entries.

Signed-off-by: Steve Capper <steve.capper@linaro.org>
Tested-by: Jungseok Lee <jays.lee@samsung.com>
[catalin.marinas@arm.com: s/prot_sect_kernel/PROT_SECT_NORMAL_EXEC/]
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-05-09 16:10:58 +01:00
Catalin Marinas
a501e32430 arm64: Clean up the default pgprot setting
The primary aim of this patchset is to remove the pgprot_default and
prot_sect_default global variables and rely strictly on predefined
values. The original goal was to be able to run SMP kernels on UP
hardware by not setting the Shareability bit. However, it is unlikely to
see UP ARMv8 hardware and even if we do, the Shareability bit is no
longer assumed to disable cacheable accesses.

A side effect is that the device mappings now have the Shareability
attribute set. The hardware, however, should ignore it since Device
accesses are always Outer Shareable.

Following the removal of the two global variables, there is some PROT_*
macro reshuffling and cleanup, including the __PAGE_* macros (replaced
by PAGE_*).

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Will Deacon <will.deacon@arm.com>
2014-05-09 15:53:37 +01:00
Catalin Marinas
bc07c2c6e9 arm64: Introduce execute-only page access permissions
The ARMv8 architecture allows execute-only user permissions by clearing
the PTE_UXN and PTE_USER bits. The kernel, however, can still access
such page, so execute-only page permission does not protect against
read(2)/write(2) etc. accesses. Systems requiring such protection must
implement/enable features like SECCOMP.

This patch changes the arm64 __P100 and __S100 protection_map[] macros
to the new __PAGE_EXECONLY attributes. A side effect is that
pte_valid_user() no longer triggers for __PAGE_EXECONLY since PTE_USER
isn't set. To work around this, the check is done on the PTE_NG bit via
the pte_valid_ng() macro. VM_READ is also checked now for page faults.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-05-09 15:53:36 +01:00
Catalin Marinas
9141300a58 arm64: Provide read/write fault information in compat signal handlers
For AArch32, bit 11 (WnR) of the FSR/ESR register is set when the fault
was caused by a write access and applications like Qemu rely on such
information being provided in sigcontext. This patch introduces the
ESR_EL1 tracking for the arm64 kernel faults and sets bit 11 accordingly
in compat sigcontext.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-05-09 15:47:47 +01:00
Catalin Marinas
6ecba8eb51 arm64: Use bus notifiers to set per-device coherent DMA ops
Recently, the default DMA ops have been changed to non-coherent for
alignment with 32-bit ARM platforms (and DT files). This patch adds bus
notifiers to be able to set the coherent DMA ops (with no cache
maintenance) for devices explicitly marked as coherent via the
"dma-coherent" DT property.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-05-03 22:20:34 +01:00
Ritesh Harjani
c7a4a7658d arm64: Make default dma_ops to be noncoherent
Currently arm64 dma_ops is by default made coherent which makes it
opposite in default policy from arm.

Make default dma_ops to be noncoherent (same as arm), as currently there
aren't any dma-capable drivers which assumes coherent ops

Signed-off-by: Ritesh Harjani <ritesh.harjani@gmail.com>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-05-03 22:20:33 +01:00
Dave Anderson
da6e4cb67c arm64: Fix for the arm64 kern_addr_valid() function
Fix for the arm64 kern_addr_valid() function to recognize
virtual addresses in the kernel logical memory map.  The
function fails as written because it does not check whether
the addresses in that region are mapped at the pmd level to
2MB or 512MB pages, continues the page table walk to the
pte level, and issues a garbage value to pfn_valid().

Tested on 4K-page and 64K-page kernels.

Signed-off-by: Dave Anderson <anderson@redhat.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-05-03 22:20:29 +01:00
Mark Salter
d7ecbddf4c arm64: Add function to create identity mappings
At boot time, before switching to a virtual UEFI memory map, firmware
expects UEFI memory and IO regions to be identity mapped whenever
kernel makes runtime services calls. The existing early boot code
creates an identity map of kernel text/data but this is not sufficient
for UEFI. This patch adds a create_id_mapping() function which reuses
the core code of the existing create_mapping().

Signed-off-by: Mark Salter <msalter@redhat.com>
[ Fixed error message formatting (%pa). ]
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
2014-04-30 19:49:52 +01:00
Rob Herring
d1552ce449 of/fdt: move memreserve and dtb memory reservations into core
Move the /memreserve/ processing and dtb memory reservations into
early_init_fdt_scan_reserved_mem. This converts arm, arm64, and powerpc
as they are the only users of early_init_fdt_scan_reserved_mem.

memblock_reserve is safe to call on the same region twice, so the
reservation check for the dtb in powerpc 32-bit reservations is safe to
remove.

Signed-off-by: Rob Herring <robh@kernel.org>
Tested-by: Michal Simek <michal.simek@xilinx.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Tested-by: Grant Likely <grant.likely@linaro.org>
Tested-by: Stephen Chivers <schivers@csc.com>
2014-04-30 00:59:17 -05:00
Linus Torvalds
e4f30545a2 - Documentation clarification on CPU topology and booting requirements
- Additional cache flushing during boot (needed in the presence of
   external caches or under virtualisation)
 - DMA range invalidation fix for non cache line aligned buffers
 - Build failure fix with !COMPAT
 - Kconfig update for STRICT_DEVMEM
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1.4.9 (GNU/Linux)
 
 iQIcBAABAgAGBQJTRDI8AAoJEGvWsS0AyF7xx2IP/jgBjjeIII7L6T45dG3BeJR1
 ph8CBfnsHK3wr5KXiFrdQ2mhAWFr44SLNOd9xuZZmzpA/QWlyqaH46+oqz2GozUf
 y2HucJoEIO+wXMFscQ/EQDiT4uUSbEXBQ6JeZzrhCgVaXeSs+wkKTtGXkxEh2gWo
 w4OI1/JX7phv4heim51aabzziQ3o9JziIs6hALv6OVZVsuPF/TX+mK8C2ejJWLnv
 ou+6E6iv69wNrgPnM3fcKj1CDisCNdFVvjd2LwzkJS7MUra74SWoXczCbfBYW6Ty
 1GgZ/t3TOluDoaLgXfGyQXxnhOUyHdV16034/k8wISfuXG59V0eT+CaCgAotftKD
 5oH+P4MfyVOvZpILrRtY/4MajlCr4V1RnzSYKnS/h3zzHW7Cx/BtYbbQEOVQnZwc
 Gh4adLqc0f8QtkD4zGI7UWmxPxiI9KX9EEpVDAU3TJw6FjVSp7qZ1ifajsWc201h
 STzQEu8LDBWQY2WKrtZxXvFjZj4eSSXNaDHNVCugODk2FBU6wNv4P5q1S+23xt+G
 rR9UI8a0mpginNvhnwHoIR6X+RW3CDaUzn9k3gaJWDWvGoTeuAIstdyxCn6Dm26c
 XF2B5xf5SdxXeNv513WfULCyLgJmLs9CzVFybgYWKLciLbLn7f3pAiisHuO1DWvu
 Gv+70wfYArAcbkGPw5Qt
 =mDkd
 -----END PGP SIGNATURE-----

Merge tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux

Pull second set of arm64 updates from Catalin Marinas:
 "A second pull request for this merging window, mainly with fixes and
  docs clarification:

   - Documentation clarification on CPU topology and booting
     requirements
   - Additional cache flushing during boot (needed in the presence of
     external caches or under virtualisation)
   - DMA range invalidation fix for non cache line aligned buffers
   - Build failure fix with !COMPAT
   - Kconfig update for STRICT_DEVMEM"

* tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
  arm64: Fix DMA range invalidation for cache line unaligned buffers
  arm64: Add missing Kconfig for CONFIG_STRICT_DEVMEM
  arm64: fix !CONFIG_COMPAT build failures
  Revert "arm64: virt: ensure visibility of __boot_cpu_mode"
  arm64: Relax the kernel cache requirements for boot
  arm64: Update the TCR_EL1 translation granule definitions for 16K pages
  ARM: topology: Make it clear that all CPUs need to be described
2014-04-08 12:06:03 -07:00
Catalin Marinas
ebf81a938d arm64: Fix DMA range invalidation for cache line unaligned buffers
If the buffer needing cache invalidation for inbound DMA does start or
end on a cache line aligned address, we need to use the non-destructive
clean&invalidate operation. This issue was introduced by commit
7363590d2c (arm64: Implement coherent DMA API based on swiotlb).

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Reported-by: Jon Medhurst (Tixy) <tixy@linaro.org>
2014-04-08 11:45:08 +01:00
Mark Salter
bf4b558eba arm64: add early_ioremap support
Add support for early IO or memory mappings which are needed before the
normal ioremap() is usable.  This also adds fixmap support for permanent
fixed mappings such as that used by the earlyprintk device register
region.

Signed-off-by: Mark Salter <msalter@redhat.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Cc: Borislav Petkov <borislav.petkov@amd.com>
Cc: Dave Young <dyoung@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-04-07 16:36:15 -07:00
Mark Salter
0bf757c73d arm64: initialize pgprot info earlier in boot
Presently, paging_init() calls init_mem_pgprot() to initialize pgprot
values used by macros such as PAGE_KERNEL, PAGE_KERNEL_EXEC, etc.

The new fixmap and early_ioremap support also needs to use these macros
before paging_init() is called.  This patch moves the init_mem_pgprot()
call out of paging_init() and into setup_arch() so that pgprot_default
gets initialized in time for fixmap and early_ioremap.

Signed-off-by: Mark Salter <msalter@redhat.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Borislav Petkov <borislav.petkov@amd.com>
Cc: Dave Young <dyoung@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-04-07 16:36:15 -07:00
Catalin Marinas
c218bca74e arm64: Relax the kernel cache requirements for boot
With system caches for the host OS or architected caches for guest OS we
cannot easily guarantee that there are no dirty or stale cache lines for
the areas of memory written by the kernel during boot with the MMU off
(therefore non-cacheable accesses).

This patch adds the necessary cache maintenance during boot and relaxes
the booting requirements.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-04-05 10:06:18 +01:00
Catalin Marinas
35a8697692 arm64: Update the TCR_EL1 translation granule definitions for 16K pages
The current TCR register setting in arch/arm64/mm/proc.S assumes that
TCR_EL1.TG* fields are one bit wide and bit 31 is RES1 (reserved, set to
1). With the addition of 16K pages (currently unsupported in the
kernel), the TCR_EL1.TG* fields have been extended to two bits. This
patch updates the corresponding Linux definitions and drops the bit 31
setting in proc.S in favour of the new macros.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Reported-by: Joe Sylve <joe.sylve@gmail.com>
2014-04-03 10:43:11 +01:00
Linus Torvalds
b9f2b21a32 Devicetree changes for v3.15
Updates to devicetree core code. This branch contains the following notable changes:
 * Add reserved memory binding
 * Make struct device_node a kobject and remove legacy /proc/device-tree
 * ePAPR conformance fixes
 * Update in-kernel DTC copy to version v1.4.0
 * Preparation changes for dynamic device tree overlays
 * minor bug fixes and documentation changes
 
 The most significant change in this branch is the conversion of struct
 device_node to be a kobject that is exposed via sysfs and removal of the
 old /proc/device-tree code. This simplifies the device tree handling
 code and tightens up the lifecycle on device tree nodes.
 
 [updated: added fix for dangling select PROC_DEVICETREE]
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1.4.14 (GNU/Linux)
 
 iQIcBAABAgAGBQJTOyNwAAoJEMWQL496c2LNZY0QAIreUrpo3/hKRau61EDPXkOA
 UFRyPUHD0k/dNXWWDbTfvKH/nAfzdVwejhePqEWiODiFOFkq7JyQlMKPA+CZuZj0
 ygN4215A1yj/hDf6JRD5Zn4WGpawDt9InlbZSps6P5dd8voV5t5dz6uzz+Y7uqaK
 CAjTDlBSmxEen5vRHiHQgKv74au/+b9yfSURjPQVWg46+wl3WJwjsdzerphm4unW
 tpEr8zkIsm51mqqAx4penIuiovh7+L2J5v4BFeg8o+kaZEuZpVxLHJPOuBd5hdom
 zeqEIj3AqHTh5suYIHe4aAbZ2wMP3kYGgkPGwfWLnwLyULxalcCtGZeaCi9nwTFj
 Fdj+7f17ocrt5mif0f5Deufi1LqJsDjhY6G9p7HuV7Y9hsMILpJIUoGENPji+TWj
 BA4L45eaPmNYdKJytEtFD7F2WnXeHZ6fDtYho/39DWW+Bt16IFX85T199irhxGG4
 byN6LRaahk2UeycSXkQHAlWOQHqzBcJJAkQLN2iahzyYRr9Dy+VI2E9clm53m49O
 YQYcONdUlMYrtfRwJpbB9XHM0HgZUvg0LT5z/iHQs9uJtoo33Oj+zxFixyZLQ9Dq
 qyLqQWEpV9gFLAo9tpf56gffkLiJRsHkX4UJ6oTtj4DY1WWU9H81jjCvv/7flzp/
 8ZyyZzANQf1DZ9kqO2v+
 =lyA5
 -----END PGP SIGNATURE-----

Merge tag 'dt-for-linus' of git://git.secretlab.ca/git/linux

Pull devicetree changes from Grant Likely:
 "Updates to devicetree core code.  This branch contains the following
  notable changes:

   - add reserved memory binding
   - make struct device_node a kobject and remove legacy
     /proc/device-tree
   - ePAPR conformance fixes
   - update in-kernel DTC copy to version v1.4.0
   - preparatory changes for dynamic device tree overlays
   - minor bug fixes and documentation changes

  The most significant change in this branch is the conversion of struct
  device_node to be a kobject that is exposed via sysfs and removal of
  the old /proc/device-tree code.  This simplifies the device tree
  handling code and tightens up the lifecycle on device tree nodes.

  [updated: added fix for dangling select PROC_DEVICETREE]"

* tag 'dt-for-linus' of git://git.secretlab.ca/git/linux: (29 commits)
  dt: Remove dangling "select PROC_DEVICETREE"
  of: Add support for ePAPR "stdout-path" property
  of: device_node kobject lifecycle fixes
  of: only scan for reserved mem when fdt present
  powerpc: add support for reserved memory defined by device tree
  arm64: add support for reserved memory defined by device tree
  of: add missing major vendors
  of: add vendor prefix for SMSC
  of: remove /proc/device-tree
  of/selftest: Add self tests for manipulation of properties
  of: Make device nodes kobjects so they show up in sysfs
  arm: add support for reserved memory defined by device tree
  drivers: of: add support for custom reserved memory drivers
  drivers: of: add initialization code for dynamic reserved memory
  drivers: of: add initialization code for static reserved memory
  of: document bindings for reserved-memory nodes
  Revert "of: fix of_update_property()"
  kbuild: dtbs_install: new make target
  ARM: mvebu: Allows to get the SoC ID even without PCI enabled
  of: Allows to use the PCI translator without the PCI core
  ...
2014-04-02 14:27:15 -07:00
Catalin Marinas
196adf2f30 arm64: Remove pgprot_dmacoherent()
Since this macro is identical to pgprot_writecombine() and is only used
in a single place, remove it completely to avoid confusion. On ARMv7+
processors, the coherent DMA mapping must be Normal NonCacheable (a.k.a.
writecombine) to avoid mismatched hardware attribute aliases (with the
kernel linear mapping as Normal Cacheable).

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-03-24 10:35:35 +00:00
Laura Abbott
214fdbe74a arm64: Support DMA_ATTR_WRITE_COMBINE
DMA_ATTR_WRITE_COMBINE is currently ignored. Set the pgprot
appropriately for non coherent opperations.

Signed-off-by: Laura Abbott <lauraa@codeaurora.org>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-03-24 10:26:10 +00:00
Laura Abbott
6e8d7968e9 arm64: Implement custom mmap functions for dma mapping
The current dma_ops do not specify an mmap function so maping
falls back to the default implementation. There are at least
two issues with using the default implementation:

1) The pgprot is always pgprot_noncached (strongly ordered)
memory even with coherent operations
2) dma_common_mmap calls virt_to_page on the remapped non-coherent
address which leads to invalid memory being mapped.

Fix both these issue by implementing a custom mmap function which
correctly accounts for remapped addresses and sets vm_pg_prot
appropriately.

Signed-off-by: Laura Abbott <lauraa@codeaurora.org>
[catalin.marinas@arm.com: replaced "arm64_" with "__" prefix for consistency]
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-03-24 10:01:17 +00:00
Radha Mohan Chintakuntla
87366d8cf7 arm64: Add boot time configuration of Intermediate Physical Address size
ARMv8 supports a range of physical address bit sizes. The PARange bits
from ID_AA64MMFR0_EL1 register are read during boot-time and the
intermediate physical address size bits are written in the translation
control registers (TCR_EL1 and VTCR_EL2).

There is no change in the VA bits and levels of translation.

Signed-off-by: Radha Mohan Chintakuntla <rchintakuntla@cavium.com>
Reviewed-by: Will Deacon <Will.deacon@arm.com>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-03-13 11:22:36 +00:00
Marek Szyprowski
9bf14b7c54 arm64: add support for reserved memory defined by device tree
Enable reserved memory initialization from device tree.

Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Grant Likely <grant.likely@linaro.org>
2014-03-13 10:53:07 +00:00
Mark Rutland
bff705950e arm64: remove unnecessary cache flush at boot
Currently we flush the entire dcache at boot within __cpu_setup, but
this is unnecessary as the booting protocol demands that the dcache is
invalid and off upon entering the kernel. The presence of the cache
flush only serves to hide bugs in bootloaders, and is not safe in the
presence of SMP.

In an SMP boot scenario the CPUs enter coherency outside of the kernel,
and the primary CPU enables its caches before bringing up secondary
CPUs. Therefore if any secondary CPU has an entry in its cache (in
violation of the boot protocol), the primary CPU might snoop it even if
the secondary CPU's cache is disabled. The boot-time cache flush only
serves to hide a firmware bug, and slows down a cpu boot unnecessarily.

This patch removes the unnecessary boot-time cache flush.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Acked-by: Will Deacon <will.deacon@arm.com>
[catalin.marinas@arm.com: make __flush_dcache_all local only]
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-03-04 01:07:22 +00:00
Catalin Marinas
7363590d2c arm64: Implement coherent DMA API based on swiotlb
This patch adds support for DMA API cache maintenance on SoCs without
hardware device cache coherency.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-02-27 17:16:59 +00:00
Catalin Marinas
3690951fc6 arm64: Use swiotlb late initialisation
Since arm64 does not support ISA, there is no need for early swiotlb
initialisation. This patch switches the DMA mapping code to
swiotlb_tlb_late_init_with_default_size(). A side effect of this is that
GFP_DMA is used for the swiotlb buffer and devices with a 32-bit
coherent mask are correctly supported.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-02-27 14:11:53 +00:00
Catalin Marinas
19e7640d1f arm64: Replace ZONE_DMA32 with ZONE_DMA
On arm64 we do not have two DMA zones, so it does not make sense to
implement ZONE_DMA32. This patch changes ZONE_DMA32 with ZONE_DMA, the
latter covering 32-bit dma address space to honour GFP_DMA allocations.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-02-27 12:09:22 +00:00
Ritesh Harjani
bb10eb7b4d arm64: Change misleading function names in dma-mapping
arm64_swiotlb_alloc/free_coherent name can be misleading
somtimes with CMA support being enabled after this
patch (c2104debc235b745265b64d610237a6833fd53)

Change this name to be more generic:
__dma_alloc/free_coherent

Signed-off-by: Ritesh Harjani <ritesh.harjani@gmail.com>
[catalin.marinas@arm.com: renamed arm64_swiotlb_dma_ops to coherent_swiotlb_dma_ops]
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-02-26 11:16:28 +00:00
Mark Rutland
883d50a0ed arm64: simplify pgd_alloc
Currently pgd_alloc has a redundant NULL check in its return path that
can be removed with no ill effects. With that removed it's also possible
to return early and eliminate the new_pgd temporary variable.

This patch applies said modifications, making the logic of pgd_alloc
correspond 1-1 with that of pgd_free.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-02-05 10:45:07 +00:00
Catalin Marinas
a55f9929a9 arm64: Invalidate the TLB when replacing pmd entries during boot
With the 64K page size configuration, __create_page_tables in head.S
maps enough memory to get started but using 64K pages rather than 512M
sections with a single pgd/pud/pmd entry pointing to a pte table.
create_mapping() may override the pgd/pud/pmd table entry with a block
(section) one if the RAM size is more than 512MB and aligned correctly.
For the end of this block to be accessible, the old TLB entry must be
invalidated.

Cc: <stable@vger.kernel.org>
Reported-by: Mark Salter <msalter@redhat.com>
Tested-by: Mark Salter <msalter@redhat.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-02-05 10:30:51 +00:00
Laura Abbott
ccc9e244eb arm64: Align CMA sizes to PAGE_SIZE
dma_alloc_from_contiguous takes number of pages for a size.
Align up the dma size passed in to page size to avoid truncation
and allocation failures on sizes less than PAGE_SIZE.

Cc: Will Deacon <will.deacon@arm.com>
Signed-off-by: Laura Abbott <lauraa@codeaurora.org>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-02-05 10:28:27 +00:00
Jingoo Han
812944e91d arm64: mm: fix the function name in comment of cpu_do_switch_mm
Fix the function name of comment of cpu_do_switch_mm,
because cpu_do_switch_mm is the correct name.

Signed-off-by: Jingoo Han <jg1.han@samsung.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2014-01-27 15:50:38 +00:00