linux/arch/arm64/mm
Will Deacon f1304f9763 arm64: mm: Don't invalidate FROM_DEVICE buffers at start of DMA transfer
commit c50f11c619 upstream.

Invalidating the buffer memory in arch_sync_dma_for_device() for
FROM_DEVICE transfers

When using the streaming DMA API to map a buffer prior to inbound
non-coherent DMA (i.e. DMA_FROM_DEVICE), we invalidate any dirty CPU
cachelines so that they will not be written back during the transfer and
corrupt the buffer contents written by the DMA. This, however, poses two
potential problems:

  (1) If the DMA transfer does not write to every byte in the buffer,
      then the unwritten bytes will contain stale data once the transfer
      has completed.

  (2) If the buffer has a virtual alias in userspace, then stale data
      may be visible via this alias during the period between performing
      the cache invalidation and the DMA writes landing in memory.

Address both of these issues by cleaning (aka writing-back) the dirty
lines in arch_sync_dma_for_device(DMA_FROM_DEVICE) instead of discarding
them using invalidation.

Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Robin Murphy <robin.murphy@arm.com>
Cc: Russell King <linux@armlinux.org.uk>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20220606152150.GA31568@willie-the-truck
Signed-off-by: Will Deacon <will@kernel.org>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Link: https://lore.kernel.org/r/20220610151228.4562-2-will@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-06-25 15:18:40 +02:00
..
cache.S arm64: mm: Don't invalidate FROM_DEVICE buffers at start of DMA transfer 2022-06-25 15:18:40 +02:00
context.c arm64: mm: Use better bitmap_zalloc() 2021-06-01 18:52:05 +01:00
copypage.c arm64: fix types in copy_highpage() 2022-06-09 10:22:54 +02:00
dma-mapping.c iommu/dma: Pass address limit rather than size to iommu_setup_dma_ops() 2021-06-25 15:02:43 +02:00
extable.c arm64/bpf: Remove 128MB limit for BPF JIT programs 2022-01-29 10:58:25 +01:00
fault.c arm64: kasan: mte: remove redundant mte_report_once logic 2021-08-02 18:15:28 +01:00
flush.c arm64: Rename arm64-internal cache maintenance functions 2021-05-25 19:27:49 +01:00
hugetlbpage.c arm64/hugetlb: fix CMA gigantic page order for non-4K PAGE_SIZE 2021-10-11 18:45:19 +01:00
init.c arm64/mm: drop HAVE_ARCH_PFN_VALID 2022-04-27 14:38:50 +02:00
ioremap.c arm[64]/memremap: don't abuse pfn_valid() to ensure presence of linear map 2022-05-18 10:26:56 +02:00
kasan_init.c Merge branches 'for-next/misc', 'for-next/kselftest', 'for-next/xntable', 'for-next/vdso', 'for-next/fiq', 'for-next/epan', 'for-next/kasan-vmalloc', 'for-next/fgt-boot-init', 'for-next/vhe-only' and 'for-next/neon-softirqs-disabled', remote-tracking branch 'arm64/for-next/perf' into for-next/core 2021-04-15 14:00:38 +01:00
Makefile RISC-V Patches for the 5.12 Merge Window 2021-02-26 10:28:35 -08:00
mmap.c arm64: Ensure execute-only permissions are not allowed without EPAN 2022-03-16 14:23:43 +01:00
mmu.c arm64/mm: avoid fixmap race condition when create pud mapping 2022-04-08 14:23:06 +02:00
mteswap.c arm64: mte: reset the page tag in page->flags 2020-12-22 12:55:07 -08:00
pageattr.c set_memory: allow querying whether set_direct_map_*() is actually enabled 2021-07-08 11:48:20 -07:00
pgd.c mm: consolidate pgtable_cache_init() and pgd_cache_init() 2019-09-24 15:54:09 -07:00
physaddr.c arm64: Do not pass tagged addresses to __is_lm_address() 2021-02-02 17:44:47 +00:00
proc.S arm64: kasan: mte: use a constant kernel GCR_EL1 value 2021-08-02 18:14:21 +01:00
ptdump_debugfs.c arm64: Add __init section marker to some functions 2021-04-08 17:45:10 +01:00
ptdump.c arm64/bpf: Remove 128MB limit for BPF JIT programs 2022-01-29 10:58:25 +01:00
trans_pgd.c arm64: trans_pgd: hibernate: idmap the single page that holds the copy page routines 2021-01-27 15:41:12 +00:00