These get allocated and freed millions of times on this kernel tree.
Use a dedicated kmem_cache pool and avoid costly dynamic memory allocations.
Most allocations' size is:
(sizeof(struct dma_buf) + sizeof(struct reservation_object)).
Put those under kmem_cache pool and distinguish them with dmabuf->from_kmem
flag.
Signed-off-by: Park Ju Hyung <qkrwngud825@gmail.com>
[@0ctobot: Adapted for 4.19]
Signed-off-by: Adam W. Willis <return.of.octobot@gmail.com>
There is a sleep-while-atomic reported when the selinux tries to
dump the file path name of a dmabuf file using d_path(). Fix it.
Flow will be like below:
flush_unauthorized_files -->
iterate_fd -->
spin_lock --> Start of the atomic section
match_file -->
file_has_perm -->
avc_has_perm -->
avc_audit -->
slow_avc_audit -->
common_lsm_audit -->
dump_common_audit_data -->
audit_log_d_path -->
d_path -->
dmabuffs_dname -->
mutex_lock--> Sleep while atomic.
Change-Id: Ieed88a9093355d160e1c194d31fe33f1c7dadf60
Signed-off-by: Charan Teja Reddy <charante@codeaurora.org>
Signed-off-by: Rahul Shahare <rshaha@codeaurora.org>
The following race occurs while reading the dma_buf:
P1 P2
dma_buf_release() dmabuffs_dname()[say from /proc/pid/fd/num]
read dmabuf stored in dentry->fsdata
Free the dmabuf object
Start accessing the dmabuf structure
As said above, we access the dmabuf structure after we free the dmabuf
object resulting into the below 'use-after-free' issue.
__mutex_lock_common+0x174/0x1020
mutex_lock_nested+0x40/0x50
dmabuffs_dname+0x48/0xc8
d_path+0x84/0x290
proc_pid_readlink+0xb4/0x1c8
vfs_readlink+0x128/0x130
do_readlinkat+0xc8/0x148
__arm64_sys_readlinkat+0x24/0x38
el0_svc_common+0xa4/0x178
el0_svc_handler+0x6c/0x88
el0_svc+0x8/0xc.
Fixes: bf93f26 ("UPSTREAM: dma-buf: add DMA_BUF_SET_NAME ioctls")
Change-Id: Ie0a833f185687f4cc7ab8189b17fde3516270572
Signed-off-by: Charan Teja Reddy <charante@codeaurora.org>
Signed-off-by: Naitik Bharadiya <bharad@codeaurora.org>
* refs/heads/tmp-2bb70f4:
ANDROID: virtio: virtio_input: Set the amount of multitouch slots in virtio input
ANDROID: dummy_cpufreq: Implement get()
rtlwifi: Fix potential overflow on P2P code
ANDROID: cpufreq: create dummy cpufreq driver
ANDROID: Allow DRM_IOCTL_MODE_*_DUMB for render clients.
ANDROID: sdcardfs: evict dentries on fscrypt key removal
ANDROID: fscrypt: add key removal notifier chain
ANDROID: Move from clang r353983c to r365631c
ANDROID: move up spin_unlock_bh() ahead of remove_proc_entry()
BACKPORT: arm64: tags: Preserve tags for addresses translated via TTBR1
UPSTREAM: arm64: memory: Implement __tag_set() as common function
UPSTREAM: arm64/mm: fix variable 'tag' set but not used
UPSTREAM: arm64: avoid clang warning about self-assignment
ANDROID: refactor build.config files to remove duplication
UPSTREAM: mm: vmalloc: show number of vmalloc pages in /proc/meminfo
BACKPORT: PM/sleep: Expose suspend stats in sysfs
UPSTREAM: power: supply: Init device wakeup after device_add()
UPSTREAM: PM / wakeup: Unexport wakeup_source_sysfs_{add,remove}()
UPSTREAM: PM / wakeup: Register wakeup class kobj after device is added
BACKPORT: PM / wakeup: Fix sysfs registration error path
BACKPORT: PM / wakeup: Show wakeup sources stats in sysfs
UPSTREAM: PM / wakeup: Print warn if device gets enabled as wakeup source during sleep
UPSTREAM: PM / wakeup: Use wakeup_source_register() in wakelock.c
UPSTREAM: PM / wakeup: Only update last time for active wakeup sources
UPSTREAM: PM / core: Add support to skip power management in device/driver model
cuttlefish-4.14: Enable CONFIG_DM_SNAPSHOT
ANDROID: cuttlefish_defconfig: Enable BPF_JIT and BPF_JIT_ALWAYS_ON
UPSTREAM: netfilter: xt_IDLETIMER: fix sysfs callback function type
UPSTREAM: mm: untag user pointers in mmap/munmap/mremap/brk
UPSTREAM: vfio/type1: untag user pointers in vaddr_get_pfn
UPSTREAM: media/v4l2-core: untag user pointers in videobuf_dma_contig_user_get
UPSTREAM: drm/radeon: untag user pointers in radeon_gem_userptr_ioctl
BACKPORT: drm/amdgpu: untag user pointers
UPSTREAM: userfaultfd: untag user pointers
UPSTREAM: fs/namespace: untag user pointers in copy_mount_options
UPSTREAM: mm: untag user pointers in get_vaddr_frames
UPSTREAM: mm: untag user pointers in mm/gup.c
BACKPORT: mm: untag user pointers passed to memory syscalls
BACKPORT: lib: untag user pointers in strn*_user
UPSTREAM: arm64: Fix reference to docs for ARM64_TAGGED_ADDR_ABI
UPSTREAM: selftests, arm64: add kernel headers path for tags_test
BACKPORT: arm64: Relax Documentation/arm64/tagged-pointers.rst
UPSTREAM: arm64: Define Documentation/arm64/tagged-address-abi.rst
UPSTREAM: arm64: Change the tagged_addr sysctl control semantics to only prevent the opt-in
UPSTREAM: arm64: Tighten the PR_{SET, GET}_TAGGED_ADDR_CTRL prctl() unused arguments
UPSTREAM: selftests, arm64: fix uninitialized symbol in tags_test.c
UPSTREAM: arm64: mm: Really fix sparse warning in untagged_addr()
UPSTREAM: selftests, arm64: add a selftest for passing tagged pointers to kernel
BACKPORT: arm64: Introduce prctl() options to control the tagged user addresses ABI
UPSTREAM: thread_info: Add update_thread_flag() helpers
UPSTREAM: arm64: untag user pointers in access_ok and __uaccess_mask_ptr
UPSTREAM: uaccess: add noop untagged_addr definition
BACKPORT: block: annotate refault stalls from IO submission
ext4: add verity flag check for dax
ANDROID: usb: gadget: Fix dependency for f_accessory
ANDROID: sched: fair: balance for single core cluster
UPSTREAM: mm/kasan: fix false positive invalid-free reports with CONFIG_KASAN_SW_TAGS=y
f2fs: add a condition to detect overflow in f2fs_ioc_gc_range()
f2fs: fix to add missing F2FS_IO_ALIGNED() condition
f2fs: fix to fallback to buffered IO in IO aligned mode
f2fs: fix to handle error path correctly in f2fs_map_blocks
f2fs: fix extent corrupotion during directIO in LFS mode
f2fs: check all the data segments against all node ones
f2fs: Add a small clarification to CONFIG_FS_F2FS_FS_SECURITY
f2fs: fix inode rwsem regression
f2fs: fix to avoid accessing uninitialized field of inode page in is_alive()
f2fs: avoid infinite GC loop due to stale atomic files
f2fs: Fix indefinite loop in f2fs_gc()
f2fs: convert inline_data in prior to i_size_write
f2fs: fix error path of f2fs_convert_inline_page()
f2fs: add missing documents of reserve_root/resuid/resgid
f2fs: fix flushing node pages when checkpoint is disabled
f2fs: enhance f2fs_is_checkpoint_ready()'s readability
f2fs: clean up __bio_alloc()'s parameter
f2fs: fix wrong error injection path in inc_valid_block_count()
f2fs: fix to writeout dirty inode during node flush
f2fs: optimize case-insensitive lookups
f2fs: introduce f2fs_match_name() for cleanup
f2fs: Fix indefinite loop in f2fs_gc()
f2fs: allocate memory in batch in build_sit_info()
f2fs: fix to avoid data corruption by forbidding SSR overwrite
f2fs: Fix build error while CONFIG_NLS=m
Revert "f2fs: avoid out-of-range memory access"
f2fs: cleanup the code in build_sit_entries.
f2fs: fix wrong available node count calculation
f2fs: remove duplicate code in f2fs_file_write_iter
f2fs: fix to migrate blocks correctly during defragment
f2fs: use wrapped f2fs_cp_error()
f2fs: fix to use more generic EOPNOTSUPP
f2fs: use wrapped IS_SWAPFILE()
f2fs: Support case-insensitive file name lookups
f2fs: include charset encoding information in the superblock
fs: Reserve flag for casefolding
f2fs: fix to avoid call kvfree under spinlock
fs: f2fs: Remove unnecessary checks of SM_I(sbi) in update_general_status()
f2fs: disallow direct IO in atomic write
f2fs: fix to handle quota_{on,off} correctly
f2fs: fix to detect cp error in f2fs_setxattr()
f2fs: fix to spread f2fs_is_checkpoint_ready()
f2fs: support fiemap() for directory inode
f2fs: fix to avoid discard command leak
f2fs: fix to avoid tagging SBI_QUOTA_NEED_REPAIR incorrectly
f2fs: fix to drop meta/node pages during umount
f2fs: disallow switching io_bits option during remount
f2fs: fix panic of IO alignment feature
f2fs: introduce {page,io}_is_mergeable() for readability
f2fs: fix livelock in swapfile writes
f2fs: add fs-verity support
ext4: update on-disk format documentation for fs-verity
ext4: add fs-verity read support
ext4: add basic fs-verity support
fs-verity: support builtin file signatures
fs-verity: add SHA-512 support
fs-verity: implement FS_IOC_MEASURE_VERITY ioctl
fs-verity: implement FS_IOC_ENABLE_VERITY ioctl
fs-verity: add data verification hooks for ->readpages()
fs-verity: add the hook for file ->setattr()
fs-verity: add the hook for file ->open()
fs-verity: add inode and superblock fields
fs-verity: add Kconfig and the helper functions for hashing
fs: uapi: define verity bit for FS_IOC_GETFLAGS
fs-verity: add UAPI header
fs-verity: add MAINTAINERS file entry
fs-verity: add a documentation file
ext4: fix kernel oops caused by spurious casefold flag
ext4: fix coverity warning on error path of filename setup
ext4: optimize case-insensitive lookups
ext4: fix dcache lookup of !casefolded directories
unicode: update to Unicode 12.1.0 final
unicode: add missing check for an error return from utf8lookup()
ext4: export /sys/fs/ext4/feature/casefold if Unicode support is present
unicode: refactor the rule for regenerating utf8data.h
ext4: Support case-insensitive file name lookups
ext4: include charset encoding information in the superblock
unicode: update unicode database unicode version 12.1.0
unicode: introduce test module for normalized utf8 implementation
unicode: implement higher level API for string handling
unicode: reduce the size of utf8data[]
unicode: introduce code for UTF-8 normalization
unicode: introduce UTF-8 character database
ext4 crypto: fix to check feature status before get policy
fscrypt: document the new ioctls and policy version
ubifs: wire up new fscrypt ioctls
f2fs: wire up new fscrypt ioctls
ext4: wire up new fscrypt ioctls
fscrypt: require that key be added when setting a v2 encryption policy
fscrypt: add FS_IOC_REMOVE_ENCRYPTION_KEY_ALL_USERS ioctl
fscrypt: allow unprivileged users to add/remove keys for v2 policies
fscrypt: v2 encryption policy support
fscrypt: add an HKDF-SHA512 implementation
fscrypt: add FS_IOC_GET_ENCRYPTION_KEY_STATUS ioctl
fscrypt: add FS_IOC_REMOVE_ENCRYPTION_KEY ioctl
fscrypt: add FS_IOC_ADD_ENCRYPTION_KEY ioctl
fscrypt: rename keyinfo.c to keysetup.c
fscrypt: move v1 policy key setup to keysetup_v1.c
fscrypt: refactor key setup code in preparation for v2 policies
fscrypt: rename fscrypt_master_key to fscrypt_direct_key
fscrypt: add ->ci_inode to fscrypt_info
fscrypt: use FSCRYPT_* definitions, not FS_*
fscrypt: use FSCRYPT_ prefix for uapi constants
fs, fscrypt: move uapi definitions to new header <linux/fscrypt.h>
fscrypt: use ENOPKG when crypto API support missing
fscrypt: improve warnings for missing crypto API support
fscrypt: improve warning messages for unsupported encryption contexts
fscrypt: make fscrypt_msg() take inode instead of super_block
fscrypt: clean up base64 encoding/decoding
fscrypt: remove loadable module related code
ANDROID: arm64: bpf: implement arch_bpf_jit_check_func
ANDROID: bpf: validate bpf_func when BPF_JIT is enabled with CFI
UPSTREAM: kcm: use BPF_PROG_RUN
UPSTREAM: psi: get poll_work to run when calling poll syscall next time
UPSTREAM: sched/psi: Do not require setsched permission from the trigger creator
UPSTREAM: sched/psi: Reduce psimon FIFO priority
BACKPORT: arm64: Add support for relocating the kernel with RELR relocations
ANDROID: Log which device failed to suspend in dpm_suspend_start()
ANDROID: Revert "ANDROID: sched: Disallow WALT with CFS bandwidth control"
ANDROID: sched: WALT: Add support for CFS_BANDWIDTH
ANDROID: sched: WALT: Refactor cumulative runnable average fixup
ANDROID: sched: Disallow WALT with CFS bandwidth control
fscrypt: document testing with xfstests
fscrypt: remove selection of CONFIG_CRYPTO_SHA256
fscrypt: remove unnecessary includes of ratelimit.h
fscrypt: don't set policy for a dead directory
fscrypt: decrypt only the needed blocks in __fscrypt_decrypt_bio()
fscrypt: support decrypting multiple filesystem blocks per page
fscrypt: introduce fscrypt_decrypt_block_inplace()
fscrypt: handle blocksize < PAGE_SIZE in fscrypt_zeroout_range()
fscrypt: support encrypting multiple filesystem blocks per page
fscrypt: introduce fscrypt_encrypt_block_inplace()
fscrypt: clean up some BUG_ON()s in block encryption/decryption
fscrypt: rename fscrypt_do_page_crypto() to fscrypt_crypt_block()
fscrypt: remove the "write" part of struct fscrypt_ctx
fscrypt: simplify bounce page handling
ANDROID: fiq_debugger: remove
UPSTREAM: lib/test_meminit.c: use GFP_ATOMIC in RCU critical section
UPSTREAM: mm: slub: Fix slab walking for init_on_free
UPSTREAM: lib/test_meminit.c: minor test fixes
UPSTREAM: lib/test_meminit.c: fix -Wmaybe-uninitialized false positive
UPSTREAM: lib: introduce test_meminit module
UPSTREAM: mm: init: report memory auto-initialization features at boot time
BACKPORT: mm: security: introduce init_on_alloc=1 and init_on_free=1 boot options
UPSTREAM: arm64: move jump_label_init() before parse_early_param()
ANDROID: Add a tracepoint for mapping inode to full path
BACKPORT: arch: add pidfd and io_uring syscalls everywhere
UPSTREAM: dma-buf: add show_fdinfo handler
UPSTREAM: dma-buf: add DMA_BUF_SET_NAME ioctls
BACKPORT: dma-buf: give each buffer a full-fledged inode
ANDROID: fix kernelci build-break
UPSTREAM: drm/virtio: Fix cache entry creation race.
UPSTREAM: drm/virtio: Wake up all waiters when capset response comes in.
UPSTREAM: drm/virtio: Ensure cached capset entries are valid before copying.
UPSTREAM: drm/virtio: use u64_to_user_ptr macro
UPSTREAM: drm/virtio: remove irrelevant DRM_UNLOCKED flag
UPSTREAM: drm/virtio: Remove redundant return type
UPSTREAM: drm/virtio: allocate fences with GFP_KERNEL
UPSTREAM: drm/virtio: add trace events for commands
UPSTREAM: drm/virtio: trace drm_fence_emit
BACKPORT: drm/virtio: set seqno for dma-fence
BACKPORT: drm/virtio: move drm_connector_update_edid_property() call
UPSTREAM: drm/virtio: add missing drm_atomic_helper_shutdown() call.
BACKPORT: drm/virtio: rework resource creation workflow.
UPSTREAM: drm/virtio: params struct for virtio_gpu_cmd_create_resource_3d()
BACKPORT: drm/virtio: params struct for virtio_gpu_cmd_create_resource()
BACKPORT: drm/virtio: use struct to pass params to virtio_gpu_object_create()
UPSTREAM: drm/virtio: add virtio-gpu-features debugfs file.
UPSTREAM: drm/virtio: remove set but not used variable 'vgdev'
BACKPORT: drm/virtio: implement prime export
UPSTREAM: drm/virtio: remove prime pin/unpin callbacks.
UPSTREAM: drm/virtio: implement prime mmap
UPSTREAM: drm/virtio: drop virtio_gpu_fence_cleanup()
UPSTREAM: drm/virtio: fix pageflip flush
UPSTREAM: drm/virtio: log error responses
UPSTREAM: drm/virtio: Add missing virtqueue reset
UPSTREAM: drm/virtio: Remove incorrect kfree()
UPSTREAM: drm/virtio: virtio_gpu_cmd_resource_create_3d: drop unused fence arg
UPSTREAM: drm/virtio: fence: pass plain pointer
BACKPORT: drm/virtio: add edid support
UPSTREAM: virtio-gpu: add VIRTIO_GPU_F_EDID feature
BACKPORT: drm/virtio: fix memory leak of vfpriv on error return path
UPSTREAM: drm/virtio: bump driver version after explicit synchronization addition
UPSTREAM: drm/virtio: add in/out fence support for explicit synchronization
UPSTREAM: drm/virtio: add uapi for in and out explicit fences
UPSTREAM: drm/virtio: add virtio_gpu_alloc_fence()
UPSTREAM: drm/virtio: Handle error from virtio_gpu_resource_id_get
UPSTREAM: gpu/drm/virtio/virtgpu_vq.c: Use kmem_cache_zalloc
UPSTREAM: drm/virtio: fix resource id handling
UPSTREAM: drm/virtio: drop resource_id argument.
UPSTREAM: drm/virtio: use virtio_gpu_object->hw_res_handle in virtio_gpu_resource_create_ioctl()
UPSTREAM: drm/virtio: use virtio_gpu_object->hw_res_handle in virtio_gpu_mode_dumb_create()
UPSTREAM: drm/virtio: use virtio_gpu_object->hw_res_handle in virtio_gpufb_create()
BACKPORT: drm/virtio: track created object state
UPSTREAM: drm/virtio: document drm_dev_set_unique workaround
UPSTREAM: virtio: Support prime objects vmap/vunmap
UPSTREAM: virtio: Rework virtio_gpu_object_kmap()
UPSTREAM: virtio: Add virtio_gpu_object_kunmap()
UPSTREAM: drm/virtio: pass virtio_gpu_object to virtio_gpu_cmd_transfer_to_host_{2d, 3d}
UPSTREAM: drm/virtio: add dma sync for dma mapped virtio gpu framebuffer pages
UPSTREAM: drm/virtio: Remove set but not used variable 'bo'
UPSTREAM: drm/virtio: add iommu support.
UPSTREAM: drm/virtio: add virtio_gpu_object_detach() function
UPSTREAM: drm/virtio: track virtual output state
UPSTREAM: drm/virtio: fix bounds check in virtio_gpu_cmd_get_capset()
UPSTREAM: gpu: drm: virtio: code cleanup
UPSTREAM: drm/virtio: Place GEM BOs in drm_framebuffer
UPSTREAM: drm/virtio: fix mode_valid's return type
UPSTREAM: drm/virtio: Add spaces around operators
UPSTREAM: drm/virtio: Remove multiple blank lines
UPSTREAM: drm/virtio: Replace 'unsigned' for 'unsigned int'
UPSTREAM: drm/virtio: Remove return from void function
UPSTREAM: drm/virtio: Add */ in block comments to separate line
UPSTREAM: drm/virtio: Add blank line after variable declarations
UPSTREAM: drm/virtio: Add tabs at the start of a line
UPSTREAM: drm/virtio: Don't return invalid caps on timeout
UPSTREAM: virtgpu: remove redundant task_comm copying
UPSTREAM: drm/virtio: add create_handle support.
UPSTREAM: drm: virtio: replace reference/unreference with get/put
UPSTREAM: drm/virtio: Replace instances of reference/unreference with get/put
UPSTREAM: drm: byteorder: add DRM_FORMAT_HOST_*
UPSTREAM: drm: add drm_connector_attach_edid_property()
BACKPORT: drm/prime: Add drm_gem_prime_mmap()
f2fs: fix build error on android tracepoints
ANDROID: cuttlefish_defconfig: Enable CAN/VCAN
UPSTREAM: pidfd: fix a poll race when setting exit_state
BACKPORT: arch: wire-up pidfd_open()
BACKPORT: pid: add pidfd_open()
UPSTREAM: pidfd: add polling support
UPSTREAM: signal: improve comments
UPSTREAM: fork: do not release lock that wasn't taken
BACKPORT: signal: support CLONE_PIDFD with pidfd_send_signal
BACKPORT: clone: add CLONE_PIDFD
UPSTREAM: Make anon_inodes unconditional
UPSTREAM: signal: use fdget() since we don't allow O_PATH
UPSTREAM: signal: don't silently convert SI_USER signals to non-current pidfd
BACKPORT: signal: add pidfd_send_signal() syscall
UPSTREAM: net-ipv6-ndisc: add support for RFC7710 RA Captive Portal Identifier
ANDROID: fix up 9p filesystem due to CFI non-upstream patches
f2fs: use EINVAL for superblock with invalid magic
f2fs: fix to read source block before invalidating it
f2fs: remove redundant check from f2fs_setflags_common()
f2fs: use generic checking function for FS_IOC_FSSETXATTR
f2fs: use generic checking and prep function for FS_IOC_SETFLAGS
ubifs, fscrypt: cache decrypted symlink target in ->i_link
vfs: use READ_ONCE() to access ->i_link
fs, fscrypt: clear DCACHE_ENCRYPTED_NAME when unaliasing directory
ANDROID: (arm64) cuttlefish_defconfig: enable CONFIG_CPU_FREQ_TIMES
ANDROID: xfrm: remove in_compat_syscall() checks
ANDROID: enable CONFIG_RTC_DRV_TEST on cuttlefish
UPSTREAM: binder: Set end of SG buffer area properly.
ANDROID: x86_64_cuttlefish_defconfig: enable CONFIG_CPU_FREQ_TIMES
ANDROID: f2fs: add android fsync tracepoint
ANDROID: f2fs: fix wrong android tracepoint
fscrypt: cache decrypted symlink target in ->i_link
fscrypt: fix race where ->lookup() marks plaintext dentry as ciphertext
fscrypt: only set dentry_operations on ciphertext dentries
fscrypt: fix race allowing rename() and link() of ciphertext dentries
fscrypt: clean up and improve dentry revalidation
fscrypt: use READ_ONCE() to access ->i_crypt_info
fscrypt: remove WARN_ON_ONCE() when decryption fails
fscrypt: drop inode argument from fscrypt_get_ctx()
f2fs: improve print log in f2fs_sanity_check_ckpt()
f2fs: avoid out-of-range memory access
f2fs: fix to avoid long latency during umount
f2fs: allow all the users to pin a file
f2fs: support swap file w/ DIO
f2fs: allocate blocks for pinned file
f2fs: fix is_idle() check for discard type
f2fs: add a rw_sem to cover quota flag changes
f2fs: set SBI_NEED_FSCK for xattr corruption case
f2fs: use generic EFSBADCRC/EFSCORRUPTED
f2fs: Use DIV_ROUND_UP() instead of open-coding
f2fs: print kernel message if filesystem is inconsistent
f2fs: introduce f2fs_<level> macros to wrap f2fs_printk()
f2fs: avoid get_valid_blocks() for cleanup
f2fs: ioctl for removing a range from F2FS
f2fs: only set project inherit bit for directory
f2fs: separate f2fs i_flags from fs_flags and ext4 i_flags
UPSTREAM: kasan: initialize tag to 0xff in __kasan_kmalloc
UPSTREAM: x86/boot: Provide KASAN compatible aliases for string routines
UPSTREAM: mm/kasan: Remove the ULONG_MAX stack trace hackery
UPSTREAM: x86/uaccess, kasan: Fix KASAN vs SMAP
UPSTREAM: x86/uaccess: Introduce user_access_{save,restore}()
UPSTREAM: kasan: fix variable 'tag' set but not used warning
UPSTREAM: Revert "x86_64: Increase stack size for KASAN_EXTRA"
UPSTREAM: kasan: fix coccinelle warnings in kasan_p*_table
UPSTREAM: kasan: fix kasan_check_read/write definitions
BACKPORT: kasan: remove use after scope bugs detection.
BACKPORT: kasan: turn off asan-stack for clang-8 and earlier
UPSTREAM: slub: fix a crash with SLUB_DEBUG + KASAN_SW_TAGS
UPSTREAM: kasan, slab: remove redundant kasan_slab_alloc hooks
UPSTREAM: kasan, slab: make freelist stored without tags
UPSTREAM: kasan, slab: fix conflicts with CONFIG_HARDENED_USERCOPY
UPSTREAM: kasan: prevent tracing of tags.c
UPSTREAM: kasan: fix random seed generation for tag-based mode
UPSTREAM: slub: fix SLAB_CONSISTENCY_CHECKS + KASAN_SW_TAGS
UPSTREAM: kasan, slub: fix more conflicts with CONFIG_SLAB_FREELIST_HARDENED
UPSTREAM: kasan, slub: fix conflicts with CONFIG_SLAB_FREELIST_HARDENED
UPSTREAM: kasan, slub: move kasan_poison_slab hook before page_address
UPSTREAM: kasan, kmemleak: pass tagged pointers to kmemleak
UPSTREAM: kasan: fix assigning tags twice
UPSTREAM: kasan: mark file common so ftrace doesn't trace it
UPSTREAM: kasan, arm64: remove redundant ARCH_SLAB_MINALIGN define
UPSTREAM: kasan: fix krealloc handling for tag-based mode
UPSTREAM: kasan: make tag based mode work with CONFIG_HARDENED_USERCOPY
UPSTREAM: kasan, arm64: use ARCH_SLAB_MINALIGN instead of manual aligning
BACKPORT: mm/memblock.c: skip kmemleak for kasan_init()
UPSTREAM: kasan: add SPDX-License-Identifier mark to source files
BACKPORT: kasan: update documentation
UPSTREAM: kasan, arm64: select HAVE_ARCH_KASAN_SW_TAGS
UPSTREAM: kasan: add __must_check annotations to kasan hooks
BACKPORT: kasan, mm, arm64: tag non slab memory allocated via pagealloc
UPSTREAM: kasan, arm64: add brk handler for inline instrumentation
UPSTREAM: kasan: add hooks implementation for tag-based mode
UPSTREAM: mm: move obj_to_index to include/linux/slab_def.h
UPSTREAM: kasan: add bug reporting routines for tag-based mode
UPSTREAM: kasan: split out generic_report.c from report.c
UPSTREAM: kasan, mm: perform untagged pointers comparison in krealloc
BACKPORT: kasan, arm64: enable top byte ignore for the kernel
BACKPORT: kasan, arm64: fix up fault handling logic
UPSTREAM: kasan: preassign tags to objects with ctors or SLAB_TYPESAFE_BY_RCU
UPSTREAM: kasan, arm64: untag address in _virt_addr_is_linear
UPSTREAM: kasan: add tag related helper functions
BACKPORT: arm64: move untagged_addr macro from uaccess.h to memory.h
BACKPORT: kasan: initialize shadow to 0xff for tag-based mode
BACKPORT: kasan: rename kasan_zero_page to kasan_early_shadow_page
BACKPORT: kasan, arm64: adjust shadow size for tag-based mode
BACKPORT: kasan: add CONFIG_KASAN_GENERIC and CONFIG_KASAN_SW_TAGS
UPSTREAM: kasan: rename source files to reflect the new naming scheme
BACKPORT: kasan: move common generic and tag-based code to common.c
UPSTREAM: kasan, slub: handle pointer tags in early_kmem_cache_node_alloc
UPSTREAM: kasan, mm: change hooks signatures
UPSTREAM: arm64: add EXPORT_SYMBOL_NOKASAN()
BACKPORT: compiler: remove __no_sanitize_address_or_inline again
UPSTREAM: mm/kasan/quarantine.c: make quarantine_lock a raw_spinlock_t
UPSTREAM: lib/test_kasan.c: add tests for several string/memory API functions
UPSTREAM: arm64: lib: use C string functions with KASAN enabled
UPSTREAM: compiler: introduce __no_sanitize_address_or_inline
UPSTREAM: arm64: Fix typo in a comment in arch/arm64/mm/kasan_init.c
BACKPORT: kernel/memremap, kasan: make ZONE_DEVICE with work with KASAN
BACKPORT: mm/mempool.c: remove unused argument in kasan_unpoison_element() and remove_element()
UPSTREAM: kasan: only select SLUB_DEBUG with SYSFS=y
UPSTREAM: kasan: depend on CONFIG_SLUB_DEBUG
UPSTREAM: KASAN: prohibit KASAN+STRUCTLEAK combination
UPSTREAM: arm64: kasan: avoid pfn_to_nid() before page array is initialized
UPSTREAM: kasan: fix invalid-free test crashing the kernel
UPSTREAM: kasan, slub: fix handling of kasan_slab_free hook
UPSTREAM: slab, slub: skip unnecessary kasan_cache_shutdown()
BACKPORT: kasan: make kasan_cache_create() work with 32-bit slab cache sizes
UPSTREAM: locking/atomics: Instrument cmpxchg_double*()
UPSTREAM: locking/atomics: Instrument xchg()
UPSTREAM: locking/atomics: Simplify cmpxchg() instrumentation
UPSTREAM: locking/atomics/x86: Reduce arch_cmpxchg64*() instrumentation
UPSTREAM: locking/atomic, asm-generic, x86: Add comments for atomic instrumentation
UPSTREAM: locking/atomic, asm-generic: Add KASAN instrumentation to atomic operations
UPSTREAM: locking/atomic/x86: Switch atomic.h to use atomic-instrumented.h
UPSTREAM: locking/atomic, asm-generic: Add asm-generic/atomic-instrumented.h
BACKPORT: kasan, arm64: clean up KASAN_SHADOW_SCALE_SHIFT usage
UPSTREAM: kasan: clean up KASAN_SHADOW_SCALE_SHIFT usage
UPSTREAM: kasan: fix prototype author email address
UPSTREAM: kasan: detect invalid frees
UPSTREAM: kasan: unify code between kasan_slab_free() and kasan_poison_kfree()
UPSTREAM: kasan: detect invalid frees for large mempool objects
UPSTREAM: kasan: don't use __builtin_return_address(1)
UPSTREAM: kasan: detect invalid frees for large objects
UPSTREAM: kasan: add functions for unpoisoning stack variables
UPSTREAM: kasan: add tests for alloca poisoning
UPSTREAM: kasan: support alloca() poisoning
UPSTREAM: kasan/Makefile: support LLVM style asan parameters
BACKPORT: kasan: add compiler support for clang
BACKPORT: fs: dcache: Revert "manually unpoison dname after allocation to shut up kasan's reports"
UPSTREAM: fs/dcache: Use read_word_at_a_time() in dentry_string_cmp()
UPSTREAM: lib/strscpy: Shut up KASAN false-positives in strscpy()
UPSTREAM: compiler.h: Add read_word_at_a_time() function.
UPSTREAM: compiler.h, kasan: Avoid duplicating __read_once_size_nocheck()
UPSTREAM: arm64/mm/kasan: don't use vmemmap_populate() to initialize shadow
UPSTREAM: Documentation/features/KASAN: mark KASAN as supported only on 64-bit on x86
f2fs: Add option to limit required GC for checkpoint=disable
f2fs: Fix accounting for unusable blocks
f2fs: Fix root reserved on remount
f2fs: Lower threshold for disable_cp_again
f2fs: fix sparse warning
f2fs: fix f2fs_show_options to show nodiscard mount option
f2fs: add error prints for debugging mount failure
f2fs: fix to do sanity check on segment bitmap of LFS curseg
f2fs: add missing sysfs entries in documentation
f2fs: fix to avoid deadloop if data_flush is on
f2fs: always assume that the device is idle under gc_urgent
f2fs: add bio cache for IPU
f2fs: allow ssr block allocation during checkpoint=disable period
f2fs: fix to check layout on last valid checkpoint park
Conflicts:
arch/arm64/configs/cuttlefish_defconfig
arch/arm64/include/asm/memory.h
arch/arm64/include/asm/thread_info.h
arch/x86/configs/x86_64_cuttlefish_defconfig
build.config.common
drivers/dma-buf/dma-buf.c
fs/crypto/Makefile
fs/crypto/bio.c
fs/crypto/fscrypt_private.h
fs/crypto/keyinfo.c
fs/ext4/page-io.c
fs/f2fs/data.c
fs/f2fs/f2fs.h
fs/f2fs/inode.c
fs/f2fs/segment.c
fs/userfaultfd.c
include/linux/dma-buf.h
include/linux/fscrypt.h
include/linux/kasan.h
include/linux/platform_data/ds2482.h
include/uapi/linux/fs.h
kernel/sched/deadline.c
kernel/sched/fair.c
kernel/sched/rt.c
kernel/sched/sched.h
kernel/sched/stop_task.c
kernel/sched/walt.c
kernel/sched/walt.h
lib/test_kasan.c
mm/kasan/common.c
mm/kasan/kasan.h
mm/kasan/report.c
mm/slub.c
mm/vmalloc.c
scripts/Makefile.kasan
Changed below files to fix build errors:
drivers/char/diag/diagchar_core.c
drivers/power/supply/qcom/battery.c
drivers/power/supply/qcom/smb1390-charger-psy.c
drivers/power/supply/qcom/smb1390-charger.c
drivers/power/supply/qcom/step-chg-jeita.c
fs/crypto/fscrypt_ice.c
fs/crypto/fscrypt_private.h
fs/f2fs/inode.c
include/uapi/linux/fscrypt.h
net/qrtr/qrtr.c
gen_headers_arm.bp
gen_headers_arm64.bp
Extra added fixes in fs/f2fs/data.c for FBE:
* Fix FBE regression with 9937c21ce1 ("f2fs: add bio cache
for IPU"). The above commit is not setting the DUN for
bio, due to which the bio's could get corrupted when FBE
is enabled.
* The f2fs_merge_page_bio() incorrectly uses the bio after
it is submitted for IO when fscrypt_mergeable_bio()
returns false. Fix it by making the submitted bio NULL
so that a new bio gets allocated for the next/new page.
Ignored the below scheduler patches as they are already present:
ANDROID: sched: WALT: Add support for CFS_BANDWIDTH
ANDROID: sched: WALT: Refactor cumulative runnable average fixup
picked below patches from 4.14.159 and 4.14.172 versions to fix issues
0e39aa9d5 "UPSTREAM: arm64: Validate tagged addresses in access_ok() called from kernel threads"
352902650 "fscrypt: support passing a keyring key to FS_IOC_ADD_ENCRYPTION_KEY"
Change-Id: I205b796ee125fa6e9d27fa30f881e4e8fe8bea29
Signed-off-by: Srinivasarao P <spathi@codeaurora.org>
Signed-off-by: Blagovest Kolenichev <bkolenichev@codeaurora.org>
rename 'name' to 'buf_name' in dma_buf since 'name' will be used
by the user specified name in upstream commit
"UPSTREAM: dma-buf: add DMA_BUF_SET_NAME ioctls".
Change-Id: I70e20bdb71e964d1da240b1d0323b26bb89d0ddc
Signed-off-by: Srinivasarao P <spathi@codeaurora.org>
This patch adds complimentary DMA_BUF_SET_NAME ioctls, which lets
userspace processes attach a free-form name to each buffer.
This information can be extremely helpful for tracking and accounting
shared buffers. For example, on Android, we know what each buffer will
be used for at allocation time: GL, multimedia, camera, etc. The
userspace allocator can use DMA_BUF_SET_NAME to associate that
information with the buffer, so we can later give developers a
breakdown of how much memory they're allocating for graphics, camera,
etc.
Signed-off-by: Greg Hackmann <ghackmann@google.com>
Signed-off-by: Chenbo Feng <fengc@google.com>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20190613223408.139221-3-fengc@google.com
(cherry picked from commit bb2bb903042517b8fb17b2bc21e00512f2dcac01)
Bug: 135613055
Test: run dmabufinfo_test.cpp and check the name is showed up in
/proc/pid/fd
Change-Id: Ib7f227b5e730c7ecfade14ad596ba3889eab1b0b
CONFIG_DEBUG_DMA_BUF_REF will collect stacktraces for
each call to dma_buf_get and dma_buf_put to help identify
leaks.
Change-Id: If299d73fbde80fc41d2a2150c4b6b8e7223734fa
Signed-off-by: Patrick Daly <pdaly@codeaurora.org>
gettimeofday() uses CLOCK_REALTIME which gives the current real-world
time. This time is discontiguous from boot till the userspace sets
the "wall clock time". Depending on such discountigous timer for
tracking buffer alive time is not right. Instead, use ktime which
uses CLOCK_MONOTONIC which is guaranteed to increase monotonically
from zero when system boots.
Change-Id: I2889be4ac7bb0e2ffd8845af1a8c91759712c891
Signed-off-by: Sudarshan Rajagopalan <sudaraja@codeaurora.org>
Debugfs node dmabuf/dmaprocs prints information about all dma-bufs
sorted by size and organized by process.
Change-Id: I7255be83d32896157fc099848cf511ae551ab527
Signed-off-by: Dylan Hatch <dhatch@codeaurora.org>
Making the dma-buf anon file name unique will allow clients to be able
to uniquely identify a dma-buf.
One advantage of this is that dma-buf memory leaks can be more easily
debugged as now we will be able to identify which dma-buf buffers a client
is using.
Change-Id: I5cff7baafdda16d2812201b73693fc32b38e86b4
Signed-off-by: Liam Mark <lmark@codeaurora.org>
In order to improve performance allow dma-buf clients the ability
to apply cache maintenance to only a subset of a dma-buf.
Kernel clients will be able to use dma_buf_begin_cpu_access_partial
and dma_buf_end_cpu_access_partial functions to only apply cache
maintenance to a range within the dma-buf.
Userspace clients will be able to restrict cache maintenance to only
the subset of the dma-buf which is mmap(ed) by setting the
DMA_BUF_SYNC_USER_MAPPED flag when calling the DMA_BUF_IOCTL_SYNC IOCT.
Change-Id: Id02e240ec43de636bb801929016c8d351b2ae96d
Signed-off-by: Liam Mark <lmark@codeaurora.org>
Allow ION kernel clients to get an ion buffer's flags.
This will allow ION clients to be able to determine if a buffer
is cached, uncached, secure ect.
Change-Id: I5460478e69f28e74906fd298f57dabd9e62c1e1f
Signed-off-by: Liam Mark <lmark@codeaurora.org>
Add support for configuring dma mapping attributes when mapping
memory through dma_buf_map_attachment.
Change-Id: I05c2133c432718bcaececb3858af19652d31b02b
Signed-off-by: Liam Mark <lmark@codeaurora.org>
- Again move the information relevant for driver writers next to the
callbacks.
- Put the overview and userspace interface documentation into a DOC:
section within the code.
- Remove the text that mmap needs to be coherent - since the
DMA_BUF_IOCTL_SYNC landed that's no longer the case. But keep the text
that for pte zapping exporters need to adjust the address space.
- Add a FIXME that kmap and the new begin/end stuff used by the SYNC
ioctl don't really mix correctly. That's something I just realized
while doing this doc rework.
- Augment function and structure docs like usual.
Cc: linux-doc@vger.kernel.org
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
[sumits: fix cosmetic issues]
Link: http://patchwork.freedesktop.org/patch/msgid/20161209185309.1682-5-daniel.vetter@ffwll.ch
- Put the initial overview for dma-buf into dma-buf.rst.
- Put all the comments about detailed semantics into the right
kernel-doc comment for functions or ops structure member.
- To allow that detail, switch the reworked kerneldoc to inline style
for dma_buf_ops.
- Tie everything together into a much more streamlined overview
comment, relying on the hyperlinks for all the details.
- Also sprinkle some links into the kerneldoc for dma_buf and
dma_buf_attachment to tie it all together.
Cc: linux-doc@vger.kernel.org
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Link: http://patchwork.freedesktop.org/patch/msgid/20161209185309.1682-4-daniel.vetter@ffwll.ch
Apparently nobody noticed that dma-buf.h wasn't actually pulled into
docbook build. And as a result the headerdoc comments bitrot a bit.
Add missing params/fields.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Add reference counting on a kernel module that exports dma-buf and
implements its operations. This prevents the module from being unloaded
while DMABUF file is in use.
The original patch [1] was submitted by Tomasz Stanislawski, but this
is a simpler way to do it.
v3: call module_put() as late as possible, per gregkh's comment.
v2: move owner to struct dma_buf, and use DEFINE_DMA_BUF_EXPORT_INFO
macro to simplify the change.
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
[1]: https://lkml.org/lkml/2012/8/8/163
At present, dma_buf_export() takes a series of parameters, which
makes it difficult to add any new parameters for exporters, if required.
Make it simpler by moving all these parameters into a struct, and pass
the struct * as parameter to dma_buf_export().
While at it, unite dma_buf_export_named() with dma_buf_export(), and
change all callers accordingly.
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
Reviewed-by: Daniel Thompson <daniel.thompson@linaro.org>
Acked-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Acked-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Russell King observed 'wierd' looking output from debugfs, and also suggested
better ways of getting device names (use KBUILD_MODNAME, dev_name())
This patch addresses these issues to make the debugfs output correct and better
looking.
While at it, replace seq_printf with seq_puts to remove the checkpatch.pl
warnings.
Reported-by: Russell King - ARM Linux <linux@arm.linux.org.uk>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Add debugfs support to make it easier to print debug information
about the dma-buf buffers.
Cc: Dave Airlie <airlied@redhat.com>
[minor fixes on init and warning fix]
Cc: Dan Carpenter <dan.carpenter@oracle.com>
[remove double unlock in fail case]
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
For debugging purposes, it is useful to have a name-string added
while exporting buffers. Hence, dma_buf_export() is replaced with
dma_buf_export_named(), which additionally takes 'exp_name' as a
parameter.
For backward compatibility, and for lazy exporters who don't wish to
name themselves, a #define dma_buf_export() is also made available,
which adds a __FILE__ instead of 'exp_name'.
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
[Thanks for the idea!]
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
All drivers which implement this need to have some sort of refcount to
allow concurrent vmap usage. Hence implement this in the dma-buf core.
To protect against concurrent calls we need a lock, which potentially
causes new funny locking inversions. But this shouldn't be a problem
for exporters with statically allocated backing storage, and more
dynamic drivers have decent issues already anyway.
Inspired by some refactoring patches from Aaron Plattner, who
implemented the same idea, but only for drm/prime drivers.
v2: Check in dma_buf_release that no dangling vmaps are left.
Suggested by Aaron Plattner. We might want to do similar checks for
attachments, but that's for another patch. Also fix up ERR_PTR return
for vmap.
v3: Check whether the passed-in vmap address matches with the cached
one for vunmap. Eventually we might want to remove that parameter -
compared to the kmap functions there's no need for the vaddr for
unmapping. Suggested by Chris Wilson.
v4: Fix a brown-paper-bag bug spotted by Aaron Plattner.
Cc: Aaron Plattner <aplattner@nvidia.com>
Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
Tested-by: Aaron Plattner <aplattner@nvidia.com>
Reviewed-by: Rob Clark <rob@ti.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Documentation says that code requiring dma-buf should add it to
select, so inline fallbacks are not going to be used. A link error
will make it obvious what went wrong, instead of silently doing
nothing at runtime.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Rob Clark <rob.clark@linaro.org>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Some minor inline documentation fixes for gaps resulting from new patches.
Signed-off-by: Sumit Semwal <sumit.semwal@ti.com>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
The main requirement I have for this interface is for scanning out
using the USB gpu devices. Since these devices have to read the
framebuffer on updates and linearly compress it, using kmaps
is a major overhead for every update.
v2: fix warn issues pointed out by Sylwester Nawrocki.
v3: fix compile !CONFIG_DMA_SHARED_BUFFER and add _GPL for now
Signed-off-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Rob Clark <rob.clark@linaro.org>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Compared to Rob Clark's RFC I've ditched the prepare/finish hooks
and corresponding ioctls on the dma_buf file. The major reason for
that is that many people seem to be under the impression that this is
also for synchronization with outstanding asynchronous processsing.
I'm pretty massively opposed to this because:
- It boils down reinventing a new rather general-purpose userspace
synchronization interface. If we look at things like futexes, this
is hard to get right.
- Furthermore a lot of kernel code has to interact with this
synchronization primitive. This smells a look like the dri1 hw_lock,
a horror show I prefer not to reinvent.
- Even more fun is that multiple different subsystems would interact
here, so we have plenty of opportunities to create funny deadlock
scenarios.
I think synchronization is a wholesale different problem from data
sharing and should be tackled as an orthogonal problem.
Now we could demand that prepare/finish may only ensure cache
coherency (as Rob intended), but that runs up into the next problem:
We not only need mmap support to facilitate sw-only processing nodes
in a pipeline (without jumping through hoops by importing the dma_buf
into some sw-access only importer), which allows for a nicer
ION->dma-buf upgrade path for existing Android userspace. We also need
mmap support for existing importing subsystems to support existing
userspace libraries. And a loot of these subsystems are expected to
export coherent userspace mappings.
So prepare/finish can only ever be optional and the exporter /needs/
to support coherent mappings. Given that mmap access is always
somewhat fallback-y in nature I've decided to drop this optimization,
instead of just making it optional. If we demonstrate a clear need for
this, supported by benchmark results, we can always add it in again
later as an optional extension.
Other differences compared to Rob's RFC is the above mentioned support
for mapping a dma-buf through facilities provided by the importer.
Which results in mmap support no longer being optional.
Note that this dma-buf mmap patch does _not_ support every possible
insanity an existing subsystem could pull of with mmap: Because it
does not allow to intercept pagefaults and shoot down ptes importing
subsystems can't add some magic of their own at these points (e.g. to
automatically synchronize with outstanding rendering or set up some
special resources). I've done a cursory read through a few mmap
implementions of various subsytems and I'm hopeful that we can avoid
this (and the complexity it'd bring with it).
Additonally I've extended the documentation a bit to explain the hows
and whys of this mmap extension.
In case we ever want to add support for explicitly cache maneged
userspace mmap with a prepare/finish ioctl pair, we could specify that
userspace needs to mmap a different part of the dma_buf, e.g. the
range starting at dma_buf->size up to dma_buf->size*2. This works
because the size of a dma_buf is invariant over it's lifetime. The
exporter would obviously need to fall back to coherent mappings for
both ranges if a legacy clients maps the coherent range and the
architecture cannot suppor conflicting caching policies. Also, this
would obviously be optional and userspace needs to be able to fall
back to coherent mappings.
v2:
- Spelling fixes from Rob Clark.
- Compile fix for !DMA_BUF from Rob Clark.
- Extend commit message to explain how explicitly cache managed mmap
support could be added later.
- Extend the documentation with implementations notes for exporters
that need to manually fake coherency.
v3:
- dma_buf pointer initialization goof-up noticed by Rebecca Schultz
Zavin.
Cc: Rob Clark <rob.clark@linaro.org>
Cc: Rebecca Schultz Zavin <rebecca@android.com>
Acked-by: Rob Clark <rob.clark@linaro.org>
Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Pull dma-buf updates from Sumit Semwal:
"This includes the following key items:
- kernel cpu access support,
- flag-passing to dma_buf_fd,
- relevant Documentation updates, and
- some minor cleanups and fixes.
These changes are needed for the drm prime/dma-buf interface code that
Dave Airlie plans to submit in this merge window."
* 'for-linus-3.4' of git://git.linaro.org/people/sumitsemwal/linux-dma-buf:
dma-buf: correct dummy function declarations.
dma-buf: document fd flags and O_CLOEXEC requirement
dma_buf: Add documentation for the new cpu access support
dma-buf: add support for kernel cpu access
dma-buf: don't hold the mutex around map/unmap calls
dma-buf: add get_dma_buf()
dma-buf: pass flags into dma_buf_fd.
dma-buf: add dma_data_direction to unmap dma_buf_op
dma-buf: Move code out of mutex-protected section in dma_buf_attach()
dma-buf: Return error instead of using a goto statement when possible
dma-buf: Remove unneeded sanity checks
dma-buf: Constify ops argument to dma_buf_export()
Dummy functions for the newly added cpu access ops need variable names for
arguments.
Also, the introduction of flags in dma_buf_fd needs to be added to dummy
functions as well.
Signed-off-by: Sumit Semwal <sumit.semwal@ti.com>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Big differences to other contenders in the field (like ion) is
that this also supports highmem, so we have to split up the cpu
access from the kernel side into a prepare and a kmap step.
Prepare is allowed to fail and should do everything required so that
the kmap calls can succeed (like swapin/backing storage allocation,
flushing, ...).
More in-depth explanations will follow in the follow-up documentation
patch.
Changes in v2:
- Clear up begin_cpu_access confusion noticed by Sumit Semwal.
- Don't automatically fallback from the _atomic variants to the
non-atomic variants. The _atomic callbacks are not allowed to
sleep, so we want exporters to make this decision explicit. The
function signatures are explicit, so simpler exporters can still
use the same function for both.
- Make the unmap functions optional. Simpler exporters with permanent
mappings don't need to do anything at unmap time.
Changes in v3:
- Adjust the WARN_ON checks for the new ->ops functions as suggested
by Rob Clark and Sumit Semwal.
- Rebased on top of latest dma-buf-next git.
Changes in v4:
- Fixup a missing - in a return -EINVAL; statement.
Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Rob Clark <rob@ti.com>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
The mutex protects the attachment list and hence needs to be held
around the callbakc to the exporters (optional) attach/detach
functions.
Holding the mutex around the map/unmap calls doesn't protect any
dma_buf state. Exporters need to properly protect any of their own
state anyway (to protect against calls from their own interfaces).
So this only makes the locking messier (and lockdep easier to anger).
Therefore let's just drop this.
v2: Rebased on top of latest dma-buf-next git.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Rob Clark <rob.clark@linaro.org>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Works in a similar way to get_file(), and is needed in cases such as
when the exporter needs to also keep a reference to the dmabuf (that
is later released with a dma_buf_put()), and possibly other similar
cases.
Signed-off-by: Rob Clark <rob@ti.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
We need to pass the flags into dma_buf_fd at this point,
so the flags end up doing the right thing for O_CLOEXEC.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Rob Clark <rob@ti.com>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Some exporters may use DMA map/unmap APIs in dma-buf ops, which require
enum dma_data_direction for both map and unmap operations.
Thus, the unmap dma_buf_op also needs to have enum dma_data_direction as
a parameter.
Reported-by: Tomasz Stanislawski <t.stanislaws@samsung.com>
Signed-off-by: Sumit Semwal <sumit.semwal@ti.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
The <linux/device.h> header includes a lot of stuff, and
it in turn gets a lot of use just for the basic "struct device"
which appears so often.
Clean up the users as follows:
1) For those headers only needing "struct device" as a pointer
in fcn args, replace the include with exactly that.
2) For headers not really using anything from device.h, simply
delete the include altogether.
3) For headers relying on getting device.h implicitly before
being included themselves, now explicitly include device.h
4) For files in which doing #1 or #2 uncovers an implicit
dependency on some other header, fix by explicitly adding
the required header(s).
Any C files that were implicitly relying on device.h to be
present have already been dealt with in advance.
Total removals from #1 and #2: 51. Total additions coming
from #3: 9. Total other implicit dependencies from #4: 7.
As of 3.3-rc1, there were 110, so a net removal of 42 gives
about a 38% reduction in device.h presence in include/*
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
This is the first step in defining a dma buffer sharing mechanism.
A new buffer object dma_buf is added, with operations and API to allow easy
sharing of this buffer object across devices.
The framework allows:
- creation of a buffer object, its association with a file pointer, and
associated allocator-defined operations on that buffer. This operation is
called the 'export' operation.
- different devices to 'attach' themselves to this exported buffer object, to
facilitate backing storage negotiation, using dma_buf_attach() API.
- the exported buffer object to be shared with the other entity by asking for
its 'file-descriptor (fd)', and sharing the fd across.
- a received fd to get the buffer object back, where it can be accessed using
the associated exporter-defined operations.
- the exporter and user to share the scatterlist associated with this buffer
object using map_dma_buf and unmap_dma_buf operations.
Atleast one 'attach()' call is required to be made prior to calling the
map_dma_buf() operation.
Couple of building blocks in map_dma_buf() are added to ease introduction
of sync'ing across exporter and users, and late allocation by the exporter.
For this first version, this framework will work with certain conditions:
- *ONLY* exporter will be allowed to mmap to userspace (outside of this
framework - mmap is not a buffer object operation),
- currently, *ONLY* users that do not need CPU access to the buffer are
allowed.
More details are there in the documentation patch.
This is based on design suggestions from many people at the mini-summits[1],
most notably from Arnd Bergmann <arnd@arndb.de>, Rob Clark <rob@ti.com> and
Daniel Vetter <daniel@ffwll.ch>.
The implementation is inspired from proof-of-concept patch-set from
Tomasz Stanislawski <t.stanislaws@samsung.com>, who demonstrated buffer sharing
between two v4l2 devices. [2]
[1]: https://wiki.linaro.org/OfficeofCTO/MemoryManagement
[2]: http://lwn.net/Articles/454389
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Signed-off-by: Sumit Semwal <sumit.semwal@ti.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-and-Tested-by: Rob Clark <rob.clark@linaro.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
