kernel/drivers/media/common/videobuf2, branch linux-rolling-stable

media: videobuf2: Fix device reference leak in vb2_dc_alloc error path

2026-01-08T09:17:04Z

commit 94de23a9aa487d7c1372efb161721d7949a177ae upstream. In vb2_dc_alloc(), get_device() is called to increment the device reference count. However, if subsequent DMA allocation fails (vb2_dc_alloc_coherent or vb2_dc_alloc_non_coherent returns error), the function returns without calling put_device(), causing a device reference leak. Add put_device() call in the error path before kfree() to properly release the device reference acquired earlier. Fixes: de27891f675e ("media: videobuf2: handle non-contiguous DMA allocations") Cc: stable@vger.kernel.org Signed-off-by: Haotian Zhang Reviewed-by: Marek Szyprowski Signed-off-by: Hans Verkuil Signed-off-by: Greg Kroah-Hartman

media: videobuf2: forbid remove_bufs when legacy fileio is active

2025-10-24T09:16:26Z

vb2_ioctl_remove_bufs() call manipulates queue internal buffer list, potentially overwriting some pointers used by the legacy fileio access mode. Forbid that ioctl when fileio is active to protect internal queue state between subsequent read/write calls. CC: stable@vger.kernel.org Fixes: a3293a85381e ("media: v4l2: Add REMOVE_BUFS ioctl") Reported-by: Shuangpeng Bai Closes: https://lore.kernel.org/linux-media/5317B590-AAB4-4F17-8EA1-621965886D49@psu.edu/ Signed-off-by: Marek Szyprowski Signed-off-by: Hans Verkuil

media: Drop V4L2_FL_USES_V4L2_FH checks

2025-08-13T06:33:46Z

Now that all drivers use v4l2_fh, we can drop the V4L2_FL_USES_V4L2_FH checks through the V4L2 core. To ensure that all new drivers use v4l2_fh, keep setting the V4L2_FL_USES_V4L2_FH flag in v4l2_fh_init(), and verify it is set after the .open() file operation returns. Signed-off-by: Laurent Pinchart Signed-off-by: Hans Verkuil

media: Wrap file->private_data access with a helper function

2025-08-13T06:33:23Z

Accessing file->private_data manually to retrieve the v4l2_fh pointer is error-prone, as the field is a void * and will happily convert implicitly to any pointer type. To avoid direct access to file->private_data, introduce a new inline function that retrieves the v4l2_fh pointer, and use it to replace common access patterns through the kernel. Changes to drivers have been generated with the following coccinelle semantic patch: @@ struct file *filp; identifier fh; @@ - struct v4l2_fh *fh = filp->private_data; + struct v4l2_fh *fh = file_to_v4l2_fh(filp); Manual changes have been applied to Documentation/ to update the usage patterns, and to include/media/v4l2-fh.h to add the new function. While at it, fix a typo in the title of v4l2-fh.rst: the file describes the "file handles" API, not "file handlers". No functional change is intended, this only paves the way to remove direct accesses to file->private_data and make V4L2 drivers safer. Other accesses to the field will be addressed separately. Signed-off-by: Laurent Pinchart Signed-off-by: Hans Verkuil

media: videobuf2: use sgtable-based scatterlist wrappers

2025-05-09T09:55:07Z

Use common wrappers operating directly on the struct sg_table objects to fix incorrect use of scatterlists sync calls. dma_sync_sg_for_*() functions have to be called with the number of elements originally passed to dma_map_sg_*() function, not the one returned in sgt->nents. Fixes: d4db5eb57cab ("media: videobuf2: add begin/end cpu_access callbacks to dma-sg") CC: stable@vger.kernel.org Signed-off-by: Marek Szyprowski Reviewed-by: Sergey Senozhatsky Acked-by: Tomasz Figa Signed-off-by: Hans Verkuil

media: videobuf2: check constants during build time

2025-04-25T13:14:29Z

There is nothing a driver author can do fix in the driver to make the global constants match. Since the assertion can be verified at build time, don't return EINVAL at runtime for it. Signed-off-by: Michał Mirosław Acked-by: Marek Szyprowski Signed-off-by: Hans Verkuil

module: Convert symbol namespace to string literal

2024-12-02T19:34:44Z

Clean up the existing export namespace code along the same lines of commit 33def8498fdd ("treewide: Convert macro and uses of __section(foo) to __section("foo")") and for the same reason, it is not desired for the namespace argument to be a macro expansion itself. Scripted using git grep -l -e MODULE_IMPORT_NS -e EXPORT_SYMBOL_NS | while read file; do awk -i inplace ' /^#define EXPORT_SYMBOL_NS/ { gsub(/__stringify$ns$/, "ns"); print; next; } /^#define MODULE_IMPORT_NS/ { gsub(/__stringify$ns$/, "ns"); print; next; } /MODULE_IMPORT_NS/ { $0 = gensub(/MODULE_IMPORT_NS$([^)]*)$/, "MODULE_IMPORT_NS(\"\\1\")", "g"); } /EXPORT_SYMBOL_NS/ { if ($0 ~ /(EXPORT_SYMBOL_NS[^(]*)$([^,]+),/) { if ($0 !~ /(EXPORT_SYMBOL_NS[^(]*)\(([^,]+), ([^)]+)$/ && $0 !~ /(EXPORT_SYMBOL_NS[^(]*)/ && $0 !~ /^my/) { getline line; gsub(/[[:space:]]*\\$/, ""); gsub(/[[:space:]]/, "", line); $0 = $0 " " line; } $0 = gensub(/(EXPORT_SYMBOL_NS[^(]*)$([^,]+), ([^)]+)$/, "\\1(\\2, \"\\3\")", "g"); } } { print }' $file; done Requested-by: Masahiro Yamada Signed-off-by: Peter Zijlstra (Intel) Link: https://mail.google.com/mail/u/2/#inbox/FMfcgzQXKWgMmjdFwwdsfgxzKpVHWPlc Acked-by: Greg KH Signed-off-by: Linus Torvalds

Merge tag 'v6.12-rc7' into __tmp-hansg-linux-tags_media_atomisp_6_13_1

2024-11-11T11:16:33Z

Linux 6.12-rc7 * tag 'v6.12-rc7': (1909 commits) Linux 6.12-rc7 filemap: Fix bounds checking in filemap_read() i2c: designware: do not hold SCL low when I2C_DYNAMIC_TAR_UPDATE is not set mailmap: add entry for Thorsten Blum ocfs2: remove entry once instead of null-ptr-dereference in ocfs2_xa_remove() signal: restore the override_rlimit logic fs/proc: fix compile warning about variable 'vmcore_mmap_ops' ucounts: fix counter leak in inc_rlimit_get_ucounts() selftests: hugetlb_dio: check for initial conditions to skip in the start mm: fix docs for the kernel parameter ``thp_anon=`` mm/damon/core: avoid overflow in damon_feed_loop_next_input() mm/damon/core: handle zero schemes apply interval mm/damon/core: handle zero {aggregation,ops_update} intervals mm/mlock: set the correct prev on failure objpool: fix to make percpu slot allocation more robust mm/page_alloc: keep track of free highatomic bcachefs: Fix UAF in __promote_alloc() error path bcachefs: Change OPT_STR max to be 1 less than the size of choices array bcachefs: btree_cache.freeable list fixes bcachefs: check the invalid parameter for perf test ...

media: videobuf2-core: copy vb planes unconditionally

2024-11-07T11:55:46Z

Copy the relevant data from userspace to the vb->planes unconditionally as it's possible some of the fields may have changed after the buffer has been validated. Keep the dma_buf_put(planes[plane].dbuf) calls in the first `if (!reacquired)` case, in order to be close to the plane validation code where the buffers were got in the first place. Cc: stable@vger.kernel.org Fixes: 95af7c00f35b ("media: videobuf2-core: release all planes first in __prepare_dmabuf()") Signed-off-by: Tudor Ambarus Tested-by: Will McVicker Acked-by: Tomasz Figa Signed-off-by: Hans Verkuil

media: vb2: fix confusing log message

2024-10-28T14:35:56Z

If the number of allocated buffers is less than q->min_queued_buffers, then a debug message was logged saying that it needs at least that many queued buffers. But the test is about allocated buffers. Update the message to say "allocated buffers". Signed-off-by: Hans Verkuil Reviewed-by: Nicolas Dufresne Acked-by: Tomasz Figa