Skip to content

K
Kernel

openeuler / Kernel
接近 2 年 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 06 7月, 2012 1 次提交
  3. 14 5月, 2012 1 次提交
    • H
      [media] v4l2-dev: add flag to have the core lock all file operations · 5126f259
      由 Hans Verkuil 提交于 
      This used to be the default if the lock pointer was set, but now that lock is by
default only used for ioctl serialization. Those drivers that already used
core locking have this flag set explicitly, except for some drivers where
it was obvious that there was no need to serialize any file operations other
than ioctl.

The drivers that didn't need this flag were:

drivers/media/radio/dsbr100.c
drivers/media/radio/radio-isa.c
drivers/media/radio/radio-keene.c
drivers/media/radio/radio-miropcm20.c
drivers/media/radio/radio-mr800.c
drivers/media/radio/radio-tea5764.c
drivers/media/radio/radio-timb.c
drivers/media/video/vivi.c
sound/i2c/other/tea575x-tuner.c

The other drivers that use core locking and where it was not immediately
obvious that this flag wasn't needed were changed so that the flag is set
together with a comment that that driver needs work to avoid having to
set that flag. This will often involve taking the core lock in the fops
themselves.

Eventually this flag should go and it should not be used in new drivers.

There are a few reasons why we want to avoid core locking of non-ioctl
fops: in the case of mmap this can lead to a deadlock in rare situations
since when mmap is called the mmap_sem is held and it is possible for
other parts of the code to take that lock as well (copy_from_user()/copy_to_user()
perform a down_read(&mm->mmap_sem) when a page fault occurs).

It is very unlikely that that happens since the core lock serializes all
fops, but the kernel warns about it if lock validation is turned on.

For poll it is also undesirable to take the core lock as that can introduce
increased latency. The same is true for read/write.

While it was possible to make flags or something to turn on/off taking the
core lock for each file operation, in practice it is much simpler to just
not take it at all except for ioctl and leave it to the driver to take the
lock. There are only a handful fops compared to the zillion ioctls we have.

I also wanted to make it obvious which drivers still take the lock for all
fops, so that's why I chose to have drivers set it explicitly.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Acked-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>
      5126f259
  5. 08 5月, 2012 1 次提交
  7. 28 7月, 2011 1 次提交
    • M
      [media] Stop using linux/version.h on most video drivers · 1990d50b
      由 Mauro Carvalho Chehab 提交于 
      All the modified drivers didn't have any version increment since
Jan, 1 2011. Several of them didn't have any version increment
for a long time, even having new features and important bug fixes
happening.

As we're now filling the QUERYCAP version with the current Kernel
Release, we don't need to maintain a per-driver version control
anymore. So, let's just use the default.

In order to preserve the Kernel module version history, a
KERNEL_VERSION() macro were added to all modified drivers, and
the extraver number were incremented.

I opted to preserve the per-driver version control to a few
pwc, pvrusb2, s2255, s5p-fimc and sh_vou.

A few drivers are still using the legacy way to handle ioctl's.
So, we can't do such change on them, otherwise, they'll break.
Those are: uvc, et61x251 and sn9c102.

The rationale is that the per-driver version control seems to be
actively maintained on those.

Yet, I think that the better for them would be to just use the
default version numbering, instead of doing that by themselves.

While here, removed a few uneeded include linux/version.h
Acked-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>
      1990d50b
  9. 26 5月, 2011 1 次提交
  11. 20 5月, 2011 1 次提交
  13. 23 3月, 2011 1 次提交
  15. 22 3月, 2011 1 次提交
  17. 19 1月, 2011 1 次提交
  19. 29 12月, 2010 1 次提交
  21. 21 10月, 2010 1 次提交
  23. 19 5月, 2010 1 次提交
  25. 16 12月, 2009 3 次提交
  27. 17 6月, 2009 1 次提交
  29. 30 3月, 2009 1 次提交
  31. 03 1月, 2009 2 次提交
  33. 30 12月, 2008 2 次提交
  35. 22 10月, 2008 1 次提交
    • H
      V4L/DVB (9327): v4l: use video_device.num instead of minor in video%d · c6330fb8
      由 Hans Verkuil 提交于 
      The kernel number of a v4l2 node (e.g. videoX, radioX or vbiX) is now
independent of the minor number. So instead of using the minor field
of the video_device struct one has to use the num field: this always
contains the kernel number of the device node.

I forgot about this when I did the v4l2 core change, so this patch
converts all drivers that use it in one go. Luckily the change is
trivial.

Cc: michael@mihu.de
Cc: mchehab@infradead.org
Cc: corbet@lwn.net
Cc: luca.risolia@studio.unibo.it
Cc: isely@pobox.com
Cc: pe1rxq@amsat.org
Cc: royale@zerezo.com
Cc: mkrufky@linuxtv.org
Cc: stoth@linuxtv.org
Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>
      c6330fb8
  37. 12 10月, 2008 1 次提交
  39. 04 9月, 2008 2 次提交
  41. 27 7月, 2008 1 次提交
  43. 26 7月, 2008 1 次提交
  45. 24 7月, 2008 1 次提交
  47. 25 4月, 2008 1 次提交
  49. 22 10月, 2007 1 次提交
  51. 10 10月, 2007 1 次提交
  53. 19 7月, 2007 1 次提交
  55. 13 2月, 2007 1 次提交
  57. 01 7月, 2006 1 次提交
  59. 25 6月, 2006 2 次提交
    • J
      V4L/DVB (4154): Fix use-after-free bug in cpia2 driver · bafefc0c
      由 Jesper Juhl 提交于 
      The coverity checker detected a use-after-free error in
drivers/media/video/cpia2/cpia2_v4l.c::cpia2_close() (coverity
error #1281).
What happens is that we lock cam->busy_lock, then proceed to free
resources, and in the case of (--cam->open_count == 0) we finish off by
doing a kfree(cam) and then at the end of the function we do a
mutex_unlock(&cam->busy_lock) which will explode since it'll dereference
the free'd `cam' :
...
mutex_lock(&cam->busy_lock);
...
if (--cam->open_count == 0) {
    ...
    if (!cam->present) {
        video_unregister_device(dev);
        kfree(cam);
    }
}
mutex_unlock(&cam->busy_lock);   <--- PROBLEM, cam no longer around.
...
Since this only happens in the case of open_count going down to zero I
don't see a problem with just releasing the mutex after unregistering the
device and just before the kfree().  In this case there is nothing around
that we can race against; we are in the release method, open_count is zero,
(!cam->present) and the device has just been unregistered, so letting go of
the mutex at this point looks safe to me.
Patch below to implement that solution.
Acked-by: NRandy Dunlap <rdunlap@xenotime.net>
Signed-off-by: NJesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NMauro Carvalho Chehab <mchehab@infradead.org>
      bafefc0c
    • T
      V4L/DVB (4118): Whitespace cleanups · 657de3cd
      由 Trent Piepho 提交于 
      Signed-off-by: NTrent Piepho <xyzzy@speakeasy.org>
Signed-off-by: NMauro Carvalho Chehab <mchehab@infradead.org>
      657de3cd
  61. 25 3月, 2006 1 次提交
  63. 08 3月, 2006 1 次提交
  65. 03 3月, 2006 1 次提交
  67. 27 2月, 2006 1 次提交
    • A
      V4L/DVB (3376): Add cpia2 camera support · ab33d507
      由 Alan Cox 提交于 
      There has been a CPIA2 driver out of kernel for a long time and it has
been pretty clean for some time too. This is an import of the
sourceforge driver which has been stripped of
- 2.4 back compatibility
- 2.4 old style MJPEG ioctls
A couple of functions have been made static and the docs have been
repackaged into Documentation/video4linux.  The rvmalloc/free functions now
match the cpia driver again.  Other than that this is the code as is.
Tested on x86-64 with a QX5 microscope.
Signed-off-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NMauro Carvalho Chehab <mchehab@infradead.org>
      ab33d507