| Message ID | 9f401f3732dd728e3d2ca508002c97b80a2eae30.1725265884.git.hverkuil-cisco@xs4all.nl (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | media: add missing wait_prepare/finish ops | expand |
Hi Hans, On 9/2/24 10:31 AM, Hans Verkuil wrote: > Without these ops the v4l2-compliance blocking wait test will fail. > These ops are required to ensure that when VIDIOC_DQBUF has to > wait for buffers to arrive, the queue lock is correctly released > and retaken. Otherwise the wait for a buffer would block all other > queue ioctls. > > Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> > Cc: Hans de Goede <hdegoede@redhat.com> Thank you for this patch. I have merged this in my media-atomisp branch: https://git.kernel.org/pub/scm/linux/kernel/git/hansg/linux.git/log/?h=media-atomisp I'll run some tests when I'm back home tonight (with access to atomisp2 hw) before included this in my upcoming atomisp pull-request for 6.12. Regards, Hans > --- > drivers/staging/media/atomisp/pci/atomisp_fops.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/drivers/staging/media/atomisp/pci/atomisp_fops.c b/drivers/staging/media/atomisp/pci/atomisp_fops.c > index 50c4123ba006..b180fcbea9b1 100644 > --- a/drivers/staging/media/atomisp/pci/atomisp_fops.c > +++ b/drivers/staging/media/atomisp/pci/atomisp_fops.c > @@ -441,6 +441,8 @@ const struct vb2_ops atomisp_vb2_ops = { > .buf_queue = atomisp_buf_queue, > .start_streaming = atomisp_start_streaming, > .stop_streaming = atomisp_stop_streaming, > + .wait_prepare = vb2_ops_wait_prepare, > + .wait_finish = vb2_ops_wait_finish, > }; > > static void atomisp_dev_init_struct(struct atomisp_device *isp)
Hi Hans, On 02/09/2024 12:39, Hans de Goede wrote: > Hi Hans, > > On 9/2/24 10:31 AM, Hans Verkuil wrote: >> Without these ops the v4l2-compliance blocking wait test will fail. >> These ops are required to ensure that when VIDIOC_DQBUF has to >> wait for buffers to arrive, the queue lock is correctly released >> and retaken. Otherwise the wait for a buffer would block all other >> queue ioctls. >> >> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> >> Cc: Hans de Goede <hdegoede@redhat.com> > > Thank you for this patch. > > I have merged this in my media-atomisp branch: > https://git.kernel.org/pub/scm/linux/kernel/git/hansg/linux.git/log/?h=media-atomisp > > I'll run some tests when I'm back home tonight (with access to > atomisp2 hw) before included this in my upcoming atomisp pull-request > for 6.12. Would you be able to post a PR today? We want to switch over to our new gitlab tree (https://gitlab.freedesktop.org/linux-media/media-staging), but we'd like to get all PRs for 6.12 merged first. Regards, Hans > > Regards, > > Hans > > >> --- >> drivers/staging/media/atomisp/pci/atomisp_fops.c | 2 ++ >> 1 file changed, 2 insertions(+) >> >> diff --git a/drivers/staging/media/atomisp/pci/atomisp_fops.c b/drivers/staging/media/atomisp/pci/atomisp_fops.c >> index 50c4123ba006..b180fcbea9b1 100644 >> --- a/drivers/staging/media/atomisp/pci/atomisp_fops.c >> +++ b/drivers/staging/media/atomisp/pci/atomisp_fops.c >> @@ -441,6 +441,8 @@ const struct vb2_ops atomisp_vb2_ops = { >> .buf_queue = atomisp_buf_queue, >> .start_streaming = atomisp_start_streaming, >> .stop_streaming = atomisp_stop_streaming, >> + .wait_prepare = vb2_ops_wait_prepare, >> + .wait_finish = vb2_ops_wait_finish, >> }; >> >> static void atomisp_dev_init_struct(struct atomisp_device *isp) > >
Hi, On 9/3/24 9:59 AM, Hans Verkuil wrote: > Hi Hans, > > On 02/09/2024 12:39, Hans de Goede wrote: >> Hi Hans, >> >> On 9/2/24 10:31 AM, Hans Verkuil wrote: >>> Without these ops the v4l2-compliance blocking wait test will fail. >>> These ops are required to ensure that when VIDIOC_DQBUF has to >>> wait for buffers to arrive, the queue lock is correctly released >>> and retaken. Otherwise the wait for a buffer would block all other >>> queue ioctls. >>> >>> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> >>> Cc: Hans de Goede <hdegoede@redhat.com> >> >> Thank you for this patch. >> >> I have merged this in my media-atomisp branch: >> https://git.kernel.org/pub/scm/linux/kernel/git/hansg/linux.git/log/?h=media-atomisp >> >> I'll run some tests when I'm back home tonight (with access to >> atomisp2 hw) before included this in my upcoming atomisp pull-request >> for 6.12. > > Would you be able to post a PR today? We want to switch over to our new gitlab > tree (https://gitlab.freedesktop.org/linux-media/media-staging), but we'd like > to get all PRs for 6.12 merged first. Yes I already reviewed + merged all the pending atomisp stuff yesterday. I just need to merge your "media: staging: atomisp: set lock before calling vb2_queue_init()" and run a quick test and then I'll send out the PR today. Regards, Hans
diff --git a/drivers/staging/media/atomisp/pci/atomisp_fops.c b/drivers/staging/media/atomisp/pci/atomisp_fops.c index 50c4123ba006..b180fcbea9b1 100644 --- a/drivers/staging/media/atomisp/pci/atomisp_fops.c +++ b/drivers/staging/media/atomisp/pci/atomisp_fops.c @@ -441,6 +441,8 @@ const struct vb2_ops atomisp_vb2_ops = { .buf_queue = atomisp_buf_queue, .start_streaming = atomisp_start_streaming, .stop_streaming = atomisp_stop_streaming, + .wait_prepare = vb2_ops_wait_prepare, + .wait_finish = vb2_ops_wait_finish, }; static void atomisp_dev_init_struct(struct atomisp_device *isp)
Without these ops the v4l2-compliance blocking wait test will fail. These ops are required to ensure that when VIDIOC_DQBUF has to wait for buffers to arrive, the queue lock is correctly released and retaken. Otherwise the wait for a buffer would block all other queue ioctls. Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Cc: Hans de Goede <hdegoede@redhat.com> --- drivers/staging/media/atomisp/pci/atomisp_fops.c | 2 ++ 1 file changed, 2 insertions(+)