| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <20250327212723.GK4861@pendragon.ideasonboard.com>
Date: Thu, 27 Mar 2025 23:27:23 +0200
From: Laurent Pinchart <laurent.pinchart@...asonboard.com>
To: Ricardo Ribalda <ribalda@...omium.org>
Cc: Hans de Goede <hdegoede@...hat.com>,
Mauro Carvalho Chehab <mchehab@...nel.org>,
Guennadi Liakhovetski <guennadi.liakhovetski@...el.com>,
linux-media@...r.kernel.org, linux-kernel@...r.kernel.org,
Mauro Carvalho Chehab <mchehab+samsung@...nel.org>
Subject: Re: [PATCH v5 3/5] media: uvcvideo: Increase/decrease the PM counter
per IOCTL
On Thu, Mar 27, 2025 at 10:05:19PM +0100, Ricardo Ribalda wrote:
> On Thu, 27 Mar 2025 at 18:57, Laurent Pinchart wrote:
> > On Thu, Mar 27, 2025 at 07:52:27PM +0200, Laurent Pinchart wrote:
> > > On Mon, Mar 03, 2025 at 07:13:40PM +0000, Ricardo Ribalda wrote:
> > > > Now we call uvc_pm_get/put from the device open/close. This low
> > > > level of granularity might leave the camera powered on in situations
> > > > where it is not needed.
> > > >
> > > > Increase the granularity by increasing and decreasing the Power
> > >
> > > You're decreasing the granularity, not increasing it.
> > >
> > > > Management counter per ioctl. There are two special cases where the
> > > > power management outlives the ioctl: async controls and streamon. Handle
> > > > those cases as well.
> > > >
> > > > In a future patch, we will remove the uvc_pm_get/put from open/close.
> > > >
> > > > Reviewed-by: Hans de Goede <hdegoede@...hat.com>
> > > > Signed-off-by: Ricardo Ribalda <ribalda@...omium.org>
> > > > ---
> > > > drivers/media/usb/uvc/uvc_ctrl.c | 13 +++++++++++--
> > > > drivers/media/usb/uvc/uvc_v4l2.c | 23 +++++++++++++++++++++--
> > > > drivers/media/usb/uvc/uvcvideo.h | 1 +
> > > > 3 files changed, 33 insertions(+), 4 deletions(-)
> > > >
> > > > diff --git a/drivers/media/usb/uvc/uvc_ctrl.c b/drivers/media/usb/uvc/uvc_ctrl.c
> > > > index 4e58476d305e..47188c7f96c7 100644
> > > > --- a/drivers/media/usb/uvc/uvc_ctrl.c
> > > > +++ b/drivers/media/usb/uvc/uvc_ctrl.c
> > > > @@ -1594,12 +1594,15 @@ static void uvc_ctrl_set_handle(struct uvc_fh *handle, struct uvc_control *ctrl,
> > > >
> > > > if (ctrl->handle) {
> > > > WARN_ON(!ctrl->handle->pending_async_ctrls);
> > > > - if (ctrl->handle->pending_async_ctrls)
> > > > + if (ctrl->handle->pending_async_ctrls) {
> > > > ctrl->handle->pending_async_ctrls--;
> > > > + uvc_pm_put(handle->chain->dev);
> > >
> > > Shouldn't this be
> > >
> > > uvc_pm_put(ctrl->handle->chain->dev);
> > >
> > > ? In practice it won't make a difference as dev will be the same for
> > > both, but it seems clearer.
> > >
> > > > + }
> > > > }
> > > >
> > > > ctrl->handle = new_handle;
> > > > handle->pending_async_ctrls++;
> > > > + uvc_pm_get(handle->chain->dev);
> > >
> > > Similarly, we should use ctrl->handle here too (including for the
> > > pending_async_ctrls++).
> > >
> > > > return;
> > > > }
> > > >
> > > > @@ -1611,6 +1614,7 @@ static void uvc_ctrl_set_handle(struct uvc_fh *handle, struct uvc_control *ctrl,
> > > > if (WARN_ON(!handle->pending_async_ctrls))
> > > > return;
> > > > handle->pending_async_ctrls--;
> > > > + uvc_pm_put(handle->chain->dev);
> > > > }
> > > >
> > > > void uvc_ctrl_status_event(struct uvc_video_chain *chain,
> > > > @@ -2815,6 +2819,7 @@ int uvc_ctrl_init_device(struct uvc_device *dev)
> > > > void uvc_ctrl_cleanup_fh(struct uvc_fh *handle)
> > > > {
> > > > struct uvc_entity *entity;
> > > > + int i;
> > > >
> > > > guard(mutex)(&handle->chain->ctrl_mutex);
> > > >
> > > > @@ -2829,7 +2834,11 @@ void uvc_ctrl_cleanup_fh(struct uvc_fh *handle)
> > > > }
> > > > }
> > > >
> > > > - WARN_ON(handle->pending_async_ctrls);
> > > > + if (!WARN_ON(handle->pending_async_ctrls))
> > > > + return;
> > > > +
> > > > + for (i = 0; i < handle->pending_async_ctrls; i++)
> > > > + uvc_pm_put(handle->stream->dev);
> > > > }
> > > >
> > > > /*
> > > > diff --git a/drivers/media/usb/uvc/uvc_v4l2.c b/drivers/media/usb/uvc/uvc_v4l2.c
> > > > index de1e105f7263..1c9ac72be58a 100644
> > > > --- a/drivers/media/usb/uvc/uvc_v4l2.c
> > > > +++ b/drivers/media/usb/uvc/uvc_v4l2.c
> > > > @@ -691,6 +691,9 @@ static int uvc_v4l2_release(struct file *file)
> > > > if (uvc_has_privileges(handle))
> > > > uvc_queue_release(&stream->queue);
> > > >
> > > > + if (handle->is_streaming)
> > > > + uvc_pm_put(stream->dev);
> > > > +
> > > > /* Release the file handle. */
> > > > uvc_dismiss_privileges(handle);
> > > > v4l2_fh_del(&handle->vfh);
> > > > @@ -857,6 +860,7 @@ static int uvc_ioctl_streamon(struct file *file, void *fh,
> > > > return ret;
> > > >
> > > > handle->is_streaming = true;
> > > > + uvc_pm_get(stream->dev);
> >
> > Another comment: shouldn't you handle the return value (here and
> > elsewhere, including where you use guards) ?
>
> Good point... I guess I got excited trying to use the guards :)
I like them too :-)
> > > >
> > > > return 0;
> > > > }
> > > > @@ -873,7 +877,10 @@ static int uvc_ioctl_streamoff(struct file *file, void *fh,
> > > > guard(mutex)(&stream->mutex);
> > > >
> > > > uvc_queue_streamoff(&stream->queue, type);
> > > > - handle->is_streaming = false;
> > > > + if (handle->is_streaming) {
> > > > + handle->is_streaming = false;
> > > > + uvc_pm_put(stream->dev);
> > > > + }
> > > >
> > > > return 0;
> > > > }
> > > > @@ -1410,6 +1417,8 @@ static long uvc_v4l2_compat_ioctl32(struct file *file,
> > > > void __user *up = compat_ptr(arg);
> > > > long ret;
> > > >
> > > > + guard(uvc_pm)(handle->stream->dev);
> > > > +
> > > > switch (cmd) {
> > > > case UVCIOC_CTRL_MAP32:
> > > > ret = uvc_v4l2_get_xu_mapping(&karg.xmap, up);
> > > > @@ -1444,6 +1453,16 @@ static long uvc_v4l2_compat_ioctl32(struct file *file,
> > > > }
> > > > #endif
> > > >
> > > > +static long uvc_v4l2_video_ioctl2(struct file *file,
> > > > + unsigned int cmd, unsigned long arg)
> > > > +{
> > > > + struct uvc_fh *handle = file->private_data;
> > > > +
> > > > + guard(uvc_pm)(handle->stream->dev);
> > > > +
> > > > + return video_ioctl2(file, cmd, arg);
> > > > +}
> > > > +
> > > > static ssize_t uvc_v4l2_read(struct file *file, char __user *data,
> > > > size_t count, loff_t *ppos)
> > > > {
> > > > @@ -1529,7 +1548,7 @@ const struct v4l2_file_operations uvc_fops = {
> > > > .owner = THIS_MODULE,
> > > > .open = uvc_v4l2_open,
> > > > .release = uvc_v4l2_release,
> > > > - .unlocked_ioctl = video_ioctl2,
> > > > + .unlocked_ioctl = uvc_v4l2_video_ioctl2,
> > >
> > > I'd have named this uvc_v4l2_unlocked_ioctl.
> > >
> > > > #ifdef CONFIG_COMPAT
> > > > .compat_ioctl32 = uvc_v4l2_compat_ioctl32,
> > > > #endif
> > > > diff --git a/drivers/media/usb/uvc/uvcvideo.h b/drivers/media/usb/uvc/uvcvideo.h
> > > > index fbe3649c7cd6..eb8e374fa4c5 100644
> > > > --- a/drivers/media/usb/uvc/uvcvideo.h
> > > > +++ b/drivers/media/usb/uvc/uvcvideo.h
> > > > @@ -766,6 +766,7 @@ void uvc_status_put(struct uvc_device *dev);
> > > > /* PM */
> > > > int uvc_pm_get(struct uvc_device *dev);
> > > > void uvc_pm_put(struct uvc_device *dev);
> > > > +DEFINE_GUARD(uvc_pm, struct uvc_device *, uvc_pm_get(_T), uvc_pm_put(_T))
> > > >
> > > > /* Controls */
> > > > extern const struct v4l2_subscribed_event_ops uvc_ctrl_sub_ev_ops;
--
Regards,
Laurent Pinchart
Powered by blists - more mailing lists