[PATCH 2/2] drm/xe/ufence: Warn if mmget_not_zero() fails

[Francois Dugast]

On Mon, Oct 14, 2024 at 12:58:29PM +0200, Maarten Lankhorst wrote:
> I believe this is triggerable from userspace, so this should be a
> XE_IOCTL_DBG

Actually it seems this code is executed from the context of the worker (which
can be kicked from userspace). Besides, XE_IOCTL_DBG is used to check IOCTL
arguments, which I think is not really the case here. See:

    $ git grep XE_IOCTL_DBG drivers/gpu/drm/xe

> 
> Den 2024-10-14 kl. 11:27, skrev Francois Dugast:
> > On Fri, Oct 11, 2024 at 01:57:15PM +0200, Nirmoy Das wrote:
> > > This shouldn't happen but seen this while debugging ufence timeout
> > > issue time to time so log it to isolate this particular case.
> > > 
> > > Link: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/1630
> > > Cc: Matthew Brost <matthew.brost at intel.com>
> > > Signed-off-by: Nirmoy Das <nirmoy.das at intel.com>
> > 
> > Reviewed-by: Francois Dugast <francois.dugast at intel.com>
> > 
> > > ---
> > >   drivers/gpu/drm/xe/xe_sync.c | 2 ++
> > >   1 file changed, 2 insertions(+)
> > > 
> > > diff --git a/drivers/gpu/drm/xe/xe_sync.c b/drivers/gpu/drm/xe/xe_sync.c
> > > index 6a2e4dd41d56..41758f528d7c 100644
> > > --- a/drivers/gpu/drm/xe/xe_sync.c
> > > +++ b/drivers/gpu/drm/xe/xe_sync.c
> > > @@ -83,6 +83,8 @@ static void user_fence_worker(struct work_struct *w)
> > >   			XE_WARN_ON("Copy to user failed");
> > >   		kthread_unuse_mm(ufence->mm);
> > >   		mmput(ufence->mm);
> > > +	} else {
> > > +		XE_WARN_ON("mmget_not_zero() failed");
> > >   	}
> > >   	wake_up_all(&ufence->xe->ufence_wq);
> > > -- 
> > > 2.46.0
> > > 
>