| Message ID | 1479324769-1532-1-git-send-email-matthew.auld@intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Wed, Nov 16, 2016 at 07:32:49PM +0000, Matthew Auld wrote: > We need to clean up the global_timeline in i915_gem_load_cleanup. > > Cc: Chris Wilson <chris@chris-wilson.co.uk> > Signed-off-by: Matthew Auld <matthew.auld@intel.com> > --- > drivers/gpu/drm/i915/i915_gem.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c > index 3fb5e66..c440e72 100644 > --- a/drivers/gpu/drm/i915/i915_gem.c > +++ b/drivers/gpu/drm/i915/i915_gem.c > @@ -4557,6 +4557,8 @@ void i915_gem_load_cleanup(struct drm_device *dev) > > WARN_ON(!llist_empty(&dev_priv->mm.free_list)); > > + i915_gem_timeline_fini(&dev_priv->gt.global_timeline); Once upon a time the global timeline didn't allocate. However, to free the timeline I assert we hold the struct_mutex for manipulatingn the timelines list (that's important for the client timelines) and we don't hold the struct_mutex here. Give it a mutex_lock() and a WARN_ON(!list_empty(&dev_priv->gt.timelines)); -Chris
On Wed, Nov 16, 2016 at 07:32:49PM +0000, Matthew Auld wrote: > We need to clean up the global_timeline in i915_gem_load_cleanup. > > Cc: Chris Wilson <chris@chris-wilson.co.uk> > Signed-off-by: Matthew Auld <matthew.auld@intel.com> Fixes: line probably missing? And why was that not caught when the original patch landed? -Daniel > --- > drivers/gpu/drm/i915/i915_gem.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c > index 3fb5e66..c440e72 100644 > --- a/drivers/gpu/drm/i915/i915_gem.c > +++ b/drivers/gpu/drm/i915/i915_gem.c > @@ -4557,6 +4557,8 @@ void i915_gem_load_cleanup(struct drm_device *dev) > > WARN_ON(!llist_empty(&dev_priv->mm.free_list)); > > + i915_gem_timeline_fini(&dev_priv->gt.global_timeline); > + > kmem_cache_destroy(dev_priv->dependencies); > kmem_cache_destroy(dev_priv->requests); > kmem_cache_destroy(dev_priv->vmas); > -- > 2.7.4 > > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/intel-gfx
On Thu, Nov 17, 2016 at 09:48:18AM +0100, Daniel Vetter wrote: > On Wed, Nov 16, 2016 at 07:32:49PM +0000, Matthew Auld wrote: > > We need to clean up the global_timeline in i915_gem_load_cleanup. > > > > Cc: Chris Wilson <chris@chris-wilson.co.uk> > > Signed-off-by: Matthew Auld <matthew.auld@intel.com> > > Fixes: line probably missing? And why was that not caught when the > original patch landed? It's a just a missing free of a few bytes. This would require kmemleak. -Chris
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 3fb5e66..c440e72 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -4557,6 +4557,8 @@ void i915_gem_load_cleanup(struct drm_device *dev) WARN_ON(!llist_empty(&dev_priv->mm.free_list)); + i915_gem_timeline_fini(&dev_priv->gt.global_timeline); + kmem_cache_destroy(dev_priv->dependencies); kmem_cache_destroy(dev_priv->requests); kmem_cache_destroy(dev_priv->vmas);
We need to clean up the global_timeline in i915_gem_load_cleanup. Cc: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Matthew Auld <matthew.auld@intel.com> --- drivers/gpu/drm/i915/i915_gem.c | 2 ++ 1 file changed, 2 insertions(+)