Message ID | 1372709154-7524-2-git-send-email-daniel.vetter@ffwll.ch (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
On Mon, Jul 01, 2013 at 10:05:54PM +0200, Daniel Vetter wrote: > When converting to the preallocated drm_mm_node interfaces in > > commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d > Author: Chris Wilson <chris@chris-wilson.co.uk> > Date: Fri Dec 7 20:37:07 2012 +0000 > > drm/i915: Preallocate the drm_mm_node prior to manipulating the GTT drm_mm > > only the allocation side was converted, but not the freeing. Fix this > up. > > Note that the only difference between put_block and remove_node is > that the former fills up the preallocation cache. Which we don't need > anyway and hence is just wasted space. > > Cc: Chris Wilson <chris@chris-wilson.co.uk> > Cc: Ben Widawsky <ben@bwidawsk.net> > Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> You learn something new every day. Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> -Chris
Hi On Mon, Jul 1, 2013 at 10:05 PM, Daniel Vetter <daniel.vetter@ffwll.ch> wrote: > When converting to the preallocated drm_mm_node interfaces in > > commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d > Author: Chris Wilson <chris@chris-wilson.co.uk> > Date: Fri Dec 7 20:37:07 2012 +0000 > > drm/i915: Preallocate the drm_mm_node prior to manipulating the GTT drm_mm > > only the allocation side was converted, but not the freeing. Fix this > up. > > Note that the only difference between put_block and remove_node is > that the former fills up the preallocation cache. Which we don't need > anyway and hence is just wasted space. > > Cc: Chris Wilson <chris@chris-wilson.co.uk> > Cc: Ben Widawsky <ben@bwidawsk.net> > Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> > --- > drivers/gpu/drm/i915/i915_gem.c | 6 +++--- > 1 file changed, 3 insertions(+), 3 deletions(-) > > diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c > index 4200c32..30fd694 100644 > --- a/drivers/gpu/drm/i915/i915_gem.c > +++ b/drivers/gpu/drm/i915/i915_gem.c > @@ -2620,7 +2620,7 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj) > /* Avoid an unnecessary call to unbind on rebind. */ > obj->map_and_fenceable = true; > > - drm_mm_put_block(obj->gtt_space); > + drm_mm_remove_node(obj->gtt_space); kfree(obj->gtt_space); > obj->gtt_space = NULL; > obj->gtt_offset = 0; > > @@ -3137,14 +3137,14 @@ search_free: > } > if (WARN_ON(!i915_gem_valid_gtt_space(dev, node, obj->cache_level))) { > i915_gem_object_unpin_pages(obj); > - drm_mm_put_block(node); > + drm_mm_remove_node(node); kfree(node); > return -EINVAL; > } > > ret = i915_gem_gtt_prepare_object(obj); > if (ret) { > i915_gem_object_unpin_pages(obj); > - drm_mm_put_block(node); > + drm_mm_remove_node(node); kfree(node); drm_mm_remove_node() does unlink the node but not remove it. Btw., I have these fixes in my series, too. I will push it later and write the git-link to #dri-devel. Cheers David > return ret; > } > > -- > 1.7.11.7 >
On Mon, Jul 01, 2013 at 09:16:56PM +0100, Chris Wilson wrote: > On Mon, Jul 01, 2013 at 10:05:54PM +0200, Daniel Vetter wrote: > > When converting to the preallocated drm_mm_node interfaces in > > > > commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d > > Author: Chris Wilson <chris@chris-wilson.co.uk> > > Date: Fri Dec 7 20:37:07 2012 +0000 > > > > drm/i915: Preallocate the drm_mm_node prior to manipulating the GTT drm_mm > > > > only the allocation side was converted, but not the freeing. Fix this > > up. > > > > Note that the only difference between put_block and remove_node is > > that the former fills up the preallocation cache. Which we don't need > > anyway and hence is just wasted space. > > > > Cc: Chris Wilson <chris@chris-wilson.co.uk> > > Cc: Ben Widawsky <ben@bwidawsk.net> > > Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> > > You learn something new every day. > Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> The lack of kfree in this patch might be a problem ... /me tries again -Daniel
On Mon, Jul 01, 2013 at 10:21:57PM +0200, David Herrmann wrote: > Hi > > On Mon, Jul 1, 2013 at 10:05 PM, Daniel Vetter <daniel.vetter@ffwll.ch> wrote: > > When converting to the preallocated drm_mm_node interfaces in > > > > commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d > > Author: Chris Wilson <chris@chris-wilson.co.uk> > > Date: Fri Dec 7 20:37:07 2012 +0000 > > > > drm/i915: Preallocate the drm_mm_node prior to manipulating the GTT drm_mm > > > > only the allocation side was converted, but not the freeing. Fix this > > up. > > > > Note that the only difference between put_block and remove_node is > > that the former fills up the preallocation cache. Which we don't need > > anyway and hence is just wasted space. > > > > Cc: Chris Wilson <chris@chris-wilson.co.uk> > > Cc: Ben Widawsky <ben@bwidawsk.net> > > Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> > > --- > > drivers/gpu/drm/i915/i915_gem.c | 6 +++--- > > 1 file changed, 3 insertions(+), 3 deletions(-) > > > > diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c > > index 4200c32..30fd694 100644 > > --- a/drivers/gpu/drm/i915/i915_gem.c > > +++ b/drivers/gpu/drm/i915/i915_gem.c > > @@ -2620,7 +2620,7 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj) > > /* Avoid an unnecessary call to unbind on rebind. */ > > obj->map_and_fenceable = true; > > > > - drm_mm_put_block(obj->gtt_space); > > + drm_mm_remove_node(obj->gtt_space); > > kfree(obj->gtt_space); > > > obj->gtt_space = NULL; > > obj->gtt_offset = 0; > > > > @@ -3137,14 +3137,14 @@ search_free: > > } > > if (WARN_ON(!i915_gem_valid_gtt_space(dev, node, obj->cache_level))) { > > i915_gem_object_unpin_pages(obj); > > - drm_mm_put_block(node); > > + drm_mm_remove_node(node); > > kfree(node); > > > return -EINVAL; > > } > > > > ret = i915_gem_gtt_prepare_object(obj); > > if (ret) { > > i915_gem_object_unpin_pages(obj); > > - drm_mm_put_block(node); > > + drm_mm_remove_node(node); > > kfree(node); Yeah, I fail ... > drm_mm_remove_node() does unlink the node but not remove it. Btw., I > have these fixes in my series, too. I will push it later and write the > git-link to #dri-devel. We have patches in-flight to convert over to embedded drm_mm_nodes anyway, so I guess that part will solve itself automatically. -Daniel
On Mon, Jul 01, 2013 at 10:39:03PM +0200, Daniel Vetter wrote: > On Mon, Jul 01, 2013 at 10:21:57PM +0200, David Herrmann wrote: > > Hi > > > > On Mon, Jul 1, 2013 at 10:05 PM, Daniel Vetter <daniel.vetter@ffwll.ch> wrote: > > > When converting to the preallocated drm_mm_node interfaces in > > > > > > commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d > > > Author: Chris Wilson <chris@chris-wilson.co.uk> > > > Date: Fri Dec 7 20:37:07 2012 +0000 > > > > > > drm/i915: Preallocate the drm_mm_node prior to manipulating the GTT drm_mm > > > > > > only the allocation side was converted, but not the freeing. Fix this > > > up. > > > > > > Note that the only difference between put_block and remove_node is > > > that the former fills up the preallocation cache. Which we don't need > > > anyway and hence is just wasted space. > > > > > > Cc: Chris Wilson <chris@chris-wilson.co.uk> > > > Cc: Ben Widawsky <ben@bwidawsk.net> > > > Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> > > > --- > > > drivers/gpu/drm/i915/i915_gem.c | 6 +++--- > > > 1 file changed, 3 insertions(+), 3 deletions(-) > > > > > > diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c > > > index 4200c32..30fd694 100644 > > > --- a/drivers/gpu/drm/i915/i915_gem.c > > > +++ b/drivers/gpu/drm/i915/i915_gem.c > > > @@ -2620,7 +2620,7 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj) > > > /* Avoid an unnecessary call to unbind on rebind. */ > > > obj->map_and_fenceable = true; > > > > > > - drm_mm_put_block(obj->gtt_space); > > > + drm_mm_remove_node(obj->gtt_space); > > > > kfree(obj->gtt_space); > > > > > obj->gtt_space = NULL; > > > obj->gtt_offset = 0; > > > > > > @@ -3137,14 +3137,14 @@ search_free: > > > } > > > if (WARN_ON(!i915_gem_valid_gtt_space(dev, node, obj->cache_level))) { > > > i915_gem_object_unpin_pages(obj); > > > - drm_mm_put_block(node); > > > + drm_mm_remove_node(node); > > > > kfree(node); > > > > > return -EINVAL; > > > } > > > > > > ret = i915_gem_gtt_prepare_object(obj); > > > if (ret) { > > > i915_gem_object_unpin_pages(obj); > > > - drm_mm_put_block(node); > > > + drm_mm_remove_node(node); > > > > kfree(node); > > Yeah, I fail ... > > > drm_mm_remove_node() does unlink the node but not remove it. Btw., I > > have these fixes in my series, too. I will push it later and write the > > git-link to #dri-devel. > > We have patches in-flight to convert over to embedded drm_mm_nodes anyway, > so I guess that part will solve itself automatically. > -Daniel I'm planning to get this out ASAP. I'm a bit confused now though what I actually need to send.
On Mon, Jul 1, 2013 at 10:46 PM, Ben Widawsky <ben@bwidawsk.net> wrote: >> > drm_mm_remove_node() does unlink the node but not remove it. Btw., I >> > have these fixes in my series, too. I will push it later and write the >> > git-link to #dri-devel. >> >> We have patches in-flight to convert over to embedded drm_mm_nodes anyway, >> so I guess that part will solve itself automatically. > > I'm planning to get this out ASAP. I'm a bit confused now though what I > actually need to send. I think for now just the top-down allocator + the create_node stuff for stolen memory. I guess the conversion to embed the gtt_space drm_mm_node will take a bit longer. -Daniel -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 4200c32..30fd694 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -2620,7 +2620,7 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj) /* Avoid an unnecessary call to unbind on rebind. */ obj->map_and_fenceable = true; - drm_mm_put_block(obj->gtt_space); + drm_mm_remove_node(obj->gtt_space); obj->gtt_space = NULL; obj->gtt_offset = 0; @@ -3137,14 +3137,14 @@ search_free: } if (WARN_ON(!i915_gem_valid_gtt_space(dev, node, obj->cache_level))) { i915_gem_object_unpin_pages(obj); - drm_mm_put_block(node); + drm_mm_remove_node(node); return -EINVAL; } ret = i915_gem_gtt_prepare_object(obj); if (ret) { i915_gem_object_unpin_pages(obj); - drm_mm_put_block(node); + drm_mm_remove_node(node); return ret; }
When converting to the preallocated drm_mm_node interfaces in commit dc9dd7a20fde95aa81a8307cde79c2dff9f83f3d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 7 20:37:07 2012 +0000 drm/i915: Preallocate the drm_mm_node prior to manipulating the GTT drm_mm only the allocation side was converted, but not the freeing. Fix this up. Note that the only difference between put_block and remove_node is that the former fills up the preallocation cache. Which we don't need anyway and hence is just wasted space. Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Ben Widawsky <ben@bwidawsk.net> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> --- drivers/gpu/drm/i915/i915_gem.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)