| Message ID | 20140320134307.GG4463@nuc-i3427.alporthouse.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
2014-03-20 9:43 GMT-04:00 Chris Wilson <chris@chris-wilson.co.uk>: > On Thu, Mar 20, 2014 at 09:38:17AM -0400, Nikolay Martynov wrote: >> (gdb) list *i915_gem_object_set_cache_level+0x8a >> 0x24e3a is in i915_gem_object_set_cache_level >> (drivers/gpu/drm/i915/i915_gem.c:3147). >> 3142 * crossing memory domains and dying. >> 3143 */ >> 3144 if (HAS_LLC(dev)) >> 3145 return true; >> 3146 >> 3147 if (!drm_mm_node_allocated(gtt_space)) >> 3148 return true; >> 3149 >> 3150 if (list_empty(>t_space->node_list)) >> 3151 return true; >> >> Please let me know if there's anything else I can do. > > Can you please try: > > > diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c > index 13fc490d1f62..4f71125493fd 100644 > --- a/drivers/gpu/drm/i915/i915_gem.c > +++ b/drivers/gpu/drm/i915/i915_gem.c > @@ -3676,7 +3676,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj, > enum i915_cache_level cache_level) > { > struct drm_device *dev = obj->base.dev; > - struct i915_vma *vma; > + struct i915_vma *vma, *next; > int ret; > > if (obj->cache_level == cache_level) > @@ -3687,7 +3687,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj, > return -EBUSY; > } > > - list_for_each_entry(vma, &obj->vma_list, vma_link) { > + list_for_each_entry_safe(vma, next, &obj->vma_list, vma_link) { > if (!i915_gem_valid_gtt_space(dev, &vma->node, cache_level)) { > ret = i915_vma_unbind(vma); > if (ret) > > -- Yes, that seem to help. It didn't freeze anymore in 15 mins I used it. Thanks!
On Thu, Mar 20, 2014 at 07:17:00PM -0400, Nikolay Martynov wrote: > Yes, that seem to help. It didn't freeze anymore in 15 mins I used it. Thanks! Thanks indeed, commit 3f5e0f06a3355a77ace053b4ffc0ac1c413cf2d0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Mar 21 07:40:56 2014 +0000 drm/i915: Fix unsafe loop iteration over vma whilst unbinding them On non-LLC platforms, when changing the cache level of an object, we may need to unbind it so that prefetching across page boundaries does not cross into a different memory domain. This requires us to unbind conflicting vma, but we did so iterating over the objects vma in an unsafe manner (as the list was being modified as we iterated). The regression was introduced in commit 3089c6f239d7d2c4cb2dd5c353e8984cf79af1d7 Author: Ben Widawsky <ben@bwidawsk.net> Date: Wed Jul 31 17:00:03 2013 -0700 drm/i915: make caching operate on all address spaces apparently as far back as v3.12-rc1, but it has only just begun to trigger real world bug reports. Reported-and-tested-by: Nikolay Martynov <mar.kolya@gmail.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=76384 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Ben Widawsky <ben@bwidawsk.net> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Now you can get back to your original bug :( -Chris
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 13fc490d1f62..4f71125493fd 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -3676,7 +3676,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj, enum i915_cache_level cache_level) { struct drm_device *dev = obj->base.dev; - struct i915_vma *vma; + struct i915_vma *vma, *next; int ret; if (obj->cache_level == cache_level) @@ -3687,7 +3687,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj, return -EBUSY; } - list_for_each_entry(vma, &obj->vma_list, vma_link) { + list_for_each_entry_safe(vma, next, &obj->vma_list, vma_link) { if (!i915_gem_valid_gtt_space(dev, &vma->node, cache_level)) { ret = i915_vma_unbind(vma); if (ret)