drm/i915: Pin the object whilst faulting it in
Subsystems (2)
A message that is not in the archive (sent off-list or not yet received): id:[email protected]
PATCH 1/1 drm/i915: Pin the object whilst faulting it in
In order to prevent reaping of the object whilst setting it up to handle the pagefault, we need to mark it as pinned. This has the nice side-effect of eliminating some special cases from the pagefault handler as well!
Signed-off-by Chris Wilson <[email protected]>
--- drivers/gpu/drm/i915/i915_gem.c | 19 +++++++------------ 1 file changed, 7 insertions(+), 12 deletions(-)
drivers/gpu/drm/i915/i915_gem.c +7 -12
@@ -1452,23 +1452,16 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)trace_i915_gem_object_fault (obj, page_offset, true, write);/* Now bind it into the GTT if needed */ -if (!obj->map_and_fenceable) { - ret =i915_gem_object_unbind (obj); -if (ret) - goto unlock; - } -if (!obj->gtt_space) { - ret =i915_gem_object_bind_to_gtt (obj,0 , true, false); -if (ret) - goto unlock; - } + ret =i915_gem_object_pin (obj,0 , true, false); +if (ret) + goto unlock;if (!obj->has_global_gtt_mapping)i915_gem_gtt_bind_object (obj, obj->cache_level); ret =i915_gem_object_get_fence (obj);if (ret) - goto unlock; + goto unpin;if (i915_gem_object_is_inactive (obj))list_move_tail (&obj->mm_list, &dev_priv->mm.inactive_list);
@@ -1481,7 +1474,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)/* Finally, remap it using the new GTT offset */ ret =vm_insert_pfn (vma, (unsigned long )vmf->virtual_address, pfn);if (ret) - goto unlock; + goto unpin; prefault =16 ;if (page_offset + prefault >= obj->base.size >>PAGE_SHIFT )
@@ -1492,6 +1485,8 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)if (vm_insert_pfn (vma, page_offset, pfn))break ; } +unpin: +i915_gem_object_unpin (obj); unlock:mutex_unlock (&dev->struct_mutex); out:
Signature · 1 line
1.7.10.4
Quoted 6 lines · Chris Wilson
In order to prevent reaping of the object whilst setting it up to handle the pagefault, we need to mark it as pinned. This has the nice side-effect of eliminating some special cases from the pagefault handler as well!
Signed-off-by Chris Wilson <[email protected]>
Can you rebase this on top of dinq pls? I don't have a 16 page prefaulter here ... ;-)
-Daniel
Quoted 61 lines
--- drivers/gpu/drm/i915/i915_gem.c | 19 +++++++------------ 1 file changed, 7 insertions(+), 12 deletions(-)drivers/gpu/drm/i915/i915_gem.c +7
-12@@ -1452,23 +1452,16 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)trace_i915_gem_object_fault (obj, page_offset, true, write);/* Now bind it into the GTT if needed */ -if (!obj->map_and_fenceable) { - ret =i915_gem_object_unbind (obj); -if (ret) - goto unlock; - } -if (!obj->gtt_space) { - ret =i915_gem_object_bind_to_gtt (obj,0 , true, false); -if (ret) - goto unlock; - } + ret =i915_gem_object_pin (obj,0 , true, false); +if (ret) + goto unlock;if (!obj->has_global_gtt_mapping)i915_gem_gtt_bind_object (obj, obj->cache_level); ret =i915_gem_object_get_fence (obj);if (ret) - goto unlock; + goto unpin;if (i915_gem_object_is_inactive (obj))list_move_tail (&obj->mm_list, &dev_priv->mm.inactive_list);@@ -1481,7 +1474,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)/* Finally, remap it using the new GTT offset */ ret =vm_insert_pfn (vma, (unsigned long )vmf->virtual_address, pfn);if (ret) - goto unlock; + goto unpin; prefault =16 ;if (page_offset + prefault >= obj->base.size >>PAGE_SHIFT )@@ -1492,6 +1485,8 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)if (vm_insert_pfn (vma, page_offset, pfn))break ; } +unpin: +i915_gem_object_unpin (obj); unlock:mutex_unlock (&dev->struct_mutex); out:Signature · 6 lines
1.7.10.4 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 3 lines
Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch