xiaomi-sm8450/android_kernel_xiaomi_sm8450
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

drm/i915; Preallocate the lazy request

Browse Source 
It is possible for us to be forced to perform an allocation for the lazy
request whilst running the shrinker. This allocation may fail, leaving
us unable to reclaim any memory leading to premature OOM. A neat
solution to the problem is to preallocate the request at the same time
as acquiring the seqno for the ring transaction. This means that we can
report ENOMEM prior to touching the rings.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This commit is contained in:
Chris Wilson
2013-09-04 10:45:52 +01:00
committed by Daniel Vetter
parent 1823521d2b
commit 3c0e234c84
3 changed files with 15 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
drivers/gpu/drm/i915/i915_gem.c
View File

@@ -2041,8 +2041,8 @@ int __i915_add_request(struct intel_ring_buffer *ring,
if (ret)
return ret;
request = kmalloc(sizeof(*request), GFP_KERNEL);
if (request == NULL)
request = ring->preallocated_lazy_request;
if (WARN_ON(request == NULL))
return -ENOMEM;
/* Record the position of the start of the request so that
@@ -2053,10 +2053,8 @@ int __i915_add_request(struct intel_ring_buffer *ring,
request_ring_position = intel_ring_get_tail(ring);
ret = ring->add_request(ring);
if (ret) {
kfree(request);
if (ret)
return ret;
}
request->seqno = intel_ring_get_seqno(ring);
request->ring = ring;
@@ -2095,6 +2093,7 @@ int __i915_add_request(struct intel_ring_buffer *ring,
trace_i915_gem_request_add(ring, request->seqno);
ring->outstanding_lazy_seqno = 0;
ring->preallocated_lazy_request = NULL;
if (!dev_priv->ums.mm_suspended) {
i915_queue_hangcheck(ring->dev);