diff options
| author | Chris Wilson <chris@chris-wilson.co.uk> | 2011-12-12 14:10:57 +0000 |
|---|---|---|
| committer | Chris Wilson <chris@chris-wilson.co.uk> | 2011-12-13 01:38:09 +0000 |
| commit | c481bec356b2e40e66a000dbaaf261bf7aae930d (patch) | |
| tree | ff2d9b5a1a0bdf07e9942d3ff81db58d00db3755 /src/sna/kgem.h | |
| parent | 6c9aa6f9cf8e59ca6aa1866b83690a1de8cfb757 (diff) | |
sna: Experiment with creating the CPU pixmap using an LLC BO
A poor cousin to vmap is to instead allocate snooped bo and use a CPU
mapping for zero-copy uploads into GPU resident memory. For maximum
performance, we still need tiled GPU buffers so CPU bo are only useful
in situations where we are frequently migrating data.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Diffstat (limited to 'src/sna/kgem.h')
| -rw-r--r-- | src/sna/kgem.h | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/src/sna/kgem.h b/src/sna/kgem.h index 0d85f643..2fd5a551 100644 --- a/src/sna/kgem.h +++ b/src/sna/kgem.h @@ -319,6 +319,8 @@ uint32_t kgem_add_reloc(struct kgem *kgem, void *kgem_bo_map(struct kgem *kgem, struct kgem_bo *bo, int prot); void kgem_bo_unmap(struct kgem *kgem, struct kgem_bo *bo); +void *kgem_bo_map__cpu(struct kgem *kgem, struct kgem_bo *bo); +void kgem_bo_unmap__cpu(struct kgem *kgem, struct kgem_bo *bo, void *ptr); uint32_t kgem_bo_flink(struct kgem *kgem, struct kgem_bo *bo); Bool kgem_bo_write(struct kgem *kgem, struct kgem_bo *bo, |