diff options
| author | Chris Wilson <chris@chris-wilson.co.uk> | 2011-02-10 20:25:51 +0000 |
|---|---|---|
| committer | Chris Wilson <chris@chris-wilson.co.uk> | 2011-02-21 12:59:35 +0000 |
| commit | 8d68a90e225d831a395ba788e425cb717eec1f9a (patch) | |
| tree | 1c46eb19125ebbb205625783dd184e0489231d01 /src/mesa/drivers/dri/i965/brw_wm_surface_state.c | |
| parent | 3f55683927278e57f3ef8a151d15f4cffdc060dc (diff) | |
intel: use pwrite for batch
It's faster. Not only is the memcpy more efficiently performed in the
kernel (making up for the system call overhead), but by not using mmap
we remove the greater overhead of tracking the vma of every batch.
And it means we can read back from the batch buffer without incurring
the cost of a uncached read through the GTT.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Diffstat (limited to 'src/mesa/drivers/dri/i965/brw_wm_surface_state.c')
| -rw-r--r-- | src/mesa/drivers/dri/i965/brw_wm_surface_state.c | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/src/mesa/drivers/dri/i965/brw_wm_surface_state.c b/src/mesa/drivers/dri/i965/brw_wm_surface_state.c index 82da6c4777..1010d9f6f9 100644 --- a/src/mesa/drivers/dri/i965/brw_wm_surface_state.c +++ b/src/mesa/drivers/dri/i965/brw_wm_surface_state.c @@ -224,7 +224,7 @@ brw_update_texture_surface( struct gl_context *ctx, GLuint unit ) } /* Emit relocation to surface contents */ - drm_intel_bo_emit_reloc(brw->intel.batch->buf, + drm_intel_bo_emit_reloc(brw->intel.batch.bo, brw->wm.surf_offset[surf_index] + offsetof(struct brw_surface_state, ss1), intelObj->mt->region->buffer, 0, @@ -268,7 +268,7 @@ brw_create_constant_surface(struct brw_context *brw, * bspec ("Data Cache") says that the data cache does not exist as * a separate cache and is just the sampler cache. */ - drm_intel_bo_emit_reloc(brw->intel.batch->buf, + drm_intel_bo_emit_reloc(brw->intel.batch.bo, (*out_offset + offsetof(struct brw_surface_state, ss1)), bo, 0, @@ -491,7 +491,7 @@ brw_update_renderbuffer_surface(struct brw_context *brw, surf->ss0.writedisable_alpha = !ctx->Color.ColorMask[unit][3]; } - drm_intel_bo_emit_reloc(brw->intel.batch->buf, + drm_intel_bo_emit_reloc(brw->intel.batch.bo, brw->wm.surf_offset[unit] + offsetof(struct brw_surface_state, ss1), region->buffer, |