On Mon, 2010-09-13 at 17:52 +0800, Chris Wilson wrote:
On Mon, 13 Sep 2010 15:17:05 +0800, Xiang, Haihao haihao.xi...@intel.com
wrote:
Some little changes:
Add set_tail hook to struct intel_ring_buffer
fix HAS_BSD with a device info flag
Don't export the initialiser of
Introduce intel_init_render_ring_buffer(), intel_init_bsd_ring_buffer
for ring initialization.
Signed-off-by: Xiang, Haihao haihao.xi...@intel.com
Reviewed-by: Chris Wilson ch...@chris-wilson.co.uk
---
drivers/gpu/drm/i915/i915_gem.c | 14 ++
This is prepared for video codec ring buffer on Sandybridge. It is
needed to read/write more than one register to move the tail pointer of
the video codec ring on Sandybridge.
Signed-off-by: Xiang, Haihao haihao.xi...@intel.com
Reviewed-by: Chris Wilson ch...@chris-wilson.co.uk
---
This ring buffer is used for video decoding/encoding on Sandybridge.
Signed-off-by: Xiang, Haihao haihao.xi...@intel.com
Reviewed-by: Chris Wilson ch...@chris-wilson.co.uk
---
drivers/gpu/drm/i915/i915_drv.c |2 +
drivers/gpu/drm/i915/i915_irq.c | 15 +++-
On 2010.09.16 10:43:12 +0800, Xiang, Haihao wrote:
This is prepared for video codec ring buffer on Sandybridge. It is
needed to read/write more than one register to move the tail pointer of
the video codec ring on Sandybridge.
Do we really need new 'set_tail'? Isn't advance_ring used for set
On Thu, 2010-09-16 at 12:21 +0800, Zhenyu Wang wrote:
On 2010.09.16 10:43:12 +0800, Xiang, Haihao wrote:
This is prepared for video codec ring buffer on Sandybridge. It is
needed to read/write more than one register to move the tail pointer of
the video codec ring on Sandybridge.
Do we
On 2010.09.16 13:10:29 +0800, Xiang, Haihao wrote:
Or can't that be done in init function by using advance_ring?
advance_ring uses ring-tail to set TAIL register, i965_reset() also
invokes ring-init() to re-init ring buffer, how to guarantee ring-tail
is 0? BTW advance_ring can be implemented
