On Thu, Mar 19, 2015 at 09:36:14AM +, Tvrtko Ursulin wrote:
Oh right, I got it, but not sure I like it that much. I don't think
batch pool implementation is well enough decoupled from the users.
batch pool Splitting it this way actually improves decoupling.
Well in a way at least where
On 03/18/2015 05:33 PM, Tvrtko Ursulin wrote:
On 03/18/2015 05:27 PM, Chris Wilson wrote:
On Wed, Mar 18, 2015 at 04:51:58PM +, Tvrtko Ursulin wrote:
On 03/09/2015 09:55 AM, Chris Wilson wrote:
diff --git a/drivers/gpu/drm/i915/i915_gem_batch_pool.c
On 03/19/2015 10:06 AM, Chris Wilson wrote:
On Thu, Mar 19, 2015 at 09:36:14AM +, Tvrtko Ursulin wrote:
Oh right, I got it, but not sure I like it that much. I don't think
batch pool implementation is well enough decoupled from the users.
batch pool Splitting it this way actually improves
On Thu, Mar 19, 2015 at 11:39:16AM +, Tvrtko Ursulin wrote:
On 03/19/2015 10:06 AM, Chris Wilson wrote:
On Thu, Mar 19, 2015 at 09:36:14AM +, Tvrtko Ursulin wrote:
Well in a way at least where when we talk about LRU ordering, it
depends on retiring working properly and that is not
On 03/19/2015 11:46 AM, Chris Wilson wrote:
On Thu, Mar 19, 2015 at 11:39:16AM +, Tvrtko Ursulin wrote:
On 03/19/2015 10:06 AM, Chris Wilson wrote:
On Thu, Mar 19, 2015 at 09:36:14AM +, Tvrtko Ursulin wrote:
Well in a way at least where when we talk about LRU ordering, it
depends on
On Thu, Mar 19, 2015 at 11:58:17AM +, Tvrtko Ursulin wrote:
How about retire all rings and then the inactive batch search with a
global pool becomes only O(num_rings) at worst? Might be worth
saving memory resource (multiple pools) vs. trivial traversal like
that?
There isn't a memory
On 03/19/2015 12:04 PM, Chris Wilson wrote:
On Thu, Mar 19, 2015 at 11:58:17AM +, Tvrtko Ursulin wrote:
How about retire all rings and then the inactive batch search with a
global pool becomes only O(num_rings) at worst? Might be worth
saving memory resource (multiple pools) vs. trivial
On Thu, Mar 19, 2015 at 02:01:37PM +, Tvrtko Ursulin wrote:
On 03/19/2015 12:04 PM, Chris Wilson wrote:
On Thu, Mar 19, 2015 at 11:58:17AM +, Tvrtko Ursulin wrote:
How about retire all rings and then the inactive batch search with a
global pool becomes only O(num_rings) at worst?
On 03/09/2015 09:55 AM, Chris Wilson wrote:
I woke up one morning and found 50k objects sitting in the batch pool
and every search seemed to iterate the entire list... Painting the
screen in oils would provide a more fluid display.
One issue with the current design is that we only check for
On Wed, Mar 18, 2015 at 04:51:58PM +, Tvrtko Ursulin wrote:
On 03/09/2015 09:55 AM, Chris Wilson wrote:
I woke up one morning and found 50k objects sitting in the batch pool
and every search seemed to iterate the entire list... Painting the
screen in oils would provide a more fluid
On 03/18/2015 05:27 PM, Chris Wilson wrote:
On Wed, Mar 18, 2015 at 04:51:58PM +, Tvrtko Ursulin wrote:
On 03/09/2015 09:55 AM, Chris Wilson wrote:
diff --git a/drivers/gpu/drm/i915/i915_gem_batch_pool.c
b/drivers/gpu/drm/i915/i915_gem_batch_pool.c
index 21f3356cc0ab..1287abf55b84 100644
I woke up one morning and found 50k objects sitting in the batch pool
and every search seemed to iterate the entire list... Painting the
screen in oils would provide a more fluid display.
One issue with the current design is that we only check for retirements
on the current ring when preparing to
12 matches
Mail list logo