On Sat 12 Sep 2015, Chris Wilson wrote:
> On Thu, Sep 10, 2015 at 08:41:37AM +0100, Chris Wilson wrote:
> > http://people.freedesktop.org/~ickle/vbo/
> >
> > ministat 63c4b7e_bench_synmark:OglBatch7 2fbce8b_bench_synmark:OglBatch7
> > x 63c4b7e_bench_synmark:OglBatch7
> > + 2fbce8b_bench_synmark:O
On Thu, Sep 10, 2015 at 08:41:37AM +0100, Chris Wilson wrote:
> http://people.freedesktop.org/~ickle/vbo/
>
> ministat 63c4b7e_bench_synmark:OglBatch7 2fbce8b_bench_synmark:OglBatch7
> x 63c4b7e_bench_synmark:OglBatch7
> + 2fbce8b_bench_synmark:OglBatch7
> +
Hi Chris,
Instead of the nop functions, could you just do:
if (vbo_context(ctx)->resolve)
vbo_context(ctx)->resolve(ctx);
Marek
On Wed, Sep 9, 2015 at 3:38 PM, Chris Wilson wrote:
> A common problem with using HiZ and multisampling is that surfaces need
> to resolved prior to use. Currently
On Wed, Sep 09, 2015 at 10:19:10AM -0700, Ian Romanick wrote:
> On 09/09/2015 10:10 AM, Kenneth Graunke wrote:
> > On Wednesday, September 09, 2015 02:38:56 PM Chris Wilson wrote:
> >> A common problem with using HiZ and multisampling is that surfaces need
> >> to resolved prior to use. Currently i
On Wednesday, September 09, 2015 10:19:10 AM Ian Romanick wrote:
> On 09/09/2015 10:10 AM, Kenneth Graunke wrote:
> > On Wednesday, September 09, 2015 02:38:56 PM Chris Wilson wrote:
> >> A common problem with using HiZ and multisampling is that surfaces need
> >> to resolved prior to use. Currentl
On 09/09/2015 10:10 AM, Kenneth Graunke wrote:
> On Wednesday, September 09, 2015 02:38:56 PM Chris Wilson wrote:
>> A common problem with using HiZ and multisampling is that surfaces need
>> to resolved prior to use. Currently i965 does this inside its state
>> update hook, but that is a comparati
On Wednesday, September 09, 2015 02:38:56 PM Chris Wilson wrote:
> A common problem with using HiZ and multisampling is that surfaces need
> to resolved prior to use. Currently i965 does this inside its state
> update hook, but that is a comparatively heavyweight operation that need
> not be perfor
A common problem with using HiZ and multisampling is that surfaces need
to resolved prior to use. Currently i965 does this inside its state
update hook, but that is a comparatively heavyweight operation that need
not be performed so frequently. The obvious solution (and therefore
fraught with drago
