Yeah, looking at it a bit longer, it seems that the unbind behavior
broke (again like Eric suggested).
When you replaced that condition :
unbind = (cctx == NULL && ddraw == NULL && rdraw == NULL);
if (unbind || dri2_dpy->core->bindContext(cctx, ddraw, rdraw)) {
By this :
if (cctx || ddraw || rdraw) {
The bindContext() used to get executed when unbinding a context, now it
reports a BAD_MATCH error.
-
Lionel
On 14/10/16 16:13, Lionel Landwerlin wrote:
Thanks a lot,
The diff is a bit too complicated for me to tell how it's different.
-
Lionel
On 14/10/16 16:11, Emil Velikov wrote:
On 14 October 2016 at 15:58, Lionel Landwerlin
<lionel.g.landwer...@intel.com> wrote:
Hi Emil,
I've not been able to run the deqp gles31 test suite after this commit.
Like Eric suggested I think this is introducing a behavior change.
v2 should not do that, yet it might be missing something. I've
reverted it and will give it another try after checking with deqp.
Thanks
Emil
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
