[Bug 109532] ir_variable has maximum access out of bounds -- but it's not out of bounds

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=109532

--- Comment #25 from asimiklit <andrey.simiklit at gmail.com> ---
(In reply to asimiklit from comment #23)
> (In reply to Mark Janes from comment #22)
> > The combination of deqp/mesa patches passed CI.  The only failure was 
> > 
> >   piglit.spec.!opengl 1_1.tex-upside-down-miptree
> > 
> > This test is new and was fixed on mesa master.  The fix is probably not in
> > the tested developer branch.
> > 
> > You should contact deqp-external-requests at google.com to make your case that
> > the deqp test is buggy.
> 
> Cool, thanks a lot.
> I will contact them tomorrow)

I don't know, is it make sense to contact deqp-external-requests at google.com
with questions about specs they used to write this test, according to the last
Ilia comment?
I think I rather won't contact them until we have 100% explanation how it
should work.
I have been trying to find some explanation how we should assign binding points
for
the ssbo array instances with packed layout case but
unfortunately didn't find any concrete answer yet.

It is possible that the deqp fix will migrate to mesa depends on answer which
we will find.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-3d-bugs/attachments/20190212/60d05cc3/attachment.html>