[Bug 90114] [SNB+ Bisected]Ogles3conform ES3-CTS.shaders.struct.uniform.sampler_array_fragment fails

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Thu May 7 23:15:42 PDT 2015


https://bugs.freedesktop.org/show_bug.cgi?id=90114

--- Comment #13 from Tapani Pälli <lemody at gmail.com> ---
(In reply to Jason Ekstrand from comment #11)
> (In reply to Tapani Pälli from comment #10)
> > Created attachment 115601 [details] [review] [review]
> > fix for the issue
> > 
> > here's another try for the fix, for me this passes all of piglit and the
> > attached 'complex' case
> 
> I'm still not convinced that's correct, but it made me look through the code
> again and I think I know what's going wrong.  On line 81 in the direct array
> access case we increment sampler_index by deref->base_offset *and* add
> "[%u]" to the string description with deref->base_offset.  Then
> sampler_index is incremented by the return value of get_sampler_index which
> should take the array offset into account *again*.  Maybe we just need to
> remove the "instr->sampler_index += deref->base_offset" line to get rid of
> the double-counting.

Ah right, that's what the first (workaround patch) causes and the second patch
too but by accident. It looks to me that deref_array->base_offset is required
*only* in the case of having array of samplers (), otherwise it should be not
taken to account. I will test this and set a patch.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-3d-bugs/attachments/20150508/fd1ea027/attachment-0001.html>


More information about the intel-3d-bugs mailing list