https://bugs.freedesktop.org/show_bug.cgi?id=109532
Bug ID: 109532
Summary: ir_variable has maximum access out of bounds -- but
it's not out of bounds
Product: Mesa
Version: git
Hardware: Other
OS: All
Status: NEW
Severity: normal
Priority: medium
Component: glsl-compiler
Assignee: mesa-dev@lists.freedesktop.org
Reporter: imir...@alum.mit.edu
QA Contact: intel-3d-b...@lists.freedesktop.org
Created attachment 143274
--> https://bugs.freedesktop.org/attachment.cgi?id=143274&action=edit
gdb session + generated shaders
This dEQP test asserts, at least on nouveau, but I don't see what would cause
it to be driver-specific.
dEQP-GLES31.functional.ssbo.layout.random.all_per_block_buffers.18
This appears to happen because it doesn't understand the interface array index
properly? This message is printed before the abort:
ir_variable has maximum access out of bounds (1 vs 0)
For some reason it thinks that the array length is 1 when it should be 2. I
recall interfaces being treated rather differently than most things, so perhaps
related.
--
You are receiving this mail because:
You are the assignee for the bug._______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
