Bug 103668

Summary: mesa-17.3.0_rc3: FAIL: glsl/tests/cache-test
Product: Mesa Reporter: erhard_f
Component: Mesa coreAssignee: Matt Turner <mattst88>
Status: RESOLVED FIXED QA Contact: mesa-dev
Severity: normal    
Priority: medium    
Version: 17.3   
Hardware: PowerPC   
OS: All   
Whiteboard:
i915 platform: i915 features:
Attachments: build.log

Description erhard_f 2017-11-10 13:11:08 UTC
Created attachment 135372 [details]
build.log

==================================================
   Mesa 17.3.0-rc3: src/compiler/test-suite.log
==================================================

# TOTAL: 10
# PASS:  9
# SKIP:  0
# XFAIL: 0
# FAIL:  1
# XPASS: 0
# ERROR: 0

.. contents:: :depth: 2

FAIL: glsl/tests/cache-test
===========================

Failed to create ./cache-test-tmp/xdg-cache-home for shader cache (No such file or directory)---disabling.
Failed to create ./cache-test-tmp/mesa-glsl-cache-dir for shader cache (No such file or directory)---disabling.
Error: Test 'put_key of a colliding key evicts from the cache' failed: Expected=0, Actual=1
Error: Test 'put_key of orginal key evicts the colliding key' failed: Expected=0, Actual=1
FAIL glsl/tests/cache-test (exit status: 1)
Comment 1 Eric Engestrom 2017-11-10 13:42:13 UTC
(In reply to erhard_f from comment #0)
> Failed to create ./cache-test-tmp/xdg-cache-home for shader cache (No such file or directory)---disabling.
> Failed to create ./cache-test-tmp/mesa-glsl-cache-dir for shader cache (No such file or directory)---disabling.

just fyi, iirc these two messages are expected, they're not part of the problem here

don't know that codebase enough to actually be of any help, though
Comment 2 Emil Velikov 2017-11-10 14:14:24 UTC
I'm suspecting something Gentoo related - either layout or FS.
We have a similar issue in bug 103122, again Gentoo.

Issue may not be as widespread as it seems - 17.3.0-rc3 is fine on pq (Pekka) setup.
Comment 3 Ilia Mirkin 2017-11-10 14:23:53 UTC
That's pretty different... the other issue is about disk management, and probably restricted by gentoo's auto-sandbox thing somehow.

This is PPC, so more likely an endian issue in the cache key computation code. Or a problem in the test.
Comment 4 Emil Velikov 2017-11-10 14:53:34 UTC
Right my bad. I missed the PPC piece in the report :-\
Comment 5 Matt Turner 2017-11-23 19:08:53 UTC
Patches on the list.

      util: Fix SHA1 implementation on big endian
      util: Add a SHA1 unit test program
      util: Fix disk_cache index calculation on big endian
Comment 6 Matt Turner 2017-11-26 05:07:33 UTC
Should be fixed by:

commit c690a7a8cdfb6425547bbb782020098405851194
Author: Matt Turner <mattst88@gmail.com>
Date:   Thu Nov 23 10:41:34 2017 -0800

    util: Fix disk_cache index calculation on big endian
Comment 7 erhard_f 2017-11-26 15:32:45 UTC
I can confirm that the test passes now. Thanks!

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.