Bug 107448

Summary: [r600g] 5 of 30 piglit sanity tests fail (PCIe, ppc64, mesa-18.3.4)
Product: Mesa Reporter: erhard_f
Component: Drivers/Gallium/r600Assignee: Default DRI bug account <dri-devel>
Status: RESOLVED MOVED QA Contact: Default DRI bug account <dri-devel>
Severity: normal    
Priority: medium    
Version: unspecified   
Hardware: PowerPC   
OS: Linux (All)   
See Also: https://bugs.freedesktop.org/show_bug.cgi?id=107470
Whiteboard:
i915 platform: i915 features:
Attachments: results from 'pigllit run sanity' (18.1.5)
html-summary from 'pigllit run sanity' (18.1.5)
html-summary from 'pigllit run sanity' (18.0.5)
html-summary from 'pigllit run sanity' (18.3.4)
results from 'pigllit run sanity' (18.3.4)

Description erhard_f 2018-08-01 18:36:08 UTC
Created attachment 140928 [details]
results from 'pigllit run sanity' (18.1.5)

Machine is a PowerMac G5 11,2 + Radeon HD 6450.

Command
/root/build/piglit/bin/shader_runner /root/build/piglit/test/spec/arb_tessellation_shader/execution/sanity.shader_test -auto -fbo

Stdout 	
Probe color at (0,0)
  Expected: 0 255 0 255
  Observed: 25 25 25 25
Test failure on line 57

Stderr 	
Mesa: User error: GL_INVALID_ENUM in lGetFramebufferAttachmentParameteriv(invalid attachment GL_NONE)


Command
/root/build/piglit/bin/glsl-1.50-transform-feedback-builtins -auto -fbo

Stdout 	
Incorrect data for 'gl_Layer' output 0.  Expected 1, got 16777216.
Incorrect data for 'gl_Layer' output 1.  Expected 3, got 50331648.
Incorrect data for 'gl_Layer' output 2.  Expected 5, got 83886080.
Incorrect data for 'gl_ViewportIndex' output 0.  Expected 1, got 16777216.
Incorrect data for 'gl_ViewportIndex' output 1.  Expected 2, got 33554432.
Incorrect data for 'gl_ViewportIndex' output 2.  Expected 3, got 50331648.
Comment 1 erhard_f 2018-08-01 18:37:09 UTC
Created attachment 140929 [details]
html-summary from 'pigllit run sanity' (18.1.5)
Comment 2 Ilia Mirkin 2018-08-01 18:51:26 UTC
Observation:

bswap32(16777216) == 1
bswap32(50331648) == 3
bswap32(83886080) == 5

For the second test, probably some TF thing that needs to either be configured for BE or get byte-swapped somewhere.
Comment 3 erhard_f 2018-08-01 18:56:43 UTC
Hm, so you think the sanity test itself got broken and not a r600g component? I am no expert, sorry.
Comment 4 Ilia Mirkin 2018-08-01 19:01:50 UTC
(In reply to erhard_f from comment #3)
> Hm, so you think the sanity test itself got broken and not a r600g
> component? I am no expert, sorry.

No clue. Just an observation for someone who will maybe at one point look into it.
Comment 5 erhard_f 2018-08-03 12:50:17 UTC
Created attachment 140950 [details]
html-summary from 'pigllit run sanity' (18.0.5)

Same test failures with mesa 18.0.5. So I guess it's piglit itself failing, as at the time I did first piglit tests with 18.0.x on ppc64 (see bug #105463) the sanity tests passed.
Comment 6 erhard_f 2018-10-19 20:05:54 UTC
Meanwhile, on another machine (Talos II + Radeon HD 6450), another mesa (18.2.2) and rebuilt piglit from git-master. But same test-failures here.
Comment 7 erhard_f 2019-03-10 17:11:06 UTC
Created attachment 143610 [details]
html-summary from 'pigllit run sanity' (18.3.4)
Comment 8 erhard_f 2019-03-10 17:23:37 UTC
Created attachment 143611 [details]
results from 'pigllit run sanity' (18.3.4)

As mesa progresses, piglit sanity tests error count on ppc64 progresses. ;)

Command 	
/root/build/piglit/bin/shader_runner /root/build/piglit/tests/spec/arb_tessellation_shader/execution/sanity.shader_test -auto -fboStderr 	
Stderr
Mesa: User error: GL_INVALID_ENUM in glGetFramebufferAttachmentParameteriv(invalid attachment GL_NONE)
Stdout 	
Probe color at (0,0)
  Expected: 0 255 0 255
  Observed: 25 25 25 25
Test failure on line 57

Command 	
/root/build/piglit/bin/shader_runner /root/build/piglit/generated_tests/spec/arb_tessellation_shader/execution/built-in-functions/tcs-any-bvec4-using-if.shader_test -auto -fbo
Stderr 	
Mesa: User error: GL_INVALID_ENUM in glGetFramebufferAttachmentParameteriv(invalid attachment GL_NONE)
Stdout 	
Probe color at (0,0)
  Expected: 0 0 255 255
  Observed: 0 0 255 0
Test failure on line 92
Probe color at (4,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 93
Probe color at (8,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 94
Probe color at (12,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 95
Probe color at (16,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 96
Probe color at (20,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 97
Probe color at (24,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 98
Probe color at (28,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 99
Probe color at (32,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 100
Probe color at (36,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 101
Probe color at (40,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 102
Probe color at (44,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 103
Probe color at (48,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 104
Probe color at (52,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 105
Probe color at (56,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 106
Probe color at (60,0)
  Expected: 1.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 107

Command 	
/root/build/piglit/bin/shader_runner /root/build/piglit/tests/spec/arb_tessellation_shader/execution/barrier-patch.shader_test -auto -fbo
Stderr 	
Mesa: User error: GL_INVALID_ENUM in glGetFramebufferAttachmentParameteriv(invalid attachment GL_NONE)
Stdout 	
Probe color at (0,0)
  Expected: 0 255 0 255
  Observed: 0 0 0 255
Test failure on line 79
Probe color at (0,0)
  Expected: 0 255 0 255
  Observed: 0 0 0 255
Test failure on line 86

Command 	
/root/build/piglit/bin/glsl-1.50-transform-feedback-builtins -auto -fbo
Stderr
Stdout 	
Incorrect data for 'gl_Layer' output 0.  Expected 1, got 16777216.
Incorrect data for 'gl_Layer' output 1.  Expected 3, got 50331648.
Incorrect data for 'gl_Layer' output 2.  Expected 5, got 83886080.
Incorrect data for 'gl_ViewportIndex' output 0.  Expected 1, got 16777216.
Incorrect data for 'gl_ViewportIndex' output 1.  Expected 2, got 33554432.
Incorrect data for 'gl_ViewportIndex' output 2.  Expected 3, got 50331648.

Command 	
/root/build/piglit/bin/shader_runner /root/build/piglit/generated_tests/spec/amd_shader_trinary_minmax/execution/built-in-functions/tcs-max3-vec4-vec4-vec4.shader_test -auto -fbo
Stderr 	
Mesa: User error: GL_INVALID_ENUM in glGetFramebufferAttachmentParameteriv(invalid attachment GL_NONE)Stdout 	
Probe color at (0,0)
  Expected: 0 255 0 255
  Observed: 0 0 255 0
Test failure on line 160
Probe color at (4,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 161
Probe color at (8,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 162
Probe color at (12,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 163
Probe color at (16,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 164
Probe color at (20,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 165
Probe color at (24,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 166
Probe color at (28,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 167
Probe color at (32,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 168
Probe color at (36,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 169
Probe color at (40,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 170
Probe color at (44,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 171
Probe color at (48,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 172
Probe color at (52,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 173
Probe color at (56,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 174
Probe color at (60,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 0.000000 0.000000 1.000000 0.000000
Test failure on line 175
Comment 9 erhard_f 2019-03-10 17:36:23 UTC
According to on Comment 1 on bug #107470 this is most probably not a piglit bug itself, but a driver bug.
Comment 10 GitLab Migration User 2019-09-18 19:26:09 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/mesa/mesa/issues/642.

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.