| Summary: | [HSW Bisected]igt/pc8/i2c fails | ||||||
|---|---|---|---|---|---|---|---|
| Product: | DRI | Reporter: | lu hua <huax.lu> | ||||
| Component: | DRM/Intel | Assignee: | Paulo Zanoni <przanoni> | ||||
| Status: | CLOSED FIXED | QA Contact: | Intel GFX Bugs mailing list <intel-gfx-bugs> | ||||
| Severity: | normal | ||||||
| Priority: | medium | CC: | intel-gfx-bugs, przanoni | ||||
| Version: | unspecified | ||||||
| Hardware: | All | ||||||
| OS: | Linux (All) | ||||||
| Whiteboard: | |||||||
| i915 platform: | i915 features: | ||||||
| Attachments: |
|
||||||
|
Description
lu hua
2013-09-18 02:54:14 UTC
Paulo's feature broken by Paulo's patch ... But the I2C test fails since it was added, I know this. Are you sure this bisect is correct? The failure in line 124 is expected since your machine is not ULT. But we should be getting a "skip" instead of a failure on that case. Can you please run ./pc8 without the "--run-subtest i2c" argument and then paste the full output and the return value here? We should really get a skip instead of a failure. I wonder if igt_fixture/igt_require are not doing the right thing. (In reply to comment #3) > The failure in line 124 is expected since your machine is not ULT. But we > should be getting a "skip" instead of a failure on that case. Run it on ULT, the result is "skip". If run it on desktop, it is fail. output on ULT: Machine is not reaching PC8+ states, please check its configuration. Test requirement not met in function main, file pc8.c:770: Test requirement: (!(setup_environment())) Subtest drm-resources-equal: SKIP Subtest batch: SKIP Subtest i2c: SKIP Subtest stress-test: SKIP Subtest register-compare: SKIP > Can you please run ./pc8 without the "--run-subtest i2c" argument and then > paste the full output and the return value here? We should really get a skip > instead of a failure. Run ./pc8 output: Test assertion failure function get_residency, file pc8.c:124: Failed assertion: rc == sizeof(ret) Subtest drm-resources-equal: FAIL Subtest batch: FAIL Subtest i2c: FAIL Subtest stress-test: FAIL Subtest register-compare: FAIL I just committed "tests/pc8: fix supports_pc8_plus_residencies" to the intel-gpu-tools repository. With this, non-ULT machines should now SKIP instead of fail. The i2c test should still fail on properly-configured ULT machines, but this problem requires a Kernel patch. Fixed. Verified. Just to check: This testcase should fail on HSW ULT. Can you confirm that and if so file a bug report about it? (In reply to comment #8) > Just to check: This testcase should fail on HSW ULT. Can you confirm that > and if so file a bug report about it? It also skip on HSW ULT. ./pc8 output: Machine is not reaching PC8+ states, please check its configuration. Test requirement not met in function main, file pc8.c:770: Test requirement: (!(setup_environment())) Subtest drm-resources-equal: SKIP Subtest batch: SKIP Subtest i2c: SKIP Subtest stress-test: SKIP Subtest register-compare: SKIP (In reply to comment #9) > (In reply to comment #8) > > Just to check: This testcase should fail on HSW ULT. Can you confirm that > > and if so file a bug report about it? > > It also skip on HSW ULT. > ./pc8 > output: > Machine is not reaching PC8+ states, please check its configuration. > Test requirement not met in function main, file pc8.c:770: > Test requirement: (!(setup_environment())) > Subtest drm-resources-equal: SKIP > Subtest batch: SKIP > Subtest i2c: SKIP > Subtest stress-test: SKIP > Subtest register-compare: SKIP That's not good ... Can you please file a new bug report about this? Paulo, I think we need to check for pci ids in the testcase and only skip it for !ULT, but fail if an ULT machine fails to reach pc8. (In reply to comment #10) > (In reply to comment #9) > > (In reply to comment #8) > > > Just to check: This testcase should fail on HSW ULT. Can you confirm that > > > and if so file a bug report about it? > > > > It also skip on HSW ULT. > > ./pc8 > > output: > > Machine is not reaching PC8+ states, please check its configuration. > > Test requirement not met in function main, file pc8.c:770: > > Test requirement: (!(setup_environment())) > > Subtest drm-resources-equal: SKIP > > Subtest batch: SKIP > > Subtest i2c: SKIP > > Subtest stress-test: SKIP > > Subtest register-compare: SKIP > > That's not good ... Can you please file a new bug report about this? Paulo, > I think we need to check for pci ids in the testcase and only skip it for > !ULT, but fail if an ULT machine fails to reach pc8. See my response on bug #69838. I think we can move the discussion to there. Closing old verified. |
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.