| Summary: | AB-INT PTEST: valgrind memcheck/tests/linux/timerfd-syscall ptest intermittent failure | ||
|---|---|---|---|
| Product: | [QA/Testing] Package Testing (ptest) | Reporter: | Köry Maincent <kory.maincent> |
| Component: | ptest | Assignee: | Tony Tascioglu <tony.tascioglu> |
| Status: | RESOLVED INVALID | QA Contact: | |
| Severity: | normal | ||
| Priority: | Medium+ | CC: | alexandre.belloni, bluelightning, chee.yang.lee, flowergom, jon.mason, naveen.kumar.saini, oobitots, randy.macleod, richard.purdie, tony.tascioglu |
| Version: | unspecified | ||
| Target Milestone: | 3.4 M3 | ||
| Hardware: | x86 | ||
| OS: | Multiple | ||
| Whiteboard: | AB-INT AB-INT-TMPFS | ||
| OS type for building Yocto: | --- | Type of Regression: | --- |
| Verified: | Documentation change: | No (bug/feature does not impact docs) | |
|
Description
Köry Maincent
2021-03-09 14:35:25 UTC
This was on ubuntu1804-arm-1 https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1556/steps/12/logs/stdio https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1548/steps/12/logs/stdio I can't reproduce this failure I think its due to the load/IO issue version: 3.16.1 === Test Summary === TOTAL: 726 PASSED: 686 FAILED: 0 SKIPPED: 40 DURATION: 1278 leaving it open for now... Until now, this only failed on ubuntu1804-arm-1 this seems to happening frequently on arm and not that intermittent in nature Another instance: https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1664/steps/12/logs/stdio Another one: https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1671/steps/12/logs/stdio This last onesA were on ubunt1804-arm-1. and the last one did contain the runqemu in tmpfs patch fixed(more a workaround), see previous commit. Sorry wrong bugzilla link.. reopening although I did submit a patch https://git.openembedded.org/openembedded-core/commit/?id=099313ef541920d4a84b801d9d8788a56ba7ec61 to print out the diff from expected result. Sorry wrong bugzilla link.. reopening although I did submit a patch https://git.openembedded.org/openembedded-core/commit/?id=099313ef541920d4a84b801d9d8788a56ba7ec61 to print out the diff from expected result. assigning it back to randy, if it happens again, we can get the test output details file at https://autobuilder.yocto.io/pub/non-release/ and make a judgement if the test needs to be disabled for timing/load reasons. Seen additional failures:
AssertionError: Failed ptests:
{'valgrind': ['drd/tests/bar_bad']}
on master (qemux86-64):
woerker: ubuntu1604-ty-1
https://autobuilder.yoctoproject.org/typhoon/#/builders/81/builds/2005
On master-next:
worker:fedora32-ty-1
https://autobuilder.yoctoproject.org/typhoon/#/builders/81/builds/2006
https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1723/steps/12/logs/stdio qemuarm64 master on ubuntu1804-arm-1 'valgrind': ['memcheck/tests/leak_cpp_interior'] https://autobuilder.yoctoproject.org/typhoon/#/builders/81/builds/2022/steps/12/logs/stdio qemuarm64 master on ubuntu1804-arm-1 leak_cpp_interior is now disabled by 984ffe3ab4b7857b6201494cd77f2261bfc6d050 happen again qemuarm64-ptest worker : ubuntu1804-arm-1 branch_poky : master-next https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1777 https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1799/steps/12/logs/stdio qemuarm64-ptest ubuntu1804-arm-1 Reoccurence https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1818/steps/12/logs/stdio qemuarm64-ptest ubuntu1804-arm-1 https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1840/steps/12/logs/stdio {'valgrind': ['memcheck/tests/linux/stack_changes']} ubuntu1804-arm-1 qemuarm64-ptest {'valgrind': ['memcheck/tests/linux/stack_changes']}
https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1843/steps/12/logs/stdio
qemuarm64-ptest
ubuntu2004-arm-1
Happend again: https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1853/steps/12/logs/stdio 'valgrind': ['memcheck/tests/linux/stack_changes'] qemuarm64-ptest Happend again: https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1852/steps/12/logs/stdio Happend again: https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1850/steps/12/logs/stdio Happend again: https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1871/steps/12/logs/stdio Happend again: https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1872/steps/12/logs/stdio https://autobuilder.yoctoproject.org/typhoon/#/builders/81/builds/2161/steps/12/logs/stdio Hello, I want to update this thread with a few findings. First, there are several different tests failing in the autobuilder logs in this thread, I'm not sure if it's due to the same underlying cause between these yet. Furthermore, some of these are failures are on x86_64 and some are on arm64. I don't think it's the same SMP issue that was affecting the helgrind tests, though it's harder to tell without the diffs or logs from the autobuilder. I tried to recreate this error by running stack_changes, timerfd-syscall and fb_test_amd64 in Qemu x86_64 in parallel with SMP enabled for over 12 hours overnight and neither of the tests failed. Arm64 on the other hand has several issues. Testing on qemu_arm64 from an x86_64 host without SMP and with 1G of RAM allocated, stack_changes failed all 10/10 times I ran it on master, and several of the other tests fail randomly a lot more frequently. I am going to try on older versions as well to check if it is a recent change. As discussed, the memcheck/tests/linux/stack_changes issue is not intermittent and I split it ou in bug 14430 This was catch-all bug with several failures grouped together. We've created more-specific targeted bugs. The issues covered here are either covered by those bugs or no longer occurring. |