Details
-
Bug
-
Resolution: Won't Do
-
P2: Important
-
None
-
6.1.0 Alpha, 6.2
-
None
-
Linux QEMU (gcc-armv7) on Linux Ubuntu_20_04 (gcc-x86_64)
Description
We've seen lots of integration failures from tst_QGlobalStatic. One example fails with:
agent:2021/02/24 10:36:41 build.go:379: 55: Config: Using QtTest library 6.2.0, Qt 6.2.0 (arm-little_endian-ilp32-eabi-hardfloat shared (dynamic) release build; by GCC 10.2.0), b2qt 3.2 agent:2021/02/24 10:36:41 build.go:379: 55: PASS : tst_QGlobalStatic::initTestCase() agent:2021/02/24 10:36:41 build.go:379: 55: PASS : tst_QGlobalStatic::beforeInitialization() agent:2021/02/24 10:36:41 build.go:379: 55: PASS : tst_QGlobalStatic::api() agent:2021/02/24 10:36:41 build.go:379: 55: PASS : tst_QGlobalStatic::constVolatile() agent:2021/02/24 10:36:41 build.go:379: 55: PASS : tst_QGlobalStatic::exception() agent:2021/02/24 10:36:41 build.go:379: 55: PASS : tst_QGlobalStatic::catchExceptionAndRetry() agent:2021/02/24 10:36:42 build.go:379: 55: realloc(): invalid next size agent:2021/02/24 10:51:42 build.go:239: Virtual Memory Total: 15734611968, Free:8296857600, UsedPercent:8.514319% [...] agent:2021/02/24 10:51:42 build.go:438: Killed process: No output received (timeout: 900 seconds)
The next test after catchExceptionAndRetry() is threadStressTest(); it's not entirely clear where the failure is happening, but the latter seems the more likely (if only given its name) to be the problem.
Attachments
For Gerrit Dashboard: QTBUG-91423 | ||||||
---|---|---|---|---|---|---|
# | Subject | Branch | Project | Status | CR | V |
337406,5 | Skip globalstatic threadStressTest on QEMU | dev | qt/qtbase | Status: MERGED | +2 | 0 |
337760,2 | Skip globalstatic threadStressTest on QEMU | 6.1 | qt/qtbase | Status: MERGED | +2 | 0 |