Details
-
Bug
-
Resolution: Done
-
P1: Critical
-
5.11
-
QEMU armv7 and arm64 on Ubuntu 16.04 in CI
-
caa5a20479bd58fda4380181691f772a1f3c1da4
Description
tst_QObjectPerformance::emitToManyReceivers is flaky in QEMU in CI.
Example failure:
********* Start testing of tst_QObjectPerformance ********* Config: Using QtTest library 5.11.0, Qt 5.11.0 (arm64-little_endian-lp64 shared (dynamic) release build; by GCC 6.3.0) PASS : tst_QObjectPerformance::initTestCase() QDEBUG : tst_QObjectPerformance::emitToManyReceivers() [ 0.001 D] default unknown - 5000 receivers, elapsed time 8 cannot be compared to previous, unmeasurable time QDEBUG : tst_QObjectPerformance::emitToManyReceivers() [ 0.090 D] default unknown - 8000 receivers, elapsed time 4 compared to previous time 8 QDEBUG : tst_QObjectPerformance::emitToManyReceivers() [ 0.175 D] default unknown - 11000 receivers, elapsed time 4 compared to previous time 4 QDEBUG : tst_QObjectPerformance::emitToManyReceivers() [ 0.283 D] default unknown - 14000 receivers, elapsed time 12 compared to previous time 4 FAIL! : tst_QObjectPerformance::emitToManyReceivers() 'double(e) / double(elapsed) <= increaseRatio * 2.0' returned FALSE. () Loc: [tst_qobjectperformance.cpp(97)] PASS : tst_QObjectPerformance::cleanupTestCase() Totals: 2 passed, 1 failed, 0 skipped, 0 blacklisted, 377ms ********* Finished testing of tst_QObjectPerformance *********
Attachments
For Gerrit Dashboard: QTBUG-66823 | ||||||
---|---|---|---|---|---|---|
# | Subject | Branch | Project | Status | CR | V |
221996,2 | Make a benchmark out of tst_QObjectPerformance::emitToManyReceivers | 5.11 | qt/qtbase | Status: MERGED | +2 | 0 |