Summary: | dev-qt/qtbase-6.6.1-r1: fails tst_QUuid::uint128 test on ppc64 | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Sam James <sam> |
Component: | Current packages | Assignee: | Qt Bug Alias <qt> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | ionen, ppc64 |
Priority: | Normal | Keywords: | TESTFAILURE |
Version: | unspecified | ||
Hardware: | PPC64 | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Bug Depends on: | |||
Bug Blocks: | 914026 | ||
Attachments: |
build.log.xz
LastTest.log.xz |
Description
Sam James
![]() ![]() ![]() ![]() Created attachment 875946 [details]
build.log.xz
Created attachment 875947 [details]
LastTest.log.xz
(In reply to Sam James from comment #0) > FAIL! : tst_QUuid::uint128() Compared values are not the same > Actual (uuid) : {d128b8c5-ee95-a43c-3644-34cc9eb569fc} > Expected (uuidA): {fc69b59e-cc34-4436-a43c-ee95d128b8c5} It looks like an endianness issue, although it's weird ppc32 passed... This test is actually new in >=qtbase-6.6.0 (6.5.3 did no uint128 checks at all in tst_quuid), I suspect 6.5.3 would've passed and that 6.6.1 "probably" fails for all BE arches now. It's quite possible that it's the test itself that's wrong (it has some odd be/le handling that does not actually verify anything?), and is not an actual runtime problem. wrt ppc32, I suspect the test was skipped, would need to check test logs for: #ifdef QT_SUPPORTS_INT128 ... #else QSKIP("This platform has no support for 128-bit integer"); #endif The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=3c920c22b4ba6280e04ed73153ed95c406f6f305 commit 3c920c22b4ba6280e04ed73153ed95c406f6f305 Author: Ionen Wolkens <ionen@gentoo.org> AuthorDate: 2023-11-30 05:21:35 +0000 Commit: Ionen Wolkens <ionen@gentoo.org> CommitDate: 2023-11-30 05:28:35 +0000 dev-qt/qtbase: skip tst_quuid for now uint128 parts for this test are new, and is likely not worth holding back keywording over. Unlikely to result in real runtime problems. Needs confirmation, but believed that this would (now) fail on other 64bit BE arches as well and is not ppc64be-specific. Closes: https://bugs.gentoo.org/918878 Signed-off-by: Ionen Wolkens <ionen@gentoo.org> dev-qt/qtbase/qtbase-6.6.1-r1.ebuild | 3 ++- dev-qt/qtbase/qtbase-6.6.9999.ebuild | 3 ++- dev-qt/qtbase/qtbase-6.9999.ebuild | 3 ++- 3 files changed, 6 insertions(+), 3 deletions(-) (In reply to Ionen Wolkens from comment #4) > This test is actually new in >=qtbase-6.6.0 (6.5.3 did no uint128 checks at > all in tst_quuid), I suspect 6.5.3 would've passed and that 6.6.1 "probably" > fails for all BE arches now. > > It's quite possible that it's the test itself that's wrong (it has some odd > be/le handling that does not actually verify anything?), and is not an > actual runtime problem. > > wrt ppc32, I suspect the test was skipped, would need to check test logs for: > > #ifdef QT_SUPPORTS_INT128 > ... > #else > QSKIP("This platform has no support for 128-bit integer"); > #endif Confirmed: 15499:SKIP : tst_QUuid::uint128() This platform has no support for 128-bit integer 15500: Loc: [/var/tmp/portage/dev-qt/qtbase-6.6.1-r1/work/qtbase-everywhere-src-6.6.1/tests/auto/corelib/plugin/quuid/tst_quuid.cpp(293)] I'll check on another BE platform with the skip reverted too. (In reply to Sam James from comment #6) > [...] > > I'll check on another BE platform with the skip reverted too. Indeed on sparc: # grep ^FAIL.*: /var/tmp/portage/dev-qt/qtbase-6.6.1-r1/work/qtbase-everywhere-src-6.6.1_build/Testing/Temporary/LastTest.log -i FAIL! : tst_QUuid::uint128() Compared values are not the same FAIL! : tst_QBuffer::writeOfMoreThan2GiB() 'buffer.buffer() == QByteArrayView{input}' returned FALSE. () FAIL! : tst_QtConcurrentIterateKernel::throttling() Compared values are not the same FAIL! : tst_QProcess::processesInMultipleThreads() 'threads[j]->wait(10000)' returned FALSE. () FAIL! : tst_QProcess::processesInMultipleThreads() Received a fatal error. Thanks for confirming, not going to worry about this new test too much. |