Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 935356 - dev-qt/qtsvg-6.7.1 fails tests on ppc: tst_QSvgRenderer::testFeColorMatrix() 'image.allGray()' returned FALSE. ()
Summary: dev-qt/qtsvg-6.7.1 fails tests on ppc: tst_QSvgRenderer::testFeColorMatrix() ...
Status: RESOLVED FIXED
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: Current packages (show other bugs)
Hardware: PPC Linux
: Normal normal (vote)
Assignee: Qt Bug Alias
URL:
Whiteboard:
Keywords: TESTFAILURE
Depends on:
Blocks:
 
Reported: 2024-07-02 16:28 UTC by matoro
Modified: 2024-07-03 03:47 UTC (History)
3 users (show)

See Also:
Package list:
Runtime testing required: ---


Attachments
build.log and emerge --info (file_935356.txt,98.89 KB, text/plain)
2024-07-02 16:28 UTC, matoro
Details
build.log and emerge --info (file_935356.txt,98.44 KB, text/plain)
2024-07-03 02:47 UTC, matoro
Details

Note You need to log in before you can comment on or make changes to this bug.
Description matoro archtester 2024-07-02 16:28:56 UTC
Created attachment 896808 [details]
build.log and emerge --info

Test project /var/tmp/portage/dev-qt/qtsvg-6.7.1/work/qtsvg-everywhere-src-6.7.1_build
    Start 1: tst_qsvgdevice
    Start 2: tst_qsvggenerator
    Start 3: tst_qsvgplugin
    Start 4: tst_qicon_svg
    Start 5: tst_qsvgrenderer
1/5 Test #1: tst_qsvgdevice ...................   Passed    0.06 sec
2/5 Test #2: tst_qsvggenerator ................   Passed    0.08 sec
3/5 Test #4: tst_qicon_svg ....................   Passed    0.08 sec
4/5 Test #3: tst_qsvgplugin ...................   Passed    0.09 sec
5/5 Test #5: tst_qsvgrenderer .................***Failed    0.14 sec
 
80% tests passed, 1 tests failed out of 5
 
Label Time Summary:
tests/auto/qicon_svg/tst_qicon_svg            =   0.08 sec*proc (1 test)
tests/auto/qsvgdevice/tst_qsvgdevice          =   0.06 sec*proc (1 test)
tests/auto/qsvggenerator/tst_qsvggenerator    =   0.08 sec*proc (1 test)
tests/auto/qsvgplugin/tst_qsvgplugin          =   0.09 sec*proc (1 test)
tests/auto/qsvgrenderer/tst_qsvgrenderer      =   0.14 sec*proc (1 test)

FAIL!  : tst_QSvgRenderer::testFeColorMatrix() 'image.allGray()' returned FALSE. ()
   Loc: [/var/tmp/portage/dev-qt/qtsvg-6.7.1/work/qtsvg-everywhere-src-6.7.1/tests/auto/qsvgrenderer/tst_qsvgrenderer.cpp(2064)]
Comment 1 Ionen Wolkens gentoo-dev 2024-07-03 00:18:20 UTC
Hm, seems testFeColorMatrix was added in 6.7.0-beta1 after 6.6.1 was keyworded for ~ppc.

May possibly be another case of broken on all BE, like some other color tests. Could you try another BE arch to confirm? I think our ppc64 stable was done on ppc64le and the ~hppa keyword may predate the test (haven't checked).

Not too worried about this test and don't want to spend time on ppc32 issues more than necessary, so figure I'll just skip this either way and not limit to ppc if it's all BE. With test being new it's at least likely not a regression.
Comment 2 matoro archtester 2024-07-03 02:47:01 UTC
Created attachment 896827 [details]
build.log and emerge --info

(In reply to Ionen Wolkens from comment #1)
> Hm, seems testFeColorMatrix was added in 6.7.0-beta1 after 6.6.1 was
> keyworded for ~ppc.
> 
> May possibly be another case of broken on all BE, like some other color
> tests. Could you try another BE arch to confirm? I think our ppc64 stable
> was done on ppc64le and the ~hppa keyword may predate the test (haven't
> checked).
> 
> Not too worried about this test and don't want to spend time on ppc32 issues
> more than necessary, so figure I'll just skip this either way and not limit
> to ppc if it's all BE. With test being new it's at least likely not a
> regression.

Aye, fails on ppc64 BE as well.  Please do drop if you can.  Even better mask in profiles/features/big-endian and cite bug.
Comment 3 Ionen Wolkens gentoo-dev 2024-07-03 03:11:57 UTC
(In reply to matoro from comment #2)
> Aye, fails on ppc64 BE as well.  Please do drop if you can.  Even better
> mask in profiles/features/big-endian and cite bug.
Thanks, and unfortunately qtsvg is kind of essential -- tons of applications will have no icons without it -- and this (new) test doesn't indicate it's heavily broken to the point of needing to mask it as far as I can tell.

Not that I'd personally mind to see ppc32 dropped or at least destabilized from *all* of qt5+qt6 and their revdeps where non-optional if someone wanted to sort that out. Right now doing the ppc dance with qt6 just for the sake of parity w/ qt5 and allowing to migrate+stabilize ppc-keyworded ebuilds to qt6 without hassle.

On that note, sparc should probably be dropped entirely from qt5+qt6 too. Both USE=qt5 and qt6 are still masked for sparc and tests been in shambles.

Anyhow, as far as this bug goes, I'll just go ahead and skip the test so BE arches don't all trip on this.
Comment 4 Larry the Git Cow gentoo-dev 2024-07-03 03:47:23 UTC
The bug has been closed via the following commit(s):

https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=84374fa574854aa4e81c77c60be57baa275ee65a

commit 84374fa574854aa4e81c77c60be57baa275ee65a
Author:     Ionen Wolkens <ionen@gentoo.org>
AuthorDate: 2024-07-03 03:30:31 +0000
Commit:     Ionen Wolkens <ionen@gentoo.org>
CommitDate: 2024-07-03 03:47:02 +0000

    dev-qt/qtsvg: skip tst_qsvgrenderer on big-endian
    
    Closes: https://bugs.gentoo.org/935356
    Signed-off-by: Ionen Wolkens <ionen@gentoo.org>

 dev-qt/qtsvg/qtsvg-6.7.1.ebuild    | 10 +++++++++-
 dev-qt/qtsvg/qtsvg-6.7.2.ebuild    | 10 +++++++++-
 dev-qt/qtsvg/qtsvg-6.7.9999.ebuild | 10 +++++++++-
 dev-qt/qtsvg/qtsvg-6.8.9999.ebuild | 10 +++++++++-
 dev-qt/qtsvg/qtsvg-6.9999.ebuild   | 10 +++++++++-
 5 files changed, 45 insertions(+), 5 deletions(-)