You are not logged in.
Falkon 3.0.1 crashing at startup after upgrading to kde 5.13
error exit:
Failed to load python support plugin "Can not load library /usr/lib/qt/plugins/falkon/PyFalkon.so: (libshiboken2.cpython-36m-x86_64-linux-gnu.so.5.11: Can not open shared object file: File or directory not found) "
Falkon: 3 extensions loaded
double free or corruption (fasttop)
Received signal 6
Received signal 11 SEGV_MAPERR 010000000000
# 0 0x7f30a370ed8e # 0 0x7f30a370ed8e <unknown>
# 1 0x7f30a370eea0 Aborted (recorded core image)
I already searched https://bugs.kde.org/ and found nothing about it.
Can someone help me?
Last edited by nubiocicarini (2018-07-02 22:23:14)
Workers of the world, unite!
Offline
Version of Falkon?
Moving to NC...
Offline
Falkon 3.0.1
Workers of the world, unite!
Offline
Another error output:
Failed to load python support plugin "Cannot load library /usr/lib/qt/plugins/falkon/PyFalkon.so: (libpyside2.cpython-36m-x86_64-linux-gnu.so.5.11: cannot open shared object file: Arquivo ou diretório não encontrado)"
Falkon: 3 extensions loaded
nouveau: kernel rejected pushbuf: Arquivo ou diretório não encontrado
nouveau: ch25: krec 0 pushes 0 bufs 2 relocs 0
nouveau: ch25: buf 00000000 00000002 00000004 00000004 00000000
nouveau: ch25: buf 00000001 00000006 00000004 00000000 00000004
falkon: ../libdrm-2.4.92/nouveau/pushbuf.c:723: nouveau_pushbuf_data: Assertion `kref' failed.
Received signal 6
#0 0x7f5d5175dd8e <unknown>
#1 0x7f5d5175dea0 <unknown>
#2 0x7f5d5175e4d7 <unknown>
#3 0x7f5d574fc8f0 <unknown>
#4 0x7f5d574fc86b __GI_raise
#5 0x7f5d574e740e __GI_abort
#6 0x7f5d574e72e0 __assert_fail_base.cold.0
#7 0x7f5d574f5112 __GI___assert_fail
#8 0x7f5d25659600 nouveau_pushbuf_data
#9 0x7f5d25659563 nouveau_pushbuf_data
#10 0x7f5d25659698 <unknown>
#11 0x7f5d2565a2af <unknown>
#12 0x7f5d2565b2d0 nouveau_pushbuf_kick
#13 0x7f5d2a371d4c <unknown>
#14 0x7f5d2a4f8690 <unknown>
#15 0x7f5d2a07d327 <unknown>
#16 0x7f5d2ba47d1a <unknown>
#17 0x7f5d377fb569 <unknown>
#18 0x7f5d4e84f6f0 QOpenGLContext::swapBuffers()
#19 0x7f5d4eae6848 QPlatformBackingStore::composeAndFlush()
#20 0x7f5d3a541da6 <unknown>
#21 0x7f5d5846e2f1 <unknown>
#22 0x7f5d5846fe0b <unknown>
#23 0x7f5d58471168 <unknown>
#24 0x7f5d58471309 <unknown>
#25 0x7f5d58488a57 QWidgetPrivate::syncBackingStore()
#26 0x7f5d584a1199 QWidget::event()
#27 0x7f5d585b98b5 QMainWindow::event()
#28 0x7f5d58461984 QApplicationPrivate::notify_helper()
#29 0x7f5d5846925b QApplication::notify()
#30 0x7f5d57e7cbc9 QCoreApplication::notifyInternal2()
#31 0x7f5d57e7fc5c QCoreApplicationPrivate::sendPostedEvents()
#32 0x7f5d57ed0864 <unknown>
#33 0x7f5d4caf1368 g_main_context_dispatch
#34 0x7f5d4caf15b1 <unknown>
#35 0x7f5d4caf163e g_main_context_iteration
#36 0x7f5d57ecfe49 QEventDispatcherGlib::processEvents()
#37 0x7f5d3a5b0b32 <unknown>
#38 0x7f5d57e7b85c QEventLoop::exec()
#39 0x7f5d57e83b56 QCoreApplication::exec()
#40 0x558b919bbf21 <unknown>
#41 0x7f5d574e906b __libc_start_main
#42 0x558b919bc23a _start
r8: 0000000000000000 r9: 00007ffd64643960 r10: 0000000000000008 r11: 0000000000000246
r12: 00007f5d2565c640 r13: 00007f5d2565c7b1 r14: 00000000000002d3 r15: 0000000000000000
di: 0000000000000002 si: 00007ffd64643960 bp: 00007f5d57647e38 bx: 0000000000000006
dx: 0000000000000000 ax: 0000000000000000 cx: 00007f5d574fc86b sp: 00007ffd64643960
ip: 00007f5d574fc86b efl: 0000000000000246 cgf: 002b000000000033 erf: 0000000000000000
trp: 0000000000000000 msk: 0000000000000000 cr2: 0000000000000000
[end of stack trace]
Calling _exit(1). Core file will not be generated.
Last edited by nubiocicarini (2018-06-18 19:33:09)
Workers of the world, unite!
Offline
Does it work if you downgrade to qt5-webengine 5.11.0-2?
Offline
Please use code tags when pasting to the boards: https://wiki.archlinux.org/index.php/Co … s_and_code
Offline
Does it work if you downgrade to qt5-webengine 5.11.0-2?
With the downgrade to the 5.11.0-2 qt5-webengine works perfectly. However, I will not close the topic, since it is not the current version. All right?
Last edited by nubiocicarini (2018-06-18 19:35:49)
Workers of the world, unite!
Offline
Please use code tags when pasting to the boards: https://wiki.archlinux.org/index.php/Co … s_and_code
Sorry for the inconvenience. I'm starting in the community.
Workers of the world, unite!
Offline
I found a similar error (at https://github.com/ColinDuquesnoy/Mello … ssues/252), the indications point to the nouveau drive. However, it also has some relation with qt5-webengine 5.11.0-3, since the downgrade solved the problem.
Last edited by nubiocicarini (2018-06-18 20:34:08)
Workers of the world, unite!
Offline
The problem was in the OpenGL acceleration of the qt5-webengine that does not work well under the nouveau drive. The workaround is to run the QT_XCB_FORCE_SOFTWARE_OPENGL = 1 workaround until qt5-webengine's OpenGL is disabled by the maintainer. Or commit the sin of installing the nvidia owner drive.
Workers of the world, unite!
Offline
Or try the modesetting driver (removal of xf86-vide-nouveau should be enough to make X use modesttting )
Last edited by Lone_Wolf (2018-06-19 13:17:59)
Disliking systemd intensely, but not satisfied with alternatives so focusing on taming systemd.
clean chroot building not flexible enough ?
Try clean chroot manager by graysky
Offline
This is an issue with the GL driver, so I doubt changing the DDX driver will do much.
Last edited by arojas (2018-06-19 13:36:33)
Offline
I may be wrong, but I think the nouveau has problems with OpenGL when it comes to multiprocessing. I may be wrong, but I think the nouveau has problems with OpenGL when it comes to multiprocessing. The point is that Upstream is supposed to have code to disable opengl for nouveau but it's obviously not working (qtwebengine + opengl + nouveau). Already with the owner drive this error does not occur.
Workers of the world, unite!
Offline
In driver modesetting lose acceleration?
Workers of the world, unite!
Offline
This is an issue with the GL driver, so I doubt changing the DDX driver will do much.
The error messages in this topic and the linked github issue both mention problems with pushbuf and appear to come from the nouveau driver.
I haven't found similar reports from users with intel, amd or modesetting drivers, trying modesetting driver seems a logical next step for troubleshooting.
In driver modesetting lose acceleration?
2D may be slower, but 3D is accelerated with the modesetting driver.
Whether modesetting performs better or worse then DDX depends on a lot of factors and is unpredictable.
Last edited by Lone_Wolf (2018-06-20 12:30:59)
Disliking systemd intensely, but not satisfied with alternatives so focusing on taming systemd.
clean chroot building not flexible enough ?
Try clean chroot manager by graysky
Offline
I notified the bug: https://bugs.archlinux.org/task/59057#comment170555
Antonio Rojas (arojas), from the arch community, referred to the Qt: https://bugreports.qt.io/browse/QTBUG-41242
Last edited by nubiocicarini (2018-06-20 14:07:55)
Workers of the world, unite!
Offline
Antonio Rojas (arojas), from the arch community, referred to the Qt: https://bugreports.qt.io/browse/QTBUG-68479
What? No, I didn't, in fact that Qt bug report is about Wayland and has nothing to do with your issue. The relevant Qt report is https://bugreports.qt.io/browse/QTBUG-41242 which they claim is fixed.
Offline
This is an issue with the GL driver, so I doubt changing the DDX driver will do much.
actually, the opengl of the nvidia owner driver also does not work on qt-webengine. I got errors when activating firefox acceleration. However, with the owner drive the function is disabled by default, not returning error as it does with nouveau.
Workers of the world, unite!
Offline
nubiocicarini wrote:Antonio Rojas (arojas), from the arch community, referred to the Qt: https://bugreports.qt.io/browse/QTBUG-68479
What? No, I didn't, in fact that Qt bug report is about Wayland and has nothing to do with your issue. The relevant Qt report is https://bugreports.qt.io/browse/QTBUG-41242 which they claim is fixed.
Thanks for the correction, I listed it wrong while copying the link. I just made the correction.
Last edited by nubiocicarini (2018-06-20 14:08:49)
Workers of the world, unite!
Offline
The relevant Qt report is https://bugreports.qt.io/browse/QTBUG-41242 which they claim is fixed.
That bug report does make sense and put the blame in the mesa nouveau driver, not the X DDX as I thought.
@nubiocicarini : switching to x modesetting should have no influence on this problem.
I've checked the tarball used by the qt5-webengine 5.11.1-1 package and it does include this fix* in
src/3rdparty/chromium/gpu/config/software_rendering_list.json
nubiocicarini, can you verify if latest qt5-webengine solves this when running nouveau ?
*
Disabling all HW acceleration is not what I would call a fix,it feels more like the lesser of two evils
Disliking systemd intensely, but not satisfied with alternatives so focusing on taming systemd.
clean chroot building not flexible enough ?
Try clean chroot manager by graysky
Offline
arojas wrote:The relevant Qt report is https://bugreports.qt.io/browse/QTBUG-41242 which they claim is fixed.
That bug report does make sense and put the blame in the mesa nouveau driver, not the X DDX as I thought.
@nubiocicarini : switching to x modesetting should have no influence on this problem.I've checked the tarball used by the qt5-webengine 5.11.1-1 package and it does include this fix* in
src/3rdparty/chromium/gpu/config/software_rendering_list.jsonnubiocicarini, can you verify if latest qt5-webengine solves this when running nouveau ?
*
Disabling all HW acceleration is not what I would call a fix,it feels more like the lesser of two evils
I just did the test (nouveau + falkon) and it did not work. The fix should refer to chromiun.
Workers of the world, unite!
Offline
The blacklist is clearly not working, so someone affected should ask for the upstream bug to be reopened. I don't have a NVIDIA card myself so can't test.
Offline
I've enforced the upstream error notification. However, I'm just a freak without technical background and do not know how to do things like, for example, ask to reopen the bug.
Workers of the world, unite!
Offline
Hi All,
Am having the same issues, although this thread is marked [SOLVED] it doesn't appear to be, any updates?
Falkon: 2 extensions loaded
nouveau: kernel rejected pushbuf: No such file or directory
nouveau: ch11: krec 0 pushes 0 bufs 1 relocs 0
nouveau: ch11: buf 00000000 00000003 00000004 00000004 00000000
nouveau: kernel rejected pushbuf: No such file or directory
nouveau: ch11: krec 0 pushes 0 bufs 1 relocs 0
nouveau: ch11: buf 00000000 00000003 00000004 00000004 00000000
falkon: ../libdrm-2.4.92/nouveau/pushbuf.c:723: nouveau_pushbuf_data: Assertion `kref' failed.
Received signal 6
#0 0x7f870c2d4e9e <unknown>
#1 0x7f870c2d4fb0 <unknown>
#2 0x7f870c2d55e7 <unknown>
#3 0x7f87120888f0 <unknown>
#4 0x7f871208886b __GI_raise
#5 0x7f871207340e __GI_abort
#6 0x7f87120732e0 __assert_fail_base.cold.0
#7 0x7f8712081112 __GI___assert_fail
#8 0x7f86dfa32600 nouveau_pushbuf_data
#9 0x7f86dfa32563 nouveau_pushbuf_data
#10 0x7f86dfa32698 <unknown>
#11 0x7f86dfa332af <unknown>
#12 0x7f86dfa342d0 nouveau_pushbuf_kick
#13 0x7f86e474ae97 <unknown>
#14 0x7f86e49abe1f <unknown>
#15 0x7f86e49abe90 <unknown>
#16 0x7f8711748404 <unknown>
#17 0x7f8711748567 <unknown>
#18 0x7f8712a3338b QObject::event()
#19 0x7f871302e76c QWidget::event()
#20 0x7f871174c256 QQuickWidget::event()
#21 0x7f8711e417f1 <unknown>
#22 0x7f8712feea74 QApplicationPrivate::notify_helper()
#23 0x7f8712ff6341 QApplication::notify()
#24 0x7f8712a08cb9 QCoreApplication::notifyInternal2()
#25 0x7f8712a5b40a QTimerInfoList::activateTimers()
Offline
It had closed, therefore I decided to install the proprietary drive to get around the problem. I'm running out of time to do more testing. I am a social scientist and I do not have much computer knowledge. It takes me a lot of reading to understand these things. But I reopened the topic for you.
Workers of the world, unite!
Offline