Steps to reproduce: 1. Open LibreOffice 2. Tools - Extension Manager 3. Check for updates 4. Close the dialog -> CRASH Reproduced in Version: 6.2.0.0.beta1 Build ID: d1b41307be3f8c19fe6f1938cf056e7ff1eb1d18 CPU threads: 1; OS: Linux 4.15; UI render: default; VCL: kde5; Locale: en-US (en_US.UTF-8); UI-Language: en-US Calc: threaded but not in Version: 6.2.0.0.beta1 Build ID: d1b41307be3f8c19fe6f1938cf056e7ff1eb1d18 CPU threads: 1; OS: Linux 4.15; UI render: default; VCL: kde4; Locale: en-US (en_US.UTF-8); UI-Language: en-US Calc: threaded
Created attachment 146822 [details] screenshot of installed extensions I cannot reproduce here with exactly the same build and a clean user profile (nor with a custom master build): Version: 6.2.0.0.beta1 Build ID: d1b41307be3f8c19fe6f1938cf056e7ff1eb1d18 CPU threads: 4; OS: Linux 4.18; UI render: default; VCL: kde5; Locale: en-GB (en_GB.UTF-8); UI-Language: en-US Calc: threaded @Xisco: Do you have any additional extensions installed besides the ones bundled with LibreOffice? I just have those, s. attached screenshot.
Created attachment 146842 [details] install extensions
it seems I have the same extension with the same versions as you...
I'm using Kubuntu 18.04.1 For the record, it hangs with GTK3. @Michael, it works fine with GTK3 in your end ?
(In reply to Xisco Faulí from comment #4) > I'm using Kubuntu 18.04.1 > > For the record, it hangs with GTK3. > @Michael, it works fine with GTK3 in your end ? I'm using Debian testing with Plasma. When using GTK3, it hangs for me as well.
I reproduced this crash for Version: 6.3.0.0.alpha0+ Build ID: b15b1a2a90fa4c239ff8a6a33e73ff50ea422abf CPU threads: 1; OS: Linux 4.14; UI render: default; VCL: kde5; TinderBox: Linux-rpm_deb-x86_64@86-TDF, Branch:master, Time: 2018-11-28_06:21:05 Locale: ru-RU (ru_RU.UTF-8); UI-Language: en-US Calc: threaded NO additional extentions installed.
Created attachment 147125 [details] gdbtrace.log
On pc Debian x86-64 with master sources updated today + kde5 rendering, I got no hang but noticed these kind of trace on console: warn:vcl.schedule:8331:8331:vcl/source/app/scheduler.cxx:367: we're too early - restart the timer (14ms)! warn:svtools:8331:8331:vcl/source/graphic/UnoGraphicProvider.cxx:433: Could not create graphic: 0x8201(Error Area:Vcl Class:General Code:1) QObject::startTimer: Timers cannot be started from another thread QObject::killTimer: Timers cannot be stopped from another thread I don't have these with gtk3 rendering.
Still reproducible in Version: 6.2.0.1 Build ID: 0412ee99e862f384c1106d0841a950c4cfaa9df1 CPU threads: 1; OS: Linux 4.15; UI render: default; VCL: kde5; Locale: en-US (en_US.UTF-8); UI-Language: en-US Calc: threaded
Created attachment 148197 [details] gdbtrace.log Crash is still reproduced Version: 6.3.0.0.alpha0+ Build ID: 32b85ec64e5c901369ce3ce5dbda0e96f2ec7fc0 CPU threads: 4; OS: Linux 4.14; UI render: default; VCL: kde5; TinderBox: Linux-rpm_deb-x86_64@86-TDF, Branch:master, Time: 2019-01-07_16:08:14 Locale: ru-RU (ru_RU.UTF-8); UI-Language: en-US Calc: threaded New (advanced) gdbtrace.log in the attachment.
I'm unable to reproduce this crash with Qt 5.11.3, but I didn't try Qt 5.9.x. Can someone confirm if it doesn't crash with Qt >= 5.11? Version: 6.3.0.0.alpha0+ Build ID: 0aff7edaa4ec4a2aa232bac27f669e42435905b3 CPU threads: 12; OS: Linux 4.19; UI render: default; VCL: kde5; Locale: en-US (C); UI-Language: en-US Calc: threaded
(In reply to Aleksei Nikiforov from comment #11) > I'm unable to reproduce this crash with Qt 5.11.3, but I didn't try Qt > 5.9.x. Can someone confirm if it doesn't crash with Qt >= 5.11? Yes, my result in comment 1 also was with Qt 5.11.x
(In reply to Michael Weghorn from comment #12) > (In reply to Aleksei Nikiforov from comment #11) > > I'm unable to reproduce this crash with Qt 5.11.3, but I didn't try Qt > > 5.9.x. Can someone confirm if it doesn't crash with Qt >= 5.11? > > Yes, my result in comment 1 also was with Qt 5.11.x ... but of course I can't say that a newer Qt version will fix it for those that have the crash...
Peeps please try again, lot of threading fixes were merged recently and the backtrace looks very much like a threading/locking issue
Hello, I've checked the behavior for 6.3.0.0.alpha0+, 6.2.3.0.0+, 6.2.2.0.0+ Crash is NOT reproduced for Version: 6.3.0.0.alpha0+ Build ID: bbc5347be70d11d187cd537dd73cdab584def719 CPU threads: 4; OS: Linux 4.19; UI render: default; VCL: kde5; TinderBox: Linux-rpm_deb-x86_64@86-TDF, Branch:master, Time: 2019-03-03_20:19:18 Locale: ru-RU (ru_RU.UTF-8); UI-Language: en-US Calc: threaded Crash is NOT reproduced for Version: 6.2.3.0.0+ Build ID: 77cc7463bcd3fe9a6528ed1b0309708608cd1cc7 CPU threads: 4; OS: Linux 4.19; UI render: default; VCL: kde5; TinderBox: Linux-rpm_deb-x86_64@86-TDF, Branch:libreoffice-6-2, Time: 2019-03-13_11:55:49 Locale: ru-RU (ru_RU.UTF-8); UI-Language: en-US Calc: threaded BUT still ACTUAL for Version: 6.2.2.0.0+ Build ID: b17bffe4eb15e4cec969aa2699ce8be60e01b1c1 CPU threads: 4; OS: Linux 4.19; UI render: default; VCL: kde5; TinderBox: Linux-rpm_deb-x86_64@86-TDF, Branch:libreoffice-6-2, Time: 2019-02-25_17:41:32 Locale: ru-RU (ru_RU.UTF-8); UI-Language: en-US Calc: threaded
Pls set to RESOLVED if 6.2.3 and 6.3.0 is enough.
(In reply to Vera Blagoveschenskaya from comment #16) > Pls set to RESOLVED if 6.2.3 and 6.3.0 is enough. it works like a charm now Version: 6.3.0.0.alpha0+ Build ID: 17b25fd3df237a64d6a28fbc57b869e080963193 CPU threads: 2; OS: Linux 4.15; UI render: default; VCL: kde5; TinderBox: Linux-rpm_deb-x86_64@86-TDF, Branch:master, Time: 2019-03-13_20:42:25 Locale: en-US (en_US.UTF-8); UI-Language: en-US Calc: threaded This week some KDE5 commits were backported to 6-2 but not to 6-2-2. Probably one of those commits fixed the issue. Closing as RESOLVED WORKSFORME @Bubli, @Michael, thanks for your work here
VERIFIED due to https://bugs.documentfoundation.org/show_bug.cgi?id=121516#c15
Hm, so we might want to backport the threading fixes we made bc of crashers in tdf#119856
(In reply to Katarina Behrens (CIB) from comment #19) > Hm, so we might want to backport the threading fixes we made bc of crashers > in tdf#119856 As far as I understand, those should be unrelated, since comment 15 mentions it's no more reproducible with 6.2.3.0.0+, and the fixes have not been backported to libreoffice-6-2 so far.
Maybe bug 123745 ?
(In reply to Xisco Faulí from comment #21) > Maybe bug 123745 ? Sounds quite possible. In that case, there's nothing to do, since the patch has been backported to libreoffice-6-2-2 already. (I quickly verified that the commit wasn't yet included in Vera's 6.2.2 build from comment 15, but was in the 6.2.3 build).
(In reply to Michael Weghorn from comment #22) > (In reply to Xisco Faulí from comment #21) > > Maybe bug 123745 ? > > Sounds quite possible. In that case, there's nothing to do, since the patch > has been backported to libreoffice-6-2-2 already. (I quickly verified that > the commit wasn't yet included in Vera's 6.2.2 build from comment 15, but > was in the 6.2.3 build). I guess it can be quickly check with a local build in KDE5 by reverting it... I don't have it though
(In reply to Xisco Faulí from comment #23) > I guess it can be quickly check with a local build in KDE5 by reverting > it... I don't have it though Unfortunately, I cannot test either, since I was never able to reproduce (s. comment 1). Alternatively, it could be checked it's actually fixed with a recent build from the 6-2-2 branch. Or one could fall back to hoping it's fixed and in case that fails, it's still going to be fixed in 6.2.3 at least...