FS#35466 - [icu] libreoffice-writer doesn't start with icu 51.2-1

Attached to Project: Arch Linux
Opened by Andrew Gaydenko (student975) - Saturday, 25 May 2013, 00:05 GMT
Last edited by Evangelos Foutras (foutrelis) - Saturday, 25 May 2013, 05:38 GMT
Task Type Bug Report
Category Packages: Extra
Status Closed
Assigned To Andreas Radke (AndyRTR)
Architecture x86_64
Severity Low
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 4
Private No

Details

"Doesn't start" means the lowriter doesn't start without any output, CPU eating and so on - just momentary and silent returning to bash prompt.

Downgrading to icu 51.1-2 resolves the issue.
This task depends upon

Closed by  Evangelos Foutras (foutrelis)
Saturday, 25 May 2013, 05:38 GMT
Reason for closing:  Fixed
Additional comments about closing:  libreoffice 4.0.3-3
Comment by Fabricio Todeschini (ginobvhc) - Saturday, 25 May 2013, 00:26 GMT
Sorry by the english :S
I can't downgrade becouse i didn't find the 51.1-2
Comment by alexander (actics) - Saturday, 25 May 2013, 01:01 GMT
I have the same problem. I downgrade icu to 51.1-1 with "downgrade" utility to solution.

journalctl:
май 25 06:29:33 actics-arch-x120e kernel: soffice.bin[3160]: segfault at 10 ip 00007fb994be2b60 sp 00007fff049637d8 error 4 in libicule.so.51.2[7fb994bb7000+55000]
май 25 06:29:33 actics-arch-x120e systemd-coredump[3164]: Process 3160 (soffice.bin) dumped core.
Comment by Fabricio Todeschini (ginobvhc) - Saturday, 25 May 2013, 01:32 GMT
downgrade is exelent (Y), temporarily solved.

shoud i cange the topic to SOLVED?

sorry my ignorance, i never report bug becouse my awful english.
Comment by alexander (actics) - Saturday, 25 May 2013, 02:03 GMT
No, I think it's bug.
Comment by Kasi (popsch) - Saturday, 25 May 2013, 02:35 GMT
I have the same problem. It happens with all libreoffice programs. Here's the output of localc --backtrace

Program received signal SIGSEGV, Segmentation fault.
0x00007fffef878b60 in icu_51::LEGlyphStorage::getGlyphs(unsigned int*, unsigned int, LEErrorCode&) const () from /usr/lib/libicule.so.51
#0 0x00007fffef878b60 in icu_51::LEGlyphStorage::getGlyphs(unsigned int*, unsigned int, LEErrorCode&) const () from /usr/lib/libicule.so.51
#1 0x00007ffff391836b in ?? () from /usr/lib/libreoffice/program/libvcllo.so
#2 0x00007ffff378de70 in ?? () from /usr/lib/libreoffice/program/libvcllo.so
#3 0x00007ffff3790d35 in OutputDevice::GetTextArray(String const&, int*, unsigned short, unsigned short) const () from /usr/lib/libreoffice/program/libvcllo.so
#4 0x00007ffff3790ecb in OutputDevice::approximate_char_width() const () from /usr/lib/libreoffice/program/libvcllo.so


Grabbing an icu-51.1-1 from an out-of-sync mirror solved the problem for me.
Comment by Evangelos Foutras (foutrelis) - Saturday, 25 May 2013, 03:06 GMT
Looks like ICU 51.2 didn't maintain ABI compatibility with 51.1:

http://upstream-tracker.org/compat_reports/icu4c/51.1_to_51.2/abi_compat_report.html#High_Risk_Problems

Loading...