retitle 877339 libreoffice-core: libreoffice SEGV in _expand_stack_to on most documents reassign 877339 src:linux forcemerge 865303 877339 thanks
Hi, On Sat, Sep 30, 2017 at 05:53:47PM +0100, Matthew Vernon wrote: > I've found stable's version of libreoffice basically unusable (I > previously found oldoldstable's just fine), crashing on nearly oldoldstables would also crash the same way if it happened to initialize someting Javaish (e.g. libreoffice-wiki-publisher). And wiki-publisher is installed by the libreoffice metapackage in stretch... (Recommends:, which is installed per default.) > everything I try and open with it. Unfortunately, most of the > Office docs I have to interact with are confidential. I've finally > found one that isn't, so I'll attach it to this bug report. > > The failure mode is always the same - libreoffice gets much of the way > through start up, often opening a window, and then core dumps: > > matthew@aragorn:~$ gdb /usr/lib/libreoffice/program/soffice.bin core > GNU gdb (Debian 7.12-6) 7.12.0.20161007-git > Copyright (C) 2016 Free Software Foundation, Inc. > License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> > This is free software: you are free to change and redistribute it. > There is NO WARRANTY, to the extent permitted by law. Type "show copying" > and "show warranty" for details. > This GDB was configured as "i686-linux-gnu". > Type "show configuration" for configuration details. > For bug reporting instructions, please see: > <http://www.gnu.org/software/gdb/bugs/>. > Find the GDB manual and other documentation resources online at: > <http://www.gnu.org/software/gdb/documentation/>. > For help, type "help". > Type "apropos word" to search for commands related to "word"... > Reading symbols from /usr/lib/libreoffice/program/soffice.bin...(no debugging > symbols found)...done. > [New LWP 25365] > [New LWP 25380] > [New LWP 25367] > [New LWP 25370] > [New LWP 25374] > [New LWP 25372] > [New LWP 25371] > [Thread debugging using libthread_db enabled] > Using host libthread_db library "/lib/i386-linux-gnu/libthread_db.so.1". > Core was generated by `/usr/lib/libreoffice/program/soffice.bin > firefox-downloads/it_manager_person_sp'. > Program terminated with signal SIGSEGV, Segmentation fault. > #0 0x51fb0f95 in _expand_stack_to(unsigned char*) () > from /usr/lib/jvm/java-8-openjdk-i386/jre/lib/i386/server/libjvm.so > [Current thread is 1 (Thread 0x483dc180 (LWP 25365))] > (gdb) bt > #0 0x51fb0f95 in _expand_stack_to(unsigned char*) () > from /usr/lib/jvm/java-8-openjdk-i386/jre/lib/i386/server/libjvm.so There we go. The infamous Stack Clash kernel regression (maybe also a bug in OpenJDK). A simple look at the mostly affected modules (Base, Writer if wiki-publisher is installed) would have shown you that bugs: Outstanding bugs -- Grave functionality bugs; More information needed (1 bug) #865303 [G|Mu|=☣] [src:linux] libreoffice: Libreoffice Java features crash with Linux 3.16.43-2+deb8u1 Outstanding bugs -- Important bugs; Patch Available (2 bugs) #876051 [i|+|☣] [openjdk-8-jre-headless] i386 applications using JNI may crash due to Hotspot workaround for Exec Shield #876069 [i|+|☣] [openjdk-9-jre-headless] i386 applications using JNI may crash due to Hotspot workaround for Exec Shield [...] #876068 [i|+|☣] [openjdk-7-jre-headless] i386 applications using JNI may crash due to Hotspot workaround for Exec Shield See the above bugs, especially on the kernel and https://lwn.net/Articles/727206/ https://lwn.net/Articles/727703/ Yes, everything here is _still_ unfixed. Regards, Rene
