OOo-2.4.1 segfaults in Slack-12.0 - Slackware

This is a discussion on OOo-2.4.1 segfaults in Slack-12.0 - Slackware ; Hi group, Openoffice-2.4.1 works fine in Slack-11.0 but segfaults in Slack-12.0: $ soffice /usr/bin/soffice: line 254: 4105 Segmentation fault "$sd_prog/$sd_binary" "$@" $ strace /opt/openoffice.org2.4/program/soffice.bin gives huge output that ends up with open("/opt/openoffice.org2.4/program/resource/svx680en-US.res", O_RDONLY) = 35 lseek(35, 0, SEEK_END) = 367930 ...

+ Reply to Thread
Results 1 to 3 of 3

Thread: OOo-2.4.1 segfaults in Slack-12.0

  1. OOo-2.4.1 segfaults in Slack-12.0

    Hi group,

    Openoffice-2.4.1 works fine in Slack-11.0 but segfaults in Slack-12.0:

    $ soffice
    /usr/bin/soffice: line 254: 4105 Segmentation fault
    "$sd_prog/$sd_binary" "$@"

    $ strace /opt/openoffice.org2.4/program/soffice.bin

    gives huge output that ends up with

    open("/opt/openoffice.org2.4/program/resource/svx680en-US.res", O_RDONLY) = 35
    lseek(35, 0, SEEK_END) = 367930
    lseek(35, 0, SEEK_CUR) = 367930
    lseek(35, 367926, SEEK_SET) = 367926
    lseek(35, 0, SEEK_CUR) = 367926
    lseek(35, 367926, SEEK_SET) = 367926
    lseek(35, 0, SEEK_CUR) = 367926
    read(35, "\0\0h\340", 1024) = 4
    lseek(35, 341082, SEEK_SET) = 341082
    lseek(35, 0, SEEK_CUR) = 341082
    mmap2(NULL, 65536, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb0914000
    lseek(35, 341082, SEEK_SET) = 341082
    lseek(35, 0, SEEK_CUR) = 341082
    read(35, "\0\0\1\2\0\0\0\310\0\5+\346\0\0\1\20\0\0\4\333\0\0 \0\0"..., 26848) = 26848
    lseek(35, 335772, SEEK_SET) = 335772
    lseek(35, 0, SEEK_CUR) = 335772
    lseek(35, 335772, SEEK_SET) = 335772
    lseek(35, 0, SEEK_CUR) = 335772
    read(35, "\0\0\'\20\0\0\1v\0\0\0J\0\0\0J\0\0\0\270\0\0\0\2\0 \0\1"..., 1024) = 1024
    --- SIGSEGV (Segmentation fault) @ 0 (0) ---
    +++ killed by SIGSEGV +++
    Process 4220 detached

    The problem is not of vital importance since OOo-2.4.0 works well
    but I'd like to get to know how this `segfaulting' can be fixed, thus
    any ideas will be appreciated!

    Cheers,
    Mikhail


  2. Re: OOo-2.4.1 segfaults in Slack-12.0

    Mikhail Zotov wrote:
    > The problem is not of vital importance since OOo-2.4.0 works well but I'd
    > like to get to know how this `segfaulting' can be fixed, thus any ideas
    > will be appreciated!


    To really debug such a problem you will have to compile the program with
    debug symbols enabled (the -g switch to gcc). Once you have a debug
    version of the program you can use a tool like gdb or a graphical frontend
    like ddd. Another useful tools for finding this kind of bugs are valgrind.

    However, compiling OOo from source is a non-trivial task. You will end up
    in a long chain of dependencies on different java related packages. Also,
    at least those days when I compiled my own OOo the installation method was
    very odd. The compilation ended up with an .rpm file which I then had to
    convert to .tgz and install.

    regards Henrik
    --
    The address in the header is only to prevent spam. My real address is:
    hc3(at)poolhem.se Examples of addresses which go to spammers:
    root@localhost postmaster@localhost


  3. Re: OOo-2.4.1 segfaults in Slack-12.0

    On Sun, 22 Jun 2008 13:17:32 +0200
    Henrik Carlqvist wrote:

    > To really debug such a problem you will have to compile the program with
    > debug symbols enabled (the -g switch to gcc)....
    > However, compiling OOo from source is a non-trivial task.


    Thank you, Henrik! I think, the problem is not worth compiling
    OOo from source. I have found a (pretty dumb) workaround:

    $ touch some.odt && soffice some.odt &

    Surprisingly, this works and soffice doesn't crash.

    Regards,
    Mikhail


+ Reply to Thread