Feedback to Kingsoft Office/WPS Office for Linux.
16 posts Page 1 of 2
I just installed the latest version for linux (kingsoft-office.i686 0:9.1.0.4244-0.1.a12p3) on Fedora 20 (Linux 3.12.6-300.fc20.x86_64 #1 SMP Mon Dec 23 16:44:31 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux) and it crashed with a seg. fault what ever I run: wps, wpp or et.

F.

$ wps
/usr/bin/wps: line 31: 10418 Segmentation fault (core dumped) ${gInstallPath}/office6/${gApp} ${gOpt} "$@"
So sorry to hear that.
We need more details, can you upload the core dump file? (You need to set ulimit -c unlimited)
Or any crash log, like trace stack.

Do you have multiple printers installed? Try to disable them see if it helps.
And if you set LD_LIBRARY_PATH to somewhere,
and there is a Qt of system, this also crash WPS (as WPS use it's own Qt)
Hi,

The core file is here : https://mega.co.nz/#!w8wVWBaK!FR7u4CwWt ... NbrLrwxKGo

I found this in /var/log/messages:
Code: Select all
Jan 14 14:47:45 arabella kernel: [ 2810.465672] wps[2949]: segfault at 0 ip 00000000f728d317 sp 00000000ffd14170 error 4 in libkxshare.so[f70f4000+3d8000]
Jan 14 14:51:52 arabella kernel: [ 3057.317034] wps[3321]: segfault at 0 ip 00000000f7287317 sp 00000000ff954710 error 4 in libkxshare.so[f70ee000+3d8000]
Jan 14 14:53:01 arabella kernel: [ 3126.552587] wps[3475]: segfault at 0 ip 00000000f7225317 sp 00000000ffab4de0 error 4 in libkxshare.so[f708c000+3d8000]
Jan 14 14:54:57 arabella kernel: [ 3243.011002] wps[3656]: segfault at 0 ip 00000000f729d317 sp 00000000ffeac4d0 error 4 in libkxshare.so[f7104000+3d8000]
Jan 14 14:55:23 arabella kernel: [ 3268.294197] wps[3767]: segfault at 0 ip 00000000f727b317 sp 00000000ff9b3580 error 4 in libkxshare.so[f70e2000+3d8000]


I have no LD_LIBRARY_PATH.

Remove all Qt rpm files:
Code: Select all
$ sudo yum history info 62
Loaded plugins: aliases, fastestmirror, filter-data, fs-snapshot, keys, langpacks, list-data, local, refresh-packagekit
Transaction ID : 62
Begin time     : Tue Jan 14 14:53:38 2014
Begin rpmdb    : 2356:80b974b5287e7df26f38d69b99c0a52ee2896e39
End time       :            14:54:32 2014 (54 seconds)
End rpmdb      : 2339:22ce73f069f15e960e4917493b1d734199506eae
Return-Code    : Success
Command Line   : remove qt
Transaction performed with:
    Installed     rpm-4.11.1-7.fc20.x86_64                       @anaconda
    Installed     yum-3.4.3-128.fc20.noarch                      @updates
    Installed     yum-plugin-aliases-1.1.31-19.fc20.noarch       @updates
    Installed     yum-plugin-fastestmirror-1.1.31-19.fc20.noarch @updates
Packages Altered:
    Erase PyQt4-4.10.2-5.fc20.x86_64                     @updates
    Erase ibus-qt-1.3.2-5.fc20.x86_64                    @anaconda
    Erase libqzeitgeist-0.8.0-10.fc20.x86_64             @fedora
    Erase liveusb-creator-3.12.0-1.fc20.noarch           @fedora
    Erase oprofile-gui-0.9.9-2.fc20.x86_64               @anaconda
    Erase phonon-4.7.1-1.fc20.x86_64                     @updates
    Erase phonon-backend-gstreamer-2:4.7.1-1.fc20.x86_64 @updates
    Erase pinentry-qt-0.8.1-11.fc20.x86_64               @anaconda
    Erase poppler-qt-0.24.3-3.fc20.x86_64                @updates
    Erase qt-1:4.8.5-11.fc20.x86_64                      @updates
    Erase qt-mobility-1.2.2-0.5.20120224git.fc20.x86_64  @fedora
    Erase qt-x11-1:4.8.5-11.fc20.x86_64                  @updates
    Erase qtwebkit-2.3.3-3.fc20.x86_64                   @updates
    Erase seahorse-3.10.1-1.fc20.x86_64                  @anaconda
    Erase texmaker-1:4.0.3-1.fc20.x86_64                 @fedora
    Erase ubuntu-sso-client-4.1.2-1.fc18.noarch          @fedora-ubuntuone
    Erase ubuntuone-client-4.1.2-3.fc18.x86_64           @fedora-ubuntuone


No printer is installed
Code: Select all
$ lpq
lpq: Error - no default destination available.


But I still get the error when running wps.

F.

Attachments

Hi,
Thanks for your core dump file, finally we found the code may cause the crash.
But still need to check how this happen.
You need provide more information.

1.
download test.c from attachments
then compile with gcc test.c -o test
run like this ./test
see if prints OK or Errno:
If is OK, recompile with: gcc test.c -o test -m32 (You need to install glibc-devel.i686)
and excute ./test again see the output

2.
from terminal, execute getconf OPEN_MAX

the source of test.c is just a test (just printf), and you can check, there is no harm to your computer
and sorry for taking so much trouble to you
if you have no time or sth, just wait for next version (it may take a long time)

Attachments

Here is what you asked for:

$ gcc test.c -o test
$ ./test
OK
$ gcc test.c -o test -m32
$ ./test
Errno: 0
$ getconf OPEN_MAX
1024
I wonder how the 32bit version returns with errno value as 0
change the code as below

#include <pwd.h>
#include <stdio.h>
#include <errno.h>

static void
test_getpwuid()
{
uid_t uid;

uid = getuid();
printf("uid: %d\n", uid);

errno = 0;
if (getpwuid(uid))
{
printf("OK\n");
}
else
{
printf("Errno: %d\n", errno);
}
}

int main(int argc, char **argv)
{
test_getpwuid();

return 0;
}


compile with gcc test.c -o test -m32
and run again
thanks
I wonder is your libc (32bit version) or your account have something wrong
check the output value uid whether same with echo $UID (execute in terminal)

You can also create a new user, login and run WPS see if it works

Weitian Leung wrote: I wonder how the 32bit version returns with errno value as 0
change the code as below

#include <pwd.h>
#include <stdio.h>
#include <errno.h>

static void
test_getpwuid()
{
uid_t uid;

uid = getuid();
printf("uid: %d\n", uid);

errno = 0;
if (getpwuid(uid))
{
printf("OK\n");
}
else
{
printf("Errno: %d\n", errno);
}
}

int main(int argc, char **argv)
{
test_getpwuid();

return 0;
}


compile with gcc test.c -o test -m32
and run again
thanks
Here you are:

$ gcc test2.c -o test2 -m32
$ ./test2
uid: 1010
Errno: 0
$ echo $UID
1010
I also ran it as root and it works:

$ gcc test2.c -o test2 -m32
$ sudo ./test2
uid: 0
OK
My account is an ldap account. So I created a local user (john, UID=1000) ans ran the test:
Code: Select all
john@arabella ~$ gcc test2.c -o test2 -m32
john@arabella ~$ ./test2
uid: 1000
OK

It seems that my ldap account is the problem.
16 posts Page 1 of 2

Who is online

Users browsing this forum: No registered users and 1 guest

cron