Segmentation fault

[Zaister]

Done! And thanks.

[Zaister]

I'm getting more spurious segmentationfaults now that the strstr issue seems to be handled. Here's a backtrace from a crash I just experienced from calling the program with two files as arguments:

Program received signal SIGSEGV, Segmentation fault.
0xf7d7303a in QPixmap_isNull () from /opt/beyondcompare/lib64/libqtc.so.1
(gdb) backtrace
#0 0xf7d7303a in QPixmap_isNull () from /opt/beyondcompare/lib64/libqtc.so.1
#1 0x080e3689 in ?? ()
#2 0x081805fc in ?? ()
#3 0x081a827e in ?? ()
#4 0x084b24c5 in ?? ()
#5 0x084dd5d3 in ?? ()
#6 0x084db2a0 in ?? ()
#7 0x08197e01 in ?? ()
#8 0xf7792dab in makecontext () from /lib32/libc.so.6
#9 0x081a8cd4 in ?? ()
#10 0x08afe858 in ?? ()
Backtrace stopped: previous frame inner to this frame (corrupt stack?)

Can you help with this?

[David]

Yes. There is a fix in the repository for this that will be in the next release.

[Zaister]

Can you project a date for that release? These crashes are really annoying.

Thank you.

[Tim]

We hope to get a release out today.

[Zaister]

That's' great!

[Zaister]

The new version seems to work fine now, thanks!

[Zaister]

Here's how I patched my glibc 2.18, so it no longer tries to use the SSE42 version of strstr:

extract the glibc-2.18.tar.xz archive, then open the file

glibc-2.18/sysdeps/i386/i686/multiarch/strstr-c.c

in an editor. In line 26, replace "HAS_SSE4_2" with "false". Then save the file and compile and install glibc as usual.