FS#29549 - [gvim] crash on mouse interaction
Attached to Project:
Arch Linux
Opened by H.pferd (stosch) - Saturday, 21 April 2012, 10:27 GMT
Last edited by Thomas Dziedzic (tomd123) - Sunday, 29 April 2012, 18:58 GMT
Opened by H.pferd (stosch) - Saturday, 21 April 2012, 10:27 GMT
Last edited by Thomas Dziedzic (tomd123) - Sunday, 29 April 2012, 18:58 GMT
|
Details
Description:
With the upgrade from gvim and vim-runtime 7.3.495-1 to 7.3.495-2 gvim (not vim!) crashes whenever I do anything with my mouse (scrolling, clicking etc) in it. Downgrading solves the problem for me. It is not related to any file extension (https://bugs.archlinux.org/task/29377). Error message: Vim: Caught deadly signal ILL Vim: Finished. |
This task depends upon
Closed by Thomas Dziedzic (tomd123)
Sunday, 29 April 2012, 18:58 GMT
Reason for closing: Fixed
Additional comments about closing: pkgrel 3
Sunday, 29 April 2012, 18:58 GMT
Reason for closing: Fixed
Additional comments about closing: pkgrel 3
I can't reproduce this, you will need to provide me with more info for this to possibly get solved.
https://wiki.archlinux.org/index.php/Step_By_Step_Debugging_Guide
I get the same behaviour with deactivated custom vimrc.
GDB gives no information.
I recompiled with CFLAG "-g -O0 -fbuiltin" and again got the same..
Strace gives the following: http://pastebin.com/afizV2mu
But sadly I cannot get any useful information from this.
The problem does not appear on my laptops which are a little newer than this old AthlonXP - could this be a reason?
If you need further information let me know.
When I enter "bt full" I receive "No stack.". There is also no core-file. Did I miss something?
By the way: My system is fully up to date.
At the gdb prompt type run
Then get the backtrace.
You will need to compile gvim w/ debug symbols.
export CFLAGS="-march=x86-64 ...
added in the last version in the PKGBUILD. I changed it to -march=native on my 32 bit system, and the problem vanished.
@Szunti:
Thank you for your tip. I tried it and it solved the problem for me.
But still it is strange, that the problem does not appear on my i686 laptop. Does this mean it is related to some processor flags?