Re: [Inkscape-user] RPM build errors
On Sat, 4 Dec 2004, Jonathan Leighton [Turnip] wrote:
Okay. I took them all out, and tried to run inkscape, but nothing happened. As in zero, zilch, nada. The cursor just sat there on the line beneath the "inkscape" command. So I quit that, and put all the extensions back in. I then took them out one by one, and when I took out the one called "gimpgrad.inx", it again wouldn't do anything when I ran the "inkscape" command. So I believe we have a culprit, although I'm not entirely sure what I should do now.
Hmm interesting, that wouldn't have been the one I'd have guessed.
So to recap, you're getting a failure during Extension initialization where it complains about an improperly freed pointer, presumably a string, that appears related to the gimpgrad.inx extension.
Ted can you help give some guidance on this?
Just out of interest, do you think this problem is specific to me, or would it happen to anyone using the RPM I built? Because if it would happen to anyone then I guess it's more serious.
Aha, this is why you shanghai a couple friends into testing it. ;-)
Go ahead and see if you can find someone to test it, and see if they get the same problem. That'll be good to know.
Let's move this over to the inkscape-devel@ list, it's gotten a bit too nuts and bolts for the user list.
Bryce
I made a post on FedoraForum.org asking for people to test my RPM. If anyone here wants to try it, I've uploaded it to http://turnipspatch.com/misc/inkscape-0.40-1.fc3.i386.rpm
I hope Ted can shed some light on this :(.
Cheers Jonathan
Bryce Harrington wrote:
On Sat, 4 Dec 2004, Jonathan Leighton [Turnip] wrote:
Okay. I took them all out, and tried to run inkscape, but nothing happened. As in zero, zilch, nada. The cursor just sat there on the line beneath the "inkscape" command. So I quit that, and put all the extensions back in. I then took them out one by one, and when I took out the one called "gimpgrad.inx", it again wouldn't do anything when I ran the "inkscape" command. So I believe we have a culprit, although I'm not entirely sure what I should do now.
Hmm interesting, that wouldn't have been the one I'd have guessed.
So to recap, you're getting a failure during Extension initialization where it complains about an improperly freed pointer, presumably a string, that appears related to the gimpgrad.inx extension.
Ted can you help give some guidance on this?
Just out of interest, do you think this problem is specific to me, or would it happen to anyone using the RPM I built? Because if it would happen to anyone then I guess it's more serious.
Aha, this is why you shanghai a couple friends into testing it. ;-)
Go ahead and see if you can find someone to test it, and see if they get the same problem. That'll be good to know.
Let's move this over to the inkscape-devel@ list, it's gotten a bit too nuts and bolts for the user list.
Bryce
On Sat, 2004-12-04 at 15:19 -0800, Bryce Harrington wrote:
On Sat, 4 Dec 2004, Jonathan Leighton [Turnip] wrote:
Okay. I took them all out, and tried to run inkscape, but nothing happened. As in zero, zilch, nada. The cursor just sat there on the line beneath the "inkscape" command. So I quit that, and put all the extensions back in. I then took them out one by one, and when I took out the one called "gimpgrad.inx", it again wouldn't do anything when I ran the "inkscape" command. So I believe we have a culprit, although I'm not entirely sure what I should do now.
Hmm interesting, that wouldn't have been the one I'd have guessed.
So to recap, you're getting a failure during Extension initialization where it complains about an improperly freed pointer, presumably a string, that appears related to the gimpgrad.inx extension.
Ted can you help give some guidance on this?
Hmm, not really sure. But, I would say that the GIMP gradient one is the only one that is based on loadable libraries. The only issue there is that loadable libraries aren't loaded unless the extension is actually used (they're loaded on the fly), so it is probably not an issue with loading the library directly.
I don't seem to be getting e-mail from the user's list -- did you post any GDB backtrace, or where inkscape is stopping? There is some code that is unique to plug-in based extensions -- not alot, but there could be something funny there.
Also, is there anything different about your setup? Russian filesystem? Chinese menus? Running on a PIC?
--Ted
Ted Gould wrote:
On Sat, 2004-12-04 at 15:19 -0800, Bryce Harrington wrote:
On Sat, 4 Dec 2004, Jonathan Leighton [Turnip] wrote:
Okay. I took them all out, and tried to run inkscape, but nothing happened. As in zero, zilch, nada. The cursor just sat there on the line beneath the "inkscape" command. So I quit that, and put all the extensions back in. I then took them out one by one, and when I took out the one called "gimpgrad.inx", it again wouldn't do anything when I ran the "inkscape" command. So I believe we have a culprit, although I'm not entirely sure what I should do now.
Hmm interesting, that wouldn't have been the one I'd have guessed.
So to recap, you're getting a failure during Extension initialization where it complains about an improperly freed pointer, presumably a string, that appears related to the gimpgrad.inx extension.
Ted can you help give some guidance on this?
Hmm, not really sure. But, I would say that the GIMP gradient one is the only one that is based on loadable libraries. The only issue there is that loadable libraries aren't loaded unless the extension is actually used (they're loaded on the fly), so it is probably not an issue with loading the library directly.
I don't seem to be getting e-mail from the user's list -- did you post any GDB backtrace, or where inkscape is stopping? There is some code that is unique to plug-in based extensions -- not alot, but there could be something funny there.
Also, is there anything different about your setup? Russian filesystem? Chinese menus? Running on a PIC?
--Ted
Hi Ted,
I can assure you I'm definitely not running on a PIC ;-). No, I'm pretty sure that don't have anything particularly out of the ordinary about my setup.
The backtrace is here: http://sourceforge.net/mailarchive/message.php?msg_id=10219286, complete thread here: http://sourceforge.net/mailarchive/forum.php?thread_id=6078715&forum_id=...,
and for your convenience I'll repost the backtrace ;-)
[turnip@...33... ~]$ gdb inkscape GNU gdb Red Hat Linux (6.1post-1.20040607.43rh) Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "i386-redhat-linux-gnu"...(no debugging symbols found)...Using host libthread_db library "/lib/tls/libthread_db.so.1".
(gdb) run Starting program: /usr/bin/inkscape (no debugging symbols found)...(no debugging symbols found)...[Thread debuggingusing libthread_db enabled] [New Thread -151128384 (LWP 7673)] (no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...*** glibc detected *** free(): invalid pointer: 0x083332b0 ***
Program received signal SIGABRT, Aborted. [Switching to Thread -151128384 (LWP 7673)] 0x00ad47a2 in _dl_sysinfo_int80 () from /lib/ld-linux.so.2 (gdb) bt #0 0x00ad47a2 in _dl_sysinfo_int80 () from /lib/ld-linux.so.2 #1 0x00b14955 in raise () from /lib/tls/libc.so.6 #2 0x00b16319 in abort () from /lib/tls/libc.so.6 #3 0x00b4da1b in malloc_printerr () from /lib/tls/libc.so.6 #4 0x00b4e4ba in free () from /lib/tls/libc.so.6 #5 0x00836445 in operator delete () from /usr/lib/libstdc++.so.6 #6 0x0081759f in std::string::_Rep::_M_destroy () from /usr/lib/libstdc++.so.6 #7 0x0828ab18 in Inkscape::Extension::Dependency::check () #8 0x082890e8 in Inkscape::Extension::Extension::check () #9 0x0828c50f in Inkscape::Extension::Input::check () #10 0x0828b0f9 in Inkscape::Extension::Dependency::check () #11 0x08289e69 in Inkscape::Extension::DB::foreach () #12 0x0828b20e in Inkscape::Extension::init () #13 0x080c53b2 in sp_main_gui () #14 0x080c55ae in main () (gdb)
Cheers.
P.S. I haven't managed to persuade anyone to test my package themselves; they want me to rebuild it according to the Fedora packaging guidelines, so it looks like I might do that. Sometime. Perhaps it will produce different results? Who knows... Anyway, I'll keep the current package, just in case.
participants (3)
-
Bryce Harrington -
Jonathan Leighton [Turnip] -
Ted Gould