make - Patches: patch #7534, Use-after-free() when eval in...
You are not allowed to post comments on this tracker with your current authentication level.
patch #7534: Use-after-free() when eval in variable assignments reassigns the same variable
Submitter: | Lubomir Rintel <lkundrak> | ||
Submitted: | Tue 26 Apr 2011 09:55:56 PM UTC | ||
Votes: | 10 | ||
Category: | None | Priority: | 7 - High |
Status: | Done | Privacy: | Public |
Assigned to: | psmith | Open/Closed: | Closed |
Fixed Release: | CVS |
Jump to the original submission
Mon 29 Aug 2011 04:25:22 PM UTC, comment #9: |
Paul D. Smith <psmith> |
Wed 24 Aug 2011 01:00:20 PM UTC, comment #8: Ping? |
Lubomir Rintel <lkundrak> |
Wed 03 Aug 2011 10:34:46 AM UTC, comment #7: /me feels like noone loves him. |
Lubomir Rintel <lkundrak> |
Sat 02 Jul 2011 09:41:23 PM UTC, comment #6: Sorry I've been on vacation; just got back. I'll try to look at this this weekend. Thanks. |
Paul D. Smith <psmith> |
Fri 01 Jul 2011 12:26:50 PM UTC, comment #5: Hello,
|
Lubomir Rintel <lkundrak> |
Wed 15 Jun 2011 08:48:20 AM UTC, comment #4: Paul, thanks for your response. The example in the comment both contains typoes and is incomplete, sorry for that. The bug is indeed still visible in the code. Also, I've tried it with fresh CVS checkout my (Fedora 14/i386) system (valgrind output filtered for clarity):
|
Lubomir Rintel <lkundrak> |
Sun 12 Jun 2011 07:22:12 PM UTC, comment #3: I've tried both the example in the comment below and also the example shown in the patch file, under valgrind, and I can't see any memory error with the current CVS HEAD version of GNU make.
|
Paul D. Smith <psmith> |
Wed 04 May 2011 02:44:18 PM UTC, comment #2: Ping? |
Lubomir Rintel <lkundrak> |
Tue 26 Apr 2011 10:17:37 PM UTC, comment #1: Needless to say; test suite passes well and the problem was checked to go away with valgrind. |
Lubomir Rintel <lkundrak> |
Tue 26 Apr 2011 09:55:56 PM UTC, original submission:
In real world, such memory corruption occurs when using eval constructs for lazy assignment:
|
Lubomir Rintel <lkundrak> |
Depends on the following items: None found
Items that depend on this one: None found
There are 10 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.
Follow 9 latest changes.
Date | Changed by | Updated Field | Previous Value | => | Replaced by |
---|---|---|---|---|---|
2013-05-17 | vshih | Carbon-Copy | - | Added vshih | |
2011-08-29 | psmith | Status | None | Done | |
Assigned to | None | psmith | |||
Open/Closed | Open | Closed | |||
Fixed Release | None | CVS | |||
2011-04-26 | lkundrak | Carbon-Copy | Removed lkundrak | - | |
2011-04-26 | lkundrak | Attached File | - | Added 0001-Always-copy-the-string-before-expanding-it.patch, #23308 | |
Attached File | - | Added make-3.81-copy-on-expand.patch, #23309 | |||
Carbon-Copy | - | Added lkundrak |
Powered by Savane 3.13-d3ae.
Corresponding source code
This really should have been filed as a bug... because it is a bug. Patches are more for new features although I would prefer those to be filed as bugs as well. I'm going to try to figure out how to close this tracker completely without losing any information in it.
Anyway I think I fixed it in the latest CVS. Thanks!