From: SourceForge.net <no...@so...> - 2008-11-18 21:37:15
|
Bugs item #2310660, was opened at 2008-11-18 18:47 Message generated for change (Comment added) made by dkf You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=112997&aid=2310660&group_id=12997 Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: 69. Events Group: obsolete: 8.4.13 Status: Open Resolution: None Priority: 5 Private: No Submitted By: Michael Smith (mtsmith) Assigned to: Jeffrey Hobbs (hobbs) Summary: Tk tests produce FMR in Purify on Linux Initial Comment: Parts of the Tk test suite from release 8.4.13 produces an Free Memory Read (FMR) when run with Purify. The specific tests are event-double-click-drag-1.1, textTag-15.2, and textTag-15.3. I've included a minimal test case derived from those tests that demonstrates the problem, and the log file from running with Purify. I ran this test on Redhat Enterprise Linux 4.0 U6, using gcc 4.2.2. I built Tcl 8.4.13 from source with a shared library, then built Tk 8.4.13 from source with a shared library and based off the Tcl I'd built. Then I rebuilt wish8.4 with IBM Rational Purify 7.0.0.0-008.P and ran the attached test file. Notes from my own debugging: the FMR occurs in SortTags called from line 1355 in tkTextTag.c, happens whether the fast sort or qsort is used, seems to happen on a tag named "sel" (the other members are "x" and "y"), happens on the 2nd call when SortTags is called twice in a row with the same tagArrayPtr pointer (first time in the entire Tk test suite that happens). ---------------------------------------------------------------------- >Comment By: Donal K. Fellows (dkf) Date: 2008-11-18 21:34 Message: That's a release that's 2.5 years old from a branch that's now EOL'ed. Care to test with something more recent? (Even 8.4.19 would be more sensible, though 8.5.5 would be more helpful; the text widget has had quite a lot of revisions in the 8.5 series, so bug reports *can't* be directly compared.) ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=112997&aid=2310660&group_id=12997 |