|
From: Julian S. <js...@ac...> - 2003-08-01 00:36:44
|
Hi. I just made a branch tag VALGRIND_2_0_REALLY for the 2.0 branch. So I guess the head is now open for more breaking commits. Don't get VALGRIND_2_0_REALLY mixed up with VALGRIND_2_0_BRANCH; the latter holds the 1.9.X line, which at one point looked like becoming the 2.0 branch, but in the end didn't. I guess the VALGRIND_2_0_BRANCH tag should be got rid of really, since its presence is confusing. J |
|
From: Madhu M K. <mm...@ya...> - 2003-08-01 00:53:14
|
Hi, Julian Seward <js...@ac...> said on August 1,2003: > Hi. I just made a branch tag VALGRIND_2_0_REALLY for the > 2.0 branch. So I guess the head is now open for more breaking > commits. Don't get VALGRIND_2_0_REALLY mixed up with Does the statistical version of the --gen-suppressions=file patch stand a better chance now? Short description: Instead of --gen-suppressions=yes/no you also have the ability to supply a filename --gen-suppresions=foo at which foo.pidXXXX is created. This file contains the list of suppressions generated sorted by the frequency at which they occured during that program. This traps the same "leak/uninitialized value" present in a library but called from multiple contexts. While this file is by itself useful (esp the low frequency suppressions), I use the smallest startup shutdown sequence in my program to grab all the noise and then test each option to find leaks that are in my control (v/s system libraries) for example. Comments, ayes, nays are welcome :) Cheerio, M Madhu M Kurup /* Nemo Me Impune Lacessit */ mmk at yahoo-inc dt com |
|
From: Nicholas N. <nj...@ca...> - 2003-08-03 10:08:55
|
On Thu, 31 Jul 2003, Madhu M Kurup wrote: > Does the statistical version of the --gen-suppressions=file patch stand > a better chance now? Short description: > > Comments, ayes, nays are welcome :) It does seem like a good idea... mostly a matter of time to get around to checking and committing it. It's not the only patch/feature/bug that hasn't been addressed :) N |