On Sunday 04 November 2007 22:45, Nicholas Nethercote wrote: > On Sun, 4 Nov 2007, Julian Seward wrote: > >> Tom, can you send me diffs for some of the Massif ones? Any idea why so > >> many tests fail on this machine? > > > > I have a vanilla Red Hat 7.3 installation on QEMU. Attached are the > > results from that. > > The single cause of all of these appears to be that each program is doing a > heap allocation of size 24 bytes very early on, ie. before any of the > allocations in the main part of the C program are done. We have to > understand why that's happening before we can fix these.
Perhaps glibc, or ld.so and its myriad function calls, etc, are doing an allocation before we get to main. Do your tests assume that there are zero allocations prior to main being called? If so, how about having a massif-specific client request which you call at the start of main, to say "forget everything and start with a clean slate" ? > It may explain > why so many Memcheck results are also going wrong for this machine. Ah, good point. J ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Valgrind-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/valgrind-developers
