Closed Bug 502972 Opened 11 years ago Closed 11 years ago
Another allocator mismatch in sdrtest
I have been running all.sh with NSPR and NSS built with different allocators on Windows - one built with the standard allocator, the other with the debug allocator. I found the following problem : ntdll.dll!7d61ea56() msvcr90d.dll!_free_dbg(void * pUserData=0x028585f8, int nBlockUse=1) Line 1258 + 0xd bytes C++ msvcr90d.dll!free(void * pUserData=0x028585f8) Line 49 + 0xb bytes C++ > sdrtest.exe!main(int argc=9, char * * argv=0x020d3410) Line 440 + 0x10 bytes C sdrtest.exe!__tmainCRTStartup() Line 586 + 0x19 bytes C sdrtest.exe!mainCRTStartup() Line 403 C kernel32.dll!7d4e7d42() This is more complex to resolve than the previous allocation mismatch in sdrtest (bug 502965), because in this case there are multiple code paths, and apparently some of them allocate/reallocate result with libc, and others with NSPR/NSS.
Assignee: nobody → julien.pierre.boogz
Priority: -- → P2
Attachment #387346 - Flags: review?(nelson) → review+
Comment on attachment 387346 [details] [diff] [review] Fix mismatches Why is this a separate bug from bug 502965? This patch is correct, but again, it would be better in several respects to use SECITEM_ZFreeItem on the SECItem object rather than acting on the .data member directly.
Nelson, This is a separate bug because it was found in a different run of tests, after I fixed the first bug. I checked in the fix using SECITEM_ZfreeItem . Checking in sdrtest.c; /cvsroot/mozilla/security/nss/cmd/sdrtest/sdrtest.c,v <-- sdrtest.c new revision: 1.16; previous revision: 1.15 done
Status: NEW → RESOLVED
Closed: 11 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.