DynamoRIO / DynamoRIO/drmemory
Should replace memcpy_s and similar functions?
- Dominant language
- C
- Stars
- 2.7k
- Forks
- 290
- PR merge metrics
- No merged PRs in 30d
Description
_From [timurrrr@google.com](https://code.google.com/u/timurrrr@google.com/) on May 05, 2011 09:33:45_
On our Vista bot I've noticed reports like this: http://build.chromium.org/p/client.drmemory/builders/win-vista_x64-drm/builds/946/steps/Run%20TSan%20tests/logs/summary%3A%2068%20report%28s%29 Error `#1`: UNADDRESSABLE ACCESS: writing 0x0fdd0048-0x0fdd004c 4 byte(s)
@0:00:05.360 in thread 2164
0x77ca0d4d ntdll.dll!RtlInitializeCriticalSectionEx
0x77c5d6f4 ntdll.dll!RtlInitializeCriticalSection
0x77c68265 ntdll.dll!RtlUnicodeToOemN
0x77c68424 ntdll.dll!RtlUnicodeToOemN
0x77cb95f3 ntdll.dll!RtlImageRvaToVa
0x77c5aab9 ntdll.dll!RtlTryEnterCriticalSection
0x77c5a4df ntdll.dll!RtlDllShutdownInProgress
0x7759a84a msvcrt.dll!memcpy_s
0x7759a8a3 msvcrt.dll!memcpy_s
0x7759a015 msvcrt.dll!lock
0x77599fc8 msvcrt.dll!lock
0x77c79354 ntdll.dll!RtlDestroyQueryDebugBuffer
While I believe no sane memcpy_s would call RtlInitializeCriticalSection (or will it?)
and the problem is likely something like wrong symbols
one question comes to surface - should we replace memcpy_s and
other string/memory-related functions with _s suffix?
_Original issue: http://code.google.com/p/drmemory/issues/detail?id=383_
Contributor guide
Assessment
This issue has not been assessed yet.