DynamoRIO / DynamoRIO/drmemory
confirm and file bugs found in cygwin1.dll
- Dominant language
- C
- Stars
- 2.7k
- Forks
- 290
- PR merge metrics
- No merged PRs in 30d
Description
_From [derek.br...@gmail.com](https://code.google.com/u/117968039472581148324/) on December 10, 2010 17:57:53_
PR 425237
hellog++.exe is "hello world" compiled with cygwin's g++
current status: 7 positives: 3 false that need per-bit (PR 408523), and 4 real bugs
should re-visit what I thought were real bugs here, with the cygwin source in hand,
and file them if confirmed. the kernel32 one needs to be re-examined: something about
init-to-0 heap in my notes but unclear.
11/11/08, after strlen/strcpy exceptions as part of and/or dynamic checks:
% grep -A 1 UNINIT drmemory.[0-9]\* | grep -vE 'UNINIT|--' | sort | uniq -c | sort
1 0x6106b5e2 f6 85 97 fd ff ff 04 test 0xfffffd97(%ebp) $0x04
=> may need per-bit checking
1 0x610c38b7 f6 c2 02 test %dl $0x02
=> definitely need per-bit checking
3 0x77c5b856 84 56 70 test 0x70(%esi) %dl
=> definitely need per-bit checking
4 0x6107e0b8 83 7d ec 01 cmp 0xffffffec(%ebp) $0x00000001
=> almost certainly real bug
5 0x610677c6 80 fb 2f cmp %bl $0x2f
=> looks like real bug
6 0x610679a9 38 e0 cmp %al %ah
=> looks like real bug
27 0x77e65e0a 66 39 7c 43 fe data16 cmp 0xfffffffe(%ebx,%eax,2) %d
=> looks like real bug
on laptop, don't have 0x70 or the last 2
_Original issue: http://code.google.com/p/drmemory/issues/detail?id=131_
Contributor guide
Assessment
This issue has not been assessed yet.