http://bugs.winehq.org/show_bug.cgi?id=36641
Sebastian Lackner sebastian@fds-team.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |sebastian@fds-team.de
--- Comment #1 from Sebastian Lackner sebastian@fds-team.de --- Created attachment 48863 --> http://bugs.winehq.org/attachment.cgi?id=48863 wininet: Fix invalid memory access in HTTP_QUERY_RAW_HEADERS.
Could you please test this patch? (for the first one)
I'm wondering about two things:
* Why didn't Valgrind complain about the 'for (...)' loop? 'len' is in bytes, so the old code should be wrong, but Valgrind doesn't notice it.
* As the memcpy(...) is now 2 bytes smaller, we only have one '\0' bytes at the end, but not two as the MSDN says. When changing the conditions for ERROR_INSUFFICIENT_BUFFER above then the wine test suite immediately complains about test failures. Does Windows maybe also write over the end of the user-provided buffer?