Kees Cook wrote:
This patch fixes a number of cases where error conditions are untested,
which cause problems when compiling with -D_FORTIFY_SOURCE=2.
Hi! Sorry for the late reply; just got back from vacation.
I'm not super-keen on diverging from upstream in general unless
absolutely necessary. Most of the changes in the patch are "tool
patches"; they don't really fix problems in the code itself, but just
issues with Fortify. I'm not inclined to include them unless they come
down from upstream.
The only two parts that fix real issues, as I see it, are:
- the daemonization code
- the pipe flush (read) in lib/arch/CArchNetworkBSD.cpp
The daemonization code is just ugly. I smell a future bug in the code
that relies on sequential file descriptor assignment by the OS. So that
needs more work.
The pipe flush code is probably a little more correct, but I'd also like
to see better handling of unexpected error conditions.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
