Vladimir Kotal wrote:
> New webrev is here:
> http://cr.opensolaris.org/~vkotal/daemon_libc-4471189.onnv.checks/
A few very minor nits:
- open(2) actually returns -1 on error, so I prefer "== -1" instead
of "< 0". Nobody's going to have 2G open descriptors, so it doesn't
matter, but still ...
- there's no need to check whether fd is equal to the dup2 target
before calling dup2(2); that system call is defined to do exactly
nothing if the fd arguments are equal. (But if you feel it's
clearer or somehow better to check first, go ahead.)
- got lint? (Missing (void) casts on those new close(2) calls.)
Other than that, +1.
--
James Carlson 42.703N 71.076W <carlsonj at workingcode.com>