Re: [PATCH] Fix gcc-4 warning in accept() call
- From
- H. Peter Anvin <hpa@zytor.com>
- Date
- Sep 30, 2005, 05:58 UTC
- Message-ID
- <433CD412.8020409@zytor.com>
- In-Reply-To
- <433CD2E7.5090909@zytor.com>
H. Peter Anvin wrote:
Show 18 quoted lines
> Junio C Hamano wrote: > >> Pavel Roskin <proski@gnu.org> writes: >> >> >>> gcc-4 warns about sign mismatch in pointers. Third argument in accept() >>> is socklen_t, which is unsigned. Since Linus doesn't like socklen_t >>> (see commit 7fa090844f7d1624c7d1ffc621aae6aec84a1110), let's use >>> unsigned int. >> >> >> Based on the list comments, I'd drop this patch -- it appears >> that gcc-4 warning is useless in this case. >> > > No, it's not. The proper type, if you don't use socklen_t, is int, not > unsigned int. >
Apparently I'm wrong; socklen_t is unsigned int at least on glibc for i386. I'll shut up now.
-hpa