Re: [PATCH 1/4] nonblocking: provide O_NONBLOCK for mingw

2011-04-19 Thread Bruno Haible
Hi Eric, Thanks for the review. Eric Blake wrote: > > --- doc/posix-functions/open.texi.orig Sun Apr 17 23:18:36 2011 > > +++ doc/posix-functions/open.texi Sun Apr 17 23:08:20 2011 > > @@ -14,6 +14,10 @@ > > directory, on some platforms: > > FreeBSD 7.2, AIX 7.1, HP-UX 11.00, Solaris 9,

Re: [PATCH 1/4] nonblocking: provide O_NONBLOCK for mingw

2011-04-18 Thread Eric Blake
On 04/17/2011 03:24 PM, Bruno Haible wrote: > Hi Eric, > >> * modules/nonblocking (Depends-on): Add open. > > This looks like a dependency that will surprise people. If a program needs the > 'nonblocking' module for pipe or socket communication, why would it need to > compile a replacement for op

Re: [PATCH 1/4] nonblocking: provide O_NONBLOCK for mingw

2011-04-17 Thread Bruno Haible
Hi Eric, > * modules/nonblocking (Depends-on): Add open. This looks like a dependency that will surprise people. If a program needs the 'nonblocking' module for pipe or socket communication, why would it need to compile a replacement for open(). If a program uses open(), then it can use the open(

Re: [PATCH 1/4] nonblocking: provide O_NONBLOCK for mingw

2011-04-17 Thread Bruno Haible
Eric Blake wrote: > * lib/fcntl.in.h (O_NONBLOCK): Guarantee non-zero definition when > nonblocking module is in use. I think this deserves some comment. (It is documented in doc/posix-headers/fcntl.texi but sometimes we are looking at the code only.) 2011-04-17 Bruno Haible nonblock

[PATCH 1/4] nonblocking: provide O_NONBLOCK for mingw

2011-04-08 Thread Eric Blake
Mingw is the only known platform that lacks O_NONBLOCK (possibly via the alternate spelling O_NDELAY). But mingw also lacks any files where open() needs to enforce non-blocking behavior, and lacks openat(), so it is relatively simple to provide a non-zero flag. A future patches will make use of O