Re: new module alignof

2009-06-07 Thread Bruno Haible
Eric Blake wrote: > I fixed the typo (CHECK only takes a single token argument): Shame on me. > > + CHECK (int64_t) > > I didn't touch this, but should we still want to cater to C89 platforms > that lack 64 bit integers Right you are. Yes, you cannot generally assume the presence of working int

Re: new module alignof

2009-06-06 Thread Eric Blake
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Bruno Haible on 5/31/2009 1:51 PM: > Eric Blake wrote: >> I noticed you just pushed test-alignof.c. Should this file also check >> 'long double', and (where available) 'long long int', particularly since >> these types often have smaller

Re: new module alignof

2009-05-31 Thread Bruno Haible
Eric Blake wrote: > I noticed you just pushed test-alignof.c. Should this file also check > 'long double', and (where available) 'long long int', particularly since > these types often have smaller alignment than size? Good point here as well. I'm adding them: 2009-05-31 Bruno Haible

Re: new module alignof

2009-05-31 Thread Bruno Haible
Eric Blake wrote: > So if I understand correctly, you are trading one set of portability > problems (the AIX workaround guesses wrong for char and short) for another > (alignof() must not be used to initialize an enumerator, but can be used > in all other contexts). Shouldn't this at least be docu

Re: sys_socket: sockaddr_storage, new module alignof

2009-05-12 Thread Eric Blake
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Bruno Haible on 5/12/2009 7:22 PM: >> New module 'alignof'. >> * lib/alignof.h: New file. >> * modules/alignof: New file. I noticed you just pushed test-alignof.c. Should this file also che

Re: sys_socket: sockaddr_storage, new module alignof

2009-05-12 Thread Eric Blake
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Bruno Haible on 5/12/2009 7:22 PM: > the special cases for AIX cc,xlc and HP-UX cc are not right in general. > They are only right in the context of malloca.h. So if I understand correctly, you are trading one set of portability problems

Re: sys_socket: sockaddr_storage, new module alignof

2009-05-12 Thread Bruno Haible
In this commit: > 2009-05-08 Bruno Haible > > New module 'alignof'. > * lib/alignof.h: New file. > * modules/alignof: New file. > * lib/malloca.h: Include alignof.h. > (sa_alignof): Remove macro. > (sa_alignment_*): Use th

Re: sys_socket: sockaddr_storage, new module alignof

2009-05-10 Thread Simon Josefsson
Bruno Haible writes: > Simon Josefsson wrote: >> +#define _SS_PADSIZE (_SS_SIZE - (max (sizeof (sa_family_t), \ >> + alignof (__ss_aligntype)) \ > > 'max' is not a predefined macro. I'm applying this fix: Thanks. /Simon

Re: sys_socket: sockaddr_storage, new module alignof

2009-05-08 Thread Bruno Haible
Simon Josefsson wrote: > +#define _SS_PADSIZE (_SS_SIZE - (max (sizeof (sa_family_t), \ > + alignof (__ss_aligntype)) \ 'max' is not a predefined macro. I'm applying this fix: 2009-05-08 Bruno Haible * lib/sys_socket.in.h (_SS_PADSIZ

Re: sys_socket: sockaddr_storage, new module alignof

2009-05-08 Thread Simon Josefsson
The test for sockaddr_storage failed incorrectly under mingw because the HAVE_WS2TCPIP_H wasn't set. The patch below moves back the test of ws2tcpip.h before the sockaddr_storage test. There is something like a catch-22 here: to test for sockaddr_storage, we need to test for ws2tcpip.h, but to te

Re: sys_socket: sockaddr_storage, new module alignof

2009-05-08 Thread Simon Josefsson
Simon Josefsson writes: > Bruno, I pushed the following trivial fix because I got 'make dist' > failures. Sigh, I should have tested more -- there is an automatic EXTRA_DIST added, and the error I got was caused by other problems. I reverted the patch. Sorry about the noise. /Simon

Re: sys_socket: sockaddr_storage, new module alignof

2009-05-08 Thread Simon Josefsson
Bruno, I pushed the following trivial fix because I got 'make dist' failures. /Simon >From 0773c46ee42a43177fa958f2437a8c45e748cc06 Mon Sep 17 00:00:00 2001 From: Simon Josefsson Date: Fri, 8 May 2009 17:06:26 +0200 Subject: [PATCH] modules/alignof (Makefile.am): Dist alignof.h. --- ChangeLog

Re: sys_socket: sockaddr_storage, new module alignof

2009-05-08 Thread Simon Josefsson
Bruno Haible writes: > Simon Josefsson wrote: >> Looks fine to me. If you push it, I can make sys_socket depend on it. > > Pushed. Thanks. I have pushed the patch below. /Simon >From 014f60069ce88c16683c533813b2463771ac2d0b Mon Sep 17 00:00:00 2001 From: Simon Josefsson Date: Fri, 8 May 200

Re: sys_socket: sockaddr_storage, new module alignof

2009-05-08 Thread Bruno Haible
Simon Josefsson wrote: > Looks fine to me. If you push it, I can make sys_socket depend on it. Pushed. Bruno

Re: sys_socket: sockaddr_storage, new module alignof

2009-05-08 Thread Simon Josefsson
Bruno Haible writes: > Simon Josefsson wrote: >> +#define _SS_PADSIZE (_SS_SIZE - (max (sizeof (sa_family_t), \ >> + alignof (__ss_aligntype)) \ >> + + sizeof (__ss_aligntype))) > > Fine, except that 'alignof' is not

Re: sys_socket: sockaddr_storage, new module alignof

2009-05-08 Thread Bruno Haible
redefined macro. We have a definition of it in "malloca.h" and in some .c files from glibc. OSF/1 has it in a file , but it contains nothing better than our alignof macro. I propose to create a module like this: 2009-05-08 Bruno Haible New module 'alignof&#