Re: include_next with OSF/1 cc

[Re-adding bug-gnulib in CC. This is a reply to a reply to http://lists.gnu.org/archive/html/bug-gnulib/2009-01/msg00313.html.] Hi Gary, Gary V. Vaughan wrote: 2009/1/24 Bruno Haible br...@clisp.org: Gary V. Vaughan wrote: !THE FOLLOWING IS A REGRESSION OVER m4-1.4.12, WHICH COMPILES AND

Re: test-c-stack2.sh failure on HP-UX

2009/1/24 Bruno Haible br...@clisp.org: Gary V. Vaughan wrote: hppa2.0w-hp-hpux11.11-hpc m4 tests pass, gnulib fails: test-c-stack.sh ./test-c-stack.sh[7]: 1752 Memory fault(coredump) FAIL: test-c-stack.sh I reproduce this on HP-UX 11.11, without libsigsegv. I'm committing the change

Re: include_next with OSF/1 cc

2009/2/19 Bruno Haible br...@clisp.org: [Re-adding bug-gnulib in CC. This is a reply to a reply to http://lists.gnu.org/archive/html/bug-gnulib/2009-01/msg00313.html.] Thanks, I'm currently catching up with over 4 months of mailing lists, and probably not paying as much attention to those

[PATCH] Add 'safe-alloc' module

Hi, attached is a patch for a new module 'safe-alloc'. The module originated with libvirt[1] and was originally written by Dan Berrange. I touched it up for inclusion into gnulib. I've found them useful for some of my projects, and thought that putting them into gnulib would be much better than

Re: [PATCH] Add 'safe-alloc' module

Hi David, Thanks for proposing such a nice module! Uses of malloc/realloc really need a lot of attention and care, and I've also produced bug #7 (memory leak if realloc fails) more than once. For a module as basic as this one, in gnulib, I think it's essential that the copyright is with the FSF.

Re: fts: make find *much* faster on dirent.d_type-challenged FS

Karl Berry wrote: In compiling the binaries for TeX Live, where we try to use fairly old compilers and systems, the only platform which had problems with decl-after-stmt was v5 of the IBM C compiler for AIX 4.3. There are various reasons why the person who does those builds doesn't want to use

Re: [PATCH] Add 'safe-alloc' module

Bruno Haible wrote: For a module as basic as this one, in gnulib, I think it's essential that the copyright is with the FSF. Can you arrange with Daniel Berrange and/or his employer (Red Hat) that the copyright gets assigned? Please excuse my (mostly-ignorant) intrusion, but are you sure Red

Re: [PATCH] Add 'safe-alloc' module

Please excuse my (mostly-ignorant) intrusion, but are you sure Red Hat does not already have an assignment on file? (I'm sure I heard somewhere that they do at least for certain GNU projects...) Indeed, Red Hat does, as an employer. I have been told that RH employees don't even

Re: [PATCH] Add 'safe-alloc' module

Karl Berry wrote: Please excuse my (mostly-ignorant) intrusion, but are you sure Red Hat does not already have an assignment on file? (I'm sure I heard somewhere that they do at least for certain GNU projects...) Indeed, Red Hat does, as an employer. I have been told that RH

Re: fts: make find *much* faster on dirent.d_type-challenged FS

Have you built gcc on AIX 4.x? Personally? Not recently. I've done so in the past. The person who builds the AIX binaries for TL is Vladimir Volovich. As I recall, vvv told me that he did have gcc on AIX, but that the resulting binaries were both slower and bigger than the IBM C compiler

Re: fts: make find *much* faster on dirent.d_type-challenged FS

Karl Berry wrote: Anyway, my point here was merely that decl-after-stmt was not an unreasonable assumption for GNU packages. I guess that depends on your definition of portable, then. I tend to define that as conforms-to-POSIX-and-C89. YMMV. Not that there is any great need to continue this

Re: [PATCH] Add 'safe-alloc' module

Hi Bruno, On Thu, 2009-02-19 at 23:59 +0100, Bruno Haible wrote: Thanks for proposing such a nice module! All glory for this lies with Dan (and Jim Meyering for doing a preliminary review) For a module as basic as this one, in gnulib, I think it's essential that the copyright is with the