"Richard W.M. Jones" <[EMAIL PROTECTED]> writes:
> I have a general question: Does Gnulib get any sort of nightly
> build/test? On [real] Windows machines?
http://autobuild.josefsson.org/gnulib/
There is a mingw+wine build, but not a real Windows build. If you or
anyone can setup and maintain
Hi Jim,
> If someone is interested enough to time things on HP-UX
> and finds that there's a file system type (probably memory backed)
> that it'd be good to exempt, then it might be worthwhile to
> pursue this.
I didn't mean to discuss for which filesystems on HP-UX which optimization
may be wor
Hi,
posix_spawn exists on AIX 5.3 and 6.1, but it is unusable. Rainer Tammer
identified two bugs:
- failure to execute any program in the child process, when the parent
program uses POSIX threads.
- when the child process fails to execute the given program, it outputs
the stdio buffers
Jim Meyering wrote:
> > Should this also depend on the errno module?
>
> I wondered, but figured it's not needed since the code is
> windows-specific.
The 'errno' module is useful on mingw, but is also useful on Unix platforms
like OpenBSD and OSF/1.
The 'errno' module is not needed here because
Richard W.M. Jones wrote:
> This patch implements fsync for Windows.
Nice and quite well done. Bravo for having understood the many idioms used
in gnulib.
Just two minor comments:
- The phrase "cross-compilers like MinGW" is misleading.
MinGW is is often used natively, not cross-compiled.
This propagates a change from coreutils and
also fixes a typo in fts.m4:
diff --git a/lib/fts.c b/lib/fts.c
index 95d0c7a..a55a98d 100644
--- a/lib/fts.c
+++ b/lib/fts.c
@@ -934,8 +934,8 @@ fts_children (register FTS *sp, int instr)
return (sp->fts_child);
}
-#if defined HAVE_SYS_VFS_H &
I have a general question: Does Gnulib get any sort of nightly
build/test? On [real] Windows machines?
And a general comment: Some work colleagues and I are working on a
Windows cross-compiler project for Fedora[1]. The number one big
porting problem has been lack of flock/lockf/fcntl file locki
On Wed, Oct 01, 2008 at 05:51:23PM +, Eric Blake wrote:
> Richard W.M. Jones redhat.com> writes:
>
> >
> > Like the earlier patch, but this relicenses to LGPLv2+, includes
> > license text, and assigns everything to the FSF.
>
> Do you have copyright on file for gnulib yet?
No, but the gen
Jim Meyering <[EMAIL PROTECTED]> wrote:
> Eric Blake <[EMAIL PROTECTED]> wrote:
>> Richard W.M. Jones redhat.com> writes:
>>> Like the earlier patch, but this relicenses to LGPLv2+, includes
>>> license text, and assigns everything to the FSF.
>
> Addressing Eric's feedback, and tweaking copyright
Eric Blake <[EMAIL PROTECTED]> wrote:
> Richard W.M. Jones redhat.com> writes:
>> Like the earlier patch, but this relicenses to LGPLv2+, includes
>> license text, and assigns everything to the FSF.
Addressing Eric's feedback, and tweaking copyright and a comment,
FYI, this is the incremental I'm
Eric Blake <[EMAIL PROTECTED]> wrote:
> Richard W.M. Jones redhat.com> writes:
>> Like the earlier patch, but this relicenses to LGPLv2+, includes
>> license text, and assigns everything to the FSF.
>
> Do you have copyright on file for gnulib yet?
No.
Technically he doesn't need it, since the wo
Richard W.M. Jones redhat.com> writes:
>
> Like the earlier patch, but this relicenses to LGPLv2+, includes
> license text, and assigns everything to the FSF.
Do you have copyright on file for gnulib yet?
> --- a/NEWS
> +++ b/NEWS
> @@ -6,6 +6,8 @@ User visible incompatible changes
>
> Date
"Richard W.M. Jones" <[EMAIL PROTECTED]> wrote:
> Like the earlier patch, but this relicenses to LGPLv2+, includes
> license text, and assigns everything to the FSF.
This looks fine. Thanks, Rich.
I'll wait for feedback, then push tomorrow morning.
Like the earlier patch, but this relicenses to LGPLv2+, includes
license text, and assigns everything to the FSF.
Rich.
--
Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones
Read my OCaml programming blog: http://camltastic.blogspot.com/
Fedora now supports 68 OCaml pack
Updated patch, including Jim's tests and also a note pointing to the
original sources in sqlite.
Rich.
--
Richard Jones, Emerging Technologies, Red Hat http://et.redhat.com/~rjones
Read my OCaml programming blog: http://camltastic.blogspot.com/
Fedora now supports 68 OCaml packages (the OPEN a
On Wed, Oct 01, 2008 at 06:06:31PM +0200, Jim Meyering wrote:
> To make gnulib-tool's --with-tests option work, you can add
> this to your change set:
OK, I tried your suggested patch and the tests passed under Linux and
under the Fedora cross-compiler / Wine environment. I'm unable to
test it on
"Richard W.M. Jones" <[EMAIL PROTECTED]> wrote:
> This patch implements fsync for Windows.
>
> I tested it using MinGW cross-compiler from a Fedora host, and wine
> instead of Windows:
>
> ./gnulib-tool --create-testdir --dir=/tmp/testdir fsync
>
> [Verify it configures & builds normally on Lin
Bruno Haible <[EMAIL PROTECTED]> wrote:
>> I noticed that the configure-time test for the f_type member
>> of struct statfs was failing on some systems.
>>
>> It didn't include .
>> I've pushed the obvious fix:
>
> This introduces a regression on HP-UX.
>
> The use of HAVE_STRUCT_STATFS_F_TYPE is p
This patch implements fsync for Windows.
I tested it using MinGW cross-compiler from a Fedora host, and wine
instead of Windows:
./gnulib-tool --create-testdir --dir=/tmp/testdir fsync
[Verify it configures & builds normally on Linux, then ...]
[Create a test program, test.c (attached).
Bruno Haible wrote:
> Paolo Bonzini wrote:
>> I agree, but it seems like pipe WaitForSingleObject is broken. You'd
>> need a thread polling with PeekNamedPipe, or a busy-waiting loop in the
>> main thread altogether.
>
> Maybe we need to reset the pipe from "signaled" to "non-signaled" state
> af
Paolo Bonzini wrote:
> I agree, but it seems like pipe WaitForSingleObject is broken. You'd
> need a thread polling with PeekNamedPipe, or a busy-waiting loop in the
> main thread altogether.
Maybe we need to reset the pipe from "signaled" to "non-signaled" state
after [Msg]WaitFor..Objects retur
> No, please. I don`t need sockets. For communication with a subprocess, all
> I need is a pipe. Sockets would be an overkill for that.
I agree, but it seems like pipe WaitForSingleObject is broken. You'd
need a thread polling with PeekNamedPipe, or a busy-waiting loop in the
main thread altoget
I'm preparing to pull this code out into its own module,
since it's already used by coreutils' remove.c.
>From 2c65e30590ff468ad1e94df2c30ca4bfa9ce98d5 Mon Sep 17 00:00:00 2001
From: Jim Meyering <[EMAIL PROTECTED]>
Date: Wed, 1 Oct 2008 09:22:11 +0200
Subject: [PATCH] fts.c: adjust a new interfac
23 matches
Mail list logo
