-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 8/22/2009 5:59 AM:
> Hi Eric,
>
>> First order of business - we need two modules based on the fcntl name, one
>> for the header ... and one for the function.
>
> Yes, absolutely.
>
> Your 3 patches look all fine, except
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 8/22/2009 5:20 PM:
> Here is a proposal to add a module 'mkostemp', using the source code changes
> from glibc. It also reduces the diffs to glibc, by incorporating parts of this
> glibc commit:
Nice to do some resyncing.
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 8/22/2009 3:39 PM:
> Hi,
>
> Here's the proposed module for function 'accept4'.
>
> It does not define the macro SOCK_CLOEXEC. I think this macro should be
> replaced by a separate module, that would then also influence s
The following patch series implements custom quotation marks in quotearg.
My main motivation is to be able to handle single-quoted strings. For
example, you can invoke this:
quotearg_custom ("'", "'", "joel's\tstring");
to produce this:
"'joel\\'s\\tstring'"
However, I figured it was pr
Eric Blake asked:
> > /* Check the supported flags. */
> > if ((flags & ~(O_CLOEXEC | O_BINARY | O_TEXT)) != 0)
>
> Is there any way to support NONBLOCK in Woe32?
Not really. The concept of asynchronous I/O in Woe32 means that the I/O
processes independently, and the caller is notified at th
Here is a proposal to add a module 'mkostemp', using the source code changes
from glibc. It also reduces the diffs to glibc, by incorporating parts of this
glibc commit:
http://sourceware.org/git/?p=glibc.git;a=commitdiff;h=d7e23b02a40384be1eaf00762ea36fd117c462cd
Opinions? Objections?
2009-0
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 8/22/2009 10:46 AM:
> /* Native Woe32 API. */
>
> # include
>
> int
> pipe2 (int fd[2], int flags)
> {
> /* Check the supported flags. */
> if ((flags & ~(O_CLOEXEC | O_BINARY | O_TEXT)) != 0)
Is there any way to
Hi,
Here's the proposed module for function 'accept4'.
It does not define the macro SOCK_CLOEXEC. I think this macro should be
replaced by a separate module, that would then also influence socket(),
socketpair() - and what about connect()?
Opinions? Objections?
2009-08-22 Bruno Haible
Eric Blake wrote:
> What's the difference between dup2_ex and dup3, other than the set of
> valid flags accepted?
It's the same as dup3. I woke up a bit late on this thread.
Bruno
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 8/22/2009 5:59 AM:
> I therefore think that instead of offering the POSIXy fcntl call,
> gnulib should offer functions like dup, dup2, dup_safer, fd_safer,
> that take a flags argument as additional parameter.
> int dup_e
> already possible to implement pipe2.
For reference, the introduction of pipe2 in Linux was here:
http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.28.y.git;a=commit;h=ed8cae8ba01348bfd8f4648dd807b04d7f08
http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.28.y.git;a=commit;h=
With the conditional definition of O_CLOEXEC in place in fcntl.h, it's
already possible to implement pipe2.
It does not yet support O_CLOEXEC on systems other than Linux and mingw,
but when O_CLOEXEC will be supported on all platforms, this module does
not need to be changed.
2009-08-22 Bruno H
Paolo Bonzini wrote:
> Also, actually the flags can be accessed; it's undocumented but it can
> be found. See for example these files, taken from Perl and Ruby
> respectively:
>
> http://downloads.activestate.com/contrib.old/ntsock.cpp
> http://www.google.com/codesearch/p#-ImLthoNrik/ruby-1.8.6
On 08/22/2009 01:59 PM, Bruno Haible wrote:
But the flags stored inside MSVCRT for fd cannot be directly accessed.
Ah okay now I understand the other message. I think we're safe if we
only support _setting_ the flag, and not clearing it. Most of the use
cases are like this.
Also, actually
On 08/22/2009 03:25 PM, Bruno Haible wrote:
Eric Blake wrote:
we also need to
tweak the O_NOINHERIT flag of an existing fd without having to use dup in
cloexec.c (I'm hoping the windows functions GetHandleInformation and
SetHandleInformation work here).
I don't see how to make it work without
Eric Blake wrote:
> we also need to
> tweak the O_NOINHERIT flag of an existing fd without having to use dup in
> cloexec.c (I'm hoping the windows functions GetHandleInformation and
> SetHandleInformation work here).
I don't see how to make it work without a lot of dup() calls and without
assumin
Eric Blake asked:
> > +/* Fix up the FD_* macros. */
> > +
> > +#ifndef FD_CLOEXEC
> > +# define FD_CLOEXEC 1
> > +#endif
>
> Is mingw the only known platform that lacks this?
Yes. And on all Unix platforms, FD_CLOEXEC has the value 1.
Bruno
Hi Eric,
> First order of business - we need two modules based on the fcntl name, one
> for the header ... and one for the function.
Yes, absolutely.
Your 3 patches look all fine, except for the mingw replacement of fcntl().
+static int
+dupfd (int fd, int desired_fd)
+{
+ /* Although this loo
Bruno Haible wrote:
>> I'll wait for feedback before pushing.
>
> OK for me in general. However, I would combine some Makefile commands
> before adding the $(AM_V_GEN) placeholder. When you put $(AM_V_GEN)
> in front of the 'rm -f $...@-t $@' commands but use '$(AM_V_at)' in front of
> the command
Jim Meyering wrote:
> I will follow up with a change to join the continued
> lines like this,
>
> - } > $...@-t && \
> - mv $...@-t $@
> + } > $...@-t && mv $...@-t $@
I will object this one, for legibility. 1 1/2 statements in one line, is
not good. Either multiple statements in a li
Hi Jim,
> I'll wait for feedback before pushing.
OK for me in general. However, I would combine some Makefile commands
before adding the $(AM_V_GEN) placeholder. When you put $(AM_V_GEN)
in front of the 'rm -f $...@-t $@' commands but use '$(AM_V_at)' in front of
the command that actually creates
First order of business - we need two modules based on the fcntl name, one
for the header (as fcntl.in.h does other useful things independently of
whether we are fixing cloexec and other fcntl(2) items), and one for the
function. If we keep the existing fcntl name as the header, then what
should
22 matches
Mail list logo
