Emacs pretest 24.0.96

Emacs pretest 24.0.96

[Chong Yidong]

Emacs pretest 24.0.96 is now available for download via FTP, at the
following location:

  ftp://alpha.gnu.org/gnu/emacs/pretest/emacs-24.0.96.tar.gz

This is the seventh pretest for what will become Emacs 24.1.
See etc/NEWS for a list of changes since Emacs 23.4.

Please send me an email reporting success or failure on your build
platform.  Please report any bugs that you come across via
M-x report-emacs-bugs, or email bug-gnu-emacs@gnu.org.
For questions, email emacs-devel@gnu.org.

Thank you for helping to test Emacs.


Note to developers: we are coming up on the release, so please exercise
great caution committing to the emacs-24 branch from now on.  If in
doubt about a commit, discuss it on emacs-devel first.  I encourage
everyone to give this pretest as much testing as possible.  Thanks.

Re: Emacs pretest 24.0.96

[Juanma Barranquero]

> Emacs pretest 24.0.96 is now available for download via FTP, at the
> following location:
>
>  ftp://alpha.gnu.org/gnu/emacs/pretest/emacs-24.0.96.tar.gz

Builds, bootstraps and runs fine on Windows 7 64-bit.

    Juanma

Re: Emacs pretest 24.0.96

[Alex Harsanyi]

2012/5/1 Juanma Barranquero <lekktu@gmail.com>:
>> Emacs pretest 24.0.96 is now available for download via FTP, at the
>> following location:
>>
>>  ftp://alpha.gnu.org/gnu/emacs/pretest/emacs-24.0.96.tar.gz
>
> Builds, bootstraps and runs fine on Windows 7 64-bit.
>
>     Juanma
>
I still get the "resource temporarily unavailable" error when trying
to spawn subprocesses using the prebuilt windows binaries.  The
problem was described here:

 http://article.gmane.org/gmane.emacs.devel/148967

A resolution for the problem was found (see
http://thread.gmane.org/gmane.emacs.devel/148994).  Are there any
plans in implementing it for Emacs 24?

Thanks,
Alex.

Re: Emacs pretest 24.0.96

[Eli Zaretskii]

> Date: Wed, 2 May 2012 09:10:05 +0800
> From: Alex Harsanyi <alexharsanyi@gmail.com>
> 
> I still get the "resource temporarily unavailable" error when trying
> to spawn subprocesses using the prebuilt windows binaries.  The
> problem was described here:
> 
>  http://article.gmane.org/gmane.emacs.devel/148967
> 
> A resolution for the problem was found (see
> http://thread.gmane.org/gmane.emacs.devel/148994).  Are there any
> plans in implementing it for Emacs 24?

You promised to report back on how well the changes we discussed work
for you, but never did.

If it works well, please show the diffs of the changes you are/were
using, and I will take care of installing them.

Thanks.

Re: Emacs pretest 24.0.96

[Alex Harsanyi]

2012/5/2 Eli Zaretskii <eliz@gnu.org>:
>> Date: Wed, 2 May 2012 09:10:05 +0800
>> From: Alex Harsanyi <alexharsanyi@gmail.com>
>>
>> I still get the "resource temporarily unavailable" error when trying
>> to spawn subprocesses using the prebuilt windows binaries.  The
>> problem was described here:
>>
>>  http://article.gmane.org/gmane.emacs.devel/148967
>>
>> A resolution for the problem was found (see
>> http://thread.gmane.org/gmane.emacs.devel/148994).  Are there any
>> plans in implementing it for Emacs 24?
>
> You promised to report back on how well the changes we discussed work
> for you, but never did.
>
> If it works well, please show the diffs of the changes you are/were
> using, and I will take care of installing them.

I didn't have a single problem with the changed CreateThread call, I
used the fixed Emacs daily in the past 7 weeks.  Here is the diff:

diff -up --tabsize=4 e\:/Packages/emacs-24.0.94/src/w32proc.c\~
e\:/Packages/emacs-24.0.94/src/w32proc.c
--- e:/Packages/emacs-24.0.94/src/w32proc.c~	2012-02-14 00:13:25.000000000 +0800
+++ e:/Packages/emacs-24.0.94/src/w32proc.c	2012-03-14 14:00:18.049600700 +0800
@@ -141,7 +141,7 @@ new_child (void)
       cp->char_consumed = CreateEvent (NULL, FALSE, FALSE, NULL);
       if (cp->char_consumed)
         {
-	  cp->thrd = CreateThread (NULL, 1024, reader_thread, cp, 0, &id);
+	  cp->thrd = CreateThread (NULL, 64 * 1024, reader_thread, cp,
0x00010000, &id);
 	  if (cp->thrd)
 	    return cp;
 	}

Diff finished.  Tue Apr 24 11:40:10 2012

Alex.

Re: Emacs pretest 24.0.96

[Eli Zaretskii]

> Date: Wed, 2 May 2012 14:15:48 +0800
> From: Alex Harsanyi <alexharsanyi@gmail.com>
> Cc: emacs-devel@gnu.org
> 
> I didn't have a single problem with the changed CreateThread call, I
> used the fixed Emacs daily in the past 7 weeks.  Here is the diff:
> 
> diff -up --tabsize=4 e\:/Packages/emacs-24.0.94/src/w32proc.c\~
> e\:/Packages/emacs-24.0.94/src/w32proc.c
> --- e:/Packages/emacs-24.0.94/src/w32proc.c~	2012-02-14 00:13:25.000000000 +0800
> +++ e:/Packages/emacs-24.0.94/src/w32proc.c	2012-03-14 14:00:18.049600700 +0800
> @@ -141,7 +141,7 @@ new_child (void)
>        cp->char_consumed = CreateEvent (NULL, FALSE, FALSE, NULL);
>        if (cp->char_consumed)
>          {
> -	  cp->thrd = CreateThread (NULL, 1024, reader_thread, cp, 0, &id);
> +	  cp->thrd = CreateThread (NULL, 64 * 1024, reader_thread, cp,
> 0x00010000, &id);
>  	  if (cp->thrd)
>  	    return cp;
>  	}
> 

Thanks.

Would people who build Emacs on MS-Windows please apply this change
and run with it for a while?  If I don't hear any complaints in a
week, I will install this on the release branch.

Re: Emacs pretest 24.0.96

[Chong Yidong]

Eli Zaretskii <eliz@gnu.org> writes:

> Would people who build Emacs on MS-Windows please apply this change
> and run with it for a while?  If I don't hear any complaints in a
> week, I will install this on the release branch.

No point---if you are planning to install it, please just go ahead and
do that now.  That maximizes the amount of testing and the chances of
getting problem reports.

Re: Emacs pretest 24.0.96

[Eli Zaretskii]

> From: Chong Yidong <cyd@gnu.org>
> Cc: Alex Harsanyi <alexharsanyi@gmail.com>,  emacs-devel@gnu.org
> Date: Thu, 03 May 2012 01:09:56 +0800
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > Would people who build Emacs on MS-Windows please apply this change
> > and run with it for a while?  If I don't hear any complaints in a
> > week, I will install this on the release branch.
> 
> No point---if you are planning to install it, please just go ahead and
> do that now.  That maximizes the amount of testing and the chances of
> getting problem reports.

I assume most people use the pretest version, not the emacs-24 branch
version.  If I'm right, then changes on the branch will not get tested
too much.

That said, I have no objections committing sooner.  Just let me run
with it for a while, since until now this change was tested on a
single machine which runs Windows 7.  It wasn't tested on XP and other
users of Windows 7 didn't suffer from the original problem at all.
I'd like to have a bit more confidence before installing.

Re: Emacs pretest 24.0.96

[Juanma Barranquero]

On Wed, May 2, 2012 at 7:09 PM, Chong Yidong <cyd@gnu.org> wrote:

> No point---if you are planning to install it, please just go ahead and
> do that now.  That maximizes the amount of testing and the chances of
> getting problem reports.

Speaking of changes to emacs-24, may I ask why the bugfix for
bug#11374 has been committed to the release branch? It is not a
regression, and though it fixes a real bug, that bug has never been
reported in more than five years, not  even in bug#11374, which is
just the result of a survey of strncpy calls in the sources, I think.
The fix is relatively small, but not trivial enough that it couldn't
hide another bug in what is a crucial part of emacsclient.c.

Confused,

    Juanma

Re: Emacs pretest 24.0.96

[Chong Yidong]

Juanma Barranquero <lekktu@gmail.com> writes:

> Speaking of changes to emacs-24, may I ask why the bugfix for
> bug#11374 has been committed to the release branch? It is not a
> regression, and though it fixes a real bug, that bug has never been
> reported in more than five years, not  even in bug#11374, which is
> just the result of a survey of strncpy calls in the sources, I think.
> The fix is relatively small, but not trivial enough that it couldn't
> hide another bug in what is a crucial part of emacsclient.c.

Hmm, it is a corner case.  I thought Jim's fix seemed pretty
self-contained and clear, and it's always good to avoid buffer
overflows.  But indeed it's not a regression.  So if you guys object to
it, I don't mind reverting the commit.

Re: Emacs pretest 24.0.96

[Juanma Barranquero]

On Thu, May 3, 2012 at 9:54 AM, Chong Yidong <cyd@gnu.org> wrote:

> But indeed it's not a regression.  So if you guys object to
> it, I don't mind reverting the commit.

Oh, I not so objected as wondered. But if you think it is correct, do
not revert it just by my asking about it.

    Juanma

Re: Emacs pretest 24.0.96

[Eli Zaretskii]

> From: Chong Yidong <cyd@gnu.org>
> Cc: Alex Harsanyi <alexharsanyi@gmail.com>,  emacs-devel@gnu.org
> Date: Thu, 03 May 2012 01:09:56 +0800
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > Would people who build Emacs on MS-Windows please apply this change
> > and run with it for a while?  If I don't hear any complaints in a
> > week, I will install this on the release branch.
> 
> No point---if you are planning to install it, please just go ahead and
> do that now.

Done (revision 107972 on the emacs-24 branch).