Emacs pretest 28.1.90 is out

Emacs pretest 28.1.90 is out

[Stefan Kangas]

[-- Attachment #1: Type: text/plain, Size: 1164 bytes --]

Hi!

The first pretest for what will be the 28.2 release of Emacs (the
extensible text editor) is available at:

  https://alpha.gnu.org/gnu/emacs/pretest/emacs-28.1.90.tar.xz

The tarball is signed; you can get the PGP signature file at:

  https://alpha.gnu.org/gnu/emacs/pretest/emacs-28.1.90.tar.xz.sig

Please give it as much testing as you can.

As always, if you encounter problems building or using Emacs,
send a report to bug-gnu-emacs@gnu.org with full details
(if possible, use M-x report-emacs-bug).

Thanks for helping to test Emacs.

--------------------------------------

To verify that the tarball is intact, download both the .sig and
the tarball, and run this command:

  gpg --verify emacs-28.1.90.tar.xz.sig

If that command fails because you don't have the required public key,
run this command to import it:

  gpg --keyserver pgp.mit.edu --recv-keys \
    CEA1DE21AB108493CC9C65742E82323B8F4353EE

You can also run sha1sum or sha256sum and confirm that these
checksums match:

SHA1  emacs-28.1.90.tar.xz
09a0d3be1cf5e24dd2c0d6381ffd85ad06e7cdfa

SHA256  emacs-28.1.90.tar.xz
3c004d099a1b80532a685134d7328f46da4c552410905653adc1476036345e38

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 511 bytes --]

Re: Emacs pretest 28.1.90 is out

[Arash Esbati]

Stefan Kangas <stefan@marxist.se> writes:

> The first pretest for what will be the 28.2 release of Emacs (the
> extensible text editor) is available at:
>
>   https://alpha.gnu.org/gnu/emacs/pretest/emacs-28.1.90.tar.xz

Thanks, builds fine on Win10 with Mingw64 (I ran only ./configure and
make).  I get only these warnings:

  CCLD     make-docfile.exe
  CCLD     make-fingerprint.exe
In function 'main':
cc1.exe: warning: '__builtin_memcpy' writing 12 bytes into a region of size between 0 and 7 [-Wstringop-overflow=]
In file included from ctags.c:2:
etags.c:1429:21: note: destination object of size [0, 7] allocated by 'xmalloc'
 1429 |         char *cmd = xmalloc (2 * strlen (tagfile) + sizeof "sort -u -o..");
      |                     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
etags.c:1435:14: warning: writing 1 byte into a region of size 0 [-Wstringop-overflow=]
 1435 |         *z++ = ' ';
      |         ~~~~~^~~~~
etags.c:1429:21: note: at offset 11 into destination object of size [0, 7] allocated by 'xmalloc'
 1429 |         char *cmd = xmalloc (2 * strlen (tagfile) + sizeof "sort -u -o..");
      |                     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
etags.c:1434:13: warning: 'stpcpy' writing 1 or more bytes into a region of size 0 overflows the destination [-Wstringop-overflow=]
 1434 |         z = stpcpy (z, tagfile);
      |             ^~~~~~~~~~~~~~~~~~~
etags.c:1429:21: note: at offset 11 into destination object of size [0, 7] allocated by 'xmalloc'
 1429 |         char *cmd = xmalloc (2 * strlen (tagfile) + sizeof "sort -u -o..");
      |                     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
etags.c:1436:9: warning: 'strcpy' writing 1 or more bytes into a region of size 0 overflows the destination [-Wstringop-overflow=]
 1436 |         strcpy (z, tagfile);
      |         ^~~~~~~~~~~~~~~~~~~
etags.c:1429:21: note: at offset 12 into destination object of size [0, 7] allocated by 'xmalloc'
 1429 |         char *cmd = xmalloc (2 * strlen (tagfile) + sizeof "sort -u -o..");
      |                     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

I can't tell if they can be ignored or not.  This is with gcc 12.1.0.

Best, Arash

Re: Emacs pretest 28.1.90 is out

[Eli Zaretskii]

> From: Arash Esbati <arash@gnu.org>
> Cc: emacs-devel@gnu.org
> Date: Fri, 01 Jul 2022 09:15:43 +0200
> 
> Stefan Kangas <stefan@marxist.se> writes:
> 
> > The first pretest for what will be the 28.2 release of Emacs (the
> > extensible text editor) is available at:
> >
> >   https://alpha.gnu.org/gnu/emacs/pretest/emacs-28.1.90.tar.xz
> 
> Thanks, builds fine on Win10 with Mingw64 (I ran only ./configure and
> make).  I get only these warnings:
> 
>   CCLD     make-docfile.exe
>   CCLD     make-fingerprint.exe
> In function 'main':
> cc1.exe: warning: '__builtin_memcpy' writing 12 bytes into a region of size between 0 and 7 [-Wstringop-overflow=]
> In file included from ctags.c:2:
> etags.c:1429:21: note: destination object of size [0, 7] allocated by 'xmalloc'
>  1429 |         char *cmd = xmalloc (2 * strlen (tagfile) + sizeof "sort -u -o..");
>       |                     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Does anyone understand why it says "size [0, 7]"?  The result sizeof
should yield 13, so even if GCC considers the case of strlen returning
zero, the result should be at least 13.  In my case (GCC 9.3)
disassembly of the 'main' function shows that indeed the code calls
strlen and then adds 13 to its result.

There are quite a few hits for -Wstringop-overflow on the Internet, so
maybe this is a GCC bug?  Maybe you should downgrade to a GCC whose
version is not XX.1.0.

I see nothing wrong with that code, FWIW.

Re: Emacs pretest 28.1.90 is out

[Andreas Schwab]

See commit 0f731c49e6.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 7578 EB47 D4E5 4D69 2510  2552 DF73 E780 A9DA AEC1
"And now for something completely different."

Re: Emacs pretest 28.1.90 is out

[Eli Zaretskii]

> From: Andreas Schwab <schwab@linux-m68k.org>
> Cc: Arash Esbati <arash@gnu.org>,  stefan@marxist.se,  emacs-devel@gnu.org
> Date: Fri, 01 Jul 2022 14:13:04 +0200
> 
> See commit 0f731c49e6.

Thanks.  So these warnings are false positives in GCC 12.

(And, of course, that commit breaks 'ctags' on Windows, where file
names cannot be quoted 'like this', sigh...)