GNU bug report logs - #79868
30.2: eval-tests/funcall-with-delayed-message fails on sparc64

Previous Next

To reply to this bug, email your comments to 79868 AT debbugs.gnu.org.
There is no need to reopen the bug first.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: John Paul Adrian Glaubitz <glaubitz <at> physik.fu-berlin.de>
To: bug-gnu-emacs <bug-gnu-emacs <at> gnu.org>
Subject: 30.2: eval-tests/funcall-with-delayed-message fails on sparc64
Date: Thu, 20 Nov 2025 08:18:55 +0100

Hello,

the test eval-tests/funcall-with-delayed-message fails on sparc64 starting
with emacs 30.2:

Test eval-tests/funcall-with-delayed-message backtrace:
  signal(ert-test-failed (((should (equal (string-trim (buffer-string)
  ert-fail(((should (equal (string-trim (buffer-string)) expected-mess
  #f(compiled-function () #<bytecode 0x193351670ad30cf7>)()
  #f(compiled-function () #<bytecode 0x1d1e697ce3d565dd>)()
  handler-bind-1(#f(compiled-function () #<bytecode 0x1d1e697ce3d565dd
  ert--run-test-internal(#s(ert--test-execution-info :test #s(ert-test
  ert-run-test(#s(ert-test :name eval-tests/funcall-with-delayed-messa
  ert-run-or-rerun-test(#s(ert--stats :selector ... :tests ... :test-m
  ert-run-tests((not (or (tag :expensive-test) (tag :unstable))) #f(co
  ert-run-tests-batch((not (or (tag :expensive-test) (tag :unstable)))
  ert-run-tests-batch-and-exit((not (or (tag :expensive-test) (tag :un
  eval((ert-run-tests-batch-and-exit '(not (or (tag :expensive-test) (
  command-line-1(("-L" ":/build/reproducible-path/emacs-30.2+1/debian/
  command-line()
  normal-top-level()
Test eval-tests/funcall-with-delayed-message condition:
    params: (0.1 0.8)
    (ert-test-failed
     ((should (equal (string-trim ...) expected-messages)) :form
      (equal "finished" "timed out\nfinished") :value nil :explanation
      (arrays-of-different-length 8 18 "finished" "timed out\nfinished"
                                  first-mismatch-at 0)))
   FAILED  26/26  eval-tests/funcall-with-delayed-message (1.201423 sec) at ../../build-src/test/src/eval-tests.el:223

Ran 26 tests, 25 results as expected, 1 unexpected (2025-11-18 13:52:51+0000, 7.466815 sec)

1 unexpected results:
   FAILED  eval-tests/funcall-with-delayed-message

Full build log available in [1].

Access to a SPARC machine, both running Solaris 11.4 and Linux, can be
obtained through the GCC Compile Farm [2].

This bug is also tracked in [3].

Thanks,
Adrian

> [1] https://buildd.debian.org/status/fetch.php?pkg=emacs&arch=sparc64&ver=1%3A30.2%2B1-1&stamp=1763474165&raw=0
> [2] https://gcc.gnu.org/wiki/CompileFarm
> [3] https://github.com/sparclinux/issues/issues/37

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer - glaubitz <at> debian.org
`. `'   Freie Universitaet Berlin - glaubitz <at> physik.fu-berlin.de
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913

Message #8 received at 79868 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: John Paul Adrian Glaubitz <glaubitz <at> physik.fu-berlin.de>
Cc: 79868 <at> debbugs.gnu.org
Subject: Re: bug#79868: 30.2: eval-tests/funcall-with-delayed-message fails on
 sparc64
Date: Thu, 20 Nov 2025 10:19:59 +0200

> From: John Paul Adrian Glaubitz <glaubitz <at> physik.fu-berlin.de>
> Date: Thu, 20 Nov 2025 08:18:55 +0100
> 
> the test eval-tests/funcall-with-delayed-message fails on sparc64 starting
> with emacs 30.2:
> 
> Test eval-tests/funcall-with-delayed-message backtrace:
>   signal(ert-test-failed (((should (equal (string-trim (buffer-string)
>   ert-fail(((should (equal (string-trim (buffer-string)) expected-mess
>   #f(compiled-function () #<bytecode 0x193351670ad30cf7>)()
>   #f(compiled-function () #<bytecode 0x1d1e697ce3d565dd>)()
>   handler-bind-1(#f(compiled-function () #<bytecode 0x1d1e697ce3d565dd
>   ert--run-test-internal(#s(ert--test-execution-info :test #s(ert-test
>   ert-run-test(#s(ert-test :name eval-tests/funcall-with-delayed-messa
>   ert-run-or-rerun-test(#s(ert--stats :selector ... :tests ... :test-m
>   ert-run-tests((not (or (tag :expensive-test) (tag :unstable))) #f(co
>   ert-run-tests-batch((not (or (tag :expensive-test) (tag :unstable)))
>   ert-run-tests-batch-and-exit((not (or (tag :expensive-test) (tag :un
>   eval((ert-run-tests-batch-and-exit '(not (or (tag :expensive-test) (
>   command-line-1(("-L" ":/build/reproducible-path/emacs-30.2+1/debian/
>   command-line()
>   normal-top-level()
> Test eval-tests/funcall-with-delayed-message condition:
>     params: (0.1 0.8)
>     (ert-test-failed
>      ((should (equal (string-trim ...) expected-messages)) :form
>       (equal "finished" "timed out\nfinished") :value nil :explanation
>       (arrays-of-different-length 8 18 "finished" "timed out\nfinished"
>                                   first-mismatch-at 0)))
>    FAILED  26/26  eval-tests/funcall-with-delayed-message (1.201423 sec) at ../../build-src/test/src/eval-tests.el:223
> 
> Ran 26 tests, 25 results as expected, 1 unexpected (2025-11-18 13:52:51+0000, 7.466815 sec)
> 
> 1 unexpected results:
>    FAILED  eval-tests/funcall-with-delayed-message
> 
> Full build log available in [1].
> 
> Access to a SPARC machine, both running Solaris 11.4 and Linux, can be
> obtained through the GCC Compile Farm [2].
> 
> This bug is also tracked in [3].

Sounds like SPARC-specific, because I cannot reproduce this, neither
on GNU/Linux nor on MS-Windows.

Let's wait what the sparclinux issue yields, and take it from there.

Alternatively, if you or someone else could investigate how well
funcall-with-delayed-message works on those systems, I'm interested to
hear the details.  E.g., perhaps the timeouts used by the test are
too short or something?

Message #11 received at 79868 <at> debbugs.gnu.org (full text, mbox):

From: John Paul Adrian Glaubitz <glaubitz <at> physik.fu-berlin.de>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 79868 <at> debbugs.gnu.org
Subject: Re: bug#79868: 30.2: eval-tests/funcall-with-delayed-message fails
 on sparc64
Date: Thu, 20 Nov 2025 09:37:14 +0100

Hi Eli,

On Thu, 2025-11-20 at 10:19 +0200, Eli Zaretskii wrote:
> 
> Sounds like SPARC-specific, because I cannot reproduce this, neither
> on GNU/Linux nor on MS-Windows.
> 
> Let's wait what the sparclinux issue yields, and take it from there.
> 
> Alternatively, if you or someone else could investigate how well
> funcall-with-delayed-message works on those systems, I'm interested to
> hear the details.  E.g., perhaps the timeouts used by the test are
> too short or something?

Is this a new feature with 30.2? Does the error bisecting make any sense?

You can use the GCC Compile Farm if you want to test it yourself on SPARC.

Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer
`. `'   Physicist
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913

Message #14 received at 79868 <at> debbugs.gnu.org (full text, mbox):

From: John Paul Adrian Glaubitz <glaubitz <at> physik.fu-berlin.de>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 79868 <at> debbugs.gnu.org
Subject: Re: bug#79868: 30.2: eval-tests/funcall-with-delayed-message fails
 on sparc64
Date: Thu, 20 Nov 2025 10:18:13 +0100

Hi Eli,

On Thu, 2025-11-20 at 09:37 +0100, John Paul Adrian Glaubitz wrote:
> On Thu, 2025-11-20 at 10:19 +0200, Eli Zaretskii wrote:
> > 
> > Sounds like SPARC-specific, because I cannot reproduce this, neither
> > on GNU/Linux nor on MS-Windows.
> > 
> > Let's wait what the sparclinux issue yields, and take it from there.
> > 
> > Alternatively, if you or someone else could investigate how well
> > funcall-with-delayed-message works on those systems, I'm interested to
> > hear the details.  E.g., perhaps the timeouts used by the test are
> > too short or something?
> 
> Is this a new feature with 30.2? Does the error bisecting make any sense?
> 
> You can use the GCC Compile Farm if you want to test it yourself on SPARC.

I just checked out Emacs from git and built and tested on SPARC like this:

$ ./autogen.sh
$ ./configure
$ make -j48
$ make -j48 check

The test results look okay:

SUMMARY OF TEST RESULTS
-----------------------
Files examined: 542
Ran 8283 tests, 7964 results as expected, 0 unexpected, 319 skipped
make[2]: Leaving directory '/home/glaubitz/emacs-git/test'
make[1]: Leaving directory '/home/glaubitz/emacs-git/test'

Was there any improvement in this regard that could be backported?

Thanks,
Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer
`. `'   Physicist
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913

Message #17 received at 79868 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: John Paul Adrian Glaubitz <glaubitz <at> physik.fu-berlin.de>
Cc: 79868 <at> debbugs.gnu.org
Subject: Re: bug#79868: 30.2: eval-tests/funcall-with-delayed-message fails
 on sparc64
Date: Thu, 20 Nov 2025 12:55:55 +0200

> From: John Paul Adrian Glaubitz <glaubitz <at> physik.fu-berlin.de>
> Cc: 79868 <at> debbugs.gnu.org
> Date: Thu, 20 Nov 2025 09:37:14 +0100
> 
> On Thu, 2025-11-20 at 10:19 +0200, Eli Zaretskii wrote:
> > 
> > Sounds like SPARC-specific, because I cannot reproduce this, neither
> > on GNU/Linux nor on MS-Windows.
> > 
> > Let's wait what the sparclinux issue yields, and take it from there.
> > 
> > Alternatively, if you or someone else could investigate how well
> > funcall-with-delayed-message works on those systems, I'm interested to
> > hear the details.  E.g., perhaps the timeouts used by the test are
> > too short or something?
> 
> Is this a new feature with 30.2?

No, it was introduced in Emacs 29.1.

Does the error bisecting make any sense?

Only if it works in Emacs 29 or Emacs 30.1.

> You can use the GCC Compile Farm if you want to test it yourself on SPARC.

I don't think I will have time, but maybe someone else will.  (And I
don't know enough about Solaris or Linux on SPARC anyway.)

Message #20 received at 79868 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: John Paul Adrian Glaubitz <glaubitz <at> physik.fu-berlin.de>
Cc: 79868 <at> debbugs.gnu.org
Subject: Re: bug#79868: 30.2: eval-tests/funcall-with-delayed-message fails
 on sparc64
Date: Thu, 20 Nov 2025 13:04:05 +0200

> From: John Paul Adrian Glaubitz <glaubitz <at> physik.fu-berlin.de>
> Cc: 79868 <at> debbugs.gnu.org
> Date: Thu, 20 Nov 2025 10:18:13 +0100
> 
> On Thu, 2025-11-20 at 09:37 +0100, John Paul Adrian Glaubitz wrote:
> > On Thu, 2025-11-20 at 10:19 +0200, Eli Zaretskii wrote:
> > > 
> > > Sounds like SPARC-specific, because I cannot reproduce this, neither
> > > on GNU/Linux nor on MS-Windows.
> > > 
> > > Let's wait what the sparclinux issue yields, and take it from there.
> > > 
> > > Alternatively, if you or someone else could investigate how well
> > > funcall-with-delayed-message works on those systems, I'm interested to
> > > hear the details.  E.g., perhaps the timeouts used by the test are
> > > too short or something?
> > 
> > Is this a new feature with 30.2? Does the error bisecting make any sense?
> > 
> > You can use the GCC Compile Farm if you want to test it yourself on SPARC.
> 
> I just checked out Emacs from git and built and tested on SPARC like this:
> 
> $ ./autogen.sh
> $ ./configure
> $ make -j48
> $ make -j48 check
> 
> The test results look okay:
> 
> SUMMARY OF TEST RESULTS
> -----------------------
> Files examined: 542
> Ran 8283 tests, 7964 results as expected, 0 unexpected, 319 skipped
> make[2]: Leaving directory '/home/glaubitz/emacs-git/test'
> make[1]: Leaving directory '/home/glaubitz/emacs-git/test'
> 
> Was there any improvement in this regard that could be backported?

Then the question is which change fixed this for you (assuming you
used the same configure-time options).  I guess bisecting could tell.

Did you build the master branch or the emacs-30 release branch?

Message #25 received at 79868-done <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: glaubitz <at> physik.fu-berlin.de
Cc: 79868-done <at> debbugs.gnu.org
Subject: Re: bug#79868: 30.2: eval-tests/funcall-with-delayed-message fails on
 sparc64
Date: Sat, 06 Dec 2025 12:23:28 +0200

> Cc: 79868 <at> debbugs.gnu.org
> Date: Thu, 20 Nov 2025 13:04:05 +0200
> From: Eli Zaretskii <eliz <at> gnu.org>
> 
> > From: John Paul Adrian Glaubitz <glaubitz <at> physik.fu-berlin.de>
> > Cc: 79868 <at> debbugs.gnu.org
> > Date: Thu, 20 Nov 2025 10:18:13 +0100
> > 
> > On Thu, 2025-11-20 at 09:37 +0100, John Paul Adrian Glaubitz wrote:
> > > On Thu, 2025-11-20 at 10:19 +0200, Eli Zaretskii wrote:
> > > > 
> > > > Sounds like SPARC-specific, because I cannot reproduce this, neither
> > > > on GNU/Linux nor on MS-Windows.
> > > > 
> > > > Let's wait what the sparclinux issue yields, and take it from there.
> > > > 
> > > > Alternatively, if you or someone else could investigate how well
> > > > funcall-with-delayed-message works on those systems, I'm interested to
> > > > hear the details.  E.g., perhaps the timeouts used by the test are
> > > > too short or something?
> > > 
> > > Is this a new feature with 30.2? Does the error bisecting make any sense?
> > > 
> > > You can use the GCC Compile Farm if you want to test it yourself on SPARC.
> > 
> > I just checked out Emacs from git and built and tested on SPARC like this:
> > 
> > $ ./autogen.sh
> > $ ./configure
> > $ make -j48
> > $ make -j48 check
> > 
> > The test results look okay:
> > 
> > SUMMARY OF TEST RESULTS
> > -----------------------
> > Files examined: 542
> > Ran 8283 tests, 7964 results as expected, 0 unexpected, 319 skipped
> > make[2]: Leaving directory '/home/glaubitz/emacs-git/test'
> > make[1]: Leaving directory '/home/glaubitz/emacs-git/test'
> > 
> > Was there any improvement in this regard that could be backported?
> 
> Then the question is which change fixed this for you (assuming you
> used the same configure-time options).  I guess bisecting could tell.
> 
> Did you build the master branch or the emacs-30 release branch?

No further comments, so I'm closing this bug.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.