Fwd: another diff to try

[Hermann Peifer]

Hmm. Fw to bug-gawk, as my GMX mailer daemon does not seem to like
arnold@skeeve.com. Just now.

The following address failed:

    arnold@skeeve.com:
        SMTP error from remote server for RCPT TO command, host:
skeeve.com (96.88.95.60) reason: 550 5.7.1 <arnold@skeeve.com>...
Rejected: 82.165.159.14 listed at zen.spamhaus.org




-------- Forwarded Message --------
Subject: Re: another diff to try
Date: Tue, 11 Oct 2022 11:22:26 +0200
From: Hermann Peifer <peifer@gmx.eu>
To: arnold@skeeve.com

[~/local/src/gawk]> git status
HEAD detached at ebf5e417
nothing to commit, working tree clean

[~/local/src/gawk]> patch -p1 < tmp/x.diff
patching file symbol.c

[~/local/src/gawk]> ./bootstrap.sh

[~/local/src/gawk]> CFLAGS="-g -O0" ./configure && make -j && make check
...
======== Starting MPFR tests ========
MPFR tests not supported on this system
======== Done with MPFR tests ========
======== Starting PMA tests ========
gcc makepmafile.c -o makepmafile
pma


# Hanging pma test, as before. 1st run was OK, as before.
[~/local/src/gawk]> cat test/_pma
1

# This is the hanging process
PID    COMMAND      %CPU  TIME
9179   gawk         99.4  07:39.17

# Backtrace from hanging process
(lldb) process attach --pid 9179
Process 9179 stopped
* thread #1, queue = 'com.apple.main-thread', stop reason =
EXC_BAD_ACCESS (code=1, address=0x100afe4c8)
    frame #0: 0x00000001047ed794 gawk`in_array(a=0x0000303800006978,
s=0x000030380001d568) at awk.h:1916:8
   1913 {
   1914         NODE **ret;
   1915
-> 1916              ret = a->aexists(a, s);
   1917
   1918         return ret ? *ret : NULL;
   1919 }
Target 0: (gawk) stopped.
Executable module set to "/Users/peifer/local/src/gawk/gawk".
Architecture set to: arm64e-apple-macosx-.
(lldb) bt
* thread #1, queue = 'com.apple.main-thread', stop reason =
EXC_BAD_ACCESS (code=1, address=0x100afe4c8)
  * frame #0: 0x00000001047ed794 gawk`in_array(a=0x0000303800006978,
s=0x000030380001d568) at awk.h:1916:8
    frame #1: 0x00000001047ed648 gawk`lookup(name="LINENO") at
symbol.c:161:7
    frame #2: 0x00000001047dff68 gawk`err(isfatal=false, s="",
emsg="fatal error: internal error", argp="") at msg.c:56:14
    frame #3: 0x00000001047e029c gawk`msg(mesg="fatal error: internal
error") at msg.c:122:2
    frame #4: 0x00000001047dd154 gawk`catchsig(sig=11) at main.c:1318:3
    frame #5: 0x00000001908974a4 libsystem_platform.dylib`_sigtramp + 56
    frame #6: 0x00000001047ed50c gawk`install(name="\xa8\xe4\xaf",
parm=0x000030380001d4b8, type=Node_param_list | Node_arrayfor | 0x3020)
at symbol.c:388:3
    frame #7: 0x00000001047ed50c gawk`install(name="ARGIND",
parm=0x0000000000000000, type=Node_var) at symbol.c:388:3
    frame #8: 0x00000001047ed338 gawk`install_symbol(name="ARGIND",
type=Node_var) at symbol.c:127:9
    frame #9: 0x00000001047ddf04 gawk`init_vars at main.c:879:21
    frame #10: 0x00000001047dc690 gawk`main(argc=2,
argv=0x000000016b683850) at main.c:430:2
    frame #11: 0x0000000104a9108c dyld`start + 520
(lldb)


On 11/10/2022 10.52, arnold@skeeve.com wrote:
> Yes, let's drop the patch into an unmodified setup.
>
> Thanks,
>
> Arnold
>
> Hermann Peifer <peifer@gmx.eu> wrote:
>
> > OK. I see the new patch is about symbol.c. What about your earlier patch
> > for main.c and support/pma.c ?
> >
> > I do a use "git restore main.c and support/pma.c " first -- or rather not?
> >
> > Hermann
> >
> > HEAD detached at ebf5e417
> > Changes not staged for commit:
> >     (use "git add <file>..." to update what will be committed)
> >     (use "git restore <file>..." to discard changes in working directory)
> >         modified:   main.c
> >         modified:   support/pma.c
> >
> > On 11/10/2022 10.05, arnold@skeeve.com wrote:
> > > Hi.
> > >
> > > The attached diff  might fix the basic failure. It won't fix the
> > > dump-vars issue.
> > >
> > > Thanks,
> > >
> > > Arnold