comint-run can't send signals to subjob

comint-run can't send signals to subjob

[Robert Andersson]

This bug report will be sent to the Free Software Foundation,
not to your local site managers!
Please write in English, because the Emacs maintainers do not have
translators to read other languages for them.

Your bug report will be posted to the bug-gnu-emacs@gnu.org mailing list,
and to the gnu.emacs.bug news group.

In GNU Emacs 21.1.1 (sparc-sun-solaris2.8, X toolkit)
 of 2001-10-22 on mozart
configured using `configure  --with-xpm --with-jpeg --with-tiff=/usr/local --with-gif --with-png'
Important settings:
  value of $LC_ALL: nil
  value of $LC_COLLATE: nil
  value of $LC_CTYPE: iso_8859_1
  value of $LC_MESSAGES: nil
  value of $LC_MONETARY: nil
  value of $LC_NUMERIC: nil
  value of $LC_TIME: nil
  value of $LANG: nil
  locale-coding-system: iso-latin-1
  default-enable-multibyte-characters: t

Please describe exactly what actions triggered the bug
and the precise symptoms of the bug:

when running a programme in comint-run it will not accept any signals
from this buffer, though it will accept something like

     pkill -INT -U $USER programme

from a external shell. note though that we have recently upgraded our
system from sun os 5.8 to 5.9 without recompiling emacs. this can not
be done because of another compile bug:

In file included from /usr/openwin/include/X11/Xos.h:79,
                 from xfaces.c:279:
/usr/include/strings.h:25: conflicting types for `memmove'
/usr/include/iso/string_iso.h:62: previous declaration of `memmove'
/usr/include/strings.h:26: parse error before `0'
/usr/include/strings.h:26: warning: conflicting types for built-in function `memset'
gcc -c  -Demacs -DHAVE_CONFIG_H -DUSE_LUCID  -I. -I/usr/local/src/installed/emacs-21.1/src       -I/usr/openwin/include    -I/usr/dt/include  -O3 -mcpu=v8 -mtune=v9 xfns.c
In file included from /usr/openwin/include/X11/Xos.h:79,
                 from xfns.c:74:
/usr/include/strings.h:25: conflicting types for `memmove'
/usr/include/iso/string_iso.h:62: previous declaration of `memmove'
/usr/include/strings.h:26: parse error before `0'
/usr/include/strings.h:26: warning: conflicting types for built-in function `memset'
gmake[1]: *** [xfaces.o] Error 1
gmake[1]: *** Waiting for unfinished jobs....
gmake[1]: *** [xfns.o] Error 1
gmake[1]: Leaving directory `/usr/local/src/installed/emacs-21.1/src'
gmake: *** [src] Error 2

Recent input:
<escape> x c o m i n <tab> r u <tab> <return> s i c 
s t s u <backspace> <backspace> u s <return> C-c C-c 
<return> <escape> x c o m i n <tab> r u <tab> <backspace> 
<backspace> <backspace> <backspace> <backspace> <backspace> 
<backspace> <backspace> <backspace> <backspace> r e 
o p <tab> <backspace> <backspace> p o <tab> t <backspace> 
r <tab> <return>

Recent messages:
Loading bytecomp...
Loading cl-macs...done
Loading bytecomp...done
Loading byte-opt...done
Loading jka-compr...done
Loading two-column...done
For information about the GNU Project and its goals, type C-h C-p.
Making completion list...
Loading view...done
Loading emacsbug...done

Re: comint-run can't send signals to subjob

[Richard Stallman]

    when running a programme in comint-run it will not accept any signals
    from this buffer, 

I am not sure what it means to "accept signals from a buffer".  But
rather than explain it, please send a precise test case.  That will
show us more precisely than any amount of description.

Please read the Bugs section in the Emacs manual, which provides
guidelines on how to write a bug report to give us the
necessary information so we can fix the bug.

    In file included from /usr/openwin/include/X11/Xos.h:79,
		     from xfaces.c:279:
    /usr/include/strings.h:25: conflicting types for `memmove'
    /usr/include/iso/string_iso.h:62: previous declaration of `memmove'
    /usr/include/strings.h:26: parse error before `0'
    /usr/include/strings.h:26: warning: conflicting types for built-in function `memset'

Would you please try the latest pretest and see if it compiles?  If it
does not, I hope you will tell us what change is needed to make it
compile.  Only a Solaris user can tell us how to fix this problem.

Re: comint-run can't send signals to subjob

[Robert Andersson]

>>>>> "RS" == Richard Stallman <rms@gnu.org> writes:
RS>     when running a programme in comint-run it will not accept any signals
RS>     from this buffer, 

RS> I am not sure what it means to "accept signals from a buffer".  But
RS> rather than explain it, please send a precise test case.  That will
RS> show us more precisely than any amount of description.

RS> Please read the Bugs section in the Emacs manual, which provides
RS> guidelines on how to write a bug report to give us the
RS> necessary information so we can fix the bug.

k, i read the section on understanding bug reporting. thanks. i see
now that my explaination was a tad vague.

the program that i run with comin-run does of course accept signals,
it's comint-run that doesn't send any to the program, eg

	M-x comint-run RET swi RET

to start swi prolog (yes this example only works if you got swi
installed and in your PATH maybe a new ispell could do as well???),
then i want to do

	C-c C-c

that is, comint-interrupt-subjob to send the swi programme an INT
signal. It does not work on emacs version 21.1.1 or 20.7 running on
solaris 9, though on a solaris 8 box i does using the same emacs
binaries.

in my example above, how do i know that it is not swi that rejects
signals? well i can send it a signal from any tty on the same system
with something like

	kill -INT swi-pid

which will work.

actually none of the comint-signals seem to work.

i got an answer from usenet saying that they have had the same problem
with hpux 11... though in this i will not speculate.

i think this was a tad more elaborative don't you think so?

RS>     In file included from /usr/openwin/include/X11/Xos.h:79,
RS> 		     from xfaces.c:279:
RS>     /usr/include/strings.h:25: conflicting types for `memmove'
RS>     /usr/include/iso/string_iso.h:62: previous declaration of `memmove'
RS>     /usr/include/strings.h:26: parse error before `0'
RS>     /usr/include/strings.h:26: warning: conflicting types for built-in function `memset'

RS> Would you please try the latest pretest and see if it compiles?  If it
RS> does not, I hope you will tell us what change is needed to make it
RS> compile.  Only a Solaris user can tell us how to fix this problem.

uhm, i can do prolog... i am terrible at c :)

yours,
/robert

-- 
Systems Administrator
Swedish National Graduate School of Language Technology