[Ecls-list] ECL HEAD + Slime HEAD

[Marko Kocić]

- rm -rf ~/.slime ~/.fasls
- start emacs and slime (evertyhing seems ok while in repl)
- restart (or quit and start slime) couple times. Doesn't fail each time
- eventually get the following message in repl

;;; Stack overflow.
;;; Jumping to the outermost toplevel prompt
;;;

and after attaching gdb to running process I get the following backtrace
(gdb) bt
#0  0x00007ff9480c30cb in read () from /lib/libpthread.so.0
#1  0x00007ff9483f3f06 in read (strm=0xf40d20, c=0x7fff63392fb0 " \r",
<incomplete sequence \364>, n=1) at /usr/include/bits/unistd.h:45
#2  io_file_read_byte8 (strm=0xf40d20, c=0x7fff63392fb0 " \r",
<incomplete sequence \364>, n=1) at
/home/markko/cvstree/ecl.git/src/c/file.d:2561
#3  0x00007ff9483f705b in utf_8_decoder (stream=0xf40d20,
read_byte8=0x7ff9483f3e80 <io_file_read_byte8>, source=0xf40d20)
    at /home/markko/cvstree/ecl.git/src/c/file.d:1068
#4  0x00007ff9483f28e5 in eformat_read_char (strm=0x0) at
/home/markko/cvstree/ecl.git/src/c/file.d:587
#5  0x00007ff9483f7c9d in generic_peek_char (strm=0x0) at
/home/markko/cvstree/ecl.git/src/c/file.d:434
#6  0x00007ff9483fc491 in cl_peek_char (narg=4) at
/home/markko/cvstree/ecl.git/src/c/read.d:1716
#7  0x00007ff9483d4521 in L26tpl_read () at
/home/markko/cvstree/ecl.git/build/clos/top.c:1113
#8  0x00007ff9483d7f93 in LC19rep (narg=0) at
/home/markko/cvstree/ecl.git/build/clos/top.c:931
#9  0x00007ff9483d925e in L24tpl (narg=0) at
/home/markko/cvstree/ecl.git/build/clos/top.c:846
#10 0x00007ff9483da4b5 in L1top_level () at
/home/markko/cvstree/ecl.git/build/clos/top.c:42
#11 0x00007ff9483e6f80 in ecl_interpret (frame=<value optimized out>,
env=<value optimized out>, bytecodes=<value optimized out>)
    at /home/markko/cvstree/ecl.git/src/c/interpreter.d:489
#12 0x00007ff9483eaa7c in si_eval_with_env (narg=1, form=<value
optimized out>) at /home/markko/cvstree/ecl.git/src/c/compiler.d:2867
#13 0x00007ff9483d69e8 in si_safe_eval (narg=2, V1=0x2acdb81, V2=0x1)
at /home/markko/cvstree/ecl.git/build/clos/top.c:3580
#14 0x0000000000400bd6 in main (argc=<value optimized out>,
argv=<value optimized out>) at /tmp/ECLINIT6OVgfK.c:57

Thanks for the tip about attaching gdb. I encountered similar problems
in usocket similar to this while trying to run hunchentoot, but those
can't be reproduced with latest ecl.

Regards,
Marko Kocić


On Tue, Feb 16, 2010 at 5:40 PM, Tobias C. Rittweiler <tcr at freebits.de> wrote:
> Marko Kocić writes:
>
>> On Tue, Feb 16, 2010 at 12:14 PM, Tobias C. Rittweiler <tcr at freebits.de> wrote:
>> >
>> > I'd like people to test current Slime HEAD with ECL HEAD.
>> >
>> > If you build ECL with --enable-threads, Slime will now use threads on
>> > ECL. If you build ecl with serve-event, Slime will use that (though I
>> > haven't tested that yet.)
>> >
>> > All in all, this should result in much better user experience. No random
>> > hangups anymore, etc.
>> >
>> > Important is that you do use ECL HEAD though, revisions of a few days
>> > ago, won't do.
>> >
>> I updated to latest slime and latest ecl (cvs and git from few minutes
>> ago) and notice the following problems
>> - first start of ecl (with clean .slime) under slime is ok
>> - stack overflow fault when doing restart, somethimes before,
>> somethimes after quit
>> - stack overflow (almost always) when doing start after quit
>> - slime is now compiled (I think that is the point)
>>
>> Platform is ubuntu 64 bit, ecl is with unicode and threads enabled
>
> I'm on 32 bit and not experiencing any such problems.
>
> Can you give exact steps what you're doing?
>
> Also, once Slime is connected to the SWANK server, you can attach gdb to
> the ecl process, then print a backtrace after stack overflow.
>
>  -T.
>
>
> ------------------------------------------------------------------------------
> SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
> Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
> http://p.sf.net/sfu/solaris-dev2dev
> _______________________________________________
> Ecls-list mailing list
> Ecls-list at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/ecls-list
>