GNU bug report logs - #11776
Guile 2.0.5 on GNU/Hurd

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 11776 in the body.
You can then email your comments to 11776 AT debbugs.gnu.org in the normal way.

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

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

From: Cyril Roelandt <tipecaml <at> gmail.com>
To: bug-guile <at> gnu.org, bug-hurd <at> gnu.org
Cc: debian-hurd <at> lists.debian.org
Subject: Guile
Date: Mon, 25 Jun 2012 03:23:04 +0200

[Message part 1 (text/plain, inline)]

(I'm sending this to both bug-guile and bug-hurd; could you please keep 
debian-hurd CC'ed ?)

Hi !

I'm trying to run Guile 2.0.5 on Debian GNU/Hurd in a qemu virtual machine.

I installed Guile from the tarball, with no configure options at all. 
I'm getting the following error:

$ ./meta/guile
GNU Guile 2.0.5
Copyright (C) 1995-2012 Free Software Foundation, Inc.

Guile comes with ABSOLUTELY NO WARRANTY; for details type `,show w'.
This program is free software, and you are welcome to redistribute it
under certain conditions; type `,show c' for details.

Enter `,help' for help.
scheme@(guile-user)> (+ 1 2)
Illegal instruction

The GDB output is attached.


When compiling with --without-threads, I can run (+ 1 2) in the 
interpreter successfully. Running ./check-guile gives me the following 
result:

Totals for this test run:
passes:                 34286
failures:               2
unexpected passes:      0
expected failures:      30
unresolved test cases:  575
untested test cases:    1
unsupported test cases: 15
errors:                 0

So I think it's probably usable this way.


WBR,
Cyril Roelandt.

[guile-sigill (text/plain, attachment)]

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

From: Samuel Thibault <samuel.thibault <at> gnu.org>
To: Cyril Roelandt <tipecaml <at> gmail.com>
Cc: bug-guile <at> gnu.org, debian-hurd <at> lists.debian.org, bug-hurd <at> gnu.org
Subject: Re: Guile
Date: Mon, 25 Jun 2012 08:00:31 -0300

Cyril Roelandt, le Mon 25 Jun 2012 03:23:04 +0200, a écrit :
> Program received signal SIGILL, Illegal instruction.
> 0x01413a5c in _hurd_intr_rpc_mach_msg () from /lib/i386-gnu/libc.so.0.3
> (gdb) bt

Please also post

thread apply all bt

Samuel

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

From: ludo <at> gnu.org (Ludovic Courtès)
To: bug-guile <at> gnu.org
Cc: debian-hurd <at> lists.debian.org, bug-hurd <at> gnu.org
Subject: Re: Guile
Date: Mon, 25 Jun 2012 23:24:39 +0200

Hi,

Cyril Roelandt <tipecaml <at> gmail.com> skribis:

> I'm trying to run Guile 2.0.5 on Debian GNU/Hurd in a qemu virtual machine.
>
> I installed Guile from the tarball, with no configure options at
> all. I'm getting the following error:
>
> $ ./meta/guile
> GNU Guile 2.0.5
> Copyright (C) 1995-2012 Free Software Foundation, Inc.
>
> Guile comes with ABSOLUTELY NO WARRANTY; for details type `,show w'.
> This program is free software, and you are welcome to redistribute it
> under certain conditions; type `,show c' for details.
>
> Enter `,help' for help.
> scheme@(guile-user)> (+ 1 2)
> Illegal instruction

For the record, I can’t reproduce it with the QEMU image at
<http://hydra.nixos.org/build/2729559/download/1/disk-image.qcow2>,
which contains Guile 2.0.5.178 (from the ‘stable-2.0’ branch),
cross-built with thread support.

Something like this works fine:

  (map (compose join-thread
                (lambda (i) (call-with-new-thread (lambda () i))))
       (iota 1000))

Thanks,
Ludo’.

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

From: Cyril Roelandt <tipecaml <at> gmail.com>
To: bug-guile <at> gnu.org, bug-hurd <at> gnu.org, debian-hurd <at> lists.debian.org
Subject: Re: Guile
Date: Tue, 26 Jun 2012 03:27:06 +0200

[Message part 1 (text/plain, inline)]

On 06/25/2012 01:00 PM, Samuel Thibault wrote:
> Cyril Roelandt, le Mon 25 Jun 2012 03:23:04 +0200, a écrit :
>> Program received signal SIGILL, Illegal instruction.
>> 0x01413a5c in _hurd_intr_rpc_mach_msg () from /lib/i386-gnu/libc.so.0.3
>> (gdb) bt
>
> Please also post
>
> thread apply all bt
>

I attached the log.

Cyril.

[gdb-all-threads (text/plain, attachment)]

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

From: Samuel Thibault <samuel.thibault <at> gnu.org>
To: Cyril Roelandt <tipecaml <at> gmail.com>
Cc: bug-guile <at> gnu.org, debian-hurd <at> lists.debian.org, bug-hurd <at> gnu.org
Subject: Re: Guile
Date: Tue, 26 Jun 2012 07:26:45 -0300

Cyril Roelandt, le Tue 26 Jun 2012 03:27:06 +0200, a écrit :
> On 06/25/2012 01:00 PM, Samuel Thibault wrote:
> >Cyril Roelandt, le Mon 25 Jun 2012 03:23:04 +0200, a écrit :
> >>Program received signal SIGILL, Illegal instruction.
> >>0x01413a5c in _hurd_intr_rpc_mach_msg () from /lib/i386-gnu/libc.so.0.3
> >>(gdb) bt
> >
> >Please also post
> >
> >thread apply all bt
> >
> 
> I attached the log.
> 
> Cyril.
> 
> Thread 6 (Thread 29520.6):
> #0  0x01412036 in ?? () from /lib/i386-gnu/libc.so.0.3

This is where SIGILL actually happens (this is the firewall function in
libc, sysdeps/mach/hurd/i386/trampoline.c). Do you have the libc0.3-dbg
package installed? it's odd we don't have debugging signals. Anyway, the
offset seems to be 0x13da000, thus the functions below.

I don't know _hurd_setup_sighandler very much and don't have time to
dive in it, but that's most probably where to look at. Just an idea: is
guile perhaps using SA_ONSTACK? It's not yet supported, we are still
waiting for moving the threadvars into TLS before enabling it.

> #1  0x015bf1a5 in io_read () from /lib/i386-gnu/libhurduser.so.0.3
> #2  0x01418321 in ?? () from /lib/i386-gnu/libc.so.0.3
readfd (nested function)

> #3  0x0141e880 in ?? () from /lib/i386-gnu/libc.so.0.3
_hurd_ctty_input

> #4  0x01417d44 in ?? () from /lib/i386-gnu/libc.so.0.3
_hurd_fd_read

> #5  0x014d75a2 in read () from /lib/i386-gnu/libc.so.0.3
> #6  0x010d21f4 in read_signal_pipe_data (data=data <at> entry=0x1bff6e4) at scmsigs.c:158
> #7  0x010fa2ac in with_gc_inactive (data=0x1bff6e4, func=0x10d21c0 <read_signal_pipe_data>) at threads.c:249
> #8  scm_without_guile (func=func <at> entry=0x10d21c0 <read_signal_pipe_data>, data=data <at> entry=0x1bff6e4) at threads.c:964
> #9  0x010d2144 in signal_delivery_thread (data=0x0) at scmsigs.c:178
> #10 0x010fc734 in apply_catch_closure (clo=0x838eef0, args=0x304) at throw.c:146
> #11 0x01109be0 in vm_debug_engine (vm=0x8390fd0, program=0x156f534, argv=0x83920c0, nargs=137948912) at vm-i-system.c:960
> #12 0x0111564a in scm_c_vm_run (vm=0x8390fd0, program=program <at> entry=0x8143a50, argv=argv <at> entry=0x1bff9a4, 
>     nargs=nargs <at> entry=3) at vm.c:567
> #13 0x0107b5cc in scm_call_3 (proc=0x8143a50, arg1=arg1 <at> entry=0x404, arg2=arg2 <at> entry=0x838eef0, arg3=arg3 <at> entry=0x838eee0)
>     at eval.c:500
> #14 0x010fcb08 in scm_catch (key=key <at> entry=0x404, thunk=thunk <at> entry=0x838eef0, handler=handler <at> entry=0x838eee0)
>     at throw.c:73
> #15 0x010fcc1c in scm_catch_with_pre_unwind_handler (key=key <at> entry=0x404, thunk=thunk <at> entry=0x838eef0, 
>     handler=handler <at> entry=0x838eee0, pre_unwind_handler=0x904) at throw.c:81
> #16 0x010fcd9c in scm_c_catch (tag=tag <at> entry=0x404, body=body <at> entry=0x10d20f0 <signal_delivery_thread>, 
>     body_data=body_data <at> entry=0x0, handler=handler <at> entry=0x10fd220 <scm_handle_by_message>, 
>     handler_data=handler_data <at> entry=0x113aba0, pre_unwind_handler=pre_unwind_handler <at> entry=0, 
>     pre_unwind_handler_data=pre_unwind_handler_data <at> entry=0x0) at throw.c:213
> #17 0x010fcdfb in scm_internal_catch (tag=tag <at> entry=0x404, body=body <at> entry=0x10d20f0 <signal_delivery_thread>, 
>     body_data=body_data <at> entry=0x0, handler=handler <at> entry=0x10fd220 <scm_handle_by_message>, 
>     handler_data=handler_data <at> entry=0x113aba0) at throw.c:222
> #18 0x010fa79e in really_spawn (d=0x17ff24c) at threads.c:1091
> #19 0x01071470 in c_body (d=0x1bffe64) at continuations.c:512
> #20 0x010fc734 in apply_catch_closure (clo=0x838efc0, args=0x304) at throw.c:146
> #21 0x01109be0 in vm_debug_engine (vm=0x8390fd0, program=0x156f534, argv=0x8392060, nargs=137949120) at vm-i-system.c:960
> #22 0x0111564a in scm_c_vm_run (vm=0x8390fd0, program=program <at> entry=0x8143a50, argv=argv <at> entry=0x1bffd70, 
>     nargs=nargs <at> entry=4) at vm.c:567
> #23 0x0107b634 in scm_call_4 (proc=0x8143a50, arg1=arg1 <at> entry=0x404, arg2=arg2 <at> entry=0x838efc0, arg3=arg3 <at> entry=0x838efb0, 
>     arg4=arg4 <at> entry=0x838efa0) at eval.c:507
> #24 0x010fcbb6 in scm_catch_with_pre_unwind_handler (key=key <at> entry=0x404, thunk=thunk <at> entry=0x838efc0, 
>     handler=handler <at> entry=0x838efb0, pre_unwind_handler=0x838efa0) at throw.c:86
> #25 0x010fcd9c in scm_c_catch (tag=tag <at> entry=0x404, body=body <at> entry=0x1071460 <c_body>, 
>     body_data=body_data <at> entry=0x1bffe64, handler=handler <at> entry=0x10718f0 <c_handler>, 
>     handler_data=handler_data <at> entry=0x1bffe64, pre_unwind_handler=pre_unwind_handler <at> entry=0x10715b0 <pre_unwind_handler>, 
>     pre_unwind_handler_data=pre_unwind_handler_data <at> entry=0x80f8be8) at throw.c:213
> #26 0x01071d03 in scm_i_with_continuation_barrier (body=body <at> entry=0x1071460 <c_body>, body_data=body_data <at> entry=0x1bffe64, 
>     handler=handler <at> entry=0x10718f0 <c_handler>, handler_data=handler_data <at> entry=0x1bffe64, 
>     pre_unwind_handler=pre_unwind_handler <at> entry=0x10715b0 <pre_unwind_handler>, pre_unwind_handler_data=0x80f8be8)
>     at continuations.c:450
> #27 0x01071de6 in scm_c_with_continuation_barrier (func=0x10fa700 <really_spawn>, data=0x17ff24c) at continuations.c:546
> #28 0x010f9fda in with_guile_trampoline (data=0x1bffef4) at threads.c:884
> #29 with_gc_active (data=0x1bffef4, func=<optimized out>) at threads.c:255
> #30 with_guile_and_parent (base=0x1bffecc, data=0x1bffef4) at threads.c:930
> #31 0x011b421e in GC_call_with_stack_base () from /usr/lib/libgc.so.1
> #32 0x010f95d1 in scm_i_with_guile_and_parent (parent=<optimized out>, data=0x17ff24c, func=0x10fa700 <really_spawn>)
>     at threads.c:945
> #33 spawn_thread (d=0x17ff24c) at threads.c:1103
> #34 0x011b896b in GC_inner_start_routine () from /usr/lib/libgc.so.1
> #35 0x011b421e in GC_call_with_stack_base () from /usr/lib/libgc.so.1
> #36 0x011b8375 in GC_start_routine () from /usr/lib/libgc.so.1
> #37 0x013d0c80 in entry_point () from /lib/i386-gnu/libpthread.so.0.3
> #38 0x00000000 in ?? ()
> 
> Thread 5 (Thread 29520.5):
> #0  0x013f67cc in ?? () from /lib/i386-gnu/libc.so.0.3
> #1  0x013f6fc9 in mach_msg () from /lib/i386-gnu/libc.so.0.3
> #2  0x013f76f9 in mach_msg_server_timeout () from /lib/i386-gnu/libc.so.0.3
> #3  0x013f77cb in mach_msg_server () from /lib/i386-gnu/libc.so.0.3
> #4  0x01407ecf in ?? () from /lib/i386-gnu/libc.so.0.3
> #5  0x013d0c80 in entry_point () from /lib/i386-gnu/libpthread.so.0.3
> #6  0x00000000 in ?? ()
> 
> Thread 4 (Thread 29520.4):
> #0  0x01413a5c in _hurd_intr_rpc_mach_msg () from /lib/i386-gnu/libc.so.0.3
> #1  0x015bb834 in dir_lookup () from /lib/i386-gnu/libhurduser.so.0.3
> #2  0x013fec38 in ?? () from /lib/i386-gnu/libc.so.0.3
> #3  0x013f90bd in _hurd_ports_use () from /lib/i386-gnu/libc.so.0.3
> #4  0x013fee7d in hurd_file_name_lookup () from /lib/i386-gnu/libc.so.0.3
> #5  0x013ff069 in file_name_lookup () from /lib/i386-gnu/libc.so.0.3
> #6  0x014d44ab in __xstat64 () from /lib/i386-gnu/libc.so.0.3
> #7  0x010a2827 in stat (__statbuf=0x17ff250, __path=<optimized out>) at /usr/include/i386-gnu/sys/stat.h:457
> #8  search_path (path=0x813fcc8, filename=filename <at> entry=0x84deda0, extensions=0x80f02c0, 
>     require_exts=require_exts <at> entry=0x4, stat_buf=stat_buf <at> entry=0x17ff250) at load.c:548
> #9  0x010a2bee in scm_primitive_load_path (args=0x84ebcd8) at load.c:863
> #10 0x01109ab4 in vm_debug_engine (vm=0x80f8c50, program=0x15d2808, argv=0x80fb428, nargs=17443584) at vm-i-system.c:892
> #11 0x0111564a in scm_c_vm_run (vm=0x80f8c50, program=program <at> entry=0x813c150, argv=argv <at> entry=0x17ff5e0, 
>     nargs=nargs <at> entry=1) at vm.c:567
> #12 0x0107bc37 in scm_primitive_eval (exp=exp <at> entry=0x8208ae0) at eval.c:685
> #13 0x0107bcb0 in scm_eval (exp=0x8208ae0, module_or_state=module_or_state <at> entry=0x81786c0) at eval.c:719
> #14 0x010d3333 in scm_shell (argc=1, argv=0x17ffd08) at script.c:441
> #15 0x0109c2ac in invoke_main_func (body_data=0x17ffc50) at init.c:336
> #16 0x01071470 in c_body (d=0x17ffb94) at continuations.c:512
> #17 0x010fc734 in apply_catch_closure (clo=0x81ec670, args=0x304) at throw.c:146
> #18 0x011140da in vm_regular_engine (vm=0x80f8c50, program=0x15d2808, argv=0x81ec670, nargs=4) at vm-i-system.c:960
> #19 0x0111564a in scm_c_vm_run (vm=0x80f8c50, program=program <at> entry=0x8143a50, argv=argv <at> entry=0x17ffaa0, 
>     nargs=nargs <at> entry=4) at vm.c:567
> #20 0x0107b634 in scm_call_4 (proc=0x8143a50, arg1=arg1 <at> entry=0x404, arg2=arg2 <at> entry=0x81ec670, arg3=arg3 <at> entry=0x81ec660, 
>     arg4=arg4 <at> entry=0x81ec650) at eval.c:507
> #21 0x010fcbb6 in scm_catch_with_pre_unwind_handler (key=key <at> entry=0x404, thunk=thunk <at> entry=0x81ec670, 
>     handler=handler <at> entry=0x81ec660, pre_unwind_handler=0x81ec650) at throw.c:86
> #22 0x010fcd9c in scm_c_catch (tag=tag <at> entry=0x404, body=body <at> entry=0x1071460 <c_body>, 
>     body_data=body_data <at> entry=0x17ffb94, handler=handler <at> entry=0x10718f0 <c_handler>, 
>     handler_data=handler_data <at> entry=0x17ffb94, pre_unwind_handler=pre_unwind_handler <at> entry=0x10715b0 <pre_unwind_handler>, 
>     pre_unwind_handler_data=pre_unwind_handler_data <at> entry=0x80f8be8) at throw.c:213
> #23 0x01071d03 in scm_i_with_continuation_barrier (body=body <at> entry=0x1071460 <c_body>, body_data=body_data <at> entry=0x17ffb94, 
>     handler=handler <at> entry=0x10718f0 <c_handler>, handler_data=handler_data <at> entry=0x17ffb94, 
>     pre_unwind_handler=pre_unwind_handler <at> entry=0x10715b0 <pre_unwind_handler>, pre_unwind_handler_data=0x80f8be8)
>     at continuations.c:450
> #24 0x01071de6 in scm_c_with_continuation_barrier (func=0x109c270 <invoke_main_func>, data=0x17ffc50) at continuations.c:546
> #25 0x010f9fda in with_guile_trampoline (data=0x17ffc24) at threads.c:884
> #26 with_gc_active (data=0x17ffc24, func=<optimized out>) at threads.c:255
> #27 with_guile_and_parent (base=0x17ffbfc, data=0x17ffc24) at threads.c:930
> #28 0x011b421e in GC_call_with_stack_base () from /usr/lib/libgc.so.1
> #29 0x010fa1ff in scm_i_with_guile_and_parent (parent=<optimized out>, data=0x17ffc50, func=0x109c270 <invoke_main_func>)
>     at threads.c:945
> #30 scm_with_guile (func=func <at> entry=0x109c270 <invoke_main_func>, data=data <at> entry=0x17ffc50) at threads.c:951
> #31 0x0109c435 in scm_boot_guile (argc=argc <at> entry=1, argv=argv <at> entry=0x17ffd08, 
>     main_func=main_func <at> entry=0x8048960 <inner_main>, closure=closure <at> entry=0x0) at init.c:319
> #32 0x08048875 in main (argc=1, argv=0x17ffd08) at guile.c:71


-- 
Samuel
<s> cool, j'ai un rapport a rendre pour le 31 decembre a minuit...
 -+- #ens-mim - bonne année ! -+-

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

From: ludo <at> gnu.org (Ludovic Courtès)
To: bug-guile <at> gnu.org
Cc: debian-hurd <at> lists.debian.org, bug-hurd <at> gnu.org
Subject: Re: Guile
Date: Thu, 28 Jun 2012 23:20:07 +0200

Samuel Thibault <samuel.thibault <at> gnu.org> skribis:

> I don't know _hurd_setup_sighandler very much and don't have time to
> dive in it, but that's most probably where to look at. Just an idea: is
> guile perhaps using SA_ONSTACK?

Neither Guile nor libgc does, AFAICS.

Ludo’.

Message #33 received at 11776-done <at> debbugs.gnu.org (full text, mbox):

From: Andy Wingo <wingo <at> pobox.com>
To: ludo <at> gnu.org (Ludovic Courtès)
Cc: Cyril Roelandt <tipecaml <at> gmail.com>, 11776-done <at> debbugs.gnu.org
Subject: Re: bug#11784: Guile on GNU/Hurd
Date: Mon, 20 Jun 2016 21:35:54 +0200

No update in some years; let's archive this one and if you find a new
build issue with 2.0.11 or 2.1.3, let's work from there :)

Andy

On Mon 25 Jun 2012 23:13, ludo <at> gnu.org (Ludovic Courtès) writes:

> Hi Cyril!
>
> Cyril Roelandt <tipecaml <at> gmail.com> skribis:
>
>> When compiling with --without-threads, I can run (+ 1 2) in the
>> interpreter successfully. Running ./check-guile gives me the following
>> result:
>>
>> Totals for this test run:
>> passes:                 34286
>> failures:               2
>> unexpected passes:      0
>> expected failures:      30
>> unresolved test cases:  575
>> untested test cases:    1
>> unsupported test cases: 15
>> errors:                 0
>
> Can you please post the ‘check-guile.log’ excerpt corresponding to these
> two failures?
>
> Thanks,
> Ludo’.

Previous Next

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