Nabble - debian-hppa

Re: Erlang VM immediately exits on hppa: execve() fails

2009-12-15T13:49:07Z

On Tue, Dec 15, 2009 at 2:24 PM, Sergei Golovan <sgolovan@...> wrote:

> On Tue, Dec 15, 2009 at 7:11 PM, <carlos@...> wrote:
>> On Tue, Dec 15, 2009 at 2:52 AM, Sergei Golovan <sgolovan@...> wrote:
>>> After the recent changes in hppa architecture Erlang virtual machine
>>> fails to start (which leads to several FTBFS bugs in particular:
>>> #559628, #559962, #559004).
>>>
>>> The problem is that after vfork(), execve() fails to start a child
>>> process. It sets errno to EMFILE. If I run erl under strace, it works
>>> fine. Is it a bug in glibc, or in Erlang itself?
>>>
>>> I'm giving a small excerpt from erts/emulator/sys/unix/sys.c file
>>> where the failed execve(), the rest could be found in erlang
>>> 12.b.2.1-1 sources.
>>
>> Are you certain that execve() fails to start a child process?
>
> Yes, after I've inserted some debug printf() right after execve() it
> was reached.

Yes, you are definitely right, I hadn't thought to try that.

I will test an instrumented kernel tonight to see what the problem is.

Cheers,
Carlos.

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: Erlang VM immediately exits on hppa: execve() fails

2009-12-15T11:24:24Z

On Tue, Dec 15, 2009 at 7:11 PM, <carlos@...> wrote:

> On Tue, Dec 15, 2009 at 2:52 AM, Sergei Golovan <sgolovan@...> wrote:
>> After the recent changes in hppa architecture Erlang virtual machine
>> fails to start (which leads to several FTBFS bugs in particular:
>> #559628, #559962, #559004).
>>
>> The problem is that after vfork(), execve() fails to start a child
>> process. It sets errno to EMFILE. If I run erl under strace, it works
>> fine. Is it a bug in glibc, or in Erlang itself?
>>
>> I'm giving a small excerpt from erts/emulator/sys/unix/sys.c file
>> where the failed execve(), the rest could be found in erlang
>> 12.b.2.1-1 sources.
>
> Are you certain that execve() fails to start a child process?

Yes, after I've inserted some debug printf() right after execve() it
was reached.

>
> While debugging emacs32 failures I discovered that execve() works, but
> that the parent which called the vfork() is corrupted by the child.

But you're right, parent process becomes seems to be corrupt after
vfork(). (At least test printf() doesn't print anything.

> We haven't been able to determine the cause of the failure yet.

OK, I see.

Cheers!
--
Sergei Golovan

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: Erlang VM immediately exits on hppa: execve() fails

2009-12-15T08:11:27Z

On Tue, Dec 15, 2009 at 2:52 AM, Sergei Golovan <sgolovan@...> wrote:

> After the recent changes in hppa architecture Erlang virtual machine
> fails to start (which leads to several FTBFS bugs in particular:
> #559628, #559962, #559004).
>
> The problem is that after vfork(), execve() fails to start a child
> process. It sets errno to EMFILE. If I run erl under strace, it works
> fine. Is it a bug in glibc, or in Erlang itself?
>
> I'm giving a small excerpt from erts/emulator/sys/unix/sys.c file
> where the failed execve(), the rest could be found in erlang
> 12.b.2.1-1 sources.

Are you certain that execve() fails to start a child process?

While debugging emacs32 failures I discovered that execve() works, but
that the parent which called the vfork() is corrupted by the child.

I posted a test case for our kernel developers to look at here:
http://article.gmane.org/gmane.linux.ports.parisc/2403

We haven't been able to determine the cause of the failure yet.

Cheers,
Carlos.

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: access to hppa machine to work on Bug#558980

2009-12-15T06:55:11Z

> On Mon, Dec 14, 2009 at 07:46:52AM -0500, Stephen Leake wrote:
> ...
> > Ludovic also suggested removing '-fstack-check' from the list of
> > compiler options. I did that, rebuilt the static and dynamic
> > libraries, and the bug went away; the test code works with both static
> > and dynamic libraries.
>
> Dave, Carlos,
> Would this imply something might be broken in the stack unwinding?

Possibly. An off by one error was recently fixed on GCC head.
This affects Ada but not Java. Java had code to adjust the program
counter by one instruction when catching signals. The change could
be backported to 4.4, but I don't intend to do it because the behavior
isn't a regression and it changes the behavior of libgcc_s.so. We
need a version bump for libgcc_s.so in 4.5.

We don't have any backend support for '-fstack-check'. Eric
Botcazou has recently done some updates which I think fix some issues
with generic stack checking. See for example,

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20548

So, it might help to try a recent 4.5.0 snapshot.

The fact that the stack grows upwards on hppa is a bit unusual and
might be a source of problems. It is possible that we could add backend
support for stack checking using the probe instruction, etc, but I don't
have time to look into it until we finish moving. There's still a few
remaining issues for 4.5.0 to be addressed.

> Or is this more likely to be something specific to Gnu ADA?

Dave
--
J. David Anglin dave.anglin@...
National Research Council of Canada (613) 990-0752 (FAX: 952-6602)

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Erlang VM immediately exits on hppa: execve() fails

2009-12-14T23:52:04Z

Hi!

After the recent changes in hppa architecture Erlang virtual machine
fails to start (which leads to several FTBFS bugs in particular:
#559628, #559962, #559004).

The problem is that after vfork(), execve() fails to start a child
process. It sets errno to EMFILE. If I run erl under strace, it works
fine. Is it a bug in glibc, or in Erlang itself?

I'm giving a small excerpt from erts/emulator/sys/unix/sys.c file
where the failed execve(), the rest could be found in erlang
12.b.2.1-1 sources.

pid = vfork();

if (pid == 0) {
/* The child! */

/* Observe!
* OTP-4389: The child setup program (implemented in
* erl_child_setup.c) will perform the necessary setup of the
* child before it execs to the user program. This because
* vfork() only allow an *immediate* execve() or _exit() in the
* child.
*/
execve(child_setup_prog, cs_argv, new_environ);
_exit(1);
}

Cheers!
--
Sergei Golovan

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: access to hppa machine to work on Bug#558980

2009-12-14T21:38:27Z

On Mon, Dec 14, 2009 at 07:46:52AM -0500, Stephen Leake wrote:
...
> Ludovic also suggested removing '-fstack-check' from the list of
> compiler options. I did that, rebuilt the static and dynamic
> libraries, and the bug went away; the test code works with both static
> and dynamic libraries.

Dave, Carlos,
Would this imply something might be broken in the stack unwinding?
Or is this more likely to be something specific to Gnu ADA?

thanks,
grant

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: SIGFPE trapping on HPPA

2009-12-14T16:01:24Z

Nice test case!

Thibaut Paumard wrote:
> Compile with
> gcc -lm -o fputest fputest.c

It's kind of funny that gcc only warns about div by zero for the integer
case:

fputest.c: In function ‘worker’:
fputest.c:73: warning: division by zero

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: openoffice.org port there ;)

2009-12-14T14:52:43Z

On Mon, Dec 14, 2009 at 4:52 PM, Rene Engelhard <rene@...> wrote:

> ene@frodo:~/Debian/Pakete/openoffice.org/openoffice.org-3.2.0/debian$ grep Architecture: control | sort | uniq
> Architecture: all
> Architecture: i386 amd64 ia64 sparc
> Architecture: i386 mips mipsel s390 armel amd64 ia64 s390x sparc hppa
> Architecture: i386 powerpc mips mipsel armel s390 m68k amd64 ia64 ppc64 s390x sparc hppa
> ^^^^
> Architecture: i386 powerpc mips mipsel s390 arm armel amd64 ia64 ppc64 s390x sparc hppa
> Architecture: i386 powerpc mips mipsel s390 armel amd64 ia64 ppc64 s390x sparc hppa
> Architecture: i386 powerpc mips mipsel s390 m68k amd64 ia64 ppc64 s390x sparc hppa
> Architecture: i386 powerpc s390 amd64 ia64 ppc64 s390x sparc
>
> Yes, RedHat created a hppa port, so there it is. Even with Java :-) (gcj-jdk)
>
> Will be enabled in the next upload of the 3.2 branch; a older snapshot of
> the packaging was also built (with some hacks, but they're not relevant for
> the overall standing) on paer successfully, you can get it from my home
> there if you want to play with it :-)

Thanks!

Cheers,
Carlos.

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: SIGFPE trapping on HPPA (was: Yorick FTBFS: DM needs access to HPPA machine)

2009-12-14T14:08:02Z

Hi,

Le 14 déc. 09 à 16:06, Carlos O'Donell a écrit :
> Could you please provide a small test case and describe the expected
> behaviour?

Attached is a rather small testcase.

Compile with
gcc -lm -o fputest fputest.c

Run with
./fputest
or with any succession of 'i's and 'f's, e.g.
./fputest iffifi

The expected behaviour (checked on i386) is that for each f or i
letter, a line is output, for instance
$ ./fputest iffi
Triggering integer SIGFPE: 1/0=(SIGFPE trapped)0
Triggering floating SIGFPE: 1./0.=(SIGFPE trapped)0.000000
Triggering floating SIGFPE: 1./0.=(SIGFPE trapped)0.000000
Triggering integer SIGFPE: 1/0=(SIGFPE trapped)0
$ echo $?
0

All the SIGFPEs should be trapped and the program should exit
gracefully (check $?).

Under HPPA, the integer SIGFPEs are trapped but the floating-point
SIGFPEs are not:
$ ./fputest iffi
Triggering integer SIGFPE: 1/0=(SIGFPE trapped)0
Triggering floating SIGFPE: 1./0.=Floating point exception
$ echo $?
136

It's great if you can check why this is.

Best regards, Thibaut.

fputest.c (2K) Download Attachment

openoffice.org port there ;)

2009-12-14T13:52:35Z

Hi,

ene@frodo:~/Debian/Pakete/openoffice.org/openoffice.org-3.2.0/debian$ grep Architecture: control | sort | uniq
Architecture: all
Architecture: i386 amd64 ia64 sparc
Architecture: i386 mips mipsel s390 armel amd64 ia64 s390x sparc hppa
Architecture: i386 powerpc mips mipsel armel s390 m68k amd64 ia64 ppc64 s390x sparc hppa
^^^^
Architecture: i386 powerpc mips mipsel s390 arm armel amd64 ia64 ppc64 s390x sparc hppa
Architecture: i386 powerpc mips mipsel s390 armel amd64 ia64 ppc64 s390x sparc hppa
Architecture: i386 powerpc mips mipsel s390 m68k amd64 ia64 ppc64 s390x sparc hppa
Architecture: i386 powerpc s390 amd64 ia64 ppc64 s390x sparc

Yes, RedHat created a hppa port, so there it is. Even with Java :-) (gcj-jdk)

Will be enabled in the next upload of the 3.2 branch; a older snapshot of
the packaging was also built (with some hacks, but they're not relevant for
the overall standing) on paer successfully, you can get it from my home
there if you want to play with it :-)

Grüße/Regards,

Rene
--
.''`. René Engelhard -- Debian GNU/Linux Developer
: :' : http://www.debian.org | http://people.debian.org/~rene/
`. `' rene@... | GnuPG-Key ID: D03E3E70
`- Fingerprint: E12D EA46 7506 70CF A960 801D 0AA0 4571 D03E 3E70

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: access to hppa machine to work on Bug#558980

2009-12-14T07:18:03Z

On Mon, Dec 14, 2009 at 7:46 AM, Stephen Leake
<stephen_leake@...> wrote:
>> If you see "transferring control:" then the dynamic loader has just
>> handed off control the real program, and any fault after that is
>> possibly related to the real program.
>
> That is there, although there are more symbols looked up after it.

This is good news, it means the dynamic loader handed control to the program.

> So I will leave hppa in the arch list, but remove the stack check
> option.

Would it be possible for you to keep open a low severity bug
indicating that this needs to be fixed?

Remember to tag the bug hppa, so that I can track it on our ports
to-do list e.g.
http://bugs.debian.org/cgi-bin/pkgreport.cgi?users=debian-hppa@...;which=tag&data=hppa&archive=no

We are working through the tagged bugs to try make progress.

Cheers,
Carlos.

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: SIGFPE trapping on HPPA (was: Yorick FTBFS: DM needs access to HPPA machine)

2009-12-14T07:06:42Z

On Mon, Dec 14, 2009 at 9:32 AM, Thibaut Paumard <mlotpot.news@...> wrote:

> Under some circumstances (when and integer SIGFPE has been raised
> previously), I have been able to have my handler output a short message, but
> then it loops (this short message is repeated indefinitely). Calling
> feclearexcepts() before outputting this message did not help: when
> attempting a floating-point division by zero, either yorick dies or the
> handler loops. The first three lines of the handler on this test case are:
> feclearexcept(FE_ALL_EXCEPT);
> printf("in u_sigfpe\n");
> fflush(stdout);
>
> The handler is setup using signal(2):
> signal(SIGFPE, &u_sigfpe)
>
> HPPA is the only architecture so far which exhibits this behaviour.
>
> Does anyone has an idea of what I could do to fix this? Is there something
> specific in the way HPPA handles SIGFPE trapping? Why is it different
> between integer SIGFPE and floating SIGFPE?

Could you please provide a small test case and describe the expected behaviour?

I'll look at the test case and see if there is any corner cases
missing from the glibc floating point exception handling code.

Cheers,
Carlos.

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

SIGFPE trapping on HPPA (was: Yorick FTBFS: DM needs access to HPPA machine)

2009-12-14T06:32:14Z

> [1] http://packages.debian.org/sid/yorick
> [2] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=559406

Hello,

Thanks to access to Frans I have been able to test a little bit more
this issue.

It turns out the Yorick code is able to trap SIGFPE when it is
triggered by an integer division (1/0) but not when it is triggered by
a floating point division (1./0.).

In the former case, Yorick correctly signals the exception to the user
with a Yorick error message and stops to wait for input, whereas in
the latter it just dies with the single line "Floating point exception".

Under some circumstances (when and integer SIGFPE has been raised
previously), I have been able to have my handler output a short
message, but then it loops (this short message is repeated
indefinitely). Calling feclearexcepts() before outputting this message
did not help: when attempting a floating-point division by zero,
either yorick dies or the handler loops. The first three lines of the
handler on this test case are:
feclearexcept(FE_ALL_EXCEPT);
printf("in u_sigfpe\n");
fflush(stdout);

The handler is setup using signal(2):
signal(SIGFPE, &u_sigfpe)

HPPA is the only architecture so far which exhibits this behaviour.

Does anyone has an idea of what I could do to fix this? Is there
something specific in the way HPPA handles SIGFPE trapping? Why is it
different between integer SIGFPE and floating SIGFPE?

Best regards, Thibaut.

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: access to hppa machine to work on Bug#558980

2009-12-14T04:46:52Z

"Carlos O'Donell" <carlos@...> writes:

> On Sun, Dec 13, 2009 at 10:03 AM, Stephen Leake
> <stephen_leake@...> wrote:
>> Frans Pop gave me access to his machine. I have some more information
>> on the bug.
>>
>> If I compile from full GNADE source (not using the GNADE dynamic or
>> static libraries), the code works.
>
> What does "from full GNADE source" mean?

Ludovic answered this nicely.

>> With the dynamic library, I get a SIGSEGV, somewhere in the program
>> startup, before any user code.
>
> Export LD_DEBUG=all, run the program, this will tell you what the
> dynamic loader is doing every step of the way.

This shows lots of symbols being resolved, nothing helpful.

> If you see "transferring control:" then the dynamic loader has just
> handed off control the real program, and any fault after that is
> possibly related to the real program.

That is there, although there are more symbols looked up after it.

>> With the static library, I get a stack overflow (caught and reported
>> by the Ada runtime), during elaboration of a compiler-provided
>> container library.
>
> Could you please define "during elaboration?"

Ludovic answered this nicely.

Ludovic also suggested removing '-fstack-check' from the list of
compiler options. I did that, rebuilt the static and dynamic
libraries, and the bug went away; the test code works with both static
and dynamic libraries.

So I will leave hppa in the arch list, but remove the stack check
option.

--
-- Stephe

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: access to hppa machine to work on Bug#558980

2009-12-14T02:52:32Z

Carlos O'Donell wrote:
> On Sun, Dec 13, 2009 at 10:03 AM, Stephen Leake wrote:
>> Frans Pop gave me access to his machine. I have some more information
>> on the bug.
>>
>> If I compile from full GNADE source (not using the GNADE dynamic or
>> static libraries), the code works.
>
> What does "from full GNADE source" mean?

I think that means "recompile the application program and the sources of
the library into a single, statically-linked program" as opposed to
"compile the library sources, create a (static|shared) library and link
into the executable".

[...]
>> With the static library, I get a stack overflow (caught and reported
>> by the Ada runtime), during elaboration of a compiler-provided
>> container library.
>
> Could you please define "during elaboration?"

It seems only Ada programmers know about elaboration :)

In a nutshell, elaboration is all that happens between the transfer of
control from the dynamic loader to the executable and the start of
execution of the main subprogram (main() in C and C++, anything in Ada).
Elaboration consists in allocating and initialalizing nonstatic global
variables, i.e. those declared outside of any subprogram but whose value is
unknown at compile time. This might involve calling user-defined functions
that return the initial value of a variable. Subprograms and Ada packages,
too, are elaborated; their elaboration consists only in making sure all the
global variables and other subprograms they depend on are already
elaborated. Example:

-*- mode: c -*-
#include <stdlib.h>
int i = foo ();
int main () { puts ("this is called after elaboration\n"); return i; }
int foo () { puts ("this is called during elaboration\n"); return rand ();
}

The example illustrates the importance of controlling elaboration order. In
this case, i must be elaborated before main() is called but after puts()
and rand () are elaborated.

Ada offers precise control of elaboration order and the compiler can report
the elaboration order it has chosen for a given executable program.

--
Ludovic Brenta.

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: access to hppa machine to work on Bug#558980

2009-12-13T08:55:05Z

Stephen Leake wrote:
> Frans Pop gave me access to his machine. I have some more information
> on the bug.

I've forwarded this mail and Carlos' reply to the bug report.

Please don't forget to CC the BR when discussing the bug itself! That's
where people will look for it later; they won't look on the list.

Cheers,
FJP

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: access to hppa machine to work on Bug#558980

2009-12-13T07:39:17Z

On Sun, Dec 13, 2009 at 10:03 AM, Stephen Leake
<stephen_leake@...> wrote:
> Frans Pop gave me access to his machine. I have some more information
> on the bug.
>
> If I compile from full GNADE source (not using the GNADE dynamic or
> static libraries), the code works.

What does "from full GNADE source" mean?

> With the dynamic library, I get a SIGSEGV, somewhere in the program
> startup, before any user code.

Export LD_DEBUG=all, run the program, this will tell you what the
dynamic loader is doing every step of the way.

If you see "transferring control:" then the dynamic loader has just
handed off control the real program, and any fault after that is
possibly related to the real program.

> With the static library, I get a stack overflow (caught and reported
> by the Ada runtime), during elaboration of a compiler-provided
> container library.

Could you please define "during elaboration?"

> So the problem is related to libraries, which means it's probably in
> the linker and/or loader somewhere.
>
> It seems clear that no user will be able to use GNADE in its current
> form on hppa, so I should remove hppa from the arch list in
> debian/control for the squeeze release.

Thanks for the help.

Cheers,
Carlos.

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: access to hppa machine to work on Bug#558980

2009-12-13T07:03:24Z

Stephen Leake <stephen_leake@...> writes:

> I'm the Debian GNADE maintainer, but not yet a Debian Developer.
> Bug#558980 is a failure of the GNADE 'make check'. It works fine on my
> x86, and several other architectures, so this appears to be an
> arch-specific bug. I'd like to work on fixing it, but I don't have
> access to an hppa machine.
>
> Can someone give me access to an hppa machine? I should have time this
> weekend to work on this.

Frans Pop gave me access to his machine. I have some more information
on the bug.

If I compile from full GNADE source (not using the GNADE dynamic or
static libraries), the code works.

With the dynamic library, I get a SIGSEGV, somewhere in the program
startup, before any user code.

With the static library, I get a stack overflow (caught and reported
by the Ada runtime), during elaboration of a compiler-provided
container library.

So the problem is related to libraries, which means it's probably in
the linker and/or loader somewhere.

It seems clear that no user will be able to use GNADE in its current
form on hppa, so I should remove hppa from the arch list in
debian/control for the squeeze release.

If there is some more debugging I could do?

--
-- Stephe

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

¡Ven y únete a Pro-Red.tk!

2009-12-12T14:00:56Z

Network Email

Participa conmigo en Pro-Red.tk

COMUNIDAD OFICIAL DEL EQUIPO PRO-RED

Hector Gòmez tiene:

39 amigos

¡2PLUS7 EL NUEVO GIGANTE DEL INTERNET!

Haz clic para unirte

Miembros en Pro-Red.tk:

alvaro manuel nieto moreno

SANDRA JIMENA BECERRA LULIGO

clemente alvarado

Patricio Eugenio Espinoza Villar

rita vieyra

Sobre Pro-Red.tk

Hemos cumplido 5 meses y 2plus7 ha pagado $1,400.000 de dólares a sus miembros con solo $27!

790 miembros
247 fotos
44 canciones
26 videos

17 discusiones
9 Eventos
29 publicaciones en el blog

Para controlar los correos electrónicos que recibes en Pro-Red.tk, haz clic aquí

Re: access to hppa machine to work on Bug#558980

2009-12-12T05:33:41Z

On Saturday 12 December 2009, Stephen Leake wrote:
> Would it help to find the bug, if I can reduce the GNADE failure to a
> small test case? It would be in Ada.

Yes, small test cases are obviously much easier to debug for a porter than
having to first spend time learning about a specific package. It would be
great if you could post a simple, targeted test case to the bug report.

> If this bug is not fixed, I have two choices for GNADE:
>
> 1) remove hppa from the list of supported architectures.
>
> 2) remove the 'check' target from debian/rules, and hope that user
> code will not hit the same bug.
>
> Which is the recommended approach?

I think this question is best asked on the debian-release list. I do know
of at least one other package that has (temporarily) disabled its test
suite for hppa.

It depends a bit on when a porter (which I'm not) could look into the
issue. Carlos does most of the work ATM and AFAIK he has a number of other
issues on his TODO list.

Cheers,
FJP

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: access to hppa machine to work on Bug#558980

2009-12-11T23:39:15Z

Frans Pop <elendil@...> writes:

>> I'm the Debian GNADE maintainer, but not yet a Debian Developer.
>> Bug#558980 is a failure of the GNADE 'make check'. It works fine on my
>> x86, and several other architectures, so this appears to be an
>> arch-specific bug. I'd like to work on fixing it, but I don't have
>> access to an hppa machine.
>
> It fails with a segfault. As there are quite a few other packages that only
> fail on HPPA in a similar way, it is likely that this is due to a general
> HPPA bug, and not an issue specific to GNAT.

That makes sense. I assume you mean a bug in the gcc backend for HPPA.

Would it help to find the bug, if I can reduce the GNADE failure to a
small test case? It would be in Ada.

If this bug is not fixed, I have two choices for GNADE:

1) remove hppa from the list of supported architectures.

2) remove the 'check' target from debian/rules, and hope that user
code will not hit the same bug.

Which is the recommended approach?

--
-- Stephe

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: Yorick FTBFS: DM needs access to HPPA machine

2009-12-11T11:55:57Z

On Friday 11 December 2009, Frans Pop wrote:

> Thibaut Paumard wrote:
> > I am the maintainer of the Yorick packages. [1]Yorick is an
> > interpreter used for data analysis and number crunching in astronomy
> > and other research fields.
> >
> > With the last upload, I introduced a test suite. One of the tests
> > fails on HPPA.
> >
> > Correcting the [2]FTBFS is easy: just need to remove the test. But I
> > would need access to a machine to actually investigate and correct the
> > bug.
> >
> > More specifically, the tests intentionally triggers a SIGFPE that
> > Yorick should, but doesn't, catch.
>
> You're welcome to use my box over the weekend. If that's OK, please send
> me a _private_ mail with your public openssh key and I'll set up an
> account and install the build deps.

SSH key received; I've created the account and sent the information how to
use it privately.

Problem is however that my private mails are not getting anywhere. I've
received "delivery has been delayed" messages from my ISP. I suspect some
kind of overly aggressive spam filtering to be the cause.

Thibaut: please let me know how I can get private mails to you.

Cheers,
FJP

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: access to hppa machine to work on Bug#558980

2009-12-11T06:47:07Z

Carlos O'Donell wrote:
> Do we have a machine available that Stephen could use?

Nice timing :-P

I've just sent a mail offering the use of my box.

Cheers,
FJP

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: access to hppa machine to work on Bug#558980

2009-12-11T06:46:42Z

On Fri, Dec 11, 2009 at 9:40 AM, Frans Pop <elendil@...> wrote:

>> I'm the Debian GNADE maintainer, but not yet a Debian Developer.
>> Bug#558980 is a failure of the GNADE 'make check'. It works fine on my
>> x86, and several other architectures, so this appears to be an
>> arch-specific bug. I'd like to work on fixing it, but I don't have
>> access to an hppa machine.
>
> It fails with a segfault. As there are quite a few other packages that only
> fail on HPPA in a similar way, it is likely that this is due to a general
> HPPA bug, and not an issue specific to GNAT.
>
>> Can someone give me access to an hppa machine? I should have time this
>> weekend to work on this.
>
> If you still want to work on it given the above, you're welcome to use my
> box. Please send me _privately_ your public SSH key in a signed email and
> I'll sent up an account and install the build deps.

Thanks Frans.

Cheers,
Carlos.

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: Yorick FTBFS: DM needs access to HPPA machine

2009-12-11T06:44:13Z

Thibaut Paumard wrote:

> I am the maintainer of the Yorick packages. [1]Yorick is an
> interpreter used for data analysis and number crunching in astronomy
> and other research fields.
>
> With the last upload, I introduced a test suite. One of the tests
> fails on HPPA.
>
> Correcting the [2]FTBFS is easy: just need to remove the test. But I
> would need access to a machine to actually investigate and correct the
> bug.
>
> More specifically, the tests intentionally triggers a SIGFPE that
> Yorick should, but doesn't, catch.

You're welcome to use my box over the weekend. If that's OK, please send me
a _private_ mail with your public openssh key and I'll set up an account
and install the build deps.

Cheers,
FJP

signature.asc (204 bytes) Download Attachment

Re: access to hppa machine to work on Bug#558980

2009-12-11T06:40:58Z

> I'm the Debian GNADE maintainer, but not yet a Debian Developer.
> Bug#558980 is a failure of the GNADE 'make check'. It works fine on my
> x86, and several other architectures, so this appears to be an
> arch-specific bug. I'd like to work on fixing it, but I don't have
> access to an hppa machine.

It fails with a segfault. As there are quite a few other packages that only
fail on HPPA in a similar way, it is likely that this is due to a general
HPPA bug, and not an issue specific to GNAT.

> Can someone give me access to an hppa machine? I should have time this
> weekend to work on this.

If you still want to work on it given the above, you're welcome to use my
box. Please send me _privately_ your public SSH key in a signed email and
I'll sent up an account and install the build deps.

Cheers,
FJP

signature.asc (204 bytes) Download Attachment

Re: access to hppa machine to work on Bug#558980

2009-12-11T06:35:21Z

On Fri, Dec 11, 2009 at 7:55 AM, Stephen Leake
<stephen_leake@...> wrote:
> I'm the Debian GNADE maintainer, but not yet a Debian Developer.
> Bug#558980 is a failure of the GNADE 'make check'. It works fine on my
> x86, and several other architectures, so this appears to be an
> arch-specific bug. I'd like to work on fixing it, but I don't have
> access to an hppa machine.
>
> Can someone give me access to an hppa machine? I should have time this
> weekend to work on this.

Dan,

Do we have a machine available that Stephen could use?

Cheers,
Carlos.

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

access to hppa machine to work on Bug#558980

2009-12-11T04:55:56Z

Hi.

I'm the Debian GNADE maintainer, but not yet a Debian Developer.
Bug#558980 is a failure of the GNADE 'make check'. It works fine on my
x86, and several other architectures, so this appears to be an
arch-specific bug. I'd like to work on fixing it, but I don't have
access to an hppa machine.

Can someone give me access to an hppa machine? I should have time this
weekend to work on this.

--
-- Stephe

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Richiesta autorizzazione

2009-12-10T03:12:07Z

Buongiorno,

sono Giorgia Ferrari, assistente Marketing di Forum Media Srl, società operante nel campo dell'editoria e della formazione professionale.

La presente per richiederLe l'autorizzazione, ai sensi della vigente normativa sulla Privacy, ad inviarLe, a titolo informativo, tramite e-mail qualche dettaglio sui nostri recenti prodotti editoriali specifici per il settore Tecnico-Amministrativo, Ambiente e Sicurezza, Risorse Umane, Management e Formazione Professionale.

A questo proposito vorremmo inviarLe, a titolo informativo, schede prodotto dettagliate in modo che Lei possa valutare un Suo possibile interesse. Nel caso in cui preferisca ricevere dette informazioni via fax, oppure ad altro indirizzo di posta elettronica, voglia cortesemente trasmetterci i riferimenti.

Teniamo a precisare che, nel rispetto rigoroso della normativa sulla privacy, valuteremo solo autorizzazioni con la modalità opt-in, ossia non invieremo nulla senza un Suo regolare ed espresso assenso.

RingraziandoLa per l'attenzione, Le porgo i miei più cordiali saluti.

Giorgia Ferrari
FORUM MEDIA
www.forum-media.it

Il vostro indirizzo e-mail è trattato in conformità all'Art. 13 del D.lgs 30 giugno 2003 n.196, garantendo la massima riservatezza ed al fine di fornire utili informazioni commerciali per la Sua/Vostra attività. Se non volete ricevere più messaggi inviati per conto dell'operatore si prega di inviare un messaggio all'indirizzo unsubscribe@....

Come join me on Government Lists - U.S. | Canada

2009-12-09T06:28:10Z

Network Email

Join me on Government Lists - U.S. | Canada

REACH YOUR KEY GOVERNMENT AUDIENCE EASILY!

government--
mailing--
lists.com has:

1 blog post

I welcome you to join me on this business network.

REACH YOUR KEY GOVERNMENT AUDIENCE EASILY!
http://www.government-mailing-lists.com

Click to Join

Members on Government Lists - U.S. | Canada:

government--
mailing--
lists.com

Nitin Chaudhry

Ryan Sharpe

Richard Furlong

Jim Kellough

About Government Lists - U.S. | Canada

government-mailing-lists.com is the key-contact database for reaching government markets. Government Lists – U.S. | Canada

10 photos

To control which emails you receive on Government Lists - U.S. | Canada, click here

Re: Bug#559638: FTBFS [hppa]: unrecognized command line option "-m32"

2009-12-08T06:57:41Z

Hi,

On Tue, 2009-12-08 at 01:23 +0100, Domenico Andreoli wrote:
> Hi,
>
> -m32 is a machine dependent option, on x86 it is used to build 32bit
> code whereas -m64 is for 64bit code. it is not available on hppa
> because userland is 32bit only.

I believe I have fixed the problem in my current working release.
Unfortunately, I cannot test it, as paer.debian.org does not allow me to
log in (Error: 'Permission denied (publickey).'). I will upload a new
revision soon, and I hope that it will build OK.

Yours,

--
___ | Christian Holm Christensen
|_| | -------------------------------------------------------------
| | Address: Sankt Hansgade 23, 4 Phone: (+45) 35 35 96 91
_| DK-2200 Copenhagen N Cell: (+45) 24 61 85 91
_| Denmark Office: (+45) 353 25 447
____| Email: cholm@... Web: http://cern.ch/cholm
| |

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Yorick FTBFS: DM needs access to HPPA machine

2009-12-08T02:03:09Z

Hello,

I am the maintainer of the Yorick packages. [1]Yorick is an
interpreter used for data analysis and number crunching in astronomy
and other research fields.

With the last upload, I introduced a test suite. One of the tests
fails on HPPA.

Correcting the [2]FTBFS is easy: just need to remove the test. But I
would need access to a machine to actually investigate and correct the
bug.

More specifically, the tests intentionally triggers a SIGFPE that
Yorick should, but doesn't, catch.

Best regards, Thibaut.

[1] http://packages.debian.org/sid/yorick
[2] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=559406

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: Bug#559628: FTBFS [hppa]: could not run sample program

2009-12-07T17:00:59Z

hi,

i'm rebuilding erlang on my hppa box, just in case anything (like
NPTL transition) has changed/improved the erl interpreter behaviour in
the meanwhile. stay tuned.

cheers,
Domenico

-----[ Domenico Andreoli, aka cavok
--[ http://www.dandreoli.com/gpgkey.asc
---[ 3A0F 2F80 F79C 678A 8936 4FEE 0677 9033 A20E BC50

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: Bug#559638: FTBFS [hppa]: unrecognized command line option "-m32"

2009-12-07T16:23:11Z

Hi,

-m32 is a machine dependent option, on x86 it is used to build 32bit
code whereas -m64 is for 64bit code. it is not available on hppa
because userland is 32bit only.

regards,
Domenico

-----[ Domenico Andreoli, aka cavok
--[ http://www.dandreoli.com/gpgkey.asc
---[ 3A0F 2F80 F79C 678A 8936 4FEE 0677 9033 A20E BC50

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...

Re: emacs23 build failure

2009-12-07T09:56:30Z

On Wed, Nov 25, 2009 at 5:24 PM, Carlos O'Donell
<carlos@...> wrote:

> On Tue, Nov 24, 2009 at 4:52 PM, Moritz Muehlenhoff <jmm@...> wrote:
>> Hi,
>> you wanted to be notified for hppa-specific failures. emacs23 fails
>> to build on hppa:
>>
>> Wrote /build/buildd/emacs23-23.1+1/lisp/mail/rfc2368.elc
>> Compiling /build/buildd/emacs23-23.1+1/lisp/mail/rfc822.el
>> Wrote /build/buildd/emacs23-23.1+1/lisp/mail/rfc822.elc
>> Compiling /build/buildd/emacs23-23.1+1/lisp/mail/rmail-spam-filter.el
>> Wrote /build/buildd/emacs23-23.1+1/lisp/mail/rmail-spam-filter.elc
>> Compiling /build/buildd/emacs23-23.1+1/lisp/mail/rmail.el
>> Wrote /build/buildd/emacs23-23.1+1/lisp/mail/rmail.elc
>> Compiling /build/buildd/emacs23-23.1+1/lisp/mail/rmailedit.el
>>
>> In toplevel form:
>> ../../../lisp/mail/rmailedit.el:28:1:Error: Doing vfork: Inappropriate ioctl for device
>> make[2]: *** [/build/buildd/emacs23-23.1+1/lisp/mail/rmailedit.elc] Error 1
>> make[2]: Leaving directory `/build/buildd/emacs23-23.1+1/debian/build-x/lisp'
>> make[1]: *** [lisp] Error 2
>> make[1]: Leaving directory `/build/buildd/emacs23-23.1+1/debian/build-x'
>> make: *** [debian/stamp/build] Error 2
>> dpkg-buildpackage: error: debian/rules build gave error exit status 2
>> ******************************************************************************
>> Build finished at 20091122-1723
>>
>> Logs at:
>> https://buildd.debian.org/build.php?arch=hppa&pkg=emacs23&ver=23.1%2B1-5
>
> Thanks! I'll add this to my list and look into the failure.

This is a linux kernel bug on hppa. I've been able to reduce a
testcase from the emacs23 source, and I've passed it on to our kernel
developers. We are looking into it.

This is bug 558985.

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=558905

Cheers,
Carlos.

--
To UNSUBSCRIBE, email to debian-hppa-REQUEST@...
with a subject of "unsubscribe". Trouble? Contact listmaster@...