[Bug ada/38400] New: Acats faillures due to undefined pthread_setrunon_np

[Bug ada/38400] New: Acats faillures due to undefined pthread_setrunon_np

[J dot J dot vanderHeijden at gmail dot com]

A whole lot of Acats tests fail for IRIX 6.2 (see
http://gcc.gnu.org/ml/gcc-testresults/2008-10/msg00710.html)

This is because the Ada runtime uses the non-portable IRIX6.5-only function
pthread_setrunon_np, as evident from this snippet of acats.log:

splitting
/build/gcc/gcc-4.3.2/obj-quick/gcc/testsuite/ada/acats/tests/a/a83a02b.ada
into:
   a83a02b.adb
BUILD a83a02b.adb
gnatmake --GCC="/build/gcc/gcc-4.3.2/obj-quick/gcc/xgcc
-B/build/gcc/gcc-4.3.2/obj-quick/gcc/" -gnatws -O2
-I/build/gcc/gcc-4.3.2/obj-quick/gcc/testsuite/ada/acats/support a83a02b.adb
-largs --GCC="/build/gcc/gcc-4.3.2/obj-quick/gcc/xgcc
-B/build/gcc/gcc-4.3.2/obj-quick/gcc/"
/build/gcc/gcc-4.3.2/obj-quick/gcc/xgcc -c
-B/build/gcc/gcc-4.3.2/obj-quick/gcc/ -gnatws -O2
-I/build/gcc/gcc-4.3.2/obj-quick/gcc/testsuite/ada/acats/support a83a02b.adb
gnatbind -I/build/gcc/gcc-4.3.2/obj-quick/gcc/testsuite/ada/acats/support -x
a83a02b.ali
gnatlink a83a02b.ali --GCC=/build/gcc/gcc-4.3.2/obj-quick/gcc/xgcc
-B/build/gcc/gcc-4.3.2/obj-quick/gcc/
ld32: WARNING 84: /build/gcc/gcc-4.3.2/obj-quick/gcc/ada/rts/libgnarl.a is not
used for resolving any symbol.
ld32: ERROR 33: Unresolved data symbol "pthread_setrunon_np" -- 1st referenced
by /build/gcc/gcc-4.3.2/obj-quick/gcc/ada/rts/s-taprop.o.
        Use linker option -v to see when and which objects, archives and dsos
are loaded.  
ld32: INFO 152: Output file removed because of error.
collect2: ld returned 2 exit status
gnatlink: error when calling /build/gcc/gcc-4.3.2/obj-quick/gcc/xgcc
gnatmake: *** link failed.
FAIL:   a83a02b

It would be nice to restrict the usage of pthread_setrunon_np to IRIX 6.5


-- 
           Summary: Acats faillures due to undefined pthread_setrunon_np
           Product: gcc
           Version: 4.3.2
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: ada
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: J dot J dot vanderHeijden at gmail dot com
 GCC build triplet: mips-sgi-irix6.2
  GCC host triplet: mips-sgi-irix6.2
GCC target triplet: mips-sgi-irix6.2


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38400

[Bug ada/38400] Acats faillures due to undefined pthread_setrunon_np

[laurent at guerby dot net]

------- Comment #1 from laurent at guerby dot net  2008-12-10 11:54 -------
The missing function is called only in one place, visibly to set thread
affinity:

gcc/ada/s-taprop-irix.adb

      if Self_ID.Common.Task_Info /= null
        and then Self_ID.Common.Task_Info.Scope = PTHREAD_SCOPE_SYSTEM
        and then Self_ID.Common.Task_Info.Runon_CPU /= ANY_CPU
      then
         Result := pthread_setrunon_np
           (To_Int (Self_ID.Common.Task_Info.Runon_CPU));
         pragma Assert (Result = 0);
      end if;

The simplest thing to do would be to add #ifdef to define this function in
gcc/ada/adaint.c as dummy returning zero when IRIX. I don't have access to an
IRIX system so I don't know what #ifdef are available (or may be configure
test):

Could you try to add in gcc/ada/adaint.c the following definition:

int pthread_setrunon_np (int cpu) { return 0; }

Then see if it fixes the ACATS failures, then try to find out #ifdef that
trigger only on IRIX 6.2.

Thanks in advance,


-- 

laurent at guerby dot net changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |laurent at guerby dot net
           Keywords|                            |link-failure


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38400

[Bug ada/38400] Acats faillures due to undefined pthread_setrunon_np

[J dot J dot vanderHeijden at gmail dot com]

------- Comment #2 from J dot J dot vanderHeijden at gmail dot com  2008-12-11 11:07 -------
Subject: Re:  Acats faillures due to undefined pthread_setrunon_np

On Wed, Dec 10, 2008 at 12:54 PM, laurent at guerby dot net
<gcc-bugzilla@gcc.gnu.org> wrote:
>
>
> Could you try to add in gcc/ada/adaint.c the following definition:
>
> int pthread_setrunon_np (int cpu) { return 0; }
>
> Then see if it fixes the ACATS failures, then try to find out #ifdef that
> trigger only on IRIX 6.2.
>
> Thanks in advance,

Sounds like a plan.

The offending piece of code is pretty useless btw, even on IRIX 6.5.
The cpu# in question is always initialized in s-tasinf-irix.ads with
CPU_ANY so the end result is the same as when you wouldn't call
pthread_setrunon_np() at all. So my initial fix was to simply remove
this functionality entirely.

I'll have to dig up an old SGI running IRIX 6.2, yet fast enough to
finish the regression test. Usually I run 6.2 regression tests inside
a sandbox on a much faster SGI running IRIX 6.5, but this sandbox is
limited: it always fails anything linked to the pthreads library.

Jan-Jaap


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38400

Re: [Bug ada/38400] Acats faillures due to undefined pthread_setrunon_np

[Arnaud Charlet]

> The offending piece of code is pretty useless btw, even on IRIX 6.5.
> The cpu# in question is always initialized in s-tasinf-irix.ads with
> CPU_ANY so the end result is the same as when you wouldn't call
> pthread_setrunon_np() at all. So my initial fix was to simply remove
> this functionality entirely.

That's a misunderstanding of the code: Task_Info can be specified by the user
via pragma Task_Info.
What you're seeing in s-tasinf-irix.ads is just a default that the user
can change.

Arno

[Bug ada/38400] Acats faillures due to undefined pthread_setrunon_np

[charlet at adacore dot com]

------- Comment #3 from charlet at adacore dot com  2008-12-11 11:15 -------
Subject: Re:  Acats faillures due to undefined
        pthread_setrunon_np

> The offending piece of code is pretty useless btw, even on IRIX 6.5.
> The cpu# in question is always initialized in s-tasinf-irix.ads with
> CPU_ANY so the end result is the same as when you wouldn't call
> pthread_setrunon_np() at all. So my initial fix was to simply remove
> this functionality entirely.

That's a misunderstanding of the code: Task_Info can be specified by the user
via pragma Task_Info.
What you're seeing in s-tasinf-irix.ads is just a default that the user
can change.

Arno


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38400

[Bug ada/38400] Acats faillures due to undefined pthread_setrunon_np

[J dot J dot vanderHeijden at gmail dot com]

------- Comment #4 from J dot J dot vanderHeijden at gmail dot com  2008-12-15 16:41 -------
Created an attachment (id=16911)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16911&action=view)
IRIX-6.2 only dummy implementation of pthread_setrunon_np()

This patch adds an empty implementation of pthread_setrunon_np() on IRIX 6.2.

Since Ada threading seems to operate independent of GCC's
'enable-threads=posix' I've chosen the existence of SGI_INITIAL_PT_SPROC as a
trigger to #include <pthreads.h> and thus prevent breakage on IRIX 5.x (which
doesn't have pthreads at all).

Testsuite results follow


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38400

[Bug ada/38400] Acats faillures due to undefined pthread_setrunon_np

[J dot J dot vanderHeijden at gmail dot com]

------- Comment #5 from J dot J dot vanderHeijden at gmail dot com  2008-12-15 16:48 -------
Test results: http://gcc.gnu.org/ml/gcc-testresults/2008-12/msg01459.html

Please approve.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38400

[Bug ada/38400] Acats faillures due to undefined pthread_setrunon_np

[ro at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38400

Rainer Orth <ro at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
                 CC|                            |ro at gcc dot gnu.org
         Resolution|                            |WONTFIX

--- Comment #6 from Rainer Orth <ro at gcc dot gnu.org> 2011-07-19 14:48:46 UTC ---
Since IRIX < 6.5 has been deprecated since GCC 4.5 and I don't have a test
system any longer, I won't fix this.