Problem compiling 4.2.4 on Solaris 10 with Sun Studio 12

[jcenter]

Hi,

I was unable to compile GMP 4.2.4 on Solaris 10 with Sun Studio 12.  The
errors were:

"ismpf.cc", line 53: Error: Could not find a match for
std::use_facet<std::Facet
 >(const std::locale) needed in operator>>(std::istream &, __mpf_struct*).
"ismpf.cc", line 53: Error: Could not find a match for
std::use_facet<std::Facet
 >(const std::locale) needed in operator>>(std::istream &, __mpf_struct*).
"ismpf.cc", line 65: Error: Could not find a match for
std::use_facet<std::Facet
 >(const std::locale) needed in operator>>(std::istream &, __mpf_struct*).
3 Error(s) and 11 Warning(s) detected.
gmake[2]: *** [ismpf.lo] Error 1

I found a patch on gmp-bugs for GMP 4.2.1 for Sun Studio 11 from July
2006 by Matthias Koeppe
(http://gmplib.org/list-archives/gmp-bugs/2006-July/000519.html) that
worked for me.  Could this, or one similar to it, be added to the patch
list on the home page & incorporated in the next release?

Thanks.

        -John

--
John Center
Villanova University
_______________________________________________
gmp-bugs mailing list
gmp-bugs@...
http://gmplib.org/mailman/listinfo/gmp-bugs

[Marc Glisse-3]

On Tue, 17 Feb 2009, John Center wrote:

> I was unable to compile GMP 4.2.4 on Solaris 10 with Sun Studio 12.  The
> errors were:
>
> "ismpf.cc", line 53: Error: Could not find a match for
> std::use_facet<std::Facet
>> (const std::locale) needed in operator>>(std::istream &, __mpf_struct*).
>
> I found a patch on gmp-bugs for GMP 4.2.1 for Sun Studio 11 from July 2006 by
> Matthias Koeppe
> (http://gmplib.org/list-archives/gmp-bugs/2006-July/000519.html) that worked
> for me.

Good that you found it.

> Could this, or one similar to it, be added to the patch list on the
> home page & incorporated in the next release?

A FAQ about typical problems with solaris+gmp would help.

But do you really still need to use the old broken version of the standard
library? If you have a choice, you should really use -library=stlport4, or
some other self-compiled standard library like Apache stdcxx (used by
default in opensolaris and soon shipped with the compiler). Both should
work fine with gmp without any patch.

I somehow doubt TG will be happy to include such a workaround, but I may
be wrong. Note that such a workaround should test that
_RWSTD_NO_TEMPLATE_ON_RETURN_TYPE is defined (hello MPIR...).

--
Marc Glisse
_______________________________________________
gmp-bugs mailing list
gmp-bugs@...
http://gmplib.org/mailman/listinfo/gmp-bugs

[jcenter]

Thanks, Marc.  I wasn't aware of the libCstd vs. libstlport issue.  I'll
back out of the patch, recompile & test.  -John

Marc Glisse wrote:

> On Tue, 17 Feb 2009, John Center wrote:
>
>> I was unable to compile GMP 4.2.4 on Solaris 10 with Sun Studio 12.  The
>> errors were:
>>
>> "ismpf.cc", line 53: Error: Could not find a match for
>> std::use_facet<std::Facet
>>> (const std::locale) needed in operator>>(std::istream &, __mpf_struct*).
>> I found a patch on gmp-bugs for GMP 4.2.1 for Sun Studio 11 from July 2006 by
>> Matthias Koeppe
>> (http://gmplib.org/list-archives/gmp-bugs/2006-July/000519.html) that worked
>> for me.
>
> Good that you found it.
>
>> Could this, or one similar to it, be added to the patch list on the
>> home page & incorporated in the next release?
>
> A FAQ about typical problems with solaris+gmp would help.
>
> But do you really still need to use the old broken version of the standard
> library? If you have a choice, you should really use -library=stlport4, or
> some other self-compiled standard library like Apache stdcxx (used by
> default in opensolaris and soon shipped with the compiler). Both should
> work fine with gmp without any patch.
>
> I somehow doubt TG will be happy to include such a workaround, but I may
> be wrong. Note that such a workaround should test that
> _RWSTD_NO_TEMPLATE_ON_RETURN_TYPE is defined (hello MPIR...).
>

_______________________________________________
gmp-bugs mailing list
gmp-bugs@...
http://gmplib.org/mailman/listinfo/gmp-bugs

[jcenter]

Hi,

To follow up, I recompiled with GMP 4.2.4 with -library=stlport4.  I
used the -enable-cxx option & the compile failed while building
osdoprnti.cc.  The error was:

"../gmp.h", line 613: Error: va_list is not defined.
...
"../gmp-impl.h", line 3258: Error: va_list is not defined.
...

The problem was cstdarg was included instead of stdarg.h.

Here is a patch to fix this:

*** gmp-4.2.4-ORIG/cxx/osdoprnti.cc     Sat Sep  1 06:09:03 2007
--- gmp-4.2.4/cxx/osdoprnti.cc  Thu Feb 19 22:50:36 2009
***************
*** 22,28 ****
   along with the GNU MP Library.  If not, see
http://www.gnu.org/licenses/.  */

   #include <iostream>
! #include <cstdarg>    /* for va_list and hence doprnt_funs_t */
   #include <cstring>    /* for strlen */

   #include "gmp.h"
--- 22,28 ----
   along with the GNU MP Library.  If not, see
http://www.gnu.org/licenses/.  */

   #include <iostream>
! #include <stdarg.h>    /* for va_list and hence doprnt_funs_t */
   #include <cstring>    /* for strlen */

   #include "gmp.h"


On a different issue, when I ran "make check" one test failed, t-locale,
& dumped core.

Here's the output from dbx:

# dbx .libs/t-locale /var/core/core_ra_t-locale_0_0_1235103740_10227
Reading t-locale
core file header read successfully
Reading ld.so.1
Reading libc.so.1
Reading libgmp.so.3.4.4
Reading libumem.so.1
Reading libstlport.so.1
Reading libCrun.so.1
Reading libm.so.1
Reading librt.so.1
Reading libaio.so.1
Reading libmd.so.1
Reading libc_psr.so.1
Reading libm.so.2
Reading libgmpxx.so.4.0.4
program terminated by signal SEGV (no mapping at the fault address)
0xffffffff7eb771f8: imbue+0x0078:       ldx      [%o0], %i5
Current function is check_output (optimized)
   141         ostringstream  got;
(dbx) where
   [1] std::basic_ios<char,std::char_traits<char>
 >::imbue(0xffffffff7ffff018, 0xffffffff7ffff3a0, 0xffffffff7ffff010,
0x5d8, 0x400, 0x400), at 0xffffffff7eb771f8
   [2] std::basic_ios<char,std::char_traits<char>
 >::init(0xffffffff7ffff3a0, 0x0, 0x158, 0xffffffff7ed194c8,
0xffffffff7ed194f0, 0xffffffff7ffff248), at 0xffffffff7eb76fc8
   [3]
std::basic_ostringstream<char,std::char_traits<char>,std::allocator<char>
 >::basic_ostringstream(0xffffffff7ffff248, 0x10, 0x6f0,
0xffffffff7ffff3a0, 0xffffffff7ed19518, 0xffffffff7ed10300), at
0xffffffff7eb796d0
=>[4] check_output() (optimized), at 0x10000658c (line ~141) in
"t-locale.cc"
   [5] main() (optimized), at 0x100006ed0 (line ~184) in "t-locale.cc"

Here's the flags it was compiled & linked with:

cc -xc99=%all -DHAVE_CONFIG_H -I. -I. -I../.. -I../.. -I../../tests
-I/opt/gnu/include -I/opt/local/include -I/usr/sfw/include
-I/usr/include  -g -xs -fast -fsimple=1 -fns=no -xtarget=ultraT1
-xarch=sparcvis2 -m64 -mt -xipo -xcode=pic32 -xpagesize=default
-xlic_lib=sunperf  -c t-locale.c

cc -xc99=%all -g -xs -fast -fsimple=1 -fns=no -xtarget=ultraT1
-xarch=sparcvis2 -m64 -mt -xipo -xcode=pic32 -xpagesize=default
-xlic_lib=sunperf -Wl,-ztext -Wl,-Bdirect -Wl,-zlazyload -Wl,-zcombreloc
-Wl,-zignore -o .libs/t-locale t-locale.o  -L/opt/gnu/lib/sparcv9
-L/opt/local/lib/sparcv9 -L/usr/sfw/lib/sparcv9 -L/usr/lib/sparcv9
../../tests/.libs/libtests.a /opt/ws/dists/gmp-4.2.4/.libs/libgmp.so
../../.libs/libgmp.so -lumem -lpthread  -R/opt/gnu/lib/sparcv9
-R/opt/local/lib/sparcv9 -R/usr/sfw/lib/sparcv9 -R/usr/lib/sparcv9

Could someone please tell me what might be wrong?

Thanks.

        -John


Marc Glisse wrote:

> On Tue, 17 Feb 2009, John Center wrote:
>
>> I was unable to compile GMP 4.2.4 on Solaris 10 with Sun Studio 12.  The
>> errors were:
>>
>> "ismpf.cc", line 53: Error: Could not find a match for
>> std::use_facet<std::Facet
>>> (const std::locale) needed in operator>>(std::istream &, __mpf_struct*).
>> I found a patch on gmp-bugs for GMP 4.2.1 for Sun Studio 11 from July 2006 by
>> Matthias Koeppe
>> (http://gmplib.org/list-archives/gmp-bugs/2006-July/000519.html) that worked
>> for me.
>
> Good that you found it.
>
>> Could this, or one similar to it, be added to the patch list on the
>> home page & incorporated in the next release?
>
> A FAQ about typical problems with solaris+gmp would help.
>
> But do you really still need to use the old broken version of the standard
> library? If you have a choice, you should really use -library=stlport4, or
> some other self-compiled standard library like Apache stdcxx (used by
> default in opensolaris and soon shipped with the compiler). Both should
> work fine with gmp without any patch.
>
> I somehow doubt TG will be happy to include such a workaround, but I may
> be wrong. Note that such a workaround should test that
> _RWSTD_NO_TEMPLATE_ON_RETURN_TYPE is defined (hello MPIR...).
>

_______________________________________________
gmp-bugs mailing list
gmp-bugs@...
http://gmplib.org/mailman/listinfo/gmp-bugs

[Marc Glisse-3]

On Fri, 20 Feb 2009, John Center wrote:

> To follow up, I recompiled with GMP 4.2.4 with -library=stlport4. I used
> the -enable-cxx option & the compile failed while building osdoprnti.cc.

Oh yes, I had forgotten to warn you about this gmp bug.

> On a different issue, when I ran "make check" one test failed, t-locale, &
> dumped core.
>
> Here's the output from dbx:
>
> # dbx .libs/t-locale /var/core/core_ra_t-locale_0_0_1235103740_10227
> Reading t-locale
> core file header read successfully
> Reading ld.so.1
> Reading libc.so.1
> Reading libgmp.so.3.4.4
> Reading libumem.so.1
> Reading libstlport.so.1
> Reading libCrun.so.1
> Reading libm.so.1
> Reading librt.so.1
> Reading libaio.so.1
> Reading libmd.so.1
> Reading libc_psr.so.1
> Reading libm.so.2
> Reading libgmpxx.so.4.0.4
> program terminated by signal SEGV (no mapping at the fault address)
> 0xffffffff7eb771f8: imbue+0x0078:       ldx      [%o0], %i5
> Current function is check_output (optimized)
>  141         ostringstream  got;
> (dbx) where
>  [1] std::basic_ios<char,std::char_traits<char> >::imbue(0xffffffff7ffff018,
> 0xffffffff7ffff3a0, 0xffffffff7ffff010, 0x5d8, 0x400, 0x400), at
> 0xffffffff7eb771f8
>  [2] std::basic_ios<char,std::char_traits<char> >::init(0xffffffff7ffff3a0,
> 0x0, 0x158, 0xffffffff7ed194c8, 0xffffffff7ed194f0, 0xffffffff7ffff248), at
> 0xffffffff7eb76fc8
>  [3]
> std::basic_ostringstream<char,std::char_traits<char>,std::allocator<char>
>> ::basic_ostringstream(0xffffffff7ffff248, 0x10, 0x6f0, 0xffffffff7ffff3a0,
> 0xffffffff7ed19518, 0xffffffff7ed10300), at 0xffffffff7eb796d0
> =>[4] check_output() (optimized), at 0x10000658c (line ~141) in "t-locale.cc"
>  [5] main() (optimized), at 0x100006ed0 (line ~184) in "t-locale.cc"
>
> Here's the flags it was compiled & linked with:
>
> cc -xc99=%all -DHAVE_CONFIG_H -I. -I. -I../.. -I../.. -I../../tests
> -I/opt/gnu/include -I/opt/local/include -I/usr/sfw/include -I/usr/include  -g
> -xs -fast -fsimple=1 -fns=no -xtarget=ultraT1 -xarch=sparcvis2 -m64 -mt -xipo
> -xcode=pic32 -xpagesize=default -xlic_lib=sunperf  -c t-locale.c
>
> cc -xc99=%all -g -xs -fast -fsimple=1 -fns=no -xtarget=ultraT1
> -xarch=sparcvis2 -m64 -mt -xipo -xcode=pic32 -xpagesize=default
> -xlic_lib=sunperf -Wl,-ztext -Wl,-Bdirect -Wl,-zlazyload -Wl,-zcombreloc
> -Wl,-zignore -o .libs/t-locale t-locale.o  -L/opt/gnu/lib/sparcv9
> -L/opt/local/lib/sparcv9 -L/usr/sfw/lib/sparcv9 -L/usr/lib/sparcv9
> ../../tests/.libs/libtests.a /opt/ws/dists/gmp-4.2.4/.libs/libgmp.so
> ../../.libs/libgmp.so -lumem -lpthread  -R/opt/gnu/lib/sparcv9
> -R/opt/local/lib/sparcv9 -R/usr/sfw/lib/sparcv9 -R/usr/lib/sparcv9

These are the flags for misc/t-locale.c, not cxx/t-locale.cc.

> Could someone please tell me what might be wrong?

I don't know. Well I do know that stlport4 only supports the C locale, but
that should not cause this segfault. Also, you are using -fast without
-fsimple=0, which could cause problems, but usually not just there.

You could try running the test in dbx with check -all to get more
information, but I fear someone more knowledgeable than me will have to
help. You can try -library=stlport4_dbg, play with the various options
used (try with the default flags for instance), check you have the latest
patches for your compiler, etc.

Not related: I see you are compiling for a T1. You may want to replace
-xarch=sparcvis2 by -xarch=sparc there (I am not sure whether gmp disables
the use of floating point for the T1). Also, -xlic_lib=sunperf is a bit
useless for gmp.

--
Marc Glisse
_______________________________________________
gmp-bugs mailing list
gmp-bugs@...
http://gmplib.org/mailman/listinfo/gmp-bugs

[jcenter]

Hi Marc,


Marc Glisse wrote:
> On Fri, 20 Feb 2009, John Center wrote:
>
>> To follow up, I recompiled with GMP 4.2.4 with -library=stlport4. I used
>> the -enable-cxx option & the compile failed while building osdoprnti.cc.
>
> Oh yes, I had forgotten to warn you about this gmp bug.
>
That's ok about the gmp bug. I missed it in my search on gmp bugs & the
"-DHAVE_CONFIG_H" flag threw me off for a while.  Since it's included
with the cc & c++ compiles, I assumed it was actually being used.  Silly
me...  ;-)


> These are the flags for misc/t-locale.c, not cxx/t-locale.cc.
>
Sorry about that, I missed that.  Here they are:

  CC -DHAVE_CONFIG_H -I. -I. -I.. -D__GMP_WITHIN_GMPXX -I..
-I/opt/gnu/include -I/opt/local/include -I/usr/sfw/include
-I/usr/include -D__GMP_NO_ATTRIBUTE_CONST_PURE -g0 -xs -fast -fsimple=1
-fns=no -xtarget=ultraT1 -xarch=sparcvis2 -m64 -mt -xipo -xcode=pic32
-xpagesize=default -features=extensions -library=stlport4,sunperf

-R/opt/gnu/lib/sparcv9 -R/opt/local/lib/sparcv9 -R/usr/sfw/lib/sparcv9
-R/usr/lib/sparcv9 -L/opt/gnu/lib/sparcv9 -L/opt/local/lib/sparcv9 -L
/usr/sfw/lib/sparcv9 -L/usr/lib/sparcv9 -lumem -lpthread -lc  -ztext
-Bdirect -zlazyload -zcombreloc -zignore


>> Could someone please tell me what might be wrong?
>
> I don't know. Well I do know that stlport4 only supports the C locale, but
> that should not cause this segfault. Also, you are using -fast without
> -fsimple=0, which could cause problems, but usually not just there.
>

> You could try running the test in dbx with check -all to get more
> information, but I fear someone more knowledgeable than me will have to
> help. You can try -library=stlport4_dbg, play with the various options
> used (try with the default flags for instance), check you have the latest
> patches for your compiler, etc.
>
> Not related: I see you are compiling for a T1. You may want to replace
> -xarch=sparcvis2 by -xarch=sparc there (I am not sure whether gmp disables
> the use of floating point for the T1). Also, -xlic_lib=sunperf is a bit
> useless for gmp.
>

As you may be able to tell, I've been trying to come up with a common
set of flags to build apps & libs for the ultraT1  platform.  So, this
is for a T1000/T2000, not a Netra T1.  I've had to tweak these a bit,
but I've gotten good results so far.  But, nothing's perfect..  :-)

Thanks for your help.

        -John

--
John Center
Villanova University
_______________________________________________
gmp-bugs mailing list
gmp-bugs@...
http://gmplib.org/mailman/listinfo/gmp-bugs