[Bug bootstrap/39019] New: Solaris and IRIX libelf cause trouble for build

[Bug bootstrap/39019] New: Solaris and IRIX libelf cause trouble for build

[ro at gcc dot gnu dot org]

Despite they seem to have the same interface, the libelf bundled with Solaris
(at
least as far back as Solaris 8) cause trouble:

* The Solaris libelf.h isn't largefile aware:

#if defined(_ILP32) && (_FILE_OFFSET_BITS != 32)
#error "large files are not supported by libelf"
#endif

  This is explained in detail in

 
http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/cmd/sgs/libelf/common/README.LFS

* IRIX 6.5 has libelf.h, too, but lacks gelf.h.

* When I use libelf 0.8.10 instead (installed into /vol/gcc/{include, lib}),
  I have to be very careful: using CPPFLAGS=-I/vol/gcc/include doesn't work
  since libelf.h is preferred and thus the copy in /usr/include is used.  Using
  -I/vol/gcc/include/libelf doesn't work either since gelf.h includes
  libelf/libelf.h which isn't found than.  One needs to use
  -I/vol/gcc/include/libelf -I/vol/gcc/include, which is intricate to get
right.
  I think the use of libelf should be handled similarly to gmp/mpfr via
  --with-libelf or some such.


-- 
           Summary: Solaris and IRIX libelf cause trouble for build
           Product: gcc
           Version: lto
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: bootstrap
        AssignedTo: dnovillo at gcc dot gnu dot org
        ReportedBy: ro at gcc dot gnu dot org
 GCC build triplet: sparc-sun-solaris2.11
  GCC host triplet: sparc-sun-solaris2.11
GCC target triplet: sparc-sun-solaris2.11


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39019

[Bug bootstrap/39019] Solaris and IRIX libelf cause trouble for build

[rob1weld at aol dot com]

------- Comment #1 from rob1weld at aol dot com  2009-02-23 01:21 -------
Confirmed on i386-pc-solaris2.11 (OpenSolaris 2009.06).

We are advised to use libelf v0.8.10, which (by default) installs in
/usr/local but the lto configury uses /usr/include/libelf.h .

+1 for "--with-libelf=/usr/local" .


# gcc/xgcc -v
Using built-in specs.
COLLECT_GCC=gcc/xgcc
Target: i386-pc-solaris2.11
Configured with: ../lto_trunk/configure --prefix=/usr/local/lto
--enable-languages=lto,c++ --enable-shared --disable-static --enable-multilib
--with-gnu-as --with-as=/usr/local/bin/as --with-gnu-ld
--with-ld=/usr/local/bin/ld --with-gmp=/usr/local --with-mpfr=/usr/local
Thread model: posix
gcc version 4.4.0 20090218 (experimental) [lto revision 144375] (lto merged
with rev 144262) 

Rob


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39019

[Bug bootstrap/39019] Solaris and IRIX libelf cause trouble for build

[rob1weld at aol dot com]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1408 bytes --]



------- Comment #2 from rob1weld at aol dot com  2009-02-23 03:51 -------
Rainer, while we wait for the patch did you wish to try this:


1. Edit the gcc/configure (Line 9562) and reverse the detection order:

- for ac_header in gelf.h libelf/gelf.h
+ for ac_header in libelf/gelf.h gelf.h


2. Edit lto_plugin/lto_plugin.c and lto_plugin/lto-symtab.c and
both add in ../gcc/auto-host.h and change the define testing order:


#undef PACKAGE_NAME
#undef PACKAGE_STRING
#undef PACKAGE_TARNAME
#undef PACKAGE_VERSION
#include "../gcc/auto-host.h"

#ifdef    __COMMENT__
#ifdef HAVE_GELF_H
# include <gelf.h>
#else
# if defined(HAVE_LIBELF_GELF_H)
#   include <libelf/gelf.h>
# else
#  error "gelf.h not available"
# endif
#endif
#endif /* __COMMENT__ */

#if defined(HAVE_LIBELF_GELF_H)
#   include <libelf/gelf.h>
#else
# ifdef HAVE_GELF_H
#   include <gelf.h>
# else
#  error "gelf.h not available"
# endif
#endif


You can hand-compile to fix the 'Werror" that will remain:

cc1: warnings being treated as errors
../../lto_trunk/lto-plugin/lto-plugin.c: In function ‘onload’:
../../lto_trunk/lto-plugin/lto-plugin.c:670: error: implicit declaration of
function ‘mkdtemp’
../../lto_trunk/lto-plugin/lto-plugin.c:670: error: assignment makes pointer
from integer without a cast
gmake[3]: *** [lto-plugin.lo] Error 1


That fixes it for me.

Rob


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39019

[Bug bootstrap/39019] Solaris and IRIX libelf cause trouble for build

[rob1weld at aol dot com]

------- Comment #3 from rob1weld at aol dot com  2009-02-23 14:52 -------
Dismal Testsuite results are here:
http://gcc.gnu.org/ml/gcc-testresults/2009-02/msg02284.html

Rob


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39019

[Bug bootstrap/39019] Solaris and IRIX libelf cause trouble for build

[rob1weld at aol dot com]

------- Comment #4 from rob1weld at aol dot com  2009-03-10 04:22 -------
(In reply to comment #3)
> Dismal Testsuite results are here:
> http://gcc.gnu.org/ml/gcc-testresults/2009-02/msg02284.html
> Rob

Great results are here:
http://gcc.gnu.org/ml/gcc-testresults/2009-02/msg02375.html


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39019

[Bug bootstrap/39019] Solaris and IRIX libelf cause trouble for build

[bje at gcc dot gnu dot org]

------- Comment #5 from bje at gcc dot gnu dot org  2009-07-09 05:09 -------
Building with --with-libelf is the right approach.  I don't think it works 100%
correctly, though, so I will take this bug and investigate.


-- 

bje at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|dnovillo at gcc dot gnu dot |bje at gcc dot gnu dot org
                   |org                         |
             Status|UNCONFIRMED                 |ASSIGNED
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2009-07-09 05:09:13
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39019

[Bug bootstrap/39019] Solaris and IRIX libelf cause trouble for build

[bje at gcc dot gnu dot org]

------- Comment #6 from bje at gcc dot gnu dot org  2009-07-15 03:29 -------
Using --with-libelf should work (I just checked it on several systems).  Please
re-open this report if your problems persist.


-- 

bje at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |WORKSFORME


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39019