public inbox for gcc@gcc.gnu.org
 help / color / mirror / Atom feed
From: Kenneth Lareau <elessar@numenor.org>
To: law@cygnus.com
Cc: egcs@cygnus.com
Subject: Re: egcs 1.1.1 on IRIX 6.5 - mips3 and R8k/R10k systems
Date: Sun, 28 Feb 1999 22:53:00 -0000	[thread overview]
Message-ID: <199902052247.RAA08497@minddrive.numenor.org> (raw)
Message-ID: <19990228225300.T7NwppVIHdNe1BGrnb1PQPxOc9cUwSjGDs5DOgMhbE4@z> (raw)
In-Reply-To: <30935.918187658@upchuck>

In message < 30935.918187658@upchuck >, Jeffrey A Law writes:
>
>  In message < 199902041900.OAA04221@minddrive.numenor.org >you write:
>  > After several unsuccessful attempts with configure, and a lot of reading,
>  > I have been unable to compile egcs 1.1.1 on an R8k/R10k based SGI system
>  > running IRIX 6.5 to create executables and libraries compiled with the
>  > mips3 libraries to enable the ability to use them on the R4k/R5k based
>  > SGI system (by default it uses the mips4 libraries).
>  > 
>  > Is it possible to do this with a properly setup configure line?  Or is
>  > this something that is unachievable, except possibly through a cross-
>  > compile procedure?  The inability to make this work is hampering other
>  > efforts and tasks I need to complete, so any suggestions would be great-
>  > ly appreciated.
>It's hard to suggest anything to you since you don't give any
>specifics about your problems.

Okay, here's a bit more detail:

If I'm on a R8k/R10k based SGI system, and use the following configure
line:

  CC=cc ../configure --enable-shared --enable-haifa --with-stabs \
  --enable-threads --prefix=/common/software_y2k/egcs-1.1.1/sgi/6.5

and then run 'make bootstrap', when I do a 'file' command on the various
executables, I get the following:

  gcc: ELF N32 MSB mips-4 dynamic executable MIPS - version 1

This is bad, because older SGIs, like the Indy, can't use these exe-
cutables; they need the mips3 libraries.

So I added "CFLAGS='-mips3'" to the above configure line, which runs
fine.  During the run of the "make bootstrap", however, the very first
build of C binaries does _not_ include the '-mips3' option.  I used
the following make line:

  make CFLAGS='-O2 -mips3' LIBCFLAGS='-g -O2 -mips3' LIBCXXFLAGS=\
  '-g -O2 -mips3 -fno-implicit-templates' bootstrap

I edited the CFLAGS line in gcc/Makefile to show '-g -mips3', and
starting the build again, everything looks fine, until it gets to the
building of the libraries, then I get the following:

  ./xgcc -B./ -O2   -DIN_GCC    -g -mips3 -I./include   -g1  -DIN_LIBGCC2 
-D__GCC_FLOAT_NOT_NEEDED    -I. -I../../gcc -I../../gcc/config -c -DL${name} \
      ../../gcc/libgcc2.c -o ${name}.o; \
  if [ $? -eq 0 ] ; then true; else exit 1; fi; \
  ar rc tmplibgcc2.a ${name}.o; \
  rm -f ${name}.o; \
done
_muldi3
In file included from ../../gcc/config/mips/xm-iris6.h:16,
                 from tconfig.h:1,
                 from ../../gcc/libgcc2.c:33:
/usr/include/string.h:67: warning: conflicting types for built-in function 
`memcpy'
/usr/include/string.h:74: warning: conflicting types for built-in function 
`memcmp'
/usr/include/string.h:95: warning: conflicting types for built-in function 
`memset'
/usr/include/string.h:97: warning: conflicting types for built-in function 
`strlen'


This occurs for every entry in the preceding for loop (for name in _muldi3
_divdi3 _moddi3 _udivdi3 _umoddi3 ...), though it doesn't occur the second
time around when the '-mabi=64' option is used.

Finally, it starts the 2nd stage of the build, giving the above string.h
conflict messages until finally:

stage1/xgcc -Bstage1/  -DIN_GCC    -O2 -O2 -mips3  -DHAVE_CONFIG_H -DHAIFA  -o 
gencheck \
 gencheck.o ` case "obstack.o" in ?*) echo obstack.o ;; esac ` ` case "" in 
?*) echo  ;; esac ` ` case "" in ?*) echo  ;; esac ` ` case "" in ?*) echo  ;; 
esac ` ` case "" in ?*) echo  ;; esac `
./gencheck > tmp-check.h
make[2]: *** [s-check] Bus error (core dumped)


I don't know if this is enough information, but up to this point I have
been unable to get it to build mips-3 binaries and libraries.


Ken Lareau
elessar@numenor.org



  reply	other threads:[~1999-02-28 22:53 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
1999-02-04 11:00 Kenneth Lareau
     [not found] ` < 199902041900.OAA04221@minddrive.numenor.org >
1999-02-04 22:50   ` Jeffrey A Law
     [not found]     ` < 30935.918187658@upchuck >
1999-02-05 14:47       ` Kenneth Lareau [this message]
1999-02-28 22:53         ` Kenneth Lareau
1999-02-28 22:53     ` Jeffrey A Law
1999-02-28 22:53 ` Kenneth Lareau
1999-02-05 18:17 N8TM
1999-02-28 22:53 ` N8TM
1999-02-05 18:22 N8TM
     [not found] ` < e472f205.36bba6ae@aol.com >
1999-02-07  7:51   ` Kenneth Lareau
1999-02-28 22:53     ` Kenneth Lareau
1999-02-28 22:53 ` N8TM

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=199902052247.RAA08497@minddrive.numenor.org \
    --to=elessar@numenor.org \
    --cc=egcs@cygnus.com \
    --cc=law@cygnus.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).