Re: 7.1.3, IRIX 6.5 and gcc - Mailing list pgsql-hackers

From Peter Eisentraut
Subject Re: 7.1.3, IRIX 6.5 and gcc
Date
Msg-id Pine.LNX.4.30.0108281146390.699-100000@peter.localdomain
Whole thread Raw
In response to 7.1.3, IRIX 6.5 and gcc  (Bruno Mattarollo <bruno@web1.greenpeace.org>)
Responses Re: 7.1.3, IRIX 6.5 and gcc
List pgsql-hackers
Bruno Mattarollo writes:

> I am trying to get 7.1.3 to compile on IRIX 6.5.13 using GCC 3.0.1 ... I
> always end up having this error:
>
> gcc -Wall -Wmissing-prototypes -Wmissing-declarations
> -I../../../../src/include -U_NO_XOPEN4 -c -o s_lock.o s_lock.c s_lock.c:
> In function `s_lock': s_lock.c:134: warning: passing arg 1 of pointer to
> function discards qualifiers from pointer target type s_lock.c: At top
> level: s_lock.c:234: warning: `tas_dummy' defined but not used as: Error:
> /var/tmp/ccLTwXmB.s, line 403: undefined assembler operation:  .global
>       .global tas gmake[4]: *** [s_lock.o] Error 1 gmake[4]: Leaving
> directory `/usr/local/src/postgresql-7.1.3/src/backend/storage/buffer'
> gmake[3]: *** [buffer-recursive] Error 2 gmake[3]: Leaving directory
> `/usr/local/src/postgresql-7.1.3/src/backend/storage'
> gmake[2]: *** [storage-recursive] Error 2

Go into the file src/backend/storage/s_lock.c and change the line that
looks like

#if defined(__mips__)

to

#if defined(__mips__) && !defined(__sgi)

(This "mips" assembly code is there for the likes of NetBSD and Linux.)

This should allow for the compilation to continue.

> I read in the archives that GCC was not very good to compile on IRIX, at
> this stage I have no choice, we don't have MIPSPro licensed.

"not very good" can also be read as "won't work", unless GCC 3 now does
better.

-- 
Peter Eisentraut   peter_e@gmx.net   http://funkturm.homeip.net/~peter



pgsql-hackers by date:

Previous
From: Joerg Hessdoerfer
Date:
Subject: Bytea/Base64 encoders for libpq - interested?
Next
From: Karel Zak
Date:
Subject: Re: Bytea/Base64 encoders for libpq - interested?