patches to fix uw7 target

Robert Lipe robertl@sco.com
Mon Mar 22 23:53:00 GMT 1999


I blew completely too many hours staring at procfs.c and bfd
handling when all I had was a stupid configuration problem.
While 'i386-pc-sysv4.2MP' (yes, it has to be capital or you get
something else) generates a configuration that basically works,
'i386-pc-unixware2' (which was the product that defined SVR4.2MP/386)
generates a target that does NOT get the 4.2MP definitions even though
that was a cornerstone of Unixware2.  How Icky!  This naming is way too
fragile...

Tue Mar 23 01:47:08 1999  Robert Lipe  <robertlipe@usa.net>

	* configure.host, configure.tgt (i?86-*-sysv5*): Treat like SVR4.2MP.

--- configure.host_     Wed Mar 10 08:01:14 1999
+++ configure.host      Tue Mar 23 01:50:30 1999
@@ -71,6 +71,7 @@
 i[3456]86-*-sysv32*)   gdb_host=i386v32 ;;
 i[3456]86-*-sysv4.2MP) gdb_host=i386v42mp ;;
 i[3456]86-*-sysv4.2uw2*)       gdb_host=i386v42mp ;;
+i[3456]86-*-sysv5*)    gdb_host=i386v42mp ;;
 i[3456]86-*-sysv4*)    gdb_host=i386v4 ;;
 i[3456]86-*-unixware2*)        gdb_host=i386v42mp ;;
 i[3456]86-*-unixware*) gdb_host=i386v4 ;;
(robertl) rjludi:/home3/gdb/gdb-4.17.86/gdb
$ diff -u configure.tgt_ configure.tgt    
--- configure.tgt_      Fri Mar 12 09:37:53 1999
+++ configure.tgt       Tue Mar 23 01:49:06 1999
@@ -101,6 +101,7 @@
 i[3456]86-*-sysv4.2MP) gdb_target=i386v42mp ;;
 i[3456]86-*-sysv4.2uw2*)       gdb_target=i386v42mp ;;
 i[3456]86-*-sysv4*)    gdb_target=i386v4 ;;
+i[3456]86-*-sysv5*)    gdb_target=i386v42mp ;;
 i[3456]86-*-unixware2*)        gdb_target=i386v42mp ;;
 i[3456]86-*-unixware*) gdb_target=i386v4 ;;
 i[3456]86-*-sco*)      gdb_target=i386v ;;



Draft these in from EGCS:

Thu Oct 22 10:55:25 1998  Robert Lipe  <robertl@dgii.com>

        * config.guess: Match any version of Unixware7.
        * config.guess: Correctly identify Pentium II sco boxes.

--- config.guess_       Tue Mar 23 02:00:50 1999
+++ config.guess        Tue Mar 23 02:00:59 1999
@@ -722,13 +722,13 @@
                echo ${UNAME_MACHINE}-pc-sysv${UNAME_RELEASE}
        fi
        exit 0 ;;
-    i?86:*:5:7)
+    i?86:*:5:7*)
        UNAME_REL=`(/bin/uname -X|egrep Release|sed -e 's/.*= //')`
        (/bin/uname -X|egrep i80486 >/dev/null) && UNAME_MACHINE=i486
        (/bin/uname -X|egrep '^Machine.*Pentium' >/dev/null) \
                && UNAME_MACHINE=i586
 # 5.0.4c returns "Pent II".  5.0.5 returns PentII
-       (/bin/uname -X|egrep '^Machine.*Pent *II' >/dev/null) \
+       (/bin/uname -X|egrep '^Machine.*Pent.*II' >/dev/null) \
                && UNAME_MACHINE=i686
        (/bin/uname -X|egrep '^Machine.*Pentium Pro' >/dev/null) \
                && UNAME_MACHINE=i686
(robertl) rjludi:/home3/gdb/gdb-4.17.86




More information about the Gdb-patches mailing list