------- Comment #3 from rob1weld at aol dot com 2009-01-18 12:52 -------
(In reply to comment #1)
> for the pass-stratcliff.c failure, change:
> #ifndef __FreeBSD__
> to
> #ifdef __gnu_linux__
>
> And this should fix the issue there.
>
> For pass47-frag.c failure, you need to either do a { dg-warning "__ctype" "" {
> target *-*-solaris* } or prune the output of the warning.
>
I will add those to my next run.
I could not decide where or how to fix the Testsuite with a trivial
patch (that addressed a single issue) to provide the greatest benefit
so for myself I tried something like this:
# svn diff libmudflap/testsuite/mfconfig.exp.in
Index: libmudflap/testsuite/mfconfig.exp.in
===================================================================
--- libmudflap/testsuite/mfconfig.exp.in (revision 143454)
+++ libmudflap/testsuite/mfconfig.exp.in (working copy)
@@ -1,4 +1,13 @@
+# We need libsocket to link successfully on OpenSolaris, set
default-extra-flags
+proc solaris_libmudflap_need_lsocket {} {
+ if { [istarget "*-*-solaris*"] } {
+ lappend uselsocket "-lsocket"
+ }
+ return $uselsocket
+}
+
global mfconfig_libs
-set mfconfig_libs "@LIBS@"
+set uselsocket [solaris_libmudflap_need_lsocket]
+set mfconfig_libs "@LIBS@ $uselsocket "
global libmudflapth
set libmudflapth "@build_libmudflapth@"
The code was derived by reading
"gcc_trunk/libjava/testsuite/libjava.jni/jni.exp"
(line 263). The file contains no copyright info whatsoever but it is yours.
That gives us this result:
=== libmudflap Summary ===
# of expected passes 1107
# of unexpected failures 317
Full results are here:
Results for 4.4.0 20090117 (experimental) [trunk revision 143454] (GCC)
testsuite on i386-pc-solaris2.11
http://gcc.gnu.org/ml/gcc-testresults/2009-01/msg01790.html
Rob
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38738
