From mboxrd@z Thu Jan  1 00:00:00 1970
From: Mark Kettenis <kettenis@wins.uva.nl>
To: gdb-patches@sources.redhat.com
Subject: [RFA] Handle expected failure in interrupt.exp for Linux/x86
Date: Sun, 28 Oct 2001 07:54:00 -0000
Message-id: <200110281554.f9SFsFe11879@delius.kettenis.local>
X-SW-Source: 2001-10/msg00346.html

Ouch!  Letting GDB call a function from an interrupted restartable
system call doesn't work on Linux/x86.  This already was an expected
failure.  Unfortunately it only deals with a SIGSEGV wheras a SIGILL
is just as likely.

Fixing the actual problem isn't easy.  I'll try to come up with
something clever.  In the meantime, I'd like to check in this patch.

OK?

Mark


Index: ChangeLog
from  Mark Kettenis  <kettenis@gnu.org>

	* gdb.base/interrupt.exp: Treat SIGILL similar to SIGSEGV such
	that we catch the expected failure under Linux/x86.

Index: gdb.base/interrupt.exp
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/gdb.base/interrupt.exp,v
retrieving revision 1.3
diff -u -p -r1.3 interrupt.exp
--- gdb.base/interrupt.exp 2001/03/06 08:21:50 1.3
+++ gdb.base/interrupt.exp 2001/10/28 15:46:38
@@ -101,7 +101,7 @@ if ![file exists $binfile] then {
 	send_gdb "p func1 ()\n"
 	gdb_expect {
 	    -re " = 4.*$gdb_prompt $" { pass "call function when asleep" }
-	    -re ".*Program received signal SIGSEGV.*$gdb_prompt $" {
+	    -re ".*Program received signal SIG(SEGV|ILL).*$gdb_prompt $" {
 		setup_xfail "i*86-pc-linux*-gnu*"
 		fail "child died when we called func1, skipped rest of tests"
 		return