http://sourceware.org/ml/gdb-patches/2016-01/msg00140.html
Subject: [testsuite patch] i386 regression for funcargs.exp  [Re: [PATCH 3/3] Remove HP-UX references fom testsuite]


--EeQfGwPcQSOJBaQU
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

On Mon, 21 Dec 2015 18:52:59 +0100, Simon Marchi wrote:
> Ok, I pushed this one in.  Thanks!

3ca22649a6dfeb71058c33be4d0542b98f1f0ff5 is the first bad commit
commit 3ca22649a6dfeb71058c33be4d0542b98f1f0ff5
Author: Simon Marchi <simon.marchi@polymtl.ca>
Date:   Mon Dec 21 12:51:54 2015 -0500
    Remove HP-UX references fom testsuite
### a/gdb/testsuite/gdb.base/funcargs.exp
### b/gdb/testsuite/gdb.base/funcargs.exp
@@ -1013,13 +1013,6 @@ proc localvars_in_indirect_call { } {
     #
 
     gdb_test_multiple "finish" "finish from indirectly called function" {
-	-re "\\(\\*pointer_to_call0a\\) \\(c, s, i, l\\);.*First.*$gdb_prompt $" {
-	    #On hppa2.0w-hp-hpux11.00, gdb finishes at one line earlier than 
-	    #hppa1.1-hp-hpux11.00. Therefore, an extra "step" is necessary 
-	    #to continue the test.
-	    send_gdb "step\n"
-	    exp_continue
-	}
 	-re ".*\\(\\*pointer_to_call0a\\) \\(c, s, i, l\\);.*Second.*$gdb_prompt $" {
 	    pass "finish from indirectly called function"
 	}

->

 finish^M
 Run till exit from #0  call0a (c=97 'a', s=1, i=2, l=3) at ./gdb.base/funcargs.c:82^M
 0x0804a189 in main () at ./gdb.base/funcargs.c:583^M
 583      (*pointer_to_call0a) (c, s, i, l);    /* First step into call0a.  */^M
-(gdb) step^M
-584      (*pointer_to_call0a) (c, s, i, l);    /* Second step into call0a.  */^M
-(gdb) PASS: gdb.base/funcargs.exp: finish from indirectly called function
+(gdb) FAIL: gdb.base/funcargs.exp: finish from indirectly called function
 step^M
-call0a (c=97 'a', s=1, i=2, l=3) at ./gdb.base/funcargs.c:82^M
-82       c = 'a';^M
-(gdb) PASS: gdb.base/funcargs.exp: stepping into indirectly called function
+584      (*pointer_to_call0a) (c, s, i, l);    /* Second step into call0a.  */^M
+(gdb) FAIL: gdb.base/funcargs.exp: stepping into indirectly called function

At least on x86_64 with testsuite in -m32 (expecting native i386 would be the
same).

OK for check-in?


Jan

--EeQfGwPcQSOJBaQU
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline; filename=1

gdb/testsuite/ChangeLog
2016-01-08  Jan Kratochvil  <jan.kratochvil@redhat.com>

	* gdb.base/funcargs.exp (finish from indirectly called function):
	Reintroduce the case for 'First'.

diff --git a/gdb/testsuite/gdb.base/funcargs.exp b/gdb/testsuite/gdb.base/funcargs.exp
index e5b6ced..a476169 100644
--- a/gdb/testsuite/gdb.base/funcargs.exp
+++ b/gdb/testsuite/gdb.base/funcargs.exp
@@ -1013,6 +1013,12 @@ proc localvars_in_indirect_call { } {
     #
 
     gdb_test_multiple "finish" "finish from indirectly called function" {
+	-re "\\(\\*pointer_to_call0a\\) \\(c, s, i, l\\);.*First.*$gdb_prompt $" {
+	    # On i386, gdb finishes at one line earlier than on x86_64.
+	    # Therefore, an extra "step" is necessary to continue the test.
+	    send_gdb "step\n"
+	    exp_continue
+	}
 	-re ".*\\(\\*pointer_to_call0a\\) \\(c, s, i, l\\);.*Second.*$gdb_prompt $" {
 	    pass "finish from indirectly called function"
 	}

--EeQfGwPcQSOJBaQU--