MI testsuite mi-until.exp failures

MI testsuite mi-until.exp failures

[Bob Rossi]

Hi,

Using gcc 4.0, I get error's with the mi-until.exp and mi2-until.exp
testcases. The first issue is trivial.

The file mi-var-child.c needed '#include <string.h>' in order for the
file to compile because of memset. Is string.h appropriate for this? I
know memset can be in different headers on different systems.

Also, for some reason, the line number was different. So, either it's OK
for the line number to be different and this patch is OK, or it's not OK
for the line number to be different and this patch is bad. Any ideas?

Thanks,
Bob Rossi

Index: src/gdb/testsuite/ChangeLog
+	* gdb.mi/mi-until.exp: Change hard coded number to regex for number.
+	* gdb.mi/mi2-until.exp: Ditto.
+	* gdb.mi/mi-var-child.c: Add #include <string.h>.

Index: gdb/testsuite/gdb.mi/mi-until.exp
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/gdb.mi/mi-until.exp,v
retrieving revision 1.9
diff -u -r1.9 mi-until.exp
--- gdb/testsuite/gdb.mi/mi-until.exp	18 May 2005 03:41:59 -0000	1.9
+++ gdb/testsuite/gdb.mi/mi-until.exp	17 Aug 2005 03:08:37 -0000
@@ -78,7 +78,7 @@
 
     send_gdb "111-exec-until\n"
     gdb_expect {
-	-re "111\\^running\r\n${mi_gdb_prompt}111\\*stopped,reason=\"end-stepping-range\",thread-id=\"\[01\]\",frame=\{addr=\"$hex\",func=\"foo\",args=\\\[\\\],file=\".*until.c\",fullname=\"${fullname_syntax}${srcfile}\",line=\"12\"\}\r\n$mi_gdb_prompt$" {
+	-re "111\\^running\r\n${mi_gdb_prompt}111\\*stopped,reason=\"end-stepping-range\",thread-id=\"\[01\]\",frame=\{addr=\"$hex\",func=\"foo\",args=\\\[\\\],file=\".*until.c\",fullname=\"${fullname_syntax}${srcfile}\",line=\"\[0-9\]+\"\}\r\n$mi_gdb_prompt$" {
 	    pass "until after while loop"
 	}
 	timeout {
Index: gdb/testsuite/gdb.mi/mi-var-child.c
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/gdb.mi/mi-var-child.c,v
retrieving revision 1.2
diff -u -r1.2 mi-var-child.c
--- gdb/testsuite/gdb.mi/mi-var-child.c	31 Jul 2005 22:10:53 -0000	1.2
+++ gdb/testsuite/gdb.mi/mi-var-child.c	17 Aug 2005 03:08:37 -0000
@@ -18,6 +18,7 @@
    Boston, MA 02111-1307, USA.  */
 
 #include <stdlib.h>
+#include <string.h>
 
 struct _simple_struct {
   int integer;
Index: gdb/testsuite/gdb.mi/mi2-until.exp
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/gdb.mi/mi2-until.exp,v
retrieving revision 1.3
diff -u -r1.3 mi2-until.exp
--- gdb/testsuite/gdb.mi/mi2-until.exp	18 May 2005 03:41:59 -0000	1.3
+++ gdb/testsuite/gdb.mi/mi2-until.exp	17 Aug 2005 03:08:37 -0000
@@ -79,7 +79,7 @@
 
     send_gdb "111-exec-until\n"
     gdb_expect {
-	-re "111\\^running\r\n${mi_gdb_prompt}111\\*stopped,reason=\"end-stepping-range\",thread-id=\"\[01\]\",frame=\{addr=\"$hex\",func=\"foo\",args=\\\[\\\],file=\".*until.c\",fullname=\"${fullname_syntax}${srcfile}\",line=\"12\"\}\r\n$mi_gdb_prompt$" {
+	-re "111\\^running\r\n${mi_gdb_prompt}111\\*stopped,reason=\"end-stepping-range\",thread-id=\"\[01\]\",frame=\{addr=\"$hex\",func=\"foo\",args=\\\[\\\],file=\".*until.c\",fullname=\"${fullname_syntax}${srcfile}\",line=\"\[0-9\]+\"\}\r\n$mi_gdb_prompt$" {
 	    pass "until after while loop"
 	}
 	timeout {

Re: MI testsuite mi-until.exp failures

[Daniel Jacobowitz]

On Tue, Aug 16, 2005 at 11:17:18PM -0400, Bob Rossi wrote:
> Hi,
> 
> Using gcc 4.0, I get error's with the mi-until.exp and mi2-until.exp
> testcases. The first issue is trivial.
> 
> The file mi-var-child.c needed '#include <string.h>' in order for the
> file to compile because of memset. Is string.h appropriate for this? I
> know memset can be in different headers on different systems.

<string.h> is fine.  I do these things mostly by example: for instance
call-ar-st.c already includes it unconditionally so it must be OK for
the testsuite, on targets anyone's cared about lately.

This bit is OK, feel free to commit it on its own.

> Also, for some reason, the line number was different. So, either it's OK
> for the line number to be different and this patch is OK, or it's not OK
> for the line number to be different and this patch is bad. Any ideas?

Well that doesn't tell me much, and I hate decoding regexps to figure
this stuff out.  What's the command?  Where's it stopping?  Where's it
supposed to stop?  Does the behavior seem reasonable to you?

I seem to recall that this sparked some contention about the correct
behavior of "until", which we discussed on gdb@ some months ago.  That
may have been a different bug, though.

-- 
Daniel Jacobowitz
CodeSourcery, LLC