* Re: A patch for gcc/testsuite
@ 1998-02-04 15:52 Mike Stump
1998-02-06 0:57 ` H.J. Lu
0 siblings, 1 reply; 6+ messages in thread
From: Mike Stump @ 1998-02-04 15:52 UTC (permalink / raw)
To: hjl, schmid; +Cc: egcs
This is the wrong solution to this problem. Please don't install this patch.
The right solution is to add a `execution test - XFAIL *-*-*' to old-deja,
in the spirit of `excess errors test - ' and then change `execution test fails'
to be like `excess errors test fails'. Then you will notice that:
egcs/gcc/testsuite/g++.old-deja/g++.jason/opeq5.C:// execution test fails -
egcs/gcc/testsuite/g++.old-deja/g++.law/code-gen3.C:// execution test fails -
egcs/gcc/testsuite/g++.old-deja/g++.mike/p7325.C:// execution test fails *-*-*
Should be change to use the new `execution test - XFAIL *-*-*' support.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: A patch for gcc/testsuite
1998-02-04 15:52 A patch for gcc/testsuite Mike Stump
@ 1998-02-06 0:57 ` H.J. Lu
1998-02-09 1:07 ` Jeffrey A Law
0 siblings, 1 reply; 6+ messages in thread
From: H.J. Lu @ 1998-02-06 0:57 UTC (permalink / raw)
To: Mike Stump; +Cc: egcs
>
> This is the wrong solution to this problem. Please don't install this patch.
> The right solution is to add a `execution test - XFAIL *-*-*' to old-deja,
> in the spirit of `excess errors test - ' and then change `execution test fails'
> to be like `excess errors test fails'. Then you will notice that:
>
> egcs/gcc/testsuite/g++.old-deja/g++.jason/opeq5.C:// execution test fails -
> egcs/gcc/testsuite/g++.old-deja/g++.law/code-gen3.C:// execution test fails -
> egcs/gcc/testsuite/g++.old-deja/g++.mike/p7325.C:// execution test fails *-*-*
>
> Should be change to use the new `execution test - XFAIL *-*-*' support.
>
How about this patch?
--
H.J. Lu (hjl@gnu.org)
----
Thu Feb 5 23:17:37 1998 H.J. Lu (hjl@gnu.org)
* g++.old-deja/g++.jason/opeq5.C: Changed
// execution test fails - ...
to
// execution test - XFAIL ...
* g++.old-deja/g++.law/code-gen3.C: Ditto.
* g++.old-deja/g++.mike/p6610a.C: Ditto.
* g++.old-deja/g++.mike/p7325.C: Ditto.
* lib/old-dejagnu.exp (old-dejagnu): Added the
"execution test - XFAIL *-*-*" handling in the spirit of
"excess errors test -". Changed the "execution test fails"
handling to be like "excess errors test fails".
diff -x CVS -ur ../../../../import/egcs/gcc/testsuite/g++.old-deja/g++.jason/opeq5.C ./g++.old-deja/g++.jason/opeq5.C
--- ../../../../import/egcs/gcc/testsuite/g++.old-deja/g++.jason/opeq5.C Mon Jan 12 17:03:54 1998
+++ ./g++.old-deja/g++.jason/opeq5.C Thu Feb 5 16:29:02 1998
@@ -1,7 +1,7 @@
// Testcase for tricky synthesized op= in complex inheritance situation.
// See discussion in g++int.texi.
-// execution test fails -
+// execution test - XFAIL *-*-*
int count = 0;
extern "C" int printf (const char *, ...);
diff -x CVS -ur ../../../../import/egcs/gcc/testsuite/g++.old-deja/g++.law/code-gen3.C ./g++.old-deja/g++.law/code-gen3.C
--- ../../../../import/egcs/gcc/testsuite/g++.old-deja/g++.law/code-gen3.C Mon Jan 12 17:04:07 1998
+++ ./g++.old-deja/g++.law/code-gen3.C Thu Feb 5 16:29:22 1998
@@ -1,5 +1,5 @@
// GROUPS passed code-generation
-// execution test fails -
+// execution test - XFAIL *-*-*
// code-gen file
// From: mscha@anne.wifo.uni-mannheim.de (Martin Schader)
// Date: Wed, 4 Aug 93 19:14:52 +0200
diff -x CVS -ur ../../../../import/egcs/gcc/testsuite/g++.old-deja/g++.mike/p6610a.C ./g++.old-deja/g++.mike/p6610a.C
--- ../../../../import/egcs/gcc/testsuite/g++.old-deja/g++.mike/p6610a.C Wed Feb 4 14:17:55 1998
+++ ./g++.old-deja/g++.mike/p6610a.C Thu Feb 5 18:26:41 1998
@@ -1,6 +1,6 @@
// prms-id: 6610
// There is a bug in vtable thunks with multiple/virtual inheritance.
-// execution test fails - XFAIL *-*-linux-gnu *-*-linux
+// execution test - XFAIL *-*-linux-gnu *-*-linux
int fail = 1;
struct B;
struct A { virtual int f(const B*) = 0; int g(const B*); };
diff -x CVS -ur ../../../../import/egcs/gcc/testsuite/g++.old-deja/g++.mike/p7325.C ./g++.old-deja/g++.mike/p7325.C
--- ../../../../import/egcs/gcc/testsuite/g++.old-deja/g++.mike/p7325.C Mon Jan 12 17:04:50 1998
+++ ./g++.old-deja/g++.mike/p7325.C Thu Feb 5 16:30:30 1998
@@ -1,7 +1,7 @@
// I hate this type of test case. I'm not sure how to code it better.
// See the PR for what this tests.
// prms-id: 7325
-// execution test fails *-*-*
+// execution test - XFAIL *-*-*
int fail = 0;
diff -x CVS -ur ../../../../import/egcs/gcc/testsuite/lib/old-dejagnu.exp ./lib/old-dejagnu.exp
--- ../../../../import/egcs/gcc/testsuite/lib/old-dejagnu.exp Mon Jan 12 17:06:11 1998
+++ ./lib/old-dejagnu.exp Thu Feb 5 16:27:34 1998
@@ -361,10 +361,16 @@
set message [concat $message $tmp]
}
- set tmp [process-option $prog "execution test fails" "an execution failure" EXEC $text]
+ set tmp [process-option $prog "execution test fails" "an execution failure" EXECO $text]
if ![string match "" $tmp] then {
set execbug_flag 1
set message [concat $message $tmp]
+ warning "please use execution test - XFAIL *-*-* in $prog instead"
+ }
+
+ set tmp [process-option $prog "execution test - " "an excess error failure" EXEC $text]
+ if ![string match "" $tmp] then {
+ set message [concat $message $tmp]
}
set tmp [process-option $prog "excess errors test fails" "an excess error failure" EXCESSO $text]
@@ -472,7 +478,9 @@
"XBADASM" {
x$uhoh "$name $pattern (test for bad assembler, line $line)"
}
- "XEXEC" { }
+ "XEXEC" {
+ set execbug_flag 1
+ }
"XEXCESS" {
set excessbug_flag 1
}
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: A patch for gcc/testsuite
1998-02-06 0:57 ` H.J. Lu
@ 1998-02-09 1:07 ` Jeffrey A Law
0 siblings, 0 replies; 6+ messages in thread
From: Jeffrey A Law @ 1998-02-09 1:07 UTC (permalink / raw)
To: H.J. Lu; +Cc: Mike Stump, egcs
In message < m0y0iCW-0004ecC@ocean.lucon.org >you write:
> Thu Feb 5 23:17:37 1998 H.J. Lu (hjl@gnu.org)
>
> * g++.old-deja/g++.jason/opeq5.C: Changed
> // execution test fails - ...
> to
> // execution test - XFAIL ...
> * g++.old-deja/g++.law/code-gen3.C: Ditto.
> * g++.old-deja/g++.mike/p6610a.C: Ditto.
> * g++.old-deja/g++.mike/p7325.C: Ditto.
>
> * lib/old-dejagnu.exp (old-dejagnu): Added the
> "execution test - XFAIL *-*-*" handling in the spirit of
> "excess errors test -". Changed the "execution test fails"
> handling to be like "excess errors test fails".
Installed.
jeff
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: A patch for gcc/testsuite
@ 1998-02-06 16:53 Mike Stump
0 siblings, 0 replies; 6+ messages in thread
From: Mike Stump @ 1998-02-06 16:53 UTC (permalink / raw)
To: hjl; +Cc: egcs
Fairly beautiful. I say install it.
> From: hjl@lucon.org (H.J. Lu)
> To: mrs@wrs.com (Mike Stump)
> Date: Thu, 5 Feb 1998 23:27:52 -0800 (PST)
> How about this patch?
> Thu Feb 5 23:17:37 1998 H.J. Lu (hjl@gnu.org)
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: A patch for gcc/testsuite
@ 1998-02-06 8:53 Peter Schmid
0 siblings, 0 replies; 6+ messages in thread
From: Peter Schmid @ 1998-02-06 8:53 UTC (permalink / raw)
To: hjl; +Cc: egcs
These are the results of the g++ testsuite after applying your patch
<egcs/1998-Feb/0225.html>.
The XPASS in g++.mike/p6610a.C is gone.
Thanks for fixing the problem.
Peter Schmid.
Native configuration is i586-pc-linux-gnulibc1
=== g++ tests ===
Running target unix
Using /usr/bin/../share/dejagnu/baseboards/unix.exp as board description file
for target.
Using /usr/bin/../share/dejagnu/config/unix.exp as generic interface file for
target.
Using /home/schmid/ftp/egcs-980205/gcc/testsuite/config/default.exp as
tool-and-target-specific interface file.
Running /home/schmid/ftp/egcs-980205/gcc/testsuite/g++.old-deja/old-deja.exp
...
FAIL: g++.other/realloc.C (test for excess errors)
=== g++ Summary ===
# of expected passes 3648
# of unexpected failures 1
# of expected failures 79
# of untested testcases 6
/home/schmid/ftp/egcs-980205/gcc/testsuite/../xgcc version egcs-2.91.07 980205
(gcc-2.8.0 release)
make: [check-g++] Error 1 (ignored)
^ permalink raw reply [flat|nested] 6+ messages in thread
* results for egcs-2.91.06-cvs i386-pc-linux-gnulibc1
@ 1998-02-04 9:21 Peter Schmid
1998-02-04 12:28 ` A patch for gcc/testsuite H.J. Lu
0 siblings, 1 reply; 6+ messages in thread
From: Peter Schmid @ 1998-02-04 9:21 UTC (permalink / raw)
To: egcs
These are the testsuite results for egcs-2.91.06-cvs on the
i386-pc-linux-gnulibc1 system.
I updated the cvs sources at Feb 4 12:06:02 1998.
Peter Schmid
g++:
XPASS: g++.mike/p6610a.C - Execution test
FAIL: g++.other/realloc.C (test for excess errors)
/home/schmid/ftp/egcs/gcc/testsuite/g++.old-deja/g++.other/realloc.C:12:
Internal compiler error.
/home/schmid/ftp/egcs/gcc/testsuite/g++.old-deja/g++.other/realloc.C:12:
Please submit a full bug report to `egcs-bugs@cygnus.com'.
g77:
According to the g77 documentation this behaviour of dnrm2 is expected
on the x86 architecture.
g77.log:FAIL: g77.f-torture/execute/dnrm2.f execution, -O2
-fomit-frame-pointer -finline-functions -funroll-loops
g77.log:FAIL: g77.f-torture/execute/dnrm2.f execution, -O2
-fomit-frame-pointer -finline-functions -funroll-all-loops
^ permalink raw reply [flat|nested] 6+ messages in thread
* A patch for gcc/testsuite
1998-02-04 9:21 results for egcs-2.91.06-cvs i386-pc-linux-gnulibc1 Peter Schmid
@ 1998-02-04 12:28 ` H.J. Lu
0 siblings, 0 replies; 6+ messages in thread
From: H.J. Lu @ 1998-02-04 12:28 UTC (permalink / raw)
To: Peter Schmid; +Cc: egcs
>
>
> These are the testsuite results for egcs-2.91.06-cvs on the
> i386-pc-linux-gnulibc1 system.
> I updated the cvs sources at Feb 4 12:06:02 1998.
>
> Peter Schmid
>
> g++:
>
> XPASS: g++.mike/p6610a.C - Execution test
What is PATTERN in process-option used for? It is kind of bogus
to me. Even if I have
// execution test fails - XFAIL *-*-linux-gnu *-*-linux
in g++.old-deja/g++.mike/p6610a.C, dejagnu still sets execbug_flag
for i586-unknown-linux-gnulibc1 since process will always return
a non-empty string if XFAIL is found. Does this patch make any senses?
Thanks.
H.J.
----
Wed Feb 4 07:19:22 1998 H.J. Lu (hjl@gnu.org)
* lib/old-dejagnu.exp (process-option): Append result only if
$triplet_match is not 0.
--- lib/old-dejagnu.exp.orig Wed Feb 4 11:41:34 1998
+++ lib/old-dejagnu.exp Wed Feb 4 12:13:18 1998
@@ -79,10 +79,12 @@
}
}
set compos [expr [llength $option] + 1] ;# Start of comment, if any
- if { $xfail_test && $triplet_match } then {
- lappend result [list [lindex $i 0] "X$flag_name" [lrange $i $compos end] "$pattern"]
- } else {
- lappend result [list [lindex $i 0] "$flag_name" [lrange $i $compos end] "$pattern"]
+ if { $triplet_match } then {
+ if { $xfail_test } then {
+ lappend result [list [lindex $i 0] "X$flag_name" [lrange $i $compos end] "$pattern"]
+ } else {
+ lappend result [list [lindex $i 0] "$flag_name" [lrange $i $compos end] "$pattern"]
+ }
}
if { $verbose > 1 } then {
if [string match "" [lrange $i $compos end]] then {
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~1998-02-09 1:07 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
1998-02-04 15:52 A patch for gcc/testsuite Mike Stump
1998-02-06 0:57 ` H.J. Lu
1998-02-09 1:07 ` Jeffrey A Law
-- strict thread matches above, loose matches on Subject: below --
1998-02-06 16:53 Mike Stump
1998-02-06 8:53 Peter Schmid
1998-02-04 9:21 results for egcs-2.91.06-cvs i386-pc-linux-gnulibc1 Peter Schmid
1998-02-04 12:28 ` A patch for gcc/testsuite H.J. Lu
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).