[RFA/testsuite/ada] KFAIL the tests that use "start"

[RFA/testsuite/ada] KFAIL the tests that use "start"

[Joel Brobecker]

[-- Attachment #1: Type: text/plain, Size: 437 bytes --]

Hello,

As suggested in http://sources.redhat.com/ml/gdb/2005-03/msg00067.html,
I am posting a patch that KFAILs the two tests that currently fail
while a patch fixing the problem is being reviewed.

2005-03-07  Joel Brobecker  <brobecker@adacore.com>

        * gdb.ada/null_record.exp: KFAIL failing "start" test.
        * gdb.ada/start.exp: Likewise.

Tested on x86-linux, changes the FAIL to a KFAIL.
Ok to apply?

Thanks,
-- 
Joel

[-- Attachment #2: ada-ts.diff --]
[-- Type: text/plain, Size: 1213 bytes --]

Index: gdb.ada/null_record.exp
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/gdb.ada/null_record.exp,v
retrieving revision 1.6
diff -u -p -r1.6 null_record.exp
--- gdb.ada/null_record.exp	24 Feb 2005 23:54:03 -0000	1.6
+++ gdb.ada/null_record.exp	7 Mar 2005 20:31:54 -0000
@@ -36,6 +36,9 @@ gdb_start
 gdb_reinitialize_dir $srcdir/$subdir
 gdb_load ${binfile}
 
+# The "start" command is not yet fully implemented for Ada.
+setup_kfail *-*-* "ada/1892"
+
 gdb_test "start" \
          "null_record \\(\\) at .*null_record.adb.*" \
          "start"
Index: gdb.ada/start.exp
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/gdb.ada/start.exp,v
retrieving revision 1.2
diff -u -p -r1.2 start.exp
--- gdb.ada/start.exp	24 Feb 2005 23:54:03 -0000	1.2
+++ gdb.ada/start.exp	7 Mar 2005 20:31:54 -0000
@@ -36,6 +36,9 @@ gdb_start
 gdb_reinitialize_dir $srcdir/$subdir
 gdb_load ${binfile}
 
+# The "start" command is not yet fully implemented for Ada.
+setup_kfail *-*-* "ada/1892"
+
 # Verify that "start" lands inside the right procedure.
 gdb_test "start" \
          "dummy \\(\\) at .*dummy.adb.*" \

Re: [RFA/testsuite/ada] KFAIL the tests that use "start"

[Daniel Jacobowitz]

On Mon, Mar 07, 2005 at 12:36:08PM -0800, Joel Brobecker wrote:
> Hello,
> 
> As suggested in http://sources.redhat.com/ml/gdb/2005-03/msg00067.html,
> I am posting a patch that KFAILs the two tests that currently fail
> while a patch fixing the problem is being reviewed.
> 
> 2005-03-07  Joel Brobecker  <brobecker@adacore.com>
> 
>         * gdb.ada/null_record.exp: KFAIL failing "start" test.
>         * gdb.ada/start.exp: Likewise.
> 
> Tested on x86-linux, changes the FAIL to a KFAIL.
> Ok to apply?

What Ada test results do you have with this patch?  I have more than
two failures; IIRC, there were four, all related.

-- 
Daniel Jacobowitz
CodeSourcery, LLC

Re: [RFA/testsuite/ada] KFAIL the tests that use "start"

[Joel Brobecker]

> What Ada test results do you have with this patch?  I have more than
> two failures; IIRC, there were four, all related.

This is what I have:

    Running ../.././gdb/testsuite/gdb.ada/fixed_points.exp ...
    PASS: gdb.ada/fixed_points.exp: p on a fixed point type
    PASS: gdb.ada/fixed_points.exp: p on a subtype fixed point type
    PASS: gdb.ada/fixed_points.exp: p on a new fixed point type
    Running ../.././gdb/testsuite/gdb.ada/null_record.exp ...
    KFAIL: gdb.ada/null_record.exp: start (PRMS: ada/1892)
    PASS: gdb.ada/null_record.exp: ptype on null record
    Running ../.././gdb/testsuite/gdb.ada/packed_array.exp ...
    PASS: gdb.ada/packed_array.exp: print var
    Running ../.././gdb/testsuite/gdb.ada/start.exp ...
    KFAIL: gdb.ada/start.exp: start (PRMS: ada/1892)

Could you send me your logs. I only had 2 failures on my side.
Perhaps there's something else going on because I use a different
compiler than you do?

-- 
Joel

Re: [RFA/testsuite/ada] KFAIL the tests that use "start"

[Joel Brobecker]

> As suggested in http://sources.redhat.com/ml/gdb/2005-03/msg00067.html,
> I am posting a patch that KFAILs the two tests that currently fail
> while a patch fixing the problem is being reviewed.
> 
> 2005-03-07  Joel Brobecker  <brobecker@adacore.com>
> 
>         * gdb.ada/null_record.exp: KFAIL failing "start" test.
>         * gdb.ada/start.exp: Likewise.

Since the fix has recently been checked in, the problem no longer
exist, and I therefore withdraw this patch. The GDB PR has also
been closed.

-- 
Joel

Re: [RFA/testsuite/ada] KFAIL the tests that use "start"

[Paul Gilliam]

On Monday 07 March 2005 22:46, Joel Brobecker wrote:
> > As suggested in http://sources.redhat.com/ml/gdb/2005-03/msg00067.html,
> > I am posting a patch that KFAILs the two tests that currently fail
> > while a patch fixing the problem is being reviewed.
> >
> > 2005-03-07  Joel Brobecker  <brobecker@adacore.com>
> >
> >         * gdb.ada/null_record.exp: KFAIL failing "start" test.
> >         * gdb.ada/start.exp: Likewise.
>
> Since the fix has recently been checked in, the problem no longer
> exist, and I therefore withdraw this patch. The GDB PR has also
> been closed.

If the testsuite is to be used for regression testing, then you should not 
remove a test because the problem it identifies has been fixed.

-=# Paul Gilliam #=-

Re: [RFA/testsuite/ada] KFAIL the tests that use "start"

[Joel Brobecker]

> > > 2005-03-07  Joel Brobecker  <brobecker@adacore.com>
> > >
> > >         * gdb.ada/null_record.exp: KFAIL failing "start" test.
> > >         * gdb.ada/start.exp: Likewise.
> >
> > Since the fix has recently been checked in, the problem no longer
> > exist, and I therefore withdraw this patch. The GDB PR has also
> > been closed.
> 
> If the testsuite is to be used for regression testing, then you should not 
> remove a test because the problem it identifies has been fixed.

Right. The testcases remain, but I no longer need to KFAIL them, as
they now PASS.

-- 
Joel

Re: [RFA/testsuite/ada] KFAIL the tests that use "start"

[Paul Gilliam]

On Wednesday 09 March 2005 09:51, Joel Brobecker wrote:
> > > > 2005-03-07  Joel Brobecker  <brobecker@adacore.com>
> > > >
> > > >         * gdb.ada/null_record.exp: KFAIL failing "start" test.
> > > >         * gdb.ada/start.exp: Likewise.
> > >
> > > Since the fix has recently been checked in, the problem no longer
> > > exist, and I therefore withdraw this patch. The GDB PR has also
> > > been closed.
> >
> > If the testsuite is to be used for regression testing, then you should
> > not remove a test because the problem it identifies has been fixed.
>
> Right. The testcases remain, but I no longer need to KFAIL them, as
> they now PASS.
Ah, I see.   Thanks for the clearification (or education ;-)

-=# Paul #=-