Re: [RFA/testsuite] New (Ada) testcase for watchpoint issue

[Joel Brobecker <brobecker at adacore dot com>]

[with the patch this time...]

Here is a new testcase that reproduces the issue presented in:

        http://www.sourceware.org/ml/gdb-patches/2006-10/msg00044.html

I tried reproducing this problem in C, but didn't manage too. Perhaps
I could have tried using a GCC extension and used a nested function,
but I thought it was cleaner to do it in pure legal Ada instead.

The only difference between the example in the message above and
the example I'm posting now is that I got read of the "delay 1.0"
statement, and replaced it with something much simpler in terms
of code generation - for targets that only have software watchpoints,
that should help shorten the time it takes to run this test (as well
as removing and unnecessary delay while waiting for the inferior
to run, actually!).

2006-10-05  Joel Brobecker  <brobecker@adacore.com>

        * gdb.ada/watch_arg/watch.adb: New file.
        * gdb.ada/watch_arg.exp: New testcase.

Tested on x86-linux, all PASSES with the patch submitted in the message
mentioned above, 1 FAIL without it:

        FAIL: gdb.ada/watch_arg.exp: Continuing to second breakpoint

OK to apply?

Thanks,
-- 
Joel

Attachment: watch.adb
Description: Text document

Attachment: watch_arg.exp
Description: Text document