Re: [patch] gdb.c++/pr-1023.exp: new test script

[Daniel Jacobowitz <drow@mvista.com>]

On Mon, Feb 03, 2003 at 12:05:52PM -0600, Michael Elizabeth Chastain wrote:
> This is a new C++ test script.  It demonstrates the bug in PR gdb/1023.
> This bug is still present in gdb HEAD%20030201, so the test script
> has a KFAIL arm in it.
> 
> Testing: native i686-pc-linux-gnu, gcc v2 and v3, dwarf-2 and stabs+.
> All results are either PASS or KFAIL.  As the PR says, it KFAIL's with
> gcc 2.95.3 dwarf-2.
> 
> I will wait 24 hours and then commit this.
> 
> This is a simple test script, so it will be a good candidate for the
> new gdb_test_multiple syntax.  *After* committing this nice working
> version, I will be open to gdb_test_multiple experiments.

Sounds good.  By the way:

> setup_kfail "gdb/1023"
> 
> send_gdb "break myClass::performBlocking\n"
> gdb_expect {
>     -re "Breakpoint $decimal at $hex: file .*$srcfile, line 12.*$gdb_prompt $" {
> 	pass "break myClass::performBlocking"
>     }
>     -re "the class myClass does not have any method named performBlocking.*$gdb_prompt $" {
> 	# fails with gcc 2.95.3 -gstabs+, native i686-pc-linux-gnu
> 	# -- chastain 2003-02-03
> 	kfail "gdb/1023" "break myClass::performBlocking"
>     }

You should be using either setup_kfail or an explicit kfail, but not
both.  I think you meant the latter in this case.

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer