Re: [RFA] stack.c: Always set current_source_{symtab,line}

[Keith Seitz <keiths at redhat dot com>]

On Fri, 30 Aug 2002, Andrew Cagney wrote:

> > Already in mi-cli.exp on the branch. (I think I submitted it, too, to this 
> > list.) The method is different, but it demonstrates the failure using 
> > "list" with the console interpreter.
> 
> For MI, the testcase needs to go in at the same time as the fix (this is 
> stronger then the rest of GDB where it is a ``should'').  It's also 
> important to check that the test actually fails without the fix.  That 
> way we know that the change at least fixed something :-)

Ok, but, as with all of this, it is IMPOSSIBLE to get this testcase to do 
anything UNTIL all the interps stuff is in. This bug only affects the 
command "-interpreter-exec console list". Well, this command doesn't yet 
exist in MI, because the infrastructure has not been approved for trunk.

Yes, I have confirmed that there is a difference with and without the 
patch on my branch. mi-cli.exp shows an additional failure without the 
patch.

I'm really at a loss about all this. Testcases, documentation, just about 
everything exists on the (public) branch. I can submit (and already did, 
in this case) a testcase to demonstrate the bug. The problem is that you 
can't run it at all in CVS head right now -- not until all the rest of the 
changes have been approved and committed.

Help me out here. I'm obviously approaching the approval process 
incorrectly with these (relatively) small, one-at-a-time changes. Is there 
something I can do to facilitate review/approval of the mega-changes 
needed to support interpreters in gdb? Can I do something better? Is 
there a better process that I can follow?

Keith