[patch] fix for PR2424

[Aleksandar Ristovski]

Daniel Jacobowitz wrote:
> On Mon, Apr 14, 2008 at 10:41:38AM -0400, Aleksandar Ristovski wrote:
>> ping?
> 
> Thanks.  The most recent version of this patch (code + testsuite) is OK.
> 
> Please mention the PR at the top of the changelog entry, like this:
> 
> DATE  NAME  <ADDR>
> 
> 	PR gdb/2424
> 	* infrun.c (some change): ....
> 
> Then if you include that line in the commit message, the changelog
> entry is automatically appended to the PR audit trail.
> 
> Shouldn't the disp="del" field be described in the manual?  If so,
> that can be a separate patch.
> 
Committed.

NOTE1 (reminder): With this patch, we have new field in the *stopped
message, 
when "reason == breakpoint-hit" - we will have "disp" field printed; the
value 
of the field will be exactly the same as the value printed when the
breakpoint 
was created.

NOTE2: Due to quite significant rework of certain functions in
mi-support.exp, 
committed diff for testsuite is not exactly the same as the one I posted. I
took 
liberty to commit anyway since the changes are functionally the same; note
the 
slightly awkward syntax for "mi_expect_stop" call, namely the "extra"
argument 
which now must be something like { "" "disp=\"keep\"" } - I don't think it's
too 
bad (it can be beautified later).

NOTE3: In mi-support.exp (mi_expect_stop), I put argument "after_reason" to
be 
after the reason (I couldn't find it being used anywhere anyway, so this
should 
be a safe change).

NOTE4: Documentation is not updated (I will try to do that now and submit
the 
diff for approval).

Thanks,

Aleksandar Ristovski
QNX Software Systems