Re: [RFA/doco] Adjust the normal_stop observer documentation

[Daniel Jacobowitz <drow at mvista dot com>]

On Mon, Mar 31, 2003 at 03:05:56PM -0800, Joel Brobecker wrote:
> Hello,
> 
> The normal_stop notification has been inserted in a slightly different
> location than the location I had in mind. As a consequence, the
> description for this notification needed to be slightly adjusted. I took
> this opportunity to fix a minor nit (same word repeated twice).

> ***************
> *** 40,48 ****
>   @cindex notification about inferior execution stop
>   
>   @value{GDBN} will notify all @code{normal_stop} observers when the
> ! inferior execution has just stopped, and all the associated internal
> ! processing (such as breakpoint commands, annotations, etc) is about to
> ! be performed before the @value{GDBN} prompt is returned to the user.
>   
>   The following interface is available to manage @code{normal_stop}
>   observers:
> --- 40,48 ----
>   @cindex notification about inferior execution stop
>   
>   @value{GDBN} will notify all @code{normal_stop} observers when the
> ! inferior execution has just stopped, the associated messages and
> ! annotations have been printed, and the control is about to be returned
> ! to the user.
>   
>   The following interface is available to manage @code{normal_stop}
>   observers:

Have user breakpoint commands been executed at this point?  How about
breakpoint conditions?  If so, it suggests that the cache management
you wanted for Ada needs to be done at an earlier observation point,
perhaps corresponding to each _actual_ stop as opposed to each user
visible stop.  Otherwise "info tasks" in a breakpoint commands list may
behave strangely...

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer