Re: [PATCH] Partial infcmd.c cleanup

[Eli Zaretskii <eliz at elta dot co dot il>]

> Date: Sun, 25 Jan 2004 18:32:48 +0100 (CET)
> From: Mark Kettenis <kettenis@chello.nl>
> 
> Since I'll be changing this bit of code I though I'd give it a cleanup
> first.  Mostly wrapping long lines and adding some missing whitespace.
> 
> Committed as obvious,

I think changing finish_command_continuation from a void function into
a static void function does not really qualify as obvious.  I'm
talking about this part of the patch:

>  /* Stuff that needs to be done by the finish command after the target
> -   has stopped.  In asynchronous mode, we wait for the target to stop in
> -   the call to poll or select in the event loop, so it is impossible to
> -   do all the stuff as part of the finish_command function itself. The
> -   only chance we have to complete this command is in
> -   fetch_inferior_event, which is called by the event loop as soon as it
> -   detects that the target has stopped. This function is called via the
> -   cmd_continuation pointer. */
> -void
> +   has stopped.  In asynchronous mode, we wait for the target to stop
> +   in the call to poll or select in the event loop, so it is
> +   impossible to do all the stuff as part of the finish_command
> +   function itself.  The only chance we have to complete this command
> +   is in fetch_inferior_event, which is called by the event loop as
> +   soon as it detects that the target has stopped. This function is
> +   called via the cmd_continuation pointer.  */
> +
> +static void
>  finish_command_continuation (struct continuation_arg *arg)

Not that I'm objected to this change, but IMHO it shouldn't have been
done without at least a call for comments, or a few days of delay, to
give people some time to react.