Re: [RFA] remote debugging patches

[Andrew Cagney <ac131313@cygnus.com>]

> Andrew Cagney wrote:
> Yes fine, er almost.  Can you change the name of this to ``DONT_WAIT_FOREVER_FLAG'' and suggest making it an enum. (so GDB can print it :-)
> 
> Why DONT_WAIT_FOREWER? IMHO wait_forewer is correct, because 0 means don't wait forewer (the default) and 1 means yes, wait forever.

But you defined:

> +#define WAIT_FOREVER_FLAG 0

Anyway, I think I misunderstood the reason for the change:

> The first one is pretty straightforward and just makes use of the last parameter to getpkt().

As far as I can tell, you've just replaced the last parameter of 
getpkt() with a hardwired value - initially a macro but now an enum.  Is 
that the intent?

When I wrote:

> If you're feeling really inspired (...), you could even introduce an enum to handle both the DO and DONT cases.

I was thinking of something more like:

enum { do_wait_forever, dont_wait_forever } wait_forever_flag;

int
getpkt (..., enum wait_forever_flag forever, ...)
{
	if (forever == do_wait_forever)
	  ...
	else
	  ...
	et.al.
}

{
	getpkt (blah, dont_wait_forever);
}
Andrew