[9fans] APE select(2)

[9fans] APE select(2)

[lucio]

I can't quite get my head around all the twists and turns of the APE
select(2) implementation, I seem to miss the big picture altogether.
Is there a brief description somewhere that will make it easier for me
to understand the source?

++L

Re: [9fans] APE select(2)

[Charles Forsyth]

/sys/src/ape/lib/ap/plan9/buf.prom
might help, if it's reasonably consistent with
the current implementation.

RE: [9fans] APE select(2)

[rog]

> Now to do a little more digging.  Promela, unfortunately, I am not at
> all familiar with, but this may be an occasion to catch up.

it's well worth it.

RE: [9fans] APE select(2)

[lucio]

> With regards to the nfds test in the code given earlier, I agree that it seems that:
> 
> 	if(nfds >= OPEN_MAX)
> 		nfds = OPEN_MAX-1;
> 	for(i = 0; i<= nfds; i++)
> 
> should change to
> 
> 	if(nfds > OPEN_MAX)
> 		nfds = OPEN_MAX;
> 	for(i = 0; i< nfds; i++)
> 
Thanks, Howard, this improves things a little, enough to show promise.
It also gives me the confidence that I'm not completely out of my
depths.

Now to do a little more digging.  Promela, unfortunately, I am not at
all familiar with, but this may be an occasion to catch up.

++L

RE: [9fans] APE select(2)

[Trickey, Howard W (Howard)]

I wrote the original APE select implementation, and the promela model below, but others have had their hands in it since, so it is likely that the promela model is out of date.

With regards to the nfds test in the code given earlier, I agree that it seems that:

	if(nfds >= OPEN_MAX)
		nfds = OPEN_MAX-1;
	for(i = 0; i<= nfds; i++)

should change to

	if(nfds > OPEN_MAX)
		nfds = OPEN_MAX;
	for(i = 0; i< nfds; i++)

- Howard

-----Original Message-----
From: Charles Forsyth [mailto:forsyth@terzarima.net]
Sent: Friday, April 02, 2004 6:56 AM
To: 9fans@cse.psu.edu
Subject: Re: [9fans] APE select(2)


/sys/src/ape/lib/ap/plan9/buf.prom
might help, if it's reasonably consistent with
the current implementation.