*BSD News Article 26012


Return to BSD News archive

Newsgroups: comp.os.386bsd.questions
Path: sserve!newshost.anu.edu.au!munnari.oz.au!sgiblab!swrinde!cs.utexas.edu!math.ohio-state.edu!jussieu.fr!univ-lyon1.fr!frmug.fr.net!fasterix.frmug.fr.net!pb
From: pb@fasterix.frmug.fr.net (Pierre Beyssac)
Subject: Re: NetBSD - routing and slip
References: <2gv4lc$l0i@darkstar.UCSC.EDU> <CJJ0w6.ADo@usenet.ucs.indiana.edu> <2h1neg$gju@pdq.coe.montana.edu> <CJJHoM.r9@usenet.ucs.indiana.edu>
Organization: considered harmful
Date: Fri, 14 Jan 1994 23:48:12 GMT
Message-ID: <1994Jan14.234812.7205@fasterix.frmug.fr.net>
Lines: 51

In article <CJJHoM.r9@usenet.ucs.indiana.edu>,
Jim Pitts <pitts@mimosa.astro.indiana.edu> wrote:
>Many moons ago I posted posted something about how the sio drivers had solved
>all the problems in my life ... how great they are.  I got several flame
>messages telling me that the NetBSD drivers were faster and I was an idiot.
>Not the latter -may- be true, I don't deny it ... but I honestly never saw
>any performance loss when I went to FreeBSD and its sio drivers.  I
>never measured with enough intrest to notice a speed -increase-.

I must be an idiot too :-)

I noticed no performance loss when I went from NetBSD/com to
FreeBSD/sio. In fact I had *noticeable* performance *improvements*.
No speed increase, but a better reliability when transferring on
a loaded machine.

On a 486DX2/66, with good old 16450 UARTs :

	1) have been able to use the 'i' protocol of Taylor UUCP
	   reliably at 14400 bps (never could with the com driver
	   even at 2400bps)

	2) have been able to make compiles and process news batches
	   in background while receiving data at 14400bps via UUCP
	   (I had *many* errors when the machine was loaded during
	    transfers with the com driver)

	3) UUCP detects carrier loss with a SIGHUP (detects nothing
	   and times out with the com driver)

The reasons are simple :

	- interrupt latency is lower with sio (sio does not use splx, and
	  interrupt processing when sending data out is faster)
	  (this is probably useful specially on slow machines)

	- probably most of all on my machine, sio handles CTS/RTS
	  flow control

	- and sio looks at CD

In fact, by looking at the sio and com code, it seems that sio is
derived from the com driver.

Now maybe I'm an idiot or I understood nothing at all in the code
I read, but I am perfectly happy with the sio driver, thank you !
-- 
Pierre Beyssac
FreeBSD@home: pb@fasterix.frmug.fr.net

NetBSD, FreeBSD, Linux -- Il y a moins bien, mais c'est plus cher.