Date: Tue, 7 Dec 1999 16:34:17 +0100 From: "Niels Chr. Bank-Pedersen" <ncbp@bank-pedersen.dk> To: FreeBSD mobile Mailing List <freebsd-mobile@FreeBSD.ORG>, Greg Lehey <grog@lemis.com> Subject: Re: 3c589d w/ freebsd 3.3 works badly. Message-ID: <19991207163417.B27105@bank-pedersen.dk> In-Reply-To: <19991207102219.22780@mojave.sitaranetworks.com>; from grog@mojave.sitaranetworks.com on Tue, Dec 07, 1999 at 10:22:19AM -0500 References: <199912060251.NAA16461@cairo.anu.edu.au> <199912062312.QAA37583@harmony.village.org> <19991206210033.19522@mojave.sitaranetworks.com> <19991207160848.A27105@bank-pedersen.dk> <19991207102219.22780@mojave.sitaranetworks.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Dec 07, 1999 at 10:22:19AM -0500, Greg Lehey wrote: > On Tuesday, 7 December 1999 at 16:08:49 +0100, Niels Chr. Bank-Pedersen wrote: > > On Mon, Dec 06, 1999 at 09:00:33PM -0500, Greg Lehey wrote: > >> [Moved to FreeBSD-mobile] > >> > >> On Monday, 6 December 1999 at 16:12:49 -0700, Warner Losh wrote: > >>> In message <199912060251.NAA16461@cairo.anu.edu.au> Darren Reed writes: > >>> : How reliable should the ep0 driver be with 3c389d pcmcia cards ? > >>> > >>> I had no problems using 3.3 and my 3C589D, but I've only done minor > >>> stuff with that. I've done most of my work on -current, however. The > >>> most likely problem is that you're using the wrong IRQ for the card. > >>> You'll want to check /etc/rc.conf to make sure that you are using the > >>> /etc/pccard.conf file. Also, you'll want to make sure that the irq > >>> line is correct. > >> > >> I don't know if this is the same issue, but I've seen terrible write > >> performance on my 3C589C under -CURRENT, and so has phk. Read > >> performance is OK. Looking at the hub, I see a short burst of > >> activity and then nothing for the rest of a second. This repeats > >> itself in this manner. No errors, but a write throughput of less than > >> 50 kB/s. I've seen this before recent changes in -CURRENT, but it > >> seems worse now (or maybe I've just paid more attention to it now :-). > > > > I've seen something similar - turned out to be changes in the pcm > > detection that resulted in irq conflicts (as mentioned above). > > After I hardwired irg/drq in my kernel, the problem was gone... > > Did this result in this one-second delay only on writes? I would > expect that the behaviour in the case of interrupt conflicts would be > the same for reads and writes, and it would occur on just about every > packet, not allowing 30 or 50 out and then delaying for a significant > period of time. Now you are getting too specific for my memory :-) You are probably right. What I saw was an unbearable performance when writing a large file over ftp. The transmission went down to what I recall as less than 8KB/s, and it can easily have been the same thing in the other direction. I don't know why i suspected pcm at that time (probably some recent commits), but after wiring pcm down, the problem was gone. Anyway, I just said I saw "something similar" :-) . > Greg /Niels Chr. -- Niels Christian Bank-Pedersen, NCB1-RIPE. Network Manager, Tele Danmark NET, IP-section. "Hey, are any of you guys out there actually *using* RFC 2549?" To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-mobile" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19991207163417.B27105>