Date: Thu, 21 Jul 2005 09:18:44 -0600 (MDT) From: "M. Warner Losh" <imp@bsdimp.com> To: ache@FreeBSD.ORG Cc: marcolz@stack.nl, wpaul@ctr.columbia.edu, current@FreeBSD.ORG Subject: Re: rl(4) is not ready for mpsafenet net enough? (silent reboots) Message-ID: <20050721.091844.10574798.imp@bsdimp.com> In-Reply-To: <20050721143443.GB10010@nagual.pp.ru> References: <20050721134816.GA8550@nagual.pp.ru> <20050721142445.GA77847@stack.nl> <20050721143443.GB10010@nagual.pp.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
In message: <20050721143443.GB10010@nagual.pp.ru>
Andrey Chernov <ache@FreeBSD.ORG> writes:
: On Thu, Jul 21, 2005 at 04:24:45PM +0200, Marc Olzheim wrote:
: > I thought rl cards were inherently MP unsafe because of design
: > flaws. Does anyone know it they can be used safely nowadays anyway ?
:
: This is not something about MP exactly, but about internal logic flaws. As
: I wrote, I have _UP_ machine.
The problem isn't that they are inherently MP unsafe (that's a
*DRIVER* concept, not a hardware one in this context). The problem
with REV C and earlier is that their performance sucks because the DMA
engine used in the cards was LAME. Realtec fixed this in newer
revisions of the chip, and the re driver is able to take advantage of
that (as well as support the newer gige chips).
Warner
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050721.091844.10574798.imp>