Date: Sun, 22 May 2005 16:45:04 -0700 From: Sam Leffler <sam@errno.com> To: Nate Lawson <nate@root.org> Cc: cvs-src@FreeBSD.org, Damien Bergamini <damien.bergamini@free.fr>, src-committers@FreeBSD.org, Damien Bergamini <damien@FreeBSD.org>, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/dev/ipw if_ipw.c Message-ID: <42911980.8020607@errno.com> In-Reply-To: <4291171E.1030307@root.org> References: <200505221831.j4MIV8lq072154@repoman.freebsd.org> <4290D2ED.6080602@errno.com> <006a01c55f01$46851520$0100a8c0@bigbang> <4291171E.1030307@root.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Nate Lawson wrote: > Damien Bergamini wrote: > >>> Why? A good freebsd commit msg explains why you make a change. >> >> >> >> Basically because the Linux driver does it and because some users >> reported that this fixes issues with ACPI. Remember that I have >> zero documentation for this chipset except the Linux driver itself >> which is very poorly commented. And Intel refuses to answer to >> any technical question regarding the hardware. > > > It's not an ACPI issue. You would have the problem with any > suspend/resume (i.e., APM). > I think he just meant "issues seen when ACPI is enabled". It appears this change originated in one of the Linux ethernet drivers (can't recall) and has been blindly propagated to other drivers because it works. I applied it when certain 5211 parts started to get PCI bus dma abort errors. Disabling retries stopped it but I was never able to get a logic analyzer in place to see exactly what was happening (this was a cardbus card). Sam
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?42911980.8020607>