Date: Mon, 6 Feb 2006 23:50:09 +0300 From: Gleb Smirnoff <glebius@FreeBSD.org> To: Paul Saab <ps@FreeBSD.org> Cc: Oleg Bulyzhin <oleg@FreeBSD.org>, src-committers@FreeBSD.org, Alan Cox <alc@cs.rice.edu>, cvs-all@FreeBSD.org, cvs-src@FreeBSD.org Subject: Re: cvs commit: src/sys/dev/bge if_bge.c Message-ID: <20060206205009.GB877@FreeBSD.org> In-Reply-To: <43E77338.5070404@freebsd.org> References: <200602020958.k129wWtc066930@repoman.freebsd.org> <20060202100637.GB24350@lath.rinet.ru> <20060205235817.GQ5499@cs.rice.edu> <20060206081519.GL60276@FreeBSD.org> <20060206082645.GS5499@cs.rice.edu> <20060206083112.GM60276@FreeBSD.org> <20060206105328.GA27193@cell.sick.ru> <43E77338.5070404@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Paul, On Mon, Feb 06, 2006 at 08:03:04AM -0800, Paul Saab wrote: P> revision 1.17 P> date: 2002/07/04 23:19:20; author: jdp; state: Exp; lines: +1 -1 P> Disable hardware checksum offloading of IP fragments in the bge P> driver. I tried a few obvious experiments, but was unable to make P> the 3c996B-T generate correct UDP checksums for transmitted fragmented P> packets. I'm not so sure the device is even capable of it. P> P> This fixes NFS over UDP. This change is correct, since spec says that chip can not do checksum calculation on transmitted UDP fragments. This change is not backed out in HEAD. The problem we observe now is broken checksum calculation on incoming UDP fragments that appeared to be broken after we turned on pseudo header calculation in the chip. -- Totus tuus, Glebius. GLEBIUS-RIPN GLEB-RIPE
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060206205009.GB877>