Date: Mon, 7 May 2012 11:40:30 -0700 From: Adrian Chadd <adrian@freebsd.org> To: Andrey Zonov <andrey@zonov.org> Cc: pyunyh@gmail.com, freebsd-net@freebsd.org, davidch@freebsd.org Subject: Re: bce: jumbo not working since r218423 Message-ID: <CAJ-VmonY9WPcE5zVQUD2RG4BOFTD0NFF682TJ=VGDzxFfe3Mgg@mail.gmail.com> In-Reply-To: <4FA6E03B.2000706@zonov.org> References: <CANU_PUGwoLSrPcGE8wT=ga3-=F_n9qN4pPXMJC%2BH72wpS9Mfcw@mail.gmail.com> <20120427230400.GB17009@michelle.cdnetworks.com> <4FA6E03B.2000706@zonov.org>
next in thread | previous in thread | raw e-mail | index | archive | help
.. I think you're misunderstanding how PMTU discovery works. The _point_ is that you refuse to send said frame in the first place. Any number of intermediary L2 devices may decide to drop your jumbo TX'ed frame and not generate an ICMP error message, thus breaking PMTU discovery anyway. In any case, someone from broadcom piped up and pointed out that at least for their driver, the "force strict RX MTU" is for compliance testing. Adrian
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAJ-VmonY9WPcE5zVQUD2RG4BOFTD0NFF682TJ=VGDzxFfe3Mgg>