Date: Fri, 16 May 2003 11:54:47 -0300 From: "Daniel C. Sobral" <dcs@tcoip.com.br> To: Robert Watson <rwatson@FreeBSD.org> Cc: cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/dev/fxp if_fxp.c Message-ID: <3EC4FBB7.6060201@tcoip.com.br> In-Reply-To: <200305160113.h4G1DHqN090638@repoman.freebsd.org> References: <200305160113.h4G1DHqN090638@repoman.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Robert Watson wrote: > rwatson 2003/05/15 18:13:17 PDT >=20 > FreeBSD src repository >=20 > Modified files: > sys/dev/fxp if_fxp.c=20 > Log: > Add a tunable/sysctl "hw.fxp_noflow" which disables flow control supp= ort > on if_fxp cards. When flow control is enabled, if the operating syst= em > doesn't acknowledge the packet buffer filling, the card will begin to= > generate ethernet quench packets, but appears to get into a feedback > loop of some sort, hosing local switches. This is a temporary workar= ound > for 5.1: the ability to configure flow control should probably be > exposed by some or another management interface on ethernet link laye= r > devices. Looks like a mediaopt candidate to me. Though mediaopt is usually only=20 used for half/full-dulpex, there's the "link0" specification on some=20 cards that does something different. Alas, all the interface "capabilities" ought to be tweakable in some=20 standard fashion. For instance, why are special words for txcsum and=20 txcsum, when there's also vlan-tagging support and other capabilities=20 that could be set/unset from a single place? I'm in favor of mediaopt/-mediaopt, but... <shrug> --=20 Daniel C. Sobral Ger=EAncia de Opera=E7=F5es Divis=E3o de Comunica=E7=E3o de Dados Coordena=E7=E3o de Seguran=E7a VIVO Centro Oeste Norte Fones: 55-61-313-7654/Cel: 55-61-9618-0904 E-mail: Daniel.Capo@tco.net.br Daniel.Sobral@tcoip.com.br dcs@tcoip.com.br
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3EC4FBB7.6060201>