Date: Mon, 3 Nov 1997 17:47:47 -0800 From: Don Lewis <Don.Lewis@tsc.tdk.com> To: Simon Shapiro <Shimon@i-Connect.Net>, Don Lewis <Don.Lewis@tsc.tdk.com> Cc: freebsd-hackers@FreeBSD.ORG, Jaye Mathisen <mrcpu@cdsnet.net>, Matt Thomas <matt@3am-software.com>, Charles Henrich <henrich@crh.cl.msu.edu> Subject: Re: de0 errors Message-ID: <199711040147.RAA10391@salsa.gv.tsc.tdk.com> In-Reply-To: Simon Shapiro <Shimon@i-Connect.Net> "Re: de0 errors" (Nov 3, 5:15pm)
next in thread | previous in thread | raw e-mail | index | archive | help
On Nov 3, 5:15pm, Simon Shapiro wrote: } Subject: Re: de0 errors } } Hi Don Lewis; On 04-Nov-97 you wrote: } } ... } } > } Humm, tcpblast shows 4.0MB/sec. Granted this isnt an empty net } > either, fairly } > } busy I would imagine. } > } > That's probably a reasonable number for a 10Mb network, but not a } > 100Mb network. BTW, I bet this NIC is also hurting your disk I/O } > throughput since it is hogging the PCI bus because it's using an } > inefficient transfer method. } } I must be missing something. 4MB/sec. on TCP/IP over Ethernet is on a good } day, more than 40MHz. No? You're not missing anything, I slipped a decimal point. Sigh ... BTW, I just did some quick calculations and a 21140 + 440FX might still be able to drive the network at full speed without running out of PCI bandwidth. It's just that if you're transmitting, you'll consume about 4x the transmit bandwidth on the PCI bus. If you're able to send at 10MB/sec, then you'll consume about 40MB/sec on the PCI bus out of the theoretical 132MB/sec. --- Truck
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199711040147.RAA10391>