Date: Fri, 19 Mar 1999 13:15:03 -0500 (EST) From: Andrew Gallatin <gallatin@cs.duke.edu> To: Matthew Dillon <dillon@apollo.backplane.com> Cc: Bill Paul <wpaul@skynet.ctr.columbia.edu>, hackers@FreeBSD.ORG Subject: Re: Gigabit ethernet revisited Message-ID: <14066.37142.414291.953962@grasshopper.cs.duke.edu> In-Reply-To: <199903191739.JAA59302@apollo.backplane.com> References: <199903191415.JAA03718@skynet.ctr.columbia.edu> <199903191739.JAA59302@apollo.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Matthew Dillon writes: > Hmm. I definitely think there's a bug due to not repopulating buffers > from inside your receive interrupt packet processing loop, but it may > not be the cause of this bug. Insufficient Rx buffer descriptors isn't the cause of this. If it were, we'd be seeing the firmware complaining about nicNoMoreRxBDs, rather than nicDmaWriteRingFull. > If the NIC is unable to complete DMA quickly enough, perhaps the burst > parameters can be tweaked. The PC certainly should be able to do DMA > writes to memory at the PCI bus speed! I agree that the pci tuning parameters are the best candidates for optimization. Drew ------------------------------------------------------------------------------ Andrew Gallatin, Sr Systems Programmer http://www.cs.duke.edu/~gallatin Duke University Email: gallatin@cs.duke.edu Department of Computer Science Phone: (919) 660-6590 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?14066.37142.414291.953962>