Date: Mon, 8 Mar 2004 12:37:39 +0100 From: Brad Knowles <brad.knowles@skynet.be> To: Matthew Dillon <dillon@apollo.backplane.com> Cc: freebsd-current@freebsd.org Subject: Re: patch for Linux NFS client Message-ID: <p06002053bc720b2505cf@[10.0.1.4]> In-Reply-To: <200403080825.i288Pg6B017066@apollo.backplane.com> References: <40446EF2.5020901@alumni.rice.edu> <200403080825.i288Pg6B017066@apollo.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
At 12:25 AM -0800 2004/03/08, Matthew Dillon wrote:
> Even so, it might be a good idea for us to use a significantly larger
> soreserve value, or to increase the buffer limit when a large data
> block size is negotiated. Instead of adding a slop of 2048
> (aka 32768 + 404 + 2048 = 35220 bytes) it might be better to set
> the soreserve value to 65535 by default.
Out of curiosity, is it possible to set this value to be twice
the negotiated block size?
> Generally speaking the TCP buffer ought to be large enough to buffer
> at least two full-sized NFS data packets to reduce NFSD interlock
> stalls when combined with read-ahead.
That's exactly what I was thinking. Hence my question.
--
Brad Knowles, <brad.knowles@skynet.be>
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
-Benjamin Franklin, Historical Review of Pennsylvania.
GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI++++$ P+>++ L+ !E-(---) W+++(--) N+
!w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++)
tv+(+++) b+(++++) DI+(++++) D+(++) G+(++++) e++>++++ h--- r---(+++)* z(+++)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?p06002053bc720b2505cf>