From: erik quanstrom <quanstro@quanstro.net>
To: 9fans@9fans.net
Subject: [9fans] ethernet queuing question
Date: Wed, 21 Apr 2010 09:33:39 -0400 [thread overview]
Message-ID: <6e88fbdb245882607a789097b4ca537c@kw.quanstro.net> (raw)
suppose you have x bytes of memory to devote
to ethernet controller buffers. (ether in preallocated
buffers or the main Pool.) and suppose you're
running an ip stack on this controller. is there any way
to insure that
qlen(ether->iq) + sum for all Conv
qlen(conv->iq) < f*x
for f <= 1.
the reason for the question is that i've noticed that
even for a reasonable number of users, it's pretty easy
to starve the ethernet controller or panic cause there's
no kernel memory to allocate. (raise your hand if you've
seen the i82563 no buffers message.) i've recently doubled
the number of buffers, halved the ethernet queue size,
and taken other steps, but more users has already overrun
this gain.
- erik
reply other threads:[~2010-04-21 13:33 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=6e88fbdb245882607a789097b4ca537c@kw.quanstro.net \
--to=quanstro@quanstro.net \
--cc=9fans@9fans.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).