Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 3 Nov 2003 21:30:25 -0800 (PST)
From:      Doug White <dwhite@gumbysoft.com>
To:        Scott Long <scottl@freebsd.org>
Cc:        current@freebsd.org
Subject:   Re: HEADSUP: Committing new interrupt code, tree will be broken 
Message-ID:  <20031103212841.A90051@carver.gumbysoft.com>
In-Reply-To: <20031103203242.A34233@pooker.samsco.home>
References:  <20031104012530.4A8D82A8EB@canning.wemm.org> <20031103203242.A34233@pooker.samsco.home>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 3 Nov 2003, Scott Long wrote:

> > My Dell PE1750 hangs up solid under network load.
> >
> > Its a serverworks of some recent variation, 2x2.4GHz xeons. BIOS A05. bge
> > ethernet x2.  Apachebench will mash it dead in a few seconds.
> >
>
> Sam just discovered some locking problems in the IPv4 stack.  Can you
> update to his recent commit and retry?  With the new interrupt routing
> code, you can either get your interrupts or you can't.  High load
> should not be affected by it.

(for the record)

Tried turning off mpsafenet, no luck.  Will try disabling ACPI tomorrow.

I'm hoping its not some silly bge bug that happens to manifest only in
-current. I ran some extensive load testing with this and another
single-proc box on 4.9 without any issues, but thats obviously not
directly comparable.

-- 
Doug White                    |  FreeBSD: The Power to Serve
dwhite@gumbysoft.com          |  www.FreeBSD.org



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20031103212841.A90051>