Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 11 Oct 2004 17:51:19 -0400 (EDT)
From:      Robert Watson <rwatson@freebsd.org>
To:        Daniel Eriksson <daniel_k_eriksson@telia.com>
Cc:        freebsd-current@freebsd.org
Subject:   Re: Giant-free ahc?
Message-ID:  <Pine.NEB.3.96L.1041011174937.55701D-100000@fledge.watson.org>
In-Reply-To: <!~!UENERkVCMDkAAQACAAAAAAAAAAAAAAAAABgAAAAAAAAA0VcX9IoJqUaXPS8MjT1PdsKAAAAQAAAASgCTacl2b0SDEgGCZoLcgAEAAAAA@telia.com>

next in thread | previous in thread | raw e-mail | index | archive | help

On Mon, 11 Oct 2004, Daniel Eriksson wrote:

> Is anyone working on fine-grained locking for the ahc driver? 
> 
> I just upgraded one of my servers from UP to SMP, and noticed that
> *Giant showed up a lot when running top for processes accessing data on
> the SCSI discs. It could be a GIANT lock for some other resource of
> course, but since the ahc driver reports [GIANT-LOCKED] during boot, and
> since all the processes were accessing data on SCSI discs I just assumed
> it was waiting for a lock in the ahc driver. 

It could well be that there's a lot of contention on Giant for VFS and
disk I/O operations; however, top is unfortunately not a very good tool
for diagnosing that, as the services top uses to monitor the system
require Giant, increasing the chances of Giant contention being observed
by top.  Patches to reduce the use of Giant in sysctl() were recently
posted to a mailing list (arch@?) and it could well be that those patches
help reduce the Giant-related noise that top sees.

Robert N M Watson             FreeBSD Core Team, TrustedBSD Projects
robert@fledge.watson.org      Principal Research Scientist, McAfee Research



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.NEB.3.96L.1041011174937.55701D-100000>