Date: Wed, 16 Apr 2003 13:29:51 +0200 From: Alexander Leidinger <Alexander@Leidinger.net> To: beaker@hot.pl.?= (Krzysztof =?ISO-8859-1?Q?J=EAdruczyk) Cc: current@freebsd.org Subject: Re: Adjusting pcm buffersize? Message-ID: <20030416132951.2d855e44.Alexander@Leidinger.net> In-Reply-To: <usmsjaa1q.fsf@iavision.com> References: <3E9BCC5F.4050603@ezri.org> <001d01c30356$9bba6260$0200000a@fireball> <3E9C1928.60206@ezri.org> <usmsjaa1q.fsf@iavision.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 15 Apr 2003 20:29:53 +0200
beaker@hot.pl (Krzysztof J=EAdruczyk) wrote:
> And let me just note, that it is not ATA's fault (I believe it was
> suggested before that the problem is connected to using PIO ATA
> modes). I have two U160 SCSI drives connected to Adaptec 29160LP
> controller + SB Live card and also suffer badly from this problem when
> for example mozilla's port is unbzipping sources...
AFAIK there are three possibilities for the problem:
- The problem is "caused" by the restructuring of the kernel because
of the SMPng work. When the kernel is completely preemptive and the
locking of the subsystems is done this problem should vanish.
- The problem is caused by some pieces in the kernel which block
interrupts for too long (this will get addressed when SMPng runs
stable).
- The problem is because of both of above...
At least this is how I had understand the problem last time it was
analyzed.
Bye,
Alexander.
--=20
Secret hacker rule #11: hackers read manuals.
http://www.Leidinger.net Alexander @ Leidinger.net
GPG fingerprint =3D C518 BC70 E67F 143F BE91 3365 79E2 9C60 B006 3FE7
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030416132951.2d855e44.Alexander>