Date: Fri, 22 Mar 2002 08:41:44 -0800 From: "Sam Leffler" <sam@errno.com> To: <sos@freebsd.dk>, "Ian Dowse" <iedowse@maths.tcd.ie> Cc: "Jon Larssen" <jonlarssen@hotmail.com>, <freebsd-stable@FreeBSD.ORG>, <sos@FreeBSD.ORG> Subject: Re: Is ATA partially broken in -STABLE? Message-ID: <407801c1d1c0$74867760$52557f42@errno.com> References: <200203221248.g2MCmib27911@freebsd.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
Actually, two races were reintroduced--one for disks and one for cdroms.
Please see my note with a tested patch.
Sam
----- Original Message -----
From: "Søren Schmidt" <sos@freebsd.dk>
To: "Ian Dowse" <iedowse@maths.tcd.ie>
Cc: "Jon Larssen" <jonlarssen@hotmail.com>; <freebsd-stable@FreeBSD.ORG>;
<sos@FreeBSD.ORG>
Sent: Friday, March 22, 2002 4:48 AM
Subject: Re: Is ATA partially broken in -STABLE?
> It seems Ian Dowse wrote:
> > In message <200203220933.g2M9XGT90073@freebsd.dk>, Søren Schmidt writes:
> > >Its a known problem, the old way of solving it was:
> >
> > Ok - is there a good reason for not using spl calls to block the
> > interrupt until the tsleep(), or is it just an oversight? I guess
> > maybe you are trying to avoid the delay associated with deferring
> > the interrupt, but it must open up a number of races, especially
> > if the ATA interrupt line is shared with something else.
>
> The driver used asleep to avoid the race before the MFC, one of
> those got lost in the MFC, sorry...
>
> -Søren
>
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with "unsubscribe freebsd-stable" in the body of the message
>
>
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?407801c1d1c0$74867760$52557f42>