Date: Fri, 29 Oct 2004 09:59:40 +0200 From: =?ISO-8859-1?Q?S=F8ren_Schmidt?= <sos@DeepCore.dk> To: Jake Khuon <khuon@NEEBU.Net> Cc: freebsd-current@freebsd.org Subject: Re: broken APM on IBM-T30 with 6.0-CURRENT Message-ID: <4181F86C.6050807@DeepCore.dk> In-Reply-To: <200410290736.i9T7aIia017214@Espresso.NEEBU.Net> References: <200410290736.i9T7aIia017214@Espresso.NEEBU.Net>
next in thread | previous in thread | raw e-mail | index | archive | help
Jake Khuon wrote:
> ### On Thu, 28 Oct 2004 17:38:52 -0700, Bruce M Simpson <bms@spc.org>
> ### casually decided to expound upon Jake Khuon <khuon@NEEBU.Net> the
> ### following thoughts about "Re: broken APM on IBM-T30 with 6.0-CURREN=
T":
>=20
> BMS> On Thu, Oct 28, 2004 at 05:33:03PM -0700, Jake Khuon wrote:
> BMS> > That patch worked perfectly. I had to do a little hand-patching =
of one of
> BMS> > the hunks that got rejected though. Any thoughts on if this wil=
l get
> BMS> > merged? Thanks.
> BMS>=20
> BMS> Please send me these patches so I can look at them...
>=20
> After applying the patch, I did a backdiff. This is against:
>=20
> /src/sys/dev/ata/ata-all.c,v 1.233 2004/10/19 20:13:38
> ***************
> *** 346,356 ****
> --- 347,359 ----
> if (ch->state =3D=3D ATA_IDLE) {
> ch->state =3D ATA_ACTIVE;
> mtx_unlock(&ch->state_mtx);
> + goto out;
> break;
> }
> mtx_unlock(&ch->state_mtx);
> tsleep(ch, PRIBIO, "atasusp", hz/10);
> }
> + out: =20
> ch->locking(ch, ATA_LF_UNLOCK);
> return 0;
> }
That part cant be needed right ?
As for the rest, it works around the problem that tsleep doesn't seem to =
work on the resume path under APM. Now that is all nice and that, but=20
I'd like to get the real problem fixed: why doesn't tsleep work there ?
--=20
-S=F8ren
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4181F86C.6050807>