Date: Thu, 3 May 2007 20:26:24 +0800 From: Ariff Abdullah <ariff@FreeBSD.org> To: Peter Jeremy <peterjeremy@optushome.com.au> Cc: freebsd-amd64@freebsd.org Subject: Re: Turion 64x2 still hanging after C1E patch Message-ID: <20070503202624.0e0966cd.ariff@FreeBSD.org> In-Reply-To: <20070503114239.GA4823@turion.vk2pj.dyndns.org> References: <200704251958.l3PJwgfZ044842@repoman.freebsd.org> <462FEB77.3000103@FreeBSD.org> <20070426101525.02bd830e.ariff@FreeBSD.org> <20070503114239.GA4823@turion.vk2pj.dyndns.org>
index | next in thread | previous in thread | raw e-mail
[-- Attachment #1 --]
On Thu, 3 May 2007 21:42:39 +1000
Peter Jeremy <peterjeremy@optushome.com.au> wrote:
> On 2007-Apr-26 10:15:25 +0800, Ariff Abdullah <ariff@freebsd.org>
> wrote:
> >On Wed, 25 Apr 2007 16:59:51 -0700
> >Maxim Sobolev <sobomax@FreeBSD.org> wrote:
> >> Ariff Abdullah wrote:
> >> > ariff 2007-04-25 19:58:42 UTC
> >> > Disable C1 Enhanced mode on AMD K8 Family Revision F and
> >above > > to keep local APIC timer alive.
> ...
> >> Is it expected to fix some Turion laptops out there, that were
> >> previously unable to run in the SMP mode due to LAPIC timer on
> >one > core being down on boot up?
> >
> >Certainly yes.
>
> My HP/Compaq V6107 still goes into a coma even with this patch. If
> I boot to single-user mode, it seems to happen immediately after the
> CD-ROM probe messages (which would be when the system becomes idle).
> Otherwise, it goes comatose whilst trying to load the entropy.
>
This probably something else. Try disabling sio, ppc through
/boot/device.hints. Several HP laptops are known to hang because of
this.
> I've re-applied idlecpu_apic_5.diff and it seems to work with
> hw.apic.idle_cpu="1" (but see below). I've hacked local_apic.c
> to save the value read from MSR_C001_0055 and that shows that
> C1eOnCmpHalt is set.
>
This is expected, and the recent "fix" is supposed to overwrite and
disable C1E on each core/cpu.
> Worse,I've noticed that the RTC interrupt occasionally stops. I
> have worked out that I can restart it by writing to sysctl
> machdep.adjkerntz (which was the easiest way to issue
> rtcin(RTC_INTR) from userland). I presume that occasionally the RTC
> interrupt is lost and so never reset (as per the comment above
> rtcintr()).
>
> I haven't yet tried using the i8254 as a wakeup source.
>
> Any ideas on where to look next?
>
Show us your processor Id from dmesg.
--
Ariff Abdullah
FreeBSD
... Recording in stereo is obviously too advanced
and confusing for us idiot ***** users :P ........
[-- Attachment #2 --]
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (FreeBSD)
iD8DBQFGOdTwlr+deMUwTNoRAoP8AKDTRYXz2ApUeYWLJFQn7uoafJNWtwCbB0Rq
tcKJX+w9y8nfd+j5AeMhgmI=
=Z6eE
-----END PGP SIGNATURE-----
home |
help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070503202624.0e0966cd.ariff>