Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 13 Apr 2005 10:10:50 +0200
From:      Divacky Roman <xdivac02@stud.fit.vutbr.cz>
To:        John Baldwin <jhb@freebsd.org>
Cc:        current@freebsd.org
Subject:   Re: slow kbd input on 6-current on amd64@nforce3
Message-ID:  <20050413081050.GA76859@stud.fit.vutbr.cz>
In-Reply-To: <200504121253.50950.jhb@FreeBSD.org>
References:  <20050406130909.GA90294@stud.fit.vutbr.cz> <200504112038.32964.jhb@FreeBSD.org> <20050412073810.GA89527@stud.fit.vutbr.cz> <200504121253.50950.jhb@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Apr 12, 2005 at 12:53:50PM -0400, John Baldwin wrote:
> On Tuesday 12 April 2005 03:38 am, Divacky Roman wrote:
> > On Mon, Apr 11, 2005 at 08:38:32PM -0400, John Baldwin wrote:
> > > On Wednesday 06 April 2005 09:09 am, Divacky Roman wrote:
> > > > as I have mentioned on the list I have very slow keyboard input. it
> > > > might be related to kbd not having an IRQ assigned. I repeat once again
> > > > that it worked on 5.3R.
> > >
> > > Actually, now that I look at this, you have a buggy BIOS.  It is lying
> > > and claiming that some PCI interrupts are active-hi rather than
> > > active-low.  Hmm, the 5.3 dmesg you gave me included APIC, while this one
> > > does not.  Does disabling ACPI make your keyboard happy on 6.0 by chance?
> >
> > It doesnt boot with acpi enabled (stops in probing ata devices, but it
> > never worked so I think ata is not the only culprit)
> 
> Ok.
> 
> > what can I do with it? would some quirk made the trick? why it worked in
> > 5.3R?
> 
> I don't know at this point.  Does 6.0 in any configuration work ok?  
> (ACPI !APIC, ACPI APIC, !ACPI APIC, !ACPI !APIC)

it doesnt boot with ACPI which eliminates first two cases, and I tried all
combinations of
# Workarounds for some known-to-be-broken chipsets (nVidia nForce3-Pro150)
device          atpic                   # 8259A compatability
options         NO_MIXED_MODE           # Don't penalize working chipsets

and none worked... ie. I think it eliminates later two cases


you said the logic of active interrupts is reverted. where in the code can I
revert this? I'll try it and if it works I can define some quirk...

thnx for answer

roman



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050413081050.GA76859>