lkml.org 
[lkml]   [2002]   [May]   [7]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [BK PATCH] PCI reorg changes for 2.5.14

On Tue, 7 May 2002, Linus Torvalds wrote:

>
> On Tue, 7 May 2002, Patrick Mochel wrote:
> >
> > Actually, there was an ordering problem, which was causing an oops on
> > boot. But, that doesn't help with IRQ routing.
> >
> > The problem is that ACPI IRQ routing doesn't work at all in 2.5.14 if you
> > have support for APICs enabled in any way.
>
> Hmm.. That may be true, but at least 2.5.14 gets the interrupt routing
> right.
>
> Which may be because it knows to fall back on the MP table parsing if it
> can't work out the ACPI stuff, and you probably broke that part.
>
> The whole notion of having just _one_ PCI interrupt routing function is
> definitely _broken_. The rule in 2.5.14 is "try ACPI if it's enabled, and
> if that works, we're fine. If it doesn't work, let's try the legacy
> stuff".
>
> In contrast, your PCI irq changes seem to say "if we found ACPI tables, we
> use ACPI routing", which is just stupid.

Oh right, the check for valid ACPI routing tables is _supposed_ to be
there...

> Please fix it to work at _least_ as well as 2.5.14 does (and yes, my home
> machine works fine with APIC and ACPI enabled on 2.5.14, and since it's a
> dual P4 HT machine it _cannot_ work any other way - and the PCI BK changes
> break it to the point of not booting).

The check is back in there now, and it works here, so IRQ routing works on
2.5.14 with an SMP kernel. Greg has verified that it also works on a P-IV
SMP system.

You can pull from bk://ldm.bkbits.net/linux-2.5-pci/

-pat

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:25    [W:0.032 / U:20.352 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site