linux-mips
[Top] [All Lists]

Re: [PATCH] Import updates from i386's i8259.c

To: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Subject: Re: [PATCH] Import updates from i386's i8259.c
From: "Maciej W. Rozycki" <macro@linux-mips.org>
Date: Wed, 6 Dec 2006 15:43:12 +0000 (GMT)
Cc: Ingo Molnar <mingo@redhat.com>, anemo@mba.sphere.ne.jp, Ralf Baechle <ralf@linux-mips.org>, linux-mips@linux-mips.org
In-reply-to: <4576DDEC.1050105@ru.mvista.com>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <20061206.103923.71086192.nemoto@toshiba-tops.co.jp> <20061206015818.GB27985@linux-mips.org> <20061206.115602.63741871.nemoto@toshiba-tops.co.jp> <20061206.133836.89067271.nemoto@toshiba-tops.co.jp> <4576C2E9.4060900@ru.mvista.com> <Pine.LNX.4.64N.0612061337220.29000@blysk.ds.pg.gda.pl> <4576CB64.2060705@ru.mvista.com> <Pine.LNX.4.64N.0612061438440.29000@blysk.ds.pg.gda.pl> <4576DDEC.1050105@ru.mvista.com>
Sender: linux-mips-bounce@linux-mips.org
On Wed, 6 Dec 2006, Sergei Shtylyov wrote:

>    I'd say they *only* looked consistent then. :-)

 Some user tools may rely on certain strings seen in /proc/interrupts 
("procinfo"?).

>    As for the I/O APIC, I think i82903AA was "the reference design" for it...

 Nope, it was the i82489DX -- the original "discrete" coupled Local & I/O 
APIC using a five-wire inter-APIC bus and a protocol different from later 
implementations.  Then there were ones included in the i82379AB (SIO.A) 
and i82374EB/SB (ESC) chipset components.  They used a three-wire bus and 
a new protocol.  And only then came the i82093AA.

  Maciej

<Prev in Thread] Current Thread [Next in Thread>