Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 27 Apr 2004 01:21:59 +0900 (JST)
From:      Takayoshi Kochi <kochi@netbsd.org>
To:        len.brown@intel.com
Cc:        kanaoka@netbsd.org
Subject:   Re: ACPI SCI flags
Message-ID:  <20040427.012159.74740435.kochi@netbsd.org>
In-Reply-To: <29AC424F54821A4FB5D7CBE081922E402FE8C6@hdsmsx403.hd.intel.com>
References:  <29AC424F54821A4FB5D7CBE081922E402FE8C6@hdsmsx403.hd.intel.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Hi,

From: "Brown, Len" <len.brown@intel.com>
Subject: RE: ACPI SCI flags
Date: Sun, 25 Apr 2004 21:44:56 -0400

> >> > PIC mode:
> >> > 	force level/low via ELCR always.
> >> 
> >> Did you find that made a difference on some systems?  If so, 
> >which ones?
> >> The opinion here was that ELCR is probably EISA-only.
> >
> >Kanaoka-san reported that if SCI is not set level-trigger,
> >the SCI doesn't work properly on his laptop.
> >It (Libretto L3) has ALi chipsets (1543 south bridge).
> >
> >When I tested on my laptop (ThinkPad X31, Intel ICH4),
> >it seemed that the setting of ELCR didn't make any difference, though.
> 
> You mean it doesn't work in either case?
> Testing it via power button?

It DID work in both cases.

> Months ago when Linux did the PIC-mode level/low thing, I had
> A bunch of users test it, and for many the hardware worked
> correctly no matter if ELCR was set to level or edge.
> 
> However, there were others which did care, and only worked
> if level triggered.  Most of these had a BIOS that had already
> set the ELCR to level, but some required the OS to do it.

Thanks for the long explanation.
I agree that setting ELCR makes sense.

---
Takayoshi Kochi



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040427.012159.74740435.kochi>