From owner-freebsd-mobile Tue Jul 23 14:40:36 2002 Delivered-To: freebsd-mobile@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id A24BC37B400 for ; Tue, 23 Jul 2002 14:40:32 -0700 (PDT) Received: from harmony.village.org (rover.bsdimp.com [204.144.255.66]) by mx1.FreeBSD.org (Postfix) with ESMTP id B50FB43E31 for ; Tue, 23 Jul 2002 14:40:31 -0700 (PDT) (envelope-from imp@bsdimp.com) Received: from localhost (warner@rover2.village.org [10.0.0.1]) by harmony.village.org (8.12.3/8.12.3) with ESMTP id g6NLeUYx012682; Tue, 23 Jul 2002 15:40:30 -0600 (MDT) (envelope-from imp@bsdimp.com) Date: Tue, 23 Jul 2002 15:39:53 -0600 (MDT) Message-Id: <20020723.153953.76074374.imp@bsdimp.com> To: tony@valemount.com Cc: mobile@FreeBSD.ORG, andrew@unfortu.net Subject: Re: PCI -> PCMCIA Adapter woes - patch From: "M. Warner Losh" In-Reply-To: <100401c23290$cf2568d0$114c35d1@tonyxp> References: <0f5001c23287$1a948530$114c35d1@tonyxp> <20020723.150649.12657684.imp@bsdimp.com> <100401c23290$cf2568d0$114c35d1@tonyxp> X-Mailer: Mew version 2.1 on Emacs 21.1 / Mule 5.0 (SAKAKI) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-mobile@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org In message: <100401c23290$cf2568d0$114c35d1@tonyxp> "Tony Toole" writes: : I don't have the docs you mentioned, however from reviewing various other : pieces of information (such as the Linux PCMCIA implementation), this is : needed. What harm would it do to a card that either doesn't support it : (like the 11xx), or ones that already have it working such as (probably) : Laptops? The cards we use are from Elan, however we have several smaller : SBC units with integrated TI1211 chips that also require this workaround. : Perhaps a reasonable thing to do is provide a sysctl to enable / disable the : fix on bootup? The harm on laptops would be if they are wired in a weird way. The harm for TI-113[01] parts would be that they might behave in unknown or unpreditable ways if we did this. Sure, we could test it, but we couldn't be sure. IIRC, the only way to get the PCI signals is to use the mutlifunction pins to get them. I think it would be safest to do the fixup iff it is 0 to start with AND the user doesn't say not to do it. The Linux pcmica code isn't always as careful with these things as it should be. There have been a number of occasions where workarounds/hacks in that code are later revised because they were too broadly applied in the initial version(s). So you have to be careful how you take workarounds from that code base. I'll try to work up a better fix that I'd be more comfortable with and post it later. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-mobile" in the body of the message