From owner-freebsd-hardware@FreeBSD.ORG Thu Aug 29 15:50:22 2013 Return-Path: Delivered-To: freebsd-hardware@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTP id 2A4ADDE8 for ; Thu, 29 Aug 2013 15:50:22 +0000 (UTC) (envelope-from imp@bsdimp.com) Received: from mail-oa0-f51.google.com (mail-oa0-f51.google.com [209.85.219.51]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by mx1.freebsd.org (Postfix) with ESMTPS id E46252267 for ; Thu, 29 Aug 2013 15:50:21 +0000 (UTC) Received: by mail-oa0-f51.google.com with SMTP id h1so794171oag.24 for ; Thu, 29 Aug 2013 08:50:15 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:sender:subject:mime-version:content-type:from :in-reply-to:date:cc:content-transfer-encoding:message-id:references :to; bh=WufppyoDJ8zP1DWHkXsFqxaCpMqK9Zlbk5hkFpuDsJQ=; b=NV63x8pNipXp1n7tJI8NRdbz84ZLAWU1poYcietxVqcuebxx+5V5onM2BcJZNQRkXV UnrS2Rvy7ScmJhn33eI+BmSsqxCcwfA3XNgFRXXThJ3P68VGd8LP1sefgpNhikak9lOK hNzy6rQJnCWQEcbKYcOS6H0uR2Rwea+IzMhExkJkNB7/g3gDYp8RcUpsQ6prm3HE3Mns V4E+BnJfwpvCxcgvOzXikqbn3OHoavRURqYHQ2kpK462/HYU7eYSmgkx7anK3VMZnAlY p3x+7573NdDzVBR/swyCLE3I0qBbY9ZQ7/kwS+qcv1U72fyHjA1AptP3cD9MqFQX0KWd 5sOg== X-Gm-Message-State: ALoCoQm6MpVyODD3bRxg1CCfOaCooeGmlP2APDDroKzoA/yJWXRWmhJyruIFvY+fhL/VkbLtqzak X-Received: by 10.182.48.194 with SMTP id o2mr2259927obn.90.1377791415343; Thu, 29 Aug 2013 08:50:15 -0700 (PDT) Received: from monkey-bot.int.fusionio.com ([209.117.142.2]) by mx.google.com with ESMTPSA id r3sm33040857oep.2.1969.12.31.16.00.00 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Thu, 29 Aug 2013 08:50:14 -0700 (PDT) Sender: Warner Losh Subject: Re: Why are cardbus drivers cbb(4) and pccard(4) still included in GENERIC? Mime-Version: 1.0 (Apple Message framework v1085) Content-Type: text/plain; charset=us-ascii From: Warner Losh In-Reply-To: <201308291054.02641.jhb@freebsd.org> Date: Thu, 29 Aug 2013 09:50:11 -0600 Content-Transfer-Encoding: quoted-printable Message-Id: References: <201308291054.02641.jhb@freebsd.org> To: John Baldwin X-Mailer: Apple Mail (2.1085) Cc: Adrian Chadd , "freebsd-stable@freebsd.org" , Kimmo Paasiala , freebsd-current@freebsd.org, freebsd-hardware@freebsd.org, Warner Losh X-BeenThere: freebsd-hardware@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: General discussion of FreeBSD hardware List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 29 Aug 2013 15:50:22 -0000 On Aug 29, 2013, at 8:54 AM, John Baldwin wrote: > On Thursday, August 29, 2013 6:56:53 am Adrian Chadd wrote: >> Hm! Are they dynamically loaded if you insert the cards? >>=20 >> (Ie, has devd been taught about them as appropriate?) >=20 > These are drivers for the bridges, not for cards you plug into the = bridges. =20 > If you autoloaded them at all you would load them during boot when you = saw an=20 > appropriate PCI device. Currently we don't autoload any PCI drivers, = so I=20 > don't think that should be a blocker for taking these out of GENERIC. >=20 > Warner is probably the best person to ask. We don't autoload them. The bridge code is tiny and can easily be = removed if you want to create a custom kernel. There's still plenty of = laptops that would be crippled if these were removed. Other drivers that = are even more obsolete that are still in GENERIC: esp sym trm adv adw aic bt asr ciss iir ida de cas dc gem hme nfe nve pcn (although many VM env like this) rl fs sis tl tx vr wb xl cs ed ex ep fe sn xe an wi urio uipaq aue cue kue rue So asking why cbb and cardbus are still there seems a little silly to = me. The basic problem is that our PCI and USB infrastructure doesn't include = the basic information necessarily to properly look at a directory of .ko = files and know what to autoload when it finds an unmatched device. PC = Card could easily do this, but doesn't (I have some doodles from a few = years ago that would put this stuff into an elf section that could (a) = be unloaded after probe (optionally) and (b) be read by automatic tools = to load just what's needed). USB is a lot closer than PCI, which is why = Hans' scripts work well enough to create uber-ugly devd config files. Rather than shooting randomly, perhaps some investigation about this = topic would be in order. We've talked it to death, but nobody has had = the time to STFU and implement something reasonable. Warner=