From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from lists.gentoo.org ([140.105.134.102] helo=robin.gentoo.org) by nuthatch.gentoo.org with esmtp (Exim 4.50) id 1Eg9UQ-00015x-OD for garchives@archives.gentoo.org; Sat, 26 Nov 2005 23:29:55 +0000 Received: from robin.gentoo.org (localhost [127.0.0.1]) by robin.gentoo.org (8.13.5/8.13.5) with SMTP id jAQNTf1t024439; Sat, 26 Nov 2005 23:29:41 GMT Received: from mail.twi-31o2.org (66-191-187-123.dhcp.gnvl.sc.charter.com [66.191.187.123]) by robin.gentoo.org (8.13.5/8.13.5) with ESMTP id jAQNTejN027002 for ; Sat, 26 Nov 2005 23:29:41 GMT Received: from localhost (localhost [127.0.0.1]) by mail.twi-31o2.org (Postfix) with ESMTP id A0C9124801F for ; Sat, 26 Nov 2005 18:20:01 -0500 (EST) Received: from mail.twi-31o2.org ([127.0.0.1]) by localhost (gravity.twi-31o2.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 25796-13 for ; Sat, 26 Nov 2005 18:19:59 -0500 (EST) Received: from vertigo.twi-31o2.org (vertigo.twi-31o2.org [192.168.0.10]) (using TLSv1 with cipher RC4-MD5 (128/128 bits)) (No client certificate requested) by mail.twi-31o2.org (Postfix) with ESMTP id 76E5324801E for ; Sat, 26 Nov 2005 18:19:59 -0500 (EST) Subject: Re: [gentoo-releng] oss in make.defaults for 2006.0? From: Chris Gianelloni To: gentoo-releng@lists.gentoo.org In-Reply-To: <200511260932.48681.m.debruijne@hccnet.nl> References: <200511251014.11425.m.debruijne@hccnet.nl> <1132927534.8482.2.camel@Darkmere.darkmere> <1132928493.19957.24.camel@vertigo.twi-31o2.org> <200511260932.48681.m.debruijne@hccnet.nl> Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-tc5vzWqCczaLhL9nyvjh" Organization: Gentoo Linux Date: Sat, 26 Nov 2005 18:28:20 -0500 Message-Id: <1133047700.19957.35.camel@vertigo.twi-31o2.org> Precedence: bulk List-Post: List-Help: List-Unsubscribe: List-Subscribe: List-Id: Gentoo Linux mail X-BeenThere: gentoo-releng@gentoo.org Reply-to: gentoo-releng@lists.gentoo.org Mime-Version: 1.0 X-Mailer: Evolution 2.4.1 X-Virus-Scanned: amavisd-new at twi-31o2.org X-Archives-Salt: 941ffc34-8619-43a1-a6f1-6f1c3c15054c X-Archives-Hash: 75fb8b1aad25a6605fd794ed50dff447 --=-tc5vzWqCczaLhL9nyvjh Content-Type: text/plain Content-Transfer-Encoding: quoted-printable On Sat, 2005-11-26 at 09:32 +0100, Michiel de Bruijne wrote: > > There's no way that we can make, for example, Return to Castle > > Wolfenstein or Enemy Territory, both of which are extremely popular, us= e > > ALSA natively. They *do* work with ALSA compiled with USE=3D"oss" or w= ith > > alsa-oss installed. As far as I know, the oss USE flag on ALSA only > > enables the alsa-oss dependency. >=20 > All the systems I maintain have -oss and don't have alsa-oss installed (I= do=20 > have activated OSS emulation in the kernel though). All the games I have=20 > installed on those systems (including RTCW and ET) work perfectly. That=20 > doesn't off course say that all Gentoo-based systems or all games work=20 > without problems, but I'm trying to say that the dependency on alsa might= not=20 > be as necessary as you seem to think. Having OSS emulation in the kernel is the same as merging alsa-oss if using alsa-driver. > > > > For the programs that are oss-only a useflag shouldn't even exists, > > > > because it's not optional. > > > > For those applications, correct. However, I have shown a good reason > > for it. Unless we simply tell anyone to always merge alsa-oss if they > > want to play games, which isn't exactly a "works out of the box" > > solution. I can think of a few scenarios we could employ to work aroun= d > > this, but they aren't nearly as clean as simply having OSS in the > > default USE. Personally, I think it should stay until it is removed > > from the kernel, and even then, it must stay so long as we are > > supporting 2.4 kernels which do not have ALSA, such as vanilla-sources. >=20 > Shouldn't 2.4 users use a 2.4 profile? (if they don't they have other=20 > "challenges" as well e.g. udev vs. devfs). oss turned on by default in a = 2.4=20 > profile makes perfect sense to me. They do. The point being that I would prefer not diverge them significantly except in locations where necessary. The real problem comes in with a game, such as enemy-territory, that uses oss. If we remove oss from the default USE, we must have the game check for either alsa-oss being installed, or check the kernel configuration itself (yuck!). With it being a default, we can just explain to the user that they need OSS support, as the default suggests. I really would not have a problem with removing oss from the default USE if someone can come up with a clean way of making sure the support is there on these older binary games. --=20 Chris Gianelloni Release Engineering - Strategic Lead x86 Architecture Team Games - Developer Gentoo Linux --=-tc5vzWqCczaLhL9nyvjh Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) iD8DBQBDiO+UkT4lNIS36YERAmEkAKCfrEtKhTCHrWaa3pl3qIXXgxwuCwCfSgKL LpBMWamrDdbBz6QYaqIIeC8= =neTP -----END PGP SIGNATURE----- --=-tc5vzWqCczaLhL9nyvjh-- -- gentoo-releng@gentoo.org mailing list