Return-path: Received: from smtp.rutgers.edu ([128.6.72.243]:63142 "EHLO annwn14.rutgers.edu" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S967344AbXEHAEF (ORCPT ); Mon, 7 May 2007 20:04:05 -0400 From: Michael Wu To: David Miller Subject: Re: [PATCH v2] Add rtl8187 wireless driver Date: Mon, 7 May 2007 20:03:24 -0400 Cc: jeff@garzik.org, netdev@vger.kernel.org, linux-wireless@vger.kernel.org, linville@tuxdriver.com, andrea.merello@gmail.com References: <200705071022.11711.flamingice@sourmilk.net> <200705071917.25414.flamingice@sourmilk.net> <20070507.162518.35013789.davem@davemloft.net> In-Reply-To: <20070507.162518.35013789.davem@davemloft.net> MIME-Version: 1.0 Content-Type: multipart/signed; boundary="nextPart1562274.oebPpi6ZBI"; protocol="application/pgp-signature"; micalg=pgp-sha1 Message-Id: <200705072003.29448.flamingice@sourmilk.net> Sender: linux-wireless-owner@vger.kernel.org List-ID: --nextPart1562274.oebPpi6ZBI Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline On Monday 07 May 2007 19:25, David Miller wrote: > I know this sounds trite, but when merging and researching up to 450 > patches at a time like I have to, this stuff starts to matter. > > Please put things as high in the directory hierachy as possible and > when you can put the entire driver into a single source file do so. > Sure. I generally prefer that too, but all the drivers I put in wireless-de= v=20 were put in separate directories because I didn't want to break the pattern. > The only significant argument you present is the code sharing one for > the radio stuff, but that isn't realized yet and you can certainly > split the code out once you make that sharing a reality. Having different modules for different radio chips really isn't the point o= f=20 keeping it in a separate file. (in fact, rtl8187_rtl8225.c actually contain= s=20 support for two versions of the rtl8225 radio) It's just that=20 rtl8187_rtl8225.c is basically a bunch of numbers with a bit of glue code t= o=20 send them to the hardware. Not much is missed if you decide not to look at= =20 how the radio tuning works, and I like to avoid looking at that code if I=20 can. It's also consistent with how the (unreleased) rtl8180 driver works,=20 which really does have support for different radios in different files. But of course, we can just make an exception for rtl8187. Would merging the= =20 files together make that much of a difference for you? =2DMichael Wu --nextPart1562274.oebPpi6ZBI Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) iD8DBQBGP75RT3Oqt9AH4aERArWhAKDIrPlqESQGRu1liJUy71aLRXlsfQCguNLG gCDrRTO/dxU5G92aSk5dAiA= =/gYw -----END PGP SIGNATURE----- --nextPart1562274.oebPpi6ZBI-- -: To unsubscribe from this list: send the line "unsubscribe linux-wireless" in the body of a message to majordomo@vger.kernel.org: More majordomo info at http: //vger.kernel.org/majordomo-info.html