From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?utf-8?Q?Am=C3=A9rico?= Wang Subject: Re: declare MODULE_FIRMWARE in libsas Date: Wed, 5 Jan 2011 13:42:07 +0800 Message-ID: <20110105054207.GC6256@cr0.nay.redhat.com> References: <4D22CF9F.1070005@oracle.com> <1294150132.4726.0.camel@mulgrave.site> <4D235CAC.5070309@garzik.org> <4D23E315.9060900@oracle.com> <20110105044733.GA6256@cr0.nay.redhat.com> <4D24017A.3060202@oracle.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: <4D24017A.3060202@oracle.com> Sender: linux-kernel-owner@vger.kernel.org To: DuanZhenzhong Cc: =?utf-8?Q?Am=C3=A9rico?= Wang , Jeff Garzik , James Bottomley , linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org List-Id: linux-scsi@vger.kernel.org On Wed, Jan 05, 2011 at 01:28:26PM +0800, DuanZhenzhong wrote: >Am=C3=A9rico Wang =E5=86=99=E9=81=93: >>On Wed, Jan 05, 2011 at 11:18:45AM +0800, DuanZhenzhong wrote: >>>Jeff Garzik =E5=86=99=E9=81=93: >>>>On 01/04/2011 09:08 AM, James Bottomley wrote: >>>>>On Tue, 2011-01-04 at 15:43 +0800, DuanZhenzhong wrote: >>>>>>Signed-off-by: Zhenzhong Duan >>>>>> >>>>>>--- >>>>>Description? Since libsas has no firmware, surely some explanatio= n is >>>>>required. >>>>Indeed. Looks bogus at first glance? >>>> >>>> >>>I was finding all requested but not exported firmware in scsi driver >>>for generating dump initrd use. >>>In sas_scsi_host.c, request_firmware is called with "sas_addr", >>>sas_addr neither builtin nor exported. Don't sas_addr a firmware? >>>Please correct me. thanks >>> >> >>But once you export it, it needs to be found in fireware/ directory..= =2E >Not exactly, firmware/ dir only have a subset of all needed firmware. >Also one can load firmware manually from anywhere > Then it should not be exported, otherwise user-space will get confused. We have a script to parse modinfo output to get the firmwares of a modu= le.