From mboxrd@z Thu Jan 1 00:00:00 1970 From: 'Christoph Hellwig' Subject: Re: [ANNOUNCE] megaraid linux driver version 2.00.7 Date: Wed, 6 Aug 2003 06:43:53 +0100 Sender: linux-kernel-owner@vger.kernel.org Message-ID: <20030806064353.A21769@infradead.org> References: <0E3FA95632D6D047BA649F95DAB60E570185F3FA@EXA-ATLANTA.se.lsil.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <0E3FA95632D6D047BA649F95DAB60E570185F3FA@EXA-ATLANTA.se.lsil.com>; from atulm@lsil.com on Tue, Aug 05, 2003 at 06:09:10PM -0400 To: "Mukker, Atul" Cc: 'Christoph Hellwig' , "'linux-kernel@vger.kernel.org'" , "'linux-scsi@vger.kernel.org'" , "'linux-megaraid-devel@dell.com'" List-Id: linux-scsi@vger.kernel.org On Tue, Aug 05, 2003 at 06:09:10PM -0400, Mukker, Atul wrote: > > + spin_lock_irqsave(adapter->host->host_lock, flags); > But all kernels do not have this lock as part of the host structure. With > 2.00.7, I have relapsed a patch which switches the lock to io_request_lock > if kernel does not support per host lock. > > And not all kernels have the per host lock named as "host->host_lock", some > simply have "host->lock"A All 2.5/2.6 kernels do have host->host_lock and the 2.5 driver is different from the 2.4 one. In fact it will hopefuloly become &host->host_lock soon but for that I have to repair the remaining abusers first.