From mboxrd@z Thu Jan 1 00:00:00 1970 From: "hch@lst.de" Subject: Re: [PATCH 14/18] scsi: remove __scsi_alloc_queue Date: Sat, 28 Jan 2017 09:23:32 +0100 Message-ID: <20170128082332.GA4728@lst.de> References: <1485365126-23210-1-git-send-email-hch@lst.de> <1485365126-23210-15-git-send-email-hch@lst.de> <1485539862.4267.17.camel@sandisk.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <1485539862.4267.17.camel@sandisk.com> Sender: linux-scsi-owner@vger.kernel.org To: Bart Van Assche Cc: "hch@lst.de" , "axboe@fb.com" , "linux-scsi@vger.kernel.org" , "linux-raid@vger.kernel.org" , "dm-devel@redhat.com" , "linux-block@vger.kernel.org" , "snitzer@redhat.com" , "j-nomura@ce.jp.nec.com" List-Id: linux-raid.ids On Fri, Jan 27, 2017 at 05:58:02PM +0000, Bart Van Assche wrote: > Since __scsi_init_queue() modifies data in the Scsi_Host structure, have you > considered to add the declaration for this function to ? > If you want to keep this declaration in please add a > direct include of that header file to drivers/scsi/scsi_lib.c such that the > declaration remains visible to the compiler if someone would minimize the > number of #include directives in SCSI header files. Feel free to send an incremental patch either way. In the long run I'd really like to kill off __scsi_init_queue and remove the transport BSG queue abuse of SCSI internals, though.