From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jens Axboe Subject: Re: [PATCH, scsi-block-2.6] don't export bio_map_kern Date: Mon, 20 Jun 2005 12:37:51 +0200 Message-ID: <20050620103750.GB15021@suse.de> References: <20050619093113.GB16872@lst.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from ns.virtualhost.dk ([195.184.98.160]:2269 "EHLO virtualhost.dk") by vger.kernel.org with ESMTP id S261158AbVFTKgb (ORCPT ); Mon, 20 Jun 2005 06:36:31 -0400 Content-Disposition: inline In-Reply-To: <20050619093113.GB16872@lst.de> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Christoph Hellwig Cc: jejb@steeleye.com, linux-scsi@vger.kernel.org On Sun, Jun 19 2005, Christoph Hellwig wrote: > Modules should use the higher-level blk_rq_map_kern, which they > currently do anyway. > > > Index: linux-2.6/fs/bio.c > =================================================================== > --- linux-2.6.orig/fs/bio.c 2005-06-18 17:26:42.000000000 +0200 > +++ linux-2.6/fs/bio.c 2005-06-18 17:49:31.000000000 +0200 > @@ -1201,7 +1201,6 @@ > EXPORT_SYMBOL(bio_add_page); > EXPORT_SYMBOL(bio_get_nr_vecs); > EXPORT_SYMBOL(bio_unmap_user); > -EXPORT_SYMBOL(bio_map_kern); > EXPORT_SYMBOL(bio_pair_release); > EXPORT_SYMBOL(bio_split); > EXPORT_SYMBOL(bio_split_pool); Why? It would be perfectly legit for stacking driver to set up their own bio with kernel mapped data. As with bio_map_user(), blk_rq_map_kern isn't the only valid user of bio_map_kern(). -- Jens Axboe