From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bart Van Assche Subject: Re: [PATCH 12/18] scsi: respect unchecked_isa_dma for blk-mq Date: Fri, 27 Jan 2017 17:45:14 +0000 Message-ID: <1485539094.4267.14.camel@sandisk.com> References: <1485365126-23210-1-git-send-email-hch@lst.de> <1485365126-23210-13-git-send-email-hch@lst.de> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: <1485365126-23210-13-git-send-email-hch@lst.de> Content-Language: en-US Content-ID: <9B1032A810FA4F4FA09AE00071699E04@sandisk.com> Sender: linux-raid-owner@vger.kernel.org To: "hch@lst.de" , "axboe@fb.com" Cc: "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 Wed, 2017-01-25 at 18:25 +0100, Christoph Hellwig wrote: > Currently blk-mq always allocates the sense buffer using normal GFP_KERNE= L > allocation. Refactor the cmd pool code to split the cmd and sense alloca= tion > and share the code to allocate the sense buffers as well as the sense buf= fer > slab caches between the legacy and blk-mq path. >=20 > Note that this switches to lazy allocation of the sense slab caches - the > slab caches (not the actual allocations) won't be destroy until the scsi > module is unloaded instead of keeping track of hosts using them. Reviewed-by: Bart Van Assche =