From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <54B5B6E4.4020400@fb.com> Date: Tue, 13 Jan 2015 17:23:00 -0700 From: Jens Axboe MIME-Version: 1.0 To: Richard Weinberger , Christoph Hellwig Subject: Re: [PATCH 2/2 v2] UBI: Block: Add blk-mq support References: <1420926734-16417-1-git-send-email-richard@nod.at> <1420926734-16417-2-git-send-email-richard@nod.at> <20150113162553.GA24351@infradead.org> <54B5A188.2000001@nod.at> <54B5A213.1090502@fb.com> <54B5A77C.9040405@nod.at> <54B5AA84.9010306@fb.com> <54B5AC10.6070102@nod.at> In-Reply-To: <54B5AC10.6070102@nod.at> Content-Type: text/plain; charset="windows-1252" Content-Transfer-Encoding: 7bit Cc: dedekind1@gmail.com, tom.leiming@gmail.com, linux-kernel@vger.kernel.org, linux-mtd@lists.infradead.org, computersforpeace@gmail.com, dwmw2@infradead.org List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On 01/13/2015 04:36 PM, Richard Weinberger wrote: > > > Am 14.01.2015 um 00:30 schrieb Jens Axboe: >>> If I understand you correctly it can happen that blk_rq_bytes() returns >>> more bytes than blk_rq_map_sg() allocated, right? >> >> No, the number of bytes will be the same, no magic is involved :-) > > Good to know. :) > >> But lets say the initial request has 4 bios, with each 2 pages, for a >> total of 8 segments. Lets further assume that the pages in each bio are >> contiguous, so that blk_rq_map_sg() will map this to 4 sg elements, each >> 2xpages long. >> >> Now, this may already be handled just fine, and you don't need to >> update/store the actual sg count. I just looked at the source, and I'm >> assuming it'll do the right thing (ubi_read_sg() will bump the active sg >> element, when that size has been consumed), but I don't have >> ubi_read_sg() in my tree to verify. > > Currently the sg count is hard coded to UBI_MAX_SG_COUNT. The max count doesn't matter, that just provides you a guarantee that you'll never receive a request that maps to more than that. The point I'm trying to make is that if you receive 8 segments and it maps to 4, then you better not look at segments 5..8 after it being mapped. Whatever the max is, doesn't matter in this conversation. > I'm sorry, I forgot to CC you and hch to this patch: Which is as I suspected, you'll do each segment to the length specified, hence you don't need to track the returned count from blk_rq_map_sg(). -- Jens Axboe