From mboxrd@z Thu Jan 1 00:00:00 1970 From: linux@arm.linux.org.uk (Russell King - ARM Linux) Date: Fri, 21 Dec 2012 10:39:55 +0000 Subject: [PATCH] mmc: mmci: Support non-power-of-two block sizes for ux500v2 variant In-Reply-To: References: <20121121165048.GO3290@n2100.arm.linux.org.uk> <20121122173708.GJ5764@n2100.arm.linux.org.uk> <50B34270.7070403@stericsson.com> <20121126102712.GC19440@n2100.arm.linux.org.uk> <50B349D3.2050806@stericsson.com> <20121128171221.GF19440@n2100.arm.linux.org.uk> Message-ID: <20121221103955.GH14363@n2100.arm.linux.org.uk> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Fri, Dec 21, 2012 at 11:36:32AM +0100, Ulf Hansson wrote: > Just wanted to conclude on the way forward. Should we fixup pio_write > according to how pio_read has been fixed, or should we adapt the check > for misaligned buffers to what Per proposed? > What do you prefer Russell? It's really silly to have pio_read doing something that pio_write doesn't do - they should have the same behaviour except for the differing data flow direction.