From mboxrd@z Thu Jan 1 00:00:00 1970 From: linux@arm.linux.org.uk (Russell King - ARM Linux) Date: Sun, 23 Jan 2011 23:09:32 +0000 Subject: [PATCH] mmci: fixup broken_blockend variant patch v2 In-Reply-To: References: <1295275071-13146-1-git-send-email-linus.walleij@stericsson.com> <4D346800.3070603@stericsson.com> <20110118121455.GB9719@n2100.arm.linux.org.uk> <20110119205158.GC6335@n2100.arm.linux.org.uk> <20110123180110.GJ30094@n2100.arm.linux.org.uk> Message-ID: <20110123230932.GK30094@n2100.arm.linux.org.uk> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Sun, Jan 23, 2011 at 10:40:05PM +0100, Linus Walleij wrote: > 2011/1/23 Russell King - ARM Linux : > > On Fri, Jan 21, 2011 at 01:53:06PM +0100, Linus Walleij wrote: > > > If we're going to move away from the blockend IRQs to using the data > > counter, can we have a patch which switches the driver to use the data > > counter and then a followup patch which removes the broken blockend > > stuff, rather than somethign that changes the broken blockend stuff > > and then removes it? > > 6631/1 fixes the broken blockend stuff into a state that works, > then 6632/1 removes it in favor of using the MMCIDATACNT. What I'm saying is why do we need to fix the blockend stuff if in the next patch we remove it - why not just combine the two patches and remove the thing in one go as a "we give up using the blockend irq"?