From mboxrd@z Thu Jan 1 00:00:00 1970 From: dirk.behme@gmail.com (Dirk Behme) Date: Sun, 23 Jun 2013 08:46:40 +0200 Subject: [PATCH] ARM: optimize memset_io()/memcpy_fromio()/memcpy_toio() In-Reply-To: <20130613171338.GI18614@n2100.arm.linux.org.uk> References: <51B76972.3040002@gmail.com> <20130613110501.GZ18614@n2100.arm.linux.org.uk> <2184363.o70cB30E6f@wuerfel> <20130613171338.GI18614@n2100.arm.linux.org.uk> Message-ID: <51C699D0.201@gmail.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Hi Arnd, On 13.06.2013 19:13, Russell King - ARM Linux wrote: > On Thu, Jun 13, 2013 at 02:47:35PM +0200, Arnd Bergmann wrote: >> On Thursday 13 June 2013 12:05:02 Russell King - ARM Linux wrote: >>> Yes, unfortunately it should be selectable by platform, though it needs >>> to be done carefully. Some ARM platforms can't cope with full 32-bit >>> accesses to MMIO. >>> >>> I don't think any of those intersect with the single zImage project, so >>> those should be fine to have it always enabled - it's only the older >>> platforms that might have issues. >> >> Are those all StrongARM and XScale based platforms? If so, I definitely >> don't expect them to go multiplatform. > > I'm not 100% positive on that, but... > >> Can we combine this with the existing CONFIG_NEED_MACH_IO_H? > > I think we probably can - because that implies that the generic set of > IO accessors are already being used, which means 8, 16 and 32 bit > accesses must be supportable without any kind of trickery. Do you like to send a patch with this CONFIG_NEED_MACH_IO_H proposal? Many thanks and best regards Dirk P.S.: Just for reference, the initial patch from Russell we are talking about: http://permalink.gmane.org/gmane.linux.ports.arm.kernel/190902