2 Commits

Author	SHA1	Message	Date
Uwe Kleine-König	a9b7a2dd21	imx: mangle addresses after adding the offset ... don't use IO_ADDRESS($base) + $offset but IO_ADDRESS($base + $offset) Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Cc: Sascha Hauer <kernel@pengutronix.de> Cc: Russell King <linux@arm.linux.org.uk> Cc: Holger Schurig <hs4233@mail.mn-solutions.de> Cc: Rabin Vincent <rabin@rab.in> Cc: "Agustín Ferrín Pozuelo" <gatoguan-os@yahoo.com> Cc: Javier Martin <javier.martin@vista-silicon.com> Cc: Valentin Longchamp <valentin.longchamp@epfl.ch> Cc: Daniel Mack <daniel@caiaq.de> Cc: Dmitriy Taychenachev <dimichxp@gmail.com>	2010-01-08 16:45:20 +01:00
Daniel Mack	52939c03e5	ARM: MX3: fix CPU revision number detection ... The macro mx31_revision() used to take the global variable system_rev to determine the CPU revision number. However, this number is expected to be set by the bootloader and is usually zero (at least on my MX31 based boards here). More than that, it is usually taken to identify the board's revision, not the CPU's. Fix that by reading the the CPU's SREV register instead. Right now, mx31_read_cpu_rev() is called from mx31_clocks_init() which is admittedly not a good place for it. However, we need to enable the IIM clock first, and the clock code also has conditional code that depends on mx31_revision() returning the right thing. Signed-off-by: Daniel Mack <daniel@caiaq.de> Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>	2009-11-23 11:51:18 +01:00