[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20130925184435.GH2954@lukather>
Date: Wed, 25 Sep 2013 21:44:35 +0300
From: Maxime Ripard <maxime.ripard@...e-electrons.com>
To: cinifr <cinifr@...il.com>
Cc: coosty@....com, Daniel Lezcano <daniel.lezcano@...aro.org>,
"linux@....linux.org.uk" <linux@....linux.org.uk>,
"tglx@...utronix.de" <tglx@...utronix.de>,
"linux-arm-kernel@...ts.infradead.org"
<linux-arm-kernel@...ts.infradead.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Mark Rutland <mark.rutland@....com>,
"pawel.moll@....co" <pawel.moll@....co>,
Rob Herring <rob.herring@...xeda.com>,
"linux-sunxi@...glegroups.com" <linux-sunxi@...glegroups.com>
Subject: Re: [PATCH V3: Add Smp support for Allwinner A20. 1/3] Add smp
support for Allwinner A20(sunxi 7i).
Hi Fan,
On Wed, Sep 25, 2013 at 12:22:44AM +0800, cinifr wrote:
> > You still haven't said why you don't want to use smp_init_cpus there.
> I find that of_funcation liking of_find_matching_node and of_iomap can
> not run well in smp_init_cpus. It cause kernel crash. So I have to use
> early_initcall.
Ah, right.
Then, can't we use smp_prepare_cpus for this? it looks to be late enough
so that we can use ioremap there.
I don't want it to be an initcall. The code will be different on the
A31, and we don't want it to be run on all the other machines whenever
we boot a multiplatform kernel anyway.
Maxime
--
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
Download attachment "signature.asc" of type "application/pgp-signature" (837 bytes)
Powered by blists - more mailing lists