MIME-Version: 1.0
In-Reply-To: <20130312172505.GA25266@htj.dyndns.org>
References: <1362663356-21151-1-git-send-email-tom.leiming@gmail.com>
	<CACVXFVOhr222CVrzkVP_oLgVkdReD9BXXSBg7fmp6Wvx9_peCw@mail.gmail.com>
	<20130312105638.GQ4977@n2100.arm.linux.org.uk>
	<CACVXFVOEvvFT805TJHoppqGRLTRuP18=Wg72LSXryffZD_WH1A@mail.gmail.com>
	<20130312113048.GS4977@n2100.arm.linux.org.uk>
	<CACVXFVP4NeZBg5yZW8q+vhgs0m3knkkdvyjv_hc9BLrFPqEBJA@mail.gmail.com>
	<20130312172505.GA25266@htj.dyndns.org>
Date: Wed, 13 Mar 2013 08:57:04 +0800
Message-ID: <CACVXFVNNzgMFQ_G7HeF39S7Pz42qsM+0rZaTCSy76y+sXUREUA@mail.gmail.com>
Subject: Re: [PATCH v1] ARM: keep __my_cpu_offset consistent with generic one
From: Ming Lei <tom.leiming@gmail.com>
To: Tejun Heo <tj@kernel.org>
Cc: Russell King - ARM Linux <linux@arm.linux.org.uk>,
        linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org,
        Peter Zijlstra <peterz@infradead.org>, Ingo Molnar <mingo@redhat.com>,
        Rob Herring <rob.herring@calxeda.com>,
        Will Deacon <will.deacon@arm.com>, Nicolas Pitre <nico@linaro.org>,
        Christoph Lameter <cl@linux-foundation.org>
Content-Type: text/plain; charset=ISO-8859-1
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1395
Lines: 35

On Wed, Mar 13, 2013 at 1:25 AM, Tejun Heo <tj@kernel.org> wrote:
> Hello,
>
> On Tue, Mar 12, 2013 at 07:44:55PM +0800, Ming Lei wrote:
>> On Tue, Mar 12, 2013 at 7:30 PM, Russell King - ARM Linux
>> <linux@arm.linux.org.uk> wrote:
>> >>
>> >> Ingo and Peter, what is your opinion on the problem?
>> >
>> > Having discussed this with Ben Herrenschmidt, it seems that we do need
>> > to have a more complex patch to sort this out - we need to setup our
>> > private pointer inside setup_per_cpu_areas().
>>
>> Suppose so, seems the patch is still needed to make CPU0 see
>> static variables in '.data..percpu' section correctly.
>
> If my memory serves me right, x86 also has places where CPU0 accesses
> its per-cpu data in .data.percpu.  While those existed (not sure
> they're still there but probably they're) mostly due to historical
> reasons rather than by design, as long as the data is in consistent
> state by and during percpu setup, nothing will break.

Tejun, thanks for your input, yes, nothing will break.  For lockdep,
the percpu variables in non-boot-CPUs may be initialized again after
percpu area is set up.


Thanks,
-- 
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/