From mboxrd@z Thu Jan  1 00:00:00 1970
From: lee.jones@linaro.org (Lee Jones)
Date: Wed, 20 Jun 2012 15:13:51 +0100
Subject: [PATCH 15/15] ARM: perf: handle muxed CPU IRQ lines
In-Reply-To: <20120620140115.GJ32666@mudshark.cambridge.arm.com>
References: <1340197011-5435-1-git-send-email-lee.jones@linaro.org>
 <1340197011-5435-16-git-send-email-lee.jones@linaro.org>
 <20120620132823.GG32666@mudshark.cambridge.arm.com>
 <4FE1D3D2.3050209@linaro.org>
 <20120620140115.GJ32666@mudshark.cambridge.arm.com>
Message-ID: <4FE1DA9F.30902@linaro.org>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On 20/06/12 15:01, Will Deacon wrote:
> On Wed, Jun 20, 2012 at 02:44:50PM +0100, Lee Jones wrote:
>>>
>>> Shouldn't this be in the platform IRQ handler, rather than the core v7 perf
>>> code?
>>
>> Well that's where it currently resides, but in the process of DT:ing the
>> platform, we have no real way to set the call-back. Normally, if it's
>> absolutely necessary, we do this with AUXDATA. However in the PMU case
>> we fall-down as a device address is required for DT look-up, which the
>> PMU does not have.
>
> Right, so I suspect this is going to be a huge problem for the OMAP guys,
> where they have to do a fair amount of jiggery-pokery to kick their PMU into
> action. We are also looking to put suspend/resume hooks into the
> arm_pmu_platdata which will undoubtedly be highly platform-specific.
>
>> I spoke with Arnd about sensible solutions and he suggested placing the
>> logic inside the generic instead, thus supporting muxed IRQ lines
>> without the requirement of providing a platform specific handler in
>> every instance this occurs (although, I think the ux500 are the only
>> platform to do this currently).
>
> Whilst this code isn't too problematic (but yes, you are the only platform
> that needs it) I think if we solve the bigger problem then you can use the
> same solution to do your IRQ ping-ponging.
>
>> I'd be happy to take any other suggestions, but this seemed fairly neat
>> and unobtrusive (not withstanding the predicate cock-up).
>
> Can we not fix the AUXDATA stuff to assume address 0 if it's not present for
> the matching node? This will work for perf as I don't expect to have
> multiple nodes with the same compatible string for PMUs that are not
> memory-mapped (for big.LITTLE, the strings will be different).

Okay, leave it with me.

I'm off now until Monday, I'll pick it up then.

Thanks Will.

-- 
Lee Jones
Linaro ST-Ericsson Landing Team Lead
M: +44 77 88 633 515
Linaro.org ? Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog