Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.31 as permitted sender) client-ip=23.128.96.31;
Message-ID: <140cff33-3421-4d78-9d58-83de5eec4dce@collabora.com>
Date:   Thu, 23 Nov 2023 13:44:21 +0100
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH v3] drm/panfrost: Ignore core_mask for poweroff and sync
 interrupts
Content-Language: en-US
To:     Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>,
        steven.price@arm.com
Cc:     boris.brezillon@collabora.com, robh@kernel.org,
        maarten.lankhorst@linux.intel.com, mripard@kernel.org,
        tzimmermann@suse.de, airlied@gmail.com, daniel@ffwll.ch,
        dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org
References: <20231123120521.147695-1-angelogioacchino.delregno@collabora.com>
 <5b24cc73-23aa-4837-abb9-b6d138b46426@linaro.org>
From:   AngeloGioacchino Del Regno 
        <angelogioacchino.delregno@collabora.com>
In-Reply-To: <5b24cc73-23aa-4837-abb9-b6d138b46426@linaro.org>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

Il 23/11/23 13:43, Krzysztof Kozlowski ha scritto:
> On 23/11/2023 13:05, AngeloGioacchino Del Regno wrote:
>> Some SoCs may be equipped with a GPU containing two core groups
>> and this is exactly the case of Samsung's Exynos 5422 featuring
>> an ARM Mali-T628 MP6 GPU: the support for this GPU in Panfrost
>> is partial, as this driver currently supports using only one
>> core group and that's reflected on all parts of it, including
>> the power on (and power off, previously to this patch) function.
>>
>> The issue with this is that even though executing the soft reset
>> operation should power off all cores unconditionally, on at least
>> one platform we're seeing a crash that seems to be happening due
>> to an interrupt firing which may be because we are calling power
>> transition only on the first core group, leaving the second one
>> unchanged, or because ISR execution was pending before entering
>> the panfrost_gpu_power_off() function and executed after powering
>> off the GPU cores, or all of the above.
>>
>> Finally, solve this by introducing a new panfrost_gpu_suspend_irq()
>> helper function and changing the panfrost_device_suspend() flow to
>>   1. Mask and clear all interrupts: we don't need nor want any, as
>>      for power_off() we are polling PWRTRANS, but we anyway don't
>>      want GPU IRQs to fire while it is suspended/powered off;
>>   2. Call synchronize_irq() after that to make sure that any pending
>>      ISR is executed before powering off the GPU Shaders/Tilers/L2
>>      hence avoiding unpowered registers R/W; and
>>   3. Ignore the core_mask and ask the GPU to poweroff both core groups
>>
>> Of course it was also necessary to add a `irq` variable to `struct
>> panfrost_device` as we need to get that in panfrost_gpu_power_off()
>> for calling synchronize_irq() on it.
>>
>> Fixes: 22aa1a209018 ("drm/panfrost: Really power off GPU cores in panfrost_gpu_power_off()")
>> [Regression detected on Odroid HC1, Exynos5422, Mali-T628 MP6]
>> Reported-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
>> Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
> 
> 
> Tested-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
> 

Thank you for the test and, more importantly, for your patience.

Cheers,
Angelo

> Best regards,
> Krzysztof
>