Re: [PATCH v5] bus: mhi: core: Add support for processing priority of event ring

From: Kalle Valo <kvalo@codeaurora.org>
To: Bhaumik Bhatt <bbhatt@codeaurora.org>
Cc: ath11k@lists.infradead.org, linux-arm-msm@vger.kernel.org,
	hemantk@codeaurora.org, linux-kernel@vger.kernel.org,
	Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>,
	linux-wireless@vger.kernel.org
Subject: Re: [PATCH v5] bus: mhi: core: Add support for processing priority of event ring
Date: Thu, 16 Sep 2021 08:28:50 +0300	[thread overview]
Message-ID: <87bl4t1431.fsf@codeaurora.org> (raw)
In-Reply-To: <9561c5a41e1707d694b1c872fea399de@codeaurora.org> (Bhaumik Bhatt's message of "Wed, 28 Jul 2021 17:06:24 -0700")

Hi Bhaumik,

Bhaumik Bhatt <bbhatt@codeaurora.org> writes:

> On 2021-07-16 11:22 AM, Bhaumik Bhatt wrote:
>> Hi Mani,
>>
>> On 2021-07-16 04:49 AM, Manivannan Sadhasivam wrote:
>>> On Fri, Jun 25, 2021 at 10:22:08AM -0700, Bhaumik Bhatt wrote:
>>>> From: Hemant Kumar <hemantk@codeaurora.org>
>>>>
>>>> Event ring priorities are currently set to 1 and are unused.
>>>> Default processing priority for event rings is set to regular
>>>> tasklet. Controllers can choose to use high priority tasklet
>>>> scheduling for certain event rings critical for processing such
>>>> as ones transporting control information if they wish to avoid
>>>> system scheduling delays for those packets. In order to support
>>>> these use cases, allow controllers to set event ring priority to
>>>> high.
>>>>
>>>
>>> Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
>>>
>>> Just curious, what are the event rings you are going to make as high
>>> priority? If you are going to do that for existing controllers, please
>>> submit a patch now itself.
>>>
>>> Thanks,
>>> Mani
>>>
>> Idea for this patch came from 914b72a6948b ("bus: mhi: Wait for M2
>> state during system resume").
>>
>> If WLAN ath11k controller driver wants to avoid the scenario
>> mentioned in
>> the above patch, it will help them to have a high priority for their
>> dedicated
>> control events ring.
>>
>> I would defer to Kalle and others in ath11k, whether or not they are
>> OK to take that route.
>> as an update to priority will just help return from resume faster.
>> It will also depend on system load/reproducibility rate of the
>> scenario.
>>
>> I can provide a patch for them to review/test since I do not have the
>> setup for it.
>>
> Would you like to try this patch out? It basically increases the
> priority
> at which the control events for M0/M1/M3 state changes are handled.
>
> Let me know if you have any questions.

At the moment I'm seriously lagging with patches so I don't really have
any free time. Can some other ath11k developer help here?

-- 
https://patchwork.kernel.org/project/linux-wireless/list/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches