Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <b1de7e87-9590-3aad-d62b-167a7685746a@quicinc.com>
Date:   Tue, 11 Oct 2022 15:34:45 +0530
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101
 Thunderbird/102.3.0
Subject: Re: Query regarding "firmware: arm_scmi: Free mailbox channels if
 probe fails"
To:     Cristian Marussi <cristian.marussi@arm.com>
CC:     <sudeep.holla@arm.com>, <linux-arm-kernel@lists.infradead.org>,
        <linux-kernel@vger.kernel.org>, <quic_rgottimu@quicinc.com>
References: <cfa26ff3-c95a-1986-58fc-b49fc9be49d5@quicinc.com>
 <Yyx3IAcMX309QEjB@e120937-lin> <Yyx/DKcc7XupQmnx@e120937-lin>
 <c81540cc-e485-0c45-9e4e-248d3279e1ea@quicinc.com>
 <YzriGCLf+MFNGO2n@e120937-lin>
Content-Language: en-US
From:   Shivnandan Kumar <quic_kshivnan@quicinc.com>
In-Reply-To: <YzriGCLf+MFNGO2n@e120937-lin>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk


Hi Cristian,

 >>Ok, just out of curiosity, once done, can you point me at your 
downstream public sources so I can see the issue and the fix that you 
are applying to your trees ?

https://source.codeaurora.org/quic/la/kernel/msm-5.10/tree/drivers/soc/qcom/qcom_rimps.c?h=KERNEL.PLATFORM.1.0.r1-07800-kernel.0

I have added lock while accessing con_priv inside irq handler and 
shutdown function.


I have one input regarding timeout from firmware, can we enable BUG on 
response  time out in function do_xfer based on some debug config 
flag,this will help to debug firmware timeout issue faster.

We will only enable that config flag during internal testing.


Thanks,

Shivnandan

On 10/3/2022 6:52 PM, Cristian Marussi wrote:
> On Fri, Sep 30, 2022 at 06:29:02PM +0530, Shivnandan Kumar wrote:
>> hi Cristian,
> Hi Shivnandan,
>> Thanks for your support in providing the patch to try.
>>
>> I found one race condition in our downstream mbox controller driver while
>> accessing con_priv, when I serialized access to this, issue is not seen on 3
>> days of testing.
> Good to hear that you find the issue.
>
>> As you rightly mentioned that your provided patch will impact all the other
>> users.
>>
>> Also if  we take your provided patch, same race still exists while accessing
>> con_priv in our downstream mbox controller so this issue will still be
>> there.
>>
> Yes indeed, even though I think that race in the mailbox core between RX path
> and chan_free could still be theoretically possible it does not seem to me
> appropriate to try to fix it now that you cannot reproduce it anymore and
> no other mailbox user has ever raised this concern (even though, as said, the
> proper solution to that race wont probably be directly in the mailbox-core as
> in my experimental two liners..)
>
>> So, we are planning to merge the patch( serialized access to con_priv) in
>> our downstream mbox controller now.
>>
> Ok, just out of curiosity, once done, can you point me at your downstream public
> sources so I can see the issue and the fix that you are applying to your trees ?
>
> Thanks,
> Cristian