linux-kernel - Re: [PATCH] remoteproc: qcom: Introduce Hexagon V5 based WCSS driver

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 6 Jun 2018 15:21:17 +0530
From:   Sricharan R <sricharan@...eaurora.org>
To:     Vinod <vkoul@...nel.org>
Cc:     bjorn.andersson@...aro.org, ohad@...ery.com, robh+dt@...nel.org,
        mark.rutland@....com, andy.gross@...aro.org,
        david.brown@...aro.org, linux-remoteproc@...r.kernel.org,
        devicetree@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-arm-msm@...r.kernel.org, linux-soc@...r.kernel.org,
        sibis@...eaurora.org
Subject: Re: [PATCH] remoteproc: qcom: Introduce Hexagon V5 based WCSS driver

Hi Vinod,

On 6/6/2018 12:19 PM, Vinod wrote:
> Hi Sricharan,
> 
> On 06-06-18, 12:09, Sricharan R wrote:
> 
>>>>>> +config QCOM_Q6V5_WCSS
>>>>>> +	tristate "Qualcomm Hexagon based WCSS Peripheral Image Loader"
>>>>>> +	depends on OF && ARCH_QCOM
>>>>>> +	depends on QCOM_SMEM
>>>>>> +	depends on RPMSG_QCOM_SMD || (COMPILE_TEST && RPMSG_QCOM_SMD=n)
>>>>>> +	depends on RPMSG_QCOM_GLINK_SMEM || RPMSG_QCOM_GLINK_SMEM=n
>>>>>
>>>>> Is there a reason why it depends on RPMSG_QCOM_GLINK_SMEM=n? What would
>>>>> happen if distro wants both this and RPMSG_QCOM_GLINK_SMEM
>>>>>
>>>>   RPMSG_QCOM_GLINK_SMEM=n should be for the COMPILE_TEST. Probably that
>>>
>>> why would that be a limitation? I am more worried about
>>> RPMSG_QCOM_GLINK_SMEM=n being the condition here. In new drivers we
>>> should not typically have dependency on some symbol being not there
>>
>> Without that, if RPMSG_QCOM_GLINK_SMEM=m is compiled as a module, then
>> it would break the build.
> 
> Okay I do not know the details, but that doesn't sound correct to me.
> Breaking build sounds a bit extreme to me. Can you give details on this
> part..
> 

 Having, just, depends on RPMSG_QCOM_GLINK_SMEM || COMPILE_TEST,
 is going to break when RPMSG_QCOM_GLINK_SMEM=m and COMPILE_TEST=y.
 Hence the COMPILE_TEST && RPMSG_QCOM_GLINK_SMEM=n.

 Having said that, COMPILE_TEST is getting tested for RPMSG_QCOM_SMD=n in
 the previous line. So that's the reason for not having it in below line for
 RPMSG_QCOM_GLINK_SMEM.

>>>>   means that it should be corrected here and for ADSP, Q6V5_PIL as well.
>>>>   Bjorn, is that correct ?, should it be, below ?
>>>>  
>>>>   depends on (RPMSG_QCOM_SMD || (COMPILE_TEST && RPMSG_QCOM_SMD=n)) || (RPMSG_QCOM_GLINK_SMEM || (COMPILE_TEST && RPMSG_QCOM_GLINK_SMEM=n))
>>>
>>> that doesnt really sound good :(
>>
>>  Hmm, but i was thinking it should functionally depend on either SMD or GLINK and not both.
> 
> If you are depedent upon a symbol provided by a module you should say
> depends on. If a machine is not supposed to have both SMD or GLINK then
> the driver will not get probed.
> 

This is where, i was thinking, it should be functional if either of SMD or GLINK
is there, but should not require both.

Regards,
 Sricharan

-- 
"QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus