linux-kernel - Re: [PATCH v4 1/3] bus: mhi: core: Clear devices when moving execution environments

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] [day] [month] [year] [list]

Message-ID: <f031ea65fa4eecb97d4b7a21ac7e52af@codeaurora.org>
Date:   Fri, 19 Feb 2021 11:13:16 -0800
From:   Bhaumik Bhatt <bbhatt@...eaurora.org>
To:     Jeffrey Hugo <jhugo@...eaurora.org>
Cc:     manivannan.sadhasivam@...aro.org, linux-arm-msm@...r.kernel.org,
        hemantk@...eaurora.org, linux-kernel@...r.kernel.org,
        loic.poulain@...aro.org, carl.yin@...ctel.com,
        naveen.kumar@...ctel.com, jhugo=codeaurora.org@...eaurora.org
Subject: Re: [PATCH v4 1/3] bus: mhi: core: Clear devices when moving
 execution environments

On 2021-02-19 08:10 AM, Jeffrey Hugo wrote:
> On 2/18/2021 7:17 PM, Bhaumik Bhatt wrote:
>> When moving from SBL to mission mode execution environment, there
>> is no remove callback notification to MHI client drivers which
>> operate on SBL mode only. Client driver devices are being created
>> in SBL or AMSS(mission mode) and only destroyed after power down
>> or SYS_ERROR. If there exist any SBL-specific channels, those are
>> left open and client drivers are thus unaware of the new execution
>> environment where those channels cannot operate. Close the gap and
>> issue remove callbacks to SBL-specific client drivers once device
>> enters mission mode.
>> 
>> Signed-off-by: Bhaumik Bhatt <bbhatt@...eaurora.org>
>> ---
> 
> I like the idea, but I question where this is limited to the
> transition to mission mode.  Feels like something that should occur on
> all EE changes.  We might not have a current usecase that is outside
> what you've implemented here, but I don't think there is anything
> preventing that in future.

You're right. It should not be limited to any single EE transition and 
that's
how the code is written. I see my commit message gives the impression 
that it's
only for SBL -> AMSS. I can correct that and mention it to be EE 
transition
agnostic with SBL to AMSS as the currently usable example.

Thanks,
Bhaumik
---
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora 
Forum,
a Linux Foundation Collaborative Project