linux-kernel - Re: [PATCH] firmware: arm_scmi: Pass shmem address to SMCCC call

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:   Fri, 17 Jul 2020 15:00:16 +0100
From:   Paul Murphy <paul.j.murphy@...ux.intel.com>
To:     Florian Fainelli <f.fainelli@...il.com>,
        Daniele Alessandrelli <daniele.alessandrelli@...ux.intel.com>,
        Sudeep Holla <sudeep.holla@....com>,
        linux-arm-kernel@...ts.infradead.org
Cc:     Peng Fan <peng.fan@....com>,
        "Paul J. Murphy" <paul.j.murphy@...el.com>,
        linux-kernel@...r.kernel.org,
        Daniele Alessandrelli <daniele.alessandrelli@...el.com>
Subject: Re: [PATCH] firmware: arm_scmi: Pass shmem address to SMCCC call

Hi Florian


On 7/16/20 20:57, Florian Fainelli wrote:

>>>
>>> Given that the kernel must somehow reserve this memory as a shared
>>> memory area for obvious reasons, and the trusted firmware must also
>>> ensure it treats this memory region with specific permissions in its
>>> translation regime, does it really make sense to give that much
>>> flexibility?
>>
>> Well, the trusted firmware might reserve a bigger region to be used for
>> other service as well. In other words, the MMU of TF-A is not necessary
>> specifically set up for this region, but, possibly, for a bigger
>> general shared region.
> 
> But presumably the Linux shared memory area should be mapped in a
> slightly different way than
> 


Sorry - could you clarify what you mean by that?

Just checking if we are doing everything correctly.

I didn't understand that there is a connection between the TF-A MMU 
tables for this region and the normal world MMU tables?

For example:

TF-A may map physical address range: 0x0 -> 0x400_000 as 'normal' memory 
for various purposes.

Linux SCMI driver could map physical address range, eg: 0x300_000 -> 
0x301_000 as IO memory for mailbox purpose only.

Is there any issue here?

Regards,

Paul