linux-kernel - Re: [PATCH v2 05/18] firmware: arm_scmi: add initial support for performance protocol

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]

Message-ID: <b7303424-d63f-5ea7-866a-bda6d1f93ff2@arm.com>
Date:   Tue, 5 Sep 2017 17:54:43 +0100
From:   Sudeep Holla <sudeep.holla@....com>
To:     Julien Thierry <julien.thierry@....com>,
        ALKML <linux-arm-kernel@...ts.infradead.org>,
        LKML <linux-kernel@...r.kernel.org>,
        DTML <devicetree@...r.kernel.org>
Cc:     Sudeep Holla <sudeep.holla@....com>, Nishanth Menon <nm@...com>,
        Harb Abdulhamid <harba@...eaurora.org>,
        Arnd Bergmann <arnd@...db.de>,
        Jassi Brar <jassisinghbrar@...il.com>,
        Ryan Harkin <Ryan.Harkin@....com>,
        Roy Franz <roy.franz@...ium.com>, Loc Ho <lho@....com>,
        Alexey Klimov <alexey.klimov@....com>
Subject: Re: [PATCH v2 05/18] firmware: arm_scmi: add initial support for
 performance protocol

Hi Julien,

On 05/09/17 16:56, Julien Thierry wrote:
> 
> 
> On 05/09/17 16:04, Julien Thierry wrote:

[...]


>>
>> This seems odd, shouldn't it be the following?
>> le64_to_cpu(attr->stats_addr_low | (__le64)attr->stats_addr_high << 32)
>>
> 
> After further reflexion, I think you are right. If I understood the
> specification, the address seems to be split into upper and lower 32bits
> and each one is stored as a uint32, which fits what you are doing to
> obtain the address.
> 
> You can ignore my previous comment.
> 

Sorry I am checking this bit late, you have figured out yourself
already. All the 64 bit values need to be dealt similarly as SCMI
specification restricts so. It's basically done so as the firmware
running is mostly 32-bit little endian system.

-- 
Regards,
Sudeep