linux-kernel - Re: [PATCH v2 07/13] mailbox: pcc: Always map the shared memory communication address

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: <d2bf35ea-487d-b940-fb3e-093e14441735@huawei.com>
Date: Tue, 11 Mar 2025 20:31:23 +0800
From: "lihuisong (C)" <lihuisong@...wei.com>
To: Sudeep Holla <sudeep.holla@....com>
CC: <linux-acpi@...r.kernel.org>, <linux-kernel@...r.kernel.org>, Jassi Brar
	<jassisinghbrar@...il.com>, Adam Young <admiyo@...amperecomputing.com>,
	Robbie King <robbiek@...ghtlabs.com>
Subject: Re: [PATCH v2 07/13] mailbox: pcc: Always map the shared memory
 communication address


在 2025/3/11 19:56, Sudeep Holla 写道:
> On Tue, Mar 11, 2025 at 07:32:34PM +0800, lihuisong (C) wrote:
>> 在 2025/3/6 0:38, Sudeep Holla 写道:
>>> Currently the shared memory communication address was mapped by the
>>> mailbox client drivers leading to all sorts of inconsistencies.
>>>
>>> It also has resulted in the inconsistent attributes used while mapping
>>> the shared memory regions.
>>>
>>> In order to remove/eliminate any issues, let us ensures the shared
>>> memory address is always mapped and unmapped when the PCC channels are
>>> requested and release.
>>>
>>> We need to map them as the ACPI PCCT associates these shared memory
>>> with each channel subspace and may need use the status or the flags in
>>> the headers of those shared memory communication address regions to
>>> manage the transport/channel.
>>>
>>> Since there are no users of pcc_chan_ioremap() and also it is mapped
>>> by default, we can stop exporting it and merge the functionality into
>>> pcc_mbox_request_channel().
>> There are two ioremap for the existing mbox client driver after this patch.
>> The existing mbox client driver would not use this variable, and no one else
>> uses it. So it is safe, right?
> IIUC yes, it should be fine.
>
>> Do we need to make a statement that the two iommaps have no impact on the
>> existing mbox client drivers?
> We can add one, but again it will be become obsolete once we change driver
> to use this new mapping. That's why I didn't add it. We can merge all the
> changes together if that is the concern. I am fine either way.
>
I also tested the case with this patch and no modification of driver.
I didn't find any other issue.
IMO, we should make a statement for this anyway.

/Huisong
>