netdev - Re: [PATCH v11 0/1] MCTP Over PCC Transport

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: <e4c5b40a906a560f9a333d8e9d7c556ad99d63fa.camel@codeconstruct.com.au>
Date: Mon, 06 Jan 2025 09:05:48 +0800
From: Jeremy Kerr <jk@...econstruct.com.au>
To: admiyo@...amperecomputing.com
Cc: netdev@...r.kernel.org, linux-kernel@...r.kernel.org, Matt Johnston
 <matt@...econstruct.com.au>, "David S . Miller" <davem@...emloft.net>, Eric
 Dumazet <edumazet@...gle.com>, Jakub Kicinski <kuba@...nel.org>, Paolo
 Abeni <pabeni@...hat.com>,  Sudeep Holla <sudeep.holla@....com>, Jonathan
 Cameron <Jonathan.Cameron@...wei.com>, Huisong Li <lihuisong@...wei.com>
Subject: Re: [PATCH v11 0/1] MCTP Over PCC Transport

Hi Adam,
> Changes in V11:
> - Switch Big Endian data types to machine local for PCC header

This seems suspicious; the concept of "machine local" may not be
consistent between channel endpoints (ie, system firmware and this
driver). Looking at DSP0292, and the PCC channel spec, it looks like
these should be explicitly little-endian, no?

The warnings you were seeing in v10 seemed to be an issue of direct
accesses to the __be32 types in struct mctp_pcc_hdr - ie., without
be32_to_cpu() accessors. If you keep the endian-annotated types (but
__le32 instead), and use the appropriate le32_to_cpu() accessors, that
should address those warnings, while keeping the driver endian-safe.

Cheers,

Jeremy