linux-kernel - Re: [PATCH 0/9] platform/surface: aggregator: Improve target/source handling in SSH messages

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <c09c9cef-14ac-2ab3-5e01-13189823a053@redhat.com>
Date:   Thu, 8 Dec 2022 17:03:38 +0100
From:   Hans de Goede <hdegoede@...hat.com>
To:     Maximilian Luz <luzmaximilian@...il.com>,
        Jiri Kosina <jikos@...nel.org>,
        Benjamin Tissoires <benjamin.tissoires@...hat.com>,
        Sebastian Reichel <sre@...nel.org>
Cc:     Mark Gross <markgross@...nel.org>,
        Jonathan Corbet <corbet@....net>,
        platform-driver-x86@...r.kernel.org, linux-doc@...r.kernel.org,
        linux-input@...r.kernel.org, linux-pm@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH 0/9] platform/surface: aggregator: Improve target/source
 handling in SSH messages

Hi Maximilian,

On 12/2/22 23:33, Maximilian Luz wrote:
> We have some new insights into the Serial Hub protocol, obtained through
> reverse engineering. In particular, regarding the command structure. The
> input/output target IDs actually represent source and target IDs of
> (what looks like) physical entities (specifically: host, SAM EC, KIP EC,
> debug connector, and SurfLink connector).
> 
> This series aims to improve handling of messages with regards to those
> new findings and, mainly, improve clarity of the documentation and usage
> around those fields.
> 
> See the discussion in
> 
>     https://github.com/linux-surface/surface-aggregator-module/issues/64
> 
> for more details.
> 
> There are a couple of standouts:
> 
> - Patch 1 ensures that we only handle commands actually intended for us.
>   It's possible that we receive messages not intended for us when we
>   enable debugging. I've kept it intentionally minimal to simplify
>   backporting. The rest of the series patch 9 focuses more on clarity
>   and documentation, which is probably too much to backport.
> 
> - Patch 8 touches on multiple subsystems. The intention is to enforce
>   proper usage and documentation of target IDs in the SSAM_SDEV() /
>   SSAM_VDEV() macros. As it directly touches those macros I
>   unfortunately can't split it up by subsystem.
> 
> - Patch 9 is a loosely connected cleanup for consistency.

Thank you for the patches. Unfortunately I don't have time atm to
review this.

And the next 2 weeks are the merge window, followed by 2 weeks
of christmas vacation.

So I'm afraid that I likely won't get around to reviewing
this until the week of January 9th.

> Hans, Jiri, Benjamin, Sebastian: While patch 8 ("platform/surface:
> aggregator: Enforce use of target-ID enum in device ID macros") touches
> multiple subsystems, it should be possible to take the whole series
> through the pdx86 tree. The changes in other subsystems are fairly
> limited.

I agree that it will be best to take all of this upstream through the
pdx86 tree. Sebastian thank you for the ack for patch 8/9.

Jiri or Benjamin may we have your ack for merging patch 7/9 + 8/9
through the pdx86 tree ?

Regards,

Hans




> Maximilian Luz (9):
>   platform/surface: aggregator: Ignore command messages not intended for
>     us
>   platform/surface: aggregator: Improve documentation and handling of
>     message target and source IDs
>   platform/surface: aggregator: Add target and source IDs to command
>     trace events
>   platform/surface: aggregator_hub: Use target-ID enum instead of
>     hard-coding values
>   platform/surface: aggregator_tabletsw: Use target-ID enum instead of
>     hard-coding values
>   platform/surface: dtx: Use target-ID enum instead of hard-coding
>     values
>   HID: surface-hid: Use target-ID enum instead of hard-coding values
>   platform/surface: aggregator: Enforce use of target-ID enum in device
>     ID macros
>   platform/surface: aggregator_registry: Fix target-ID of base-hub
> 
>  .../driver-api/surface_aggregator/client.rst  |  4 +-
>  .../driver-api/surface_aggregator/ssh.rst     | 36 ++++-----
>  drivers/hid/surface-hid/surface_hid.c         |  2 +-
>  drivers/hid/surface-hid/surface_kbd.c         |  2 +-
>  .../platform/surface/aggregator/controller.c  | 12 +--
>  .../platform/surface/aggregator/ssh_msgb.h    |  4 +-
>  .../surface/aggregator/ssh_request_layer.c    | 15 ++++
>  drivers/platform/surface/aggregator/trace.h   | 73 +++++++++++++++++--
>  .../platform/surface/surface_aggregator_hub.c |  8 +-
>  .../surface/surface_aggregator_registry.c     |  2 +-
>  .../surface/surface_aggregator_tabletsw.c     | 10 +--
>  drivers/platform/surface/surface_dtx.c        | 20 ++---
>  .../surface/surface_platform_profile.c        |  2 +-
>  drivers/power/supply/surface_battery.c        |  4 +-
>  drivers/power/supply/surface_charger.c        |  2 +-
>  include/linux/surface_aggregator/controller.h |  4 +-
>  include/linux/surface_aggregator/device.h     | 50 ++++++-------
>  include/linux/surface_aggregator/serial_hub.h | 40 ++++++----
>  18 files changed, 191 insertions(+), 99 deletions(-)
>