linux-kernel - [PATCH v3 0/5] platform/chrome: Fix MEC concurrency problems for Framework Laptop

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-next>] [day] [month] [year] [list]

Message-ID: <20240605063351.14836-1-ben@jubnut.com>
Date: Wed,  5 Jun 2024 07:33:46 +0100
From: Ben Walsh <ben@...nut.com>
To: Benson Leung <bleung@...omium.org>,
	Tzung-Bi Shih <tzungbi@...nel.org>,
	Guenter Roeck <groeck@...omium.org>,
	"Dustin L. Howett" <dustin@...ett.net>,
	Kieran Levin <ktl@...me.work>,
	Thomas Weißschuh <linux@...ssschuh.net>,
	Mario Limonciello <mario.limonciello@....com>,
	chrome-platform@...ts.linux.dev,
	linux-kernel@...r.kernel.org
Cc: Ben Walsh <ben@...nut.com>
Subject: [PATCH v3 0/5] platform/chrome: Fix MEC concurrency problems for Framework Laptop

Framework Laptops with the Microchip EC have a problem where the EC
"stops working" after a while. Symptoms include the Fn key not
working, and "bad packet checksum" errors appearing in the system log.

The problem is caused by ACPI code which accesses the Microchip EC
(MEC) memory using the Microchip EMI protocol. It uses an AML mutex to
prevent concurrent access. But the cros_ec_lpc driver is not aware of
this mutex. The ACPI code and LPC driver both attempt to talk to the
EC at the same time, messing up communication with the EC.

The solution is to have the cros_ec_lpc_mec code find and use the AML
mutex. But to make it all work we have to do a few more things:

  * Allow the cros_ec_lpc_mec code to return error codes in case it
    can't lock the mutex.

  * Have the cros_ec_lpc code find the correct ACPI device (PNP0C09)
    and then the AML mutex itself.

  * Use the quirks mechanism to specify the AML mutex name.

Tested-on: link, hx20, azalea
Tested-by: Dustin L. Howett <dustin@...ett.net>

Changes in v3:

  * Use ACPI_COMPANION instead of adev variable.
  * Stricter error checking in probe.
  * Remove a few blank lines.
  * Rebase to latest ChromeOS for-next branch.

Changes in v2:

  * Put ACPI id in a quirk, not the acpi_match_table.
  * Add return code check for mutex unlock.
  * Remove "in_range" check which belongs in a separate patch.
  * Use "ret" not "sum" variable (same value but clearer).
  * Remove excessive error logging.
  * Rebase to latest ChromeOS for-next branch.

Ben Walsh (5):
  platform/chrome: cros_ec_lpc: MEC access can return error code
  platform/chrome: cros_ec_lpc: MEC access can use an AML mutex
  platform/chrome: cros_ec_lpc: Add a new quirk for ACPI id
  platform/chrome: cros_ec_lpc: Add a new quirk for AML mutex
  platform/chrome: cros_ec_lpc: Add quirks for Framework Laptop

 drivers/platform/chrome/cros_ec_lpc.c      | 206 +++++++++++++++------
 drivers/platform/chrome/cros_ec_lpc_mec.c  |  85 ++++++++-
 drivers/platform/chrome/cros_ec_lpc_mec.h  |  18 +-
 drivers/platform/chrome/wilco_ec/mailbox.c |  22 ++-
 4 files changed, 261 insertions(+), 70 deletions(-)

base-commit: 106d6739823369c734a8fc3b13634274eee4f60e
-- 
2.45.1