Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Enrik Berkhan <Enrik.Berkhan@inka.de>
To:     linux-input@vger.kernel.org
Cc:     linux-i2c@vger.kernel.org, linux-kernel@vger.kernel.org,
        Rishi Gupta <gupt21@gmail.com>,
        Enrik Berkhan <Enrik.Berkhan@inka.de>
Subject: [PATCH v2 0/3] Fixes for the mcp2221 HID-to-I2C-bridge driver
Date:   Thu,  3 Nov 2022 23:27:11 +0100
Message-Id: <20221103222714.21566-1-Enrik.Berkhan@inka.de>
In-Reply-To: <20220926202239.16379-1-Enrik.Berkhan@inka.de>
References: <20220926202239.16379-1-Enrik.Berkhan@inka.de>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

In this patch series are fixes for issues I found during recent tests of
an MCP2221 board.

- you can confuse the kernel driver when using the chip from user mode
  via /dev/hidrawX, typically leading to a NULL pointer dereference in
  the driver's HID raw event handler

- the driver needs > 15s to initialize because the HID raw handling is
  not enabled during initialization of the GPIO part

- the rxbuf pointer can become invalid or even stale if the device would
  send unsolicited reports

Changes in v2:

- removed: data shared with the bottom half code is not protected from
  concurrent access

Feedback if this is actually needed or not would be appreciated.

- rebased on linux-hid/for-6.2/mcp2221

Enrik Berkhan (3):
  HID: mcp2221: don't connect hidraw
  HID: mcp2221: enable HID I/O during GPIO probe
  HID: mcp2221: avoid stale rxbuf pointer

 drivers/hid/hid-mcp2221.c | 51 +++++++++++++++++++++++++++++++--------
 1 file changed, 41 insertions(+), 10 deletions(-)

base-commit: 3d74c9eca1a2bda03e45f18d13154ac3e0dfba85
-- 
2.34.1