linux-kernel - [PATCH 1/2] Input: cros-ec-keyb: Only register keyboard if rows/columns exist

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: <20220413033334.1514008-2-swboyd@chromium.org>
Date:   Tue, 12 Apr 2022 20:33:33 -0700
From:   Stephen Boyd <swboyd@...omium.org>
To:     Dmitry Torokhov <dmitry.torokhov@...il.com>
Cc:     linux-kernel@...r.kernel.org, patches@...ts.linux.dev,
        linux-input@...r.kernel.org, chrome-platform@...ts.linux.dev,
        Benson Leung <bleung@...omium.org>,
        Guenter Roeck <groeck@...omium.org>,
        Douglas Anderson <dianders@...omium.org>,
        Hsin-Yi Wang <hsinyi@...omium.org>
Subject: [PATCH 1/2] Input: cros-ec-keyb: Only register keyboard if rows/columns exist

If the device is a detachable, we may still probe this device because
there are some button switches, e.g. volume buttons and power buttons,
registered by this driver. Let's allow the device node to be missing row
and column device properties to indicate that the keyboard matrix
shouldn't be registered. This removes an input device on Trogdor devices
such as Wormdingler that don't have a matrix keyboard, but still have
power and volume buttons. That helps userspace understand there isn't
a keyboard present when the detachable keyboard is disconnected.

Cc: Benson Leung <bleung@...omium.org>
Cc: Guenter Roeck <groeck@...omium.org>
Cc: Douglas Anderson <dianders@...omium.org>
Cc: Hsin-Yi Wang <hsinyi@...omium.org>
Signed-off-by: Stephen Boyd <swboyd@...omium.org>
---

I tried to use mkbp info to query the number of rows and columns, but my
EC firmware doesn't have commit 8505881ed0b9 ("mkbp: Separate MKBP_INFO
host command from the keyboard driver") so it always returns 8 and 13
for the rows and columns. Sigh. With updated firmware we could query it,
or we could rely on DT like we do already.

Originally I was setting the properties to 0, but
matrix_keypad_parse_properties() spits out an error message in that case
and so it seems better to delete the properties and check for their
existence instead. Another alternative would be to change the compatible
to be "google,cros-ec-keyb-switches" or something that indicates there
are only switches and no matrix keyboard.

 drivers/input/keyboard/cros_ec_keyb.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/input/keyboard/cros_ec_keyb.c b/drivers/input/keyboard/cros_ec_keyb.c
index 6534dfca60b4..ac9a953bff02 100644
--- a/drivers/input/keyboard/cros_ec_keyb.c
+++ b/drivers/input/keyboard/cros_ec_keyb.c
@@ -537,6 +537,15 @@ static int cros_ec_keyb_register_matrix(struct cros_ec_keyb *ckdev)
 	u32 key_pos;
 	unsigned int row, col, scancode, n_physmap;

+	/*
+	 * No rows and columns? There isn't a matrix but maybe there are
+	 * switches to register in cros_ec_keyb_register_bs() because this is a
+	 * detachable device.
+	 */
+	if (!device_property_read_bool(dev, "keypad,num-rows") &&
+	    !device_property_read_bool(dev, "keypad,num-cols"))
+		return 0;
+
 	err = matrix_keypad_parse_properties(dev, &ckdev->rows, &ckdev->cols);
 	if (err)
 		return err;
-- 
https://chromeos.dev