linux-kernel - [PATCH] Input: elan_i2c - do not clobber interrupt trigger on x86

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: <20170928170902.GA18201@dtor-ws>
Date:   Thu, 28 Sep 2017 10:09:02 -0700
From:   Dmitry Torokhov <dmitry.torokhov@...il.com>
To:     linux-input@...r.kernel.org
Cc:     KT Liao <kt.liao@....com.tw>, Guenter Roeck <linux@...ck-us.net>,
        Aaron Durbin <adurbin@...gle.com>,
        Craig Bergstrom <craigb@...gle.com>,
        linux-kernel@...r.kernel.org
Subject: [PATCH] Input: elan_i2c - do not clobber interrupt trigger on x86

On x86 we historically used falling edge interrupts in the driver
because that's how first Chrome devices were configured. They also
did not use ACPI to enumerate I2C devices (because back then there
was no kernel support for that), so trigger was hard-coded in the
driver. However the controller behavior is much more reliable if
we use level triggers, and that is how we configured ARM devices,
and how want to configure newer x86 devices as well. All newer
x86 boxes have their I2C devices enumerated in ACPI.

Let's see if platform code (ACPI, DT) described interrupt and
specified particular trigger type, and if so, let's use it instead
of always clobbering trigger with IRQF_TRIGGER_FALLING. We will
still use this trigger type as a fallback if platform code left
interrupt trigger unconfigured.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=196761
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@...il.com>
---
 drivers/input/mouse/elan_i2c_core.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/input/mouse/elan_i2c_core.c b/drivers/input/mouse/elan_i2c_core.c
index a2d2077303cb..c33f2cce6ba9 100644
--- a/drivers/input/mouse/elan_i2c_core.c
+++ b/drivers/input/mouse/elan_i2c_core.c
@@ -26,6 +26,7 @@
 #include <linux/init.h>
 #include <linux/input/mt.h>
 #include <linux/interrupt.h>
+#include <linux/irq.h>
 #include <linux/module.h>
 #include <linux/slab.h>
 #include <linux/kernel.h>
@@ -1141,10 +1142,13 @@ static int elan_probe(struct i2c_client *client,
 		return error;

 	/*
-	 * Systems using device tree should set up interrupt via DTS,
-	 * the rest will use the default falling edge interrupts.
+	 * Platform code (ACPI, DTS) should normally set up interrupt
+	 * for us, but in case it did not let's fall back to using falling
+	 * edge to be compatible with older Chromebooks.
 	 */
-	irqflags = dev->of_node ? 0 : IRQF_TRIGGER_FALLING;
+	irqflags = irq_get_trigger_type(client->irq);
+	if (!irqflags)
+		irqflags = IRQF_TRIGGER_FALLING;

 	error = devm_request_threaded_irq(dev, client->irq, NULL, elan_isr,
 					  irqflags | IRQF_ONESHOT,
-- 
2.14.2.822.g60be5d43e6-goog

-- 
Dmitry