public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Joshua Crofts via B4 Relay <devnull+joshua.crofts1.gmail.com@kernel.org>
To: "Jonathan Cameron" <jic23@kernel.org>,
	"David Lechner" <dlechner@baylibre.com>,
	"Nuno Sá" <nuno.sa@analog.com>,
	"Andy Shevchenko" <andy@kernel.org>
Cc: linux-iio@vger.kernel.org, linux-kernel@vger.kernel.org,
	 Joshua Crofts <joshua.crofts1@gmail.com>
Subject: [PATCH v3 8/8] iio: light: si1133: prevent race condition on timeout
Date: Wed, 29 Apr 2026 17:04:56 +0200	[thread overview]
Message-ID: <20260429-si1133-checkup-v3-8-469f21d960eb@gmail.com> (raw)
In-Reply-To: <20260429-si1133-checkup-v3-0-469f21d960eb@gmail.com>

From: Joshua Crofts <joshua.crofts1@gmail.com>

Sashiko reported a bug where the si1133_command exits on timeout
without halting the sensor or masking the interrupt. If the sensor
completes the command later, any subsequent command to the sensor
will cause the IRQ handler to complete immediately, returning stale
data to the driver all while the command hasn't finished yet, shifting
all potential reads in the future.

Fix this by masking the IRQ if wait_for_completion_timeout() fails.
When initiating a new command, do a dummy read of the IRQ_STATUS
register and turn the IRQ back on.

Closes: https://sashiko.dev/#/message/20260428-si1133-checkup-v2-5-70ad14bfefe2%40gmail.com
Assisted-by: gemini:gemini-3.1-pro-preview
Signed-off-by: Joshua Crofts <joshua.crofts1@gmail.com>
---
 drivers/iio/light/si1133.c | 19 +++++++++++++++----
 1 file changed, 15 insertions(+), 4 deletions(-)

diff --git a/drivers/iio/light/si1133.c b/drivers/iio/light/si1133.c
index 842a59bc68c20b3206e0c826f86c913f6c66bd7c..91c90f2873ae053eb7cbc8c9c79ce6d91a1ec9dc 100644
--- a/drivers/iio/light/si1133.c
+++ b/drivers/iio/light/si1133.c
@@ -401,8 +401,14 @@ static int si1133_command(struct si1133_data *data, u8 cmd)
 
 	expected_seq = (data->rsp_seq + 1) & SI1133_MAX_CMD_CTR;
 
-	if (cmd == SI1133_CMD_FORCE)
+	if (cmd == SI1133_CMD_FORCE) {
+		/* Flush pending IRQs from a previous timeout. */
+		regmap_read(data->regmap, SI1133_REG_IRQ_STATUS, &resp);
+		regmap_write(data->regmap, SI1133_REG_IRQ_ENABLE,
+			     SI1133_IRQ_CHANNEL_ENABLE);
+
 		reinit_completion(&data->completion);
+	}
 
 	err = regmap_write(data->regmap, SI1133_REG_COMMAND, cmd);
 	if (err) {
@@ -413,8 +419,13 @@ static int si1133_command(struct si1133_data *data, u8 cmd)
 
 	if (cmd == SI1133_CMD_FORCE) {
 		/* wait for irq */
-		if (!wait_for_completion_timeout(&data->completion, timeout))
+		if (!wait_for_completion_timeout(&data->completion, timeout)) {
+			/* Mask the IRQ to prevent delayed interrupt waking up
+			 * any subsequent command.
+			 */
+			regmap_write(data->regmap, SI1133_REG_IRQ_ENABLE, 0);
 			return -ETIMEDOUT;
+		}
 		err = regmap_read(data->regmap, SI1133_REG_RESPONSE0, &resp);
 		if (err)
 			return err;
@@ -431,8 +442,8 @@ static int si1133_command(struct si1133_data *data, u8 cmd)
 				 "Failed to read command 0x%02x, ret=%d\n",
 				 cmd, err);
 			/*
-			 * reset counter on err to prevent sofware and hardware
-			 * counters being out of sync
+			 * Reset counter on err to prevent sofware and hardware
+			 * counters being out of sync.
 			 */
 			si1133_cmd_reset_counter(data);
 

-- 
2.47.3



  parent reply	other threads:[~2026-04-29 15:05 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-04-29 15:04 [PATCH v3 0/8] iio: light: si1133: driver cleanup Joshua Crofts via B4 Relay
2026-04-29 15:04 ` [PATCH v3 1/8] iio: light: si1133: remove unused macros Joshua Crofts via B4 Relay
2026-04-29 19:14   ` Andy Shevchenko
2026-04-30  7:50     ` Joshua Crofts
2026-04-29 15:04 ` [PATCH v3 2/8] iio: light: si1133: prefer complex macros enclosed in parenthesis Joshua Crofts via B4 Relay
2026-04-29 15:04 ` [PATCH v3 3/8] iio: light: si1133: add missing include headers Joshua Crofts via B4 Relay
2026-04-29 19:16   ` Andy Shevchenko
2026-04-30  7:52     ` Joshua Crofts
2026-04-29 15:04 ` [PATCH v3 4/8] iio: light: si1133: group generic <linux/*> headers Joshua Crofts via B4 Relay
2026-04-29 15:04 ` [PATCH v3 5/8] iio: light: si1133: use guard(mutex)() macro Joshua Crofts via B4 Relay
2026-04-29 19:19   ` Andy Shevchenko
2026-04-30  7:55     ` Joshua Crofts
2026-04-29 15:04 ` [PATCH v3 6/8] iio: light: si1133: add local variable for timeout Joshua Crofts via B4 Relay
2026-04-29 19:21   ` Andy Shevchenko
2026-04-29 15:04 ` [PATCH v3 7/8] iio: light: si1133: reset counter to prevent race condition Joshua Crofts via B4 Relay
2026-04-29 19:23   ` Andy Shevchenko
2026-04-30  7:49     ` Joshua Crofts
2026-04-29 15:04 ` Joshua Crofts via B4 Relay [this message]
2026-04-29 19:27   ` [PATCH v3 8/8] iio: light: si1133: prevent race condition on timeout Andy Shevchenko
2026-04-30  7:48     ` Joshua Crofts
2026-04-29 19:28 ` [PATCH v3 0/8] iio: light: si1133: driver cleanup Andy Shevchenko
2026-04-30  7:58   ` Joshua Crofts

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260429-si1133-checkup-v3-8-469f21d960eb@gmail.com \
    --to=devnull+joshua.crofts1.gmail.com@kernel.org \
    --cc=andy@kernel.org \
    --cc=dlechner@baylibre.com \
    --cc=jic23@kernel.org \
    --cc=joshua.crofts1@gmail.com \
    --cc=linux-iio@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nuno.sa@analog.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox