Linux Input/HID development
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Andreas Kemnade" <andreas@kemnade.info>
Cc: linux-input@vger.kernel.org, dmitry.torokhov@gmail.com
Subject: Re: [PATCH] Input: cyttsp5 - Use dev_err_probe() to hide EPROBE_DEFER
Date: Thu, 19 Nov 2026 18:40:33 +0000	[thread overview]
Message-ID: <sashiko-outbox-223799@kernel.org> (raw)
In-Reply-To: <20261119-cyt-errporbe-v1-1-7faa44e821b4@kemnade.info>

Thank you for your contribution! Sashiko AI review found no new issues
in this patch, but noticed 1 related pre-existing issue it does not
introduce:
- [High] Input: cyttsp5 - concurrent regmap reads in cyttsp5_probe() [1]
--

commit c1590d49f726ba5943ef8a369ba24293879c0745
Author: Andreas Kemnade <andreas@kemnade.info>

Input: cyttsp5 - Use dev_err_probe() to hide EPROBE_DEFER

This commit replaces dev_err() calls in cyttsp5_probe() with
dev_err_probe() to suppress error messages on deferred probe.

> diff --git a/drivers/input/touchscreen/cyttsp5.c b/drivers/input/touchscreen/cyttsp5.c
> index d4a78facfa815..82209bfdf828d 100644
> --- a/drivers/input/touchscreen/cyttsp5.c
> +++ b/drivers/input/touchscreen/cyttsp5.c

[ ... ]

> @@ -887,22 +882,16 @@ static int cyttsp5_probe(struct device *dev, struct regmap *regmap, int irq,
>  
>  	error = devm_request_threaded_irq(dev, irq, NULL, cyttsp5_handle_irq,
>  					  IRQF_ONESHOT, name, ts);
> -	if (error) {
> -		dev_err(dev, "unable to request IRQ\n");
> -		return error;
> -	}
> +	if (error)
> +		return dev_err_probe(dev, error, "unable to request IRQ\n");
>  
>  	error = cyttsp5_startup(ts);
> -	if (error) {
> -		dev_err(ts->dev, "Fail initial startup r=%d\n", error);
> -		return error;
> -	}
> +	if (error)
> +		return dev_err_probe(ts->dev, error, "Fail initial startup\n");

[Severity: High]
This problem wasn't introduced by this patch, but can cyttsp5_handle_irq()
race with cyttsp5_startup() on HID_INPUT_REG reads [1]?

devm_request_threaded_irq() unmasks the interrupt before cyttsp5_startup()
runs.  If a pending event or touch triggers the IRQ immediately, both paths
perform unserialized regmap_bulk_read() transfers against HID_INPUT_REG:

CPU0                              CPU1
cyttsp5_probe()
  devm_request_threaded_irq()
                                  cyttsp5_handle_irq()
                                    cyttsp5_read()
                                      regmap_bulk_read() /* read size */
  cyttsp5_startup()
    cyttsp5_deassert_int()
      regmap_bulk_read()          <-- interleaves with two-step read
                                      regmap_bulk_read() /* read data */

This corrupts the sequential HID report read protocol.  In addition, if
cyttsp5_deassert_int() reads a non-empty touch report (size > 2), it returns
-EINVAL and aborts probe:

drivers/input/touchscreen/cyttsp5.c:cyttsp5_deassert_int() {
	...
	error = regmap_bulk_read(ts->regmap, HID_INPUT_REG, buf, sizeof(buf));
	if (error < 0)
		return error;

	size = get_unaligned_le16(&buf[0]);
	if (size == 2 || size == 0)
		return 0;

	return -EINVAL;
}

[1] https://sashiko.dev/#/bug/linux-bab102b6-93cb-4ea5-b362-6d8c8b8866ea

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261119-cyt-errporbe-v1-1-7faa44e821b4@kemnade.info?part=1

      reply	other threads:[~2026-11-19 18:40 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-19 18:34 [PATCH] Input: cyttsp5 - Use dev_err_probe() to hide EPROBE_DEFER Andreas Kemnade
2026-11-19 18:40 ` sashiko-bot [this message]

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=sashiko-outbox-223799@kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=andreas@kemnade.info \
    --cc=dmitry.torokhov@gmail.com \
    --cc=linux-input@vger.kernel.org \
    --cc=sashiko-reviews@lists.linux.dev \
    /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