mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Jonathan Cameron <jic23@kernel.org>
To: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Cc: jikos@kernel.org, bentiss@kernel.org,
	linux-input@vger.kernel.org, linux-iio@vger.kernel.org,
	linux-kernel@vger.kernel.org, simeon@maryasin.name,
	lixu.zhang@intel.com, crofts@kernel.org
Subject: Re: [PATCH v2 4/7] HID: sensor: custom: Fix state lifetime during removal
Date: Sat, 7 Nov 2026 20:21:45 +0000	[thread overview]
Message-ID: <20261107202145.0348a033@jic23-hlaptop> (raw)
In-Reply-To: <20261104201403.2039921-5-srinivas.pandruvada@linux.intel.com>

On Wed,  4 Nov 2026 12:14:00 -0800
Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com> wrote:

> Open misc device files can outlive platform-device removal. Since the
> devm-allocated state can be freed while a reader is queued on its wait
> queue, subsequent file operations may access freed memory.
> 
> Allocate the state independently of devres and give every open
> file a reference to keep the FIFO and state alive after removal.
> 
> On removal, mark the device as disconnected and wake readers; make reads
> return -ENODEV and polls report hangup and error. Free the FIFO and state
> when the final reference is released. Unregister the sensor hub callback
> before tearing down the interface so in-flight raw-event processing
> finishes before the sensor state can be freed.
> 
> Fixes: 4a7de0519df5 ("HID: sensor: Custom and Generic sensor support")
> Reported-by: Семён Марьясин <simeon@maryasin.name>
> Closes: https://lore.kernel.org/all/CAM2dg7ewfVJSP4rFyhkCB1Rmd1+X56sAg=eUdbPthG2xrLqwQQ@mail.gmail.com/
> Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>

Looks like fairly standard fix for this style of lifetime issue.

One trivial comment inline.


>  
>  /*
> @@ -978,8 +1004,7 @@ static int hid_sensor_custom_probe(struct platform_device *pdev)
>  	int ret;
>  	const struct hid_sensor_custom_match *match;
>  
> -	sensor_inst = devm_kzalloc(&pdev->dev, sizeof(*sensor_inst),
> -				   GFP_KERNEL);
> +	sensor_inst = kzalloc_obj(*sensor_inst);
>  	if (!sensor_inst)
>  		return -ENOMEM;
>  
> @@ -989,6 +1014,7 @@ static int hid_sensor_custom_probe(struct platform_device *pdev)
>  	sensor_inst->hsdev = hsdev;
>  	sensor_inst->pdev = pdev;
>  	mutex_init(&sensor_inst->mutex);
> +	kref_init(&sensor_inst->ref);
>  	platform_set_drvdata(pdev, sensor_inst);
>  
>  	ret = hid_sensor_custom_get_known(hsdev, &match);
> @@ -1000,7 +1026,7 @@ static int hid_sensor_custom_probe(struct platform_device *pdev)
>  		if (ret) {
>  			dev_err(&pdev->dev,
>  				"register_platform_device failed\n");
> -			return ret;
> +			goto err_free;
>  		}
>  
>  		return 0;
> @@ -1010,7 +1036,7 @@ static int hid_sensor_custom_probe(struct platform_device *pdev)
>  					   &sensor_inst->callbacks);
>  	if (ret < 0) {
>  		dev_err(&pdev->dev, "callback reg failed\n");
> -		return ret;
> +		goto err_free;

Trivial but given it might free I'd all that label err_put

>  	}
>  
>  	ret = hid_sensor_custom_add_attributes(sensor_inst);
> @@ -1035,6 +1061,8 @@ static int hid_sensor_custom_probe(struct platform_device *pdev)
>  	hid_sensor_custom_remove_attributes(sensor_inst);
>  err_remove_callback:
>  	sensor_hub_remove_callback(hsdev, hsdev->usage);
> +err_free:
> +	kref_put(&sensor_inst->ref, hid_sensor_custom_free);
>  
>  	return ret;
>  }
> @@ -1046,9 +1074,10 @@ static void hid_sensor_custom_remove(struct platform_device *pdev)
>  
>  	if (sensor_inst->custom_pdev) {
>  		platform_device_unregister(sensor_inst->custom_pdev);
> -		return;
> +		goto out;
>  	}
>  
> +	sensor_hub_remove_callback(hsdev, hsdev->usage);
>  	hid_sensor_custom_dev_if_remove(sensor_inst);
>  	/* Remove enable_sensor first as it uses fields via power_state/report_state. */
>  	sysfs_remove_group(&sensor_inst->pdev->dev.kobj,
> @@ -1061,7 +1090,8 @@ static void hid_sensor_custom_remove(struct platform_device *pdev)
>  	}
>  	mutex_unlock(&sensor_inst->mutex);
>  	hid_sensor_custom_remove_attributes(sensor_inst);
> -	sensor_hub_remove_callback(hsdev, hsdev->usage);
> +out:
> +	kref_put(&sensor_inst->ref, hid_sensor_custom_free);
>  }
>  
>  static const struct platform_device_id hid_sensor_custom_ids[] = {


  reply	other threads:[~2026-11-07 20:21 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-04 20:13 [PATCH v2 0/7] HID: sensor: custom: use-after-freec after remove Srinivas Pandruvada
2026-11-04 20:13 ` [PATCH v2 1/7] HID: sensor: hub: Synchronize callback removal with raw events Srinivas Pandruvada
2026-11-05  9:56   ` Joshua Crofts
2026-11-07 20:13   ` Jonathan Cameron
2026-11-04 20:13 ` [PATCH v2 2/7] HID: sensor: custom: Set owner for misc device Srinivas Pandruvada
2026-11-05  9:57   ` Joshua Crofts
2026-11-04 20:13 ` [PATCH v2 3/7] HID: sensor: custom: Disable sensor before hub close on removal Srinivas Pandruvada
2026-11-05 10:21   ` Joshua Crofts
2026-11-04 20:14 ` [PATCH v2 4/7] HID: sensor: custom: Fix state lifetime during removal Srinivas Pandruvada
2026-11-07 20:21   ` Jonathan Cameron [this message]
2026-11-04 20:14 ` [PATCH v2 5/7] HID: sensor: custom: Serialize misc device reads Srinivas Pandruvada
2026-11-04 20:14 ` [PATCH v2 6/7] HID: sensor: custom: Register enable attribute after misc device Srinivas Pandruvada
2026-11-07 20:16   ` Jonathan Cameron
2026-11-04 20:14 ` [PATCH v2 7/7] HID: sensor: hub: Remove clients before stopping hardware Srinivas Pandruvada
2026-11-07 20:08 ` [PATCH v2 0/7] HID: sensor: custom: use-after-freec after remove Jonathan Cameron

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=20261107202145.0348a033@jic23-hlaptop \
    --to=jic23@kernel.org \
    --cc=bentiss@kernel.org \
    --cc=crofts@kernel.org \
    --cc=jikos@kernel.org \
    --cc=linux-iio@vger.kernel.org \
    --cc=linux-input@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lixu.zhang@intel.com \
    --cc=simeon@maryasin.name \
    --cc=srinivas.pandruvada@linux.intel.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

all inboxes | Powered by JetHome®