[PATCH] comedi: comedi_bond: Track opened devices independently of subdevices

[PATCH] comedi: comedi_bond: Track opened devices independently of subdevices

[Wentao Liang]

bonding_detach() closes devices by walking the bonded subdevice array.
If a device has no digital I/O subdevices, do_dev_config() never adds it
to that array. Its open reference is then lost. The same happens if
configuration fails before the first subdevice entry for a device has
been added, for example when allocating that entry fails.

Keep the device returned by each successful comedi_open_from() in a
separate array indexed by the configuration option. Close those devices
on detach, including attach failure cleanup. This also removes the need
to deduplicate closes for devices with multiple bonded subdevices.

Fixes: 3cf840f65ad5 ("Staging: comedi: add comedi_bond driver")
Cc: stable@vger.kernel.org
Assisted-by: DeepSeek-V4.1-Flash and OpenAI Codex
Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>
---
 drivers/comedi/drivers/comedi_bond.c | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

diff --git a/drivers/comedi/drivers/comedi_bond.c b/drivers/comedi/drivers/comedi_bond.c
index a10cfbd04c72..ac8937550ce9 100644
--- a/drivers/comedi/drivers/comedi_bond.c
+++ b/drivers/comedi/drivers/comedi_bond.c
@@ -55,6 +55,7 @@ struct bonded_device {
 struct comedi_bond_private {
 	char name[256];
 	struct bonded_device **devs;
+	struct comedi_device *opened_devs[COMEDI_NDEVCONFOPTS];
 	unsigned int ndevs;
 	unsigned int nchans;
 };
@@ -215,6 +216,8 @@ static int do_dev_config(struct comedi_device *dev, struct comedi_devconfig *it)
 			return -ENODEV;
 		}
 
+		devpriv->opened_devs[i] = d;
+
 		/* Do DIO, as that's all we support now.. */
 		while ((sdev = comedi_find_subdevice_by_type(d, COMEDI_SUBD_DIO,
 							     sdev + 1)) > -1) {
@@ -312,24 +315,28 @@ static int bonding_attach(struct comedi_device *dev,
 static void bonding_detach(struct comedi_device *dev)
 {
 	struct comedi_bond_private *devpriv = dev->private;
+	unsigned int i;
 
-	if (devpriv && devpriv->devs) {
-		DECLARE_BITMAP(devs_closed, COMEDI_NUM_BOARD_MINORS);
+	if (!devpriv)
+		return;
 
-		memset(&devs_closed, 0, sizeof(devs_closed));
+	if (devpriv->devs) {
 		while (devpriv->ndevs--) {
 			struct bonded_device *bdev;
 
 			bdev = devpriv->devs[devpriv->ndevs];
 			if (!bdev)
 				continue;
-			if (!test_and_set_bit(bdev->minor, devs_closed))
-				comedi_close_from(bdev->dev, dev->minor);
 			kfree(bdev);
 		}
 		kfree(devpriv->devs);
 		devpriv->devs = NULL;
 	}
+
+	for (i = 0; i < ARRAY_SIZE(devpriv->opened_devs); i++) {
+		if (devpriv->opened_devs[i])
+			comedi_close_from(devpriv->opened_devs[i], dev->minor);
+	}
 }
 
 static struct comedi_driver bonding_driver = {
-- 
2.34.1

Re: [PATCH] comedi: comedi_bond: Track opened devices independently of subdevices

[Krzysztof Kozlowski]

On Tue, 03 Nov 2026 06:33:01 +0000, Wentao Liang wrote:
> bonding_detach() closes devices by walking the bonded subdevice array.
> If a device has no digital I/O subdevices, do_dev_config() never adds it
> to that array. Its open reference is then lost. The same happens if
> configuration fails before the first subdevice entry for a device has
> been added, for example when allocating that entry fails.
> 
> Keep the device returned by each successful comedi_open_from() in a
> separate array indexed by the configuration option. Close those devices
> on detach, including attach failure cleanup. This also removes the need
> to deduplicate closes for devices with multiple bonded subdevices.
> 
> Fixes: 3cf840f65ad5 ("Staging: comedi: add comedi_bond driver")
> Cc: stable@vger.kernel.org
> Assisted-by: DeepSeek-V4.1-Flash and OpenAI Codex
> Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>
> ---
>  drivers/comedi/drivers/comedi_bond.c | 17 ++++++++++++-----
>  1 file changed, 12 insertions(+), 5 deletions(-)
> 



Multiple things here:
1. Your team ignored completely previous feedback. Multiple times.

2. You use multiple identities with this email, thus I doubt
   that we speak with an actual person.

3. Your bot does not even bother running checkpatch, right?

4. Nothing of this was reproduced nor tested.

5. Finally, the same feedback:
You sent multiple independent patches, to multiple independent
subsystems. The amount of these patches clearly suggest this was
AI generated and most likely not tested.

More importantly, you sent all this work without properly organizing
relevant patches into patchsets. This makes reviewing difficult
and might cause multiple reviewers to address the same issue.
Replying to the entire set is impossible and requires handling each
patch independently, instead of applying or discarding the set.
Maintainers also won't see the bigger picture of your work. Quite
worrying.

This is on the verge of hostile patch: bomb us with so many
contributions, we won't be able to handle them in efficient manner,
like responding ONCE to ask you to slow down.  Considering all this
is untested and LLM generated, I have even more doubts whether this
should be considered for review.

Please read kernel documentation BEFORE posting more work. It will
explain you how to identify subsystems, how to organize your work per
subsystem (so a patchset grouping multiple patches with a short cover
letter), how to document usage of LLM and what you should not do
if this was posted in a good faith.

Best regards,
Krzysztof

Re: [PATCH] comedi: comedi_bond: Track opened devices independently of subdevices

[Ian Abbott]

On 03/11/2026 06:33, Wentao Liang wrote:
> bonding_detach() closes devices by walking the bonded subdevice array.
> If a device has no digital I/O subdevices, do_dev_config() never adds it
> to that array. Its open reference is then lost. The same happens if
> configuration fails before the first subdevice entry for a device has
> been added, for example when allocating that entry fails.
> 
> Keep the device returned by each successful comedi_open_from() in a
> separate array indexed by the configuration option. Close those devices
> on detach, including attach failure cleanup. This also removes the need
> to deduplicate closes for devices with multiple bonded subdevices.
> 
> Fixes: 3cf840f65ad5 ("Staging: comedi: add comedi_bond driver")
> Cc: stable@vger.kernel.org
> Assisted-by: DeepSeek-V4.1-Flash and OpenAI Codex
> Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>
> ---
>   drivers/comedi/drivers/comedi_bond.c | 17 ++++++++++++-----
>   1 file changed, 12 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/comedi/drivers/comedi_bond.c b/drivers/comedi/drivers/comedi_bond.c
> index a10cfbd04c72..ac8937550ce9 100644
> --- a/drivers/comedi/drivers/comedi_bond.c
> +++ b/drivers/comedi/drivers/comedi_bond.c
> @@ -55,6 +55,7 @@ struct bonded_device {
>   struct comedi_bond_private {
>   	char name[256];
>   	struct bonded_device **devs;
> +	struct comedi_device *opened_devs[COMEDI_NDEVCONFOPTS];
>   	unsigned int ndevs;
>   	unsigned int nchans;
>   };
> @@ -215,6 +216,8 @@ static int do_dev_config(struct comedi_device *dev, struct comedi_devconfig *it)
>   			return -ENODEV;
>   		}
>   
> +		devpriv->opened_devs[i] = d;
> +
>   		/* Do DIO, as that's all we support now.. */
>   		while ((sdev = comedi_find_subdevice_by_type(d, COMEDI_SUBD_DIO,
>   							     sdev + 1)) > -1) {
> @@ -312,24 +315,28 @@ static int bonding_attach(struct comedi_device *dev,
>   static void bonding_detach(struct comedi_device *dev)
>   {
>   	struct comedi_bond_private *devpriv = dev->private;
> +	unsigned int i;
>   
> -	if (devpriv && devpriv->devs) {
> -		DECLARE_BITMAP(devs_closed, COMEDI_NUM_BOARD_MINORS);
> +	if (!devpriv)
> +		return;
>   
> -		memset(&devs_closed, 0, sizeof(devs_closed));
> +	if (devpriv->devs) {
>   		while (devpriv->ndevs--) {
>   			struct bonded_device *bdev;
>   
>   			bdev = devpriv->devs[devpriv->ndevs];
>   			if (!bdev)
>   				continue;
> -			if (!test_and_set_bit(bdev->minor, devs_closed))
> -				comedi_close_from(bdev->dev, dev->minor);
>   			kfree(bdev);
>   		}
>   		kfree(devpriv->devs);
>   		devpriv->devs = NULL;
>   	}
> +
> +	for (i = 0; i < ARRAY_SIZE(devpriv->opened_devs); i++) {
> +		if (devpriv->opened_devs[i])
> +			comedi_close_from(devpriv->opened_devs[i], dev->minor);
> +	}
>   }
>   
>   static struct comedi_driver bonding_driver = {

The problem with the lost open references could also be fixed without 
adding extra storage for an array in struct comedi_bond_private by 
calling comedi_close_from(d, dev->minor) from bonding_attach() if the 
device opened by comedi_open_from() does not have a suitable subdevice, 
or an error is found in the inner 'while' loop.

-- 
-=( Ian Abbott <abbotti@mev.co.uk> || MEV Ltd. is a company  )=-
-=( registered in England & Wales.  Regd. number: 02862268.  )=-
-=( Regd. addr.: S11 & 12 Building 67, Europa Business Park, )=-
-=( Bird Hall Lane, STOCKPORT, SK3 0XA, UK. || www.mev.co.uk )=-