Re: [PATCH v4 2/4] Bluetooth: Rework le_scan_restart for hci_sync

Re: [PATCH v4 2/4] Bluetooth: Rework le_scan_restart for hci_sync

[Stefan Agner]

Hi Brian, hi all,

We experienced quite some Bluetooth issues after moving from Linux 5.15
to 6.1 on Home Assistant OS, especially on Intel NUC type systems (which
is a popular choice in our community, so it might just be that). When
continuously scanning/listening for BLE packets, the packet flow
suddenly ends. Depending on which and how many devices (possibly also
other factors) within minutes or hours.

Jan (in cc) was able to bisect the issue, and was able to pinpoint the
problem to this change.

Meanwhile I was able to confirm, that reverting this single commit on
the latest 6.1.34 seems to resolve the issue.

I've reviewed the change and surrounding code, and one thing I've
noticed is that the if statement to set cp.filter_dup in
hci_le_set_ext_scan_enable_sync and hci_le_set_scan_enable_sync are
different. Not sure if that needs to be the way it is, but my outside
gut feeling says hci_le_set_ext_scan_enable_sync should use "if (val &&
hci_dev_test_flag(hdev, HCI_MESH))" as well. 

However, that did not fix the problem (but maybe it is wrong
nonetheless?).

Anyone has an idea what could be the problem here?

--
Stefan

On 2022-07-27 15:58, Brian Gix wrote:
> le_scan_restart delayed work queue was running as a deprecated
> hci_request instead of on the newer thread-safe hci_sync mechanism.
> 
> Signed-off-by: Brian Gix <brian.gix@intel.com>
> ---
>  net/bluetooth/hci_request.c | 89 -------------------------------------
>  net/bluetooth/hci_sync.c    | 75 +++++++++++++++++++++++++++++++
>  2 files changed, 75 insertions(+), 89 deletions(-)
> 
> diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c
> index 32fefaa0d3ca..114af7350363 100644
> --- a/net/bluetooth/hci_request.c
> +++ b/net/bluetooth/hci_request.c
> @@ -1975,92 +1975,6 @@ int hci_abort_conn(struct hci_conn *conn, u8 reason)
>  	return 0;
>  }
>  
> -static int le_scan_restart(struct hci_request *req, unsigned long opt)
> -{
> -	struct hci_dev *hdev = req->hdev;
> -
> -	/* If controller is not scanning we are done. */
> -	if (!hci_dev_test_flag(hdev, HCI_LE_SCAN))
> -		return 0;
> -
> -	if (hdev->scanning_paused) {
> -		bt_dev_dbg(hdev, "Scanning is paused for suspend");
> -		return 0;
> -	}
> -
> -	hci_req_add_le_scan_disable(req, false);
> -
> -	if (use_ext_scan(hdev)) {
> -		struct hci_cp_le_set_ext_scan_enable ext_enable_cp;
> -
> -		memset(&ext_enable_cp, 0, sizeof(ext_enable_cp));
> -		ext_enable_cp.enable = LE_SCAN_ENABLE;
> -		ext_enable_cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE;
> -
> -		hci_req_add(req, HCI_OP_LE_SET_EXT_SCAN_ENABLE,
> -			    sizeof(ext_enable_cp), &ext_enable_cp);
> -	} else {
> -		struct hci_cp_le_set_scan_enable cp;
> -
> -		memset(&cp, 0, sizeof(cp));
> -		cp.enable = LE_SCAN_ENABLE;
> -		cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE;
> -		hci_req_add(req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(cp), &cp);
> -	}
> -
> -	return 0;
> -}
> -
> -static void le_scan_restart_work(struct work_struct *work)
> -{
> -	struct hci_dev *hdev = container_of(work, struct hci_dev,
> -					    le_scan_restart.work);
> -	unsigned long timeout, duration, scan_start, now;
> -	u8 status;
> -
> -	bt_dev_dbg(hdev, "");
> -
> -	hci_req_sync(hdev, le_scan_restart, 0, HCI_CMD_TIMEOUT, &status);
> -	if (status) {
> -		bt_dev_err(hdev, "failed to restart LE scan: status %d",
> -			   status);
> -		return;
> -	}
> -
> -	hci_dev_lock(hdev);
> -
> -	if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) ||
> -	    !hdev->discovery.scan_start)
> -		goto unlock;
> -
> -	/* When the scan was started, hdev->le_scan_disable has been queued
> -	 * after duration from scan_start. During scan restart this job
> -	 * has been canceled, and we need to queue it again after proper
> -	 * timeout, to make sure that scan does not run indefinitely.
> -	 */
> -	duration = hdev->discovery.scan_duration;
> -	scan_start = hdev->discovery.scan_start;
> -	now = jiffies;
> -	if (now - scan_start <= duration) {
> -		int elapsed;
> -
> -		if (now >= scan_start)
> -			elapsed = now - scan_start;
> -		else
> -			elapsed = ULONG_MAX - scan_start + now;
> -
> -		timeout = duration - elapsed;
> -	} else {
> -		timeout = 0;
> -	}
> -
> -	queue_delayed_work(hdev->req_workqueue,
> -			   &hdev->le_scan_disable, timeout);
> -
> -unlock:
> -	hci_dev_unlock(hdev);
> -}
> -
>  bool hci_req_stop_discovery(struct hci_request *req)
>  {
>  	struct hci_dev *hdev = req->hdev;
> @@ -2158,7 +2072,6 @@ int hci_req_configure_datapath(struct hci_dev
> *hdev, struct bt_codec *codec)
>  
>  void hci_request_setup(struct hci_dev *hdev)
>  {
> -	INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work);
>  	INIT_DELAYED_WORK(&hdev->adv_instance_expire, adv_timeout_expire);
>  	INIT_DELAYED_WORK(&hdev->interleave_scan, interleave_scan_work);
>  }
> @@ -2167,8 +2080,6 @@ void hci_request_cancel_all(struct hci_dev *hdev)
>  {
>  	__hci_cmd_sync_cancel(hdev, ENODEV);
>  
> -	cancel_delayed_work_sync(&hdev->le_scan_restart);
> -
>  	if (hdev->adv_instance_timeout) {
>  		cancel_delayed_work_sync(&hdev->adv_instance_expire);
>  		hdev->adv_instance_timeout = 0;
> diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
> index 7dae2ee1bb82..19d57ec0feb8 100644
> --- a/net/bluetooth/hci_sync.c
> +++ b/net/bluetooth/hci_sync.c
> @@ -392,6 +392,79 @@ static void le_scan_disable(struct work_struct *work)
>  	hci_dev_unlock(hdev);
>  }
>  
> +static int hci_le_set_scan_enable_sync(struct hci_dev *hdev, u8 val,
> +				       u8 filter_dup);
> +static int hci_le_scan_restart_sync(struct hci_dev *hdev)
> +{
> +	/* If controller is not scanning we are done. */
> +	if (!hci_dev_test_flag(hdev, HCI_LE_SCAN))
> +		return 0;
> +
> +	if (hdev->scanning_paused) {
> +		bt_dev_dbg(hdev, "Scanning is paused for suspend");
> +		return 0;
> +	}
> +
> +	hci_le_set_scan_enable_sync(hdev, LE_SCAN_DISABLE, 0x00);
> +	return hci_le_set_scan_enable_sync(hdev, LE_SCAN_ENABLE,
> +					   LE_SCAN_FILTER_DUP_ENABLE);
> +}
> +
> +static int le_scan_restart_sync(struct hci_dev *hdev, void *data)
> +{
> +	return hci_le_scan_restart_sync(hdev);
> +}
> +
> +static void le_scan_restart(struct work_struct *work)
> +{
> +	struct hci_dev *hdev = container_of(work, struct hci_dev,
> +					    le_scan_restart.work);
> +	unsigned long timeout, duration, scan_start, now;
> +	int status;
> +
> +	bt_dev_dbg(hdev, "");
> +
> +	hci_dev_lock(hdev);
> +
> +	status = hci_cmd_sync_queue(hdev, le_scan_restart_sync, NULL, NULL);
> +	if (status) {
> +		bt_dev_err(hdev, "failed to restart LE scan: status %d",
> +			   status);
> +		goto unlock;
> +	}
> +
> +	if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) ||
> +	    !hdev->discovery.scan_start)
> +		goto unlock;
> +
> +	/* When the scan was started, hdev->le_scan_disable has been queued
> +	 * after duration from scan_start. During scan restart this job
> +	 * has been canceled, and we need to queue it again after proper
> +	 * timeout, to make sure that scan does not run indefinitely.
> +	 */
> +	duration = hdev->discovery.scan_duration;
> +	scan_start = hdev->discovery.scan_start;
> +	now = jiffies;
> +	if (now - scan_start <= duration) {
> +		int elapsed;
> +
> +		if (now >= scan_start)
> +			elapsed = now - scan_start;
> +		else
> +			elapsed = ULONG_MAX - scan_start + now;
> +
> +		timeout = duration - elapsed;
> +	} else {
> +		timeout = 0;
> +	}
> +
> +	queue_delayed_work(hdev->req_workqueue,
> +			   &hdev->le_scan_disable, timeout);
> +
> +unlock:
> +	hci_dev_unlock(hdev);
> +}
> +
>  void hci_cmd_sync_init(struct hci_dev *hdev)
>  {
>  	INIT_WORK(&hdev->cmd_sync_work, hci_cmd_sync_work);
> @@ -400,6 +473,7 @@ void hci_cmd_sync_init(struct hci_dev *hdev)
>  
>  	INIT_WORK(&hdev->cmd_sync_cancel_work, hci_cmd_sync_cancel_work);
>  	INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable);
> +	INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart);
>  }
>  
>  void hci_cmd_sync_clear(struct hci_dev *hdev)
> @@ -4488,6 +4562,7 @@ int hci_dev_close_sync(struct hci_dev *hdev)
>  	cancel_delayed_work(&hdev->power_off);
>  	cancel_delayed_work(&hdev->ncmd_timer);
>  	cancel_delayed_work(&hdev->le_scan_disable);
> +	cancel_delayed_work(&hdev->le_scan_restart);
>  
>  	hci_request_cancel_all(hdev);

Re: [PATCH v4 2/4] Bluetooth: Rework le_scan_restart for hci_sync

[Linux regression tracking #adding (Thorsten Leemhuis)]

[TLDR: I'm adding this report to the list of tracked Linux kernel
regressions; the text you find below is based on a few templates
paragraphs you might have encountered already in similar form.
See link in footer if these mails annoy you.]

On 15.06.23 14:06, Stefan Agner wrote:
> Hi Brian, hi all,
> 
> We experienced quite some Bluetooth issues after moving from Linux 5.15
> to 6.1 on Home Assistant OS, especially on Intel NUC type systems (which
> is a popular choice in our community, so it might just be that). When
> continuously scanning/listening for BLE packets, the packet flow
> suddenly ends. Depending on which and how many devices (possibly also
> other factors) within minutes or hours.
> 
> Jan (in cc) was able to bisect the issue, and was able to pinpoint the
> problem to this change.
> 
> Meanwhile I was able to confirm, that reverting this single commit on
> the latest 6.1.34 seems to resolve the issue.

FWIW & BTW: might be wise to also check if latest mainline still shows
the problem, as explained on this page:

https://linux-regtracking.leemhuis.info/post/frequent-reasons-why-linux-kernel-bug-reports-are-ignored/

> I've reviewed the change and surrounding code, and one thing I've
> noticed is that the if statement to set cp.filter_dup in
> hci_le_set_ext_scan_enable_sync and hci_le_set_scan_enable_sync are
> different. Not sure if that needs to be the way it is, but my outside
> gut feeling says hci_le_set_ext_scan_enable_sync should use "if (val &&
> hci_dev_test_flag(hdev, HCI_MESH))" as well. 
> 
> However, that did not fix the problem (but maybe it is wrong
> nonetheless?).
> 
> Anyone has an idea what could be the problem here?

Thanks for the report. To be sure the issue doesn't fall through the
cracks unnoticed, I'm adding it to regzbot, the Linux kernel regression
tracking bot:

#regzbot ^introduced 27d54b778ad
#regzbot title net/bluetooth: packet flow suddenly ends when
continuously scanning/listening for BLE packets
#regzbot ignore-activity

This isn't a regression? This issue or a fix for it are already
discussed somewhere else? It was fixed already? You want to clarify when
the regression started to happen? Or point out I got the title or
something else totally wrong? Then just reply and tell me -- ideally
while also telling regzbot about it, as explained by the page listed in
the footer of this mail.

Developers: When fixing the issue, remember to add 'Link:' tags pointing
to the report (the parent of this mail). See page linked in footer for
details.

Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
--
Everything you wanna know about Linux kernel regression tracking:
https://linux-regtracking.leemhuis.info/about/#tldr
That page also explains what to do if mails like this annoy you.

> On 2022-07-27 15:58, Brian Gix wrote:
>> le_scan_restart delayed work queue was running as a deprecated
>> hci_request instead of on the newer thread-safe hci_sync mechanism.
>>
>> Signed-off-by: Brian Gix <brian.gix@intel.com>
>> ---
>>  net/bluetooth/hci_request.c | 89 -------------------------------------
>>  net/bluetooth/hci_sync.c    | 75 +++++++++++++++++++++++++++++++
>>  2 files changed, 75 insertions(+), 89 deletions(-)
>>
>> diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c
>> index 32fefaa0d3ca..114af7350363 100644
>> --- a/net/bluetooth/hci_request.c
>> +++ b/net/bluetooth/hci_request.c
>> @@ -1975,92 +1975,6 @@ int hci_abort_conn(struct hci_conn *conn, u8 reason)
>>  	return 0;
>>  }
>>  
>> -static int le_scan_restart(struct hci_request *req, unsigned long opt)
>> -{
>> -	struct hci_dev *hdev = req->hdev;
>> -
>> -	/* If controller is not scanning we are done. */
>> -	if (!hci_dev_test_flag(hdev, HCI_LE_SCAN))
>> -		return 0;
>> -
>> -	if (hdev->scanning_paused) {
>> -		bt_dev_dbg(hdev, "Scanning is paused for suspend");
>> -		return 0;
>> -	}
>> -
>> -	hci_req_add_le_scan_disable(req, false);
>> -
>> -	if (use_ext_scan(hdev)) {
>> -		struct hci_cp_le_set_ext_scan_enable ext_enable_cp;
>> -
>> -		memset(&ext_enable_cp, 0, sizeof(ext_enable_cp));
>> -		ext_enable_cp.enable = LE_SCAN_ENABLE;
>> -		ext_enable_cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE;
>> -
>> -		hci_req_add(req, HCI_OP_LE_SET_EXT_SCAN_ENABLE,
>> -			    sizeof(ext_enable_cp), &ext_enable_cp);
>> -	} else {
>> -		struct hci_cp_le_set_scan_enable cp;
>> -
>> -		memset(&cp, 0, sizeof(cp));
>> -		cp.enable = LE_SCAN_ENABLE;
>> -		cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE;
>> -		hci_req_add(req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(cp), &cp);
>> -	}
>> -
>> -	return 0;
>> -}
>> -
>> -static void le_scan_restart_work(struct work_struct *work)
>> -{
>> -	struct hci_dev *hdev = container_of(work, struct hci_dev,
>> -					    le_scan_restart.work);
>> -	unsigned long timeout, duration, scan_start, now;
>> -	u8 status;
>> -
>> -	bt_dev_dbg(hdev, "");
>> -
>> -	hci_req_sync(hdev, le_scan_restart, 0, HCI_CMD_TIMEOUT, &status);
>> -	if (status) {
>> -		bt_dev_err(hdev, "failed to restart LE scan: status %d",
>> -			   status);
>> -		return;
>> -	}
>> -
>> -	hci_dev_lock(hdev);
>> -
>> -	if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) ||
>> -	    !hdev->discovery.scan_start)
>> -		goto unlock;
>> -
>> -	/* When the scan was started, hdev->le_scan_disable has been queued
>> -	 * after duration from scan_start. During scan restart this job
>> -	 * has been canceled, and we need to queue it again after proper
>> -	 * timeout, to make sure that scan does not run indefinitely.
>> -	 */
>> -	duration = hdev->discovery.scan_duration;
>> -	scan_start = hdev->discovery.scan_start;
>> -	now = jiffies;
>> -	if (now - scan_start <= duration) {
>> -		int elapsed;
>> -
>> -		if (now >= scan_start)
>> -			elapsed = now - scan_start;
>> -		else
>> -			elapsed = ULONG_MAX - scan_start + now;
>> -
>> -		timeout = duration - elapsed;
>> -	} else {
>> -		timeout = 0;
>> -	}
>> -
>> -	queue_delayed_work(hdev->req_workqueue,
>> -			   &hdev->le_scan_disable, timeout);
>> -
>> -unlock:
>> -	hci_dev_unlock(hdev);
>> -}
>> -
>>  bool hci_req_stop_discovery(struct hci_request *req)
>>  {
>>  	struct hci_dev *hdev = req->hdev;
>> @@ -2158,7 +2072,6 @@ int hci_req_configure_datapath(struct hci_dev
>> *hdev, struct bt_codec *codec)
>>  
>>  void hci_request_setup(struct hci_dev *hdev)
>>  {
>> -	INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work);
>>  	INIT_DELAYED_WORK(&hdev->adv_instance_expire, adv_timeout_expire);
>>  	INIT_DELAYED_WORK(&hdev->interleave_scan, interleave_scan_work);
>>  }
>> @@ -2167,8 +2080,6 @@ void hci_request_cancel_all(struct hci_dev *hdev)
>>  {
>>  	__hci_cmd_sync_cancel(hdev, ENODEV);
>>  
>> -	cancel_delayed_work_sync(&hdev->le_scan_restart);
>> -
>>  	if (hdev->adv_instance_timeout) {
>>  		cancel_delayed_work_sync(&hdev->adv_instance_expire);
>>  		hdev->adv_instance_timeout = 0;
>> diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
>> index 7dae2ee1bb82..19d57ec0feb8 100644
>> --- a/net/bluetooth/hci_sync.c
>> +++ b/net/bluetooth/hci_sync.c
>> @@ -392,6 +392,79 @@ static void le_scan_disable(struct work_struct *work)
>>  	hci_dev_unlock(hdev);
>>  }
>>  
>> +static int hci_le_set_scan_enable_sync(struct hci_dev *hdev, u8 val,
>> +				       u8 filter_dup);
>> +static int hci_le_scan_restart_sync(struct hci_dev *hdev)
>> +{
>> +	/* If controller is not scanning we are done. */
>> +	if (!hci_dev_test_flag(hdev, HCI_LE_SCAN))
>> +		return 0;
>> +
>> +	if (hdev->scanning_paused) {
>> +		bt_dev_dbg(hdev, "Scanning is paused for suspend");
>> +		return 0;
>> +	}
>> +
>> +	hci_le_set_scan_enable_sync(hdev, LE_SCAN_DISABLE, 0x00);
>> +	return hci_le_set_scan_enable_sync(hdev, LE_SCAN_ENABLE,
>> +					   LE_SCAN_FILTER_DUP_ENABLE);
>> +}
>> +
>> +static int le_scan_restart_sync(struct hci_dev *hdev, void *data)
>> +{
>> +	return hci_le_scan_restart_sync(hdev);
>> +}
>> +
>> +static void le_scan_restart(struct work_struct *work)
>> +{
>> +	struct hci_dev *hdev = container_of(work, struct hci_dev,
>> +					    le_scan_restart.work);
>> +	unsigned long timeout, duration, scan_start, now;
>> +	int status;
>> +
>> +	bt_dev_dbg(hdev, "");
>> +
>> +	hci_dev_lock(hdev);
>> +
>> +	status = hci_cmd_sync_queue(hdev, le_scan_restart_sync, NULL, NULL);
>> +	if (status) {
>> +		bt_dev_err(hdev, "failed to restart LE scan: status %d",
>> +			   status);
>> +		goto unlock;
>> +	}
>> +
>> +	if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) ||
>> +	    !hdev->discovery.scan_start)
>> +		goto unlock;
>> +
>> +	/* When the scan was started, hdev->le_scan_disable has been queued
>> +	 * after duration from scan_start. During scan restart this job
>> +	 * has been canceled, and we need to queue it again after proper
>> +	 * timeout, to make sure that scan does not run indefinitely.
>> +	 */
>> +	duration = hdev->discovery.scan_duration;
>> +	scan_start = hdev->discovery.scan_start;
>> +	now = jiffies;
>> +	if (now - scan_start <= duration) {
>> +		int elapsed;
>> +
>> +		if (now >= scan_start)
>> +			elapsed = now - scan_start;
>> +		else
>> +			elapsed = ULONG_MAX - scan_start + now;
>> +
>> +		timeout = duration - elapsed;
>> +	} else {
>> +		timeout = 0;
>> +	}
>> +
>> +	queue_delayed_work(hdev->req_workqueue,
>> +			   &hdev->le_scan_disable, timeout);
>> +
>> +unlock:
>> +	hci_dev_unlock(hdev);
>> +}
>> +
>>  void hci_cmd_sync_init(struct hci_dev *hdev)
>>  {
>>  	INIT_WORK(&hdev->cmd_sync_work, hci_cmd_sync_work);
>> @@ -400,6 +473,7 @@ void hci_cmd_sync_init(struct hci_dev *hdev)
>>  
>>  	INIT_WORK(&hdev->cmd_sync_cancel_work, hci_cmd_sync_cancel_work);
>>  	INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable);
>> +	INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart);
>>  }
>>  
>>  void hci_cmd_sync_clear(struct hci_dev *hdev)
>> @@ -4488,6 +4562,7 @@ int hci_dev_close_sync(struct hci_dev *hdev)
>>  	cancel_delayed_work(&hdev->power_off);
>>  	cancel_delayed_work(&hdev->ncmd_timer);
>>  	cancel_delayed_work(&hdev->le_scan_disable);
>> +	cancel_delayed_work(&hdev->le_scan_restart);
>>  
>>  	hci_request_cancel_all(hdev);
> 
> 

Re: [PATCH v4 2/4] Bluetooth: Rework le_scan_restart for hci_sync

[Jan Čermák]

Hi Thorsten, hi everyone,

I confirm the regression is reproducible in my environment also on
latest tagged mainline (6.4.0-rc6).

Cheers,
Jan

On Thu, 15 Jun 2023 at 14:47, Linux regression tracking #adding
(Thorsten Leemhuis) <regressions@leemhuis.info> wrote:
>
> [TLDR: I'm adding this report to the list of tracked Linux kernel
> regressions; the text you find below is based on a few templates
> paragraphs you might have encountered already in similar form.
> See link in footer if these mails annoy you.]
>
> On 15.06.23 14:06, Stefan Agner wrote:
> > Hi Brian, hi all,
> >
> > We experienced quite some Bluetooth issues after moving from Linux 5.15
> > to 6.1 on Home Assistant OS, especially on Intel NUC type systems (which
> > is a popular choice in our community, so it might just be that). When
> > continuously scanning/listening for BLE packets, the packet flow
> > suddenly ends. Depending on which and how many devices (possibly also
> > other factors) within minutes or hours.
> >
> > Jan (in cc) was able to bisect the issue, and was able to pinpoint the
> > problem to this change.
> >
> > Meanwhile I was able to confirm, that reverting this single commit on
> > the latest 6.1.34 seems to resolve the issue.
>
> FWIW & BTW: might be wise to also check if latest mainline still shows
> the problem, as explained on this page:
>
> https://linux-regtracking.leemhuis.info/post/frequent-reasons-why-linux-kernel-bug-reports-are-ignored/
>
> > I've reviewed the change and surrounding code, and one thing I've
> > noticed is that the if statement to set cp.filter_dup in
> > hci_le_set_ext_scan_enable_sync and hci_le_set_scan_enable_sync are
> > different. Not sure if that needs to be the way it is, but my outside
> > gut feeling says hci_le_set_ext_scan_enable_sync should use "if (val &&
> > hci_dev_test_flag(hdev, HCI_MESH))" as well.
> >
> > However, that did not fix the problem (but maybe it is wrong
> > nonetheless?).
> >
> > Anyone has an idea what could be the problem here?
>
> Thanks for the report. To be sure the issue doesn't fall through the
> cracks unnoticed, I'm adding it to regzbot, the Linux kernel regression
> tracking bot:
>
> #regzbot ^introduced 27d54b778ad
> #regzbot title net/bluetooth: packet flow suddenly ends when
> continuously scanning/listening for BLE packets
> #regzbot ignore-activity
>
> This isn't a regression? This issue or a fix for it are already
> discussed somewhere else? It was fixed already? You want to clarify when
> the regression started to happen? Or point out I got the title or
> something else totally wrong? Then just reply and tell me -- ideally
> while also telling regzbot about it, as explained by the page listed in
> the footer of this mail.
>
> Developers: When fixing the issue, remember to add 'Link:' tags pointing
> to the report (the parent of this mail). See page linked in footer for
> details.
>
> Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
> --
> Everything you wanna know about Linux kernel regression tracking:
> https://linux-regtracking.leemhuis.info/about/#tldr
> That page also explains what to do if mails like this annoy you.
>
> > On 2022-07-27 15:58, Brian Gix wrote:
> >> le_scan_restart delayed work queue was running as a deprecated
> >> hci_request instead of on the newer thread-safe hci_sync mechanism.
> >>
> >> Signed-off-by: Brian Gix <brian.gix@intel.com>
> >> ---
> >>  net/bluetooth/hci_request.c | 89 -------------------------------------
> >>  net/bluetooth/hci_sync.c    | 75 +++++++++++++++++++++++++++++++
> >>  2 files changed, 75 insertions(+), 89 deletions(-)
> >>
> >> diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c
> >> index 32fefaa0d3ca..114af7350363 100644
> >> --- a/net/bluetooth/hci_request.c
> >> +++ b/net/bluetooth/hci_request.c
> >> @@ -1975,92 +1975,6 @@ int hci_abort_conn(struct hci_conn *conn, u8 reason)
> >>      return 0;
> >>  }
> >>
> >> -static int le_scan_restart(struct hci_request *req, unsigned long opt)
> >> -{
> >> -    struct hci_dev *hdev = req->hdev;
> >> -
> >> -    /* If controller is not scanning we are done. */
> >> -    if (!hci_dev_test_flag(hdev, HCI_LE_SCAN))
> >> -            return 0;
> >> -
> >> -    if (hdev->scanning_paused) {
> >> -            bt_dev_dbg(hdev, "Scanning is paused for suspend");
> >> -            return 0;
> >> -    }
> >> -
> >> -    hci_req_add_le_scan_disable(req, false);
> >> -
> >> -    if (use_ext_scan(hdev)) {
> >> -            struct hci_cp_le_set_ext_scan_enable ext_enable_cp;
> >> -
> >> -            memset(&ext_enable_cp, 0, sizeof(ext_enable_cp));
> >> -            ext_enable_cp.enable = LE_SCAN_ENABLE;
> >> -            ext_enable_cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE;
> >> -
> >> -            hci_req_add(req, HCI_OP_LE_SET_EXT_SCAN_ENABLE,
> >> -                        sizeof(ext_enable_cp), &ext_enable_cp);
> >> -    } else {
> >> -            struct hci_cp_le_set_scan_enable cp;
> >> -
> >> -            memset(&cp, 0, sizeof(cp));
> >> -            cp.enable = LE_SCAN_ENABLE;
> >> -            cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE;
> >> -            hci_req_add(req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(cp), &cp);
> >> -    }
> >> -
> >> -    return 0;
> >> -}
> >> -
> >> -static void le_scan_restart_work(struct work_struct *work)
> >> -{
> >> -    struct hci_dev *hdev = container_of(work, struct hci_dev,
> >> -                                        le_scan_restart.work);
> >> -    unsigned long timeout, duration, scan_start, now;
> >> -    u8 status;
> >> -
> >> -    bt_dev_dbg(hdev, "");
> >> -
> >> -    hci_req_sync(hdev, le_scan_restart, 0, HCI_CMD_TIMEOUT, &status);
> >> -    if (status) {
> >> -            bt_dev_err(hdev, "failed to restart LE scan: status %d",
> >> -                       status);
> >> -            return;
> >> -    }
> >> -
> >> -    hci_dev_lock(hdev);
> >> -
> >> -    if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) ||
> >> -        !hdev->discovery.scan_start)
> >> -            goto unlock;
> >> -
> >> -    /* When the scan was started, hdev->le_scan_disable has been queued
> >> -     * after duration from scan_start. During scan restart this job
> >> -     * has been canceled, and we need to queue it again after proper
> >> -     * timeout, to make sure that scan does not run indefinitely.
> >> -     */
> >> -    duration = hdev->discovery.scan_duration;
> >> -    scan_start = hdev->discovery.scan_start;
> >> -    now = jiffies;
> >> -    if (now - scan_start <= duration) {
> >> -            int elapsed;
> >> -
> >> -            if (now >= scan_start)
> >> -                    elapsed = now - scan_start;
> >> -            else
> >> -                    elapsed = ULONG_MAX - scan_start + now;
> >> -
> >> -            timeout = duration - elapsed;
> >> -    } else {
> >> -            timeout = 0;
> >> -    }
> >> -
> >> -    queue_delayed_work(hdev->req_workqueue,
> >> -                       &hdev->le_scan_disable, timeout);
> >> -
> >> -unlock:
> >> -    hci_dev_unlock(hdev);
> >> -}
> >> -
> >>  bool hci_req_stop_discovery(struct hci_request *req)
> >>  {
> >>      struct hci_dev *hdev = req->hdev;
> >> @@ -2158,7 +2072,6 @@ int hci_req_configure_datapath(struct hci_dev
> >> *hdev, struct bt_codec *codec)
> >>
> >>  void hci_request_setup(struct hci_dev *hdev)
> >>  {
> >> -    INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work);
> >>      INIT_DELAYED_WORK(&hdev->adv_instance_expire, adv_timeout_expire);
> >>      INIT_DELAYED_WORK(&hdev->interleave_scan, interleave_scan_work);
> >>  }
> >> @@ -2167,8 +2080,6 @@ void hci_request_cancel_all(struct hci_dev *hdev)
> >>  {
> >>      __hci_cmd_sync_cancel(hdev, ENODEV);
> >>
> >> -    cancel_delayed_work_sync(&hdev->le_scan_restart);
> >> -
> >>      if (hdev->adv_instance_timeout) {
> >>              cancel_delayed_work_sync(&hdev->adv_instance_expire);
> >>              hdev->adv_instance_timeout = 0;
> >> diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
> >> index 7dae2ee1bb82..19d57ec0feb8 100644
> >> --- a/net/bluetooth/hci_sync.c
> >> +++ b/net/bluetooth/hci_sync.c
> >> @@ -392,6 +392,79 @@ static void le_scan_disable(struct work_struct *work)
> >>      hci_dev_unlock(hdev);
> >>  }
> >>
> >> +static int hci_le_set_scan_enable_sync(struct hci_dev *hdev, u8 val,
> >> +                                   u8 filter_dup);
> >> +static int hci_le_scan_restart_sync(struct hci_dev *hdev)
> >> +{
> >> +    /* If controller is not scanning we are done. */
> >> +    if (!hci_dev_test_flag(hdev, HCI_LE_SCAN))
> >> +            return 0;
> >> +
> >> +    if (hdev->scanning_paused) {
> >> +            bt_dev_dbg(hdev, "Scanning is paused for suspend");
> >> +            return 0;
> >> +    }
> >> +
> >> +    hci_le_set_scan_enable_sync(hdev, LE_SCAN_DISABLE, 0x00);
> >> +    return hci_le_set_scan_enable_sync(hdev, LE_SCAN_ENABLE,
> >> +                                       LE_SCAN_FILTER_DUP_ENABLE);
> >> +}
> >> +
> >> +static int le_scan_restart_sync(struct hci_dev *hdev, void *data)
> >> +{
> >> +    return hci_le_scan_restart_sync(hdev);
> >> +}
> >> +
> >> +static void le_scan_restart(struct work_struct *work)
> >> +{
> >> +    struct hci_dev *hdev = container_of(work, struct hci_dev,
> >> +                                        le_scan_restart.work);
> >> +    unsigned long timeout, duration, scan_start, now;
> >> +    int status;
> >> +
> >> +    bt_dev_dbg(hdev, "");
> >> +
> >> +    hci_dev_lock(hdev);
> >> +
> >> +    status = hci_cmd_sync_queue(hdev, le_scan_restart_sync, NULL, NULL);
> >> +    if (status) {
> >> +            bt_dev_err(hdev, "failed to restart LE scan: status %d",
> >> +                       status);
> >> +            goto unlock;
> >> +    }
> >> +
> >> +    if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) ||
> >> +        !hdev->discovery.scan_start)
> >> +            goto unlock;
> >> +
> >> +    /* When the scan was started, hdev->le_scan_disable has been queued
> >> +     * after duration from scan_start. During scan restart this job
> >> +     * has been canceled, and we need to queue it again after proper
> >> +     * timeout, to make sure that scan does not run indefinitely.
> >> +     */
> >> +    duration = hdev->discovery.scan_duration;
> >> +    scan_start = hdev->discovery.scan_start;
> >> +    now = jiffies;
> >> +    if (now - scan_start <= duration) {
> >> +            int elapsed;
> >> +
> >> +            if (now >= scan_start)
> >> +                    elapsed = now - scan_start;
> >> +            else
> >> +                    elapsed = ULONG_MAX - scan_start + now;
> >> +
> >> +            timeout = duration - elapsed;
> >> +    } else {
> >> +            timeout = 0;
> >> +    }
> >> +
> >> +    queue_delayed_work(hdev->req_workqueue,
> >> +                       &hdev->le_scan_disable, timeout);
> >> +
> >> +unlock:
> >> +    hci_dev_unlock(hdev);
> >> +}
> >> +
> >>  void hci_cmd_sync_init(struct hci_dev *hdev)
> >>  {
> >>      INIT_WORK(&hdev->cmd_sync_work, hci_cmd_sync_work);
> >> @@ -400,6 +473,7 @@ void hci_cmd_sync_init(struct hci_dev *hdev)
> >>
> >>      INIT_WORK(&hdev->cmd_sync_cancel_work, hci_cmd_sync_cancel_work);
> >>      INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable);
> >> +    INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart);
> >>  }
> >>
> >>  void hci_cmd_sync_clear(struct hci_dev *hdev)
> >> @@ -4488,6 +4562,7 @@ int hci_dev_close_sync(struct hci_dev *hdev)
> >>      cancel_delayed_work(&hdev->power_off);
> >>      cancel_delayed_work(&hdev->ncmd_timer);
> >>      cancel_delayed_work(&hdev->le_scan_disable);
> >> +    cancel_delayed_work(&hdev->le_scan_restart);
> >>
> >>      hci_request_cancel_all(hdev);
> >
> >

Re: [PATCH v4 2/4] Bluetooth: Rework le_scan_restart for hci_sync

[Luiz Augusto von Dentz]

Hi Stefan,

On Thu, Jun 15, 2023 at 5:06 AM Stefan Agner <stefan@agner.ch> wrote:
>
> Hi Brian, hi all,
>
> We experienced quite some Bluetooth issues after moving from Linux 5.15
> to 6.1 on Home Assistant OS, especially on Intel NUC type systems (which
> is a popular choice in our community, so it might just be that). When
> continuously scanning/listening for BLE packets, the packet flow
> suddenly ends. Depending on which and how many devices (possibly also
> other factors) within minutes or hours.
>
> Jan (in cc) was able to bisect the issue, and was able to pinpoint the
> problem to this change.
>
> Meanwhile I was able to confirm, that reverting this single commit on
> the latest 6.1.34 seems to resolve the issue.
>
> I've reviewed the change and surrounding code, and one thing I've
> noticed is that the if statement to set cp.filter_dup in
> hci_le_set_ext_scan_enable_sync and hci_le_set_scan_enable_sync are
> different. Not sure if that needs to be the way it is, but my outside
> gut feeling says hci_le_set_ext_scan_enable_sync should use "if (val &&
> hci_dev_test_flag(hdev, HCI_MESH))" as well.
>
> However, that did not fix the problem (but maybe it is wrong
> nonetheless?).
>
> Anyone has an idea what could be the problem here?

Are there any logs of the problem? Does any HCI command fails or
anything so that we can track down what could be wrong?

> --
> Stefan
>
> On 2022-07-27 15:58, Brian Gix wrote:
> > le_scan_restart delayed work queue was running as a deprecated
> > hci_request instead of on the newer thread-safe hci_sync mechanism.
> >
> > Signed-off-by: Brian Gix <brian.gix@intel.com>
> > ---
> >  net/bluetooth/hci_request.c | 89 -------------------------------------
> >  net/bluetooth/hci_sync.c    | 75 +++++++++++++++++++++++++++++++
> >  2 files changed, 75 insertions(+), 89 deletions(-)
> >
> > diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c
> > index 32fefaa0d3ca..114af7350363 100644
> > --- a/net/bluetooth/hci_request.c
> > +++ b/net/bluetooth/hci_request.c
> > @@ -1975,92 +1975,6 @@ int hci_abort_conn(struct hci_conn *conn, u8 reason)
> >       return 0;
> >  }
> >
> > -static int le_scan_restart(struct hci_request *req, unsigned long opt)
> > -{
> > -     struct hci_dev *hdev = req->hdev;
> > -
> > -     /* If controller is not scanning we are done. */
> > -     if (!hci_dev_test_flag(hdev, HCI_LE_SCAN))
> > -             return 0;
> > -
> > -     if (hdev->scanning_paused) {
> > -             bt_dev_dbg(hdev, "Scanning is paused for suspend");
> > -             return 0;
> > -     }
> > -
> > -     hci_req_add_le_scan_disable(req, false);
> > -
> > -     if (use_ext_scan(hdev)) {
> > -             struct hci_cp_le_set_ext_scan_enable ext_enable_cp;
> > -
> > -             memset(&ext_enable_cp, 0, sizeof(ext_enable_cp));
> > -             ext_enable_cp.enable = LE_SCAN_ENABLE;
> > -             ext_enable_cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE;
> > -
> > -             hci_req_add(req, HCI_OP_LE_SET_EXT_SCAN_ENABLE,
> > -                         sizeof(ext_enable_cp), &ext_enable_cp);
> > -     } else {
> > -             struct hci_cp_le_set_scan_enable cp;
> > -
> > -             memset(&cp, 0, sizeof(cp));
> > -             cp.enable = LE_SCAN_ENABLE;
> > -             cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE;
> > -             hci_req_add(req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(cp), &cp);
> > -     }
> > -
> > -     return 0;
> > -}
> > -
> > -static void le_scan_restart_work(struct work_struct *work)
> > -{
> > -     struct hci_dev *hdev = container_of(work, struct hci_dev,
> > -                                         le_scan_restart.work);
> > -     unsigned long timeout, duration, scan_start, now;
> > -     u8 status;
> > -
> > -     bt_dev_dbg(hdev, "");
> > -
> > -     hci_req_sync(hdev, le_scan_restart, 0, HCI_CMD_TIMEOUT, &status);
> > -     if (status) {
> > -             bt_dev_err(hdev, "failed to restart LE scan: status %d",
> > -                        status);
> > -             return;
> > -     }
> > -
> > -     hci_dev_lock(hdev);
> > -
> > -     if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) ||
> > -         !hdev->discovery.scan_start)
> > -             goto unlock;
> > -
> > -     /* When the scan was started, hdev->le_scan_disable has been queued
> > -      * after duration from scan_start. During scan restart this job
> > -      * has been canceled, and we need to queue it again after proper
> > -      * timeout, to make sure that scan does not run indefinitely.
> > -      */
> > -     duration = hdev->discovery.scan_duration;
> > -     scan_start = hdev->discovery.scan_start;
> > -     now = jiffies;
> > -     if (now - scan_start <= duration) {
> > -             int elapsed;
> > -
> > -             if (now >= scan_start)
> > -                     elapsed = now - scan_start;
> > -             else
> > -                     elapsed = ULONG_MAX - scan_start + now;
> > -
> > -             timeout = duration - elapsed;
> > -     } else {
> > -             timeout = 0;
> > -     }
> > -
> > -     queue_delayed_work(hdev->req_workqueue,
> > -                        &hdev->le_scan_disable, timeout);
> > -
> > -unlock:
> > -     hci_dev_unlock(hdev);
> > -}
> > -
> >  bool hci_req_stop_discovery(struct hci_request *req)
> >  {
> >       struct hci_dev *hdev = req->hdev;
> > @@ -2158,7 +2072,6 @@ int hci_req_configure_datapath(struct hci_dev
> > *hdev, struct bt_codec *codec)
> >
> >  void hci_request_setup(struct hci_dev *hdev)
> >  {
> > -     INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work);
> >       INIT_DELAYED_WORK(&hdev->adv_instance_expire, adv_timeout_expire);
> >       INIT_DELAYED_WORK(&hdev->interleave_scan, interleave_scan_work);
> >  }
> > @@ -2167,8 +2080,6 @@ void hci_request_cancel_all(struct hci_dev *hdev)
> >  {
> >       __hci_cmd_sync_cancel(hdev, ENODEV);
> >
> > -     cancel_delayed_work_sync(&hdev->le_scan_restart);
> > -
> >       if (hdev->adv_instance_timeout) {
> >               cancel_delayed_work_sync(&hdev->adv_instance_expire);
> >               hdev->adv_instance_timeout = 0;
> > diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
> > index 7dae2ee1bb82..19d57ec0feb8 100644
> > --- a/net/bluetooth/hci_sync.c
> > +++ b/net/bluetooth/hci_sync.c
> > @@ -392,6 +392,79 @@ static void le_scan_disable(struct work_struct *work)
> >       hci_dev_unlock(hdev);
> >  }
> >
> > +static int hci_le_set_scan_enable_sync(struct hci_dev *hdev, u8 val,
> > +                                    u8 filter_dup);
> > +static int hci_le_scan_restart_sync(struct hci_dev *hdev)
> > +{
> > +     /* If controller is not scanning we are done. */
> > +     if (!hci_dev_test_flag(hdev, HCI_LE_SCAN))
> > +             return 0;
> > +
> > +     if (hdev->scanning_paused) {
> > +             bt_dev_dbg(hdev, "Scanning is paused for suspend");
> > +             return 0;
> > +     }
> > +
> > +     hci_le_set_scan_enable_sync(hdev, LE_SCAN_DISABLE, 0x00);
> > +     return hci_le_set_scan_enable_sync(hdev, LE_SCAN_ENABLE,
> > +                                        LE_SCAN_FILTER_DUP_ENABLE);
> > +}
> > +
> > +static int le_scan_restart_sync(struct hci_dev *hdev, void *data)
> > +{
> > +     return hci_le_scan_restart_sync(hdev);
> > +}
> > +
> > +static void le_scan_restart(struct work_struct *work)
> > +{
> > +     struct hci_dev *hdev = container_of(work, struct hci_dev,
> > +                                         le_scan_restart.work);
> > +     unsigned long timeout, duration, scan_start, now;
> > +     int status;
> > +
> > +     bt_dev_dbg(hdev, "");
> > +
> > +     hci_dev_lock(hdev);
> > +
> > +     status = hci_cmd_sync_queue(hdev, le_scan_restart_sync, NULL, NULL);
> > +     if (status) {
> > +             bt_dev_err(hdev, "failed to restart LE scan: status %d",
> > +                        status);
> > +             goto unlock;
> > +     }
> > +
> > +     if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) ||
> > +         !hdev->discovery.scan_start)
> > +             goto unlock;
> > +
> > +     /* When the scan was started, hdev->le_scan_disable has been queued
> > +      * after duration from scan_start. During scan restart this job
> > +      * has been canceled, and we need to queue it again after proper
> > +      * timeout, to make sure that scan does not run indefinitely.
> > +      */
> > +     duration = hdev->discovery.scan_duration;
> > +     scan_start = hdev->discovery.scan_start;
> > +     now = jiffies;
> > +     if (now - scan_start <= duration) {
> > +             int elapsed;
> > +
> > +             if (now >= scan_start)
> > +                     elapsed = now - scan_start;
> > +             else
> > +                     elapsed = ULONG_MAX - scan_start + now;
> > +
> > +             timeout = duration - elapsed;
> > +     } else {
> > +             timeout = 0;
> > +     }
> > +
> > +     queue_delayed_work(hdev->req_workqueue,
> > +                        &hdev->le_scan_disable, timeout);
> > +
> > +unlock:
> > +     hci_dev_unlock(hdev);
> > +}
> > +
> >  void hci_cmd_sync_init(struct hci_dev *hdev)
> >  {
> >       INIT_WORK(&hdev->cmd_sync_work, hci_cmd_sync_work);
> > @@ -400,6 +473,7 @@ void hci_cmd_sync_init(struct hci_dev *hdev)
> >
> >       INIT_WORK(&hdev->cmd_sync_cancel_work, hci_cmd_sync_cancel_work);
> >       INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable);
> > +     INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart);
> >  }
> >
> >  void hci_cmd_sync_clear(struct hci_dev *hdev)
> > @@ -4488,6 +4562,7 @@ int hci_dev_close_sync(struct hci_dev *hdev)
> >       cancel_delayed_work(&hdev->power_off);
> >       cancel_delayed_work(&hdev->ncmd_timer);
> >       cancel_delayed_work(&hdev->le_scan_disable);
> > +     cancel_delayed_work(&hdev->le_scan_restart);
> >
> >       hci_request_cancel_all(hdev);



-- 
Luiz Augusto von Dentz

Re: [PATCH v4 2/4] Bluetooth: Rework le_scan_restart for hci_sync

[Luiz Augusto von Dentz]

+Brian Gix

On Thu, Jun 15, 2023 at 10:27 AM Luiz Augusto von Dentz
<luiz.dentz@gmail.com> wrote:
>
> Hi Stefan,
>
> On Thu, Jun 15, 2023 at 5:06 AM Stefan Agner <stefan@agner.ch> wrote:
> >
> > Hi Brian, hi all,
> >
> > We experienced quite some Bluetooth issues after moving from Linux 5.15
> > to 6.1 on Home Assistant OS, especially on Intel NUC type systems (which
> > is a popular choice in our community, so it might just be that). When
> > continuously scanning/listening for BLE packets, the packet flow
> > suddenly ends. Depending on which and how many devices (possibly also
> > other factors) within minutes or hours.
> >
> > Jan (in cc) was able to bisect the issue, and was able to pinpoint the
> > problem to this change.
> >
> > Meanwhile I was able to confirm, that reverting this single commit on
> > the latest 6.1.34 seems to resolve the issue.
> >
> > I've reviewed the change and surrounding code, and one thing I've
> > noticed is that the if statement to set cp.filter_dup in
> > hci_le_set_ext_scan_enable_sync and hci_le_set_scan_enable_sync are
> > different. Not sure if that needs to be the way it is, but my outside
> > gut feeling says hci_le_set_ext_scan_enable_sync should use "if (val &&
> > hci_dev_test_flag(hdev, HCI_MESH))" as well.
> >
> > However, that did not fix the problem (but maybe it is wrong
> > nonetheless?).
> >
> > Anyone has an idea what could be the problem here?
>
> Are there any logs of the problem? Does any HCI command fails or
> anything so that we can track down what could be wrong?

@Brian Gix perhaps you have a better idea what is going wrong here?

> > --
> > Stefan
> >
> > On 2022-07-27 15:58, Brian Gix wrote:
> > > le_scan_restart delayed work queue was running as a deprecated
> > > hci_request instead of on the newer thread-safe hci_sync mechanism.
> > >
> > > Signed-off-by: Brian Gix <brian.gix@intel.com>
> > > ---
> > >  net/bluetooth/hci_request.c | 89 -------------------------------------
> > >  net/bluetooth/hci_sync.c    | 75 +++++++++++++++++++++++++++++++
> > >  2 files changed, 75 insertions(+), 89 deletions(-)
> > >
> > > diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c
> > > index 32fefaa0d3ca..114af7350363 100644
> > > --- a/net/bluetooth/hci_request.c
> > > +++ b/net/bluetooth/hci_request.c
> > > @@ -1975,92 +1975,6 @@ int hci_abort_conn(struct hci_conn *conn, u8 reason)
> > >       return 0;
> > >  }
> > >
> > > -static int le_scan_restart(struct hci_request *req, unsigned long opt)
> > > -{
> > > -     struct hci_dev *hdev = req->hdev;
> > > -
> > > -     /* If controller is not scanning we are done. */
> > > -     if (!hci_dev_test_flag(hdev, HCI_LE_SCAN))
> > > -             return 0;
> > > -
> > > -     if (hdev->scanning_paused) {
> > > -             bt_dev_dbg(hdev, "Scanning is paused for suspend");
> > > -             return 0;
> > > -     }
> > > -
> > > -     hci_req_add_le_scan_disable(req, false);
> > > -
> > > -     if (use_ext_scan(hdev)) {
> > > -             struct hci_cp_le_set_ext_scan_enable ext_enable_cp;
> > > -
> > > -             memset(&ext_enable_cp, 0, sizeof(ext_enable_cp));
> > > -             ext_enable_cp.enable = LE_SCAN_ENABLE;
> > > -             ext_enable_cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE;
> > > -
> > > -             hci_req_add(req, HCI_OP_LE_SET_EXT_SCAN_ENABLE,
> > > -                         sizeof(ext_enable_cp), &ext_enable_cp);
> > > -     } else {
> > > -             struct hci_cp_le_set_scan_enable cp;
> > > -
> > > -             memset(&cp, 0, sizeof(cp));
> > > -             cp.enable = LE_SCAN_ENABLE;
> > > -             cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE;
> > > -             hci_req_add(req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(cp), &cp);
> > > -     }
> > > -
> > > -     return 0;
> > > -}
> > > -
> > > -static void le_scan_restart_work(struct work_struct *work)
> > > -{
> > > -     struct hci_dev *hdev = container_of(work, struct hci_dev,
> > > -                                         le_scan_restart.work);
> > > -     unsigned long timeout, duration, scan_start, now;
> > > -     u8 status;
> > > -
> > > -     bt_dev_dbg(hdev, "");
> > > -
> > > -     hci_req_sync(hdev, le_scan_restart, 0, HCI_CMD_TIMEOUT, &status);
> > > -     if (status) {
> > > -             bt_dev_err(hdev, "failed to restart LE scan: status %d",
> > > -                        status);
> > > -             return;
> > > -     }
> > > -
> > > -     hci_dev_lock(hdev);
> > > -
> > > -     if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) ||
> > > -         !hdev->discovery.scan_start)
> > > -             goto unlock;
> > > -
> > > -     /* When the scan was started, hdev->le_scan_disable has been queued
> > > -      * after duration from scan_start. During scan restart this job
> > > -      * has been canceled, and we need to queue it again after proper
> > > -      * timeout, to make sure that scan does not run indefinitely.
> > > -      */
> > > -     duration = hdev->discovery.scan_duration;
> > > -     scan_start = hdev->discovery.scan_start;
> > > -     now = jiffies;
> > > -     if (now - scan_start <= duration) {
> > > -             int elapsed;
> > > -
> > > -             if (now >= scan_start)
> > > -                     elapsed = now - scan_start;
> > > -             else
> > > -                     elapsed = ULONG_MAX - scan_start + now;
> > > -
> > > -             timeout = duration - elapsed;
> > > -     } else {
> > > -             timeout = 0;
> > > -     }
> > > -
> > > -     queue_delayed_work(hdev->req_workqueue,
> > > -                        &hdev->le_scan_disable, timeout);
> > > -
> > > -unlock:
> > > -     hci_dev_unlock(hdev);
> > > -}
> > > -
> > >  bool hci_req_stop_discovery(struct hci_request *req)
> > >  {
> > >       struct hci_dev *hdev = req->hdev;
> > > @@ -2158,7 +2072,6 @@ int hci_req_configure_datapath(struct hci_dev
> > > *hdev, struct bt_codec *codec)
> > >
> > >  void hci_request_setup(struct hci_dev *hdev)
> > >  {
> > > -     INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work);
> > >       INIT_DELAYED_WORK(&hdev->adv_instance_expire, adv_timeout_expire);
> > >       INIT_DELAYED_WORK(&hdev->interleave_scan, interleave_scan_work);
> > >  }
> > > @@ -2167,8 +2080,6 @@ void hci_request_cancel_all(struct hci_dev *hdev)
> > >  {
> > >       __hci_cmd_sync_cancel(hdev, ENODEV);
> > >
> > > -     cancel_delayed_work_sync(&hdev->le_scan_restart);
> > > -
> > >       if (hdev->adv_instance_timeout) {
> > >               cancel_delayed_work_sync(&hdev->adv_instance_expire);
> > >               hdev->adv_instance_timeout = 0;
> > > diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
> > > index 7dae2ee1bb82..19d57ec0feb8 100644
> > > --- a/net/bluetooth/hci_sync.c
> > > +++ b/net/bluetooth/hci_sync.c
> > > @@ -392,6 +392,79 @@ static void le_scan_disable(struct work_struct *work)
> > >       hci_dev_unlock(hdev);
> > >  }
> > >
> > > +static int hci_le_set_scan_enable_sync(struct hci_dev *hdev, u8 val,
> > > +                                    u8 filter_dup);
> > > +static int hci_le_scan_restart_sync(struct hci_dev *hdev)
> > > +{
> > > +     /* If controller is not scanning we are done. */
> > > +     if (!hci_dev_test_flag(hdev, HCI_LE_SCAN))
> > > +             return 0;
> > > +
> > > +     if (hdev->scanning_paused) {
> > > +             bt_dev_dbg(hdev, "Scanning is paused for suspend");
> > > +             return 0;
> > > +     }
> > > +
> > > +     hci_le_set_scan_enable_sync(hdev, LE_SCAN_DISABLE, 0x00);
> > > +     return hci_le_set_scan_enable_sync(hdev, LE_SCAN_ENABLE,
> > > +                                        LE_SCAN_FILTER_DUP_ENABLE);
> > > +}
> > > +
> > > +static int le_scan_restart_sync(struct hci_dev *hdev, void *data)
> > > +{
> > > +     return hci_le_scan_restart_sync(hdev);
> > > +}
> > > +
> > > +static void le_scan_restart(struct work_struct *work)
> > > +{
> > > +     struct hci_dev *hdev = container_of(work, struct hci_dev,
> > > +                                         le_scan_restart.work);
> > > +     unsigned long timeout, duration, scan_start, now;
> > > +     int status;
> > > +
> > > +     bt_dev_dbg(hdev, "");
> > > +
> > > +     hci_dev_lock(hdev);
> > > +
> > > +     status = hci_cmd_sync_queue(hdev, le_scan_restart_sync, NULL, NULL);
> > > +     if (status) {
> > > +             bt_dev_err(hdev, "failed to restart LE scan: status %d",
> > > +                        status);
> > > +             goto unlock;
> > > +     }
> > > +
> > > +     if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) ||
> > > +         !hdev->discovery.scan_start)
> > > +             goto unlock;
> > > +
> > > +     /* When the scan was started, hdev->le_scan_disable has been queued
> > > +      * after duration from scan_start. During scan restart this job
> > > +      * has been canceled, and we need to queue it again after proper
> > > +      * timeout, to make sure that scan does not run indefinitely.
> > > +      */
> > > +     duration = hdev->discovery.scan_duration;
> > > +     scan_start = hdev->discovery.scan_start;
> > > +     now = jiffies;
> > > +     if (now - scan_start <= duration) {
> > > +             int elapsed;
> > > +
> > > +             if (now >= scan_start)
> > > +                     elapsed = now - scan_start;
> > > +             else
> > > +                     elapsed = ULONG_MAX - scan_start + now;
> > > +
> > > +             timeout = duration - elapsed;
> > > +     } else {
> > > +             timeout = 0;
> > > +     }
> > > +
> > > +     queue_delayed_work(hdev->req_workqueue,
> > > +                        &hdev->le_scan_disable, timeout);
> > > +
> > > +unlock:
> > > +     hci_dev_unlock(hdev);
> > > +}
> > > +
> > >  void hci_cmd_sync_init(struct hci_dev *hdev)
> > >  {
> > >       INIT_WORK(&hdev->cmd_sync_work, hci_cmd_sync_work);
> > > @@ -400,6 +473,7 @@ void hci_cmd_sync_init(struct hci_dev *hdev)
> > >
> > >       INIT_WORK(&hdev->cmd_sync_cancel_work, hci_cmd_sync_cancel_work);
> > >       INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable);
> > > +     INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart);
> > >  }
> > >
> > >  void hci_cmd_sync_clear(struct hci_dev *hdev)
> > > @@ -4488,6 +4562,7 @@ int hci_dev_close_sync(struct hci_dev *hdev)
> > >       cancel_delayed_work(&hdev->power_off);
> > >       cancel_delayed_work(&hdev->ncmd_timer);
> > >       cancel_delayed_work(&hdev->le_scan_disable);
> > > +     cancel_delayed_work(&hdev->le_scan_restart);
> > >
> > >       hci_request_cancel_all(hdev);
>
>
>
> --
> Luiz Augusto von Dentz



-- 
Luiz Augusto von Dentz

Re: [PATCH v4 2/4] Bluetooth: Rework le_scan_restart for hci_sync

[Stefan Agner]

On 2023-06-16 03:22, Brian Gix wrote:

> On Thu, Jun 15, 2023 at 11:28 AM Luiz Augusto von Dentz <luiz.dentz@gmail.com> wrote: 
> 
>> +Brian Gix
>> 
>> On Thu, Jun 15, 2023 at 10:27 AM Luiz Augusto von Dentz
>> <luiz.dentz@gmail.com> wrote:
>>> 
>>> Hi Stefan,
>>> 
>>> On Thu, Jun 15, 2023 at 5:06 AM Stefan Agner <stefan@agner.ch> wrote:
>>>> 
>>>> Hi Brian, hi all,
>>>> 
>>>> We experienced quite some Bluetooth issues after moving from Linux 5.15
>>>> to 6.1 on Home Assistant OS, especially on Intel NUC type systems (which
>>>> is a popular choice in our community, so it might just be that). When
>>>> continuously scanning/listening for BLE packets, the packet flow
>>>> suddenly ends. Depending on which and how many devices (possibly also
>>>> other factors) within minutes or hours.
>>>> 
>>>> Jan (in cc) was able to bisect the issue, and was able to pinpoint the
>>>> problem to this change.
>>>> 
>>>> Meanwhile I was able to confirm, that reverting this single commit on
>>>> the latest 6.1.34 seems to resolve the issue.
>>>> 
>>>> I've reviewed the change and surrounding code, and one thing I've
>>>> noticed is that the if statement to set cp.filter_dup in
>>>> hci_le_set_ext_scan_enable_sync and hci_le_set_scan_enable_sync are
>>>> different. Not sure if that needs to be the way it is, but my outside
>>>> gut feeling says hci_le_set_ext_scan_enable_sync should use "if (val &&
>>>> hci_dev_test_flag(hdev, HCI_MESH))" as well.
>>>> 
>>>> However, that did not fix the problem (but maybe it is wrong
>>>> nonetheless?).
>>>> 
>>>> Anyone has an idea what could be the problem here?
>>> 
>>> Are there any logs of the problem? Does any HCI command fails or
>>> anything so that we can track down what could be wrong?

No HCI command fails, there is also no issue reported in the kernel log.
BlueZ just stops receiving BLE packets, at least from certain devices.

>> 
>> @Brian Gix perhaps you have a better idea what is going wrong here?
> 
> It seems unlikely that this is Mesh related. Mesh does need for filtering to
> be FALSE, and Mesh does not use extended scanning in any case. 
> 
> But this was part of the final rewrite to retire the hci_req mechanism in
> favor of the hci_sync mechanism. So my best guess off the top of my head is
> that there was an unintended race condition that worked better than the
> synchronous single-threading mechanism?  Filtering (or not) should not

After review the code I concluded the same. What is a bit surprising to
me is that it is so well reproducible. I guess it is nicer to have a
reproducible one than a hard to reproduce one :)

> prevent advertising packets from permanently wedging.  Does anyone have an
> HCI flow log with and without the offending patch?  Ideally they should be
> identical...  If they are not then I obviously did something wrong. As this
> was not specifically Mesh related, I may have missed some non-mesh corner
> cases.


I've taken two btmon captures, I created them using:
btmon -i hci0 -w /config/hcidump-hci-req-working.log

You can find them at:
https://os-builds.home-assistant.io/hcidump-hci-req-working.log
https://os-builds.home-assistant.io/hcidump-hci-sync-non-working.log

This is while running our user space software (Home Assistant with
Bluetooth integration). Besides some BLE devices (e.g. Xioami Mi
Temperature & Humidity sensor) I have a ESP32 running which sends SPAM
advertisements every 100ms (this accelerates the issue). In the
non-working case you'll see that the system doesn't receive any SPAM
advertisements after around 27 seconds. The working log shows that it
continuously receives the same packets (capture 120s).

Hope this helps.

--
Stefan



> 
>>>> --
>>>> Stefan
>>>> 
>>>> On 2022-07-27 15:58, Brian Gix wrote:
>>>>> le_scan_restart delayed work queue was running as a deprecated
>>>>> hci_request instead of on the newer thread-safe hci_sync mechanism.
>>>>>
>>>>> Signed-off-by: Brian Gix <brian.gix@intel.com>
>>>>> ---
>>>>>  net/bluetooth/hci_request.c | 89 -------------------------------------
>>>>>  net/bluetooth/hci_sync.c    | 75 +++++++++++++++++++++++++++++++
>>>>>  2 files changed, 75 insertions(+), 89 deletions(-)
>>>>>
>>>>> diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c
>>>>> index 32fefaa0d3ca..114af7350363 100644
>>>>> --- a/net/bluetooth/hci_request.c
>>>>> +++ b/net/bluetooth/hci_request.c
>>>>> @@ -1975,92 +1975,6 @@ int hci_abort_conn(struct hci_conn *conn, u8 reason)
>>>>>       return 0;
>>>>>  }
>>>>>
>>>>> -static int le_scan_restart(struct hci_request *req, unsigned long opt)
>>>>> -{
>>>>> -     struct hci_dev *hdev = req->hdev;
>>>>> -
>>>>> -     /* If controller is not scanning we are done. */
>>>>> -     if (!hci_dev_test_flag(hdev, HCI_LE_SCAN))
>>>>> -             return 0;
>>>>> -
>>>>> -     if (hdev->scanning_paused) {
>>>>> -             bt_dev_dbg(hdev, "Scanning is paused for suspend");
>>>>> -             return 0;
>>>>> -     }
>>>>> -
>>>>> -     hci_req_add_le_scan_disable(req, false);
>>>>> -
>>>>> -     if (use_ext_scan(hdev)) {
>>>>> -             struct hci_cp_le_set_ext_scan_enable ext_enable_cp;
>>>>> -
>>>>> -             memset(&ext_enable_cp, 0, sizeof(ext_enable_cp));
>>>>> -             ext_enable_cp.enable = LE_SCAN_ENABLE;
>>>>> -             ext_enable_cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE;
>>>>> -
>>>>> -             hci_req_add(req, HCI_OP_LE_SET_EXT_SCAN_ENABLE,
>>>>> -                         sizeof(ext_enable_cp), &ext_enable_cp);
>>>>> -     } else {
>>>>> -             struct hci_cp_le_set_scan_enable cp;
>>>>> -
>>>>> -             memset(&cp, 0, sizeof(cp));
>>>>> -             cp.enable = LE_SCAN_ENABLE;
>>>>> -             cp.filter_dup = LE_SCAN_FILTER_DUP_ENABLE;
>>>>> -             hci_req_add(req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(cp), &cp);
>>>>> -     }
>>>>> -
>>>>> -     return 0;
>>>>> -}
>>>>> -
>>>>> -static void le_scan_restart_work(struct work_struct *work)
>>>>> -{
>>>>> -     struct hci_dev *hdev = container_of(work, struct hci_dev,
>>>>> -                                         le_scan_restart.work);
>>>>> -     unsigned long timeout, duration, scan_start, now;
>>>>> -     u8 status;
>>>>> -
>>>>> -     bt_dev_dbg(hdev, "");
>>>>> -
>>>>> -     hci_req_sync(hdev, le_scan_restart, 0, HCI_CMD_TIMEOUT, &status);
>>>>> -     if (status) {
>>>>> -             bt_dev_err(hdev, "failed to restart LE scan: status %d",
>>>>> -                        status);
>>>>> -             return;
>>>>> -     }
>>>>> -
>>>>> -     hci_dev_lock(hdev);
>>>>> -
>>>>> -     if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) ||
>>>>> -         !hdev->discovery.scan_start)
>>>>> -             goto unlock;
>>>>> -
>>>>> -     /* When the scan was started, hdev->le_scan_disable has been queued
>>>>> -      * after duration from scan_start. During scan restart this job
>>>>> -      * has been canceled, and we need to queue it again after proper
>>>>> -      * timeout, to make sure that scan does not run indefinitely.
>>>>> -      */
>>>>> -     duration = hdev->discovery.scan_duration;
>>>>> -     scan_start = hdev->discovery.scan_start;
>>>>> -     now = jiffies;
>>>>> -     if (now - scan_start <= duration) {
>>>>> -             int elapsed;
>>>>> -
>>>>> -             if (now >= scan_start)
>>>>> -                     elapsed = now - scan_start;
>>>>> -             else
>>>>> -                     elapsed = ULONG_MAX - scan_start + now;
>>>>> -
>>>>> -             timeout = duration - elapsed;
>>>>> -     } else {
>>>>> -             timeout = 0;
>>>>> -     }
>>>>> -
>>>>> -     queue_delayed_work(hdev->req_workqueue,
>>>>> -                        &hdev->le_scan_disable, timeout);
>>>>> -
>>>>> -unlock:
>>>>> -     hci_dev_unlock(hdev);
>>>>> -}
>>>>> -
>>>>>  bool hci_req_stop_discovery(struct hci_request *req)
>>>>>  {
>>>>>       struct hci_dev *hdev = req->hdev;
>>>>> @@ -2158,7 +2072,6 @@ int hci_req_configure_datapath(struct hci_dev
>>>>> *hdev, struct bt_codec *codec)
>>>>>
>>>>>  void hci_request_setup(struct hci_dev *hdev)
>>>>>  {
>>>>> -     INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart_work);
>>>>>       INIT_DELAYED_WORK(&hdev->adv_instance_expire, adv_timeout_expire);
>>>>>       INIT_DELAYED_WORK(&hdev->interleave_scan, interleave_scan_work);
>>>>>  }
>>>>> @@ -2167,8 +2080,6 @@ void hci_request_cancel_all(struct hci_dev *hdev)
>>>>>  {
>>>>>       __hci_cmd_sync_cancel(hdev, ENODEV);
>>>>>
>>>>> -     cancel_delayed_work_sync(&hdev->le_scan_restart);
>>>>> -
>>>>>       if (hdev->adv_instance_timeout) {
>>>>>               cancel_delayed_work_sync(&hdev->adv_instance_expire);
>>>>>               hdev->adv_instance_timeout = 0;
>>>>> diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
>>>>> index 7dae2ee1bb82..19d57ec0feb8 100644
>>>>> --- a/net/bluetooth/hci_sync.c
>>>>> +++ b/net/bluetooth/hci_sync.c
>>>>> @@ -392,6 +392,79 @@ static void le_scan_disable(struct work_struct *work)
>>>>>       hci_dev_unlock(hdev);
>>>>>  }
>>>>>
>>>>> +static int hci_le_set_scan_enable_sync(struct hci_dev *hdev, u8 val,
>>>>> +                                    u8 filter_dup);
>>>>> +static int hci_le_scan_restart_sync(struct hci_dev *hdev)
>>>>> +{
>>>>> +     /* If controller is not scanning we are done. */
>>>>> +     if (!hci_dev_test_flag(hdev, HCI_LE_SCAN))
>>>>> +             return 0;
>>>>> +
>>>>> +     if (hdev->scanning_paused) {
>>>>> +             bt_dev_dbg(hdev, "Scanning is paused for suspend");
>>>>> +             return 0;
>>>>> +     }
>>>>> +
>>>>> +     hci_le_set_scan_enable_sync(hdev, LE_SCAN_DISABLE, 0x00);
>>>>> +     return hci_le_set_scan_enable_sync(hdev, LE_SCAN_ENABLE,
>>>>> +                                        LE_SCAN_FILTER_DUP_ENABLE);
>>>>> +}
>>>>> +
>>>>> +static int le_scan_restart_sync(struct hci_dev *hdev, void *data)
>>>>> +{
>>>>> +     return hci_le_scan_restart_sync(hdev);
>>>>> +}
>>>>> +
>>>>> +static void le_scan_restart(struct work_struct *work)
>>>>> +{
>>>>> +     struct hci_dev *hdev = container_of(work, struct hci_dev,
>>>>> +                                         le_scan_restart.work);
>>>>> +     unsigned long timeout, duration, scan_start, now;
>>>>> +     int status;
>>>>> +
>>>>> +     bt_dev_dbg(hdev, "");
>>>>> +
>>>>> +     hci_dev_lock(hdev);
>>>>> +
>>>>> +     status = hci_cmd_sync_queue(hdev, le_scan_restart_sync, NULL, NULL);
>>>>> +     if (status) {
>>>>> +             bt_dev_err(hdev, "failed to restart LE scan: status %d",
>>>>> +                        status);
>>>>> +             goto unlock;
>>>>> +     }
>>>>> +
>>>>> +     if (!test_bit(HCI_QUIRK_STRICT_DUPLICATE_FILTER, &hdev->quirks) ||
>>>>> +         !hdev->discovery.scan_start)
>>>>> +             goto unlock;
>>>>> +
>>>>> +     /* When the scan was started, hdev->le_scan_disable has been queued
>>>>> +      * after duration from scan_start. During scan restart this job
>>>>> +      * has been canceled, and we need to queue it again after proper
>>>>> +      * timeout, to make sure that scan does not run indefinitely.
>>>>> +      */
>>>>> +     duration = hdev->discovery.scan_duration;
>>>>> +     scan_start = hdev->discovery.scan_start;
>>>>> +     now = jiffies;
>>>>> +     if (now - scan_start <= duration) {
>>>>> +             int elapsed;
>>>>> +
>>>>> +             if (now >= scan_start)
>>>>> +                     elapsed = now - scan_start;
>>>>> +             else
>>>>> +                     elapsed = ULONG_MAX - scan_start + now;
>>>>> +
>>>>> +             timeout = duration - elapsed;
>>>>> +     } else {
>>>>> +             timeout = 0;
>>>>> +     }
>>>>> +
>>>>> +     queue_delayed_work(hdev->req_workqueue,
>>>>> +                        &hdev->le_scan_disable, timeout);
>>>>> +
>>>>> +unlock:
>>>>> +     hci_dev_unlock(hdev);
>>>>> +}
>>>>> +
>>>>>  void hci_cmd_sync_init(struct hci_dev *hdev)
>>>>>  {
>>>>>       INIT_WORK(&hdev->cmd_sync_work, hci_cmd_sync_work);
>>>>> @@ -400,6 +473,7 @@ void hci_cmd_sync_init(struct hci_dev *hdev)
>>>>>
>>>>>       INIT_WORK(&hdev->cmd_sync_cancel_work, hci_cmd_sync_cancel_work);
>>>>>       INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable);
>>>>> +     INIT_DELAYED_WORK(&hdev->le_scan_restart, le_scan_restart);
>>>>>  }
>>>>>
>>>>>  void hci_cmd_sync_clear(struct hci_dev *hdev)
>>>>> @@ -4488,6 +4562,7 @@ int hci_dev_close_sync(struct hci_dev *hdev)
>>>>>       cancel_delayed_work(&hdev->power_off);
>>>>>       cancel_delayed_work(&hdev->ncmd_timer);
>>>>>       cancel_delayed_work(&hdev->le_scan_disable);
>>>>> +     cancel_delayed_work(&hdev->le_scan_restart);
>>>>>
>>>>>       hci_request_cancel_all(hdev);
>>> 
>>> 
>>> 
>>> --
>>> Luiz Augusto von Dentz
>> 
>> -- 
>> Luiz Augusto von Dentz

Re: [PATCH v4 2/4] Bluetooth: Rework le_scan_restart for hci_sync

[Stefan Agner]

Hi Brian,

Gentle ping on the issue below.

On 2023-06-20 16:41, Stefan Agner wrote:
> On 2023-06-16 03:22, Brian Gix wrote:
> 
>> On Thu, Jun 15, 2023 at 11:28 AM Luiz Augusto von Dentz <luiz.dentz@gmail.com> wrote:
>>
>>> +Brian Gix
>>>
>>> On Thu, Jun 15, 2023 at 10:27 AM Luiz Augusto von Dentz
>>> <luiz.dentz@gmail.com> wrote:
>>>>
>>>> Hi Stefan,
>>>>
>>>> On Thu, Jun 15, 2023 at 5:06 AM Stefan Agner <stefan@agner.ch> wrote:
>>>>>
>>>>> Hi Brian, hi all,
>>>>>
>>>>> We experienced quite some Bluetooth issues after moving from Linux 5.15
>>>>> to 6.1 on Home Assistant OS, especially on Intel NUC type systems (which
>>>>> is a popular choice in our community, so it might just be that). When
>>>>> continuously scanning/listening for BLE packets, the packet flow
>>>>> suddenly ends. Depending on which and how many devices (possibly also
>>>>> other factors) within minutes or hours.
>>>>>
>>>>> Jan (in cc) was able to bisect the issue, and was able to pinpoint the
>>>>> problem to this change.
>>>>>
>>>>> Meanwhile I was able to confirm, that reverting this single commit on
>>>>> the latest 6.1.34 seems to resolve the issue.
>>>>>
>>>>> I've reviewed the change and surrounding code, and one thing I've
>>>>> noticed is that the if statement to set cp.filter_dup in
>>>>> hci_le_set_ext_scan_enable_sync and hci_le_set_scan_enable_sync are
>>>>> different. Not sure if that needs to be the way it is, but my outside
>>>>> gut feeling says hci_le_set_ext_scan_enable_sync should use "if (val &&
>>>>> hci_dev_test_flag(hdev, HCI_MESH))" as well.
>>>>>
>>>>> However, that did not fix the problem (but maybe it is wrong
>>>>> nonetheless?).
>>>>>
>>>>> Anyone has an idea what could be the problem here?
>>>>
>>>> Are there any logs of the problem? Does any HCI command fails or
>>>> anything so that we can track down what could be wrong?
> 
> No HCI command fails, there is also no issue reported in the kernel log.
> BlueZ just stops receiving BLE packets, at least from certain devices.
> 
>>>
>>> @Brian Gix perhaps you have a better idea what is going wrong here?
>>
>> It seems unlikely that this is Mesh related. Mesh does need for filtering to
>> be FALSE, and Mesh does not use extended scanning in any case.
>>
>> But this was part of the final rewrite to retire the hci_req mechanism in
>> favor of the hci_sync mechanism. So my best guess off the top of my head is
>> that there was an unintended race condition that worked better than the
>> synchronous single-threading mechanism?  Filtering (or not) should not
> 
> After review the code I concluded the same. What is a bit surprising to
> me is that it is so well reproducible. I guess it is nicer to have a
> reproducible one than a hard to reproduce one :)
> 
>> prevent advertising packets from permanently wedging.  Does anyone have an
>> HCI flow log with and without the offending patch?  Ideally they should be
>> identical...  If they are not then I obviously did something wrong. As this
>> was not specifically Mesh related, I may have missed some non-mesh corner
>> cases.
> 
> 
> I've taken two btmon captures, I created them using:
> btmon -i hci0 -w /config/hcidump-hci-req-working.log
> 
> You can find them at:
> https://os-builds.home-assistant.io/hcidump-hci-req-working.log
> https://os-builds.home-assistant.io/hcidump-hci-sync-non-working.log

Could you gain any insights from these logs?

--
Stefan


> 
> This is while running our user space software (Home Assistant with
> Bluetooth integration). Besides some BLE devices (e.g. Xioami Mi
> Temperature & Humidity sensor) I have a ESP32 running which sends SPAM
> advertisements every 100ms (this accelerates the issue). In the
> non-working case you'll see that the system doesn't receive any SPAM
> advertisements after around 27 seconds. The working log shows that it
> continuously receives the same packets (capture 120s).
> 
> Hope this helps.
> 
> --
> Stefan
> 
> 

Re: [PATCH v4 2/4] Bluetooth: Rework le_scan_restart for hci_sync

[Linux regression tracking (Thorsten Leemhuis)]

Hi, Thorsten here, the Linux kernel's regression tracker. Top-posting
for once, to make this easily accessible to everyone.

Stefan, was this regression ever addressed? Doesn't look like it from
here, but maybe I'm missing something.

Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
--
Everything you wanna know about Linux kernel regression tracking:
https://linux-regtracking.leemhuis.info/about/#tldr
If I did something stupid, please tell me, as explained on that page.

#regzbot poke

On 30.06.23 12:59, Stefan Agner wrote:
> Hi Brian,
> 
> Gentle ping on the issue below.
> 
> On 2023-06-20 16:41, Stefan Agner wrote:
>> On 2023-06-16 03:22, Brian Gix wrote:
>>
>>> On Thu, Jun 15, 2023 at 11:28 AM Luiz Augusto von Dentz <luiz.dentz@gmail.com> wrote:
>>>
>>>> +Brian Gix
>>>>
>>>> On Thu, Jun 15, 2023 at 10:27 AM Luiz Augusto von Dentz
>>>> <luiz.dentz@gmail.com> wrote:
>>>>>
>>>>> Hi Stefan,
>>>>>
>>>>> On Thu, Jun 15, 2023 at 5:06 AM Stefan Agner <stefan@agner.ch> wrote:
>>>>>>
>>>>>> Hi Brian, hi all,
>>>>>>
>>>>>> We experienced quite some Bluetooth issues after moving from Linux 5.15
>>>>>> to 6.1 on Home Assistant OS, especially on Intel NUC type systems (which
>>>>>> is a popular choice in our community, so it might just be that). When
>>>>>> continuously scanning/listening for BLE packets, the packet flow
>>>>>> suddenly ends. Depending on which and how many devices (possibly also
>>>>>> other factors) within minutes or hours.
>>>>>>
>>>>>> Jan (in cc) was able to bisect the issue, and was able to pinpoint the
>>>>>> problem to this change.
>>>>>>
>>>>>> Meanwhile I was able to confirm, that reverting this single commit on
>>>>>> the latest 6.1.34 seems to resolve the issue.
>>>>>>
>>>>>> I've reviewed the change and surrounding code, and one thing I've
>>>>>> noticed is that the if statement to set cp.filter_dup in
>>>>>> hci_le_set_ext_scan_enable_sync and hci_le_set_scan_enable_sync are
>>>>>> different. Not sure if that needs to be the way it is, but my outside
>>>>>> gut feeling says hci_le_set_ext_scan_enable_sync should use "if (val &&
>>>>>> hci_dev_test_flag(hdev, HCI_MESH))" as well.
>>>>>>
>>>>>> However, that did not fix the problem (but maybe it is wrong
>>>>>> nonetheless?).
>>>>>>
>>>>>> Anyone has an idea what could be the problem here?
>>>>>
>>>>> Are there any logs of the problem? Does any HCI command fails or
>>>>> anything so that we can track down what could be wrong?
>>
>> No HCI command fails, there is also no issue reported in the kernel log.
>> BlueZ just stops receiving BLE packets, at least from certain devices.
>>
>>>>
>>>> @Brian Gix perhaps you have a better idea what is going wrong here?
>>>
>>> It seems unlikely that this is Mesh related. Mesh does need for filtering to
>>> be FALSE, and Mesh does not use extended scanning in any case.
>>>
>>> But this was part of the final rewrite to retire the hci_req mechanism in
>>> favor of the hci_sync mechanism. So my best guess off the top of my head is
>>> that there was an unintended race condition that worked better than the
>>> synchronous single-threading mechanism?  Filtering (or not) should not
>>
>> After review the code I concluded the same. What is a bit surprising to
>> me is that it is so well reproducible. I guess it is nicer to have a
>> reproducible one than a hard to reproduce one :)
>>
>>> prevent advertising packets from permanently wedging.  Does anyone have an
>>> HCI flow log with and without the offending patch?  Ideally they should be
>>> identical...  If they are not then I obviously did something wrong. As this
>>> was not specifically Mesh related, I may have missed some non-mesh corner
>>> cases.
>>
>>
>> I've taken two btmon captures, I created them using:
>> btmon -i hci0 -w /config/hcidump-hci-req-working.log
>>
>> You can find them at:
>> https://os-builds.home-assistant.io/hcidump-hci-req-working.log
>> https://os-builds.home-assistant.io/hcidump-hci-sync-non-working.log
> 
> Could you gain any insights from these logs?
> 
> --
> Stefan
> 
> 
>>
>> This is while running our user space software (Home Assistant with
>> Bluetooth integration). Besides some BLE devices (e.g. Xioami Mi
>> Temperature & Humidity sensor) I have a ESP32 running which sends SPAM
>> advertisements every 100ms (this accelerates the issue). In the
>> non-working case you'll see that the system doesn't receive any SPAM
>> advertisements after around 27 seconds. The working log shows that it
>> continuously receives the same packets (capture 120s).
>>
>> Hope this helps.
>>
>> --
>> Stefan
>>
>>
> 
> 

Re: [PATCH v4 2/4] Bluetooth: Rework le_scan_restart for hci_sync

[Stefan Agner]

Hi Thorsten,

No, this hasn't been addressed so far. I am also not sure how we can
help solving that particular issue.

Besides this, we have other Bluetooth issues which seem to be Kernel
regressions (where downgrading to Linux 5.15 also helps), folks see
"hci0: unexpected event for opcode" on Intel but also other systems. We
haven't bisected that issue yet. But it seems that the Bluetooth stack
is really somewhat unstable in recent releases.

--
Stefan


On 2023-08-29 13:22, Linux regression tracking (Thorsten Leemhuis)
wrote:
> Hi, Thorsten here, the Linux kernel's regression tracker. Top-posting
> for once, to make this easily accessible to everyone.
> 
> Stefan, was this regression ever addressed? Doesn't look like it from
> here, but maybe I'm missing something.
> 
> Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
> --
> Everything you wanna know about Linux kernel regression tracking:
> https://linux-regtracking.leemhuis.info/about/#tldr
> If I did something stupid, please tell me, as explained on that page.
> 
> #regzbot poke
> 
> On 30.06.23 12:59, Stefan Agner wrote:
>> Hi Brian,
>>
>> Gentle ping on the issue below.
>>
>> On 2023-06-20 16:41, Stefan Agner wrote:
>>> On 2023-06-16 03:22, Brian Gix wrote:
>>>
>>>> On Thu, Jun 15, 2023 at 11:28 AM Luiz Augusto von Dentz <luiz.dentz@gmail.com> wrote:
>>>>
>>>>> +Brian Gix
>>>>>
>>>>> On Thu, Jun 15, 2023 at 10:27 AM Luiz Augusto von Dentz
>>>>> <luiz.dentz@gmail.com> wrote:
>>>>>>
>>>>>> Hi Stefan,
>>>>>>
>>>>>> On Thu, Jun 15, 2023 at 5:06 AM Stefan Agner <stefan@agner.ch> wrote:
>>>>>>>
>>>>>>> Hi Brian, hi all,
>>>>>>>
>>>>>>> We experienced quite some Bluetooth issues after moving from Linux 5.15
>>>>>>> to 6.1 on Home Assistant OS, especially on Intel NUC type systems (which
>>>>>>> is a popular choice in our community, so it might just be that). When
>>>>>>> continuously scanning/listening for BLE packets, the packet flow
>>>>>>> suddenly ends. Depending on which and how many devices (possibly also
>>>>>>> other factors) within minutes or hours.
>>>>>>>
>>>>>>> Jan (in cc) was able to bisect the issue, and was able to pinpoint the
>>>>>>> problem to this change.
>>>>>>>
>>>>>>> Meanwhile I was able to confirm, that reverting this single commit on
>>>>>>> the latest 6.1.34 seems to resolve the issue.
>>>>>>>
>>>>>>> I've reviewed the change and surrounding code, and one thing I've
>>>>>>> noticed is that the if statement to set cp.filter_dup in
>>>>>>> hci_le_set_ext_scan_enable_sync and hci_le_set_scan_enable_sync are
>>>>>>> different. Not sure if that needs to be the way it is, but my outside
>>>>>>> gut feeling says hci_le_set_ext_scan_enable_sync should use "if (val &&
>>>>>>> hci_dev_test_flag(hdev, HCI_MESH))" as well.
>>>>>>>
>>>>>>> However, that did not fix the problem (but maybe it is wrong
>>>>>>> nonetheless?).
>>>>>>>
>>>>>>> Anyone has an idea what could be the problem here?
>>>>>>
>>>>>> Are there any logs of the problem? Does any HCI command fails or
>>>>>> anything so that we can track down what could be wrong?
>>>
>>> No HCI command fails, there is also no issue reported in the kernel log.
>>> BlueZ just stops receiving BLE packets, at least from certain devices.
>>>
>>>>>
>>>>> @Brian Gix perhaps you have a better idea what is going wrong here?
>>>>
>>>> It seems unlikely that this is Mesh related. Mesh does need for filtering to
>>>> be FALSE, and Mesh does not use extended scanning in any case.
>>>>
>>>> But this was part of the final rewrite to retire the hci_req mechanism in
>>>> favor of the hci_sync mechanism. So my best guess off the top of my head is
>>>> that there was an unintended race condition that worked better than the
>>>> synchronous single-threading mechanism?  Filtering (or not) should not
>>>
>>> After review the code I concluded the same. What is a bit surprising to
>>> me is that it is so well reproducible. I guess it is nicer to have a
>>> reproducible one than a hard to reproduce one :)
>>>
>>>> prevent advertising packets from permanently wedging.  Does anyone have an
>>>> HCI flow log with and without the offending patch?  Ideally they should be
>>>> identical...  If they are not then I obviously did something wrong. As this
>>>> was not specifically Mesh related, I may have missed some non-mesh corner
>>>> cases.
>>>
>>>
>>> I've taken two btmon captures, I created them using:
>>> btmon -i hci0 -w /config/hcidump-hci-req-working.log
>>>
>>> You can find them at:
>>> https://os-builds.home-assistant.io/hcidump-hci-req-working.log
>>> https://os-builds.home-assistant.io/hcidump-hci-sync-non-working.log
>>
>> Could you gain any insights from these logs?
>>
>> --
>> Stefan
>>
>>
>>>
>>> This is while running our user space software (Home Assistant with
>>> Bluetooth integration). Besides some BLE devices (e.g. Xioami Mi
>>> Temperature & Humidity sensor) I have a ESP32 running which sends SPAM
>>> advertisements every 100ms (this accelerates the issue). In the
>>> non-working case you'll see that the system doesn't receive any SPAM
>>> advertisements after around 27 seconds. The working log shows that it
>>> continuously receives the same packets (capture 120s).
>>>
>>> Hope this helps.
>>>
>>> --
>>> Stefan
>>>
>>>
>>
>>

Re: [PATCH v4 2/4] Bluetooth: Rework le_scan_restart for hci_sync

[Linux regression tracking (Thorsten Leemhuis)]

On 29.08.23 15:27, Stefan Agner wrote:
> 
> No, this hasn't been addressed so far.

Thx and aggh. It's vacation time, so sometimes things take longer, but
that doesn't explain why nothing seems to have happened for 9 weeks now
(at least that how it looks from here, but maybe I missed something).

Luiz, what's up here? What do you need to get down to this?

CCing the other Bluetooth maintainers just to be sure. FWIW, the thread
starts here:
https://lore.kernel.org/linux-bluetooth/578e6d7afd676129decafba846a933f5@agner.ch/#t

Jan saw similar problems:
https://lore.kernel.org/linux-bluetooth/CAPa5EdBSzkuMRoHDJ5w9ESckvNvs68nAfvixyetKcQ5+YD50wA@mail.gmail.com/

> I am also not sure how we can
> help solving that particular issue.

Let's see if this prodding helps to get things rolling. If not, I'll
have to get higher level maintainers involved.

> Besides this, we have other Bluetooth issues which seem to be Kernel
> regressions (where downgrading to Linux 5.15 also helps), folks see
> "hci0: unexpected event for opcode" on Intel but also other systems. We
> haven't bisected that issue yet. But it seems that the Bluetooth stack
> is really somewhat unstable in recent releases.

Might be wise to create a separate thread for those and asking the
bluetooth maintainers if they might have an idea (please CC the
regressions lists as well), maybe we are lucky; if not someone has to
bisect this to get closer to a solution.

Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
--
Everything you wanna know about Linux kernel regression tracking:
https://linux-regtracking.leemhuis.info/about/#tldr
If I did something stupid, please tell me, as explained on that page.

> On 2023-08-29 13:22, Linux regression tracking (Thorsten Leemhuis)
> wrote:
>> Hi, Thorsten here, the Linux kernel's regression tracker. Top-posting
>> for once, to make this easily accessible to everyone.
>>
>> Stefan, was this regression ever addressed? Doesn't look like it from
>> here, but maybe I'm missing something.
>>
>> Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
>> --
>> Everything you wanna know about Linux kernel regression tracking:
>> https://linux-regtracking.leemhuis.info/about/#tldr
>> If I did something stupid, please tell me, as explained on that page.
>>
>> #regzbot poke
>>
>> On 30.06.23 12:59, Stefan Agner wrote:
>>> Hi Brian,
>>>
>>> Gentle ping on the issue below.
>>>
>>> On 2023-06-20 16:41, Stefan Agner wrote:
>>>> On 2023-06-16 03:22, Brian Gix wrote:
>>>>
>>>>> On Thu, Jun 15, 2023 at 11:28 AM Luiz Augusto von Dentz <luiz.dentz@gmail.com> wrote:
>>>>>
>>>>>> +Brian Gix
>>>>>>
>>>>>> On Thu, Jun 15, 2023 at 10:27 AM Luiz Augusto von Dentz
>>>>>> <luiz.dentz@gmail.com> wrote:
>>>>>>>
>>>>>>> Hi Stefan,
>>>>>>>
>>>>>>> On Thu, Jun 15, 2023 at 5:06 AM Stefan Agner <stefan@agner.ch> wrote:
>>>>>>>>
>>>>>>>> Hi Brian, hi all,
>>>>>>>>
>>>>>>>> We experienced quite some Bluetooth issues after moving from Linux 5.15
>>>>>>>> to 6.1 on Home Assistant OS, especially on Intel NUC type systems (which
>>>>>>>> is a popular choice in our community, so it might just be that). When
>>>>>>>> continuously scanning/listening for BLE packets, the packet flow
>>>>>>>> suddenly ends. Depending on which and how many devices (possibly also
>>>>>>>> other factors) within minutes or hours.
>>>>>>>>
>>>>>>>> Jan (in cc) was able to bisect the issue, and was able to pinpoint the
>>>>>>>> problem to this change.
>>>>>>>>
>>>>>>>> Meanwhile I was able to confirm, that reverting this single commit on
>>>>>>>> the latest 6.1.34 seems to resolve the issue.
>>>>>>>>
>>>>>>>> I've reviewed the change and surrounding code, and one thing I've
>>>>>>>> noticed is that the if statement to set cp.filter_dup in
>>>>>>>> hci_le_set_ext_scan_enable_sync and hci_le_set_scan_enable_sync are
>>>>>>>> different. Not sure if that needs to be the way it is, but my outside
>>>>>>>> gut feeling says hci_le_set_ext_scan_enable_sync should use "if (val &&
>>>>>>>> hci_dev_test_flag(hdev, HCI_MESH))" as well.
>>>>>>>>
>>>>>>>> However, that did not fix the problem (but maybe it is wrong
>>>>>>>> nonetheless?).
>>>>>>>>
>>>>>>>> Anyone has an idea what could be the problem here?
>>>>>>>
>>>>>>> Are there any logs of the problem? Does any HCI command fails or
>>>>>>> anything so that we can track down what could be wrong?
>>>>
>>>> No HCI command fails, there is also no issue reported in the kernel log.
>>>> BlueZ just stops receiving BLE packets, at least from certain devices.
>>>>
>>>>>>
>>>>>> @Brian Gix perhaps you have a better idea what is going wrong here?
>>>>>
>>>>> It seems unlikely that this is Mesh related. Mesh does need for filtering to
>>>>> be FALSE, and Mesh does not use extended scanning in any case.
>>>>>
>>>>> But this was part of the final rewrite to retire the hci_req mechanism in
>>>>> favor of the hci_sync mechanism. So my best guess off the top of my head is
>>>>> that there was an unintended race condition that worked better than the
>>>>> synchronous single-threading mechanism?  Filtering (or not) should not
>>>>
>>>> After review the code I concluded the same. What is a bit surprising to
>>>> me is that it is so well reproducible. I guess it is nicer to have a
>>>> reproducible one than a hard to reproduce one :)
>>>>
>>>>> prevent advertising packets from permanently wedging.  Does anyone have an
>>>>> HCI flow log with and without the offending patch?  Ideally they should be
>>>>> identical...  If they are not then I obviously did something wrong. As this
>>>>> was not specifically Mesh related, I may have missed some non-mesh corner
>>>>> cases.
>>>>
>>>>
>>>> I've taken two btmon captures, I created them using:
>>>> btmon -i hci0 -w /config/hcidump-hci-req-working.log
>>>>
>>>> You can find them at:
>>>> https://os-builds.home-assistant.io/hcidump-hci-req-working.log
>>>> https://os-builds.home-assistant.io/hcidump-hci-sync-non-working.log
>>>
>>> Could you gain any insights from these logs?
>>>
>>> --
>>> Stefan
>>>
>>>
>>>>
>>>> This is while running our user space software (Home Assistant with
>>>> Bluetooth integration). Besides some BLE devices (e.g. Xioami Mi
>>>> Temperature & Humidity sensor) I have a ESP32 running which sends SPAM
>>>> advertisements every 100ms (this accelerates the issue). In the
>>>> non-working case you'll see that the system doesn't receive any SPAM
>>>> advertisements after around 27 seconds. The working log shows that it
>>>> continuously receives the same packets (capture 120s).
>>>>
>>>> Hope this helps.
>>>>
>>>> --
>>>> Stefan
>>>>
>>>>
>>>
>>>
> 
> 

Re: [PATCH v4 2/4] Bluetooth: Rework le_scan_restart for hci_sync

[Luiz Augusto von Dentz]

Hi Stefan, Brian,

On Tue, Aug 29, 2023 at 6:27 AM Stefan Agner <stefan@agner.ch> wrote:
>
> Hi Thorsten,
>
> No, this hasn't been addressed so far. I am also not sure how we can
> help solving that particular issue.
>
> Besides this, we have other Bluetooth issues which seem to be Kernel
> regressions (where downgrading to Linux 5.15 also helps), folks see
> "hci0: unexpected event for opcode" on Intel but also other systems. We
> haven't bisected that issue yet. But it seems that the Bluetooth stack
> is really somewhat unstable in recent releases.


I suspect the following change shall make it behave as before, the use
of hci_cmd_sync_queue is not equivalent to hci_req_sync:

https://gist.github.com/Vudentz/b78f34e3775c8cd2db55b868e5c8ef42

That said, I'm considering removing the whole custom handling for
HCI_QUIRK_STRICT_DUPLICATE_FILTER and just disable duplicate filtering
when this flag is set.

> --
> Stefan
>
>
> On 2023-08-29 13:22, Linux regression tracking (Thorsten Leemhuis)
> wrote:
> > Hi, Thorsten here, the Linux kernel's regression tracker. Top-posting
> > for once, to make this easily accessible to everyone.
> >
> > Stefan, was this regression ever addressed? Doesn't look like it from
> > here, but maybe I'm missing something.
> >
> > Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
> > --
> > Everything you wanna know about Linux kernel regression tracking:
> > https://linux-regtracking.leemhuis.info/about/#tldr
> > If I did something stupid, please tell me, as explained on that page.
> >
> > #regzbot poke
> >
> > On 30.06.23 12:59, Stefan Agner wrote:
> >> Hi Brian,
> >>
> >> Gentle ping on the issue below.
> >>
> >> On 2023-06-20 16:41, Stefan Agner wrote:
> >>> On 2023-06-16 03:22, Brian Gix wrote:
> >>>
> >>>> On Thu, Jun 15, 2023 at 11:28 AM Luiz Augusto von Dentz <luiz.dentz@gmail.com> wrote:
> >>>>
> >>>>> +Brian Gix
> >>>>>
> >>>>> On Thu, Jun 15, 2023 at 10:27 AM Luiz Augusto von Dentz
> >>>>> <luiz.dentz@gmail.com> wrote:
> >>>>>>
> >>>>>> Hi Stefan,
> >>>>>>
> >>>>>> On Thu, Jun 15, 2023 at 5:06 AM Stefan Agner <stefan@agner.ch> wrote:
> >>>>>>>
> >>>>>>> Hi Brian, hi all,
> >>>>>>>
> >>>>>>> We experienced quite some Bluetooth issues after moving from Linux 5.15
> >>>>>>> to 6.1 on Home Assistant OS, especially on Intel NUC type systems (which
> >>>>>>> is a popular choice in our community, so it might just be that). When
> >>>>>>> continuously scanning/listening for BLE packets, the packet flow
> >>>>>>> suddenly ends. Depending on which and how many devices (possibly also
> >>>>>>> other factors) within minutes or hours.
> >>>>>>>
> >>>>>>> Jan (in cc) was able to bisect the issue, and was able to pinpoint the
> >>>>>>> problem to this change.
> >>>>>>>
> >>>>>>> Meanwhile I was able to confirm, that reverting this single commit on
> >>>>>>> the latest 6.1.34 seems to resolve the issue.
> >>>>>>>
> >>>>>>> I've reviewed the change and surrounding code, and one thing I've
> >>>>>>> noticed is that the if statement to set cp.filter_dup in
> >>>>>>> hci_le_set_ext_scan_enable_sync and hci_le_set_scan_enable_sync are
> >>>>>>> different. Not sure if that needs to be the way it is, but my outside
> >>>>>>> gut feeling says hci_le_set_ext_scan_enable_sync should use "if (val &&
> >>>>>>> hci_dev_test_flag(hdev, HCI_MESH))" as well.
> >>>>>>>
> >>>>>>> However, that did not fix the problem (but maybe it is wrong
> >>>>>>> nonetheless?).
> >>>>>>>
> >>>>>>> Anyone has an idea what could be the problem here?
> >>>>>>
> >>>>>> Are there any logs of the problem? Does any HCI command fails or
> >>>>>> anything so that we can track down what could be wrong?
> >>>
> >>> No HCI command fails, there is also no issue reported in the kernel log.
> >>> BlueZ just stops receiving BLE packets, at least from certain devices.
> >>>
> >>>>>
> >>>>> @Brian Gix perhaps you have a better idea what is going wrong here?
> >>>>
> >>>> It seems unlikely that this is Mesh related. Mesh does need for filtering to
> >>>> be FALSE, and Mesh does not use extended scanning in any case.
> >>>>
> >>>> But this was part of the final rewrite to retire the hci_req mechanism in
> >>>> favor of the hci_sync mechanism. So my best guess off the top of my head is
> >>>> that there was an unintended race condition that worked better than the
> >>>> synchronous single-threading mechanism?  Filtering (or not) should not
> >>>
> >>> After review the code I concluded the same. What is a bit surprising to
> >>> me is that it is so well reproducible. I guess it is nicer to have a
> >>> reproducible one than a hard to reproduce one :)
> >>>
> >>>> prevent advertising packets from permanently wedging.  Does anyone have an
> >>>> HCI flow log with and without the offending patch?  Ideally they should be
> >>>> identical...  If they are not then I obviously did something wrong. As this
> >>>> was not specifically Mesh related, I may have missed some non-mesh corner
> >>>> cases.
> >>>
> >>>
> >>> I've taken two btmon captures, I created them using:
> >>> btmon -i hci0 -w /config/hcidump-hci-req-working.log
> >>>
> >>> You can find them at:
> >>> https://os-builds.home-assistant.io/hcidump-hci-req-working.log
> >>> https://os-builds.home-assistant.io/hcidump-hci-sync-non-working.log
> >>
> >> Could you gain any insights from these logs?
> >>
> >> --
> >> Stefan
> >>
> >>
> >>>
> >>> This is while running our user space software (Home Assistant with
> >>> Bluetooth integration). Besides some BLE devices (e.g. Xioami Mi
> >>> Temperature & Humidity sensor) I have a ESP32 running which sends SPAM
> >>> advertisements every 100ms (this accelerates the issue). In the
> >>> non-working case you'll see that the system doesn't receive any SPAM
> >>> advertisements after around 27 seconds. The working log shows that it
> >>> continuously receives the same packets (capture 120s).
> >>>
> >>> Hope this helps.
> >>>
> >>> --
> >>> Stefan
> >>>
> >>>
> >>
> >>



-- 
Luiz Augusto von Dentz

Re: [PATCH v4 2/4] Bluetooth: Rework le_scan_restart for hci_sync

[Luiz Augusto von Dentz]

Hi Stefan,

On Tue, Aug 29, 2023 at 1:42 PM Luiz Augusto von Dentz
<luiz.dentz@gmail.com> wrote:
>
> Hi Stefan, Brian,
>
> On Tue, Aug 29, 2023 at 6:27 AM Stefan Agner <stefan@agner.ch> wrote:
> >
> > Hi Thorsten,
> >
> > No, this hasn't been addressed so far. I am also not sure how we can
> > help solving that particular issue.
> >
> > Besides this, we have other Bluetooth issues which seem to be Kernel
> > regressions (where downgrading to Linux 5.15 also helps), folks see
> > "hci0: unexpected event for opcode" on Intel but also other systems. We
> > haven't bisected that issue yet. But it seems that the Bluetooth stack
> > is really somewhat unstable in recent releases.
>
>
> I suspect the following change shall make it behave as before, the use
> of hci_cmd_sync_queue is not equivalent to hci_req_sync:
>
> https://gist.github.com/Vudentz/b78f34e3775c8cd2db55b868e5c8ef42
>
> That said, I'm considering removing the whole custom handling for
> HCI_QUIRK_STRICT_DUPLICATE_FILTER and just disable duplicate filtering
> when this flag is set.

Any chance to tests the following changes:

https://patchwork.kernel.org/project/bluetooth/patch/20230829205936.766544-1-luiz.dentz@gmail.com/

> > --
> > Stefan
> >
> >
> > On 2023-08-29 13:22, Linux regression tracking (Thorsten Leemhuis)
> > wrote:
> > > Hi, Thorsten here, the Linux kernel's regression tracker. Top-posting
> > > for once, to make this easily accessible to everyone.
> > >
> > > Stefan, was this regression ever addressed? Doesn't look like it from
> > > here, but maybe I'm missing something.
> > >
> > > Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
> > > --
> > > Everything you wanna know about Linux kernel regression tracking:
> > > https://linux-regtracking.leemhuis.info/about/#tldr
> > > If I did something stupid, please tell me, as explained on that page.
> > >
> > > #regzbot poke
> > >
> > > On 30.06.23 12:59, Stefan Agner wrote:
> > >> Hi Brian,
> > >>
> > >> Gentle ping on the issue below.
> > >>
> > >> On 2023-06-20 16:41, Stefan Agner wrote:
> > >>> On 2023-06-16 03:22, Brian Gix wrote:
> > >>>
> > >>>> On Thu, Jun 15, 2023 at 11:28 AM Luiz Augusto von Dentz <luiz.dentz@gmail.com> wrote:
> > >>>>
> > >>>>> +Brian Gix
> > >>>>>
> > >>>>> On Thu, Jun 15, 2023 at 10:27 AM Luiz Augusto von Dentz
> > >>>>> <luiz.dentz@gmail.com> wrote:
> > >>>>>>
> > >>>>>> Hi Stefan,
> > >>>>>>
> > >>>>>> On Thu, Jun 15, 2023 at 5:06 AM Stefan Agner <stefan@agner.ch> wrote:
> > >>>>>>>
> > >>>>>>> Hi Brian, hi all,
> > >>>>>>>
> > >>>>>>> We experienced quite some Bluetooth issues after moving from Linux 5.15
> > >>>>>>> to 6.1 on Home Assistant OS, especially on Intel NUC type systems (which
> > >>>>>>> is a popular choice in our community, so it might just be that). When
> > >>>>>>> continuously scanning/listening for BLE packets, the packet flow
> > >>>>>>> suddenly ends. Depending on which and how many devices (possibly also
> > >>>>>>> other factors) within minutes or hours.
> > >>>>>>>
> > >>>>>>> Jan (in cc) was able to bisect the issue, and was able to pinpoint the
> > >>>>>>> problem to this change.
> > >>>>>>>
> > >>>>>>> Meanwhile I was able to confirm, that reverting this single commit on
> > >>>>>>> the latest 6.1.34 seems to resolve the issue.
> > >>>>>>>
> > >>>>>>> I've reviewed the change and surrounding code, and one thing I've
> > >>>>>>> noticed is that the if statement to set cp.filter_dup in
> > >>>>>>> hci_le_set_ext_scan_enable_sync and hci_le_set_scan_enable_sync are
> > >>>>>>> different. Not sure if that needs to be the way it is, but my outside
> > >>>>>>> gut feeling says hci_le_set_ext_scan_enable_sync should use "if (val &&
> > >>>>>>> hci_dev_test_flag(hdev, HCI_MESH))" as well.
> > >>>>>>>
> > >>>>>>> However, that did not fix the problem (but maybe it is wrong
> > >>>>>>> nonetheless?).
> > >>>>>>>
> > >>>>>>> Anyone has an idea what could be the problem here?
> > >>>>>>
> > >>>>>> Are there any logs of the problem? Does any HCI command fails or
> > >>>>>> anything so that we can track down what could be wrong?
> > >>>
> > >>> No HCI command fails, there is also no issue reported in the kernel log.
> > >>> BlueZ just stops receiving BLE packets, at least from certain devices.
> > >>>
> > >>>>>
> > >>>>> @Brian Gix perhaps you have a better idea what is going wrong here?
> > >>>>
> > >>>> It seems unlikely that this is Mesh related. Mesh does need for filtering to
> > >>>> be FALSE, and Mesh does not use extended scanning in any case.
> > >>>>
> > >>>> But this was part of the final rewrite to retire the hci_req mechanism in
> > >>>> favor of the hci_sync mechanism. So my best guess off the top of my head is
> > >>>> that there was an unintended race condition that worked better than the
> > >>>> synchronous single-threading mechanism?  Filtering (or not) should not
> > >>>
> > >>> After review the code I concluded the same. What is a bit surprising to
> > >>> me is that it is so well reproducible. I guess it is nicer to have a
> > >>> reproducible one than a hard to reproduce one :)
> > >>>
> > >>>> prevent advertising packets from permanently wedging.  Does anyone have an
> > >>>> HCI flow log with and without the offending patch?  Ideally they should be
> > >>>> identical...  If they are not then I obviously did something wrong. As this
> > >>>> was not specifically Mesh related, I may have missed some non-mesh corner
> > >>>> cases.
> > >>>
> > >>>
> > >>> I've taken two btmon captures, I created them using:
> > >>> btmon -i hci0 -w /config/hcidump-hci-req-working.log
> > >>>
> > >>> You can find them at:
> > >>> https://os-builds.home-assistant.io/hcidump-hci-req-working.log
> > >>> https://os-builds.home-assistant.io/hcidump-hci-sync-non-working.log
> > >>
> > >> Could you gain any insights from these logs?
> > >>
> > >> --
> > >> Stefan
> > >>
> > >>
> > >>>
> > >>> This is while running our user space software (Home Assistant with
> > >>> Bluetooth integration). Besides some BLE devices (e.g. Xioami Mi
> > >>> Temperature & Humidity sensor) I have a ESP32 running which sends SPAM
> > >>> advertisements every 100ms (this accelerates the issue). In the
> > >>> non-working case you'll see that the system doesn't receive any SPAM
> > >>> advertisements after around 27 seconds. The working log shows that it
> > >>> continuously receives the same packets (capture 120s).
> > >>>
> > >>> Hope this helps.
> > >>>
> > >>> --
> > >>> Stefan
> > >>>
> > >>>
> > >>
> > >>
>
>
>
> --
> Luiz Augusto von Dentz



-- 
Luiz Augusto von Dentz

Re: [PATCH v4 2/4] Bluetooth: Rework le_scan_restart for hci_sync

[Stefan Agner]

Hi Luiz,

On 2023-08-30 19:28, Luiz Augusto von Dentz wrote:
> Hi Stefan,
> 
> On Tue, Aug 29, 2023 at 1:42 PM Luiz Augusto von Dentz
> <luiz.dentz@gmail.com> wrote:
>>
>> Hi Stefan, Brian,
>>
>> On Tue, Aug 29, 2023 at 6:27 AM Stefan Agner <stefan@agner.ch> wrote:
>> >
>> > Hi Thorsten,
>> >
>> > No, this hasn't been addressed so far. I am also not sure how we can
>> > help solving that particular issue.
>> >
>> > Besides this, we have other Bluetooth issues which seem to be Kernel
>> > regressions (where downgrading to Linux 5.15 also helps), folks see
>> > "hci0: unexpected event for opcode" on Intel but also other systems. We
>> > haven't bisected that issue yet. But it seems that the Bluetooth stack
>> > is really somewhat unstable in recent releases.
>>
>>
>> I suspect the following change shall make it behave as before, the use
>> of hci_cmd_sync_queue is not equivalent to hci_req_sync:
>>
>> https://gist.github.com/Vudentz/b78f34e3775c8cd2db55b868e5c8ef42
>>
>> That said, I'm considering removing the whole custom handling for
>> HCI_QUIRK_STRICT_DUPLICATE_FILTER and just disable duplicate filtering
>> when this flag is set.
> 
> Any chance to tests the following changes:
> 
> https://patchwork.kernel.org/project/bluetooth/patch/20230829205936.766544-1-luiz.dentz@gmail.com/

I've tested this with my SPAM test device, and I can confirm that this
indeed fixes the problem we are seeing: The BLE advertisements continue
to come in just fine with the patch applied!

Thanks for the fix!

--
Stefan

> 
>> > --
>> > Stefan
>> >
>> >
>> > On 2023-08-29 13:22, Linux regression tracking (Thorsten Leemhuis)
>> > wrote:
>> > > Hi, Thorsten here, the Linux kernel's regression tracker. Top-posting
>> > > for once, to make this easily accessible to everyone.
>> > >
>> > > Stefan, was this regression ever addressed? Doesn't look like it from
>> > > here, but maybe I'm missing something.
>> > >
>> > > Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
>> > > --
>> > > Everything you wanna know about Linux kernel regression tracking:
>> > > https://linux-regtracking.leemhuis.info/about/#tldr
>> > > If I did something stupid, please tell me, as explained on that page.
>> > >
>> > > #regzbot poke
>> > >
>> > > On 30.06.23 12:59, Stefan Agner wrote:
>> > >> Hi Brian,
>> > >>
>> > >> Gentle ping on the issue below.
>> > >>
>> > >> On 2023-06-20 16:41, Stefan Agner wrote:
>> > >>> On 2023-06-16 03:22, Brian Gix wrote:
>> > >>>
>> > >>>> On Thu, Jun 15, 2023 at 11:28 AM Luiz Augusto von Dentz <luiz.dentz@gmail.com> wrote:
>> > >>>>
>> > >>>>> +Brian Gix
>> > >>>>>
>> > >>>>> On Thu, Jun 15, 2023 at 10:27 AM Luiz Augusto von Dentz
>> > >>>>> <luiz.dentz@gmail.com> wrote:
>> > >>>>>>
>> > >>>>>> Hi Stefan,
>> > >>>>>>
>> > >>>>>> On Thu, Jun 15, 2023 at 5:06 AM Stefan Agner <stefan@agner.ch> wrote:
>> > >>>>>>>
>> > >>>>>>> Hi Brian, hi all,
>> > >>>>>>>
>> > >>>>>>> We experienced quite some Bluetooth issues after moving from Linux 5.15
>> > >>>>>>> to 6.1 on Home Assistant OS, especially on Intel NUC type systems (which
>> > >>>>>>> is a popular choice in our community, so it might just be that). When
>> > >>>>>>> continuously scanning/listening for BLE packets, the packet flow
>> > >>>>>>> suddenly ends. Depending on which and how many devices (possibly also
>> > >>>>>>> other factors) within minutes or hours.
>> > >>>>>>>
>> > >>>>>>> Jan (in cc) was able to bisect the issue, and was able to pinpoint the
>> > >>>>>>> problem to this change.
>> > >>>>>>>
>> > >>>>>>> Meanwhile I was able to confirm, that reverting this single commit on
>> > >>>>>>> the latest 6.1.34 seems to resolve the issue.
>> > >>>>>>>
>> > >>>>>>> I've reviewed the change and surrounding code, and one thing I've
>> > >>>>>>> noticed is that the if statement to set cp.filter_dup in
>> > >>>>>>> hci_le_set_ext_scan_enable_sync and hci_le_set_scan_enable_sync are
>> > >>>>>>> different. Not sure if that needs to be the way it is, but my outside
>> > >>>>>>> gut feeling says hci_le_set_ext_scan_enable_sync should use "if (val &&
>> > >>>>>>> hci_dev_test_flag(hdev, HCI_MESH))" as well.
>> > >>>>>>>
>> > >>>>>>> However, that did not fix the problem (but maybe it is wrong
>> > >>>>>>> nonetheless?).
>> > >>>>>>>
>> > >>>>>>> Anyone has an idea what could be the problem here?
>> > >>>>>>
>> > >>>>>> Are there any logs of the problem? Does any HCI command fails or
>> > >>>>>> anything so that we can track down what could be wrong?
>> > >>>
>> > >>> No HCI command fails, there is also no issue reported in the kernel log.
>> > >>> BlueZ just stops receiving BLE packets, at least from certain devices.
>> > >>>
>> > >>>>>
>> > >>>>> @Brian Gix perhaps you have a better idea what is going wrong here?
>> > >>>>
>> > >>>> It seems unlikely that this is Mesh related. Mesh does need for filtering to
>> > >>>> be FALSE, and Mesh does not use extended scanning in any case.
>> > >>>>
>> > >>>> But this was part of the final rewrite to retire the hci_req mechanism in
>> > >>>> favor of the hci_sync mechanism. So my best guess off the top of my head is
>> > >>>> that there was an unintended race condition that worked better than the
>> > >>>> synchronous single-threading mechanism?  Filtering (or not) should not
>> > >>>
>> > >>> After review the code I concluded the same. What is a bit surprising to
>> > >>> me is that it is so well reproducible. I guess it is nicer to have a
>> > >>> reproducible one than a hard to reproduce one :)
>> > >>>
>> > >>>> prevent advertising packets from permanently wedging.  Does anyone have an
>> > >>>> HCI flow log with and without the offending patch?  Ideally they should be
>> > >>>> identical...  If they are not then I obviously did something wrong. As this
>> > >>>> was not specifically Mesh related, I may have missed some non-mesh corner
>> > >>>> cases.
>> > >>>
>> > >>>
>> > >>> I've taken two btmon captures, I created them using:
>> > >>> btmon -i hci0 -w /config/hcidump-hci-req-working.log
>> > >>>
>> > >>> You can find them at:
>> > >>> https://os-builds.home-assistant.io/hcidump-hci-req-working.log
>> > >>> https://os-builds.home-assistant.io/hcidump-hci-sync-non-working.log
>> > >>
>> > >> Could you gain any insights from these logs?
>> > >>
>> > >> --
>> > >> Stefan
>> > >>
>> > >>
>> > >>>
>> > >>> This is while running our user space software (Home Assistant with
>> > >>> Bluetooth integration). Besides some BLE devices (e.g. Xioami Mi
>> > >>> Temperature & Humidity sensor) I have a ESP32 running which sends SPAM
>> > >>> advertisements every 100ms (this accelerates the issue). In the
>> > >>> non-working case you'll see that the system doesn't receive any SPAM
>> > >>> advertisements after around 27 seconds. The working log shows that it
>> > >>> continuously receives the same packets (capture 120s).
>> > >>>
>> > >>> Hope this helps.
>> > >>>
>> > >>> --
>> > >>> Stefan
>> > >>>
>> > >>>
>> > >>
>> > >>
>>
>>
>>
>> --
>> Luiz Augusto von Dentz