2015-05-01 21:12:44

by Yanbo Li

[permalink] [raw]
Subject: [PATCH] ath10k: Debugfs entry to enable/disable BTC feature

As some radio have no connection with BT modules, enable the BTC feature
will has some side effect if the radio's GPIO connect with any other HW
modules. Add the control switcher "btc_feature" at debugfs and set the
feature as disable by default to avoid such case.

To enable this feature, execute:
echo 1 > /sys/kernel/debug/ieee80211/phyX/ath10k/btc_feature
To disable:
echo 0 > /sys/kernel/debug/ieee80211/phyX/ath10k/btc_feature

Signed-off-by: Yanbo Li <[email protected]>

diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index 8444adf42195..6852e7fc5d5f 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -720,6 +720,8 @@ struct ath10k {
u32 fw_cold_reset_counter;
} stats;

+ bool btc_feature;
+
struct ath10k_thermal thermal;
struct ath10k_wow wow;

diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c
index 8fa606a9c4dd..d686aff2e0cc 100644
--- a/drivers/net/wireless/ath/ath10k/debug.c
+++ b/drivers/net/wireless/ath/ath10k/debug.c
@@ -2092,6 +2092,54 @@ static const struct file_operations fops_quiet_period = {
.open = simple_open
};

+static ssize_t ath10k_write_btc_feature(struct file *file,
+ const char __user *ubuf,
+ size_t count, loff_t *ppos)
+{
+ struct ath10k *ar = file->private_data;
+ char buf[32];
+ size_t buf_size;
+ bool val;
+
+ buf_size = min(count, (sizeof(buf) - 1));
+ if (copy_from_user(buf, ubuf, buf_size))
+ return -EFAULT;
+
+ buf[buf_size] = '\0';
+ if (strtobool(buf, &val) != 0) {
+ ath10k_warn(ar, "Wrong BTC feature setting\n");
+ return -EINVAL;
+ }
+
+ mutex_lock(&ar->conf_mutex);
+ ar->btc_feature = val;
+ queue_work(ar->workqueue, &ar->restart_work);
+ mutex_unlock(&ar->conf_mutex);
+
+ return count;
+}
+
+static ssize_t ath10k_read_btc_feature(struct file *file, char __user *ubuf,
+ size_t count, loff_t *ppos)
+{
+ char buf[32];
+ struct ath10k *ar = file->private_data;
+ int len = 0;
+
+ mutex_lock(&ar->conf_mutex);
+ len = scnprintf(buf, sizeof(buf) - len, "%d\n",
+ ar->btc_feature);
+ mutex_unlock(&ar->conf_mutex);
+
+ return simple_read_from_buffer(ubuf, count, ppos, buf, len);
+}
+
+static const struct file_operations fops_btc_feature = {
+ .read = ath10k_read_btc_feature,
+ .write = ath10k_write_btc_feature,
+ .open = simple_open
+};
+
int ath10k_debug_create(struct ath10k *ar)
{
ar->debug.fw_crash_data = vzalloc(sizeof(*ar->debug.fw_crash_data));
@@ -2195,6 +2243,8 @@ int ath10k_debug_register(struct ath10k *ar)
debugfs_create_file("quiet_period", S_IRUGO | S_IWUSR,
ar->debug.debugfs_phy, ar, &fops_quiet_period);

+ debugfs_create_file("btc_feature", S_IRUGO | S_IWUSR,
+ ar->debug.debugfs_phy, ar, &fops_btc_feature);
return 0;
}

diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index ebaa096cf200..0d46168fbbb3 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -3933,7 +3933,8 @@ static struct sk_buff *ath10k_wmi_10_2_op_gen_init(struct ath10k *ar)
cmd = (struct wmi_init_cmd_10_2 *)buf->data;

features = WMI_10_2_RX_BATCH_MODE;
- if (test_bit(WMI_SERVICE_COEX_GPIO, ar->wmi.svc_map))
+ if (ar->btc_feature &&
+ test_bit(WMI_SERVICE_COEX_GPIO, ar->wmi.svc_map))
features |= WMI_10_2_COEX_GPIO;
cmd->resource_config.feature_mask = __cpu_to_le32(features);

--
1.9.1



2015-05-27 12:57:12

by Kalle Valo

[permalink] [raw]
Subject: Re: [PATCH] ath10k: Debugfs entry to enable/disable BTC feature

Yanbo Li <[email protected]> writes:

> As some radio have no connection with BT modules, enable the BTC feature
> will has some side effect if the radio's GPIO connect with any other HW
> modules. Add the control switcher "btc_feature" at debugfs and set the
> feature as disable by default to avoid such case.
>
> To enable this feature, execute:
> echo 1 > /sys/kernel/debug/ieee80211/phyX/ath10k/btc_feature
> To disable:
> echo 0 > /sys/kernel/debug/ieee80211/phyX/ath10k/btc_feature

I suspect "BTC" does not really tell much for most of the people and
acronyms should be always spelled out at least once.

> Signed-off-by: Yanbo Li <[email protected]>
>
> diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
> index 8444adf42195..6852e7fc5d5f 100644
> --- a/drivers/net/wireless/ath/ath10k/core.h
> +++ b/drivers/net/wireless/ath/ath10k/core.h
> @@ -720,6 +720,8 @@ struct ath10k {
> u32 fw_cold_reset_counter;
> } stats;
>
> + bool btc_feature;

Could we use ar->dev_flags instead?

> +static ssize_t ath10k_write_btc_feature(struct file *file,
> + const char __user *ubuf,
> + size_t count, loff_t *ppos)
> +{
> + struct ath10k *ar = file->private_data;
> + char buf[32];
> + size_t buf_size;
> + bool val;
> +
> + buf_size = min(count, (sizeof(buf) - 1));
> + if (copy_from_user(buf, ubuf, buf_size))
> + return -EFAULT;
> +
> + buf[buf_size] = '\0';
> + if (strtobool(buf, &val) != 0) {
> + ath10k_warn(ar, "Wrong BTC feature setting\n");
> + return -EINVAL;
> + }
> +
> + mutex_lock(&ar->conf_mutex);
> + ar->btc_feature = val;
> + queue_work(ar->workqueue, &ar->restart_work);
> + mutex_unlock(&ar->conf_mutex);

Shouldn't we test ATH10K_STATE_ON first?

> int ath10k_debug_create(struct ath10k *ar)
> {
> ar->debug.fw_crash_data = vzalloc(sizeof(*ar->debug.fw_crash_data));
> @@ -2195,6 +2243,8 @@ int ath10k_debug_register(struct ath10k *ar)
> debugfs_create_file("quiet_period", S_IRUGO | S_IWUSR,
> ar->debug.debugfs_phy, ar, &fops_quiet_period);
>
> + debugfs_create_file("btc_feature", S_IRUGO | S_IWUSR,
> + ar->debug.debugfs_phy, ar,
> &fops_btc_feature);

At least some of the other drivers use term btcoex for this, I think we
should be consistent and use the same.

I can do the changes and send v2 for this.

--
Kalle Valo

2015-05-28 01:07:48

by Yanbo Li

[permalink] [raw]
Subject: RE: [PATCH] ath10k: Debugfs entry to enable/disable BTC feature



> -----Original Message-----
> From: Valo, Kalle
> Sent: Wednesday, May 27, 2015 5:57 AM
> To: Li, Yanbo
> Cc: [email protected]; [email protected];
> [email protected]; [email protected]
> Subject: Re: [PATCH] ath10k: Debugfs entry to enable/disable BTC feature
>
> Yanbo Li <[email protected]> writes:
>
> > As some radio have no connection with BT modules, enable the BTC
> > feature will has some side effect if the radio's GPIO connect with any
> > other HW modules. Add the control switcher "btc_feature" at debugfs
> > and set the feature as disable by default to avoid such case.
> >
> > To enable this feature, execute:
> > echo 1 > /sys/kernel/debug/ieee80211/phyX/ath10k/btc_feature
> > To disable:
> > echo 0 > /sys/kernel/debug/ieee80211/phyX/ath10k/btc_feature
>
> I suspect "BTC" does not really tell much for most of the people and
> acronyms should be always spelled out at least once.
>
> > Signed-off-by: Yanbo Li <[email protected]>
> >
> > diff --git a/drivers/net/wireless/ath/ath10k/core.h
> > b/drivers/net/wireless/ath/ath10k/core.h
> > index 8444adf42195..6852e7fc5d5f 100644
> > --- a/drivers/net/wireless/ath/ath10k/core.h
> > +++ b/drivers/net/wireless/ath/ath10k/core.h
> > @@ -720,6 +720,8 @@ struct ath10k {
> > u32 fw_cold_reset_counter;
> > } stats;
> >
> > + bool btc_feature;
>
> Could we use ar->dev_flags instead?

This dev_flags currently used to mark some status, like the cradh, CAC running,
The BTcoex feature is more likely a HW feature, there are seems different set.

Maybe a separately hw_feature_flag is needed as there are some other HW feature
That we want to enable/disable from user space.
>
> > +static ssize_t ath10k_write_btc_feature(struct file *file,
> > + const char __user *ubuf,
> > + size_t count, loff_t *ppos)
> > +{
> > + struct ath10k *ar = file->private_data;
> > + char buf[32];
> > + size_t buf_size;
> > + bool val;
> > +
> > + buf_size = min(count, (sizeof(buf) - 1));
> > + if (copy_from_user(buf, ubuf, buf_size))
> > + return -EFAULT;
> > +
> > + buf[buf_size] = '\0';
> > + if (strtobool(buf, &val) != 0) {
> > + ath10k_warn(ar, "Wrong BTC feature setting\n");
> > + return -EINVAL;
> > + }
> > +
> > + mutex_lock(&ar->conf_mutex);
> > + ar->btc_feature = val;
> > + queue_work(ar->workqueue, &ar->restart_work);
> > + mutex_unlock(&ar->conf_mutex);
>
> Shouldn't we test ATH10K_STATE_ON first?

The restart process will judge by itself whether the device is on or not, and print below warning in that case:

[797424.496190] ath10k_pci 0000:05:00.0: cannot restart a device that hasn't been started

>
> > int ath10k_debug_create(struct ath10k *ar) {
> > ar->debug.fw_crash_data = vzalloc(sizeof(*ar-
> >debug.fw_crash_data));
> > @@ -2195,6 +2243,8 @@ int ath10k_debug_register(struct ath10k *ar)
> > debugfs_create_file("quiet_period", S_IRUGO | S_IWUSR,
> > ar->debug.debugfs_phy, ar, &fops_quiet_period);
> >
> > + debugfs_create_file("btc_feature", S_IRUGO | S_IWUSR,
> > + ar->debug.debugfs_phy, ar,
> > &fops_btc_feature);
>
> At least some of the other drivers use term btcoex for this, I think we should
> be consistent and use the same.
>
> I can do the changes and send v2 for this.

Thanks and agree, I will send the 2v as there are some other tiny change at my side.

BR /Yanbo

2015-06-08 13:44:15

by Kalle Valo

[permalink] [raw]
Subject: Re: [PATCH] ath10k: Debugfs entry to enable/disable BTC feature

"Li, Yanbo" <[email protected]> writes:

>> -----Original Message-----
>> From: Valo, Kalle
>> Sent: Wednesday, May 27, 2015 5:57 AM
>> To: Li, Yanbo
>> Cc: [email protected]; [email protected];
>> [email protected]; [email protected]
>> Subject: Re: [PATCH] ath10k: Debugfs entry to enable/disable BTC feature
>>
>> Yanbo Li <[email protected]> writes:
>>
>> > As some radio have no connection with BT modules, enable the BTC
>> > feature will has some side effect if the radio's GPIO connect with any
>> > other HW modules. Add the control switcher "btc_feature" at debugfs
>> > and set the feature as disable by default to avoid such case.
>> >
>> > To enable this feature, execute:
>> > echo 1 > /sys/kernel/debug/ieee80211/phyX/ath10k/btc_feature
>> > To disable:
>> > echo 0 > /sys/kernel/debug/ieee80211/phyX/ath10k/btc_feature
>>
>> I suspect "BTC" does not really tell much for most of the people and
>> acronyms should be always spelled out at least once.
>>
>> > Signed-off-by: Yanbo Li <[email protected]>
>> >
>> > diff --git a/drivers/net/wireless/ath/ath10k/core.h
>> > b/drivers/net/wireless/ath/ath10k/core.h
>> > index 8444adf42195..6852e7fc5d5f 100644
>> > --- a/drivers/net/wireless/ath/ath10k/core.h
>> > +++ b/drivers/net/wireless/ath/ath10k/core.h
>> > @@ -720,6 +720,8 @@ struct ath10k {
>> > u32 fw_cold_reset_counter;
>> > } stats;
>> >
>> > + bool btc_feature;
>>
>> Could we use ar->dev_flags instead?
>
> This dev_flags currently used to mark some status, like the cradh, CAC running,
> The BTcoex feature is more likely a HW feature, there are seems different set.
>
> Maybe a separately hw_feature_flag is needed as there are some other HW feature
> That we want to enable/disable from user space.

I think we don't need a separate bitmap, we can use dev_flags just fine
for this.

>> > +static ssize_t ath10k_write_btc_feature(struct file *file,
>> > + const char __user *ubuf,
>> > + size_t count, loff_t *ppos)
>> > +{
>> > + struct ath10k *ar = file->private_data;
>> > + char buf[32];
>> > + size_t buf_size;
>> > + bool val;
>> > +
>> > + buf_size = min(count, (sizeof(buf) - 1));
>> > + if (copy_from_user(buf, ubuf, buf_size))
>> > + return -EFAULT;
>> > +
>> > + buf[buf_size] = '\0';
>> > + if (strtobool(buf, &val) != 0) {
>> > + ath10k_warn(ar, "Wrong BTC feature setting\n");
>> > + return -EINVAL;
>> > + }
>> > +
>> > + mutex_lock(&ar->conf_mutex);
>> > + ar->btc_feature = val;
>> > + queue_work(ar->workqueue, &ar->restart_work);
>> > + mutex_unlock(&ar->conf_mutex);
>>
>> Shouldn't we test ATH10K_STATE_ON first?
>
> The restart process will judge by itself whether the device is on or not, and print below warning in that case:
>
> [797424.496190] ath10k_pci 0000:05:00.0: cannot restart a device that
> hasn't been started

That's just buggy, ath10k should not print anything if a setting is
changed while interface is down. It's much better we have the check for
ATH10K_STATE_ON here.

--
Kalle Valo

2015-06-09 18:58:57

by YanBo

[permalink] [raw]
Subject: Re: [PATCH] ath10k: Debugfs entry to enable/disable BTC feature

On Mon, Jun 8, 2015 at 6:44 AM, Kalle Valo <[email protected]> wrote:
> "Li, Yanbo" <[email protected]> writes:
>
>>> -----Original Message-----
>>> From: Valo, Kalle
>>> Sent: Wednesday, May 27, 2015 5:57 AM
>>> To: Li, Yanbo
>>> Cc: [email protected]; [email protected];
>>> [email protected]; [email protected]
>>> Subject: Re: [PATCH] ath10k: Debugfs entry to enable/disable BTC feature
>>>
>>> Yanbo Li <[email protected]> writes:
>>>
>>> > As some radio have no connection with BT modules, enable the BTC
>>> > feature will has some side effect if the radio's GPIO connect with any
>>> > other HW modules. Add the control switcher "btc_feature" at debugfs
>>> > and set the feature as disable by default to avoid such case.
>>> >
>>> > To enable this feature, execute:
>>> > echo 1 > /sys/kernel/debug/ieee80211/phyX/ath10k/btc_feature
>>> > To disable:
>>> > echo 0 > /sys/kernel/debug/ieee80211/phyX/ath10k/btc_feature
>>>
>>> I suspect "BTC" does not really tell much for most of the people and
>>> acronyms should be always spelled out at least once.
>>>
>>> > Signed-off-by: Yanbo Li <[email protected]>
>>> >
>>> > diff --git a/drivers/net/wireless/ath/ath10k/core.h
>>> > b/drivers/net/wireless/ath/ath10k/core.h
>>> > index 8444adf42195..6852e7fc5d5f 100644
>>> > --- a/drivers/net/wireless/ath/ath10k/core.h
>>> > +++ b/drivers/net/wireless/ath/ath10k/core.h
>>> > @@ -720,6 +720,8 @@ struct ath10k {
>>> > u32 fw_cold_reset_counter;
>>> > } stats;
>>> >
>>> > + bool btc_feature;
>>>
>>> Could we use ar->dev_flags instead?
>>
>> This dev_flags currently used to mark some status, like the cradh, CAC running,
>> The BTcoex feature is more likely a HW feature, there are seems different set.
>>
>> Maybe a separately hw_feature_flag is needed as there are some other HW feature
>> That we want to enable/disable from user space.
>
> I think we don't need a separate bitmap, we can use dev_flags just fine
> for this.
>
>>> > +static ssize_t ath10k_write_btc_feature(struct file *file,
>>> > + const char __user *ubuf,
>>> > + size_t count, loff_t *ppos)
>>> > +{
>>> > + struct ath10k *ar = file->private_data;
>>> > + char buf[32];
>>> > + size_t buf_size;
>>> > + bool val;
>>> > +
>>> > + buf_size = min(count, (sizeof(buf) - 1));
>>> > + if (copy_from_user(buf, ubuf, buf_size))
>>> > + return -EFAULT;
>>> > +
>>> > + buf[buf_size] = '\0';
>>> > + if (strtobool(buf, &val) != 0) {
>>> > + ath10k_warn(ar, "Wrong BTC feature setting\n");
>>> > + return -EINVAL;
>>> > + }
>>> > +
>>> > + mutex_lock(&ar->conf_mutex);
>>> > + ar->btc_feature = val;
>>> > + queue_work(ar->workqueue, &ar->restart_work);
>>> > + mutex_unlock(&ar->conf_mutex);
>>>
>>> Shouldn't we test ATH10K_STATE_ON first?
>>
>> The restart process will judge by itself whether the device is on or not, and print below warning in that case:
>>
>> [797424.496190] ath10k_pci 0000:05:00.0: cannot restart a device that
>> hasn't been started
>
> That's just buggy, ath10k should not print anything if a setting is
> changed while interface is down. It's much better we have the check for
> ATH10K_STATE_ON here.
>
Agree, will send v3 include these changes.
BR /Yanbo