* Re: [PATCH 1/4] mt76: mt7921e: fix race issue between reset and suspend/resume
[not found] <YtXZwKyAh/MZ73rf@lore-desk--annotate>
@ 2022-07-19 20:44 ` sean.wang
0 siblings, 0 replies; 3+ messages in thread
From: sean.wang @ 2022-07-19 20:44 UTC (permalink / raw)
To: lorenzo.bianconi
Cc: nbd, sean.wang, Soul.Huang, YN.Chen, Leon.Yen, Eric-SY.Chang,
Deren.Wu, km.lin, jenhao.yang, robin.chiu, Eddie.Chen, ch.yeh,
posh.sun, ted.huang, Stella.Chang, Tom.Chou, steve.lee, jsiuda,
frankgor, jemele, abhishekpandit, shawnku, linux-wireless,
linux-mediatek
From: Sean Wang <sean.wang@mediatek.com>
>> From: Sean Wang <sean.wang@mediatek.com>
>>
>> It is unexpected that the reset work is running simultaneously with
>> the suspend or resume context and it is possible that reset work is
<snip>
>> @@ -363,6 +363,7 @@ static int mt7921_pci_suspend(struct device *device)
>> int i, err;
>>
>> pm->suspended = true;
>> + cancel_work_sync(&dev->reset_work);
>
>should we just wait for the reset to complete here instead of cancelling? (e.g. flush_work)
>
yes, that seemed look better to me. It would wait until the queued reset request finishes.
>Regards,
>Lorenzo
>
>> cancel_delayed_work_sync(&pm->ps_work);
>> cancel_work_sync(&pm->wake_work);
>>
>> @@ -424,6 +425,9 @@ static int mt7921_pci_suspend(struct device
>> *device)
>> restore_suspend:
>> pm->suspended = false;
>>
>> + if (err < 0)
>> + mt7921_reset(&dev->mt76);
>> +
>> return err;
>> }
>>
>> @@ -437,7 +441,7 @@ static int mt7921_pci_resume(struct device
>> *device)
>>
>> err = mt7921_mcu_drv_pmctrl(dev);
>> if (err < 0)
>> - return err;
>> + goto failed;
>>
>> mt7921_wpdma_reinit_cond(dev);
>>
>> @@ -467,11 +471,12 @@ static int mt7921_pci_resume(struct device *device)
>> mt76_connac_mcu_set_deep_sleep(&dev->mt76, false);
>>
>> err = mt76_connac_mcu_set_hif_suspend(mdev, false);
>> - if (err)
>> - return err;
>> -
>> +failed:
>> pm->suspended = false;
>>
>> + if (err < 0)
>> + mt7921_reset(&dev->mt76);
>> +
>> return err;
>> }
>>
>> --
>> 2.25.1
>>
>
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* [PATCH 1/4] mt76: mt7921e: fix race issue between reset and suspend/resume
@ 2022-07-18 20:51 sean.wang
2022-07-18 22:08 ` Lorenzo Bianconi
0 siblings, 1 reply; 3+ messages in thread
From: sean.wang @ 2022-07-18 20:51 UTC (permalink / raw)
To: nbd, lorenzo.bianconi
Cc: sean.wang, Soul.Huang, YN.Chen, Leon.Yen, Eric-SY.Chang, Deren.Wu,
km.lin, jenhao.yang, robin.chiu, Eddie.Chen, ch.yeh, posh.sun,
ted.huang, Stella.Chang, Tom.Chou, steve.lee, jsiuda, frankgor,
kuabhs, druth, abhishekpandit, shawnku, linux-wireless,
linux-mediatek
From: Sean Wang <sean.wang@mediatek.com>
It is unexpected that the reset work is running simultaneously with
the suspend or resume context and it is possible that reset work is still
running even after mt7921 is suspended if we don't fix the race issue.
Thus, the suspend procedure should be waiting until the reset is completed
at the beginning and ignore the subsequent the reset requests.
In case there is an error that happens during either suspend or resume
handler, we will schedule a reset task to recover the error before
returning the error code to ensure we can immediately fix the error there.
Fixes: 0c1ce9884607 ("mt76: mt7921: add wifi reset support")
Co-developed-by: YN Chen <YN.Chen@mediatek.com>
Signed-off-by: YN Chen <YN.Chen@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
---
drivers/net/wireless/mediatek/mt76/mt7921/mac.c | 5 +++++
drivers/net/wireless/mediatek/mt76/mt7921/pci.c | 13 +++++++++----
2 files changed, 14 insertions(+), 4 deletions(-)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
index 47f0aa81ab02..6bd9fc9228a2 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
@@ -780,6 +780,7 @@ void mt7921_mac_reset_work(struct work_struct *work)
void mt7921_reset(struct mt76_dev *mdev)
{
struct mt7921_dev *dev = container_of(mdev, struct mt7921_dev, mt76);
+ struct mt76_connac_pm *pm = &dev->pm;
if (!dev->hw_init_done)
return;
@@ -787,8 +788,12 @@ void mt7921_reset(struct mt76_dev *mdev)
if (dev->hw_full_reset)
return;
+ if (pm->suspended)
+ return;
+
queue_work(dev->mt76.wq, &dev->reset_work);
}
+EXPORT_SYMBOL_GPL(mt7921_reset);
void mt7921_mac_update_mib_stats(struct mt7921_phy *phy)
{
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
index 9d1ba838e54f..07573ea55389 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
@@ -363,6 +363,7 @@ static int mt7921_pci_suspend(struct device *device)
int i, err;
pm->suspended = true;
+ cancel_work_sync(&dev->reset_work);
cancel_delayed_work_sync(&pm->ps_work);
cancel_work_sync(&pm->wake_work);
@@ -424,6 +425,9 @@ static int mt7921_pci_suspend(struct device *device)
restore_suspend:
pm->suspended = false;
+ if (err < 0)
+ mt7921_reset(&dev->mt76);
+
return err;
}
@@ -437,7 +441,7 @@ static int mt7921_pci_resume(struct device *device)
err = mt7921_mcu_drv_pmctrl(dev);
if (err < 0)
- return err;
+ goto failed;
mt7921_wpdma_reinit_cond(dev);
@@ -467,11 +471,12 @@ static int mt7921_pci_resume(struct device *device)
mt76_connac_mcu_set_deep_sleep(&dev->mt76, false);
err = mt76_connac_mcu_set_hif_suspend(mdev, false);
- if (err)
- return err;
-
+failed:
pm->suspended = false;
+ if (err < 0)
+ mt7921_reset(&dev->mt76);
+
return err;
}
--
2.25.1
^ permalink raw reply related [flat|nested] 3+ messages in thread* Re: [PATCH 1/4] mt76: mt7921e: fix race issue between reset and suspend/resume
2022-07-18 20:51 sean.wang
@ 2022-07-18 22:08 ` Lorenzo Bianconi
0 siblings, 0 replies; 3+ messages in thread
From: Lorenzo Bianconi @ 2022-07-18 22:08 UTC (permalink / raw)
To: sean.wang
Cc: nbd, lorenzo.bianconi, Soul.Huang, YN.Chen, Leon.Yen,
Eric-SY.Chang, Deren.Wu, km.lin, jenhao.yang, robin.chiu,
Eddie.Chen, ch.yeh, posh.sun, ted.huang, Stella.Chang, Tom.Chou,
steve.lee, jsiuda, frankgor, kuabhs, druth, abhishekpandit,
shawnku, linux-wireless, linux-mediatek
[-- Attachment #1: Type: text/plain, Size: 3407 bytes --]
> From: Sean Wang <sean.wang@mediatek.com>
>
> It is unexpected that the reset work is running simultaneously with
> the suspend or resume context and it is possible that reset work is still
> running even after mt7921 is suspended if we don't fix the race issue.
>
> Thus, the suspend procedure should be waiting until the reset is completed
> at the beginning and ignore the subsequent the reset requests.
>
> In case there is an error that happens during either suspend or resume
> handler, we will schedule a reset task to recover the error before
> returning the error code to ensure we can immediately fix the error there.
>
> Fixes: 0c1ce9884607 ("mt76: mt7921: add wifi reset support")
> Co-developed-by: YN Chen <YN.Chen@mediatek.com>
> Signed-off-by: YN Chen <YN.Chen@mediatek.com>
> Signed-off-by: Sean Wang <sean.wang@mediatek.com>
> ---
> drivers/net/wireless/mediatek/mt76/mt7921/mac.c | 5 +++++
> drivers/net/wireless/mediatek/mt76/mt7921/pci.c | 13 +++++++++----
> 2 files changed, 14 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
> index 47f0aa81ab02..6bd9fc9228a2 100644
> --- a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
> +++ b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
> @@ -780,6 +780,7 @@ void mt7921_mac_reset_work(struct work_struct *work)
> void mt7921_reset(struct mt76_dev *mdev)
> {
> struct mt7921_dev *dev = container_of(mdev, struct mt7921_dev, mt76);
> + struct mt76_connac_pm *pm = &dev->pm;
>
> if (!dev->hw_init_done)
> return;
> @@ -787,8 +788,12 @@ void mt7921_reset(struct mt76_dev *mdev)
> if (dev->hw_full_reset)
> return;
>
> + if (pm->suspended)
> + return;
> +
> queue_work(dev->mt76.wq, &dev->reset_work);
> }
> +EXPORT_SYMBOL_GPL(mt7921_reset);
>
> void mt7921_mac_update_mib_stats(struct mt7921_phy *phy)
> {
> diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
> index 9d1ba838e54f..07573ea55389 100644
> --- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
> +++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
> @@ -363,6 +363,7 @@ static int mt7921_pci_suspend(struct device *device)
> int i, err;
>
> pm->suspended = true;
> + cancel_work_sync(&dev->reset_work);
should we just wait for the reset to complete here instead of cancelling? (e.g. flush_work)
Regards,
Lorenzo
> cancel_delayed_work_sync(&pm->ps_work);
> cancel_work_sync(&pm->wake_work);
>
> @@ -424,6 +425,9 @@ static int mt7921_pci_suspend(struct device *device)
> restore_suspend:
> pm->suspended = false;
>
> + if (err < 0)
> + mt7921_reset(&dev->mt76);
> +
> return err;
> }
>
> @@ -437,7 +441,7 @@ static int mt7921_pci_resume(struct device *device)
>
> err = mt7921_mcu_drv_pmctrl(dev);
> if (err < 0)
> - return err;
> + goto failed;
>
> mt7921_wpdma_reinit_cond(dev);
>
> @@ -467,11 +471,12 @@ static int mt7921_pci_resume(struct device *device)
> mt76_connac_mcu_set_deep_sleep(&dev->mt76, false);
>
> err = mt76_connac_mcu_set_hif_suspend(mdev, false);
> - if (err)
> - return err;
> -
> +failed:
> pm->suspended = false;
>
> + if (err < 0)
> + mt7921_reset(&dev->mt76);
> +
> return err;
> }
>
> --
> 2.25.1
>
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2022-07-19 21:05 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <YtXZwKyAh/MZ73rf@lore-desk--annotate>
2022-07-19 20:44 ` [PATCH 1/4] mt76: mt7921e: fix race issue between reset and suspend/resume sean.wang
2022-07-18 20:51 sean.wang
2022-07-18 22:08 ` Lorenzo Bianconi
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox