Re: [PATCH net-next 2/2] net: wwan: t7xx: Add NAPI support

Hi Andy,

On 10/10/2022 8:51 PM, Andy Shevchenko wrote:

On Mon, Oct 10, 2022 at 07:52:49PM +0530, Kancharla, Sreehari wrote:

quoted

On 9/12/2022 6:23 PM, Loic Poulain wrote:

quoted

On Fri, 9 Sept 2022 at 18:40, Sreehari Kancharla
[off-list ref] wrote:

...

quoted

quoted

quoted

          if (!rxq->que_started) {
                  atomic_set(&rxq->rx_processing, 0);
-               dev_err(dpmaif_ctrl->dev, "Work RXQ: %d has not been started\n", rxq->index);
-               return;
+               dev_err(rxq->dpmaif_ctrl->dev, "Work RXQ: %d has not been started\n", rxq->index);
+               return work_done;
          }

-       ret = pm_runtime_resume_and_get(dpmaif_ctrl->dev);
-       if (ret < 0 && ret != -EACCES)
-               return;
+       if (!rxq->sleep_lock_pending) {
+               ret = pm_runtime_resume_and_get(rxq->dpmaif_ctrl->dev);

AFAIK, polling is not called in a context allowing you to sleep (e.g.
performing a synced pm runtime operation).

Device will be in resumed state when NAPI poll is invoked from IRQ context,
but host/driver can trigger RPM suspend to device. so we are using pm_runtime_resume_and_get
here to prevent runtime suspend.

If it's known that device is always in power on state here, there is no need to
call all this, but what you need is to bump the reference counter. Perhaps you
need pm_runtime_get_noresume(). I.o.w. find the proper one and check that is
not sleeping.

Agree, incrementing the reference counter will be sufficient. we will replace
pm_runtime_resume_and_get with pm_runtime_get_noresume in v2 submission.

Regards,
Sreehari