2024-05-28 15:06:56

by Paul Barker

[permalink] [raw]
Subject: [net-next PATCH v4 2/7] net: ravb: Consider busypolling status when re-enabling interrupts

Make use of the busypolling status returned from NAPI complete to decide
if interrupts shall be re-enabled or not. This is useful to reduce the
interrupt overhead.

While at it switch to using napi_complete_done() as it take into account
the work done when providing the busypolling status.

Signed-off-by: Paul Barker <[email protected]>
---
Changes v3->v4:
* Used Niklas' suggested commit message.

drivers/net/ethernet/renesas/ravb_main.c | 26 ++++++++++--------------
1 file changed, 11 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index 193ad05383a8..472aa80002be 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -1341,23 +1341,19 @@ static int ravb_poll(struct napi_struct *napi, int budget)
if (priv->rx_fifo_errors != ndev->stats.rx_fifo_errors)
ndev->stats.rx_fifo_errors = priv->rx_fifo_errors;

- if (work_done == budget)
- goto out;
-
- napi_complete(napi);
-
- /* Re-enable RX/TX interrupts */
- spin_lock_irqsave(&priv->lock, flags);
- if (!info->irq_en_dis) {
- ravb_modify(ndev, RIC0, mask, mask);
- ravb_modify(ndev, TIC, mask, mask);
- } else {
- ravb_write(ndev, mask, RIE0);
- ravb_write(ndev, mask, TIE);
+ if (work_done < budget && napi_complete_done(napi, work_done)) {
+ /* Re-enable RX/TX interrupts */
+ spin_lock_irqsave(&priv->lock, flags);
+ if (!info->irq_en_dis) {
+ ravb_modify(ndev, RIC0, mask, mask);
+ ravb_modify(ndev, TIC, mask, mask);
+ } else {
+ ravb_write(ndev, mask, RIE0);
+ ravb_write(ndev, mask, TIE);
+ }
+ spin_unlock_irqrestore(&priv->lock, flags);
}
- spin_unlock_irqrestore(&priv->lock, flags);

-out:
return work_done;
}

--
2.39.2



2024-05-28 16:44:44

by Sergey Shtylyov

[permalink] [raw]
Subject: Re: [net-next PATCH v4 2/7] net: ravb: Consider busypolling status when re-enabling interrupts

On 5/28/24 6:03 PM, Paul Barker wrote:

> Make use of the busypolling status returned from NAPI complete to decide

My spellchecker/translator trip over "busypolling" -- consider using
"busy-polling"?
And did you actually mean napi_complete_done()?

> if interrupts shall be re-enabled or not. This is useful to reduce the
> interrupt overhead.
>
> While at it switch to using napi_complete_done() as it take into account

Takes.

> the work done when providing the busypolling status.

Again, "busy-polling"?

> Signed-off-by: Paul Barker <[email protected]>

[...]

MBR, Sergey

2024-05-28 16:47:46

by Sergey Shtylyov

[permalink] [raw]
Subject: Re: [net-next PATCH v4 2/7] net: ravb: Consider busypolling status when re-enabling interrupts

On 5/28/24 7:44 PM, Sergey Shtylyov wrote:

>> Make use of the busypolling status returned from NAPI complete to decide
>
> My spellchecker/translator trip over "busypolling" -- consider using
> "busy-polling"?
> And did you actually mean napi_complete_done()?

Ah, napi_complete() also returns a result... maybe this should be reworded
as "NAPI completion"?

>> if interrupts shall be re-enabled or not. This is useful to reduce the
>> interrupt overhead.
>>
>> While at it switch to using napi_complete_done() as it take into account
>
> Takes.
>
>> the work done when providing the busypolling status.
>
> Again, "busy-polling"?
>
>> Signed-off-by: Paul Barker <[email protected]>

[...]

MBR, Sergey

2024-05-29 19:09:30

by Paul Barker

[permalink] [raw]
Subject: Re: [net-next PATCH v4 2/7] net: ravb: Consider busypolling status when re-enabling interrupts

On 28/05/2024 17:47, Sergey Shtylyov wrote:
> On 5/28/24 7:44 PM, Sergey Shtylyov wrote:
>
>>> Make use of the busypolling status returned from NAPI complete to decide
>>
>> My spellchecker/translator trip over "busypolling" -- consider using
>> "busy-polling"?
>> And did you actually mean napi_complete_done()?
>
> Ah, napi_complete() also returns a result... maybe this should be reworded
> as "NAPI completion"?
>
>>> if interrupts shall be re-enabled or not. This is useful to reduce the
>>> interrupt overhead.
>>>
>>> While at it switch to using napi_complete_done() as it take into account
>>
>> Takes.
>>
>>> the work done when providing the busypolling status.
>>
>> Again, "busy-polling"?

Ack to all of the above.

I used the commit message suggested by Niklas here. I'll revise it a
little for v5...

Thanks,

--
Paul Barker


Attachments:
OpenPGP_0x27F4B3459F002257.asc (3.49 kB)
OpenPGP public key
OpenPGP_signature.asc (243.00 B)
OpenPGP digital signature
Download all attachments