Hi David,

On Wednesday, October 06, 2010 3:45 AM, David Miller wrote,

We encourage all new drivers that can use NAPI to do so, just because
recent driver submissions do not do this doesn't mean we should
continue such a mistake ;-)

I understand NAPI's merit.
But, since we have already implemented without NAPI.
If possible, we don't want to use NAPI.

Does your mail mean, for accepting upstream, NAPI is essential for CAN driver ?
If essential, we will just do.

Thanks, Ohtake(OKISemi)
----- Original Message ----- 
From: "David Miller" <davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
To: <masa-korg-ECg8zkTtlr0C6LszWs/t0g@public.gmane.org>
Cc: <redacted>; <sameo-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>; <redacted>; <redacted>;
[off-list ref]; [off-list ref]; [off-list ref]; [off-list ref];
[off-list ref]; [off-list ref]; [off-list ref]; [off-list ref];
[off-list ref]; [off-list ref]; [off-list ref]
Sent: Wednesday, October 06, 2010 3:45 AM
Subject: Re: [MeeGo-Dev][PATCH v3] Topcliff: Update PCH_CAN driver to 2.6.35

From: "Masayuki Ohtake" <masa-korg-ECg8zkTtlr0C6LszWs/t0g@public.gmane.org>
Date: Tue, 5 Oct 2010 21:09:30 +0900

quoted

On Tuesday, October 05, 2010 8:08 PM, Marc Kleine-Budde wrote:

quoted

If FIFO is working you might also think about NAPI.

I think NAPI isn't necessary for our CAN driver.
NAPI is for high-speed networking.
CAN is NOT high-speed.

In fact, some accepted CAN drivers don't have NAPI.

NAPI is not just for performance concerns.

It greatly simplifies the locking since all packet processing paths
run only in software interrupt context, never in hardware interrupt
context.

We encourage all new drivers that can use NAPI to do so, just because
recent driver submissions do not do this doesn't mean we should
continue such a mistake ;-)
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html