From: David Miller <davem@davemloft.net>
To: mans@mansr.com
Cc: romieu@fr.zoreil.com, linux-kernel@vger.kernel.org,
netdev@vger.kernel.org, slash.tmp@free.fr
Subject: Re: [PATCH v5] net: ethernet: add driver for Aurora VLSI NB8800 Ethernet controller
Date: Tue, 10 Nov 2015 21:11:21 -0500 (EST) [thread overview]
Message-ID: <20151110.211121.422193831316296487.davem@davemloft.net> (raw)
In-Reply-To: <yw1xio59pdw6.fsf@unicorn.mansr.com>
From: Måns Rullgård <mans@mansr.com>
Date: Wed, 11 Nov 2015 00:40:09 +0000
> When the DMA complete interrupt arrives, the next chain should be
> kicked off as quickly as possible, and I don't see why that would
> benefit from being done in napi context.
NAPI isn't about low latency, it's about fairness and interrupt
mitigation.
You probably don't even realize that all of the TX SKB freeing you do
in the hardware interrupt handler end up being actually processed by a
scheduled software interrupt anyways.
So you are gaining almost nothing by not doing TX completion in NAPI
context, whereas by doing so you would be gaining a lot including
more simplified locking or even the ability to do no locking at all.
next prev parent reply other threads:[~2015-11-11 2:11 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-11-10 16:14 Mans Rullgard
2015-11-10 17:55 ` Eric Dumazet
2015-11-10 18:05 ` Måns Rullgård
2015-11-10 20:04 ` David Miller
2015-11-10 20:53 ` Måns Rullgård
2015-11-10 21:06 ` David Miller
2015-11-10 21:21 ` Måns Rullgård
2015-11-10 21:51 ` Eric Dumazet
2015-11-11 13:41 ` Mason
2015-11-11 13:54 ` Måns Rullgård
2015-11-11 14:10 ` Eric Dumazet
2015-11-11 14:06 ` Eric Dumazet
2015-11-10 17:58 ` Eric Dumazet
2015-11-10 18:07 ` Måns Rullgård
2015-11-10 19:13 ` Mason
2015-11-10 19:25 ` Måns Rullgård
2015-11-12 13:33 ` Mason
2015-11-12 14:04 ` Måns Rullgård
2015-11-12 16:19 ` Mason
2015-11-12 16:57 ` Måns Rullgård
2015-11-12 17:20 ` Måns Rullgård
2015-11-10 22:09 ` Andy Shevchenko
2015-11-10 22:34 ` Måns Rullgård
2015-11-10 22:40 ` Andy Shevchenko
2015-11-10 23:07 ` Måns Rullgård
2015-11-11 0:36 ` Andy Shevchenko
2015-11-11 0:44 ` Måns Rullgård
2015-11-10 23:34 ` Francois Romieu
2015-11-11 0:40 ` Måns Rullgård
2015-11-11 2:11 ` David Miller [this message]
2015-11-11 12:22 ` Måns Rullgård
2015-11-11 13:04 ` Måns Rullgård
2015-11-11 13:29 ` Eric Dumazet
2015-11-11 13:48 ` Måns Rullgård
2015-11-11 14:09 ` Eric Dumazet
2015-11-11 14:15 ` Måns Rullgård
2015-11-11 14:35 ` Måns Rullgård
2015-11-11 14:44 ` Eric Dumazet
2015-11-11 14:42 ` Eric Dumazet
2015-11-11 16:24 ` David Miller
2015-11-11 18:25 ` Måns Rullgård
2015-11-11 19:02 ` David Miller
2015-11-11 19:09 ` Måns Rullgård
2015-11-11 19:13 ` David Miller
2015-11-11 19:17 ` Måns Rullgård
2015-11-11 19:19 ` David Miller
2015-11-11 19:25 ` Måns Rullgård
2015-11-11 19:26 ` David Miller
2015-11-11 19:35 ` Måns Rullgård
2015-11-11 19:48 ` David Miller
2015-11-11 20:47 ` Måns Rullgård
2015-11-11 16:20 ` David Miller
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20151110.211121.422193831316296487.davem@davemloft.net \
--to=davem@davemloft.net \
--cc=linux-kernel@vger.kernel.org \
--cc=mans@mansr.com \
--cc=netdev@vger.kernel.org \
--cc=romieu@fr.zoreil.com \
--cc=slash.tmp@free.fr \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®