From: Arnd Bergmann <arnd@arndb.de>
To: Peppe CAVALLARO <peppe.cavallaro@st.com>
Cc: Thomas Gleixner <tglx@linutronix.de>,
Stuart MENEFY <stuart.menefy@st.com>,
"linux-sh@vger.kernel.org" <linux-sh@vger.kernel.org>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
John Stultz <johnstul@us.ibm.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux@arm.linux.org.uk" <linux@arm.linux.org.uk>
Subject: Re: [PATCH (sh-2.6) 1/4] clksource: Generic timer infrastructure
Date: Thu, 03 Mar 2011 13:55:24 +0000 [thread overview]
Message-ID: <201103031455.25089.arnd@arndb.de> (raw)
In-Reply-To: <4D6F6C96.1010303@st.com>
On Thursday 03 March 2011, Peppe CAVALLARO wrote:
> This logic is already in the driver, indeed.
> What I've seen on our embedded systems is that the
> cost of RX interrupts is very hight and NAPI partially helps.
> Typically, in an IP-STB, I receive a burst of UDP pkt
> and this means that many interrupts occur (~99% of CPU
> usage on slow platforms).
> With the ext timer I was able to reduce the CPU usage in
> these kind of scenarios to ~50%.
I don't understand. Shouldn't the interrupts be stopped as long
as the system is busy? This sounds like a bug in your NAPI
handling, or maybe you just need to use a lower NAPI_WEIGHT
so you stay in polling mode longer.
Arnd
WARNING: multiple messages have this Message-ID (diff)
From: Arnd Bergmann <arnd@arndb.de>
To: Peppe CAVALLARO <peppe.cavallaro@st.com>
Cc: Thomas Gleixner <tglx@linutronix.de>,
Stuart MENEFY <stuart.menefy@st.com>,
"linux-sh@vger.kernel.org" <linux-sh@vger.kernel.org>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
John Stultz <johnstul@us.ibm.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux@arm.linux.org.uk" <linux@arm.linux.org.uk>
Subject: Re: [PATCH (sh-2.6) 1/4] clksource: Generic timer infrastructure
Date: Thu, 3 Mar 2011 14:55:24 +0100 [thread overview]
Message-ID: <201103031455.25089.arnd@arndb.de> (raw)
In-Reply-To: <4D6F6C96.1010303@st.com>
On Thursday 03 March 2011, Peppe CAVALLARO wrote:
> This logic is already in the driver, indeed.
> What I've seen on our embedded systems is that the
> cost of RX interrupts is very hight and NAPI partially helps.
> Typically, in an IP-STB, I receive a burst of UDP pkt
> and this means that many interrupts occur (~99% of CPU
> usage on slow platforms).
> With the ext timer I was able to reduce the CPU usage in
> these kind of scenarios to ~50%.
I don't understand. Shouldn't the interrupts be stopped as long
as the system is busy? This sounds like a bug in your NAPI
handling, or maybe you just need to use a lower NAPI_WEIGHT
so you stay in polling mode longer.
Arnd
next prev parent reply other threads:[~2011-03-03 13:55 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-02-22 10:17 [PATCH 0/4] simple generic timer infrastructure and stmmac example Peppe CAVALLARO
2011-02-22 10:17 ` Peppe CAVALLARO
2011-02-22 10:17 ` [PATCH (sh-2.6) 1/4] clksource: Generic timer infrastructure Peppe CAVALLARO
2011-02-22 10:17 ` Peppe CAVALLARO
2011-02-24 17:20 ` Arnd Bergmann
2011-02-24 17:20 ` Arnd Bergmann
2011-03-01 15:20 ` Stuart Menefy
2011-03-01 15:20 ` Stuart Menefy
2011-03-01 16:43 ` Arnd Bergmann
2011-03-01 16:43 ` Arnd Bergmann
2011-03-01 20:26 ` Russell King - ARM Linux
2011-03-01 20:26 ` Russell King - ARM Linux
2011-03-01 20:41 ` Arnd Bergmann
2011-03-01 20:41 ` Arnd Bergmann
2011-03-01 16:48 ` Thomas Gleixner
2011-03-01 16:48 ` Thomas Gleixner
2011-03-02 17:35 ` Peppe CAVALLARO
2011-03-02 17:35 ` Peppe CAVALLARO
2011-03-03 8:45 ` Arnd Bergmann
2011-03-03 8:45 ` Arnd Bergmann
2011-03-03 10:25 ` Peppe CAVALLARO
2011-03-03 10:25 ` Peppe CAVALLARO
2011-03-03 13:55 ` Arnd Bergmann [this message]
2011-03-03 13:55 ` Arnd Bergmann
2011-03-04 6:53 ` Peppe CAVALLARO
2011-03-04 6:53 ` Peppe CAVALLARO
2012-06-12 3:04 ` Paul Mundt
2012-06-12 3:04 ` Paul Mundt
2011-02-22 10:17 ` [PATCH (sh-2.6) 2/4] sh_timer: add the support to use the generic Peppe CAVALLARO
2011-02-22 10:17 ` [PATCH (sh-2.6) 2/4] sh_timer: add the support to use the generic timer Peppe CAVALLARO
2011-02-22 10:17 ` [PATCH (net-2.6) 3/4] stmmac: switch to use the new " Peppe CAVALLARO
2011-02-22 10:17 ` [PATCH (net-2.6) 3/4] stmmac: switch to use the new generic timer interface Peppe CAVALLARO
2011-02-22 10:17 ` [PATCH (net-2.6) 4/4] stmmac: rework and improvement the stmmac Peppe CAVALLARO
2011-02-22 10:17 ` [PATCH (net-2.6) 4/4] stmmac: rework and improvement the stmmac timer Peppe CAVALLARO
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=201103031455.25089.arnd@arndb.de \
--to=arnd@arndb.de \
--cc=johnstul@us.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-sh@vger.kernel.org \
--cc=linux@arm.linux.org.uk \
--cc=netdev@vger.kernel.org \
--cc=peppe.cavallaro@st.com \
--cc=stuart.menefy@st.com \
--cc=tglx@linutronix.de \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.