All of lore.kernel.org
 help / color / mirror / Atom feed
From: Michael Neuling <mikey@neuling.org>
To: Shilpasri G Bhat <shilpa.bhat@linux.vnet.ibm.com>,
	linuxppc-dev@ozlabs.org, linux-kernel@vger.kernel.org
Cc: ego@linux.vnet.ibm.com, linux-pm@vger.kernel.org,
	viresh.kumar@linaro.org, rjw@rjwysocki.net, pc@us.ibm.com,
	shreyas@linux.vnet.ibm.com, anton@samba.org
Subject: Re: [PATCH v8 3/6] cpufreq: powernv: Remove cpu_to_chip_id() from hot-path
Date: Fri, 18 Mar 2016 15:11:48 +1100	[thread overview]
Message-ID: <1458274308.6622.79.camel@neuling.org> (raw)
In-Reply-To: <1458273857.6622.75.camel@neuling.org>

On Fri, 2016-03-18 at 15:04 +1100, Michael Neuling wrote:

> On Wed, 2016-02-03 at 01:11 +0530, Shilpasri G Bhat wrote:
> 

> > cpu_to_chip_id() does a DT walk through to find out the chip id by
> > taking a contended device tree lock. This adds an unnecessary
> > overhead
> > in a hot path. So instead of calling cpu_to_chip_id() everytime
> > cache
> > the chip ids for all cores in the array 'core_to_chip_map' and use
> > it
> > in the hotpath.
> > 
> > Reported-by: Anton Blanchard <anton@samba.org>
> > Signed-off-by: Shilpasri G Bhat <shilpa.bhat@linux.vnet.ibm.com>
> > Reviewed-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com>
> > Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
> > ---
> > No changes from v7.
> 
> How about this instead?  It removes the linear lookup and seems a lot
> less complex.

BTW we never init nr_chips before using it.  We also need something
like.

diff --git a/drivers/cpufreq/powernv-cpufreq.c b/drivers/cpufreq/powernv-cpufreq.c
index d63d2cb..c819ed4 100644
--- a/drivers/cpufreq/powernv-cpufreq.c
+++ b/drivers/cpufreq/powernv-cpufreq.c
@@ -556,6 +556,8 @@ static int init_chip_info(void)
 	unsigned int cpu, i;
 	unsigned int prev_chip_id = UINT_MAX;
 
+	nr_chips = 0;
+
 	for_each_possible_cpu(cpu) {
 		unsigned int id = cpu_to_chip_id(cpu);
 



WARNING: multiple messages have this Message-ID (diff)
From: Michael Neuling <mikey@neuling.org>
To: Shilpasri G Bhat <shilpa.bhat@linux.vnet.ibm.com>,
	linuxppc-dev@ozlabs.org, linux-kernel@vger.kernel.org
Cc: ego@linux.vnet.ibm.com, linux-pm@vger.kernel.org,
	viresh.kumar@linaro.org,  rjw@rjwysocki.net, pc@us.ibm.com,
	shreyas@linux.vnet.ibm.com, anton@samba.org
Subject: Re: [PATCH v8 3/6] cpufreq: powernv: Remove cpu_to_chip_id() from hot-path
Date: Fri, 18 Mar 2016 15:11:48 +1100	[thread overview]
Message-ID: <1458274308.6622.79.camel@neuling.org> (raw)
In-Reply-To: <1458273857.6622.75.camel@neuling.org>

On Fri, 2016-03-18 at 15:04 +1100, Michael Neuling wrote:

> On Wed, 2016-02-03 at 01:11 +0530, Shilpasri G Bhat wrote:
>=20

> > cpu_to_chip_id() does a DT walk through to find out the chip id by
> > taking a contended device tree lock. This adds an unnecessary
> > overhead
> > in a hot path. So instead of calling cpu_to_chip_id() everytime
> > cache
> > the chip ids for all cores in the array 'core_to_chip_map' and use
> > it
> > in the hotpath.
> >=20
> > Reported-by: Anton Blanchard <anton@samba.org>
> > Signed-off-by: Shilpasri G Bhat <shilpa.bhat@linux.vnet.ibm.com>
> > Reviewed-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com>
> > Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
> > ---
> > No changes from v7.
>=20
> How about this instead?  It removes the linear lookup and seems a lot
> less complex.

BTW we never init nr_chips before using it.  We also need something
like.

diff --git a/drivers/cpufreq/powernv-cpufreq.c b/drivers/cpufreq/powernv-cp=
ufreq.c
index d63d2cb..c819ed4 100644
--- a/drivers/cpufreq/powernv-cpufreq.c
+++ b/drivers/cpufreq/powernv-cpufreq.c
@@ -556,6 +556,8 @@ static int init_chip_info(void)
 	unsigned int cpu, i;
 	unsigned int prev_chip_id =3D UINT_MAX;
=20
+	nr_chips =3D 0;
+
 	for_each_possible_cpu(cpu) {
 		unsigned int id =3D cpu_to_chip_id(cpu);
=20

  reply	other threads:[~2016-03-18  4:11 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-02-02 19:41 [PATCH v8 0/6] cpufreq: powernv: Redesign the presentation of throttle notification and solve bug-fixes in the driver Shilpasri G Bhat
2016-02-02 19:41 ` [PATCH v8 1/6] cpufreq: powernv: Free 'chips' on module exit Shilpasri G Bhat
2016-02-02 19:41 ` [PATCH v8 2/6] cpufreq: powernv: Hot-plug safe the kworker thread Shilpasri G Bhat
2016-02-02 19:41 ` [PATCH v8 3/6] cpufreq: powernv: Remove cpu_to_chip_id() from hot-path Shilpasri G Bhat
2016-03-18  4:04   ` Michael Neuling
2016-03-18  4:04     ` Michael Neuling
2016-03-18  4:11     ` Michael Neuling [this message]
2016-03-18  4:11       ` Michael Neuling
2016-03-18 13:13       ` Rafael J. Wysocki
2016-03-18 14:58       ` [PATCH] cpufreq: powernv: Define per_cpu chip pointer to optimize hot-path Shilpasri G Bhat
2016-03-21  7:22         ` Viresh Kumar
2016-03-21 14:13           ` Rafael J. Wysocki
2016-03-18 22:37     ` [PATCH v8 3/6] cpufreq: powernv: Remove cpu_to_chip_id() from hot-path Benjamin Herrenschmidt
2016-03-18 23:20       ` Michael Neuling
2016-03-18 23:20         ` Michael Neuling
2016-02-02 19:41 ` [PATCH v8 4/6] cpufreq: powernv/tracing: Add powernv_throttle tracepoint Shilpasri G Bhat
2016-02-02 19:41 ` [PATCH v8 5/6] cpufreq: powernv: Replace pr_info with trace print for throttle event Shilpasri G Bhat
2016-02-02 19:41 ` [PATCH v8 6/6] cpufreq: powernv: Add sysfs attributes to show throttle stats Shilpasri G Bhat
2016-02-03  8:27   ` Viresh Kumar
2016-02-03  8:42     ` Shilpasri G Bhat
2016-02-03  8:42       ` Shilpasri G Bhat
2016-02-03  9:03       ` Viresh Kumar
2016-02-03 12:02         ` Gautham R Shenoy
2016-02-03 12:02           ` Gautham R Shenoy
2016-02-03 14:06           ` Viresh Kumar
2016-02-03 16:24             ` Shilpasri G Bhat
     [not found]               ` <56B229C9.4010909-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org>
2016-02-04  1:51                 ` Viresh Kumar
2016-02-04  1:51                   ` Viresh Kumar
2016-02-03 13:40 ` [PATCH v8 0/6] cpufreq: powernv: Redesign the presentation of throttle notification and solve bug-fixes in the driver Rafael J. Wysocki
2016-02-03 14:01   ` Viresh Kumar
2016-02-03 16:14   ` Shilpasri G Bhat

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=1458274308.6622.79.camel@neuling.org \
    --to=mikey@neuling.org \
    --cc=anton@samba.org \
    --cc=ego@linux.vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=linuxppc-dev@ozlabs.org \
    --cc=pc@us.ibm.com \
    --cc=rjw@rjwysocki.net \
    --cc=shilpa.bhat@linux.vnet.ibm.com \
    --cc=shreyas@linux.vnet.ibm.com \
    --cc=viresh.kumar@linaro.org \
    /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.