From: Morten Rasmussen <morten.rasmussen@arm.com>
To: peterz@infradead.org, mingo@redhat.com
Cc: vincent.guittot@linaro.org,
Dietmar Eggemann <Dietmar.Eggemann@arm.com>,
yuyang.du@intel.com, preeti@linux.vnet.ibm.com,
mturquette@linaro.org, rjw@rjwysocki.net,
Juri Lelli <Juri.Lelli@arm.com>,
sgurrappadi@nvidia.com, pang.xunlei@zte.com.cn,
linux-kernel@vger.kernel.org, linux-pm@vger.kernel.org,
morten.rasmussen@arm.com
Subject: [RFCv4 PATCH 17/34] sched: Introduce SD_SHARE_CAP_STATES sched_domain flag
Date: Tue, 12 May 2015 20:38:52 +0100 [thread overview]
Message-ID: <1431459549-18343-18-git-send-email-morten.rasmussen@arm.com> (raw)
In-Reply-To: <1431459549-18343-1-git-send-email-morten.rasmussen@arm.com>
cpufreq is currently keeping it a secret which cpus are sharing
clock source. The scheduler needs to know about clock domains as well
to become more energy aware. The SD_SHARE_CAP_STATES domain flag
indicates whether cpus belonging to the sched_domain share capacity
states (P-states).
There is no connection with cpufreq (yet). The flag must be set by
the arch specific topology code.
cc: Russell King <linux@arm.linux.org.uk>
cc: Ingo Molnar <mingo@redhat.com>
cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Morten Rasmussen <morten.rasmussen@arm.com>
---
arch/arm/kernel/topology.c | 3 ++-
include/linux/sched.h | 1 +
kernel/sched/core.c | 10 +++++++---
3 files changed, 10 insertions(+), 4 deletions(-)
diff --git a/arch/arm/kernel/topology.c b/arch/arm/kernel/topology.c
index 5867587..b35d3e5 100644
--- a/arch/arm/kernel/topology.c
+++ b/arch/arm/kernel/topology.c
@@ -276,7 +276,8 @@ void store_cpu_topology(unsigned int cpuid)
static inline int cpu_corepower_flags(void)
{
- return SD_SHARE_PKG_RESOURCES | SD_SHARE_POWERDOMAIN;
+ return SD_SHARE_PKG_RESOURCES | SD_SHARE_POWERDOMAIN | \
+ SD_SHARE_CAP_STATES;
}
static struct sched_domain_topology_level arm_topology[] = {
diff --git a/include/linux/sched.h b/include/linux/sched.h
index cf79eef..fe77e54 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -981,6 +981,7 @@ extern void wake_up_q(struct wake_q_head *head);
#define SD_PREFER_SIBLING 0x1000 /* Prefer to place tasks in a sibling domain */
#define SD_OVERLAP 0x2000 /* sched_domains of this level overlap */
#define SD_NUMA 0x4000 /* cross-node balancing */
+#define SD_SHARE_CAP_STATES 0x8000 /* Domain members share capacity state */
#ifdef CONFIG_SCHED_SMT
static inline int cpu_smt_flags(void)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index ab42515..5499b2c 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -5555,7 +5555,8 @@ static int sd_degenerate(struct sched_domain *sd)
SD_BALANCE_EXEC |
SD_SHARE_CPUCAPACITY |
SD_SHARE_PKG_RESOURCES |
- SD_SHARE_POWERDOMAIN)) {
+ SD_SHARE_POWERDOMAIN |
+ SD_SHARE_CAP_STATES)) {
if (sd->groups != sd->groups->next)
return 0;
}
@@ -5587,7 +5588,8 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent)
SD_SHARE_CPUCAPACITY |
SD_SHARE_PKG_RESOURCES |
SD_PREFER_SIBLING |
- SD_SHARE_POWERDOMAIN);
+ SD_SHARE_POWERDOMAIN |
+ SD_SHARE_CAP_STATES);
if (nr_node_ids == 1)
pflags &= ~SD_SERIALIZE;
}
@@ -6241,6 +6243,7 @@ static int sched_domains_curr_level;
* SD_SHARE_PKG_RESOURCES - describes shared caches
* SD_NUMA - describes NUMA topologies
* SD_SHARE_POWERDOMAIN - describes shared power domain
+ * SD_SHARE_CAP_STATES - describes shared capacity states
*
* Odd one out:
* SD_ASYM_PACKING - describes SMT quirks
@@ -6250,7 +6253,8 @@ static int sched_domains_curr_level;
SD_SHARE_PKG_RESOURCES | \
SD_NUMA | \
SD_ASYM_PACKING | \
- SD_SHARE_POWERDOMAIN)
+ SD_SHARE_POWERDOMAIN | \
+ SD_SHARE_CAP_STATES)
static struct sched_domain *
sd_init(struct sched_domain_topology_level *tl, int cpu)
--
1.9.1
next prev parent reply other threads:[~2015-05-12 19:38 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-05-12 19:38 [RFCv4 PATCH 00/34] sched: Energy cost model for energy-aware scheduling Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 01/34] arm: Frequency invariant scheduler load-tracking support Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 02/34] sched: Make load tracking frequency scale-invariant Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 03/34] arm: vexpress: Add CPU clock-frequencies to TC2 device-tree Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 04/34] sched: Convert arch_scale_cpu_capacity() from weak function to #define Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 05/34] arm: Update arch_scale_cpu_capacity() to reflect change to define Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 06/34] sched: Make usage tracking cpu scale-invariant Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 07/34] arm: Cpu invariant scheduler load-tracking support Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 08/34] sched: Get rid of scaling usage by cpu_capacity_orig Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 09/34] sched: Track blocked utilization contributions Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 10/34] sched: Include blocked utilization in usage tracking Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 11/34] sched: Remove blocked load and utilization contributions of dying tasks Morten Rasmussen
2015-05-13 0:33 ` Sai Gurrappadi
2015-05-13 13:49 ` Morten Rasmussen
2015-05-19 14:22 ` Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 12/34] sched: Initialize CFS task load and usage before placing task on rq Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 13/34] sched: Documentation for scheduler energy cost model Morten Rasmussen
2015-05-20 4:04 ` Kamalesh Babulal
2015-05-20 9:27 ` Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 14/34] sched: Make energy awareness a sched feature Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 15/34] sched: Introduce energy data structures Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 16/34] sched: Allocate and initialize " Morten Rasmussen
2015-05-12 19:38 ` Morten Rasmussen [this message]
2015-05-12 19:38 ` [RFCv4 PATCH 18/34] arm: topology: Define TC2 energy and provide it to the scheduler Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 19/34] sched: Compute cpu capacity available at current frequency Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 20/34] sched: Relocated get_cpu_usage() and change return type Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 21/34] sched: Highest energy aware balancing sched_domain level pointer Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 22/34] sched: Calculate energy consumption of sched_group Morten Rasmussen
2015-05-21 7:57 ` Kamalesh Babulal
2015-05-22 15:38 ` Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 23/34] sched: Extend sched_group_energy to test load-balancing decisions Morten Rasmussen
2015-05-12 19:38 ` [RFCv4 PATCH 24/34] sched: Estimate energy impact of scheduling decisions Morten Rasmussen
2015-05-12 19:39 ` [RFCv4 PATCH 25/34] sched: Add over-utilization/tipping point indicator Morten Rasmussen
2015-05-22 19:48 ` [PATCH] sched: Fix compiler errors for NO_SMP machines Abel Vesa
2015-05-23 14:52 ` Ingo Molnar
2015-05-23 19:22 ` Abel Vesa
2015-06-30 9:35 ` [RFCv4 PATCH 25/34] sched: Add over-utilization/tipping point indicator pang.xunlei
2015-05-12 19:39 ` [RFCv4 PATCH 26/34] sched: Store system-wide maximum cpu capacity in root domain Morten Rasmussen
2015-05-12 19:39 ` [RFCv4 PATCH 27/34] sched, cpuidle: Track cpuidle state index in the scheduler Morten Rasmussen
2015-05-12 19:39 ` [RFCv4 PATCH 28/34] sched: Count number of shallower idle-states in struct sched_group_energy Morten Rasmussen
2015-05-12 19:39 ` [RFCv4 PATCH 29/34] sched: Determine the current sched_group idle-state Morten Rasmussen
2015-05-12 19:39 ` [RFCv4 PATCH 30/34] sched: Add cpu capacity awareness to wakeup balancing Morten Rasmussen
2015-05-12 19:39 ` [RFCv4 PATCH 31/34] sched: Energy-aware wake-up task placement Morten Rasmussen
2015-05-14 14:03 ` Dietmar Eggemann
[not found] ` <OF168B7415.9556008C-ON48257E45.003388D7-48257E45.00349D8D@zte.com.cn>
2015-05-14 15:10 ` Morten Rasmussen
2015-05-12 19:39 ` [RFCv4 PATCH 32/34] sched: Consider a not over-utilized energy-aware system as balanced Morten Rasmussen
2015-05-12 19:39 ` [RFCv4 PATCH 33/34] sched: Enable idle balance to pull single task towards cpu with higher capacity Morten Rasmussen
2015-05-12 19:39 ` [RFCv4 PATCH 34/34] sched: Disable energy-unfriendly nohz kicks Morten Rasmussen
2015-05-12 22:07 ` [RFCv4 PATCH 00/34] sched: Energy cost model for energy-aware scheduling Sai Gurrappadi
2015-05-13 13:47 ` Morten Rasmussen
2015-06-28 20:26 ` Abel Vesa
2015-06-29 9:06 ` pang.xunlei
2015-06-29 10:19 ` Dietmar Eggemann
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=1431459549-18343-18-git-send-email-morten.rasmussen@arm.com \
--to=morten.rasmussen@arm.com \
--cc=Dietmar.Eggemann@arm.com \
--cc=Juri.Lelli@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=mturquette@linaro.org \
--cc=pang.xunlei@zte.com.cn \
--cc=peterz@infradead.org \
--cc=preeti@linux.vnet.ibm.com \
--cc=rjw@rjwysocki.net \
--cc=sgurrappadi@nvidia.com \
--cc=vincent.guittot@linaro.org \
--cc=yuyang.du@intel.com \
/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;
as well as URLs for NNTP newsgroup(s).