From: tip-bot for Vikas Shivappa <tipbot@zytor.com>
To: linux-tip-commits@vger.kernel.org
Cc: hpa@zytor.com, mingo@kernel.org, linux-kernel@vger.kernel.org,
tglx@linutronix.de, vikas.shivappa@linux.intel.com
Subject: [tip:x86/cache] x86/intel_rdt/cqm: Add sched_in support
Date: Tue, 1 Aug 2017 13:55:33 -0700 [thread overview]
Message-ID: <tip-748b6b881ccdda8f0663c68605f431279e06f49a@git.kernel.org> (raw)
In-Reply-To: <1501017287-28083-25-git-send-email-vikas.shivappa@linux.intel.com>
Commit-ID: 748b6b881ccdda8f0663c68605f431279e06f49a
Gitweb: http://git.kernel.org/tip/748b6b881ccdda8f0663c68605f431279e06f49a
Author: Vikas Shivappa <vikas.shivappa@linux.intel.com>
AuthorDate: Tue, 25 Jul 2017 14:14:43 -0700
Committer: Thomas Gleixner <tglx@linutronix.de>
CommitDate: Tue, 1 Aug 2017 22:41:28 +0200
x86/intel_rdt/cqm: Add sched_in support
OS associates an RMID/CLOSid to a task by writing the per CPU
IA32_PQR_ASSOC MSR when a task is scheduled in.
The sched_in code will stay as no-op unless we are running on Intel SKU
which supports either resource control or monitoring and we also enable
them by mounting the resctrl fs. The per cpu CLOSid/RMID values are
cached and the write is performed only when a task with a different
CLOSid/RMID is scheduled in.
Signed-off-by: Vikas Shivappa <vikas.shivappa@linux.intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: ravi.v.shankar@intel.com
Cc: tony.luck@intel.com
Cc: fenghua.yu@intel.com
Cc: peterz@infradead.org
Cc: eranian@google.com
Cc: vikas.shivappa@intel.com
Cc: ak@linux.intel.com
Cc: davidcc@google.com
Cc: reinette.chatre@intel.com
Link: http://lkml.kernel.org/r/1501017287-28083-25-git-send-email-vikas.shivappa@linux.intel.com
---
arch/x86/include/asm/intel_rdt_sched.h | 50 ++++++++++++++++++++--------------
arch/x86/kernel/cpu/intel_rdt.h | 4 ---
2 files changed, 29 insertions(+), 25 deletions(-)
diff --git a/arch/x86/include/asm/intel_rdt_sched.h b/arch/x86/include/asm/intel_rdt_sched.h
index 8c5be01..3badc0a 100644
--- a/arch/x86/include/asm/intel_rdt_sched.h
+++ b/arch/x86/include/asm/intel_rdt_sched.h
@@ -15,7 +15,8 @@
*
* The upper 32 bits of IA32_PQR_ASSOC contain closid and the
* lower 10 bits rmid. The update to IA32_PQR_ASSOC always
- * contains both parts, so we need to cache them.
+ * contains both parts, so we need to cache them. This also
+ * stores the user configured per cpu CLOSID and RMID.
*
* The cache also helps to avoid pointless updates if the value does
* not change.
@@ -30,38 +31,45 @@ DECLARE_PER_CPU_READ_MOSTLY(struct intel_pqr_state, rdt_cpu_default);
DECLARE_STATIC_KEY_FALSE(rdt_enable_key);
DECLARE_STATIC_KEY_FALSE(rdt_alloc_enable_key);
+DECLARE_STATIC_KEY_FALSE(rdt_mon_enable_key);
/*
- * __intel_rdt_sched_in() - Writes the task's CLOSid to IA32_PQR_MSR
+ * __intel_rdt_sched_in() - Writes the task's CLOSid/RMID to IA32_PQR_MSR
*
* Following considerations are made so that this has minimal impact
* on scheduler hot path:
* - This will stay as no-op unless we are running on an Intel SKU
- * which supports resource control and we enable by mounting the
- * resctrl file system.
- * - Caches the per cpu CLOSid values and does the MSR write only
- * when a task with a different CLOSid is scheduled in.
- *
+ * which supports resource control or monitoring and we enable by
+ * mounting the resctrl file system.
+ * - Caches the per cpu CLOSid/RMID values and does the MSR write only
+ * when a task with a different CLOSid/RMID is scheduled in.
+ * - We allocate RMIDs/CLOSids globally in order to keep this as
+ * simple as possible.
* Must be called with preemption disabled.
*/
-static inline void __intel_rdt_sched_in(void)
+static void __intel_rdt_sched_in(void)
{
+ struct intel_pqr_state newstate = this_cpu_read(rdt_cpu_default);
+ struct intel_pqr_state *curstate = this_cpu_ptr(&pqr_state);
+
+ /*
+ * If this task has a closid/rmid assigned, use it.
+ * Else use the closid/rmid assigned to this cpu.
+ */
if (static_branch_likely(&rdt_alloc_enable_key)) {
- struct intel_pqr_state *state = this_cpu_ptr(&pqr_state);
- u32 closid;
+ if (current->closid)
+ newstate.closid = current->closid;
+ }
- /*
- * If this task has a closid assigned, use it.
- * Else use the closid assigned to this cpu.
- */
- closid = current->closid;
- if (closid == 0)
- closid = this_cpu_read(rdt_cpu_default.closid);
+ if (static_branch_likely(&rdt_mon_enable_key)) {
+ if (current->rmid)
+ newstate.rmid = current->rmid;
+ }
- if (closid != state->closid) {
- state->closid = closid;
- wrmsr(IA32_PQR_ASSOC, state->rmid, closid);
- }
+ if (newstate.closid != curstate->closid ||
+ newstate.rmid != curstate->rmid) {
+ *curstate = newstate;
+ wrmsr(IA32_PQR_ASSOC, newstate.rmid, newstate.closid);
}
}
diff --git a/arch/x86/kernel/cpu/intel_rdt.h b/arch/x86/kernel/cpu/intel_rdt.h
index 92a5d30..7fcaa5f 100644
--- a/arch/x86/kernel/cpu/intel_rdt.h
+++ b/arch/x86/kernel/cpu/intel_rdt.h
@@ -22,8 +22,6 @@
#define RMID_VAL_ERROR BIT_ULL(63)
#define RMID_VAL_UNAVAIL BIT_ULL(62)
-DECLARE_STATIC_KEY_FALSE(rdt_enable_key);
-
/**
* struct mon_evt - Entry in the event list of a resource
* @evtid: event id
@@ -61,8 +59,6 @@ extern bool rdt_alloc_capable;
extern bool rdt_mon_capable;
extern unsigned int rdt_mon_features;
-DECLARE_STATIC_KEY_FALSE(rdt_mon_enable_key);
-
enum rdt_group_type {
RDTCTRL_GROUP = 0,
RDTMON_GROUP,
next prev parent reply other threads:[~2017-08-01 20:56 UTC|newest]
Thread overview: 68+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-07-25 21:14 [PATCH 00/28 V2] Cqm3 patch series(along with MBM support) Vikas Shivappa
2017-07-25 21:14 ` [PATCH 01/28] x86/perf/cqm: Wipe out perf based cqm Vikas Shivappa
2017-08-01 20:46 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 02/28] x86/intel_rdt/cqm: Documentation for resctrl based RDT Monitoring Vikas Shivappa
2017-08-01 20:46 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 03/28] x86/intel_rdt: Introduce a common compile option for RDT Vikas Shivappa
2017-08-01 20:47 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 04/28] x86/intel_rdt: Change file names to accommodate RDT monitor code Vikas Shivappa
2017-08-01 20:47 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 05/28] x86/intel_rdt: Mark rdt_root and closid_alloc as static Vikas Shivappa
2017-08-01 20:48 ` [tip:x86/cache] " tip-bot for Reinette Chatre
2017-07-25 21:14 ` [PATCH 06/28] x86/intel_rdt: Cleanup namespace to support RDT monitoring Vikas Shivappa
2017-08-01 20:48 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 07/28] x86/intel_rdt: make rdt_resources_all more readable Vikas Shivappa
2017-08-01 20:48 ` [tip:x86/cache] x86/intel_rdt: Make " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 08/28] x86/intel_rdt/cqm: Add RDT monitoring initialization Vikas Shivappa
2017-08-01 20:49 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 09/28] x86/intel_rdt/cqm: Add RMID(Resource monitoring ID) management Vikas Shivappa
2017-08-01 20:14 ` Thomas Gleixner
2017-08-02 7:41 ` Shivappa Vikas
2017-08-01 20:49 ` [tip:x86/cache] x86/intel_rdt/cqm: Add RMID (Resource " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 10/28] x86/intel_rdt: Simplify info and base file lists Vikas Shivappa
2017-08-01 20:34 ` Thomas Gleixner
2017-08-01 20:50 ` [tip:x86/cache] " tip-bot for Tony luck
2017-07-25 21:14 ` [PATCH 11/28] x86/intel_rdt/cqm: Add info files for RDT monitoring Vikas Shivappa
2017-08-01 20:40 ` Thomas Gleixner
2017-08-01 20:50 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 12/28] x86/intel_rdt: Prepare for RDT monitoring mkdir support Vikas Shivappa
2017-08-01 20:50 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 13/28] x86/intel_rdt/cqm: Add mkdir support for RDT monitoring Vikas Shivappa
2017-08-01 20:51 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 14/28] x86/intel_rdt: Change closid type from int to u32 Vikas Shivappa
2017-08-01 20:51 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 15/28] x86/intel_rdt/cqm: Add tasks file support Vikas Shivappa
2017-08-01 20:52 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 16/28] x86/intel_rdt: Prepare to add RDT monitor cpus " Vikas Shivappa
2017-08-01 19:49 ` Thomas Gleixner
2017-08-01 21:35 ` Shivappa Vikas
2017-08-01 20:52 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 17/28] x86/intel_rdt/cqm: Add " Vikas Shivappa
2017-08-01 20:52 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 18/28] x86/intel_rdt: Prepare for RDT monitor data support Vikas Shivappa
2017-08-01 20:53 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 19/28] x86/intel_rdt/cqm: Add mon_data Vikas Shivappa
2017-08-01 20:53 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 20/28] x86/intel_rdt: Separate the ctrl bits from rmdir Vikas Shivappa
2017-08-01 20:54 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 21/28] x86/intel_rdt/cqm: Add rmdir support Vikas Shivappa
2017-08-01 20:54 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 22/28] x86/intel_rdt/cqm: Add mount,umount support Vikas Shivappa
2017-08-01 20:54 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 23/28] x86/intel_rdt: Introduce rdt_enable_key for scheduling Vikas Shivappa
2017-08-01 20:55 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 24/28] x86/intel_rdt/cqm: Add sched_in support Vikas Shivappa
2017-08-01 20:55 ` tip-bot for Vikas Shivappa [this message]
2017-07-25 21:14 ` [PATCH 25/28] x86/intel_rdt/cqm: Add hotcpu support Vikas Shivappa
2017-08-01 20:55 ` [tip:x86/cache] x86/intel_rdt/cqm: Add CPU hotplug support tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 26/28] x86/intel_rdt/mbm: Basic counting of MBM events (total and local) Vikas Shivappa
2017-08-01 20:56 ` [tip:x86/cache] " tip-bot for Tony Luck
2017-07-25 21:14 ` [PATCH 27/28] x86/intel_rdt/mbm: Add mbm counter initialization Vikas Shivappa
2017-08-01 20:56 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 21:14 ` [PATCH 28/28] x86/intel_rdt/mbm: Handle counter overflow Vikas Shivappa
2017-08-01 20:57 ` [tip:x86/cache] " tip-bot for Vikas Shivappa
2017-07-25 22:39 ` [PATCH] x86/intel_rdt: Show bitmask of shareable resource with other executing units Luck, Tony
2017-08-01 20:57 ` [tip:x86/cache] " tip-bot for Fenghua Yu
2017-07-25 22:42 ` [RFC PATCH] Add "-f" and "-F" flags to watch a "/sys? style file (single number) Luck, Tony
2017-08-01 20:51 ` [PATCH 00/28 V2] Cqm3 patch series(along with MBM support) Thomas Gleixner
2017-08-01 21:39 ` Shivappa Vikas
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=tip-748b6b881ccdda8f0663c68605f431279e06f49a@git.kernel.org \
--to=tipbot@zytor.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-tip-commits@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=tglx@linutronix.de \
--cc=vikas.shivappa@linux.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
Powered by JetHome