From: Vikas Shivappa <vikas.shivappa@linux.intel.com>
To: linux-kernel@vger.kernel.org
Cc: vikas.shivappa@intel.com, x86@kernel.org, hpa@zytor.com,
tglx@linutronix.de, mingo@kernel.org, peterz@infradead.org,
matt.fleming@intel.com, will.auld@intel.com,
linux-rdt@eclists.intel.com, vikas.shivappa@linux.intel.com
Subject: [PATCH 09/10] x86/intel_rdt: Hot cpu support for Cache Allocation
Date: Fri, 12 Jun 2015 11:17:16 -0700 [thread overview]
Message-ID: <1434133037-25189-10-git-send-email-vikas.shivappa@linux.intel.com> (raw)
In-Reply-To: <1434133037-25189-1-git-send-email-vikas.shivappa@linux.intel.com>
This patch adds hot cpu support for Intel Cache allocation. Support
includes updating the cache bitmask MSRs IA32_L3_QOS_n when a new CPU
package comes online. The IA32_L3_QOS_n MSRs are one per Class of
service on each CPU package. The new package's MSRs are synchronized
with the values of existing MSRs. Also the software cache for
IA32_PQR_ASSOC MSRs are updated during hot cpu notifications.
Signed-off-by: Vikas Shivappa <vikas.shivappa@linux.intel.com>
---
arch/x86/kernel/cpu/intel_rdt.c | 84 ++++++++++++++++++++++++++++++++++++++++-
1 file changed, 82 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kernel/cpu/intel_rdt.c b/arch/x86/kernel/cpu/intel_rdt.c
index f90e7ab..d0be46d 100644
--- a/arch/x86/kernel/cpu/intel_rdt.c
+++ b/arch/x86/kernel/cpu/intel_rdt.c
@@ -25,6 +25,7 @@
#include <linux/slab.h>
#include <linux/err.h>
#include <linux/spinlock.h>
+#include <linux/cpu.h>
#include <asm/intel_rdt.h>
/*
@@ -313,13 +314,84 @@ out:
return err;
}
-static inline void rdt_cpumask_update(int cpu)
+static inline bool rdt_cpumask_update(int cpu)
{
cpumask_t tmp;
cpumask_and(&tmp, &rdt_cpumask, topology_core_cpumask(cpu));
- if (cpumask_empty(&tmp))
+ if (cpumask_empty(&tmp)) {
cpumask_set_cpu(cpu, &rdt_cpumask);
+ return true;
+ }
+
+ return false;
+}
+
+/*
+ * cbm_update_msrs() - Updates all the existing IA32_L3_MASK_n MSRs
+ * which are one per CLOSid except IA32_L3_MASK_0 on the current package.
+ */
+static inline void cbm_update_msrs(void)
+{
+ int maxid = boot_cpu_data.x86_rdt_max_closid;
+ unsigned int i;
+
+ /*
+ * At cpureset, all bits of IA32_L3_MASK_n are set.
+ * The index starts from one as there is no need
+ * to update IA32_L3_MASK_0 as it belongs to root cgroup
+ * whose cache mask is all 1s always.
+ */
+ for (i = 1; i < maxid; i++) {
+ if (ccmap[i].clos_refcnt)
+ cbm_cpu_update((void *)i);
+ }
+}
+
+static inline void intel_rdt_cpu_start(int cpu)
+{
+ struct intel_pqr_state *state = &per_cpu(pqr_state, cpu);
+
+ state->closid = 0;
+ mutex_lock(&rdt_group_mutex);
+ if (rdt_cpumask_update(cpu))
+ cbm_update_msrs();
+ mutex_unlock(&rdt_group_mutex);
+}
+
+static void intel_rdt_cpu_exit(unsigned int cpu)
+{
+ int i;
+
+ mutex_lock(&rdt_group_mutex);
+ if (!cpumask_test_and_clear_cpu(cpu, &rdt_cpumask)) {
+ mutex_unlock(&rdt_group_mutex);
+ return;
+ }
+
+ i = cpumask_any_online_but(topology_core_cpumask(cpu), cpu);
+ if (i < nr_cpu_ids)
+ cpumask_set_cpu(i, &rdt_cpumask);
+ mutex_unlock(&rdt_group_mutex);
+}
+
+static int intel_rdt_cpu_notifier(struct notifier_block *nb,
+ unsigned long action, void *hcpu)
+{
+ unsigned int cpu = (unsigned long)hcpu;
+
+ switch (action) {
+ case CPU_STARTING:
+ intel_rdt_cpu_start(cpu);
+ break;
+ case CPU_DOWN_PREPARE:
+ intel_rdt_cpu_exit(cpu);
+ break;
+ default:
+ break;
+ }
+
+ return NOTIFY_OK;
}
static int __init intel_rdt_late_init(void)
@@ -358,8 +430,16 @@ static int __init intel_rdt_late_init(void)
ccm->cache_mask = (1ULL << max_cbm_len) - 1;
ccm->clos_refcnt = 1;
+ cpu_notifier_register_begin();
+
+ mutex_lock(&rdt_group_mutex);
for_each_online_cpu(i)
rdt_cpumask_update(i);
+ mutex_unlock(&rdt_group_mutex);
+
+ __hotcpu_notifier(intel_rdt_cpu_notifier, 0);
+
+ cpu_notifier_register_done();
static_key_slow_inc(&rdt_enable_key);
pr_info("Intel cache allocation enabled\n");
--
1.9.1
next prev parent reply other threads:[~2015-06-12 18:21 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-06-12 18:17 [PATCH V9 00/10] New cpumask API and Intel Cache Allocation support Vikas Shivappa
2015-06-12 18:17 ` [PATCH 01/10] cpumask: Introduce cpumask_any_online_but Vikas Shivappa
2015-06-15 12:36 ` Peter Zijlstra
2015-06-15 16:48 ` Vikas Shivappa
2015-06-16 8:20 ` Thomas Gleixner
2015-06-16 19:00 ` Vikas Shivappa
2015-06-12 18:17 ` [PATCH 02/10] x86/intel_cqm: Modify hot cpu notification handling Vikas Shivappa
2015-06-15 12:38 ` Peter Zijlstra
2015-06-15 16:52 ` Vikas Shivappa
2015-06-12 18:17 ` [PATCH 03/10] x86/intel_rapl: Modify hot cpu notification handling for RAPL Vikas Shivappa
2015-06-12 18:17 ` [PATCH 04/10] x86/intel_rdt: Cache Allocation documentation and cgroup usage guide Vikas Shivappa
2015-06-12 18:17 ` [PATCH 05/10] x86/intel_rdt: Add support for Cache Allocation detection Vikas Shivappa
2015-06-15 12:48 ` Peter Zijlstra
2015-06-15 17:04 ` Vikas Shivappa
2015-06-12 18:17 ` [PATCH 06/10] x86/intel_rdt: Add new cgroup and Class of service management Vikas Shivappa
2015-06-12 18:17 ` [PATCH 07/10] x86/intel_rdt: Add support for cache bit mask management Vikas Shivappa
2015-06-12 18:17 ` [PATCH 08/10] x86/intel_rdt: Implement scheduling support for Intel RDT Vikas Shivappa
2015-06-12 18:17 ` Vikas Shivappa [this message]
2015-06-16 8:52 ` [PATCH 09/10] x86/intel_rdt: Hot cpu support for Cache Allocation Thomas Gleixner
2015-06-16 19:01 ` Vikas Shivappa
2015-06-16 19:25 ` Thomas Gleixner
2015-06-19 20:42 ` Vikas Shivappa
2015-06-12 18:17 ` [PATCH 10/10] x86/intel_rdt: Intel haswell Cache Allocation enumeration Vikas Shivappa
2015-06-15 14:05 ` Peter Zijlstra
2015-06-15 21:44 ` Vikas Shivappa
2015-06-16 8:23 ` Thomas Gleixner
2015-06-16 9:18 ` Peter Zijlstra
2015-06-17 16:23 ` Vikas Shivappa
-- strict thread matches above, loose matches on Subject: below --
2015-06-23 22:56 [PATCH V10 00/10] New cpumask API and Intel Cache Allocation support Vikas Shivappa
2015-06-23 22:56 ` [PATCH 09/10] x86/intel_rdt: Hot cpu support for Cache Allocation Vikas Shivappa
2015-06-24 8:21 ` Thomas Gleixner
2015-06-24 21:15 ` Vikas Shivappa
2015-06-05 0:01 [PATCH V8 00/10] New cpumask API and Intel Cache Allocation support Vikas Shivappa
2015-06-05 0:01 ` [PATCH 09/10] x86/intel_rdt: Hot cpu support for Cache Allocation Vikas Shivappa
2015-06-03 19:09 [PATCH V8 00/10] New cpumask API and Intel Cache Allocation support Vikas Shivappa
2015-06-03 19:10 ` [PATCH 09/10] x86/intel_rdt: Hot cpu support for Cache Allocation Vikas Shivappa
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=1434133037-25189-10-git-send-email-vikas.shivappa@linux.intel.com \
--to=vikas.shivappa@linux.intel.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rdt@eclists.intel.com \
--cc=matt.fleming@intel.com \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=vikas.shivappa@intel.com \
--cc=will.auld@intel.com \
--cc=x86@kernel.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 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).