From: Len Brown <lenb@kernel.org>
To: x86@kernel.org
Cc: linux-kernel@vger.kernel.org,
Kan Liang <kan.liang@linux.intel.com>,
Peter Zijlstra <peterz@infradead.org>,
Len Brown <len.brown@intel.com>
Subject: [PATCH 18/22] perf/x86/intel/cstate: Support multi-die/package
Date: Mon, 6 May 2019 17:26:13 -0400 [thread overview]
Message-ID: <ada9f26858dceb7484b51c23cc8028401379b4c8.1557177585.git.len.brown@intel.com> (raw)
In-Reply-To: <6f53f0e494d743c79e18f6e3a98085711e6ddd0c.1557177585.git.len.brown@intel.com>
From: Kan Liang <kan.liang@linux.intel.com>
Some cstate counters becomes die-scope on Xeon Cascade Lake-AP. Perf
cstate driver needs to support die-scope cstate counters.
Use topology_die_cpumask() to replace topology_core_cpumask().
For previous platforms which doesn't have multi-die,
topology_die_cpumask() is identical as topology_core_cpumask().
There is no functional change for previous platforms.
Name the die-scope PMU "cstate_die".
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Len Brown <len.brown@intel.com>
---
arch/x86/events/intel/cstate.c | 14 ++++++++++----
1 file changed, 10 insertions(+), 4 deletions(-)
diff --git a/arch/x86/events/intel/cstate.c b/arch/x86/events/intel/cstate.c
index 94a4b7fc75d0..52c5fea29457 100644
--- a/arch/x86/events/intel/cstate.c
+++ b/arch/x86/events/intel/cstate.c
@@ -302,7 +302,7 @@ static int cstate_pmu_event_init(struct perf_event *event)
return -EINVAL;
event->hw.event_base = pkg_msr[cfg].msr;
cpu = cpumask_any_and(&cstate_pkg_cpu_mask,
- topology_core_cpumask(event->cpu));
+ topology_die_cpumask(event->cpu));
} else {
return -ENOENT;
}
@@ -385,7 +385,7 @@ static int cstate_cpu_exit(unsigned int cpu)
if (has_cstate_pkg &&
cpumask_test_and_clear_cpu(cpu, &cstate_pkg_cpu_mask)) {
- target = cpumask_any_but(topology_core_cpumask(cpu), cpu);
+ target = cpumask_any_but(topology_die_cpumask(cpu), cpu);
/* Migrate events if there is a valid target */
if (target < nr_cpu_ids) {
cpumask_set_cpu(target, &cstate_pkg_cpu_mask);
@@ -414,7 +414,7 @@ static int cstate_cpu_init(unsigned int cpu)
* in the package cpu mask as the designated reader.
*/
target = cpumask_any_and(&cstate_pkg_cpu_mask,
- topology_core_cpumask(cpu));
+ topology_die_cpumask(cpu));
if (has_cstate_pkg && target >= nr_cpu_ids)
cpumask_set_cpu(cpu, &cstate_pkg_cpu_mask);
@@ -661,7 +661,13 @@ static int __init cstate_init(void)
}
if (has_cstate_pkg) {
- err = perf_pmu_register(&cstate_pkg_pmu, cstate_pkg_pmu.name, -1);
+ if (topology_max_die_per_package() > 1) {
+ err = perf_pmu_register(&cstate_pkg_pmu,
+ "cstate_die", -1);
+ } else {
+ err = perf_pmu_register(&cstate_pkg_pmu,
+ cstate_pkg_pmu.name, -1);
+ }
if (err) {
has_cstate_pkg = false;
pr_info("Failed to register cstate pkg pmu\n");
--
2.18.0-rc0
next prev parent reply other threads:[~2019-05-06 21:26 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-06 21:25 [PATCH 0/22] v5 multi-die/package topology support Len Brown
2019-05-06 21:25 ` [PATCH 01/22] x86 topology: Fix doc typo Len Brown
2019-05-06 21:25 ` [PATCH 02/22] topology: Simplify cputopology.txt formatting and wording Len Brown
2019-05-06 21:25 ` [PATCH 03/22] x86 smpboot: Rename match_die() to match_pkg() Len Brown
2019-05-06 21:25 ` [PATCH 04/22] x86 topology: Add CPUID.1F multi-die/package support Len Brown
2019-05-06 21:26 ` [PATCH 05/22] x86 topology: Create topology_max_die_per_package() Len Brown
2019-05-06 21:26 ` [PATCH 06/22] cpu topology: Export die_id Len Brown
2019-05-06 21:26 ` [PATCH 07/22] x86 topology: Define topology_die_id() Len Brown
2019-05-06 21:26 ` [PATCH 08/22] x86 topology: Define topology_logical_die_id() Len Brown
2019-05-06 21:26 ` [PATCH 09/22] powercap/intel_rapl: Simplify rapl_find_package() Len Brown
2019-05-06 21:26 ` [PATCH 10/22] powercap/intel_rapl: Support multi-die/package Len Brown
2019-05-07 12:14 ` Peter Zijlstra
2019-05-08 19:45 ` Len Brown
2019-05-06 21:26 ` [PATCH 11/22] thermal/x86_pkg_temp_thermal: " Len Brown
2019-05-06 21:26 ` [PATCH 12/22] powercap/intel_rapl: update rapl domain name and debug messages Len Brown
2019-05-06 21:26 ` [PATCH 13/22] hwmon/coretemp: Support multi-die/package Len Brown
2019-05-06 21:26 ` [PATCH 14/22] topology: Create package_cpus sysfs attribute Len Brown
2019-05-06 21:26 ` [PATCH 15/22] topology: Create core_cpus and die_cpus sysfs attributes Len Brown
2019-05-06 21:26 ` [PATCH 16/22] perf/x86/intel/uncore: Support multi-die/package Len Brown
2019-05-07 12:21 ` Peter Zijlstra
2019-05-08 19:51 ` Len Brown
2019-05-07 12:22 ` Peter Zijlstra
2019-05-08 19:51 ` Len Brown
2019-05-06 21:26 ` [PATCH 17/22] perf/x86/intel/rapl: " Len Brown
2019-05-06 21:26 ` Len Brown [this message]
2019-05-06 21:26 ` [PATCH 19/22] thermal/x86_pkg_temp_thermal: rename internal variables to zones from packages Len Brown
2019-05-06 21:26 ` [PATCH 20/22] hwmon/coretemp: " Len Brown
2019-05-06 21:26 ` [PATCH 21/22] perf/x86/intel/uncore: renames in response to multi-die/pkg support Len Brown
2019-05-09 15:02 ` Liang, Kan
2019-05-13 17:30 ` Len Brown
2019-05-06 21:26 ` [PATCH 22/22] perf/x86/intel/rapl: rename internal variables " Len Brown
2019-05-09 15:04 ` Liang, Kan
2019-05-13 17:29 ` Len Brown
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=ada9f26858dceb7484b51c23cc8028401379b4c8.1557177585.git.len.brown@intel.com \
--to=lenb@kernel.org \
--cc=kan.liang@linux.intel.com \
--cc=len.brown@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=peterz@infradead.org \
--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).