All of lore.kernel.org
 help / color / mirror / Atom feed
From: Leo Yan <leo.yan@linaro.org>
To: Peter Zijlstra <peterz@infradead.org>,
	Will Deacon <will@kernel.org>, Marc Zyngier <maz@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Ingo Molnar <mingo@redhat.com>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Jiri Olsa <jolsa@redhat.com>, Namhyung Kim <namhyung@kernel.org>,
	Catalin Marinas <catalin.marinas@arm.com>,
	Daniel Lezcano <daniel.lezcano@linaro.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Allison Randal <allison@lohutok.net>,
	Alexios Zavras <alexios.zavras@intel.com>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Kate Stewart <kstewart@linuxfoundation.org>,
	Enrico Weigelt <info@metux.net>,
	"Ahmed S. Darwish" <a.darwish@linutronix.de>,
	Paul Cercueil <paul@crapouillou.net>,
	"Ben Dooks (Codethink)" <ben.dooks@codethink.co.uk>,
	linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org
Cc: Leo Yan <leo.yan@linaro.org>
Subject: [PATCH v2 1/3] time/sched_clock: Add new variant sched_clock_register_epoch()
Date: Tue,  5 May 2020 21:55:42 +0800	[thread overview]
Message-ID: <20200505135544.6003-2-leo.yan@linaro.org> (raw)
In-Reply-To: <20200505135544.6003-1-leo.yan@linaro.org>

Except the sched clock's raw counter is used by sched clock itself, it
also can be used by other purposes in the same system, e.g. the raw
counter can be injected into hardware tracing data (like Arm's SPE) and
Perf tool can capture trace data and extract the raw counter from it
which finally can be used to generate timestamp.

Perf tool needs a way to convert sched clock's raw counter cycles into a
nanosecond that can be compared against values coming out of sched_clock.

To do this accurately, this patch adds a new variant API
sched_clock_register_epoch() with introducing an extra argument
'epoch_offset', as its naming indicates, this argument contains the
offset time (in nanosecond) for the clock source has been enabled prior
to epoch.

Signed-off-by: Leo Yan <leo.yan@linaro.org>
---
 include/linux/sched_clock.h | 10 ++++++++++
 kernel/time/sched_clock.c   | 13 ++++++++++++-
 2 files changed, 22 insertions(+), 1 deletion(-)

diff --git a/include/linux/sched_clock.h b/include/linux/sched_clock.h
index 0bb04a96a6d4..98965c0c7cd4 100644
--- a/include/linux/sched_clock.h
+++ b/include/linux/sched_clock.h
@@ -10,6 +10,10 @@ extern void generic_sched_clock_init(void);
 
 extern void sched_clock_register(u64 (*read)(void), int bits,
 				 unsigned long rate);
+
+extern void sched_clock_register_epoch(u64 (*read)(void), int bits,
+				       unsigned long rate,
+				       u64 *epoch_offset);
 #else
 static inline void generic_sched_clock_init(void) { }
 
@@ -17,6 +21,12 @@ static inline void sched_clock_register(u64 (*read)(void), int bits,
 					unsigned long rate)
 {
 }
+
+static inline void sched_clock_register_epoch(u64 (*read)(void), int bits,
+					      unsigned long rate,
+					      u64 *epoch_offset)
+{
+}
 #endif
 
 #endif
diff --git a/kernel/time/sched_clock.c b/kernel/time/sched_clock.c
index fa3f800d7d76..b402196afc3f 100644
--- a/kernel/time/sched_clock.c
+++ b/kernel/time/sched_clock.c
@@ -165,7 +165,8 @@ static enum hrtimer_restart sched_clock_poll(struct hrtimer *hrt)
 }
 
 void __init
-sched_clock_register(u64 (*read)(void), int bits, unsigned long rate)
+sched_clock_register_epoch(u64 (*read)(void), int bits, unsigned long rate,
+			   u64 *epoch_offset)
 {
 	u64 res, wrap, new_mask, new_epoch, cyc, ns;
 	u32 new_mult, new_shift;
@@ -204,6 +205,10 @@ sched_clock_register(u64 (*read)(void), int bits, unsigned long rate)
 	rd.epoch_cyc		= new_epoch;
 	rd.epoch_ns		= ns;
 
+	/* Output epoch offset (ns) to clock event driver */
+	if (epoch_offset)
+		*epoch_offset = cyc_to_ns(new_epoch & new_mask, new_mult, new_shift) - ns;
+
 	update_clock_read_data(&rd);
 
 	if (sched_clock_timer.function != NULL) {
@@ -240,6 +245,12 @@ sched_clock_register(u64 (*read)(void), int bits, unsigned long rate)
 	pr_debug("Registered %pS as sched_clock source\n", read);
 }
 
+void __init
+sched_clock_register(u64 (*read)(void), int bits, unsigned long rate)
+{
+	sched_clock_register_epoch(read, bits, rate, NULL);
+}
+
 void __init generic_sched_clock_init(void)
 {
 	/*
-- 
2.17.1


WARNING: multiple messages have this Message-ID (diff)
From: Leo Yan <leo.yan@linaro.org>
To: Peter Zijlstra <peterz@infradead.org>,
	Will Deacon <will@kernel.org>, Marc Zyngier <maz@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Ingo Molnar <mingo@redhat.com>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Jiri Olsa <jolsa@redhat.com>, Namhyung Kim <namhyung@kernel.org>,
	Catalin Marinas <catalin.marinas@arm.com>,
	Daniel Lezcano <daniel.lezcano@linaro.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Allison Randal <allison@lohutok.net>,
	Alexios Zavras <alexios.zavras@intel.com>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Kate Stewart <kstewart@linuxfoundation.org>,
	Enrico Weigelt <info@metux.net>,
	"Ahmed S. Darwish" <a.darwish@linutronix.de>,
	Paul Cercueil <paul@crapouillou.net>,
	"Ben Dooks (Codethink)" <ben.dooks@codethink.co.uk>,
	linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org
Cc: Leo Yan <leo.yan@linaro.org>
Subject: [PATCH v2 1/3] time/sched_clock: Add new variant sched_clock_register_epoch()
Date: Tue,  5 May 2020 21:55:42 +0800	[thread overview]
Message-ID: <20200505135544.6003-2-leo.yan@linaro.org> (raw)
In-Reply-To: <20200505135544.6003-1-leo.yan@linaro.org>

Except the sched clock's raw counter is used by sched clock itself, it
also can be used by other purposes in the same system, e.g. the raw
counter can be injected into hardware tracing data (like Arm's SPE) and
Perf tool can capture trace data and extract the raw counter from it
which finally can be used to generate timestamp.

Perf tool needs a way to convert sched clock's raw counter cycles into a
nanosecond that can be compared against values coming out of sched_clock.

To do this accurately, this patch adds a new variant API
sched_clock_register_epoch() with introducing an extra argument
'epoch_offset', as its naming indicates, this argument contains the
offset time (in nanosecond) for the clock source has been enabled prior
to epoch.

Signed-off-by: Leo Yan <leo.yan@linaro.org>
---
 include/linux/sched_clock.h | 10 ++++++++++
 kernel/time/sched_clock.c   | 13 ++++++++++++-
 2 files changed, 22 insertions(+), 1 deletion(-)

diff --git a/include/linux/sched_clock.h b/include/linux/sched_clock.h
index 0bb04a96a6d4..98965c0c7cd4 100644
--- a/include/linux/sched_clock.h
+++ b/include/linux/sched_clock.h
@@ -10,6 +10,10 @@ extern void generic_sched_clock_init(void);
 
 extern void sched_clock_register(u64 (*read)(void), int bits,
 				 unsigned long rate);
+
+extern void sched_clock_register_epoch(u64 (*read)(void), int bits,
+				       unsigned long rate,
+				       u64 *epoch_offset);
 #else
 static inline void generic_sched_clock_init(void) { }
 
@@ -17,6 +21,12 @@ static inline void sched_clock_register(u64 (*read)(void), int bits,
 					unsigned long rate)
 {
 }
+
+static inline void sched_clock_register_epoch(u64 (*read)(void), int bits,
+					      unsigned long rate,
+					      u64 *epoch_offset)
+{
+}
 #endif
 
 #endif
diff --git a/kernel/time/sched_clock.c b/kernel/time/sched_clock.c
index fa3f800d7d76..b402196afc3f 100644
--- a/kernel/time/sched_clock.c
+++ b/kernel/time/sched_clock.c
@@ -165,7 +165,8 @@ static enum hrtimer_restart sched_clock_poll(struct hrtimer *hrt)
 }
 
 void __init
-sched_clock_register(u64 (*read)(void), int bits, unsigned long rate)
+sched_clock_register_epoch(u64 (*read)(void), int bits, unsigned long rate,
+			   u64 *epoch_offset)
 {
 	u64 res, wrap, new_mask, new_epoch, cyc, ns;
 	u32 new_mult, new_shift;
@@ -204,6 +205,10 @@ sched_clock_register(u64 (*read)(void), int bits, unsigned long rate)
 	rd.epoch_cyc		= new_epoch;
 	rd.epoch_ns		= ns;
 
+	/* Output epoch offset (ns) to clock event driver */
+	if (epoch_offset)
+		*epoch_offset = cyc_to_ns(new_epoch & new_mask, new_mult, new_shift) - ns;
+
 	update_clock_read_data(&rd);
 
 	if (sched_clock_timer.function != NULL) {
@@ -240,6 +245,12 @@ sched_clock_register(u64 (*read)(void), int bits, unsigned long rate)
 	pr_debug("Registered %pS as sched_clock source\n", read);
 }
 
+void __init
+sched_clock_register(u64 (*read)(void), int bits, unsigned long rate)
+{
+	sched_clock_register_epoch(read, bits, rate, NULL);
+}
+
 void __init generic_sched_clock_init(void)
 {
 	/*
-- 
2.17.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

  reply	other threads:[~2020-05-05 13:56 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-05 13:55 [PATCH v2 0/3] arm64: perf_event: Fix time offset prior to epoch Leo Yan
2020-05-05 13:55 ` Leo Yan
2020-05-05 13:55 ` Leo Yan [this message]
2020-05-05 13:55   ` [PATCH v2 1/3] time/sched_clock: Add new variant sched_clock_register_epoch() Leo Yan
2020-05-05 13:55 ` [PATCH v2 2/3] clocksource/drivers/arm_arch_timer: Handle time offset prior to epoch Leo Yan
2020-05-05 13:55   ` Leo Yan
2020-05-05 13:55 ` [PATCH v2 3/3] arm64: perf_event: Fix time_offset for arch timer Leo Yan
2020-05-05 13:55   ` Leo Yan
2020-05-11  9:22 ` [PATCH v2 0/3] arm64: perf_event: Fix time offset prior to epoch Peter Zijlstra
2020-05-11  9:22   ` Peter Zijlstra
2020-05-11  9:25   ` Peter Zijlstra
2020-05-11  9:25     ` Peter Zijlstra
2020-05-12  6:38     ` Leo Yan
2020-05-12  6:38       ` Leo Yan
2020-05-12  8:57       ` Peter Zijlstra
2020-05-12  8:57         ` Peter Zijlstra
2020-05-12  9:19       ` Peter Zijlstra
2020-05-12  9:19         ` Peter Zijlstra
2020-05-12 10:01         ` Mark Rutland
2020-05-12 10:01           ` Mark Rutland
2020-05-12 11:21         ` Peter Zijlstra
2020-05-12 11:21           ` Peter Zijlstra

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=20200505135544.6003-2-leo.yan@linaro.org \
    --to=leo.yan@linaro.org \
    --cc=a.darwish@linutronix.de \
    --cc=acme@kernel.org \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=alexios.zavras@intel.com \
    --cc=allison@lohutok.net \
    --cc=ben.dooks@codethink.co.uk \
    --cc=catalin.marinas@arm.com \
    --cc=daniel.lezcano@linaro.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=info@metux.net \
    --cc=jolsa@redhat.com \
    --cc=kstewart@linuxfoundation.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=maz@kernel.org \
    --cc=mingo@redhat.com \
    --cc=namhyung@kernel.org \
    --cc=paul@crapouillou.net \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --cc=will@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 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.