linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Marc Zyngier <marc.zyngier@arm.com>
To: Thomas Gleixner <tglx@linutronix.de>,
	John Stultz <john.stultz@linaro.org>,
	Pavel Tatashin <pasha.tatashin@soleen.com>,
	Petr Mladek <pmladek@suse.com>,
	Sergey Senozhatsky <sergey.senozhatsky@gmail.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Will Deacon <will.deacon@arm.com>,
	Catalin Marinas <catalin.marinas@arm.com>,
	Mark Rutland <mark.rutland@arm.com>
Cc: linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org
Subject: [PATCH 3/3] arm64: Allow early time stamping
Date: Mon, 22 Jul 2019 11:33:30 +0100	[thread overview]
Message-ID: <20190722103330.255312-4-marc.zyngier@arm.com> (raw)
In-Reply-To: <20190722103330.255312-1-marc.zyngier@arm.com>

In order to provide early timestamping on arm64 systems, we provide
a timestamp_clock() function that is available as early as possible.
This function simply returns the current counter value scales in
nanoseconds, and 0-based.

In order to deal with the idiosyncrasies of some broken platforms,
we condition this on the frequency being set in the CNTFRQ_EL0
register, and revert back to using local_clock() as soon as it starts
returning non-zero values.

Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
 arch/arm64/Kconfig        |  3 +++
 arch/arm64/kernel/setup.c | 44 +++++++++++++++++++++++++++++++++++++++
 2 files changed, 47 insertions(+)

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 3adcec05b1f6..ac3882ddc1c2 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -911,6 +911,9 @@ config ARCH_WANT_HUGE_PMD_SHARE
 config ARCH_HAS_CACHE_LINE_SIZE
 	def_bool y
 
+config ARCH_HAS_TIMESTAMP_CLOCK
+	def_bool y
+
 config ARCH_ENABLE_SPLIT_PMD_PTLOCK
 	def_bool y if PGTABLE_LEVELS > 2
 
diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c
index 9c4bad7d7131..cf21e3df7165 100644
--- a/arch/arm64/kernel/setup.c
+++ b/arch/arm64/kernel/setup.c
@@ -15,6 +15,7 @@
 #include <linux/initrd.h>
 #include <linux/console.h>
 #include <linux/cache.h>
+#include <linux/clocksource.h>
 #include <linux/screen_info.h>
 #include <linux/init.h>
 #include <linux/kexec.h>
@@ -28,10 +29,12 @@
 #include <linux/of_fdt.h>
 #include <linux/efi.h>
 #include <linux/psci.h>
+#include <linux/sched/clock.h>
 #include <linux/sched/task.h>
 #include <linux/mm.h>
 
 #include <asm/acpi.h>
+#include <asm/arch_timer.h>
 #include <asm/fixmap.h>
 #include <asm/cpu.h>
 #include <asm/cputype.h>
@@ -269,8 +272,49 @@ arch_initcall(reserve_memblock_reserved_regions);
 
 u64 __cpu_logical_map[NR_CPUS] = { [0 ... NR_CPUS-1] = INVALID_HWID };
 
+static u64 notrace ts_counter_read_cc(const struct cyclecounter *cc)
+{
+	return __arch_counter_get_cntvct();
+}
+
+static struct cyclecounter ts_cc __ro_after_init = {
+	.read	= ts_counter_read_cc,
+	.mask	= CLOCKSOURCE_MASK(56),
+};
+
+static struct timecounter ts_tc;
+
+static bool ts_cc_valid __ro_after_init = false;
+
+static __init void timestamp_clock_init(void)
+{
+	u64 frq = arch_timer_get_cntfrq();
+
+	if (!frq)
+		return;
+
+	clocks_calc_mult_shift(&ts_cc.mult, &ts_cc.shift,
+			       frq, NSEC_PER_SEC, 3600);
+	/* timestamp starts at 0 (local_clock is a good enough approximation) */
+	timecounter_init(&ts_tc, &ts_cc, local_clock());
+	ts_cc_valid = true;
+	pr_info("Using timestamp clock @%lluMHz\n", frq / 1000 / 1000);
+}
+
+u64 timestamp_clock(void)
+{
+	u64 ns = local_clock();
+
+	if (likely(ns || !ts_cc_valid))
+		return ns;
+
+	return timecounter_read(&ts_tc);
+}
+
 void __init setup_arch(char **cmdline_p)
 {
+	timestamp_clock_init();
+
 	init_mm.start_code = (unsigned long) _text;
 	init_mm.end_code   = (unsigned long) _etext;
 	init_mm.end_data   = (unsigned long) _edata;
-- 
2.20.1


  parent reply	other threads:[~2019-07-22 10:33 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-07-22 10:33 [PATCH 0/3] arm64: Allow early timestamping of kernel log Marc Zyngier
2019-07-22 10:33 ` [PATCH 1/3] printk: Allow architecture-specific timestamping function Marc Zyngier
2019-07-22 11:25   ` Petr Mladek
2019-07-22 12:47     ` Marc Zyngier
2019-07-22 13:03       ` Russell King - ARM Linux admin
2019-07-22 13:26         ` Marc Zyngier
2019-07-22 10:33 ` [PATCH 2/3] sched/clock: Allow sched_clock to inherit timestamp_clock epoch Marc Zyngier
2019-07-22 10:33 ` Marc Zyngier [this message]
2019-07-22 20:52 ` [PATCH 0/3] arm64: Allow early timestamping of kernel log Pavel Tatashin
2019-07-23  7:17   ` Marc Zyngier
2019-09-23 19:13     ` Pavel Tatashin
2019-07-23  2:42 ` Hanjun Guo

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=20190722103330.255312-4-marc.zyngier@arm.com \
    --to=marc.zyngier@arm.com \
    --cc=catalin.marinas@arm.com \
    --cc=john.stultz@linaro.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=pasha.tatashin@soleen.com \
    --cc=pmladek@suse.com \
    --cc=rostedt@goodmis.org \
    --cc=sergey.senozhatsky@gmail.com \
    --cc=tglx@linutronix.de \
    --cc=will.deacon@arm.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).