[linux-yocto] [PATCH 2/2] perf/x86: Fix time_shift in perf_event_mmap_page

[jonathan.yong at intel.com]

From: Adrian Hunter <adrian.hunter at intel.com>

Commit:

  b20112edeadf ("perf/x86: Improve accuracy of perf/sched clock")

allowed the time_shift value in perf_event_mmap_page to be as much
as 32.  Unfortunately the documented algorithms for using time_shift
have it shifting an integer, whereas to work correctly with the value
32, the type must be u64.

In the case of perf tools, Intel PT decodes correctly but the timestamps
that are output (for example by perf script) have lost 32-bits of
granularity so they look like they are not changing at all.

Fix by limiting the shift to 31 and adjusting the multiplier accordingly.

Also update the documentation of perf_event_mmap_page so that new code
based on it will be more future-proof.

Signed-off-by: Adrian Hunter <adrian.hunter at intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz at infradead.org>
Cc: Andy Lutomirski <luto at amacapital.net>
Cc: Arnaldo Carvalho de Melo <acme at kernel.org>
Cc: Arnaldo Carvalho de Melo <acme at redhat.com>
Cc: David Ahern <dsahern at gmail.com>
Cc: Jiri Olsa <jolsa at redhat.com>
Cc: Linus Torvalds <torvalds at linux-foundation.org>
Cc: Namhyung Kim <namhyung at kernel.org>
Cc: Peter Zijlstra <peterz at infradead.org>
Cc: Stephane Eranian <eranian at google.com>
Cc: Thomas Gleixner <tglx at linutronix.de>
Cc: Vince Weaver <vincent.weaver at maine.edu>
Fixes: b20112edeadf ("perf/x86: Improve accuracy of perf/sched clock")
Link: http://lkml.kernel.org/r/1445001845-13688-2-git-send-email-adrian.hunter@intel.com
Signed-off-by: Ingo Molnar <mingo at kernel.org>
(cherry picked from commit b9511cd761faafca7a1acc059e792c1399f9d7c6)
Signed-off-by: Yong, Jonathan <jonathan.yong at intel.com>
---
 arch/x86/kernel/tsc.c           | 11 +++++++++++
 include/uapi/linux/perf_event.h |  4 ++--
 2 files changed, 13 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index d927c74..8aba162 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -264,6 +264,17 @@ static void set_cyc2ns_scale(unsigned long cpu_khz, int cpu)
 	clocks_calc_mult_shift(&data->cyc2ns_mul, &data->cyc2ns_shift, cpu_khz,
 			       NSEC_PER_MSEC, 0);
 
+	/*
+	 * cyc2ns_shift is exported via arch_perf_update_userpage() where it is
+	 * not expected to be greater than 31 due to the original published
+	 * conversion algorithm shifting a 32-bit value (now specifies a 64-bit
+	 * value) - refer perf_event_mmap_page documentation in perf_event.h.
+	 */
+	if (data->cyc2ns_shift == 32) {
+		data->cyc2ns_shift = 31;
+		data->cyc2ns_mul >>= 1;
+	}
+
 	data->cyc2ns_offset = ns_now -
 		mul_u64_u32_shr(tsc_now, data->cyc2ns_mul, data->cyc2ns_shift);
 
diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h
index 309211b..9107e36 100644
--- a/include/uapi/linux/perf_event.h
+++ b/include/uapi/linux/perf_event.h
@@ -473,7 +473,7 @@ struct perf_event_mmap_page {
 	 *   u64 delta;
 	 *
 	 *   quot = (cyc >> time_shift);
-	 *   rem = cyc & ((1 << time_shift) - 1);
+	 *   rem = cyc & (((u64)1 << time_shift) - 1);
 	 *   delta = time_offset + quot * time_mult +
 	 *              ((rem * time_mult) >> time_shift);
 	 *
@@ -504,7 +504,7 @@ struct perf_event_mmap_page {
 	 * And vice versa:
 	 *
 	 *   quot = cyc >> time_shift;
-	 *   rem  = cyc & ((1 << time_shift) - 1);
+	 *   rem  = cyc & (((u64)1 << time_shift) - 1);
 	 *   timestamp = time_zero + quot * time_mult +
 	 *               ((rem * time_mult) >> time_shift);
 	 */
-- 
2.7.3