Minor tweaks for pg_test_timing.

Increase the size of the "direct" histogram to 10K elements,
so that we can precisely track loop times up to 10 microseconds.
(Going further than that seems pretty uninteresting, even for
very old and slow machines.)

Relabel "Per loop time" as "Average loop time" for clarity.

Pre-zero the histogram arrays to make sure that they are loaded
into processor cache and any copy-on-write overhead has happened
before we enter the timing loop.  Also use unlikely() to keep
the compiler from thinking that the clock-went-backwards case
is part of the hot loop.  Neither of these hacks made a lot of
difference on my own machine, but they seem like they might help
on some platforms.

Discussion: https://postgr.es/m/[email protected]

2 files changed, 15 insertions, 5 deletions

diff --git a/doc/src/sgml/ref/pgtesttiming.sgml b/doc/src/sgml/ref/pgtesttiming.sgml
index 1fcdbf7f06e..afe6a12be4b 100644
--- a/doc/src/sgml/ref/pgtesttiming.sgml
+++ b/doc/src/sgml/ref/pgtesttiming.sgml
@@ -161,7 +161,7 @@ PostgreSQL documentation
   <para>
 <screen><![CDATA[
 Testing timing overhead for 3 seconds.
-Per loop time including overhead: 16.40 ns
+Average loop time including overhead: 16.40 ns
 Histogram of timing durations:
    <= ns   % of total  running %      count
        0       0.0000     0.0000          0
diff --git a/src/bin/pg_test_timing/pg_test_timing.c b/src/bin/pg_test_timing/pg_test_timing.c
index 64d080335eb..a5621251afc 100644
--- a/src/bin/pg_test_timing/pg_test_timing.c
+++ b/src/bin/pg_test_timing/pg_test_timing.c
@@ -20,8 +20,8 @@ static double max_rprct = 99.99;
 /* record duration in powers of 2 nanoseconds */
 static long long int histogram[32];
 
-/* record counts of first 1024 durations directly */
-#define NUM_DIRECT 1024
+/* record counts of first 10K durations directly */
+#define NUM_DIRECT 10000
 static long long int direct_histogram[NUM_DIRECT];
 
 /* separately record highest observed duration */
@@ -161,6 +161,16 @@ test_timing(unsigned int duration)
 				end_time,
 				temp;
 
+	/*
+	 * Pre-zero the statistics data structures.  They're already zero by
+	 * default, but this helps bring them into processor cache and avoid
+	 * possible timing glitches due to COW behavior.
+	 */
+	memset(direct_histogram, 0, sizeof(direct_histogram));
+	memset(histogram, 0, sizeof(histogram));
+	largest_diff = 0;
+	largest_diff_count = 0;
+
 	total_time = duration > 0 ? duration * INT64CONST(1000000000) : 0;
 
 	INSTR_TIME_SET_CURRENT(start_time);
@@ -177,7 +187,7 @@ test_timing(unsigned int duration)
 		diff = cur - prev;
 
 		/* Did time go backwards? */
-		if (diff < 0)
+		if (unlikely(diff < 0))
 		{
 			fprintf(stderr, _("Detected clock going backwards in time.\n"));
 			fprintf(stderr, _("Time warp: %d ms\n"), diff);
@@ -215,7 +225,7 @@ test_timing(unsigned int duration)
 
 	INSTR_TIME_SUBTRACT(end_time, start_time);
 
-	printf(_("Per loop time including overhead: %0.2f ns\n"),
+	printf(_("Average loop time including overhead: %0.2f ns\n"),
 		   INSTR_TIME_GET_DOUBLE(end_time) * 1e9 / loop_count);
 
 	return loop_count;