X-Git-Url: https://git.kernel.dk/?p=fio.git;a=blobdiff_plain;f=stat.c;h=40ca1d3881e4cf2d0f1552549c509d8361118b84;hp=71eca48e9bf23a731bb9956e2af6cca0b3cf684e;hb=a269790cd25788dd4226641a3ceab1b3c8fda14b;hpb=5b9babb7fb9ac46c0e960ccd88c2d85ba3065c01

diff --git a/stat.c b/stat.c
index 71eca48e..40ca1d38 100644
--- a/stat.c
+++ b/stat.c
@@ -9,7 +9,7 @@
 
 #include "fio.h"
 #include "diskutil.h"
-#include "ieee754.h"
+#include "lib/ieee754.h"
 
 void update_rusage_stat(struct thread_data *td)
 {
@@ -114,20 +114,26 @@ static int double_cmp(const void *a, const void *b)
 	return cmp;
 }
 
-/*
- * Find and display the p-th percentile of clat
- */
-static void show_clat_percentiles(unsigned int *io_u_plat, unsigned long nr,
-				  fio_fp64_t *plist)
+unsigned int calc_clat_percentiles(unsigned int *io_u_plat, unsigned long nr,
+				   fio_fp64_t *plist, unsigned int **output,
+				   unsigned int *maxv, unsigned int *minv)
 {
 	unsigned long sum = 0;
 	unsigned int len, i, j = 0;
-	int is_last = 0;
+	unsigned int oval_len = 0;
+	unsigned int *ovals = NULL;
+	int is_last;
+
+	*minv = -1U;
+	*maxv = 0;
 
 	len = 0;
 	while (len < FIO_IO_U_LIST_MAX_LEN && plist[len].u.f != 0.0)
 		len++;
 
+	if (!len)
+		return 0;
+
 	/*
 	 * Sort the percentile list. Note that it may already be sorted if
 	 * we are using the default values, but since it's a short list this
@@ -136,35 +142,98 @@ static void show_clat_percentiles(unsigned int *io_u_plat, unsigned long nr,
 	if (len > 1)
 		qsort((void*)plist, len, sizeof(plist[0]), double_cmp);
 
-	log_info("    clat percentiles (usec) :");
-
+	/*
+	 * Calculate bucket values, note down max and min values
+	 */
+	is_last = 0;
 	for (i = 0; i < FIO_IO_U_PLAT_NR && !is_last; i++) {
 		sum += io_u_plat[i];
 		while (sum >= (plist[j].u.f / 100.0 * nr)) {
 			assert(plist[j].u.f <= 100.0);
 
-			/* for formatting */
-			if (j != 0 && (j % 4) == 0)
-				log_info("                             ");
-
-			/* end of the list */
-			is_last = (j == len - 1);
+			if (j == oval_len) {
+				oval_len += 100;
+				ovals = realloc(ovals, oval_len * sizeof(unsigned int));
+			}
 
-			log_info(" %2.2fth=%u%c", plist[j], plat_idx_to_val(i),
-				 (is_last? '\n' : ','));
+			ovals[j] = plat_idx_to_val(i);
+			if (ovals[j] < *minv)
+				*minv = ovals[j];
+			if (ovals[j] > *maxv)
+				*maxv = ovals[j];
 
+			is_last = (j == len - 1);
 			if (is_last)
 				break;
 
-			if (j % 4 == 3)	/* for formatting */
-				log_info("\n");
 			j++;
 		}
 	}
+
+	*output = ovals;
+	return len;
+}
+
+/*
+ * Find and display the p-th percentile of clat
+ */
+static void show_clat_percentiles(unsigned int *io_u_plat, unsigned long nr,
+				  fio_fp64_t *plist)
+{
+	unsigned int len, j = 0, minv, maxv;
+	unsigned int *ovals;
+	int is_last, scale_down;
+
+	len = calc_clat_percentiles(io_u_plat, nr, plist, &ovals, &maxv, &minv);
+	if (!len)
+		goto out;
+
+	/*
+	 * We default to usecs, but if the value range is such that we
+	 * should scale down to msecs, do that.
+	 */
+	if (minv > 2000 && maxv > 99999) {
+		scale_down = 1;
+		log_info("    clat percentiles (msec):\n     |");
+	} else {
+		scale_down = 0;
+		log_info("    clat percentiles (usec):\n     |");
+	}
+
+	for (j = 0; j < len; j++) {
+		char fbuf[8];
+
+		/* for formatting */
+		if (j != 0 && (j % 4) == 0)
+			log_info("     |");
+
+		/* end of the list */
+		is_last = (j == len - 1);
+
+		if (plist[j].u.f < 10.0)
+			sprintf(fbuf, " %2.2f", plist[j].u.f);
+		else
+			sprintf(fbuf, "%2.2f", plist[j].u.f);
+
+		if (scale_down)
+			ovals[j] = (ovals[j] + 999) / 1000;
+
+		log_info(" %sth=[%5u]%c", fbuf, ovals[j], is_last ? '\n' : ',');
+
+		if (is_last)
+			break;
+
+		if (j % 4 == 3)	/* for formatting */
+			log_info("\n");
+	}
+
+out:
+	if (ovals)
+		free(ovals);
 }
 
-static int calc_lat(struct io_stat *is, unsigned long *min, unsigned long *max,
-		    double *mean, double *dev)
+int calc_lat(struct io_stat *is, unsigned long *min, unsigned long *max,
+	     double *mean, double *dev)
 {
 	double n = is->samples;
 
@@ -268,18 +337,23 @@ static void stat_calc_lat_m(struct thread_stat *ts, double *io_u_lat)
 	stat_calc_lat(ts, io_u_lat, ts->io_u_lat_m, FIO_IO_U_LAT_M_NR);
 }
 
-static int usec_to_msec(unsigned long *min, unsigned long *max, double *mean,
-			double *dev)
+static void display_lat(const char *name, unsigned long min, unsigned long max,
+			double mean, double dev)
 {
-	if (*min > 1000 && *max > 1000 && *mean > 1000.0 && *dev > 1000.0) {
-		*min /= 1000;
-		*max /= 1000;
-		*mean /= 1000.0;
-		*dev /= 1000.0;
-		return 0;
-	}
+	const char *base = "(usec)";
+	char *minp, *maxp;
 
-	return 1;
+	if (!usec_to_msec(&min, &max, &mean, &dev))
+		base = "(msec)";
+
+	minp = num2str(min, 6, 1, 0);
+	maxp = num2str(max, 6, 1, 0);
+
+	log_info("    %s %s: min=%s, max=%s, avg=%5.02f,"
+		 " stdev=%5.02f\n", name, base, minp, maxp, mean, dev);
+
+	free(minp);
+	free(maxp);
 }
 
 static void show_ddir_status(struct group_run_stats *rs, struct thread_stat *ts,
@@ -315,81 +389,55 @@ static void show_ddir_status(struct group_run_stats *rs, struct thread_stat *ts,
 	free(bw_p);
 	free(iops_p);
 
-	if (calc_lat(&ts->slat_stat[ddir], &min, &max, &mean, &dev)) {
-		const char *base = "(usec)";
-		char *minp, *maxp;
-
-		if (!usec_to_msec(&min, &max, &mean, &dev))
-			base = "(msec)";
-
-		minp = num2str(min, 6, 1, 0);
-		maxp = num2str(max, 6, 1, 0);
-
-		log_info("    slat %s: min=%s, max=%s, avg=%5.02f,"
-			 " stdev=%5.02f\n", base, minp, maxp, mean, dev);
-
-		free(minp);
-		free(maxp);
-	}
-	if (calc_lat(&ts->clat_stat[ddir], &min, &max, &mean, &dev)) {
-		const char *base = "(usec)";
-		char *minp, *maxp;
-
-		if (!usec_to_msec(&min, &max, &mean, &dev))
-			base = "(msec)";
-
-		minp = num2str(min, 6, 1, 0);
-		maxp = num2str(max, 6, 1, 0);
-
-		log_info("    clat %s: min=%s, max=%s, avg=%5.02f,"
-			 " stdev=%5.02f\n", base, minp, maxp, mean, dev);
-
-		free(minp);
-		free(maxp);
-	}
-	if (calc_lat(&ts->lat_stat[ddir], &min, &max, &mean, &dev)) {
-		const char *base = "(usec)";
-		char *minp, *maxp;
-
-		if (!usec_to_msec(&min, &max, &mean, &dev))
-			base = "(msec)";
-
-		minp = num2str(min, 6, 1, 0);
-		maxp = num2str(max, 6, 1, 0);
-
-		log_info("     lat %s: min=%s, max=%s, avg=%5.02f,"
-			 " stdev=%5.02f\n", base, minp, maxp, mean, dev);
+	if (calc_lat(&ts->slat_stat[ddir], &min, &max, &mean, &dev))
+		display_lat("slat", min, max, mean, dev);
+	if (calc_lat(&ts->clat_stat[ddir], &min, &max, &mean, &dev))
+		display_lat("clat", min, max, mean, dev);
+	if (calc_lat(&ts->lat_stat[ddir], &min, &max, &mean, &dev))
+		display_lat(" lat", min, max, mean, dev);
 
-		free(minp);
-		free(maxp);
-	}
 	if (ts->clat_percentiles) {
 		show_clat_percentiles(ts->io_u_plat[ddir],
 					ts->clat_stat[ddir].samples,
 					ts->percentile_list);
 	}
 	if (calc_lat(&ts->bw_stat[ddir], &min, &max, &mean, &dev)) {
-		double p_of_agg;
+		double p_of_agg = 100.0;
+		const char *bw_str = "KB";
 
-		p_of_agg = mean * 100 / (double) rs->agg[ddir];
-		log_info("     bw (KB/s) : min=%5lu, max=%5lu, per=%3.2f%%,"
-			 " avg=%5.02f, stdev=%5.02f\n", min, max, p_of_agg,
-							mean, dev);
+		if (rs->agg[ddir]) {
+			p_of_agg = mean * 100 / (double) rs->agg[ddir];
+			if (p_of_agg > 100.0)
+				p_of_agg = 100.0;
+		}
+
+		if (mean > 999999.9) {
+			min /= 1000.0;
+			max /= 1000.0;
+			mean /= 1000.0;
+			dev /= 1000.0;
+			bw_str = "MB";
+		}
+
+		log_info("    bw (%s/s)  : min=%5lu, max=%5lu, per=%3.2f%%,"
+			 " avg=%5.02f, stdev=%5.02f\n", bw_str, min, max,
+							p_of_agg, mean, dev);
 	}
 }
 
-static void show_lat(double *io_u_lat, int nr, const char **ranges,
-		     const char *msg)
+static int show_lat(double *io_u_lat, int nr, const char **ranges,
+		    const char *msg)
 {
-	int new_line = 1, i, line = 0;
+	int new_line = 1, i, line = 0, shown = 0;
 
 	for (i = 0; i < nr; i++) {
 		if (io_u_lat[i] <= 0.0)
 			continue;
+		shown = 1;
 		if (new_line) {
 			if (line)
 				log_info("\n");
-			log_info("     lat (%s): ", msg);
+			log_info("    lat (%s) : ", msg);
 			new_line = 0;
 			line = 0;
 		}
@@ -400,6 +448,11 @@ static void show_lat(double *io_u_lat, int nr, const char **ranges,
 		if (line == 5)
 			new_line = 1;
 	}
+
+	if (shown)
+		log_info("\n");
+
+	return shown;
 }
 
 static void show_lat_u(double *io_u_lat_u)
@@ -419,12 +472,14 @@ static void show_lat_m(double *io_u_lat_m)
 	show_lat(io_u_lat_m, FIO_IO_U_LAT_M_NR, ranges, "msec");
 }
 
-static void show_latencies(double *io_u_lat_u, double *io_u_lat_m)
+static void show_latencies(struct thread_stat *ts, double *io_u_lat_u,
+			   double *io_u_lat_m)
 {
+	stat_calc_lat_u(ts, io_u_lat_u);
+	stat_calc_lat_m(ts, io_u_lat_m);
+
 	show_lat_u(io_u_lat_u);
-	log_info("\n");
 	show_lat_m(io_u_lat_m);
-	log_info("\n");
 }
 
 void show_thread_status(struct thread_stat *ts, struct group_run_stats *rs)
@@ -449,7 +504,7 @@ void show_thread_status(struct thread_stat *ts, struct group_run_stats *rs)
 					ts->error, ts->verror, (int) ts->pid);
 	}
 
-	if (ts->description)
+	if (strlen(ts->description))
 		log_info("  Description  : [%s]\n", ts->description);
 
 	if (ts->io_bytes[DDIR_READ])
@@ -457,6 +512,8 @@ void show_thread_status(struct thread_stat *ts, struct group_run_stats *rs)
 	if (ts->io_bytes[DDIR_WRITE])
 		show_ddir_status(rs, ts, DDIR_WRITE);
 
+	show_latencies(ts, io_u_lat_u, io_u_lat_m);
+
 	runtime = ts->total_run_time;
 	if (runtime) {
 		double runt = (double) runtime;
@@ -490,14 +547,12 @@ void show_thread_status(struct thread_stat *ts, struct group_run_stats *rs)
 					io_u_dist[1], io_u_dist[2],
 					io_u_dist[3], io_u_dist[4],
 					io_u_dist[5], io_u_dist[6]);
-	log_info("     issued r/w/d: total=%lu/%lu/%lu, short=%lu/%lu/%lu\n",
+	log_info("     issued    : total=r=%lu/w=%lu/d=%lu,"
+				 " short=r=%lu/w=%lu/d=%lu\n",
 					ts->total_io_u[0], ts->total_io_u[1],
 					ts->total_io_u[2],
 					ts->short_io_u[0], ts->short_io_u[1],
 					ts->short_io_u[2]);
-	stat_calc_lat_u(ts, io_u_lat_u);
-	stat_calc_lat_m(ts, io_u_lat_m);
-	show_latencies(io_u_lat_u, io_u_lat_m);
 	if (ts->continue_on_error) {
 		log_info("     errors    : total=%lu, first_error=%d/<%s>\n",
 					ts->total_err_count,
@@ -510,16 +565,23 @@ static void show_ddir_status_terse(struct thread_stat *ts,
 				   struct group_run_stats *rs, int ddir)
 {
 	unsigned long min, max;
-	unsigned long long bw;
+	unsigned long long bw, iops;
+	unsigned int *ovals = NULL;
 	double mean, dev;
+	unsigned int len, minv, maxv;
+	int i;
 
 	assert(ddir_rw(ddir));
 
-	bw = 0;
-	if (ts->runtime[ddir])
-		bw = ts->io_bytes[ddir] / ts->runtime[ddir];
+	iops = bw = 0;
+	if (ts->runtime[ddir]) {
+		uint64_t runt = ts->runtime[ddir];
 
-	log_info(";%llu;%llu;%llu", ts->io_bytes[ddir] >> 10, bw,
+		bw = ts->io_bytes[ddir] / runt;
+		iops = (1000 * (uint64_t) ts->total_io_u[ddir]) / runt;
+	}
+
+	log_info(";%llu;%llu;%llu;%llu", ts->io_bytes[ddir] >> 10, bw, iops,
 							ts->runtime[ddir]);
 
 	if (calc_lat(&ts->slat_stat[ddir], &min, &max, &mean, &dev))
@@ -532,24 +594,46 @@ static void show_ddir_status_terse(struct thread_stat *ts,
 	else
 		log_info(";%lu;%lu;%f;%f", 0UL, 0UL, 0.0, 0.0);
 
+	if (ts->clat_percentiles) {
+		len = calc_clat_percentiles(ts->io_u_plat[ddir],
+					ts->clat_stat[ddir].samples,
+					ts->percentile_list, &ovals, &maxv,
+					&minv);
+	} else
+		len = 0;
+
+	for (i = 0; i < FIO_IO_U_LIST_MAX_LEN; i++) {
+		if (i >= len) {
+			log_info(";0%%=0");
+			continue;
+		}
+		log_info(";%2.2f%%=%u", ts->percentile_list[i].u.f, ovals[i]);
+	}
+
 	if (calc_lat(&ts->lat_stat[ddir], &min, &max, &mean, &dev))
 		log_info(";%lu;%lu;%f;%f", min, max, mean, dev);
 	else
 		log_info(";%lu;%lu;%f;%f", 0UL, 0UL, 0.0, 0.0);
 
+	if (ovals)
+		free(ovals);
+
 	if (calc_lat(&ts->bw_stat[ddir], &min, &max, &mean, &dev)) {
-		double p_of_agg;
+		double p_of_agg = 100.0;
+
+		if (rs->agg[ddir]) {
+			p_of_agg = mean * 100 / (double) rs->agg[ddir];
+			if (p_of_agg > 100.0)
+				p_of_agg = 100.0;
+		}
 
-		p_of_agg = mean * 100 / (double) rs->agg[ddir];
 		log_info(";%lu;%lu;%f%%;%f;%f", min, max, p_of_agg, mean, dev);
 	} else
 		log_info(";%lu;%lu;%f%%;%f;%f", 0UL, 0UL, 0.0, 0.0, 0.0);
 }
 
-#define FIO_TERSE_VERSION	"2"
-
-static void show_thread_status_terse(struct thread_stat *ts,
-				     struct group_run_stats *rs)
+static void show_thread_status_terse_v2(struct thread_stat *ts,
+				        struct group_run_stats *rs)
 {
 	double io_u_dist[FIO_IO_U_MAP_NR];
 	double io_u_lat_u[FIO_IO_U_LAT_U_NR];
@@ -558,8 +642,7 @@ static void show_thread_status_terse(struct thread_stat *ts,
 	int i;
 
 	/* General Info */
-	log_info("%s;%s;%d;%d", FIO_TERSE_VERSION, ts->name, ts->groupid,
-				ts->error);
+	log_info("2;%s;%d;%d", ts->name, ts->groupid, ts->error);
 	/* Log Read Status */
 	show_ddir_status_terse(ts, rs, 0);
 	/* Log Write Status */
@@ -607,6 +690,80 @@ static void show_thread_status_terse(struct thread_stat *ts,
 	log_info("\n");
 }
 
+#define FIO_TERSE_VERSION	"3"
+
+static void show_thread_status_terse_v3(struct thread_stat *ts,
+					struct group_run_stats *rs)
+{
+	double io_u_dist[FIO_IO_U_MAP_NR];
+	double io_u_lat_u[FIO_IO_U_LAT_U_NR];
+	double io_u_lat_m[FIO_IO_U_LAT_M_NR];
+	double usr_cpu, sys_cpu;
+	int i;
+
+	/* General Info */
+	log_info("%s;%s;%s;%d;%d", FIO_TERSE_VERSION, fio_version_string,
+					ts->name, ts->groupid, ts->error);
+	/* Log Read Status */
+	show_ddir_status_terse(ts, rs, 0);
+	/* Log Write Status */
+	show_ddir_status_terse(ts, rs, 1);
+
+	/* CPU Usage */
+	if (ts->total_run_time) {
+		double runt = (double) ts->total_run_time;
+
+		usr_cpu = (double) ts->usr_time * 100 / runt;
+		sys_cpu = (double) ts->sys_time * 100 / runt;
+	} else {
+		usr_cpu = 0;
+		sys_cpu = 0;
+	}
+
+	log_info(";%f%%;%f%%;%lu;%lu;%lu", usr_cpu, sys_cpu, ts->ctx, ts->majf,
+								ts->minf);
+
+	/* Calc % distribution of IO depths, usecond, msecond latency */
+	stat_calc_dist(ts->io_u_map, ts_total_io_u(ts), io_u_dist);
+	stat_calc_lat_u(ts, io_u_lat_u);
+	stat_calc_lat_m(ts, io_u_lat_m);
+
+	/* Only show fixed 7 I/O depth levels*/
+	log_info(";%3.1f%%;%3.1f%%;%3.1f%%;%3.1f%%;%3.1f%%;%3.1f%%;%3.1f%%",
+			io_u_dist[0], io_u_dist[1], io_u_dist[2], io_u_dist[3],
+			io_u_dist[4], io_u_dist[5], io_u_dist[6]);
+
+	/* Microsecond latency */
+	for (i = 0; i < FIO_IO_U_LAT_U_NR; i++)
+		log_info(";%3.2f%%", io_u_lat_u[i]);
+	/* Millisecond latency */
+	for (i = 0; i < FIO_IO_U_LAT_M_NR; i++)
+		log_info(";%3.2f%%", io_u_lat_m[i]);
+
+	/* disk util stats, if any */
+	show_disk_util(1);
+
+	/* Additional output if continue_on_error set - default off*/
+	if (ts->continue_on_error)
+		log_info(";%lu;%d", ts->total_err_count, ts->first_error);
+	log_info("\n");
+
+	/* Additional output if description is set */
+	if (strlen(ts->description))
+		log_info(";%s", ts->description);
+}
+
+static void show_thread_status_terse(struct thread_stat *ts,
+				     struct group_run_stats *rs)
+{
+	if (terse_version == 2)
+		show_thread_status_terse_v2(ts, rs);
+	else if (terse_version == 3)
+		show_thread_status_terse_v3(ts, rs);
+	else
+		log_err("fio: bad terse version!? %d\n", terse_version);
+}
+
 static void sum_stat(struct io_stat *dst, struct io_stat *src, int nr)
 {
 	double mean, S;
@@ -642,6 +799,26 @@ static void sum_stat(struct io_stat *dst, struct io_stat *src, int nr)
 	dst->S.u.f = S;
 }
 
+void sum_group_stats(struct group_run_stats *dst, struct group_run_stats *src)
+{
+	int i;
+
+	for (i = 0; i < 2; i++) {
+		if (dst->max_run[i] < src->max_run[i])
+			dst->max_run[i] = src->max_run[i];
+		if (dst->min_run[i] && dst->min_run[i] > src->min_run[i])
+			dst->min_run[i] = src->min_run[i];
+		if (dst->max_bw[i] < src->max_bw[i])
+			dst->max_bw[i] = src->max_bw[i];
+		if (dst->min_bw[i] && dst->min_bw[i] > src->min_bw[i])
+			dst->min_bw[i] = src->min_bw[i];
+
+		dst->io_kb[i] += src->io_kb[i];
+		dst->agg[i] += src->agg[i];
+	}
+
+}
+
 void sum_thread_stats(struct thread_stat *dst, struct thread_stat *src, int nr)
 {
 	int l, k;
@@ -691,6 +868,28 @@ void sum_thread_stats(struct thread_stat *dst, struct thread_stat *src, int nr)
 	dst->total_complete += src->total_complete;
 }
 
+void init_group_run_stat(struct group_run_stats *gs)
+{
+	memset(gs, 0, sizeof(*gs));
+	gs->min_bw[0] = gs->min_run[0] = ~0UL;
+	gs->min_bw[1] = gs->min_run[1] = ~0UL;
+}
+
+void init_thread_stat(struct thread_stat *ts)
+{
+	int j;
+
+	memset(ts, 0, sizeof(*ts));
+
+	for (j = 0; j <= DDIR_WRITE; j++) {
+		ts->lat_stat[j].min_val = -1UL;
+		ts->clat_stat[j].min_val = -1UL;
+		ts->slat_stat[j].min_val = -1UL;
+		ts->bw_stat[j].min_val = -1UL;
+	}
+	ts->groupid = -1;
+}
+
 void show_run_stats(void)
 {
 	struct group_run_stats *runstats, *rs;
@@ -701,13 +900,8 @@ void show_run_stats(void)
 
 	runstats = malloc(sizeof(struct group_run_stats) * (groupid + 1));
 
-	for (i = 0; i < groupid + 1; i++) {
-		rs = &runstats[i];
-
-		memset(rs, 0, sizeof(*rs));
-		rs->min_bw[0] = rs->min_run[0] = ~0UL;
-		rs->min_bw[1] = rs->min_run[1] = ~0UL;
-	}
+	for (i = 0; i < groupid + 1; i++)
+		init_group_run_stat(&runstats[i]);
 
 	/*
 	 * find out how many threads stats we need. if group reporting isn't
@@ -729,18 +923,8 @@ void show_run_stats(void)
 
 	threadstats = malloc(nr_ts * sizeof(struct thread_stat));
 
-	for (i = 0; i < nr_ts; i++) {
-		ts = &threadstats[i];
-
-		memset(ts, 0, sizeof(*ts));
-		for (j = 0; j <= DDIR_WRITE; j++) {
-			ts->lat_stat[j].min_val = -1UL;
-			ts->clat_stat[j].min_val = -1UL;
-			ts->slat_stat[j].min_val = -1UL;
-			ts->bw_stat[j].min_val = -1UL;
-		}
-		ts->groupid = -1;
-	}
+	for (i = 0; i < nr_ts; i++)
+		init_thread_stat(&threadstats[i]);
 
 	j = 0;
 	last_ts = -1;
@@ -869,20 +1053,23 @@ void show_run_stats(void)
 			show_thread_status(ts, rs);
 	}
 
-	if (!terse_output) {
-		for (i = 0; i < groupid + 1; i++) {
-			rs = &runstats[i];
-
-			rs->groupid = i;
-			if (is_backend)
-				fio_server_send_gs(rs);
-			else
-				show_group_stats(rs);
-		}
+	for (i = 0; i < groupid + 1; i++) {
+		rs = &runstats[i];
 
-		show_disk_util();
+		rs->groupid = i;
+		if (is_backend)
+			fio_server_send_gs(rs);
+		else if (!terse_output)
+			show_group_stats(rs);
 	}
 
+	if (is_backend)
+		fio_server_send_du();
+	else if (!terse_output)
+		show_disk_util(0);
+
+	free_disk_util();
+
 	free(runstats);
 	free(threadstats);
 }
@@ -912,6 +1099,9 @@ static void __add_log_sample(struct io_log *iolog, unsigned long val,
 {
 	const int nr_samples = iolog->nr_samples;
 
+	if (!iolog->nr_samples)
+		iolog->avg_last = t;
+
 	if (iolog->nr_samples == iolog->max_samples) {
 		int new_size = sizeof(struct io_sample) * iolog->max_samples*2;
 
@@ -926,14 +1116,66 @@ static void __add_log_sample(struct io_log *iolog, unsigned long val,
 	iolog->nr_samples++;
 }
 
+static inline void reset_io_stat(struct io_stat *ios)
+{
+	ios->max_val = ios->min_val = ios->samples = 0;
+	ios->mean.u.f = ios->S.u.f = 0;
+}
+
 static void add_log_sample(struct thread_data *td, struct io_log *iolog,
 			   unsigned long val, enum fio_ddir ddir,
 			   unsigned int bs)
 {
+	unsigned long elapsed, this_window;
+
 	if (!ddir_rw(ddir))
 		return;
 
-	__add_log_sample(iolog, val, ddir, bs, mtime_since_now(&td->epoch));
+	elapsed = mtime_since_now(&td->epoch);
+
+	/*
+	 * If no time averaging, just add the log sample.
+	 */
+	if (!iolog->avg_msec) {
+		__add_log_sample(iolog, val, ddir, bs, elapsed);
+		return;
+	}
+
+	/*
+	 * Add the sample. If the time period has passed, then
+	 * add that entry to the log and clear.
+	 */
+	add_stat_sample(&iolog->avg_window[ddir], val);
+
+	/*
+	 * If period hasn't passed, adding the above sample is all we
+	 * need to do.
+	 */
+	this_window = elapsed - iolog->avg_last;
+	if (this_window < iolog->avg_msec)
+		return;
+
+	/*
+	 * Note an entry in the log. Use the mean from the logged samples,
+	 * making sure to properly round up. Only write a log entry if we
+	 * had actual samples done.
+	 */
+	if (iolog->avg_window[DDIR_READ].samples) {
+		unsigned long mr;
+
+		mr = iolog->avg_window[DDIR_READ].mean.u.f + 0.50;
+		__add_log_sample(iolog, mr, DDIR_READ, 0, elapsed);
+	}
+	if (iolog->avg_window[DDIR_WRITE].samples) {
+		unsigned long mw;
+
+		mw = iolog->avg_window[DDIR_WRITE].mean.u.f + 0.50;
+		__add_log_sample(iolog, mw, DDIR_WRITE, 0, elapsed);
+	}
+
+	reset_io_stat(&iolog->avg_window[DDIR_READ]);
+	reset_io_stat(&iolog->avg_window[DDIR_WRITE]);
+	iolog->avg_last = elapsed;
 }
 
 void add_agg_sample(unsigned long val, enum fio_ddir ddir, unsigned int bs)
@@ -1014,15 +1256,26 @@ void add_bw_sample(struct thread_data *td, enum fio_ddir ddir, unsigned int bs,
 	if (spent < td->o.bw_avg_time)
 		return;
 
-	rate = (td->this_io_bytes[ddir] - td->stat_io_bytes[ddir]) *
-			1000 / spent / 1024;
-	add_stat_sample(&ts->bw_stat[ddir], rate);
+	/*
+	 * Compute both read and write rates for the interval.
+	 */
+	for (ddir = DDIR_READ; ddir <= DDIR_WRITE; ddir++) {
+		uint64_t delta;
+
+		delta = td->this_io_bytes[ddir] - td->stat_io_bytes[ddir];
+		if (!delta)
+			continue; /* No entries for interval */
 
-	if (td->bw_log)
-		add_log_sample(td, td->bw_log, rate, ddir, bs);
+		rate = delta * 1000 / spent / 1024;
+		add_stat_sample(&ts->bw_stat[ddir], rate);
+
+		if (td->bw_log)
+			add_log_sample(td, td->bw_log, rate, ddir, bs);
+
+		td->stat_io_bytes[ddir] = td->this_io_bytes[ddir];
+	}
 
 	fio_gettime(&td->bw_sample_time, NULL);
-	td->stat_io_bytes[ddir] = td->this_io_bytes[ddir];
 }
 
 void add_iops_sample(struct thread_data *td, enum fio_ddir ddir,
@@ -1038,15 +1291,24 @@ void add_iops_sample(struct thread_data *td, enum fio_ddir ddir,
 	if (spent < td->o.iops_avg_time)
 		return;
 
-	iops = ((td->this_io_blocks[ddir] - td->stat_io_blocks[ddir]) * 1000) / spent;
+	/*
+	 * Compute both read and write rates for the interval.
+	 */
+	for (ddir = DDIR_READ; ddir <= DDIR_WRITE; ddir++) {
+		uint64_t delta;
+
+		delta = td->this_io_blocks[ddir] - td->stat_io_blocks[ddir];
+		if (!delta)
+			continue; /* No entries for interval */
+
+		iops = (delta * 1000) / spent;
+		add_stat_sample(&ts->iops_stat[ddir], iops);
 
-	add_stat_sample(&ts->iops_stat[ddir], iops);
+		if (td->iops_log)
+			add_log_sample(td, td->iops_log, iops, ddir, 0);
 
-	if (td->iops_log) {
-		assert(iops);
-		add_log_sample(td, td->iops_log, iops, ddir, 0);
+		td->stat_io_blocks[ddir] = td->this_io_blocks[ddir];
 	}
 
 	fio_gettime(&td->iops_sample_time, NULL);
-	td->stat_io_blocks[ddir] = td->this_io_blocks[ddir];
 }