X-Git-Url: https://git.kernel.dk/?p=fio.git;a=blobdiff_plain;f=eta.c;h=a90f1fb4763758eaedc9130036a2d60b720fc9bc;hp=600b046ced8c86e16ff90b1d2dd05ac59e637d5a;hb=256bc54c13ba8b37eb0e08fdbbd1bbbcfe8a86cc;hpb=f982d35f5ae88f728c6853aaf57448db8f5fcf98

diff --git a/eta.c b/eta.c
index 600b046c..a90f1fb4 100644
--- a/eta.c
+++ b/eta.c
@@ -7,14 +7,33 @@
 
 #include "fio.h"
 
-static char run_str[REAL_MAX_JOBS + 1];
+static char __run_str[REAL_MAX_JOBS + 1];
+static char run_str[__THREAD_RUNSTR_SZ(REAL_MAX_JOBS)];
+
+static void update_condensed_str(char *rstr, char *run_str_condensed)
+{
+	if (*rstr) {
+		while (*rstr) {
+			int nr = 1;
+
+			*run_str_condensed++ = *rstr++;
+			while (*(rstr - 1) == *rstr) {
+				rstr++;
+				nr++;
+			}
+			run_str_condensed += sprintf(run_str_condensed, "(%u),", nr);
+		}
+		run_str_condensed--;
+	}
+	*run_str_condensed = '\0';
+}
 
 /*
  * Sets the status of the 'td' in the printed status map.
  */
 static void check_str_update(struct thread_data *td)
 {
-	char c = run_str[td->thread_number - 1];
+	char c = __run_str[td->thread_number - 1];
 
 	switch (td->runstate) {
 	case TD_REAPED:
@@ -74,10 +93,14 @@ static void check_str_update(struct thread_data *td)
 	case TD_FSYNCING:
 		c = 'F';
 		break;
+	case TD_FINISHING:
+		c = 'f';
+		break;
 	case TD_CREATED:
 		c = 'C';
 		break;
 	case TD_INITIALIZED:
+	case TD_SETTING_UP:
 		c = 'I';
 		break;
 	case TD_NOT_CREATED:
@@ -87,7 +110,8 @@ static void check_str_update(struct thread_data *td)
 		log_err("state %d\n", td->runstate);
 	}
 
-	run_str[td->thread_number - 1] = c;
+	__run_str[td->thread_number - 1] = c;
+	update_condensed_str(__run_str, run_str);
 }
 
 /*
@@ -126,8 +150,10 @@ static int thread_eta(struct thread_data *td)
 	unsigned long long bytes_total, bytes_done;
 	unsigned long eta_sec = 0;
 	unsigned long elapsed;
+	uint64_t timeout;
 
 	elapsed = (mtime_since_now(&td->epoch) + 999) / 1000;
+	timeout = td->o.timeout / 1000000UL;
 
 	bytes_total = td->total_io_size;
 
@@ -138,41 +164,66 @@ static int thread_eta(struct thread_data *td)
 		bytes_total = td->fill_device_size;
 	}
 
+	if (td->o.zone_size && td->o.zone_skip && bytes_total) {
+		unsigned int nr_zones;
+		uint64_t zone_bytes;
+
+		zone_bytes = bytes_total + td->o.zone_size + td->o.zone_skip;
+		nr_zones = (zone_bytes - 1) / (td->o.zone_size + td->o.zone_skip);
+		bytes_total -= nr_zones * td->o.zone_skip;
+	}
+
 	/*
-	 * if writing, bytes_total will be twice the size. If mixing,
-	 * assume a 50/50 split and thus bytes_total will be 50% larger.
+	 * if writing and verifying afterwards, bytes_total will be twice the
+	 * size. In a mixed workload, verify phase will be the size of the
+	 * first stage writes.
 	 */
 	if (td->o.do_verify && td->o.verify && td_write(td)) {
-		if (td_rw(td))
-			bytes_total = bytes_total * 3 / 2;
-		else
+		if (td_rw(td)) {
+			unsigned int perc = 50;
+
+			if (td->o.rwmix[DDIR_WRITE])
+				perc = td->o.rwmix[DDIR_WRITE];
+
+			bytes_total += (bytes_total * perc) / 100;
+		} else
 			bytes_total <<= 1;
 	}
 
-	if (td->o.zone_size && td->o.zone_skip)
-		bytes_total /= (td->o.zone_skip / td->o.zone_size);
-
 	if (td->runstate == TD_RUNNING || td->runstate == TD_VERIFYING) {
 		double perc, perc_t;
 
 		bytes_done = ddir_rw_sum(td->io_bytes);
-		perc = (double) bytes_done / (double) bytes_total;
-		if (perc > 1.0)
-			perc = 1.0;
+
+		if (bytes_total) {
+			perc = (double) bytes_done / (double) bytes_total;
+			if (perc > 1.0)
+				perc = 1.0;
+		} else
+			perc = 0.0;
 
 		if (td->o.time_based) {
-			perc_t = (double) elapsed / (double) td->o.timeout;
-			if (perc_t < perc)
-				perc = perc_t;
+			if (timeout) {
+				perc_t = (double) elapsed / (double) timeout;
+				if (perc_t < perc)
+					perc = perc_t;
+			} else {
+				/*
+				 * Will never hit, we can't have time_based
+				 * without a timeout set.
+				 */
+				perc = 0.0;
+			}
 		}
 
 		eta_sec = (unsigned long) (elapsed * (1.0 / perc)) - elapsed;
 
 		if (td->o.timeout &&
-		    eta_sec > (td->o.timeout + done_secs - elapsed))
-			eta_sec = td->o.timeout + done_secs - elapsed;
+		    eta_sec > (timeout + done_secs - elapsed))
+			eta_sec = timeout + done_secs - elapsed;
 	} else if (td->runstate == TD_NOT_CREATED || td->runstate == TD_CREATED
 			|| td->runstate == TD_INITIALIZED
+			|| td->runstate == TD_SETTING_UP
 			|| td->runstate == TD_RAMP
 			|| td->runstate == TD_PRE_READING) {
 		int t_eta = 0, r_eta = 0;
@@ -183,8 +234,12 @@ static int thread_eta(struct thread_data *td)
 		 * if given, otherwise assume it'll run at the specified rate.
 		 */
 		if (td->o.timeout) {
-			t_eta = td->o.timeout + td->o.start_delay +
-					td->o.ramp_time;
+			uint64_t __timeout = td->o.timeout;
+			uint64_t start_delay = td->o.start_delay;
+			uint64_t ramp_time = td->o.ramp_time;
+
+			t_eta = __timeout + start_delay + ramp_time;
+			t_eta /= 1000000ULL;
 
 			if (in_ramp_time(td)) {
 				unsigned long ramp_left;
@@ -198,7 +253,7 @@ static int thread_eta(struct thread_data *td)
 		rate_bytes = ddir_rw_sum(td->o.rate);
 		if (rate_bytes) {
 			r_eta = (bytes_total / 1024) / rate_bytes;
-			r_eta += td->o.start_delay;
+			r_eta += (td->o.start_delay / 1000000ULL);
 		}
 
 		if (r_eta && t_eta)
@@ -219,7 +274,8 @@ static int thread_eta(struct thread_data *td)
 	return eta_sec;
 }
 
-static void calc_rate(unsigned long mtime, unsigned long long *io_bytes,
+static void calc_rate(int unified_rw_rep, unsigned long mtime,
+		      unsigned long long *io_bytes,
 		      unsigned long long *prev_io_bytes, unsigned int *rate)
 {
 	int i;
@@ -228,19 +284,32 @@ static void calc_rate(unsigned long mtime, unsigned long long *io_bytes,
 		unsigned long long diff;
 
 		diff = io_bytes[i] - prev_io_bytes[i];
-		rate[i] = ((1000 * diff) / mtime) / 1024;
+		if (unified_rw_rep) {
+			rate[i] = 0;
+			rate[0] += ((1000 * diff) / mtime) / 1024;
+		} else
+			rate[i] = ((1000 * diff) / mtime) / 1024;
 
 		prev_io_bytes[i] = io_bytes[i];
 	}
 }
 
-static void calc_iops(unsigned long mtime, unsigned long long *io_iops,
+static void calc_iops(int unified_rw_rep, unsigned long mtime,
+		      unsigned long long *io_iops,
 		      unsigned long long *prev_io_iops, unsigned int *iops)
 {
 	int i;
 
 	for (i = 0; i < DDIR_RWDIR_CNT; i++) {
-		iops[i] = ((io_iops[i] - prev_io_iops[i]) * 1000) / mtime;
+		unsigned long long diff;
+
+		diff = io_iops[i] - prev_io_iops[i];
+		if (unified_rw_rep) {
+			iops[i] = 0;
+			iops[0] += (diff * 1000) / mtime;
+		} else
+			iops[i] = (diff * 1000) / mtime;
+
 		prev_io_iops[i] = io_iops[i];
 	}
 }
@@ -252,7 +321,7 @@ static void calc_iops(unsigned long mtime, unsigned long long *io_iops,
 int calc_thread_status(struct jobs_eta *je, int force)
 {
 	struct thread_data *td;
-	int i;
+	int i, unified_rw_rep;
 	unsigned long rate_time, disp_time, bw_avg_time, *eta_secs;
 	unsigned long long io_bytes[DDIR_RWDIR_CNT];
 	unsigned long long io_iops[DDIR_RWDIR_CNT];
@@ -264,7 +333,8 @@ int calc_thread_status(struct jobs_eta *je, int force)
 	static struct timeval rate_prev_time, disp_prev_time;
 
 	if (!force) {
-		if (output_format != FIO_OUTPUT_NORMAL)
+		if (output_format != FIO_OUTPUT_NORMAL &&
+		    f_out == stdout)
 			return 0;
 		if (temp_stall_ts || eta_print == FIO_ETA_NEVER)
 			return 0;
@@ -286,14 +356,18 @@ int calc_thread_status(struct jobs_eta *je, int force)
 	io_bytes[DDIR_READ] = io_bytes[DDIR_WRITE] = io_bytes[DDIR_TRIM] = 0;
 	io_iops[DDIR_READ] = io_iops[DDIR_WRITE] = io_iops[DDIR_TRIM] = 0;
 	bw_avg_time = ULONG_MAX;
+	unified_rw_rep = 0;
 	for_each_td(td, i) {
+		unified_rw_rep += td->o.unified_rw_rep;
 		if (is_power_of_2(td->o.kb_base))
 			je->is_pow2 = 1;
+		je->unit_base = td->o.unit_base;
 		if (td->o.bw_avg_time < bw_avg_time)
 			bw_avg_time = td->o.bw_avg_time;
 		if (td->runstate == TD_RUNNING || td->runstate == TD_VERIFYING
 		    || td->runstate == TD_FSYNCING
-		    || td->runstate == TD_PRE_READING) {
+		    || td->runstate == TD_PRE_READING
+		    || td->runstate == TD_FINISHING) {
 			je->nr_running++;
 			if (td_read(td)) {
 				je->t_rate[0] += td->o.rate[DDIR_READ];
@@ -318,7 +392,9 @@ int calc_thread_status(struct jobs_eta *je, int force)
 		} else if (td->runstate == TD_RAMP) {
 			je->nr_running++;
 			je->nr_ramp++;
-		} else if (td->runstate < TD_RUNNING)
+		} else if (td->runstate == TD_SETTING_UP)
+			je->nr_setting_up++;
+		else if (td->runstate < TD_RUNNING)
 			je->nr_pending++;
 
 		if (je->elapsed_sec >= 3)
@@ -328,11 +404,17 @@ int calc_thread_status(struct jobs_eta *je, int force)
 
 		check_str_update(td);
 
-		if (td->runstate > TD_RAMP) {
+		if (td->runstate > TD_SETTING_UP) {
 			int ddir;
+
 			for (ddir = DDIR_READ; ddir < DDIR_RWDIR_CNT; ddir++) {
-				io_bytes[ddir] += td->io_bytes[ddir];
-				io_iops[ddir] += td->io_blocks[ddir];
+				if (unified_rw_rep) {
+					io_bytes[0] += td->io_bytes[ddir];
+					io_iops[0] += td->io_blocks[ddir];
+				} else {
+					io_bytes[ddir] += td->io_bytes[ddir];
+					io_iops[ddir] += td->io_blocks[ddir];
+				}
 			}
 		}
 	}
@@ -358,7 +440,8 @@ int calc_thread_status(struct jobs_eta *je, int force)
 	rate_time = mtime_since(&rate_prev_time, &now);
 
 	if (write_bw_log && rate_time > bw_avg_time && !in_ramp_time(td)) {
-		calc_rate(rate_time, io_bytes, rate_io_bytes, je->rate);
+		calc_rate(unified_rw_rep, rate_time, io_bytes, rate_io_bytes,
+				je->rate);
 		memcpy(&rate_prev_time, &now, sizeof(now));
 		add_agg_sample(je->rate[DDIR_READ], DDIR_READ, 0);
 		add_agg_sample(je->rate[DDIR_WRITE], DDIR_WRITE, 0);
@@ -373,8 +456,8 @@ int calc_thread_status(struct jobs_eta *je, int force)
 	if (!force && disp_time < 900)
 		return 0;
 
-	calc_rate(disp_time, io_bytes, disp_io_bytes, je->rate);
-	calc_iops(disp_time, io_iops, disp_io_iops, je->iops);
+	calc_rate(unified_rw_rep, disp_time, io_bytes, disp_io_bytes, je->rate);
+	calc_iops(unified_rw_rep, disp_time, io_iops, disp_io_iops, je->iops);
 
 	memcpy(&disp_prev_time, &now, sizeof(now));
 
@@ -382,13 +465,15 @@ int calc_thread_status(struct jobs_eta *je, int force)
 		return 0;
 
 	je->nr_threads = thread_number;
-	memcpy(je->run_str, run_str, thread_number * sizeof(char));
-
+	update_condensed_str(__run_str, run_str);
+	memcpy(je->run_str, run_str, strlen(run_str));
 	return 1;
 }
 
 void display_thread_status(struct jobs_eta *je)
 {
+	static struct timeval disp_eta_new_line;
+	static int eta_new_line_init, eta_new_line_pending;
 	static int linelen_last;
 	static int eta_good;
 	char output[REAL_MAX_JOBS + 512], *p = output;
@@ -400,12 +485,17 @@ void display_thread_status(struct jobs_eta *je)
 		eta_to_str(eta_str, je->eta_sec);
 	}
 
+	if (eta_new_line_pending) {
+		eta_new_line_pending = 0;
+		p += sprintf(p, "\n");
+	}
+
 	p += sprintf(p, "Jobs: %d (f=%d)", je->nr_running, je->files_open);
 	if (je->m_rate[0] || je->m_rate[1] || je->t_rate[0] || je->t_rate[1]) {
 		char *tr, *mr;
 
-		mr = num2str(je->m_rate[0] + je->m_rate[1], 4, 0, je->is_pow2);
-		tr = num2str(je->t_rate[0] + je->t_rate[1], 4, 0, je->is_pow2);
+		mr = num2str(je->m_rate[0] + je->m_rate[1], 4, 0, je->is_pow2, 8);
+		tr = num2str(je->t_rate[0] + je->t_rate[1], 4, 0, je->is_pow2, 8);
 		p += sprintf(p, ", CR=%s/%s KB/s", tr, mr);
 		free(tr);
 		free(mr);
@@ -425,15 +515,20 @@ void display_thread_status(struct jobs_eta *je)
 		if ((!je->eta_sec && !eta_good) || je->nr_ramp == je->nr_running)
 			strcpy(perc_str, "-.-% done");
 		else {
+			double mult = 100.0;
+
+			if (je->nr_setting_up && je->nr_running)
+				mult *= (1.0 - (double) je->nr_setting_up / (double) je->nr_running);
+
 			eta_good = 1;
-			perc *= 100.0;
+			perc *= mult;
 			sprintf(perc_str, "%3.1f%% done", perc);
 		}
 
 		for (ddir = DDIR_READ; ddir < DDIR_RWDIR_CNT; ddir++) {
 			rate_str[ddir] = num2str(je->rate[ddir], 5,
-						1024, je->is_pow2);
-			iops_str[ddir] = num2str(je->iops[ddir], 4, 1, 0);
+						1024, je->is_pow2, je->unit_base);
+			iops_str[ddir] = num2str(je->iops[ddir], 4, 1, 0, 0);
 		}
 
 		left = sizeof(output) - (p - output) - 1;
@@ -456,22 +551,45 @@ void display_thread_status(struct jobs_eta *je)
 	p += sprintf(p, "\r");
 
 	printf("%s", output);
+
+	if (!eta_new_line_init) {
+		fio_gettime(&disp_eta_new_line, NULL);
+		eta_new_line_init = 1;
+	} else if (eta_new_line && mtime_since_now(&disp_eta_new_line) > eta_new_line) {
+		fio_gettime(&disp_eta_new_line, NULL);
+		eta_new_line_pending = 1;
+	}
+
 	fflush(stdout);
 }
 
-void print_thread_status(void)
+struct jobs_eta *get_jobs_eta(int force, size_t *size)
 {
 	struct jobs_eta *je;
-	size_t size;
 
 	if (!thread_number)
-		return;
+		return NULL;
+
+	*size = sizeof(*je) + THREAD_RUNSTR_SZ;
+	je = malloc(*size);
+	memset(je, 0, *size);
+
+	if (!calc_thread_status(je, force)) {
+		free(je);
+		return NULL;
+	}
+
+	*size = sizeof(*je) + strlen((char *) je->run_str) + 1;
+	return je;
+}
 
-	size = sizeof(*je) + thread_number * sizeof(char) + 1;
-	je = malloc(size);
-	memset(je, 0, size);
+void print_thread_status(void)
+{
+	struct jobs_eta *je;
+	size_t size;
 
-	if (calc_thread_status(je, 0))
+	je = get_jobs_eta(0, &size);
+	if (je)
 		display_thread_status(je);
 
 	free(je);
@@ -479,5 +597,6 @@ void print_thread_status(void)
 
 void print_status_init(int thr_number)
 {
-	run_str[thr_number] = 'P';
+	__run_str[thr_number] = 'P';
+	update_condensed_str(__run_str, run_str);
 }