Merge branch 'axboe:master' into directory-operation

HOWTO.rst

@@ -756,8 +756,9 @@ Time related parameters
 	CPU mask of other jobs.
 
 .. option:: job_start_clock_id=int
-   The clock_id passed to the call to `clock_gettime` used to record job_start
-   in the `json` output format. Default is 0, or CLOCK_REALTIME.
+
+        The clock_id passed to the call to `clock_gettime` used to record
+        job_start in the `json` output format. Default is 0, or CLOCK_REALTIME.
 
 
 Target file/device
@@ -1630,7 +1631,7 @@ Block size
 	Comma-separated ranges may be specified for reads, writes, and trims as
 	described in :option:`blocksize`.
 
-	Example: ``bsrange=1k-4k,2k-8k``.
+	Example: ``bsrange=1k-4k,2k-8k`` also the ':' delimiter ``bsrange=1k:4k,2k:8k``.
 
 .. option:: bssplit=str[,str][,str]
 
@@ -2506,11 +2507,11 @@ with the caveat that when used on the command line, they must come after the
         want fio to use placement identifier only at indices 0, 2 and 5 specify
         ``fdp_pli=0,2,5``.
 
-.. option:: md_per_io_size=int : [io_uring_cmd]
+.. option:: md_per_io_size=int : [io_uring_cmd] [xnvme]
 
 	Size in bytes for separate metadata buffer per IO. Default: 0.
 
-.. option:: pi_act=int : [io_uring_cmd]
+.. option:: pi_act=int : [io_uring_cmd] [xnvme]
 
 	Action to take when nvme namespace is formatted with protection
 	information. If this is set to 1 and namespace is formatted with
@@ -2526,7 +2527,7 @@ with the caveat that when used on the command line, they must come after the
 	it will use the default slower generator.
 	(see: https://github.com/intel/isa-l)
 
-.. option:: pi_chk=str[,str][,str] : [io_uring_cmd]
+.. option:: pi_chk=str[,str][,str] : [io_uring_cmd] [xnvme]
 
 	Controls the protection information check. This can take one or more
 	of these values. Default: none.
@@ -2539,16 +2540,25 @@ with the caveat that when used on the command line, they must come after the
 	**APPTAG**
 		Enables protection information checking of application tag field.
 
-.. option:: apptag=int : [io_uring_cmd]
+.. option:: apptag=int : [io_uring_cmd] [xnvme]
 
 	Specifies logical block application tag value, if namespace is
 	formatted to use end to end protection information. Default: 0x1234.
 
-.. option:: apptag_mask=int : [io_uring_cmd]
+.. option:: apptag_mask=int : [io_uring_cmd] [xnvme]
 
 	Specifies logical block application tag mask value, if namespace is
 	formatted to use end to end protection information. Default: 0xffff.
 
+.. option:: num_range=int : [io_uring_cmd]
+
+	For trim command this will be the number of ranges to trim per I/O
+	request. The number of logical blocks per range is determined by the
+	:option:`bs` option which should be a multiple of logical block size.
+	This cannot be used with read or write. Note that setting this
+	option > 1, :option:`log_offset` will not be able to log all the
+	offsets. Default: 1.
+
 .. option:: cpuload=int : [cpuio]
 
 	Attempt to use the specified percentage of CPU cycles. This is a mandatory
@@ -2641,10 +2651,13 @@ with the caveat that when used on the command line, they must come after the
 		User datagram protocol V6.
 	**unix**
 		UNIX domain socket.
+	**vsock**
+		VSOCK protocol.
 
-	When the protocol is TCP or UDP, the port must also be given, as well as the
-	hostname if the job is a TCP listener or UDP reader. For unix sockets, the
+	When the protocol is TCP, UDP or VSOCK, the port must also be given, as well as the
+	hostname if the job is a TCP or VSOCK listener or UDP reader. For unix sockets, the
 	normal :option:`filename` option should be used and the port is invalid.
+	When the protocol is VSOCK, the :option:`hostname` is the CID of the remote VM.
 
 .. option:: listen : [netsplice] [net]
 
@@ -2788,7 +2801,7 @@ with the caveat that when used on the command line, they must come after the
 
 .. option:: sg_write_mode=str : [sg]
 
-	Specify the type of write commands to issue. This option can take three values:
+	Specify the type of write commands to issue. This option can take ten values:
 
 	**write**
 		This is the default where write opcodes are issued as usual.
@@ -3325,8 +3338,8 @@ I/O rate
 
 .. option:: rate_cycle=int
 
-	Average bandwidth for :option:`rate` and :option:`rate_min` over this number
-	of milliseconds. Defaults to 1000.
+        Average bandwidth for :option:`rate_min` and :option:`rate_iops_min`
+        over this number of milliseconds. Defaults to 1000.
 
 
 I/O latency
@@ -4078,12 +4091,15 @@ Measurements and reporting
 
 .. option:: log_avg_msec=int
 
-	By default, fio will log an entry in the iops, latency, or bw log for every
-	I/O that completes. When writing to the disk log, that can quickly grow to a
-	very large size. Setting this option makes fio average the each log entry
-	over the specified period of time, reducing the resolution of the log.  See
-	:option:`log_max_value` as well. Defaults to 0, logging all entries.
-	Also see `Log File Formats`_.
+        By default, fio will log an entry in the iops, latency, or bw log for
+        every I/O that completes. When writing to the disk log, that can
+        quickly grow to a very large size. Setting this option directs fio to
+        instead record an average over the specified duration for each log
+        entry, reducing the resolution of the log. When the job completes, fio
+        will flush any accumulated latency log data, so the final log interval
+        may not match the value specified by this option and there may even be
+        duplicate timestamps. See :option:`log_window_value` as well. Defaults
+        to 0, logging entries for each I/O. Also see `Log File Formats`_.
 
 .. option:: log_hist_msec=int
 
@@ -4103,11 +4119,28 @@ Measurements and reporting
 	histogram logs contain 1216 latency bins. See :option:`write_hist_log`
 	and `Log File Formats`_.
 
-.. option:: log_max_value=bool
+.. option:: log_window_value=str, log_max_value=str
+
+	If :option:`log_avg_msec` is set, fio by default logs the average over that
+	window. This option determines whether fio logs the average, maximum or
+	both the values over the window. This only affects the latency logging,
+	as both average and maximum values for iops or bw log will be same.
+	Accepted values are:
+
+		**avg**
+			Log average value over the window. The default.
+
+		**max**
+			Log maximum value in the window.
+
+		**both**
+			Log both average and maximum value over the window.
+
+		**0**
+			Backward-compatible alias for **avg**.
 
-	If :option:`log_avg_msec` is set, fio logs the average over that window. If
-	you instead want to log the maximum value, set this option to 1. Defaults to
-	0, meaning that averaged values are logged.
+		**1**
+			Backward-compatible alias for **max**.
 
 .. option:: log_offset=bool
 
@@ -5076,11 +5109,19 @@ toggled with :option:`log_offset`.
 by the ioengine specific :option:`cmdprio_percentage`.
 
 Fio defaults to logging every individual I/O but when windowed logging is set
-through :option:`log_avg_msec`, either the average (by default) or the maximum
-(:option:`log_max_value` is set) *value* seen over the specified period of time
-is recorded. Each *data direction* seen within the window period will aggregate
-its values in a separate row. Further, when using windowed logging the *block
-size* and *offset* entries will always contain 0.
+through :option:`log_avg_msec`, either the average (by default), the maximum
+(:option:`log_window_value` is set to max) *value* seen over the specified period
+of time, or both the average *value* and maximum *value1* (:option:`log_window_value`
+is set to both) is recorded. The log file format when both the values are reported
+takes this form:
+
+    *time* (`msec`), *value*, *value1*, *data direction*, *block size* (`bytes`),
+    *offset* (`bytes`), *command priority*
+
+
+Each *data direction* seen within the window period will aggregate its values in a
+separate row. Further, when using windowed logging the *block size* and *offset*
+entries will always contain 0.
 
 
 Client/Server

arch/arch-x86_64.h

@@ -26,6 +26,11 @@ static inline unsigned long arch_ffz(unsigned long bitmask)
 	return bitmask;
 }
 
+static inline void tsc_barrier(void)
+{
+	__asm__ __volatile__("mfence":::"memory");
+}
+
 static inline unsigned long long get_cpu_clock(void)
 {
 	unsigned int lo, hi;

arch/arch.h

@@ -108,6 +108,13 @@ extern unsigned long arch_flags;
 #include "arch-generic.h"
 #endif
 
+#if !defined(__x86_64__) && defined(CONFIG_SYNC_SYNC)
+static inline void tsc_barrier(void)
+{
+	__sync_synchronize();
+}
+#endif
+
 #include "../lib/ffz.h"
 /* IWYU pragma: end_exports */
 

backend.c

@@ -1333,7 +1333,7 @@ static int init_io_u(struct thread_data *td)
 int init_io_u_buffers(struct thread_data *td)
 {
 	struct io_u *io_u;
-	unsigned long long max_bs, min_write;
+	unsigned long long max_bs, min_write, trim_bs = 0;
 	int i, max_units;
 	int data_xfer = 1;
 	char *p;
@@ -1344,7 +1344,18 @@ int init_io_u_buffers(struct thread_data *td)
 	td->orig_buffer_size = (unsigned long long) max_bs
 					* (unsigned long long) max_units;
 
-	if (td_ioengine_flagged(td, FIO_NOIO) || !(td_read(td) || td_write(td)))
+	if (td_trim(td) && td->o.num_range > 1) {
+		trim_bs = td->o.num_range * sizeof(struct trim_range);
+		td->orig_buffer_size = trim_bs
+					* (unsigned long long) max_units;
+	}
+
+	/*
+	 * For reads, writes, and multi-range trim operations we need a
+	 * data buffer
+	 */
+	if (td_ioengine_flagged(td, FIO_NOIO) ||
+	    !(td_read(td) || td_write(td) || (td_trim(td) && td->o.num_range > 1)))
 		data_xfer = 0;
 
 	/*
@@ -1396,7 +1407,10 @@ int init_io_u_buffers(struct thread_data *td)
 				fill_verify_pattern(td, io_u->buf, max_bs, io_u, 0, 0);
 			}
 		}
-		p += max_bs;
+		if (td_trim(td) && td->o.num_range > 1)
+			p += trim_bs;
+		else
+			p += max_bs;
 	}
 
 	return 0;
@@ -2080,14 +2094,14 @@ static void reap_threads(unsigned int *nr_running, uint64_t *t_rate,
 			 uint64_t *m_rate)
 {
 	unsigned int cputhreads, realthreads, pending;
-	int status, ret;
+	int ret;
 
 	/*
 	 * reap exited threads (TD_EXITED -> TD_REAPED)
 	 */
 	realthreads = pending = cputhreads = 0;
 	for_each_td(td) {
-		int flags = 0;
+		int flags = 0, status;
 
 		if (!strcmp(td->o.ioengine, "cpuio"))
 			cputhreads++;

cconv.c

@@ -111,6 +111,7 @@ int convert_thread_options_to_cpu(struct thread_options *o,
 	o->serialize_overlap = le32_to_cpu(top->serialize_overlap);
 	o->size = le64_to_cpu(top->size);
 	o->io_size = le64_to_cpu(top->io_size);
+	o->num_range = le32_to_cpu(top->num_range);
 	o->size_percent = le32_to_cpu(top->size_percent);
 	o->io_size_percent = le32_to_cpu(top->io_size_percent);
 	o->fill_device = le32_to_cpu(top->fill_device);
@@ -609,6 +610,7 @@ void convert_thread_options_to_net(struct thread_options_pack *top,
 
 	top->size = __cpu_to_le64(o->size);
 	top->io_size = __cpu_to_le64(o->io_size);
+	top->num_range = __cpu_to_le32(o->num_range);
 	top->verify_backlog = __cpu_to_le64(o->verify_backlog);
 	top->start_delay = __cpu_to_le64(o->start_delay);
 	top->start_delay_high = __cpu_to_le64(o->start_delay_high);

ci/actions-install.sh

@@ -86,9 +86,8 @@ install_macos() {
     #echo "Updating homebrew..."
     #brew update >/dev/null 2>&1
     echo "Installing packages..."
-    HOMEBREW_NO_AUTO_UPDATE=1 brew install cunit libnfs sphinx-doc pygments python-certifi
-    brew link sphinx-doc --force
-    pip3 install scipy six statsmodels
+    HOMEBREW_NO_AUTO_UPDATE=1 brew install cunit libnfs
+    pip3 install scipy six statsmodels sphinx
 }
 
 install_windows() {

client.c

@@ -1718,8 +1718,10 @@ static struct cmd_iolog_pdu *convert_iolog(struct fio_net_cmd *cmd,
 			s = (struct io_sample *)((char *)s + sizeof(struct io_u_plat_entry) * i);
 
 		s->time		= le64_to_cpu(s->time);
-		if (ret->log_type != IO_LOG_TYPE_HIST)
-			s->data.val	= le64_to_cpu(s->data.val);
+		if (ret->log_type != IO_LOG_TYPE_HIST) {
+			s->data.val.val0	= le64_to_cpu(s->data.val.val0);
+			s->data.val.val1	= le64_to_cpu(s->data.val.val1);
+		}
 		s->__ddir	= __le32_to_cpu(s->__ddir);
 		s->bs		= le64_to_cpu(s->bs);
 		s->priority	= le16_to_cpu(s->priority);

configure

@@ -1728,6 +1728,25 @@ elif compile_prog "" "-lws2_32" "TCP_NODELAY"; then
 fi
 print_config "TCP_NODELAY" "$tcp_nodelay"
 
+##########################################
+# Check whether we have vsock
+if test "$vsock" != "yes" ; then
+  vsock="no"
+fi
+cat > $TMPC << EOF
+#include <sys/types.h>
+#include <sys/socket.h>
+#include <linux/vm_sockets.h>
+int main(int argc, char **argv)
+{
+  return socket(AF_VSOCK, SOCK_STREAM, 0);
+}
+EOF
+if compile_prog "" "" "vsock"; then
+  vsock="yes"
+fi
+print_config "vsock" "$vsock"
+
 ##########################################
 # Check whether we have SO_SNDBUF
 if test "$window_size" != "yes" ; then
@@ -2678,7 +2697,7 @@ fi
 ##########################################
 # Check if we have xnvme
 if test "$xnvme" != "no" ; then
-  if check_min_lib_version xnvme 0.7.0; then
+  if check_min_lib_version xnvme 0.7.4; then
     xnvme="yes"
     xnvme_cflags=$(pkg-config --cflags xnvme)
     xnvme_libs=$(pkg-config --libs xnvme)
@@ -3192,6 +3211,9 @@ fi
 if test "$ipv6" = "yes" ; then
   output_sym "CONFIG_IPV6"
 fi
+if test "$vsock" = "yes"; then
+  output_sym "CONFIG_VSOCK"
+fi
 if test "$http" = "yes" ; then
   output_sym "CONFIG_HTTP"
 fi