summary refs log tree commit diff stats
diff options
context:
space:
mode:
authorStefan Hajnoczi <stefanha@redhat.com>2023-10-10 10:21:43 -0400
committerStefan Hajnoczi <stefanha@redhat.com>2023-10-10 10:21:43 -0400
commit9a8981e68c8de5c13b4842d7cd169ba028c6ff2a (patch)
tree5829eba619e59c19bd696ea7242a5ba90de6a999
parentcea3ea670fe265421131aad90c36fbb87bc4d206 (diff)
parent320a6ccc769dc09ae7ad0b4838e322018f334bf4 (diff)
downloadfocaccia-qemu-9a8981e68c8de5c13b4842d7cd169ba028c6ff2a.tar.gz
focaccia-qemu-9a8981e68c8de5c13b4842d7cd169ba028c6ff2a.zip
Merge tag 'dirtylimit-dirtyrate-pull-request-20231010' of https://github.com/newfriday/qemu into staging
Dirtylimit and dirtyrate 20231010 patches PULL request

Dirty page rate measurement optimization.
Please apply, thanks, Yong.

# -----BEGIN PGP SIGNATURE-----
#
# iQJKBAABCAA0FiEEaF0CINwmSCgVLlfC3/Ij1rP+y5wFAmUklg0WHHlvbmcuaHVh
# bmdAc21hcnR4LmNvbQAKCRDf8iPWs/7LnIw6D/sECFML6dIDSckhKe1kRBT2oXRd
# lYz4/RRdxPJIJP0zS0yLYXd2d5vHzXdC3hETv7//QiWB2OP/UQnsZ70JCgF4DxIq
# bGL4BUHgaQmyUsyIQXceFznJJQOLs5DczDFJBR2zlQbu3YOAGeNJJmfxmVIEPfcL
# w5NK++g3nVZ3pLJBNblDBUwIW5uoOj6z85rCTc6pvddoTBcAqS0er1aTkuyx9jbB
# VsDdFNJumF6+VnE6QUITHX2knr3UmXc5dfXCJi4CLKRfx3nyK8vYydNawhPtobGD
# 0G5MZAPLO3JxdM67EccKj3I/kcAXU4iHu7rV5AscSI/rlfneGjfCup2Xd0we1GCR
# TD07AVDRuW+cS76nEtvDSRj6+8KarZEa3lVbvoPaXIazoHg3GjKylIMTAcGjFlzL
# AnGornOEZSfwNxT3BRvNHFdUNdA9ICZ90sEpWjeu80UNOT2JASOB6JE2VJBFnW81
# 4gaoIT74hpI8H2k/x3R8REPnw+SLMI+7VpcA2XcXuOQOdfk0+8zlvxPsJRBaKBzS
# d2es+CpUcmBxZdEQNOi905qxfCFLOhwcstJXyCvFQBp4f8l2SJfIE4liI29qpuma
# hubbOEo/EAVe8ywToHSYj2RU5hnj6gu0n3hvSeye76hS/K+bfvI+HZ3AX+rcRmP8
# vX6Vqs4wdNl5khEnNg==
# =ixu5
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 09 Oct 2023 20:08:45 EDT
# gpg:                using RSA key 685D0220DC264828152E57C2DFF223D6B3FECB9C
# gpg:                issuer "yong.huang@smartx.com"
# gpg: Good signature from "Yong Huang <yong.huang@smartx.com>" [unknown]
# gpg: WARNING: The key's User ID is not certified with a trusted signature!
# gpg:          There is no indication that the signature belongs to the owner.
# Primary key fingerprint: 685D 0220 DC26 4828 152E  57C2 DFF2 23D6 B3FE CB9C

* tag 'dirtylimit-dirtyrate-pull-request-20231010' of https://github.com/newfriday/qemu:
  migration/dirtyrate: use QEMU_CLOCK_HOST to report start-time
  migration/calc-dirty-rate: millisecond-granularity period

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
-rw-r--r--migration/dirtyrate.c122
-rw-r--r--migration/dirtyrate.h12
-rw-r--r--qapi/migration.json58
3 files changed, 134 insertions, 58 deletions
diff --git a/migration/dirtyrate.c b/migration/dirtyrate.c
index bccb3515e3..036ac017fc 100644
--- a/migration/dirtyrate.c
+++ b/migration/dirtyrate.c
@@ -189,10 +189,9 @@ retry:
     return duration;
 }
 
-static bool is_sample_period_valid(int64_t sec)
+static bool is_calc_time_valid(int64_t msec)
 {
-    if (sec < MIN_FETCH_DIRTYRATE_TIME_SEC ||
-        sec > MAX_FETCH_DIRTYRATE_TIME_SEC) {
+    if ((msec < MIN_CALC_TIME_MS) || (msec > MAX_CALC_TIME_MS)) {
         return false;
     }
 
@@ -216,7 +215,39 @@ static int dirtyrate_set_state(int *state, int old_state, int new_state)
     }
 }
 
-static struct DirtyRateInfo *query_dirty_rate_info(void)
+/* Decimal power of given time unit relative to one second */
+static int time_unit_to_power(TimeUnit time_unit)
+{
+    switch (time_unit) {
+    case TIME_UNIT_SECOND:
+        return 0;
+    case TIME_UNIT_MILLISECOND:
+        return -3;
+    default:
+        assert(false); /* unreachable */
+        return 0;
+    }
+}
+
+static int64_t convert_time_unit(int64_t value, TimeUnit unit_from,
+                                 TimeUnit unit_to)
+{
+    int power = time_unit_to_power(unit_from) -
+                time_unit_to_power(unit_to);
+    while (power < 0) {
+        value /= 10;
+        power += 1;
+    }
+    while (power > 0) {
+        value *= 10;
+        power -= 1;
+    }
+    return value;
+}
+
+
+static struct DirtyRateInfo *
+query_dirty_rate_info(TimeUnit calc_time_unit)
 {
     int i;
     int64_t dirty_rate = DirtyStat.dirty_rate;
@@ -225,7 +256,10 @@ static struct DirtyRateInfo *query_dirty_rate_info(void)
 
     info->status = CalculatingState;
     info->start_time = DirtyStat.start_time;
-    info->calc_time = DirtyStat.calc_time;
+    info->calc_time = convert_time_unit(DirtyStat.calc_time_ms,
+                                        TIME_UNIT_MILLISECOND,
+                                        calc_time_unit);
+    info->calc_time_unit = calc_time_unit;
     info->sample_pages = DirtyStat.sample_pages;
     info->mode = dirtyrate_mode;
 
@@ -259,12 +293,11 @@ static struct DirtyRateInfo *query_dirty_rate_info(void)
     return info;
 }
 
-static void init_dirtyrate_stat(int64_t start_time,
-                                struct DirtyRateConfig config)
+static void init_dirtyrate_stat(struct DirtyRateConfig config)
 {
     DirtyStat.dirty_rate = -1;
-    DirtyStat.start_time = start_time;
-    DirtyStat.calc_time = config.sample_period_seconds;
+    DirtyStat.start_time = qemu_clock_get_ms(QEMU_CLOCK_HOST) / 1000;
+    DirtyStat.calc_time_ms = config.calc_time_ms;
     DirtyStat.sample_pages = config.sample_pages_per_gigabytes;
 
     switch (config.mode) {
@@ -574,7 +607,6 @@ static inline void dirtyrate_manual_reset_protect(void)
 
 static void calculate_dirtyrate_dirty_bitmap(struct DirtyRateConfig config)
 {
-    int64_t msec = 0;
     int64_t start_time;
     DirtyPageRecord dirty_pages;
 
@@ -600,11 +632,9 @@ static void calculate_dirtyrate_dirty_bitmap(struct DirtyRateConfig config)
     record_dirtypages_bitmap(&dirty_pages, true);
 
     start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
-    DirtyStat.start_time = start_time / 1000;
+    DirtyStat.start_time = qemu_clock_get_ms(QEMU_CLOCK_HOST) / 1000;
 
-    msec = config.sample_period_seconds * 1000;
-    msec = dirty_stat_wait(msec, start_time);
-    DirtyStat.calc_time = msec / 1000;
+    DirtyStat.calc_time_ms = dirty_stat_wait(config.calc_time_ms, start_time);
 
     /*
      * do two things.
@@ -615,12 +645,12 @@ static void calculate_dirtyrate_dirty_bitmap(struct DirtyRateConfig config)
 
     record_dirtypages_bitmap(&dirty_pages, false);
 
-    DirtyStat.dirty_rate = do_calculate_dirtyrate(dirty_pages, msec);
+    DirtyStat.dirty_rate = do_calculate_dirtyrate(dirty_pages,
+                                                  DirtyStat.calc_time_ms);
 }
 
 static void calculate_dirtyrate_dirty_ring(struct DirtyRateConfig config)
 {
-    int64_t duration;
     uint64_t dirtyrate = 0;
     uint64_t dirtyrate_sum = 0;
     int i = 0;
@@ -628,15 +658,13 @@ static void calculate_dirtyrate_dirty_ring(struct DirtyRateConfig config)
     /* start log sync */
     global_dirty_log_change(GLOBAL_DIRTY_DIRTY_RATE, true);
 
-    DirtyStat.start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME) / 1000;
+    DirtyStat.start_time = qemu_clock_get_ms(QEMU_CLOCK_HOST) / 1000;
 
     /* calculate vcpu dirtyrate */
-    duration = vcpu_calculate_dirtyrate(config.sample_period_seconds * 1000,
-                                        &DirtyStat.dirty_ring,
-                                        GLOBAL_DIRTY_DIRTY_RATE,
-                                        true);
-
-    DirtyStat.calc_time = duration / 1000;
+    DirtyStat.calc_time_ms = vcpu_calculate_dirtyrate(config.calc_time_ms,
+                                                      &DirtyStat.dirty_ring,
+                                                      GLOBAL_DIRTY_DIRTY_RATE,
+                                                      true);
 
     /* calculate vm dirtyrate */
     for (i = 0; i < DirtyStat.dirty_ring.nvcpu; i++) {
@@ -652,27 +680,25 @@ static void calculate_dirtyrate_sample_vm(struct DirtyRateConfig config)
 {
     struct RamblockDirtyInfo *block_dinfo = NULL;
     int block_count = 0;
-    int64_t msec = 0;
     int64_t initial_time;
 
     rcu_read_lock();
     initial_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
+    DirtyStat.start_time = qemu_clock_get_ms(QEMU_CLOCK_HOST) / 1000;
     if (!record_ramblock_hash_info(&block_dinfo, config, &block_count)) {
         goto out;
     }
     rcu_read_unlock();
 
-    msec = config.sample_period_seconds * 1000;
-    msec = dirty_stat_wait(msec, initial_time);
-    DirtyStat.start_time = initial_time / 1000;
-    DirtyStat.calc_time = msec / 1000;
+    DirtyStat.calc_time_ms = dirty_stat_wait(config.calc_time_ms,
+                                             initial_time);
 
     rcu_read_lock();
     if (!compare_page_hash_info(block_dinfo, block_count)) {
         goto out;
     }
 
-    update_dirtyrate(msec);
+    update_dirtyrate(DirtyStat.calc_time_ms);
 
 out:
     rcu_read_unlock();
@@ -718,6 +744,8 @@ void *get_dirtyrate_thread(void *arg)
 }
 
 void qmp_calc_dirty_rate(int64_t calc_time,
+                         bool has_calc_time_unit,
+                         TimeUnit calc_time_unit,
                          bool has_sample_pages,
                          int64_t sample_pages,
                          bool has_mode,
@@ -727,7 +755,6 @@ void qmp_calc_dirty_rate(int64_t calc_time,
     static struct DirtyRateConfig config;
     QemuThread thread;
     int ret;
-    int64_t start_time;
 
     /*
      * If the dirty rate is already being measured, don't attempt to start.
@@ -737,10 +764,15 @@ void qmp_calc_dirty_rate(int64_t calc_time,
         return;
     }
 
-    if (!is_sample_period_valid(calc_time)) {
-        error_setg(errp, "calc-time is out of range[%d, %d].",
-                         MIN_FETCH_DIRTYRATE_TIME_SEC,
-                         MAX_FETCH_DIRTYRATE_TIME_SEC);
+    int64_t calc_time_ms = convert_time_unit(
+        calc_time,
+        has_calc_time_unit ? calc_time_unit : TIME_UNIT_SECOND,
+        TIME_UNIT_MILLISECOND
+    );
+
+    if (!is_calc_time_valid(calc_time_ms)) {
+        error_setg(errp, "Calculation time is out of range [%dms, %dms].",
+                         MIN_CALC_TIME_MS, MAX_CALC_TIME_MS);
         return;
     }
 
@@ -787,7 +819,7 @@ void qmp_calc_dirty_rate(int64_t calc_time,
         return;
     }
 
-    config.sample_period_seconds = calc_time;
+    config.calc_time_ms = calc_time_ms;
     config.sample_pages_per_gigabytes = sample_pages;
     config.mode = mode;
 
@@ -799,21 +831,24 @@ void qmp_calc_dirty_rate(int64_t calc_time,
      **/
     dirtyrate_mode = mode;
 
-    start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME) / 1000;
-    init_dirtyrate_stat(start_time, config);
+    init_dirtyrate_stat(config);
 
     qemu_thread_create(&thread, "get_dirtyrate", get_dirtyrate_thread,
                        (void *)&config, QEMU_THREAD_DETACHED);
 }
 
-struct DirtyRateInfo *qmp_query_dirty_rate(Error **errp)
+
+struct DirtyRateInfo *qmp_query_dirty_rate(bool has_calc_time_unit,
+                                           TimeUnit calc_time_unit,
+                                           Error **errp)
 {
-    return query_dirty_rate_info();
+    return query_dirty_rate_info(
+        has_calc_time_unit ? calc_time_unit : TIME_UNIT_SECOND);
 }
 
 void hmp_info_dirty_rate(Monitor *mon, const QDict *qdict)
 {
-    DirtyRateInfo *info = query_dirty_rate_info();
+    DirtyRateInfo *info = query_dirty_rate_info(TIME_UNIT_SECOND);
 
     monitor_printf(mon, "Status: %s\n",
                    DirtyRateStatus_str(info->status));
@@ -873,8 +908,11 @@ void hmp_calc_dirty_rate(Monitor *mon, const QDict *qdict)
         mode = DIRTY_RATE_MEASURE_MODE_DIRTY_RING;
     }
 
-    qmp_calc_dirty_rate(sec, has_sample_pages, sample_pages, true,
-                        mode, &err);
+    qmp_calc_dirty_rate(sec, /* calc-time */
+                        false, TIME_UNIT_SECOND, /* calc-time-unit */
+                        has_sample_pages, sample_pages,
+                        true, mode,
+                        &err);
     if (err) {
         hmp_handle_error(mon, err);
         return;
diff --git a/migration/dirtyrate.h b/migration/dirtyrate.h
index 594a5c0bb6..869c060941 100644
--- a/migration/dirtyrate.h
+++ b/migration/dirtyrate.h
@@ -31,10 +31,12 @@
 #define MIN_RAMBLOCK_SIZE                         128
 
 /*
- * Take 1s as minimum time for calculation duration
+ * Allowed range for dirty page rate calculation (in milliseconds).
+ * Lower limit relates to the smallest realistic downtime it
+ * makes sense to impose on migration.
  */
-#define MIN_FETCH_DIRTYRATE_TIME_SEC              1
-#define MAX_FETCH_DIRTYRATE_TIME_SEC              60
+#define MIN_CALC_TIME_MS                          50
+#define MAX_CALC_TIME_MS                       60000
 
 /*
  * Take 1/16 pages in 1G as the maxmum sample page count
@@ -44,7 +46,7 @@
 
 struct DirtyRateConfig {
     uint64_t sample_pages_per_gigabytes; /* sample pages per GB */
-    int64_t sample_period_seconds; /* time duration between two sampling */
+    int64_t calc_time_ms; /* desired calculation time (in milliseconds) */
     DirtyRateMeasureMode mode; /* mode of dirtyrate measurement */
 };
 
@@ -73,7 +75,7 @@ typedef struct SampleVMStat {
 struct DirtyRateStat {
     int64_t dirty_rate; /* dirty rate in MB/s */
     int64_t start_time; /* calculation start time in units of second */
-    int64_t calc_time; /* time duration of two sampling in units of second */
+    int64_t calc_time_ms; /* actual calculation time (in milliseconds) */
     uint64_t sample_pages; /* sample pages per GB */
     union {
         SampleVMStat page_sampling;
diff --git a/qapi/migration.json b/qapi/migration.json
index 8843e74b59..d8f3bbd7b0 100644
--- a/qapi/migration.json
+++ b/qapi/migration.json
@@ -1837,6 +1837,21 @@
   'data': ['page-sampling', 'dirty-ring', 'dirty-bitmap'] }
 
 ##
+# @TimeUnit:
+#
+# Specifies unit in which time-related value is specified.
+#
+# @second: value is in seconds
+#
+# @millisecond: value is in milliseconds
+#
+# Since 8.2
+#
+##
+{ 'enum': 'TimeUnit',
+  'data': ['second', 'millisecond'] }
+
+##
 # @DirtyRateInfo:
 #
 # Information about measured dirty page rate.
@@ -1848,8 +1863,10 @@
 #
 # @start-time: start time in units of second for calculation
 #
-# @calc-time: time period for which dirty page rate was measured
-#     (in seconds)
+# @calc-time: time period for which dirty page rate was measured,
+#     expressed and rounded down to @calc-time-unit.
+#
+# @calc-time-unit: time unit of @calc-time  (Since 8.2)
 #
 # @sample-pages: number of sampled pages per GiB of guest memory.
 #     Valid only in page-sampling mode (Since 6.1)
@@ -1866,6 +1883,7 @@
            'status': 'DirtyRateStatus',
            'start-time': 'int64',
            'calc-time': 'int64',
+           'calc-time-unit': 'TimeUnit',
            'sample-pages': 'uint64',
            'mode': 'DirtyRateMeasureMode',
            '*vcpu-dirty-rate': [ 'DirtyRateVcpu' ] } }
@@ -1901,12 +1919,16 @@
 #    This mode tracks page modification per each vCPU separately.  It
 #    requires that KVM accelerator property "dirty-ring-size" is set.
 #
-# @calc-time: time period in units of second for which dirty page rate
-#     is calculated.  Note that larger @calc-time values will
-#     typically result in smaller dirty page rates because page
-#     dirtying is a one-time event.  Once some page is counted as
-#     dirty during @calc-time period, further writes to this page will
-#     not increase dirty page rate anymore.
+# @calc-time: time period for which dirty page rate is calculated.
+#     By default it is specified in seconds, but the unit can be set
+#     explicitly with @calc-time-unit.  Note that larger @calc-time
+#     values will typically result in smaller dirty page rates because
+#     page dirtying is a one-time event.  Once some page is counted
+#     as dirty during @calc-time period, further writes to this page
+#     will not increase dirty page rate anymore.
+#
+# @calc-time-unit: time unit in which @calc-time is specified.
+#     By default it is seconds. (Since 8.2)
 #
 # @sample-pages: number of sampled pages per each GiB of guest memory.
 #     Default value is 512.  For 4KiB guest pages this corresponds to
@@ -1924,8 +1946,16 @@
 # -> {"execute": "calc-dirty-rate", "arguments": {"calc-time": 1,
 #                                                 'sample-pages': 512} }
 # <- { "return": {} }
+#
+# Measure dirty rate using dirty bitmap for 500 milliseconds:
+#
+# -> {"execute": "calc-dirty-rate", "arguments": {"calc-time": 500,
+#     "calc-time-unit": "millisecond", "mode": "dirty-bitmap"} }
+#
+# <- { "return": {} }
 ##
 { 'command': 'calc-dirty-rate', 'data': {'calc-time': 'int64',
+                                         '*calc-time-unit': 'TimeUnit',
                                          '*sample-pages': 'int',
                                          '*mode': 'DirtyRateMeasureMode'} }
 
@@ -1934,6 +1964,9 @@
 #
 # Query results of the most recent invocation of @calc-dirty-rate.
 #
+# @calc-time-unit: time unit in which to report calculation time.
+#     By default it is reported in seconds. (Since 8.2)
+#
 # Since: 5.2
 #
 # Examples:
@@ -1941,14 +1974,17 @@
 # 1. Measurement is in progress:
 #
 # <- {"status": "measuring", "sample-pages": 512,
-#     "mode": "page-sampling", "start-time": 3665220, "calc-time": 10}
+#     "mode": "page-sampling", "start-time": 1693900454, "calc-time": 10,
+#     "calc-time-unit": "second"}
 #
 # 2. Measurement has been completed:
 #
 # <- {"status": "measured", "sample-pages": 512, "dirty-rate": 108,
-#     "mode": "page-sampling", "start-time": 3665220, "calc-time": 10}
+#     "mode": "page-sampling", "start-time": 1693900454, "calc-time": 10,
+#     "calc-time-unit": "second"}
 ##
-{ 'command': 'query-dirty-rate', 'returns': 'DirtyRateInfo' }
+{ 'command': 'query-dirty-rate', 'data': {'*calc-time-unit': 'TimeUnit' },
+                                 'returns': 'DirtyRateInfo' }
 
 ##
 # @DirtyLimitInfo: