vldpyatkov commented on code in PR #4808:
URL: https://github.com/apache/ignite-3/pull/4808#discussion_r1863078618
##########
modules/core/src/main/java/org/apache/ignite/internal/hlc/HybridClockImpl.java:
##########
@@ -111,20 +111,30 @@ public final HybridTimestamp current() {
*/
@Override
public final HybridTimestamp update(HybridTimestamp requestTime) {
- while (true) {
- long now = currentTime();
+ long requestTimeLong = requestTime.longValue();
+ while (true) {
// Read the latest time after accessing UTC time to reduce
contention.
- long oldLatestTime = latestTime;
+ long oldLatestTime = this.latestTime;
- long newLatestTime = max(requestTime.longValue() + 1, max(now,
oldLatestTime + 1));
+ if (oldLatestTime >= requestTimeLong) {
+ return hybridTimestamp(LATEST_TIME.incrementAndGet(this));
+ }
- // TODO https://issues.apache.org/jira/browse/IGNITE-23707 avoid
CAS on logical part update.
+ long now = currentTime();
+
+ if (now > requestTimeLong) {
+ if (LATEST_TIME.compareAndSet(this, oldLatestTime, now)) {
+ return hybridTimestamp(now);
+ }
+ } else {
+ long newLatestTime = requestTimeLong + 1;
- if (LATEST_TIME.compareAndSet(this, oldLatestTime, newLatestTime))
{
- notifyUpdateListeners(newLatestTime);
+ if (LATEST_TIME.compareAndSet(this, oldLatestTime,
newLatestTime)) {
Review Comment:
> Do we have any benchmark results here?
No, I did not run a benchmark.
We already estimated that the incrementAndGet has a less impact, either on
latency or throughput.
##########
modules/core/src/main/java/org/apache/ignite/internal/hlc/HybridClockImpl.java:
##########
@@ -111,20 +111,30 @@ public final HybridTimestamp current() {
*/
@Override
public final HybridTimestamp update(HybridTimestamp requestTime) {
- while (true) {
- long now = currentTime();
+ long requestTimeLong = requestTime.longValue();
+ while (true) {
// Read the latest time after accessing UTC time to reduce
contention.
- long oldLatestTime = latestTime;
+ long oldLatestTime = this.latestTime;
- long newLatestTime = max(requestTime.longValue() + 1, max(now,
oldLatestTime + 1));
+ if (oldLatestTime >= requestTimeLong) {
+ return hybridTimestamp(LATEST_TIME.incrementAndGet(this));
+ }
- // TODO https://issues.apache.org/jira/browse/IGNITE-23707 avoid
CAS on logical part update.
+ long now = currentTime();
+
+ if (now > requestTimeLong) {
+ if (LATEST_TIME.compareAndSet(this, oldLatestTime, now)) {
+ return hybridTimestamp(now);
+ }
+ } else {
+ long newLatestTime = requestTimeLong + 1;
- if (LATEST_TIME.compareAndSet(this, oldLatestTime, newLatestTime))
{
- notifyUpdateListeners(newLatestTime);
+ if (LATEST_TIME.compareAndSet(this, oldLatestTime,
newLatestTime)) {
Review Comment:
> Do we have any benchmark results here?
No, I did not run a benchmark.
We already estimated that the incrementAndGet has a less impact, either on
latency or throughput.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscr...@ignite.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
