Skip to content

Cache Tracer Host Entry to reduce allocation in RemoteHostnameAdder #10968

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
gh-worker-dd-mergequeue-cf854d merged 9 commits into from
Apr 2, 2026
+19 −2
Merged

Cache Tracer Host Entry to reduce allocation in RemoteHostnameAdder #10968

Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
f3d5093
Caching host entry to reduce allocation
dougqh Mar 25, 2026
29f5205
spotless
dougqh Mar 25, 2026
d8d553e
Merge branch 'master' into dougqh/cache-tracer-host
dougqh Mar 27, 2026
dd4a4c7
Merge branch 'master' into dougqh/cache-tracer-host
dougqh Apr 1, 2026
e452935
Flipped equals order
dougqh Apr 1, 2026
722fd7a
Merge branch 'master' into dougqh/cache-tracer-host
dougqh Apr 1, 2026
3d18d1d
Merge branch 'master' into dougqh/cache-tracer-host
dougqh Apr 1, 2026
6831f69
Merge branch 'master' into dougqh/cache-tracer-host
dougqh Apr 2, 2026
9e331ec
Merge branch 'master' into dougqh/cache-tracer-host
dougqh Apr 2, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 19 additions & 2 deletions dd-trace-core/src/main/java/datadog/trace/core/tagprocessor/RemoteHostnameAdder.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,15 +9,32 @@
public final class RemoteHostnameAdder extends TagsPostProcessor {
private final Supplier<String> hostnameSupplier;

private TagMap.Entry cachedHostEntry = null;

public RemoteHostnameAdder(Supplier<String> hostnameSupplier) {
this.hostnameSupplier = hostnameSupplier;
}

@Override
public void processTags(
TagMap unsafeTags, DDSpanContext spanContext, AppendableSpanLinks spanLinks) {
if (spanContext.getSpanId() == spanContext.getRootSpanId()) {
unsafeTags.put(DDTags.TRACER_HOST, hostnameSupplier.get());
if (spanContext.getSpanId() != spanContext.getRootSpanId()) {
return;
}

String hostname = hostnameSupplier.get();
if (hostname == null) {
Copy link
Copy Markdown
Contributor Author

@dougqh dougqh Mar 25, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is subtle change in behavior from before.
TagMap.put would have failed with an exception when given null.

I've decided to just ignore hostnameSupplier.get returning null instead.

return;
}

TagMap.Entry cachedHostEntry = this.cachedHostEntry;
if (cachedHostEntry != null && hostname.equals(cachedHostEntry.objectValue())) {
unsafeTags.set(cachedHostEntry);
return;
}
Comment on lines +30 to +34
Copy link
Copy Markdown
Contributor

@mhlidd mhlidd Apr 1, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Out of curiosity, is it better to check to cache before checking if hostname is null? Wonder which case would happen more frequently and is the one that we should optimize for.

Copy link
Copy Markdown
Contributor Author

@dougqh dougqh Apr 1, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That's a good point. My interpretation of the code is that a null return from hostNameSupplier isn't really expected, so checking the cache first probably does make sense.

Copy link
Copy Markdown
Contributor Author

@dougqh dougqh Apr 1, 2026 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ah, I see now. I was being conservative in my approach.

I wasn't assuming that hostnameSupplier.get() always produces the same value, so I have to call it anyway to check if the cached value matches the current value.
And because I didn't want to assume that it wasn't null, I'd have to call Object.equals which would incur a null check anyway.

I suppose as written, I could let a null hostName flow into cachedHostEntry.objectValue().equals.
But arguably, I'd be better reversing that equal check to take advantage of the more detailed type information on hostname (being String rather than Object).

That would allow for static devirtualization rather than profiled based devirtualization, but I have to keep the null check first.

Copy link
Copy Markdown
Contributor

@mhlidd mhlidd Apr 1, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for the explanation, makes sense to me!


TagMap.Entry newEntry = TagMap.Entry.create(DDTags.TRACER_HOST, hostname);
unsafeTags.set(newEntry);
this.cachedHostEntry = newEntry;
}
}
Loading