buffer: improve performance of multiple Buffer operations by thisalihassan · Pull Request #61871 · nodejs/node

thisalihassan · 2026-02-17T21:33:58Z

Multiple performance improvements to Buffer operations, verified with benchmarks (15-30 runs, comparing old vs new binaries built from same tree).

Buffer.copyBytesFrom() (+100-210%)
Avoid intermediate TypedArrayPrototypeSlice allocation by calculating byte offsets directly into the source TypedArray's underlying ArrayBuffer.

Buffer.prototype.fill("t", "ascii") (+26-37%)

ASCII indexOf (+14-46%)
Call indexOfString directly for ASCII encoding instead of first converting the search value to a Buffer via fromStringFast and then calling indexOfBuffer. ASCII and Latin-1 share the same byte values for characters 0-127.

swap16/32/64 (+3-38%)
Add V8 Fast API C++ functions (FastSwap16/32/64) alongside the existing slow path. Largest gains at len=256 (+35%).

Benchmark results

Key results (15-30 runs, *** = p < 0.001):

Benchmark	Improvement
copyBytesFrom (offset, Uint8Array, len=256)	+210% ***
copyBytesFrom (offset+length, Uint8Array, len=256)	+206% ***
swap16 len=256	+38% ***
fill("t", "ascii") size=8192	+37% ***
indexOf ASCII 'Alice'	+46% ***
indexOf ASCII '@'	+31% ***
fill("t", "ascii") size=65536	+26% ***
swap64 len=768 aligned	+12% ***

Attaching Details below:
detail.pdf -- visual breakdown
buffer-benchmark-all-rebased.csv -- raw benchmark CSV
compare-R-output.txt-- full output

nodejs-github-bot · 2026-02-17T21:34:03Z

Review requested:

@nodejs/performance

lib/buffer.js

Renegade334 · 2026-02-17T21:50:59Z

src/node_buffer.cc

+void FastSwap16(Local<Value> receiver,
+                Local<Value> buffer_obj,
+                // NOLINTNEXTLINE(runtime/references)
+                FastApiCallbackOptions& options) {
+  HandleScope scope(options.isolate);
+  ArrayBufferViewContents<char> buffer(buffer_obj);
+  CHECK(nbytes::SwapBytes16(const_cast<char*>(buffer.data()), buffer.length()));
+}


These fast callbacks are non-identical to the conventional callbacks they shadow.

The existing callbacks validate their argument and throws to JS if invalid, whereas your fast callbacks hard-crash the process. It might be better to validate in the JS layer, then use the same unwrapping logic on both sides.

Your fast callback cannot have a different return convention to the conventional callback. You will need to remove the return value from the conventional callback.

Was removing the validation instead of moving it to js intentional?

Also fast paths can keep the validation and fall back to slow i think, so we can still validate on the src side?

Also fast paths can keep the validation and fall back to slow i think, so we can still validate on the src side?

This is outdated, the fast API doesn't use fallback any more (since Node.js v23.x).

However, any validation in a JS wrapper should be shadowed by a CHECK or DCHECK in the C++ binding.

SPREAD_BUFFER_ARG already contains CHECK((val)->IsArrayBufferView()) on all Slow & Fast swap methods

lib/buffer.js

Renegade334 · 2026-02-17T22:37:01Z

src/node_buffer.cc

+void FastSwap16(Local<Value> receiver,
+                Local<Value> buffer_obj,
+                // NOLINTNEXTLINE(runtime/references)
+                FastApiCallbackOptions& options) {
+  HandleScope scope(options.isolate);
+  ArrayBufferViewContents<char> buffer(buffer_obj);
+  CHECK(nbytes::SwapBytes16(const_cast<char*>(buffer.data()), buffer.length()));
+}


Fast callbacks should include debug tracking and call tests.

Thanks for sharing these I will update the code

codecov · 2026-02-17T23:10:45Z

Codecov Report

❌ Patch coverage is 92.39130% with 7 lines in your changes missing coverage. Please review.
✅ Project coverage is 89.69%. Comparing base (7547e79) to head (85fb7f1).
⚠️ Report is 36 commits behind head on main.

Files with missing lines	Patch %	Lines
src/node_buffer.cc	72.00%	0 Missing and 7 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #61871      +/-   ##
==========================================
- Coverage   91.60%   89.69%   -1.91%     
==========================================
  Files         337      676     +339     
  Lines      140745   206763   +66018     
  Branches    21802    39605   +17803     
==========================================
+ Hits       128925   185452   +56527     
- Misses      11595    13444    +1849     
- Partials      225     7867    +7642

Files with missing lines	Coverage Δ
lib/buffer.js	`99.16% <100.00%> (+12.78%)`	⬆️
src/node_buffer.cc	`68.19% <72.00%> (ø)`

... and 459 files with indirect coverage changes

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

Renegade334 · 2026-02-18T00:08:19Z

src/node_buffer.cc

+                FastApiCallbackOptions& options) {
+  TRACK_V8_FAST_API_CALL("buffer.swap16");
+  HandleScope scope(options.isolate);
+  ArrayBufferViewContents<char> buffer(buffer_obj);


ArrayBufferViewContents is wrong here, as buffer.data() may be a stack-allocated copy of the byte data rather than the data itself. SPREAD_BUFFER_ARG is the correct macro to use here, as per the conventional callback.

@Renegade334 replaced ArrayBufferViewContents with SPREAD_BUFFER_ARG in all three fast swap callbacks

lib/buffer.js

ChALkeR

For toHex, wait until #61609, which improves native perf significantly (more than Uint8Array.prototype.toHex)

See also #60249 (comment)

ChALkeR · 2026-02-18T03:34:33Z

lib/buffer.js

    } else if (value.length === 1) {
      // Fast path: If `value` fits into a single byte, use that numeric value.
-      if (normalizedEncoding === 'utf8') {
+      if (normalizedEncoding === 'utf8' || normalizedEncoding === 'ascii') {


Currently, ascii behaves exactly like latin1
Unsure if by design or accidentally

yes, this is safe by design I am just extending the existing single byte numeric optimization to cover ASCII, since the guard already constrains it to the valid ASCII range.

thisalihassan · 2026-02-18T07:21:23Z

Note on toBase64 / toBase64url:

I also tried replacing the C++ base64Slice/base64urlSlice bindings with V8's Uint8Array.prototype.toBase64() (similar to the toHex change) but it caused a 35-54% regression across all buffer sizes so I reverted base64/base64url and kept only the toHex optimization which showed a clear +26-37% win.

ChALkeR · 2026-02-18T07:31:15Z

@thisalihassan toHex doesn't show a win anymore with nbytes update which should soon land (as it landed in nbytes)

Instead, it's ~3x slower.

See nodejs/nbytes#12

thisalihassan · 2026-02-18T08:06:14Z

Hi @ChALkeR thanks for flagging I was not aware. I benchmarked the nibble approach locally and it's indeed a much bigger win (~3x vs my ~30% with toHex). Reverted the toHex path entirely the other changes in this PR are unaffected.

Should I include the nbytes nibble HexEncode optimization in this PR or keep them as separate PRs?

~~PS: One test is failing /test/parallel/test-debugger-restart-message.js I believe it's known mac issue and unrelated to my changes~~

lib/buffer.js

src/node_buffer.cc

test/parallel/test-buffer-swap-fast.js

lib/buffer.js

thisalihassan · 2026-02-22T22:26:29Z

For toHex, wait until #61609, which improves native perf significantly (more than Uint8Array.prototype.toHex)

See also #60249 (comment)

Hi @ChALkeR I see that your changes landed, congratualtions on that huge win. Is there benchmark-ci that we can run on this PR? I can rebase it

thisalihassan · 2026-02-24T15:36:51Z

I re-ran the benchmark against the latest main (which contains the nibble improvement) and found out there are few regressions caused by my code:

buffers/buffer-indexof.js n=50000 type='buffer' encoding='undefined' search='aaaaaaaaaaaaaaaaa' *** -9.41 %
buffers/buffer-indexof.js n=50000 type='string' encoding='ascii' search='aaaaaaaaaaaaaaaaa' *** -10.93
buffers/buffer-indexof.js n=50000 type='string' encoding='latin1' search='aaaaaaaaaaaaaaaaa' *** -9.44 %
buffers/buffer-indexof.js n=50000 type='string' encoding='utf8' search='aaaaaaaaaaaaaaaaa' *** -9.18 %

I tried fixing this regression but unavoidable

Attaching Details below:
detail.pdf -- visual breakdown

buffer-benchmark-all-rebased.csv -- raw benchmark CSV

thisalihassan · 2026-02-24T19:54:41Z

compare-R-output.txt-- full output from Rscript benchmark/compare.R (shared as a file to avoid cluttering the PR comment).

thisalihassan · 2026-02-27T23:11:00Z

PS: I resolved some conflicts

thisalihassan · 2026-02-28T21:31:13Z

Hi @ChALkeR @anonrig @Renegade334 is this PR ready to land? can you also please check the latest benchmarks i have posted, I have compiled PDF for this benchmark in one my comments above for more detail

Qard

Generally LGTM, but one small nit.

src/node_buffer.cc

lib/buffer.js

aduh95 · 2026-03-25T11:40:50Z

Can you rename the other const len = this.length / const len = TypedArrayPrototypeGetLength(this) to bufferLength please?

thisalihassan · 2026-03-25T11:55:12Z

Can you rename the other const len = this.length / const len = TypedArrayPrototypeGetLength(this) to bufferLength please?

Done! Hoping this would be the last of the changes :')

Another reference spotted @aduh95

Buffer.prototype.subarray = function subarray(start, end) {
  const srcLength = this.length;

should also be?

Buffer.prototype.subarray = function subarray(start, end) {
  const srcLength = TypedArrayPrototypeGetLength(this);

nodejs-github-bot · 2026-03-27T09:30:20Z

CI: https://ci.nodejs.org/job/node-test-pull-request/72069/

thisalihassan · 2026-03-27T12:41:32Z

this test-runner.test-run-watch-run-duration is the only one failing on the CI

nodejs-github-bot · 2026-03-27T13:02:32Z

CI: https://ci.nodejs.org/job/node-test-pull-request/72079/

nodejs-github-bot · 2026-03-27T15:18:48Z

CI: https://ci.nodejs.org/job/node-test-pull-request/72094/

nodejs-github-bot · 2026-03-27T21:50:59Z

Landed in 9e0dc8b

- copyBytesFrom: calculate byte offsets directly instead of slicing into an intermediate typed array - toString('hex'): use V8 Uint8Array.prototype.toHex() builtin - fill: add single-char ASCII fast path - indexOf: use indexOfString directly for ASCII encoding - swap16/32/64: add V8 Fast API functions PR-URL: #61871 Reviewed-By: Stephen Belanger <admin@stephenbelanger.com> Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Yagiz Nizipli <yagiz@nizipli.com> Reviewed-By: James M Snell <jasnell@gmail.com>

nodejs-github-bot added buffer Issues and PRs related to the buffer subsystem. c++ Issues and PRs that require attention from people who are familiar with C++. needs-ci PRs that need a full CI run. labels Feb 17, 2026

thisalihassan force-pushed the buffer-perf-improvements branch from d2ba38f to 495feb5 Compare February 17, 2026 21:41

Renegade334 reviewed Feb 17, 2026

View reviewed changes

Renegade334 added performance Issues and PRs related to the performance of Node.js. needs-benchmark-ci PR that need a benchmark CI run. labels Feb 17, 2026

thisalihassan force-pushed the buffer-perf-improvements branch 4 times, most recently from 1395d2f to 01ba74f Compare February 17, 2026 23:42

Renegade334 reviewed Feb 18, 2026

View reviewed changes

ChALkeR reviewed Feb 18, 2026

View reviewed changes

thisalihassan requested review from ChALkeR and Renegade334 February 18, 2026 08:10

anonrig reviewed Feb 18, 2026

View reviewed changes

thisalihassan requested a review from anonrig February 18, 2026 21:42

thisalihassan force-pushed the buffer-perf-improvements branch from 59f5d09 to 48abfc5 Compare February 27, 2026 23:09

Qard approved these changes Mar 3, 2026

View reviewed changes

src/node_buffer.cc Show resolved Hide resolved

thisalihassan force-pushed the buffer-perf-improvements branch from 1d16ad4 to f1d938b Compare March 21, 2026 04:33

jasnell reviewed Mar 22, 2026

View reviewed changes

lib/buffer.js Show resolved Hide resolved

jasnell reviewed Mar 22, 2026

View reviewed changes

lib/buffer.js Show resolved Hide resolved

jasnell reviewed Mar 22, 2026

View reviewed changes

lib/buffer.js Show resolved Hide resolved

improve variable names and add Refs

7c8b68d

thisalihassan force-pushed the buffer-perf-improvements branch from f1d938b to 7c8b68d Compare March 25, 2026 09:52

aduh95 reviewed Mar 25, 2026

View reviewed changes

lib/buffer.js Outdated Show resolved Hide resolved

aduh95 reviewed Mar 25, 2026

View reviewed changes

lib/buffer.js Outdated Show resolved Hide resolved

used primordial getter for safety

85fb7f1

thisalihassan force-pushed the buffer-perf-improvements branch from c49a144 to 85fb7f1 Compare March 25, 2026 12:31

thisalihassan requested review from ChALkeR, anonrig and jasnell March 27, 2026 00:55

aduh95 approved these changes Mar 27, 2026

View reviewed changes

anonrig approved these changes Mar 27, 2026

View reviewed changes

aduh95 added commit-queue Add this label to land a pull request using GitHub Actions. commit-queue-squash Add this label to instruct the Commit Queue to squash all the PR commits into the first one. labels Mar 27, 2026

jasnell approved these changes Mar 27, 2026

View reviewed changes

nodejs-github-bot removed the commit-queue Add this label to land a pull request using GitHub Actions. label Mar 27, 2026

nodejs-github-bot merged commit 9e0dc8b into nodejs:main Mar 27, 2026
70 of 71 checks passed

Uh oh!

Conversation

thisalihassan commented Feb 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Benchmark results

Uh oh!

nodejs-github-bot commented Feb 17, 2026

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

ChALkeR Mar 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

codecov bot commented Feb 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

ChALkeR left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

thisalihassan Feb 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

thisalihassan commented Feb 18, 2026

Uh oh!

ChALkeR commented Feb 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

thisalihassan commented Feb 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

thisalihassan commented Feb 22, 2026

Uh oh!

thisalihassan commented Feb 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

thisalihassan commented Feb 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

thisalihassan commented Feb 27, 2026

Uh oh!

thisalihassan commented Feb 28, 2026

Uh oh!

Qard left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

aduh95 commented Mar 25, 2026

thisalihassan commented Feb 17, 2026 •

edited

Loading

ChALkeR Mar 5, 2026 •

edited

Loading

codecov bot commented Feb 17, 2026 •

edited

Loading

ChALkeR left a comment •

edited

Loading

thisalihassan Feb 18, 2026 •

edited

Loading

ChALkeR commented Feb 18, 2026 •

edited

Loading

thisalihassan commented Feb 18, 2026 •

edited

Loading

thisalihassan commented Feb 24, 2026 •

edited

Loading

thisalihassan commented Feb 24, 2026 •

edited

Loading

thisalihassan commented Mar 25, 2026 •

edited

Loading