factor:performance tuning #9261

mattsu2020 · 2025-11-13T14:19:09Z

Performance improvement for large numbers

fix this issue
https://bugs.launchpad.net/ubuntu/+source/rust-coreutils/+bug/2131212

github-actions · 2025-11-13T14:40:22Z

GNU testsuite comparison:

Skip an intermittent issue tests/misc/tee (fails in this run but passes in the 'main' branch)
Skip an intermittent issue tests/tail/overlay-headers (fails in this run but passes in the 'main' branch)

src/uu/factor/src/factor.rs

github-actions · 2025-11-14T00:02:49Z

GNU testsuite comparison:

Skipping an intermittent issue tests/tail/overlay-headers (passes in this run but fails in the 'main' branch)

github-actions · 2025-11-14T03:40:14Z

GNU testsuite comparison:

Skipping an intermittent issue tests/tail/overlay-headers (passes in this run but fails in the 'main' branch)

sylvestre · 2025-11-14T08:16:13Z

could you please run hyperfine with the three programs? gnu, without the patch and with the patch
and share the full results here? thanks :)

github-actions · 2025-11-14T10:40:24Z

GNU testsuite comparison:

Skip an intermittent issue tests/tail/overlay-headers (fails in this run but passes in the 'main' branch)

mattsu2020 · 2025-11-14T11:13:07Z

could you please run hyperfine with the three programs? gnu, without the patch and with the patch and share the full results here? thanks :)

Implementation Details

GMP 6.3.0 and GNU coreutils 9.5 were built and installed from source

Created factor_numbers_u128_repeat.txt (60 lines) as benchmark input, containing 6 composite numbers ranging from 64 to 128 bits repeated 10 times. Confirmed factorization completion across all 3 implementations and reran Hyperfine.
All commands used the release profile (target/profiling/factor).
Hyperfine execution results
Command: hyperfine --warmup 3 --runs 12 “ < factor_numbers_u128_repeat.txt”

Implementation	Average time (s)	Standard deviation (s)	Minimum–Maximum (s)
GNU coreutils 9.5 ( local-gnu/bin/factor)	6.718	0.106	6.594 – 7.020
Old implementation (prev_worktree/target/profiling/factor)	6.125	1.942	2.648 – 8.508
After patch application (target/profiling/factor)	6.993	1.585	4.299 – 9.457

To reduce variance, we adjusted to 3 warm-ups + 12 measurements, but the Rust version still shows relatively high dispersion due to its randomized algorithm. For greater stability, consider running at times of low system load or using CPU pinning.
Behavior with inputs exceeding 128 bits

For factor_numbers.txt (max ~260 bits), both the GNU version and the patched version achieved complete factorization. The old implementation returned factor: Factorization incomplete. Remainders exist. and exited with exit code 1. This confirms the improvement in support for large integers.
factor_numbers_u128_repeat.txt

sylvestre · 2025-11-15T12:52:34Z

src/uu/factor/src/factor.rs

+        return true;
+    }
+    // even check: candidate % 2 == 0
+    if (candidate & BigUint::from_u32(1).unwrap()).is_zero() {


maybe create a function is_even

sylvestre · 2025-11-15T12:52:53Z

src/uu/factor/src/factor.rs

+    let mut odd_component = candidate - &one;
+    let mut power_of_two = 0u32;
+    // while odd_component is even
+    while (&odd_component & BigUint::from_u32(1).unwrap()).is_zero() {


esp as it is done here too

sylvestre · 2025-11-15T12:53:22Z

src/uu/factor/src/factor.rs

+    // Use a deterministic LCG to generate parameter sequences.
+    fn lcg_next(x: &mut u128) {
+        *x = x
+            .wrapping_mul(6364136223846793005)


please move this magic number into a variable

sylvestre · 2025-11-15T12:53:28Z

src/uu/factor/src/factor.rs

+    fn lcg_next(x: &mut u128) {
+        *x = x
+            .wrapping_mul(6364136223846793005)
+            .wrapping_add(1442695040888963407);


sylvestre · 2025-11-15T12:53:53Z

src/uu/factor/src/factor.rs

+    // Search parameters: choose bounds based on bit length.
+    // Avoid overly large limits; when exhausted, treat as failure to find a factor.
+    let max_tries: u64 = 16;
+    let max_iter: u64 = (bits * bits).clamp(10_000, 200_000);


why these values ?

also, could this overflow ?

We're setting this number for now as we fine-tune and determine the value.
Since the maximum number of times is set, it will stop.

sylvestre · 2025-11-15T12:54:03Z

src/uu/factor/src/factor.rs

+    let max_tries: u64 = 16;
+    let max_iter: u64 = (bits * bits).clamp(10_000, 200_000);
+
+    let mut seed: u128 = 0x9e3779b97f4a7c15;


please, add comment explain what it is

sylvestre · 2025-11-15T12:54:44Z

src/uu/factor/src/factor.rs

+
+        while current_gcd == one && iter < max_iter {
+            // Brent variant: use batched gcd.
+            let mut inner_iter = 0;


please rename this variable for something more meaningful
like
batch_iter

sylvestre · 2025-11-15T12:55:14Z

src/uu/factor/src/factor.rs

+
+    // If n is small enough, use num_prime's factorize128 for speed.
+    if n.bits() <= 128 {
+        if let Ok(x128) = n.to_string().parse::<u128>() {


maybe investigate using a BigUint function directly here

codspeed-hq · 2025-11-16T14:04:59Z

CodSpeed Performance Report

Merging #9261 will improve performance by 19.21%

_{Comparing mattsu2020:factor_fix (c0333f3) with main (502f3b1)}

Summary

⚡ 1 improvement
✅ 126 untouched
⏩ 6 skipped¹

Benchmarks breakdown

	Benchmark	`BASE`	`HEAD`	Efficiency
⚡	`factor_multiple_u64s[2]`	212.4 ms	178.2 ms	+19.21%

6 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

github-actions · 2025-11-16T14:40:36Z

GNU testsuite comparison:

Skipping an intermittent issue tests/tail/overlay-headers (passes in this run but fails in the 'main' branch)

github-actions · 2025-11-17T09:52:38Z

GNU testsuite comparison:

Skipping an intermittent issue tests/misc/tee (passes in this run but fails in the 'main' branch)

github-actions · 2025-12-01T12:31:12Z

GNU testsuite comparison:

Skip an intermittent issue tests/tail/overlay-headers (fails in this run but passes in the 'main' branch)

sylvestre · 2025-12-07T13:53:19Z

Any idea why codspeed does not detect it?

mattsu2020 · 2025-12-07T14:01:10Z

Any idea why codspeed does not detect it?

If I were to consider it, I would create test cases with large integers.

github-actions · 2025-12-07T14:05:49Z

GNU testsuite comparison:

Skip an intermittent issue tests/tail/overlay-headers (fails in this run but passes in the 'main' branch)

github-actions · 2025-12-08T10:39:41Z

GNU testsuite comparison:

Skip an intermittent issue tests/tail/overlay-headers (fails in this run but passes in the 'main' branch)

- Add num-integer dependency to support enhanced numeric operations. - Refactor factorization logic to avoid redundant parsing and optimize u64/u128 paths. - Improve handling of non-positive and invalid inputs to align with GNU factor behavior. - Enhance large BigUint factoring with additional algorithms and clearer limitations.

- Integrate jemalloc allocator in factor benchmark suite for better memory profiling - Add jemalloc-ctl and jemallocator dependencies with OS-specific dev-dependencies - Implement logging of allocated and resident memory stats before benchmark runs - Update CI workflow to show output for uu_factor benchmarks without suppressing it - Enables precise memory usage tracking on Linux, macOS, and FreeBSD during benchmarking

Add technical terms for memory allocation libraries to the cspell dictionary to prevent false positives in spellchecking.

github-actions · 2025-12-24T23:38:16Z

GNU testsuite comparison:

Congrats! The gnu test tests/tail/follow-name is no longer failing!

sylvestre reviewed Nov 13, 2025

View reviewed changes

src/uu/factor/src/factor.rs Outdated Show resolved Hide resolved

sylvestre reviewed Nov 15, 2025

View reviewed changes

sylvestre force-pushed the factor_fix branch from 09d5c51 to fe34cf2 Compare November 17, 2025 09:21

mattsu2020 added 3 commits December 25, 2025 08:03

refactor(factor): readability and small perf tweaks

2bde8f7

docs(factor): translate comments and note spellchecker

aea625a

mattsu2020 force-pushed the factor_fix branch from 0dfc79b to aea625a Compare December 24, 2025 23:10

mattsu2020 added 2 commits December 25, 2025 08:19

chore(cspell): add jemalloc and jemallocator to jargon wordlist

c0333f3

Add technical terms for memory allocation libraries to the cspell dictionary to prevent false positives in spellchecking.

Uh oh!

factor:performance tuning #9261

Are you sure you want to change the base?

factor:performance tuning #9261

Conversation

mattsu2020 commented Nov 13, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Nov 13, 2025

Uh oh!

Uh oh!

github-actions bot commented Nov 14, 2025

Uh oh!

github-actions bot commented Nov 14, 2025

Uh oh!

sylvestre commented Nov 14, 2025

Uh oh!

github-actions bot commented Nov 14, 2025

Uh oh!

mattsu2020 commented Nov 14, 2025

Implementation Details

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!

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!

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!

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!

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!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

codspeed-hq bot commented Nov 16, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Merging #9261 will improve performance by 19.21%

Summary

Benchmarks breakdown

Footnotes

Uh oh!

github-actions bot commented Nov 16, 2025

Uh oh!

github-actions bot commented Nov 17, 2025

Uh oh!

github-actions bot commented Dec 1, 2025

Uh oh!

sylvestre commented Dec 7, 2025

Uh oh!

mattsu2020 commented Dec 7, 2025

Uh oh!

github-actions bot commented Dec 7, 2025

Uh oh!

github-actions bot commented Dec 8, 2025

Uh oh!

github-actions bot commented Dec 24, 2025

Uh oh!

Reviewers

mattsu2020 commented Nov 13, 2025 •

edited

Loading

codspeed-hq bot commented Nov 16, 2025 •

edited

Loading