cksum: symlink to md5sum,b2sum, and sha*sum

[oech3]

Alt #9776 . but without individual bins.

[codspeed]

CodSpeed Performance Report

Merging #9984 will improve performance by ×8,800

_{Comparing oech3:cksum-ln (a4c2c4f) with main (a4f6e1a)}

Summary

⚡ 14 improvements
✅ 125 untouched
⏩ 37 skipped¹

Benchmarks breakdown

	Benchmark	BASE	HEAD	Efficiency
⚡	cksum_blake2b	192,851.4 µs	208.6 µs	×920
⚡	cksum_sha2	1,727,099.3 µs	206.1 µs	×8,400
⚡	cksum_crc	16,990.8 µs	205.9 µs	×83
⚡	cksum_sha1	723,076.3 µs	206.1 µs	×3,500
⚡	cksum_md5	318,011.7 µs	206.2 µs	×1,500
⚡	cksum_sha3	1,828,114.8 µs	206.9 µs	×8,800
⚡	cksum_sha512	1,105,575.8 µs	207.2 µs	×5,300
⚡	cksum_sha256	1,727,122.1 µs	207.8 µs	×8,300
⚡	cksum_bsd	128,397.7 µs	206.2 µs	×620
⚡	cksum_sha384	1,105,538.7 µs	207.7 µs	×5,300
⚡	cksum_sha224	1,727,116.3 µs	207 µs	×8,300
⚡	cksum_crc32b	13,245.3 µs	207.2 µs	×64
⚡	cksum_sysv	48,258.8 µs	206.5 µs	×230
⚡	cksum_sm3	1,551,347.7 µs	205.8 µs	×7,500

Footnotes

1. 37 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. ↩

[oech3]

I considered that CodSpeed was broken, but it seems reproducible...

[oech3]

cksum related GNU tests are passing.

[oech3]

@naoNao89 Any idea about CodSpeed breakage?

[oech3]

We should fix or remove #10002 before discarding hashsum.

[naoNao89]

The file creation happens outside the benchmark loop, so the benchmarks are measuring almost nothing.

uu/cksum/benches/cksum_bench.rs:14-24

fn $algo_name(bencher: Bencher) {
    let data = text_data::generate_by_size(100, 80);
    let file_path = setup_test_file(&data);  // File created ONCE, outside loop
    
    bencher.bench(|| {
        black_box(run_util_function(
            uumain,
            &["--algorithm", $algo_str, file_path.to_str().unwrap()],
        ));
    });  // This loop only measures reading a cached file
}

I think we should:

• generate fresh data inside the benchmark loop, or
• clear file system caches between iterations, or
• benchmark the hash computation directly without file I/O

[github-actions]

GNU testsuite comparison:

GNU test failed: tests/cksum/md5sum-bsd. tests/cksum/md5sum-bsd is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/shuf/shuf-reservoir. tests/shuf/shuf-reservoir is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/sort/sort-stale-thread-mem. tests/sort/sort-stale-thread-mem is passing on 'main'. Maybe you have to rebase?
Skip an intermittent issue tests/timeout/timeout (fails in this run but passes in the 'main' branch)

[github-actions]

GNU testsuite comparison:

GNU test failed: tests/cksum/cksum-a. tests/cksum/cksum-a is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/cksum/md5sum-bsd. tests/cksum/md5sum-bsd is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/tty/tty-eof. tests/tty/tty-eof is passing on 'main'. Maybe you have to rebase?
Skip an intermittent issue tests/timeout/timeout (fails in this run but passes in the 'main' branch)

[github-actions]

GNU testsuite comparison:

GNU test failed: tests/cksum/cksum-a. tests/cksum/cksum-a is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/cksum/md5sum-bsd. tests/cksum/md5sum-bsd is passing on 'main'. Maybe you have to rebase?
Skip an intermittent issue tests/timeout/timeout (fails in this run but passes in the 'main' branch)

[github-actions]

GNU testsuite comparison:

GNU test failed: tests/cksum/cksum-a. tests/cksum/cksum-a is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/cksum/md5sum-bsd. tests/cksum/md5sum-bsd is passing on 'main'. Maybe you have to rebase?
Skip an intermittent issue tests/timeout/timeout (fails in this run but passes in the 'main' branch)

[github-actions]

GNU testsuite comparison:

GNU test failed: tests/cksum/md5sum-bsd. tests/cksum/md5sum-bsd is passing on 'main'. Maybe you have to rebase?
Skip an intermittent issue tests/timeout/timeout (fails in this run but passes in the 'main' branch)

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/timeout/timeout (fails in this run but passes in the 'main' branch)

[oech3]

*sums are really bad legacy program :|

[github-actions]

GNU testsuite comparison:

GNU test failed: tests/cksum/md5sum-bsd. tests/cksum/md5sum-bsd is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/shuf/shuf-reservoir. tests/shuf/shuf-reservoir is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/sort/sort-stale-thread-mem. tests/sort/sort-stale-thread-mem is passing on 'main'. Maybe you have to rebase?
Skip an intermittent issue tests/timeout/timeout (fails in this run but passes in the 'main' branch)

[github-actions]

GNU testsuite comparison:

GNU test failed: tests/cksum/b2sum. tests/cksum/b2sum is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/cksum/md5sum-bsd. tests/cksum/md5sum-bsd is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/cksum/md5sum-newline. tests/cksum/md5sum-newline is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/cksum/sha1sum. tests/cksum/sha1sum is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/cksum/sha1sum-vec. tests/cksum/sha1sum-vec is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/cksum/sha224sum. tests/cksum/sha224sum is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/cksum/sha256sum. tests/cksum/sha256sum is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/cksum/sha384sum. tests/cksum/sha384sum is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/cksum/sha512sum. tests/cksum/sha512sum is passing on 'main'. Maybe you have to rebase?
Skip an intermittent issue tests/timeout/timeout (fails in this run but passes in the 'main' branch)

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/timeout/timeout (fails in this run but passes in the 'main' branch)

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/timeout/timeout (fails in this run but passes in the 'main' branch)

[oech3]

@cakebaker @RenjiSann I made a symlink plan. I think we can make individual bins after hashsum deletion.
Should fix #10002 at 1st at least, and I'd like to merge clap related PRs to reduce size of a PR at both case (1 bin or N bins).

Failing test is uptime.

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/timeout/timeout (fails in this run but passes in the 'main' branch)

[oech3]

This looks very dangerous option if we don't have an extension --untagged on md5sum. Closing.