div_mod_knuth added

[mcseemk]

Added div_mod_knuth division for two large >128bit numbers. Resolves #16.

[nlordell]

Wow! This is awesome!

Would you be able to run the benchmarks before and after this change to get an idea for the speed improvement?

Also, I will be busy with personal things for couple of weeks, so don't be surprised if I am a little quiet on this PR.

[mcseemk]

Sure,
bench before:

U256::add time: [1.6429 ns 1.6462 ns 1.6499 ns]
Found 15 outliers among 100 measurements (15.00%)
7 (7.00%) low mild
7 (7.00%) high mild
1 (1.00%) high severe

U256::div (lo/lo) time: [8.1225 ns 8.1331 ns 8.1452 ns]
Found 15 outliers among 100 measurements (15.00%)
5 (5.00%) high mild
10 (10.00%) high severe

U256::div (hi/lo) time: [19.067 ns 19.083 ns 19.100 ns]
Found 6 outliers among 100 measurements (6.00%)
2 (2.00%) low mild
4 (4.00%) high mild

U256::div (hi/hi) time: [171.46 ns 172.14 ns 172.87 ns]

U256::mul time: [2.6050 ns 2.6064 ns 2.6078 ns]
Found 9 outliers among 100 measurements (9.00%)
1 (1.00%) low severe
1 (1.00%) low mild
5 (5.00%) high mild
2 (2.00%) high severe

U256::sub time: [1.6185 ns 1.6207 ns 1.6232 ns]
Found 5 outliers among 100 measurements (5.00%)
4 (4.00%) high mild
1 (1.00%) high severe

U256::shl time: [1.9257 ns 1.9313 ns 1.9371 ns]
Found 6 outliers among 100 measurements (6.00%)
4 (4.00%) high mild
2 (2.00%) high severe

U256::shr time: [1.7386 ns 1.7402 ns 1.7424 ns]
Found 17 outliers among 100 measurements (17.00%)
4 (4.00%) high mild
13 (13.00%) high severe

U256::ctlz time: [798.04 ps 798.59 ps 799.28 ps]
Found 13 outliers among 100 measurements (13.00%)
2 (2.00%) low mild
3 (3.00%) high mild
8 (8.00%) high severe

U256::cttz time: [881.57 ps 882.25 ps 883.15 ps]
Found 8 outliers among 100 measurements (8.00%)
4 (4.00%) high mild
4 (4.00%) high severe

U256::rotate_left time: [2.9088 ns 2.9195 ns 2.9321 ns]
Found 16 outliers among 100 measurements (16.00%)
2 (2.00%) high mild
14 (14.00%) high severe

U256::rotate_right time: [2.7154 ns 2.7197 ns 2.7249 ns]
Found 14 outliers among 100 measurements (14.00%)
2 (2.00%) low mild
2 (2.00%) high mild
10 (10.00%) high severe

and after:

U256::add time: [1.6201 ns 1.6232 ns 1.6267 ns]
change: [-1.5315% -1.1987% -0.8861%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 4 outliers among 100 measurements (4.00%)
4 (4.00%) high mild

U256::div (lo/lo) time: [8.1195 ns 8.1333 ns 8.1489 ns]
change: [-0.2028% -0.0507% +0.1022%] (p = 0.52 > 0.05)
No change in performance detected.
Found 12 outliers among 100 measurements (12.00%)
5 (5.00%) high mild
7 (7.00%) high severe

U256::div (hi/lo) time: [19.281 ns 19.295 ns 19.309 ns]
change: [+1.0802% +1.2256% +1.3774%] (p = 0.00 < 0.05)
Performance has regressed.
Found 8 outliers among 100 measurements (8.00%)
4 (4.00%) high mild
4 (4.00%) high severe

U256::div (hi/hi) time: [24.055 ns 24.095 ns 24.141 ns]
change: [-86.074% -86.006% -85.935%] (p = 0.00 < 0.05)
Performance has improved.
Found 4 outliers among 100 measurements (4.00%)
1 (1.00%) low mild
1 (1.00%) high mild
2 (2.00%) high severe

U256::mul time: [2.6083 ns 2.6114 ns 2.6154 ns]
change: [+0.3169% +0.7328% +1.2796%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 6 outliers among 100 measurements (6.00%)
4 (4.00%) high mild
2 (2.00%) high severe

U256::sub time: [1.6206 ns 1.6233 ns 1.6262 ns]
change: [-0.0591% +0.1864% +0.4386%] (p = 0.15 > 0.05)
No change in performance detected.
Found 2 outliers among 100 measurements (2.00%)
1 (1.00%) high mild
1 (1.00%) high severe

U256::shl time: [1.9175 ns 1.9225 ns 1.9283 ns]
change: [-0.6044% -0.1814% +0.2108%] (p = 0.39 > 0.05)
No change in performance detected.
Found 6 outliers among 100 measurements (6.00%)
2 (2.00%) high mild
4 (4.00%) high severe

U256::shr time: [1.7385 ns 1.7405 ns 1.7439 ns]
change: [-0.4575% -0.1798% +0.0966%] (p = 0.21 > 0.05)
No change in performance detected.
Found 11 outliers among 100 measurements (11.00%)
3 (3.00%) high mild
8 (8.00%) high severe

U256::ctlz time: [799.83 ps 801.46 ps 803.33 ps]
change: [+0.8143% +1.2245% +1.6966%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 4 outliers among 100 measurements (4.00%)
1 (1.00%) high mild
3 (3.00%) high severe

U256::cttz time: [884.29 ps 885.95 ps 887.80 ps]
change: [+0.1541% +0.3471% +0.5903%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 7 outliers among 100 measurements (7.00%)
6 (6.00%) high mild
1 (1.00%) high severe

U256::rotate_left time: [2.9030 ns 2.9095 ns 2.9165 ns]
change: [-0.4245% -0.1878% +0.0451%] (p = 0.13 > 0.05)
No change in performance detected.
Found 9 outliers among 100 measurements (9.00%)
9 (9.00%) high severe

U256::rotate_right time: [2.7006 ns 2.7032 ns 2.7068 ns]
change: [-0.6416% -0.4652% -0.3043%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 12 outliers among 100 measurements (12.00%)
5 (5.00%) high mild
7 (7.00%) high severe

[mcseemk]

Previous post was on my desktop PC Intel i12900K. Below is the Graviton 3 (c7g amazon aws instance) bench:
Before:
U256::add time: [5.4664 ns 5.4688 ns 5.4718 ns]
Found 4 outliers among 100 measurements (4.00%)
1 (1.00%) low severe
2 (2.00%) low mild
1 (1.00%) high severe

U256::div (lo/lo) time: [20.475 ns 20.477 ns 20.479 ns]
Found 19 outliers among 100 measurements (19.00%)
1 (1.00%) low severe
10 (10.00%) low mild
7 (7.00%) high mild
1 (1.00%) high severe

U256::div (hi/lo) time: [94.612 ns 94.762 ns 94.916 ns]

U256::div (hi/hi) time: [336.59 ns 336.61 ns 336.63 ns]
Found 7 outliers among 100 measurements (7.00%)
1 (1.00%) low severe
5 (5.00%) high mild
1 (1.00%) high severe

U256::mul time: [5.5886 ns 5.5908 ns 5.5931 ns]
Found 4 outliers among 100 measurements (4.00%)
1 (1.00%) low severe
1 (1.00%) low mild
2 (2.00%) high mild

U256::sub time: [5.4582 ns 5.4601 ns 5.4619 ns]
Found 8 outliers among 100 measurements (8.00%)
1 (1.00%) low severe
3 (3.00%) low mild
4 (4.00%) high mild

U256::shl time: [4.2159 ns 4.2164 ns 4.2169 ns]
Found 5 outliers among 100 measurements (5.00%)
1 (1.00%) low mild
4 (4.00%) high severe

U256::shr time: [4.1342 ns 4.1354 ns 4.1369 ns]
Found 13 outliers among 100 measurements (13.00%)
7 (7.00%) high mild
6 (6.00%) high severe

U256::ctlz time: [2.4419 ns 2.4420 ns 2.4420 ns]
Found 4 outliers among 100 measurements (4.00%)
2 (2.00%) high mild
2 (2.00%) high severe

U256::cttz time: [2.4857 ns 2.4858 ns 2.4858 ns]
Found 2 outliers among 100 measurements (2.00%)
1 (1.00%) high mild
1 (1.00%) high severe

U256::rotate_left time: [5.7390 ns 5.7577 ns 5.7763 ns]
Found 4 outliers among 100 measurements (4.00%)
1 (1.00%) low mild
2 (2.00%) high mild
1 (1.00%) high severe

U256::rotate_right time: [5.5316 ns 5.5386 ns 5.5461 ns]
Found 3 outliers among 100 measurements (3.00%)
1 (1.00%) high mild
2 (2.00%) high severe

After:

U256::add time: [5.4514 ns 5.4521 ns 5.4529 ns]
change: [-0.3786% -0.3263% -0.2762%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 9 outliers among 100 measurements (9.00%)
4 (4.00%) low severe
4 (4.00%) low mild
1 (1.00%) high severe

U256::div (lo/lo) time: [20.474 ns 20.480 ns 20.485 ns]
change: [-0.0804% -0.0553% -0.0291%] (p = 0.00 < 0.05)
Change within noise threshold.

U256::div (hi/lo) time: [94.189 ns 94.202 ns 94.215 ns]
change: [-0.7320% -0.6301% -0.5264%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 3 outliers among 100 measurements (3.00%)
1 (1.00%) low mild
2 (2.00%) high mild

U256::div (hi/hi) time: [68.181 ns 68.192 ns 68.204 ns]
change: [-79.746% -79.742% -79.738%] (p = 0.00 < 0.05)
Performance has improved.
Found 5 outliers among 100 measurements (5.00%)
3 (3.00%) low mild
1 (1.00%) high mild
1 (1.00%) high severe

U256::mul time: [5.5856 ns 5.5879 ns 5.5902 ns]
change: [-0.1402% -0.0720% -0.0009%] (p = 0.05 < 0.05)
Change within noise threshold.
Found 6 outliers among 100 measurements (6.00%)
5 (5.00%) low mild
1 (1.00%) high mild

U256::sub time: [5.4708 ns 5.4722 ns 5.4736 ns]
change: [+0.2187% +0.2696% +0.3182%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 4 outliers among 100 measurements (4.00%)
2 (2.00%) low severe
2 (2.00%) low mild

U256::shl time: [4.2157 ns 4.2162 ns 4.2167 ns]
change: [-0.0453% -0.0197% +0.0044%] (p = 0.11 > 0.05)
No change in performance detected.
Found 6 outliers among 100 measurements (6.00%)
1 (1.00%) low severe
2 (2.00%) low mild
3 (3.00%) high mild

U256::shr time: [4.1333 ns 4.1340 ns 4.1348 ns]
change: [-0.0832% -0.0447% -0.0078%] (p = 0.02 < 0.05)
Change within noise threshold.
Found 6 outliers among 100 measurements (6.00%)
3 (3.00%) low mild
3 (3.00%) high severe

U256::ctlz time: [2.4419 ns 2.4420 ns 2.4421 ns]
change: [-0.0089% -0.0009% +0.0079%] (p = 0.84 > 0.05)
No change in performance detected.
Found 8 outliers among 100 measurements (8.00%)
1 (1.00%) low severe
1 (1.00%) low mild
5 (5.00%) high mild
1 (1.00%) high severe

U256::cttz time: [2.4824 ns 2.4834 ns 2.4842 ns]
change: [-0.2846% -0.2477% -0.2093%] (p = 0.00 < 0.05)
Change within noise threshold.

U256::rotate_left time: [5.7638 ns 5.7770 ns 5.7908 ns]
change: [+0.3731% +0.9748% +1.6579%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 11 outliers among 100 measurements (11.00%)
4 (4.00%) low mild
3 (3.00%) high mild
4 (4.00%) high severe

U256::rotate_right time: [5.5470 ns 5.5548 ns 5.5626 ns]
change: [-0.1547% +0.0932% +0.3309%] (p = 0.46 > 0.05)
No change in performance detected.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high mild

[mcseemk]

Graviton 2 (c6g* amazon aws):
Before:

U256::add time: [8.3676 ns 8.3697 ns 8.3718 ns]
Found 7 outliers among 100 measurements (7.00%)
4 (4.00%) high mild
3 (3.00%) high severe

U256::div (lo/lo) time: [31.313 ns 31.330 ns 31.345 ns]
Found 3 outliers among 100 measurements (3.00%)
1 (1.00%) low mild
1 (1.00%) high mild
1 (1.00%) high severe

U256::div (hi/lo) time: [142.58 ns 142.62 ns 142.66 ns]
Found 6 outliers among 100 measurements (6.00%)
2 (2.00%) low mild
3 (3.00%) high mild
1 (1.00%) high severe

U256::div (hi/hi) time: [432.31 ns 432.42 ns 432.55 ns]
Found 8 outliers among 100 measurements (8.00%)
5 (5.00%) high mild
3 (3.00%) high severe

U256::mul time: [21.649 ns 21.654 ns 21.661 ns]
Found 6 outliers among 100 measurements (6.00%)
5 (5.00%) high mild
1 (1.00%) high severe

U256::sub time: [8.3902 ns 8.3923 ns 8.3945 ns]
Found 3 outliers among 100 measurements (3.00%)
3 (3.00%) high mild

U256::shl time: [5.9219 ns 5.9233 ns 5.9248 ns]
Found 9 outliers among 100 measurements (9.00%)
7 (7.00%) high mild
2 (2.00%) high severe

U256::shr time: [5.8867 ns 5.8869 ns 5.8871 ns]
Found 4 outliers among 100 measurements (4.00%)
2 (2.00%) low mild
2 (2.00%) high severe

U256::ctlz time: [3.2917 ns 3.2931 ns 3.2947 ns]
Found 12 outliers among 100 measurements (12.00%)
10 (10.00%) high mild
2 (2.00%) high severe

U256::cttz time: [3.2852 ns 3.2862 ns 3.2872 ns]
Found 8 outliers among 100 measurements (8.00%)
2 (2.00%) low severe
1 (1.00%) low mild
4 (4.00%) high mild
1 (1.00%) high severe

U256::rotate_left time: [8.6338 ns 8.6409 ns 8.6480 ns]
Found 3 outliers among 100 measurements (3.00%)
3 (3.00%) high mild

U256::rotate_right time: [8.5947 ns 8.5984 ns 8.6023 ns]
Found 4 outliers among 100 measurements (4.00%)
4 (4.00%) high mild

After:

U256::add time: [8.3776 ns 8.3796 ns 8.3818 ns]
change: [+0.0877% +0.1233% +0.1592%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 8 outliers among 100 measurements (8.00%)
1 (1.00%) low mild
6 (6.00%) high mild
1 (1.00%) high severe

U256::div (lo/lo) time: [31.350 ns 31.386 ns 31.423 ns]
change: [+0.2172% +0.3445% +0.4766%] (p = 0.00 < 0.05)
Change within noise threshold.

U256::div (hi/lo) time: [143.55 ns 143.63 ns 143.72 ns]
change: [+0.5793% +0.6375% +0.7054%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 8 outliers among 100 measurements (8.00%)
2 (2.00%) low mild
6 (6.00%) high mild

U256::div (hi/hi) time: [132.81 ns 132.88 ns 132.96 ns]
change: [-69.311% -69.290% -69.270%] (p = 0.00 < 0.05)
Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high mild

U256::mul time: [21.653 ns 21.658 ns 21.664 ns]
change: [-0.0171% +0.0175% +0.0521%] (p = 0.32 > 0.05)
No change in performance detected.
Found 5 outliers among 100 measurements (5.00%)
4 (4.00%) high mild
1 (1.00%) high severe

U256::sub time: [8.3743 ns 8.3750 ns 8.3757 ns]
change: [-0.1994% -0.1761% -0.1541%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 3 outliers among 100 measurements (3.00%)
3 (3.00%) high mild

U256::shl time: [5.9165 ns 5.9166 ns 5.9168 ns]
change: [-0.1599% -0.1259% -0.0897%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 8 outliers among 100 measurements (8.00%)
1 (1.00%) low severe
2 (2.00%) low mild
1 (1.00%) high mild
4 (4.00%) high severe

U256::shr time: [5.8862 ns 5.8863 ns 5.8865 ns]
change: [-0.0270% -0.0121% +0.0038%] (p = 0.13 > 0.05)
No change in performance detected.
Found 5 outliers among 100 measurements (5.00%)
1 (1.00%) low severe
2 (2.00%) low mild
1 (1.00%) high mild
1 (1.00%) high severe

U256::ctlz time: [3.2861 ns 3.2862 ns 3.2863 ns]
change: [-0.2516% -0.2076% -0.1648%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 9 outliers among 100 measurements (9.00%)
1 (1.00%) low severe
1 (1.00%) low mild
4 (4.00%) high mild
3 (3.00%) high severe

U256::cttz time: [3.2849 ns 3.2857 ns 3.2865 ns]
change: [-0.0756% -0.0139% +0.0411%] (p = 0.65 > 0.05)
No change in performance detected.
Found 3 outliers among 100 measurements (3.00%)
1 (1.00%) low severe
2 (2.00%) high mild

U256::rotate_left time: [8.6495 ns 8.6562 ns 8.6631 ns]
change: [+0.0940% +0.2308% +0.3666%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 4 outliers among 100 measurements (4.00%)
3 (3.00%) high mild
1 (1.00%) high severe

[mcseemk]

Intel(R) Xeon(R) CPU E3-1270 v6 @ 3.80GHz
Before:
U256::add time: [3.7044 ns 3.7158 ns 3.7301 ns]
Found 6 outliers among 100 measurements (6.00%)
1 (1.00%) low severe
1 (1.00%) low mild
3 (3.00%) high mild
1 (1.00%) high severe

U256::div (lo/lo) time: [34.656 ns 34.746 ns 34.849 ns]
Found 11 outliers among 100 measurements (11.00%)
5 (5.00%) high mild
6 (6.00%) high severe

U256::div (hi/lo) time: [101.12 ns 101.34 ns 101.57 ns]
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high mild

U256::div (hi/hi) time: [260.59 ns 261.64 ns 263.05 ns]
Found 4 outliers among 100 measurements (4.00%)
2 (2.00%) high mild
2 (2.00%) high severe

U256::mul time: [7.2249 ns 7.2486 ns 7.2787 ns]
Found 5 outliers among 100 measurements (5.00%)
2 (2.00%) high mild
3 (3.00%) high severe

U256::sub time: [3.6266 ns 3.6375 ns 3.6507 ns]
Found 8 outliers among 100 measurements (8.00%)
1 (1.00%) low mild
2 (2.00%) high mild
5 (5.00%) high severe

U256::shl time: [4.6821 ns 4.7187 ns 4.7582 ns]
Found 8 outliers among 100 measurements (8.00%)
7 (7.00%) high mild
1 (1.00%) high severe

U256::shr time: [5.4234 ns 5.6343 ns 5.8779 ns]
Found 16 outliers among 100 measurements (16.00%)
2 (2.00%) high mild
14 (14.00%) high severe

U256::ctlz time: [1.7979 ns 1.8071 ns 1.8172 ns]
Found 2 outliers among 100 measurements (2.00%)
1 (1.00%) high mild
1 (1.00%) high severe

U256::cttz time: [1.6234 ns 1.6275 ns 1.6317 ns]
Found 11 outliers among 100 measurements (11.00%)
4 (4.00%) high mild
7 (7.00%) high severe

U256::rotate_left time: [7.6074 ns 7.6350 ns 7.6708 ns]
Found 7 outliers among 100 measurements (7.00%)
3 (3.00%) high mild
4 (4.00%) high severe

U256::rotate_right time: [6.9703 ns 6.9921 ns 7.0137 ns]
Found 2 outliers among 100 measurements (2.00%)
2 (2.00%) high mild

After:

U256::add time: [3.8436 ns 4.0444 ns 4.2842 ns]
change: [+0.9428% +3.2698% +6.2352%] (p = 0.01 < 0.05)
Change within noise threshold.
Found 14 outliers among 100 measurements (14.00%)
1 (1.00%) low mild
13 (13.00%) high severe

U256::div (lo/lo) time: [34.616 ns 34.722 ns 34.836 ns]
change: [-1.8507% -1.2031% -0.6344%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 7 outliers among 100 measurements (7.00%)
6 (6.00%) high mild
1 (1.00%) high severe

U256::div (hi/lo) time: [102.35 ns 102.50 ns 102.67 ns]
change: [+0.9068% +1.1183% +1.3322%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 3 outliers among 100 measurements (3.00%)
2 (2.00%) high mild
1 (1.00%) high severe

U256::div (hi/hi) time: [76.856 ns 80.566 ns 84.570 ns]
change: [-71.760% -71.055% -70.233%] (p = 0.00 < 0.05)
Performance has improved.
Found 13 outliers among 100 measurements (13.00%)
13 (13.00%) high severe

U256::mul time: [7.2813 ns 7.3206 ns 7.3653 ns]
change: [-0.3070% +0.5551% +1.2446%] (p = 0.16 > 0.05)
No change in performance detected.
Found 9 outliers among 100 measurements (9.00%)
5 (5.00%) high mild
4 (4.00%) high severe

U256::sub time: [3.6143 ns 3.6161 ns 3.6180 ns]
change: [-0.8392% -0.5666% -0.3273%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 4 outliers among 100 measurements (4.00%)
1 (1.00%) low severe
3 (3.00%) high mild

U256::shl time: [4.9900 ns 5.2624 ns 5.5863 ns]
change: [+4.5167% +8.7856% +13.977%] (p = 0.00 < 0.05)
Performance has regressed.
Found 15 outliers among 100 measurements (15.00%)
1 (1.00%) high mild
14 (14.00%) high severe

U256::shr time: [5.2371 ns 5.3298 ns 5.4537 ns]
change: [-7.4553% -4.3827% -1.3965%] (p = 0.01 < 0.05)
Performance has improved.
Found 12 outliers among 100 measurements (12.00%)
4 (4.00%) high mild
8 (8.00%) high severe

U256::ctlz time: [1.9459 ns 2.0639 ns 2.2043 ns]
change: [+6.6873% +11.117% +15.950%] (p = 0.00 < 0.05)
Performance has regressed.
Found 14 outliers among 100 measurements (14.00%)
1 (1.00%) high mild
13 (13.00%) high severe

U256::cttz time: [1.6350 ns 1.6419 ns 1.6494 ns]
change: [-1.1075% -0.4105% +0.3172%] (p = 0.27 > 0.05)
No change in performance detected.
Found 6 outliers among 100 measurements (6.00%)
3 (3.00%) high mild
3 (3.00%) high severe

U256::rotate_left time: [7.7551 ns 7.7892 ns 7.8216 ns]
change: [+0.7183% +1.2633% +1.7584%] (p = 0.00 < 0.05)
Change within noise threshold.

U256::rotate_right time: [6.8823 ns 6.8993 ns 6.9163 ns]
change: [-1.9033% -1.4820% -1.0720%] (p = 0.00 < 0.05)
Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high severe

[mcseemk]

Please note that I did not test big endian arch. I don't know how to test it to be honest. I'm reasonably sure that div_mod_knuth is architecture-agnostic, but it is better to test.

[nlordell]

Sorry, just getting back to this now. The changes look great. I also ran the more comprehensive benchmarks from the intx-division branch and there are some very noticeable improvements, noticeably:

U256::div/####/####     time:   [36.128 ns 36.137 ns 36.148 ns]                                 
                        change: [-68.388% -68.374% -68.360%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 7 outliers among 100 measurements (7.00%)
  1 (1.00%) low mild
  6 (6.00%) high severe

U256::div/####/###      time:   [57.154 ns 57.161 ns 57.168 ns]                               
                        change: [-86.374% -86.366% -86.359%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 13 outliers among 100 measurements (13.00%)
  2 (2.00%) low mild
  3 (3.00%) high mild
  8 (8.00%) high severe

U256::div/###/###       time:   [36.128 ns 36.132 ns 36.136 ns]                               
                        change: [-77.010% -76.966% -76.867%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 11 outliers among 100 measurements (11.00%)
  5 (5.00%) high mild
  6 (6.00%) high severe

[nlordell]

I'm just going to run the fuzzer for a bit to see if there are any regressions with division, then this should be good to merge.