Verify invariants about number of elements and length

[ameligrana]

No description provided.

[github-actions]

Benchmark Results

	3a053b9	e0935ec	3a053b9 / e0935ec
TTFX excluding time to load	0.0533 ± 0 s	0.0523 ± 0 s	1 ± 0,0.981 ± 0,1.02 ± 0
code size in bytes	1.52e+04 ± 0 h	1.52e+04 ± 0 h	1 ± 0
code size in lines	557 ± 0 h	557 ± 0 h	1 ± 0
code size in syntax nodes	4.11e+03 ± 0 h	4.11e+03 ± 0 h	1 ± 0
constructor n=100 σ=0.1	5.77 ± 0.13 μs	5.74 ± 0.24 μs	0.991 ± 0.066,1.18 ± 0.054,1.01 ± 0.048
constructor n=100 σ=1.0	6.27 ± 0.15 μs	6.24 ± 0.18 μs	0.999 ± 0.079,1.14 ± 0.055,1 ± 0.037
constructor n=100 σ=10.0	6.38 ± 0.13 μs	6.38 ± 0.13 μs	0.96 ± 0.37,1.16 ± 0.041,1 ± 0.029
constructor n=100 σ=100.0	9.83 ± 6 μs	9.49 ± 1.4 μs	0.984 ± 0.84,1.2 ± 0.18,1.04 ± 0.64
constructor n=1000 σ=0.1	0.0424 ± 0.0011 ms	0.0427 ± 0.0011 ms	0.975 ± 0.069,1.26 ± 0.045,0.994 ± 0.036
constructor n=1000 σ=1.0	0.045 ± 0.0013 ms	0.0457 ± 0.0016 ms	1 ± 0.059,1.23 ± 0.045,0.984 ± 0.045
constructor n=1000 σ=10.0	0.0517 ± 0.0011 ms	0.051 ± 0.00083 ms	0.983 ± 0.047,1.18 ± 0.028,1.01 ± 0.027
constructor n=1000 σ=100.0	0.0593 ± 0.00085 ms	0.059 ± 0.00073 ms	0.992 ± 0.058,1.17 ± 0.028,1 ± 0.019
constructor n=10000 σ=0.1	0.427 ± 0.019 ms	0.422 ± 0.022 ms	0.989 ± 0.084,1.26 ± 0.069,1.01 ± 0.069
constructor n=10000 σ=1.0	0.434 ± 0.023 ms	0.438 ± 0.017 ms	0.956 ± 0.087,1.17 ± 0.08,0.991 ± 0.065
constructor n=10000 σ=10.0	0.517 ± 0.023 ms	0.49 ± 0.027 ms	1.03 ± 0.13,1.23 ± 0.1,1.06 ± 0.074
constructor n=10000 σ=100.0	0.576 ± 0.023 ms	0.576 ± 0.025 ms	0.99 ± 0.054,1.15 ± 0.062,1 ± 0.059
delete ∘ rand n=100 σ=0.1	4.39 ± 0.2 μs	4.44 ± 0.2 μs	0.998 ± 0.062,1.38 ± 0.073,0.989 ± 0.063
delete ∘ rand n=100 σ=1.0	4.66 ± 0.21 μs	4.73 ± 0.2 μs	0.993 ± 0.061,1.3 ± 0.068,0.985 ± 0.061
delete ∘ rand n=100 σ=10.0	4.81 ± 0.19 μs	4.9 ± 0.19 μs	0.982 ± 0.054,1.3 ± 0.062,0.981 ± 0.054
delete ∘ rand n=100 σ=100.0	7.16 ± 0.47 μs	7.33 ± 0.47 μs	0.979 ± 0.088,1.28 ± 0.1,0.977 ± 0.09
delete ∘ rand n=1000 σ=0.1	0.0436 ± 0.0008 ms	0.0439 ± 0.00083 ms	0.994 ± 0.027,1.39 ± 0.03,0.995 ± 0.026
delete ∘ rand n=1000 σ=1.0	0.0473 ± 0.0012 ms	0.048 ± 0.0013 ms	0.993 ± 0.033,1.29 ± 0.034,0.986 ± 0.037
delete ∘ rand n=1000 σ=10.0	0.0484 ± 0.00099 ms	0.0493 ± 0.00099 ms	0.991 ± 0.028,1.24 ± 0.031,0.981 ± 0.028
delete ∘ rand n=1000 σ=100.0	0.0531 ± 0.001 ms	0.0543 ± 0.0011 ms	0.982 ± 0.024,1.26 ± 0.028,0.978 ± 0.027
delete ∘ rand n=10000 σ=0.1	0.474 ± 0.012 ms	0.48 ± 0.015 ms	1.09 ± 0.14,1.34 ± 0.039,0.988 ± 0.039
delete ∘ rand n=10000 σ=1.0	0.506 ± 0.014 ms	0.511 ± 0.011 ms	0.987 ± 0.043,1.25 ± 0.036,0.99 ± 0.034
delete ∘ rand n=10000 σ=10.0	0.512 ± 0.011 ms	0.523 ± 0.012 ms	0.99 ± 0.042,1.21 ± 0.042,0.978 ± 0.03
delete ∘ rand n=10000 σ=100.0	0.507 ± 0.016 ms	0.514 ± 0.013 ms	0.989 ± 0.04,1.22 ± 0.038,0.987 ± 0.04
empty constructor	1.63 ± 0.28 μs	1.76 ± 0.29 μs	0.914 ± 0.22,0.95 ± 0.24,0.926 ± 0.22
intermixed_h n=100 σ=0.1	11.1 ± 0.83 μs	11.1 ± 0.83 μs	0.995 ± 0.11,1.7 ± 0.24,0.999 ± 0.11
intermixed_h n=100 σ=1.0	11.3 ± 0.95 μs	11.4 ± 0.98 μs	1 ± 0.13,1.66 ± 0.24,0.99 ± 0.12
intermixed_h n=100 σ=10.0	11.4 ± 1.5 μs	11.3 ± 1.3 μs	0.998 ± 0.34,1.78 ± 0.49,1.01 ± 0.18
intermixed_h n=100 σ=100.0	12.1 ± 0.92 μs	12.2 ± 0.98 μs	0.998 ± 0.15,1.65 ± 0.26,0.991 ± 0.11
intermixed_h n=1000 σ=0.1	0.106 ± 0.0097 ms	0.107 ± 0.0083 ms	0.983 ± 0.12,2.06 ± 0.57,0.99 ± 0.12
intermixed_h n=1000 σ=1.0	0.109 ± 0.0082 ms	0.111 ± 0.0089 ms	0.997 ± 0.11,1.93 ± 0.59,0.987 ± 0.11
intermixed_h n=1000 σ=10.0	0.104 ± 0.0099 ms	0.104 ± 0.0091 ms	1 ± 0.13,2.05 ± 0.62,1 ± 0.13
intermixed_h n=1000 σ=100.0	0.112 ± 0.012 ms	0.113 ± 0.013 ms	0.997 ± 0.15,1.97 ± 0.57,0.999 ± 0.16
intermixed_h n=10000 σ=0.1	1.14 ± 0.19 ms	1.14 ± 0.18 ms	0.983 ± 0.2,2.1 ± 0.82,1 ± 0.23
intermixed_h n=10000 σ=1.0	1.14 ± 0.17 ms	1.15 ± 0.16 ms	0.989 ± 0.21,2.13 ± 0.92,0.985 ± 0.2
intermixed_h n=10000 σ=10.0	1.08 ± 0.15 ms	1.11 ± 0.18 ms	1.03 ± 0.21,2.15 ± 0.77,0.972 ± 0.21
intermixed_h n=10000 σ=100.0	1.12 ± 0.16 ms	1.16 ± 0.22 ms	1.02 ± 0.2,2.1 ± 0.82,0.97 ± 0.23
pathological 1	0.0456 ± 0.00025 μs	0.0456 ± 0.00019 μs	1 ± 0.0071,1.97 ± 0.21,1 ± 0.0069
pathological 1′	0.166 ± 0.0028 μs	0.161 ± 0.0017 μs	1.05 ± 0.015,1.23 ± 0.019,1.03 ± 0.021
pathological 2	0.0629 ± 0.0002 μs	0.0629 ± 0.00019 μs	1 ± 0.0056,1.73 ± 0.11,1 ± 0.0044
pathological 2′	0.175 ± 0.0012 μs	0.169 ± 0.0017 μs	1.04 ± 0.013,1.13 ± 0.018,1.04 ± 0.013
pathological 2′′	0.193 ± 0.033 μs	0.189 ± 0.0064 μs	1.06 ± 0.18,1.17 ± 0.094,1.02 ± 0.18
pathological 3	16.8 ± 0.29 ns	16.9 ± 0.23 ns	0.986 ± 0.024,0.984 ± 0.019,0.993 ± 0.022
pathological 4	0.0624 ± 0.0003 μs	0.0623 ± 0.00019 μs	0.999 ± 0.006,1.68 ± 0.13,1 ± 0.0057
pathological 4′	0.186 ± 0.0015 μs	0.205 ± 0.0016 μs	0.944 ± 0.0098,1.14 ± 0.02,0.91 ± 0.01
pathological 4′′	0.228 ± 0.012 μs	0.237 ± 0.0057 μs	0.964 ± 0.032,1.16 ± 0.046,0.963 ± 0.058
pathological 5a	0.0446 ± 0.00018 μs	0.0446 ± 0.00014 μs	1 ± 0.0052,2.04 ± 0.25,1 ± 0.0052
pathological 5b	0.0447 ± 0.00038 μs	0.0446 ± 0.00017 μs	1 ± 0.0051,2.15 ± 0.35,1 ± 0.0094
pathological 5b′	0.349 ± 0.0032 μs	0.328 ± 0.0037 μs	1.06 ± 0.015,1.09 ± 0.021,1.06 ± 0.015
pathological 5b′′	0.36 ± 0.01 μs	0.344 ± 0.0068 μs	1.02 ± 0.018,1.14 ± 0.033,1.05 ± 0.036
pathological compaction	3.64 ± 0.026 ms	3.55 ± 0.068 ms	1.05 ± 0.021,1.03 ± 0.01,1.03 ± 0.021
sample (bulk) n=1000 k=10000 σ=1	0.161 ± 0.021 ms	0.163 ± 0.021 ms	0.998 ± 0.18,1.4 ± 0.18,0.985 ± 0.18
sample (bulk) n=1000 k=10000 σ=100	0.171 ± 0.039 ms	0.174 ± 0.036 ms	1 ± 0.3,0.995 ± 0.3,0.982 ± 0.3
sample (bulk) n=1000 k=1000000 σ=1	16.1 ± 2.1 ms	15.8 ± 2.1 ms	1.01 ± 0.18,1.44 ± 0.19,1.02 ± 0.19
sample (bulk) n=1000 k=1000000 σ=100	8 ± 3.1 ms	9.04 ± 4 ms	1.04 ± 0.55,1.07 ± 0.7,0.885 ± 0.53
sample (bulk) n=1000000 k=10000 σ=1	0.324 ± 0.046 ms	0.39 ± 0.023 ms	1 ± 0.14,1.13 ± 0.16,0.831 ± 0.13
sample (bulk) n=1000000 k=10000 σ=100	0.147 ± 0.028 ms	0.158 ± 0.029 ms	1 ± 0.25,1.97 ± 0.32,0.932 ± 0.25
sample (bulk) n=1000000 k=1000000 σ=1	20 ± 1.3 ms	21 ± 0.44 ms	0.977 ± 0.041,1.32 ± 0.047,0.952 ± 0.067
sample (bulk) n=1000000 k=1000000 σ=100	9.27 ± 2.7 ms	9.94 ± 2.9 ms	0.749 ± 0.22,1.02 ± 0.64,0.932 ± 0.39
sample n=100 σ=0.1	25 ± 0.65 ns	25.4 ± 0.61 ns	0.998 ± 0.04,0.991 ± 0.036,0.986 ± 0.035
sample n=100 σ=1.0	29.5 ± 2.5 ns	29.4 ± 2.3 ns	0.999 ± 0.11,0.991 ± 0.1,1 ± 0.12
sample n=100 σ=10.0	19.2 ± 4.5 ns	19.3 ± 4.8 ns	1.01 ± 0.37,1 ± 0.34,0.993 ± 0.34
sample n=100 σ=100.0	16.7 ± 3.8 ns	16.9 ± 3.7 ns	0.983 ± 0.32,0.973 ± 0.31,0.99 ± 0.31
sample n=1000 σ=0.1	23 ± 4.7 ns	26.5 ± 7.2 ns	0.998 ± 0.28,0.987 ± 0.28,0.867 ± 0.3
sample n=1000 σ=1.0	0.0319 ± 0.0023 μs	0.0321 ± 0.0021 μs	1 ± 0.11,1.01 ± 0.1,0.996 ± 0.097
sample n=1000 σ=10.0	20.3 ± 5.1 ns	20.2 ± 5.4 ns	0.99 ± 0.4,0.979 ± 0.36,1 ± 0.37
sample n=1000 σ=100.0	16.7 ± 3.7 ns	17 ± 4 ns	0.984 ± 0.33,0.978 ± 0.32,0.982 ± 0.32
sample n=10000 σ=0.1	31.5 ± 1.2 ns	0.032 ± 0.0013 μs	1.03 ± 0.18,1.01 ± 0.055,0.984 ± 0.055
sample n=10000 σ=1.0	0.0347 ± 0.0016 μs	0.0345 ± 0.0016 μs	1.01 ± 0.057,1.01 ± 0.061,1.01 ± 0.065
sample n=10000 σ=10.0	21.9 ± 6.6 ns	21.8 ± 6.3 ns	1.02 ± 0.46,0.972 ± 0.4,1 ± 0.42
sample n=10000 σ=100.0	16.8 ± 4.1 ns	17.6 ± 3.8 ns	1.05 ± 0.38,0.961 ± 0.35,0.954 ± 0.31
summarysize n=100 σ=0.1	1.13e+05 ± 0 h	1.13e+05 ± 0 h	1 ± 0
summarysize n=100 σ=1.0	1.13e+05 ± 0 h	1.13e+05 ± 0 h	1 ± 0
summarysize n=100 σ=10.0	1.13e+05 ± 0 h	1.13e+05 ± 0 h	1 ± 0
summarysize n=100 σ=100.0	1.13e+05 ± 0 h	1.13e+05 ± 0 h	1 ± 0
summarysize n=1000 σ=0.1	1.42e+05 ± 0 h	1.42e+05 ± 0 h	1 ± 0
summarysize n=1000 σ=1.0	1.42e+05 ± 0 h	1.42e+05 ± 0 h	1 ± 0
summarysize n=1000 σ=10.0	1.42e+05 ± 0 h	1.42e+05 ± 0 h	1 ± 0
summarysize n=1000 σ=100.0	1.42e+05 ± 0 h	1.42e+05 ± 0 h	1 ± 0
summarysize n=10000 σ=0.1	1e+06 ± 0 h	1e+06 ± 0 h	1 ± 0
summarysize n=10000 σ=1.0	1e+06 ± 0 h	1e+06 ± 0 h	1 ± 0
summarysize n=10000 σ=10.0	1e+06 ± 0 h	1e+06 ± 0 h	1 ± 0
summarysize n=10000 σ=100.0	1e+06 ± 0 h	1e+06 ± 0 h	1 ± 0
update ∘ rand n=100 σ=0.1	0.0821 ± 0.0021 μs	0.0844 ± 0.0022 μs	0.966 ± 0.036,1.8 ± 0.06,0.973 ± 0.036
update ∘ rand n=100 σ=1.0	0.0887 ± 0.003 μs	0.0907 ± 0.0029 μs	0.974 ± 0.045,1.67 ± 0.06,0.978 ± 0.046
update ∘ rand n=100 σ=10.0	0.0962 ± 0.0036 μs	0.0995 ± 0.0036 μs	0.967 ± 0.051,1.58 ± 0.065,0.967 ± 0.05
update ∘ rand n=100 σ=100.0	0.172 ± 0.015 μs	0.176 ± 0.015 μs	0.989 ± 0.12,1.32 ± 0.14,0.979 ± 0.12
update ∘ rand n=1000 σ=0.1	0.0826 ± 0.0022 μs	0.0852 ± 0.0023 μs	0.96 ± 0.036,1.79 ± 0.067,0.97 ± 0.037
update ∘ rand n=1000 σ=1.0	0.089 ± 0.0033 μs	0.091 ± 0.0031 μs	0.976 ± 0.048,1.67 ± 0.062,0.978 ± 0.049
update ∘ rand n=1000 σ=10.0	0.0945 ± 0.0023 μs	0.0971 ± 0.0022 μs	0.969 ± 0.028,1.59 ± 0.033,0.973 ± 0.032
update ∘ rand n=1000 σ=100.0	0.182 ± 0.0067 μs	0.187 ± 0.007 μs	0.985 ± 0.052,1.26 ± 0.067,0.976 ± 0.051
update ∘ rand n=10000 σ=0.1	0.0907 ± 0.0018 μs	0.0929 ± 0.0023 μs	0.975 ± 0.021,1.67 ± 0.029,0.977 ± 0.031
update ∘ rand n=10000 σ=1.0	0.0944 ± 0.0021 μs	0.0956 ± 0.0012 μs	1.02 ± 0.071,1.61 ± 0.041,0.988 ± 0.025
update ∘ rand n=10000 σ=10.0	0.0937 ± 0.00088 μs	0.0961 ± 0.0014 μs	0.984 ± 0.026,1.59 ± 0.037,0.974 ± 0.017
update ∘ rand n=10000 σ=100.0	0.162 ± 0.0018 μs	0.165 ± 0.002 μs	0.991 ± 0.025,1.33 ± 0.019,0.981 ± 0.016
time_to_load	0.0743 ± 0.0013 s	0.0738 ± 0.00024 s	0.987 ± 0.015,0.993 ± 0.023,1.01 ± 0.017

Benchmark Plots

A plot of the benchmark results have been uploaded as an artifact to the workflow run for this PR.
Go to "Actions"->"Benchmark a pull request"->[the most recent run]->"Artifacts" (at the bottom).

[LilithHafner]

It's concerning that this PR doesn't break any tests. We should add a code path to tests that exercises resizing a ResizableWeights smaller and then performing operations including checking invariants.

[LilithHafner]

We only maintain and require <=, not full ==. This allows over-allocation which is needed for repeated resizing to be performant:

Suggested change

	verify_m1(m::Memory) = m[1] == _length_from_memory(length(m))
	verify_m1(m::Memory) = m[1] <= _length_from_memory(length(m))