perf(flock): bound ARM Round1 conversion traversal

[kevaundray]

This PR was created with Claude Code, prompted by @kevaundray.

Change

Keep the ARM Round1 medium-to-F192 conversion in bounded groups of four medium rows, visiting all 64 lanes before advancing to the next group. Two stack-resident lane arrays preserve the unweighted sums; the equality weight is applied after the complete window, including a partial final group. Extend the existing arithmetic regression test to cover every tail count, empty accumulation and repeated full windows.

The packed AB sweep and C transpose are unchanged. There are no protocol, guest, transcript, proof-layout or version changes, no runtime feature detection, and no new ISA requirement. Non-ARM conversion paths are unchanged. The existing module documentation describes the traversal.

Native measurements

Paired baseline is f52bd991c8623894e7b03a2e50bede6c248ea1ad, which includes #549/#550 and excludes #552. Final head is f19e7b084abad98b6e9d7fc51f07c5e96532c004. Both sides use Rust 1.99, -C target-cpu=native, release thin LTO and one codegen unit on the same actual V2 or Xeon. Five alternating pairs per worker count. Default pools are 16 and 32 respectively. Negative percentages mean faster; percentages below are medians of paired changes, not ratios of independent medians.

Rate1 candidate screen and exact-head rate2/aggregation confirmation:

V2 leanXMSS100	Workers	Round1 baseline/head ms	Paired Round1	Proof baseline/head ms	Paired proof
log-inv-rate1	1	307/263	-14.33%	1941/1895	-2.37%
log-inv-rate1	4	78.5/67.1	-14.49%	519/508	-2.31%
log-inv-rate1	8	41.5/36	-13.25%	290/283	-2.08%
log-inv-rate1	default	22.9/20.3	-11.35%	176/173	-1.70%
log-inv-rate2	1	308/262	-14.66%	2472/2426	-1.74%
log-inv-rate2	4	78.8/67	-14.97%	654/643	-1.83%
log-inv-rate2	8	41.6/36	-13.49%	358/353	-1.67%
log-inv-rate2	default	22.9/20.3	-11.35%	211/209	-0.95%

At rate2, a genuine four-leaf tree over leanXMSS2, first/higher arity2, improves the actual whole-tree proving median by1.17%,1.09%,0.87%,0.28% at1/4/8/default workers. These are measured whole-tree runs after leaf proofs, not sums of separate component timings.

Small default-worker limitations

The initial rate2 cohort has real measured default-worker regressions: the separately measured small leaf is +4.07% (paired range -0.74% to +23.22%), and the first aggregation node is +2.50% (-0.83% to +16.67%). Higher-node proof is -0.83% (-3.23% to +16.53%), and whole-tree proof is -0.28% (-0.56% to +15.79%). These samples are retained, not discarded or declared an external-noise effect. Round1 itself improves, but this does not establish a uniform small-tree whole-proof speedup.

A single predeclared control cohort repeats only that exact small tree at default workers: 12 balanced triplets of head, baseline and an identical-baseline-executable control. Every ordering appears twice, shuffled with a fixed seed; each side occupies each execution position four times. All samples are retained. The control measures ordinary run variability, not permission to erase regressions.

Machine/stage	Head/base paired median (range)	Head/control paired median (range)	Identical control/base paired median (range)
arm64/leaf	-0.73% (-2.18%, +4.03%)	-0.37% (-1.81%, +4.41%)	+0.00% (-1.45%, +1.12%)
arm64/first	-0.83% (-3.23%, +0.83%)	-0.83% (-6.25%, +1.67%)	+0.00% (-1.63%, +5.79%)
arm64/higher	-0.82% (-7.63%, +1.64%)	+0.00% (-3.97%, +1.64%)	+0.00% (-7.63%, +4.13%)
arm64/tree	-0.55% (-1.38%, +0.55%)	-0.55% (-1.92%, +0.83%)	+0.00% (-0.55%, +0.83%)
x86-64/leaf	-0.00% (-2.91%, +3.30%)	+0.37% (-10.86%, +3.30%)	-0.19% (-2.91%, +14.29%)
x86-64/first	+0.28% (-1.99%, +3.88%)	+0.05% (-1.11%, +3.16%)	+0.17% (-2.95%, +3.76%)
x86-64/higher	-1.00% (-6.80%, +0.11%)	-0.06% (-5.91%, +1.46%)	-0.67% (-7.01%, +3.14%)
x86-64/tree	-0.94% (-3.32%, +1.53%)	-0.94% (-2.96%, +1.15%)	-0.18% (-1.89%, +1.91%)

V2 first/higher Round1 improves about11% against both references in the control cohort, while the unchanged Basis medians are0% against baseline. The original regressions and the overlapping control ranges remain limitations. No further variants or favorable-result retries were run.

Xeon, memory and compiler shape

Xeon is not an optimization claim: its conversion disassembly is literally unchanged, and the enclosing fold-reduce and drain-into-slots instruction bodies differ only in anonymous LLVM symbol hashes in disassembler comments. Observed leaf whole-proof paired medians at1/4/8/default workers are -0.10%/-0.67%/0%/0% at rate1 and +0.52%/-0.27%/+0.91%/-0.74% at rate2. Rate1 eight-worker samples range from -8.76% to +3.91%; rate2 eight-worker samples range from -6.44% to +4.11%. Positive observations are retained.

The source adds3072B of logical stack scratch, not heap allocation. WorkerState remains8192B. Actual native V2 enclosing-frame size is6464B baseline and8016B selected, an emitted increase of1552B. The full AB medium loop retains128 packed-byte loads,208 PMULL-family instructions and64B spill stores/reloads. Conversion emits bounded vector-pair/scalar loads and EOR3; no hardware-counter attribution is claimed because perf access is denied.

Rate1 V2 RSS medians in MiB, baseline/head at1/4/8/default:1591.62/1593.57,1653.70/1654.05,1691.23/1689.54,1752.68/1752.71. In the controlled default small tree they are1256.59/1253.35, with identical-base control1252.52. Heap measurements and allocation counts are recorded in the artifacts; some higher-worker allocation counts vary between runs, so global allocation-count equality is not claimed.

Rejected alternatives

Only the four-row conversion is retained. Full medium-major conversion regresses default-worker Round1 by3.49%. A separate Horner AB candidate removes the loop spills and lowers208 to192 PMULL instructions, but regresses V2 Round1 by5.21%/4.84%/3.61%/2.18% at1/4/8/default. Optional SVE2 BGRP improves the isolated C transpose from12.34ns to8.60ns without a useful whole-proof win, adds worker state and a branch per medium, and is not included. This is distinct from the previously rejected quarter-tiled AB mechanism. Standalone microkernel timings are not added to span or whole-proof gains.

Standard PR benchmark confirmation

The complete automatic PR Bench passed all 46 native jobs on the exact shipping head. Its five-pair comparisons independently support the retained improvement: V2 leanXMSS100 whole-proof paired medians are -2.00%, -1.92%, -1.75% and -1.11% at 1/4/8/default workers. Every V2 tracked CLI whole-proof row has an improved median, including the 16 first/higher aggregation rows, whose medians range from -0.80% to -1.92%. Other V2 CLI medians range from -1.20% to -1.89%. Positive individual samples remain, including default leanXMSS100's range of -1.79% to +1.71%.

This does not imply every benchmark improves. V2 wrapping-add batch is +0.55% (-2.41% to +4.68%), widening-multiply batch is +0.58% (+0.26% to +2.95%), and PCS-open is +0.36% (-0.13% to +6.04%). These observed regressions are retained. V2 wrapping-multiply and hash batches improve by 3.00% and 2.30% respectively. No uniform microbenchmark benefit or external-noise explanation is claimed.

The unchanged Xeon backend continues to show mixed observations: leanXMSS100 paired medians are -0.72%, -0.03%, -0.13% and +0.37% at 1/4/8/default workers; shielded is +1.50% (-0.51% to +2.68%), and eight-worker 4-to-1 first-node aggregation is +1.09% (-1.58% to +6.38%). Eight-worker 2-to-1 higher-node aggregation includes a +17.71% sample despite a +0.33% median. All 90 timing rows, including standalone unchanged kernels, were inspected. These results supplement, rather than replace, the earlier small-tree regressions and fixed-control limitations.

Correctness and checks

• The exact shipping-head PR check rollup is fully green: 62 successful checks, including all 46 native Bench jobs, Rust, RISC-V ACT4, counts and Python verifier checks.
• 950 captured proof files across the three cohorts were literally compared per workload/rate/shape across baseline/head, worker counts, repeated runs and architectures. Production verification passed. Coverage includes representative leaf, first aggregation node, higher node and root at log-inv-rates1/2/3/4. The control proofs also match the prior rate2-tree cohort. Captures and comparison manifests are retained in workflow artifacts.
• Exact final-head Rust CI: all13 jobs passed, including full tests, Miri, MSRV, NEON runtime, clippy, fmt, docs and dependency checks. Hosted GFNI runtime steps were skipped for unavailable CPU features; the actual Xeon screen separately passed all8 Round1 tests.
• Local Rust1.97 fmt, clippyall, docall and49 release flock tests passed, plus native proof+verify and real aggregation smoke.
• Final-head generic AArch64 fallback tests passed7/7 on each of Rust1.97 and1.99 under QEMU Cortex-A72. ARM compiler/ISA checks also passed. QEMU is correctness evidence only, never hardware timing evidence.
• No temporary capture hooks, diagnostic workflow changes or profiling toggles remain in the shipping diff.

[github-actions]

Benchmarks: base against this PR, on one runner

f19e7b0 merged into its base, against the base f52bd99: each benchmark run in turns on one runner, base and PR 5 times each. A size is shown when it changed; a time or a peak memory when it moved the same way in every round and its median by at least 1% (the change is the median ratio, with the rounds' range).

Proving time

benchmark	runner	base	this PR	change
aggregate-leanxmss-100-2to1-1thread-first	arm64	998.63 ms	982.90 ms	-1.45% (-4.35% to -1.25%)
aggregate-leanxmss-100-2to1-1thread-node	arm64	1.05 s	1.03 s	-1.92% (-6.54% to -1.57%)
aggregate-leanxmss-100-2to1-4thread-node	arm64	296.01 ms	291.77 ms	-1.42% (-3.46% to -0.75%)
aggregate-leanxmss-100-2to1-8thread-first	arm64	165.58 ms	163.76 ms	-1.11% (-1.58% to -0.86%)
aggregate-leanxmss-100-2to1-8thread-node	arm64	174.73 ms	171.81 ms	-1.38% (-1.86% to -1.13%)
aggregate-leanxmss-100-2to1-node	x86-64	89.97 ms	88.31 ms	-1.84% (-2.65% to -0.61%)
aggregate-leanxmss-100-4to1-1thread-first	arm64	2.00 s	1.97 s	-1.68% (-1.85% to -0.11%)
aggregate-leanxmss-100-4to1-1thread-node	arm64	2.18 s	2.14 s	-1.88% (-2.99% to -1.46%)
aggregate-leanxmss-100-4to1-4thread-first	arm64	555.87 ms	547.78 ms	-1.43% (-2.33% to -0.74%)
aggregate-leanxmss-100-4to1-4thread-node	arm64	601.68 ms	595.74 ms	-1.25% (-2.17% to -0.22%)
aggregate-leanxmss-100-4to1-8thread-first	arm64	319.72 ms	315.55 ms	-1.03% (-1.30% to -0.45%)
aggregate-leanxmss-100-4to1-8thread-node	arm64	348.22 ms	341.98 ms	-1.65% (-2.54% to -0.85%)
aggregate-leanxmss-100-4to1-first	arm64	202.83 ms	200.94 ms	-1.06% (-1.25% to -0.81%)
aggregate-leanxmss-100-4to1-node	arm64	221.19 ms	219.31 ms	-1.10% (-1.35% to -0.49%)
falcon-7	arm64	577.27 ms	567.87 ms	-1.43% (-1.99% to -0.76%)
fibonacci-asm-2000000	arm64	883.68 ms	865.39 ms	-1.89% (-2.07% to -1.52%)
flock-hash-batch-262144	arm64	841.85 ms	814.84 ms	-2.30% (-3.94% to -1.56%)
flock-mul-wrapping-batch-262144	arm64	206.80 ms	200.75 ms	-3.00% (-3.34% to -2.18%)
hash-50000	arm64	548.79 ms	541.92 ms	-1.23% (-1.36% to -0.67%)
leanxmss-100-1thread	arm64	1.94 s	1.91 s	-2.00% (-2.42% to -1.89%)
leanxmss-100-4thread	arm64	513.46 ms	502.76 ms	-1.92% (-2.65% to -1.40%)
leanxmss-100-8thread	arm64	286.72 ms	281.70 ms	-1.75% (-2.43% to -1.21%)
pcs-commit-22	arm64	7.52 ms	7.20 ms	-3.16% (-8.37% to -0.75%)
shielded-258	arm64	776.63 ms	764.61 ms	-1.52% (-1.77% to -1.39%)
stateproof-5	arm64	1.28 s	1.27 s	-1.20% (-2.46% to -0.52%)

Prover stages

benchmark	runner	base	this PR	change
aggregate-leanxmss-100-2to1-1thread-first	arm64	998.63 ms	982.90 ms	-1.45% (-4.35% to -1.25%)
↳ build-circuit		12.22 ms	12.23 ms	-0.02% (-4.82% to +0.51%)
↳ prove-recursion		962.68 ms	947.08 ms	-1.49% (-4.38% to -1.34%)
↳ reduce-claims		15.58 ms	15.61 ms	-0.55% (-2.51% to +2.37%)
↳ reduce-in-rows		1.37 ms	1.34 ms	-2.46% (-4.75% to +1.43%)
↳ verify		6.83 ms	6.87 ms	+0.20% (-4.57% to +0.78%)
aggregate-leanxmss-100-2to1-1thread-first	x86-64	605.86 ms	608.95 ms	-0.01% (-0.25% to +1.63%)
↳ build-circuit		14.35 ms	14.39 ms	+0.60% (-5.23% to +3.73%)
↳ prove-recursion		563.88 ms	566.51 ms	+0.07% (-0.31% to +1.72%)
↳ reduce-claims		18.28 ms	18.27 ms	+0.01% (-1.85% to +0.64%)
↳ reduce-in-rows		1.43 ms	1.48 ms	+1.85% (+0.77% to +6.42%)
↳ verify		7.79 ms	7.82 ms	+0.24% (-0.40% to +1.00%)
aggregate-leanxmss-100-2to1-1thread-node	arm64	1.05 s	1.03 s	-1.92% (-6.54% to -1.57%)
↳ build-circuit		11.92 ms	11.94 ms	+0.03% (-5.23% to +0.53%)
↳ prove-recursion		999.57 ms	979.62 ms	-2.05% (-6.85% to -1.67%)
↳ reduce-claims		30.19 ms	30.33 ms	+0.85% (-3.74% to +1.44%)
↳ reduce-in-rows		1.43 ms	1.45 ms	+1.23% (-3.30% to +7.93%)
aggregate-leanxmss-100-2to1-4thread-first	arm64	281.95 ms	279.44 ms	-1.03% (-1.46% to +0.16%)
↳ build-circuit		10.76 ms	10.82 ms	+1.14% (-1.39% to +3.52%)
↳ prove-recursion		257.80 ms	255.06 ms	-1.23% (-1.59% to -0.08%)
↳ reduce-claims		5.24 ms	5.27 ms	+0.52% (-1.15% to +1.98%)
↳ reduce-in-rows		1.42 ms	1.40 ms	-1.87% (-4.81% to +5.16%)
↳ verify		6.83 ms	6.83 ms	+0.42% (-0.42% to +1.90%)
aggregate-leanxmss-100-2to1-4thread-node	arm64	296.01 ms	291.77 ms	-1.42% (-3.46% to -0.75%)
↳ build-circuit		11.60 ms	11.43 ms	-2.39% (-2.83% to +1.39%)
↳ prove-recursion		266.79 ms	262.48 ms	-1.62% (-3.52% to -0.91%)
↳ reduce-claims		10.31 ms	10.22 ms	-0.81% (-3.84% to +0.06%)
↳ reduce-in-rows		1.45 ms	1.42 ms	-2.18% (-3.55% to +4.03%)
aggregate-leanxmss-100-2to1-8thread-first	arm64	165.58 ms	163.76 ms	-1.11% (-1.58% to -0.86%)
↳ build-circuit		10.50 ms	10.51 ms	-0.11% (-1.13% to +0.79%)
↳ prove-recursion		142.73 ms	140.68 ms	-1.50% (-1.67% to -0.98%)
↳ reduce-claims		4.10 ms	4.13 ms	+0.70% (-1.46% to +8.19%)
↳ reduce-in-rows		1.40 ms	1.41 ms	+0.87% (-0.06% to +2.31%)
↳ verify		6.99 ms	6.93 ms	-0.61% (-2.12% to +1.30%)
aggregate-leanxmss-100-2to1-8thread-node	arm64	174.73 ms	171.81 ms	-1.38% (-1.86% to -1.13%)
↳ build-circuit		11.42 ms	11.41 ms	-0.05% (-3.36% to +1.49%)
↳ prove-recursion		148.09 ms	145.46 ms	-1.81% (-1.90% to -1.30%)
↳ reduce-claims		7.66 ms	7.72 ms	+0.81% (-1.49% to +2.25%)
↳ reduce-in-rows		1.44 ms	1.44 ms	-0.48% (-1.71% to +1.32%)
aggregate-leanxmss-100-2to1-node	arm64	112.26 ms	110.91 ms	-0.96% (-2.93% to +0.00%)
↳ build-circuit		11.01 ms	11.21 ms	+0.38% (-0.57% to +2.34%)
↳ prove-recursion		87.15 ms	86.08 ms	-1.11% (-3.45% to -0.50%)
↳ reduce-claims		6.67 ms	6.54 ms	-1.85% (-3.26% to +6.46%)
↳ reduce-in-rows		1.45 ms	1.43 ms	+0.55% (-2.38% to +2.48%)
aggregate-leanxmss-100-2to1-node	x86-64	89.97 ms	88.31 ms	-1.84% (-2.65% to -0.61%)
↳ build-circuit		14.46 ms	14.13 ms	-1.51% (-5.16% to -0.03%)
↳ prove-recursion		56.71 ms	56.00 ms	-1.24% (-1.26% to +0.51%)
↳ reduce-claims		10.74 ms	9.56 ms	-9.80% (-14.86% to +2.77%)
↳ reduce-in-rows		1.73 ms	1.74 ms	+1.10% (-2.65% to +2.48%)
aggregate-leanxmss-100-4to1-1thread-first	arm64	2.00 s	1.97 s	-1.68% (-1.85% to -0.11%)
↳ build-circuit		25.23 ms	25.20 ms	-0.12% (-1.12% to +5.68%)
↳ prove-recursion		1.94 s	1.91 s	-1.73% (-1.92% to -0.21%)
↳ reduce-claims		21.26 ms	21.44 ms	+0.71% (-0.17% to +1.47%)
↳ reduce-in-rows		2.73 ms	2.72 ms	-0.47% (-1.82% to +5.07%)
↳ verify		13.67 ms	13.66 ms	-0.10% (-0.45% to +0.79%)
aggregate-leanxmss-100-4to1-1thread-node	arm64	2.18 s	2.14 s	-1.88% (-2.99% to -1.46%)
↳ build-circuit		29.03 ms	29.06 ms	+0.48% (-0.66% to +2.60%)
↳ prove-recursion		2.08 s	2.04 s	-2.01% (-3.17% to -1.56%)
↳ reduce-claims		55.04 ms	55.34 ms	+0.31% (-1.59% to +0.96%)
↳ reduce-in-rows		2.92 ms	2.91 ms	+1.18% (-3.32% to +4.35%)
aggregate-leanxmss-100-4to1-4thread-first	arm64	555.87 ms	547.78 ms	-1.43% (-2.33% to -0.74%)
↳ build-circuit		21.82 ms	21.39 ms	-0.35% (-5.55% to +0.06%)
↳ prove-recursion		509.77 ms	502.55 ms	-1.50% (-2.53% to -0.44%)
↳ reduce-claims		7.08 ms	6.92 ms	-2.34% (-5.58% to +1.30%)
↳ reduce-in-rows		2.74 ms	2.67 ms	-1.86% (-3.69% to +0.41%)
↳ verify		13.52 ms	13.28 ms	-0.78% (-2.88% to +0.01%)
aggregate-leanxmss-100-4to1-4thread-node	arm64	601.68 ms	595.74 ms	-1.25% (-2.17% to -0.22%)
↳ build-circuit		25.72 ms	25.47 ms	-0.53% (-1.57% to +3.49%)
↳ prove-recursion		543.10 ms	536.38 ms	-1.24% (-2.36% to -1.10%)
↳ reduce-claims		17.04 ms	17.01 ms	-0.15% (-1.80% to +0.87%)
↳ reduce-in-rows		2.92 ms	2.92 ms	-0.96% (-3.12% to +4.70%)
aggregate-leanxmss-100-4to1-8thread-first	arm64	319.72 ms	315.55 ms	-1.03% (-1.30% to -0.45%)
↳ build-circuit		20.86 ms	21.09 ms	+1.09% (-3.21% to +5.61%)
↳ prove-recursion		276.39 ms	272.81 ms	-1.25% (-1.99% to -0.16%)
↳ reduce-claims		5.37 ms	5.23 ms	+0.12% (-6.71% to +11.00%)
↳ reduce-in-rows		2.68 ms	2.71 ms	+1.33% (-2.46% to +5.39%)
↳ verify		13.72 ms	13.61 ms	+0.38% (-2.32% to +3.26%)
aggregate-leanxmss-100-4to1-8thread-node	arm64	348.22 ms	341.98 ms	-1.65% (-2.54% to -0.85%)
↳ build-circuit		24.86 ms	24.77 ms	-0.81% (-9.99% to +2.30%)
↳ prove-recursion		294.63 ms	289.39 ms	-1.77% (-2.52% to -1.22%)
↳ reduce-claims		11.87 ms	11.87 ms	+0.72% (-0.30% to +1.40%)
↳ reduce-in-rows		2.92 ms	2.89 ms	-1.44% (-18.31% to +1.95%)
aggregate-leanxmss-100-4to1-first	arm64	202.83 ms	200.94 ms	-1.06% (-1.25% to -0.81%)
↳ build-circuit		21.12 ms	21.09 ms	-0.17% (-0.98% to +0.87%)
↳ prove-recursion		160.09 ms	158.00 ms	-1.36% (-1.52% to -1.10%)
↳ reduce-claims		4.47 ms	4.64 ms	+4.61% (-4.93% to +7.46%)
↳ reduce-in-rows		2.74 ms	2.72 ms	+0.04% (-2.49% to +0.68%)
↳ verify		14.41 ms	14.37 ms	-0.26% (-0.51% to +0.29%)
aggregate-leanxmss-100-4to1-node	arm64	221.19 ms	219.31 ms	-1.10% (-1.35% to -0.49%)
↳ build-circuit		24.62 ms	24.78 ms	+0.18% (-0.33% to +1.61%)
↳ prove-recursion		171.05 ms	169.16 ms	-1.37% (-1.76% to -0.51%)
↳ reduce-claims		9.29 ms	9.41 ms	-0.02% (-1.65% to +1.83%)
↳ reduce-in-rows		2.91 ms	2.91 ms	+0.49% (-1.00% to +0.97%)
falcon-7	arm64	577.27 ms	567.87 ms	-1.43% (-1.99% to -0.76%)
↳ build-witness		24.99 ms	25.13 ms	+0.04% (-2.27% to +5.12%)
↳ commit		84.36 ms	84.05 ms	-0.25% (-0.83% to +1.52%)
↳ execute-program		31.30 ms	31.40 ms	-0.34% (-5.03% to +12.58%)
↳ flock-reductions		188.28 ms	178.57 ms	-5.06% (-5.79% to -3.85%)
↳ pcs-open		205.09 ms	206.19 ms	+0.51% (-0.29% to +1.37%)
↳ prove-bus		33.42 ms	33.12 ms	-0.40% (-1.98% to +2.88%)
↳ prove-constraints		7.72 ms	7.77 ms	+1.42% (-1.70% to +4.65%)
fibonacci-asm-2000000	arm64	883.68 ms	865.39 ms	-1.89% (-2.07% to -1.52%)
↳ build-witness		35.96 ms	35.96 ms	-0.53% (-8.20% to +0.34%)
↳ commit		128.31 ms	128.34 ms	-0.07% (-1.14% to +0.15%)
↳ execute-program		63.94 ms	65.00 ms	+2.44% (-1.39% to +3.65%)
↳ flock-reductions		309.08 ms	291.98 ms	-5.63% (-6.18% to -5.52%)
↳ pcs-open		280.41 ms	280.71 ms	+0.21% (-0.68% to +1.11%)
↳ prove-bus		53.25 ms	53.29 ms	+0.08% (-0.46% to +10.16%)
↳ prove-constraints		10.77 ms	10.77 ms	+0.01% (-1.05% to +8.66%)
hash-50000	arm64	548.79 ms	541.92 ms	-1.23% (-1.36% to -0.67%)
↳ build-witness		24.51 ms	24.51 ms	-0.03% (-0.37% to +0.09%)
↳ commit		83.66 ms	83.60 ms	-0.03% (-0.61% to +0.18%)
↳ execute-program		23.28 ms	23.15 ms	-0.61% (-1.11% to +1.24%)
↳ flock-reductions		176.14 ms	169.80 ms	-3.59% (-3.89% to -3.19%)
↳ pcs-open		205.61 ms	205.96 ms	+0.22% (-0.07% to +1.13%)
↳ prove-bus		27.81 ms	27.80 ms	-0.38% (-0.70% to +1.41%)
↳ prove-constraints		7.16 ms	7.11 ms	-1.51% (-1.95% to +4.06%)
leansphincs-26	arm64	193.09 ms	190.51 ms	-1.41% (-1.76% to +1.17%)
↳ build-witness		7.10 ms	7.06 ms	-0.54% (-1.58% to -0.08%)
↳ commit		24.28 ms	24.22 ms	-0.14% (-1.34% to +0.70%)
↳ execute-program		11.56 ms	11.54 ms	-0.32% (-1.66% to +2.20%)
↳ flock-reductions		65.73 ms	63.00 ms	-4.32% (-4.95% to -3.11%)
↳ pcs-open		65.54 ms	65.81 ms	+0.44% (-0.05% to +5.88%)
↳ prove-bus		14.31 ms	14.26 ms	-0.04% (-0.88% to +3.73%)
↳ prove-constraints		4.22 ms	4.27 ms	+1.33% (-1.90% to +2.44%)
leanxmss-100	arm64	173.72 ms	171.85 ms	-1.11% (-1.79% to +1.71%)
↳ build-witness		6.03 ms	6.09 ms	+0.66% (-0.19% to +22.89%)
↳ commit		20.84 ms	20.89 ms	+0.42% (-0.37% to +4.57%)
↳ execute-program		10.20 ms	10.16 ms	-0.21% (-0.52% to +0.68%)
↳ flock-reductions		60.94 ms	58.43 ms	-4.08% (-6.06% to -3.76%)
↳ pcs-open		59.21 ms	59.94 ms	+1.58% (-0.44% to +6.37%)
↳ prove-bus		12.60 ms	12.55 ms	-0.21% (-3.49% to -0.13%)
↳ prove-constraints		3.82 ms	3.79 ms	-1.51% (-12.00% to +0.44%)
leanxmss-100	x86-64	111.58 ms	112.05 ms	+0.37% (-0.03% to +0.90%)
↳ build-witness		6.32 ms	6.32 ms	+0.39% (-1.24% to +2.64%)
↳ commit		11.65 ms	11.71 ms	+0.50% (+0.33% to +1.54%)
↳ execute-program		13.20 ms	13.25 ms	+0.33% (-2.15% to +3.01%)
↳ flock-reductions		33.62 ms	33.88 ms	+1.02% (-0.51% to +1.38%)
↳ pcs-open		30.49 ms	30.80 ms	+1.13% (-0.95% to +1.46%)
↳ prove-bus		11.35 ms	11.18 ms	-1.36% (-1.65% to -0.43%)
↳ prove-constraints		4.64 ms	4.64 ms	+0.22% (-3.04% to +0.77%)
leanxmss-100-1thread	arm64	1.94 s	1.91 s	-2.00% (-2.42% to -1.89%)
↳ build-witness		66.14 ms	66.46 ms	+0.23% (+0.01% to +0.68%)
↳ commit		303.51 ms	304.21 ms	+0.70% (-0.95% to +0.74%)
↳ execute-program		10.32 ms	10.29 ms	+0.68% (-0.71% to +16.13%)
↳ flock-reductions		700.47 ms	657.94 ms	-6.05% (-7.25% to -5.92%)
↳ pcs-open		731.01 ms	732.54 ms	+0.51% (-0.05% to +0.97%)
↳ prove-bus		118.46 ms	118.56 ms	+0.18% (-3.17% to +0.54%)
↳ prove-constraints		14.89 ms	14.95 ms	-0.48% (-1.03% to +2.46%)
leanxmss-100-4thread	arm64	513.46 ms	502.76 ms	-1.92% (-2.65% to -1.40%)
↳ build-witness		17.76 ms	17.71 ms	-0.50% (-0.75% to +0.55%)
↳ commit		76.17 ms	76.61 ms	+0.49% (+0.10% to +0.90%)
↳ execute-program		10.20 ms	10.18 ms	-0.09% (-2.90% to +0.45%)
↳ flock-reductions		184.48 ms	173.34 ms	-6.10% (-6.58% to -5.32%)
↳ pcs-open		186.24 ms	187.01 ms	-0.09% (-1.12% to +1.77%)
↳ prove-bus		32.39 ms	32.41 ms	+0.25% (-0.28% to +0.64%)
↳ prove-constraints		5.50 ms	5.49 ms	-0.18% (-0.70% to +0.18%)
leanxmss-100-8thread	arm64	286.72 ms	281.70 ms	-1.75% (-2.43% to -1.21%)
↳ build-witness		9.90 ms	9.90 ms	-0.13% (-1.54% to +0.36%)
↳ commit		39.28 ms	39.47 ms	-0.00% (-0.40% to +2.22%)
↳ execute-program		10.18 ms	10.22 ms	-0.54% (-1.98% to +2.24%)
↳ flock-reductions		101.51 ms	96.23 ms	-5.17% (-5.36% to -5.13%)
↳ pcs-open		102.41 ms	102.40 ms	+0.29% (-1.65% to +1.39%)
↳ prove-bus		19.08 ms	19.12 ms	-0.40% (-1.05% to +1.13%)
↳ prove-constraints		4.16 ms	4.21 ms	+1.45% (-1.07% to +3.04%)
shielded-258	arm64	776.63 ms	764.61 ms	-1.52% (-1.77% to -1.39%)
↳ build-witness		31.37 ms	31.34 ms	-0.09% (-2.88% to +0.48%)
↳ commit		115.35 ms	115.31 ms	-0.04% (-0.42% to +0.24%)
↳ execute-program		56.72 ms	56.67 ms	-0.04% (-0.80% to +0.24%)
↳ flock-reductions		249.57 ms	237.58 ms	-4.87% (-4.99% to -4.62%)
↳ pcs-open		258.72 ms	258.91 ms	+0.20% (-0.46% to +0.39%)
↳ prove-bus		54.31 ms	54.34 ms	-0.01% (-0.50% to +1.08%)
↳ prove-constraints		9.91 ms	10.08 ms	+0.97% (+0.01% to +2.37%)
shielded-258	x86-64	435.86 ms	442.94 ms	+1.50% (-0.51% to +2.68%)
↳ build-witness		32.85 ms	32.64 ms	+0.20% (-0.87% to +1.68%)
↳ commit		56.99 ms	57.17 ms	+0.78% (-2.46% to +2.34%)
↳ execute-program		66.10 ms	67.77 ms	+2.48% (+1.89% to +6.82%)
↳ flock-reductions		109.32 ms	110.04 ms	+0.44% (-0.56% to +1.75%)
↳ pcs-open		122.26 ms	124.14 ms	+1.64% (-1.63% to +2.92%)
↳ prove-bus		36.84 ms	37.90 ms	+2.17% (+0.45% to +14.55%)
↳ prove-constraints		11.87 ms	12.32 ms	+3.61% (+1.22% to +4.64%)
stateproof-5	arm64	1.28 s	1.27 s	-1.20% (-2.46% to -0.52%)
↳ build-witness		51.98 ms	52.33 ms	+0.24% (-1.96% to +3.01%)
↳ commit		192.27 ms	192.31 ms	+0.35% (-2.01% to +1.21%)
↳ execute-program		76.73 ms	76.07 ms	+0.88% (-13.54% to +10.21%)
↳ flock-reductions		389.21 ms	367.98 ms	-4.45% (-5.75% to -2.67%)
↳ pcs-open		468.65 ms	471.18 ms	+0.11% (-2.16% to +3.09%)
↳ prove-bus		84.18 ms	83.99 ms	-0.11% (-1.67% to +2.30%)
↳ prove-constraints		15.94 ms	15.58 ms	-0.95% (-3.89% to +1.25%)

Verifying time

benchmark	runner	base	this PR	change
aggregate-leanxmss-100-2to1-1thread-node	arm64	7.27 ms	7.22 ms	-2.54% (-23.80% to -0.66%)
stateproof-5	x86-64	5.10 ms	5.62 ms	+4.96% (+0.32% to +14.65%)

Every result

Proving time

benchmark	runner	CPU	base	this PR	change
aggregate-leanxmss-100-2to1-1thread-first	arm64	Neoverse-V2	998.63 ms	982.90 ms	-1.45% (-4.35% to -1.25%)
aggregate-leanxmss-100-2to1-1thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	605.86 ms	608.95 ms	-0.01% (-0.25% to +1.63%)
aggregate-leanxmss-100-2to1-1thread-node	arm64	Neoverse-V2	1.05 s	1.03 s	-1.92% (-6.54% to -1.57%)
aggregate-leanxmss-100-2to1-1thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	639.61 ms	640.48 ms	+0.08% (-1.06% to +0.93%)
aggregate-leanxmss-100-2to1-4thread-first	arm64	Neoverse-V2	281.95 ms	279.44 ms	-1.03% (-1.46% to +0.16%)
aggregate-leanxmss-100-2to1-4thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	189.65 ms	190.63 ms	-0.07% (-1.05% to +5.68%)
aggregate-leanxmss-100-2to1-4thread-node	arm64	Neoverse-V2	296.01 ms	291.77 ms	-1.42% (-3.46% to -0.75%)
aggregate-leanxmss-100-2to1-4thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	199.02 ms	199.73 ms	+0.34% (-0.48% to +0.57%)
aggregate-leanxmss-100-2to1-8thread-first	arm64	Neoverse-V2	165.58 ms	163.76 ms	-1.11% (-1.58% to -0.86%)
aggregate-leanxmss-100-2to1-8thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	120.15 ms	119.14 ms	-0.95% (-1.52% to +1.92%)
aggregate-leanxmss-100-2to1-8thread-node	arm64	Neoverse-V2	174.73 ms	171.81 ms	-1.38% (-1.86% to -1.13%)
aggregate-leanxmss-100-2to1-8thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	125.51 ms	126.18 ms	+0.33% (-0.83% to +17.71%)
aggregate-leanxmss-100-2to1-first	arm64	Neoverse-V2	107.62 ms	106.76 ms	-0.80% (-1.64% to +3.36%)
aggregate-leanxmss-100-2to1-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	86.35 ms	85.31 ms	-1.82% (-2.57% to +3.49%)
aggregate-leanxmss-100-2to1-node	arm64	Neoverse-V2	112.26 ms	110.91 ms	-0.96% (-2.93% to +0.00%)
aggregate-leanxmss-100-2to1-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	89.97 ms	88.31 ms	-1.84% (-2.65% to -0.61%)
aggregate-leanxmss-100-4to1-1thread-first	arm64	Neoverse-V2	2.00 s	1.97 s	-1.68% (-1.85% to -0.11%)
aggregate-leanxmss-100-4to1-1thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	1.11 s	1.12 s	-0.28% (-0.90% to +3.33%)
aggregate-leanxmss-100-4to1-1thread-node	arm64	Neoverse-V2	2.18 s	2.14 s	-1.88% (-2.99% to -1.46%)
aggregate-leanxmss-100-4to1-1thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	1.22 s	1.22 s	+0.37% (-2.03% to +2.27%)
aggregate-leanxmss-100-4to1-4thread-first	arm64	Neoverse-V2	555.87 ms	547.78 ms	-1.43% (-2.33% to -0.74%)
aggregate-leanxmss-100-4to1-4thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	350.65 ms	352.08 ms	+0.41% (-0.22% to +1.25%)
aggregate-leanxmss-100-4to1-4thread-node	arm64	Neoverse-V2	601.68 ms	595.74 ms	-1.25% (-2.17% to -0.22%)
aggregate-leanxmss-100-4to1-4thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	383.54 ms	382.96 ms	-0.11% (-7.52% to +1.99%)
aggregate-leanxmss-100-4to1-8thread-first	arm64	Neoverse-V2	319.72 ms	315.55 ms	-1.03% (-1.30% to -0.45%)
aggregate-leanxmss-100-4to1-8thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	236.63 ms	238.95 ms	+1.09% (-1.58% to +6.38%)
aggregate-leanxmss-100-4to1-8thread-node	arm64	Neoverse-V2	348.22 ms	341.98 ms	-1.65% (-2.54% to -0.85%)
aggregate-leanxmss-100-4to1-8thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	257.49 ms	258.35 ms	+0.25% (-0.27% to +4.96%)
aggregate-leanxmss-100-4to1-first	arm64	Neoverse-V2	202.83 ms	200.94 ms	-1.06% (-1.25% to -0.81%)
aggregate-leanxmss-100-4to1-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	151.49 ms	151.78 ms	-0.26% (-0.43% to +0.64%)
aggregate-leanxmss-100-4to1-node	arm64	Neoverse-V2	221.19 ms	219.31 ms	-1.10% (-1.35% to -0.49%)
aggregate-leanxmss-100-4to1-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	162.77 ms	162.19 ms	-0.01% (-2.53% to +0.75%)
falcon-7	arm64	Neoverse-V2	577.27 ms	567.87 ms	-1.43% (-1.99% to -0.76%)
falcon-7	x86-64	INTEL(R) XEON(R) GOLD 6548N	321.08 ms	321.01 ms	-0.30% (-2.59% to +0.18%)
fibonacci-asm-2000000	arm64	Neoverse-V2	883.68 ms	865.39 ms	-1.89% (-2.07% to -1.52%)
fibonacci-asm-2000000	x86-64	INTEL(R) XEON(R) GOLD 6548N	468.43 ms	469.13 ms	+0.55% (-0.50% to +2.38%)
flock-add-wrapping-batch-262144	arm64	Neoverse-V2	16.35 ms	16.57 ms	+0.55% (-2.41% to +4.68%)
flock-add-wrapping-batch-262144	x86-64	INTEL(R) XEON(R) GOLD 6548N	9.08 ms	8.33 ms	-2.84% (-8.89% to +0.25%)
flock-hash-batch-262144	arm64	Neoverse-V2	841.85 ms	814.84 ms	-2.30% (-3.94% to -1.56%)
flock-hash-batch-262144	x86-64	INTEL(R) XEON(R) GOLD 6548N	346.73 ms	342.68 ms	+0.28% (-1.42% to +2.68%)
flock-mul-widening-batch-262144	arm64	Neoverse-V2	369.92 ms	372.09 ms	+0.58% (+0.26% to +2.95%)
flock-mul-widening-batch-262144	x86-64	INTEL(R) XEON(R) GOLD 6548N	158.39 ms	158.75 ms	+0.47% (-0.82% to +1.18%)
flock-mul-wrapping-batch-262144	arm64	Neoverse-V2	206.80 ms	200.75 ms	-3.00% (-3.34% to -2.18%)
flock-mul-wrapping-batch-262144	x86-64	INTEL(R) XEON(R) GOLD 6548N	85.09 ms	85.21 ms	+0.28% (-0.68% to +0.79%)
hash-50000	arm64	Neoverse-V2	548.79 ms	541.92 ms	-1.23% (-1.36% to -0.67%)
hash-50000	x86-64	INTEL(R) XEON(R) GOLD 6548N	311.38 ms	310.99 ms	-0.13% (-1.43% to +4.96%)
leansphincs-26	arm64	Neoverse-V2	193.09 ms	190.51 ms	-1.41% (-1.76% to +1.17%)
leansphincs-26	x86-64	INTEL(R) XEON(R) GOLD 6548N	127.66 ms	127.82 ms	-0.14% (-0.57% to +1.87%)
leanxmss-100	arm64	Neoverse-V2	173.72 ms	171.85 ms	-1.11% (-1.79% to +1.71%)
leanxmss-100	x86-64	INTEL(R) XEON(R) GOLD 6548N	111.58 ms	112.05 ms	+0.37% (-0.03% to +0.90%)
leanxmss-100-1thread	arm64	Neoverse-V2	1.94 s	1.91 s	-2.00% (-2.42% to -1.89%)
leanxmss-100-1thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	1.04 s	1.04 s	-0.72% (-1.58% to +0.90%)
leanxmss-100-4thread	arm64	Neoverse-V2	513.46 ms	502.76 ms	-1.92% (-2.65% to -1.40%)
leanxmss-100-4thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	308.96 ms	307.00 ms	-0.03% (-0.86% to +0.20%)
leanxmss-100-8thread	arm64	Neoverse-V2	286.72 ms	281.70 ms	-1.75% (-2.43% to -1.21%)
leanxmss-100-8thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	179.15 ms	178.96 ms	-0.13% (-6.46% to +0.51%)
pcs-commit-22	arm64	Neoverse-V2	7.52 ms	7.20 ms	-3.16% (-8.37% to -0.75%)
pcs-commit-22	x86-64	INTEL(R) XEON(R) GOLD 6548N	3.95 ms	3.96 ms	+0.77% (-10.63% to +11.01%)
pcs-open-22	arm64	Neoverse-V2	16.03 ms	16.10 ms	+0.36% (-0.13% to +6.04%)
pcs-open-22	x86-64	INTEL(R) XEON(R) GOLD 6548N	14.83 ms	14.92 ms	+0.79% (+0.22% to +3.34%)
shielded-258	arm64	Neoverse-V2	776.63 ms	764.61 ms	-1.52% (-1.77% to -1.39%)
shielded-258	x86-64	INTEL(R) XEON(R) GOLD 6548N	435.86 ms	442.94 ms	+1.50% (-0.51% to +2.68%)
stateproof-5	arm64	Neoverse-V2	1.28 s	1.27 s	-1.20% (-2.46% to -0.52%)
stateproof-5	x86-64	INTEL(R) XEON(R) GOLD 6548N	708.73 ms	708.96 ms	-0.84% (-1.51% to +0.43%)

Prover stages

benchmark	runner	CPU	base	this PR	change
aggregate-leanxmss-100-2to1-1thread-first	arm64	Neoverse-V2	998.63 ms	982.90 ms	-1.45% (-4.35% to -1.25%)
↳ build-circuit			12.22 ms	12.23 ms	-0.02% (-4.82% to +0.51%)
↳ prove-recursion			962.68 ms	947.08 ms	-1.49% (-4.38% to -1.34%)
↳ reduce-claims			15.58 ms	15.61 ms	-0.55% (-2.51% to +2.37%)
↳ reduce-in-rows			1.37 ms	1.34 ms	-2.46% (-4.75% to +1.43%)
↳ verify			6.83 ms	6.87 ms	+0.20% (-4.57% to +0.78%)
aggregate-leanxmss-100-2to1-1thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	605.86 ms	608.95 ms	-0.01% (-0.25% to +1.63%)
↳ build-circuit			14.35 ms	14.39 ms	+0.60% (-5.23% to +3.73%)
↳ prove-recursion			563.88 ms	566.51 ms	+0.07% (-0.31% to +1.72%)
↳ reduce-claims			18.28 ms	18.27 ms	+0.01% (-1.85% to +0.64%)
↳ reduce-in-rows			1.43 ms	1.48 ms	+1.85% (+0.77% to +6.42%)
↳ verify			7.79 ms	7.82 ms	+0.24% (-0.40% to +1.00%)
aggregate-leanxmss-100-2to1-1thread-node	arm64	Neoverse-V2	1.05 s	1.03 s	-1.92% (-6.54% to -1.57%)
↳ build-circuit			11.92 ms	11.94 ms	+0.03% (-5.23% to +0.53%)
↳ prove-recursion			999.57 ms	979.62 ms	-2.05% (-6.85% to -1.67%)
↳ reduce-claims			30.19 ms	30.33 ms	+0.85% (-3.74% to +1.44%)
↳ reduce-in-rows			1.43 ms	1.45 ms	+1.23% (-3.30% to +7.93%)
aggregate-leanxmss-100-2to1-1thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	639.61 ms	640.48 ms	+0.08% (-1.06% to +0.93%)
↳ build-circuit			14.91 ms	14.93 ms	+0.11% (-4.07% to +8.83%)
↳ prove-recursion			581.76 ms	582.77 ms	+0.18% (-0.90% to +0.86%)
↳ reduce-claims			33.18 ms	33.15 ms	+0.31% (-2.17% to +2.97%)
↳ reduce-in-rows			1.55 ms	1.65 ms	+6.13% (-1.73% to +11.50%)
aggregate-leanxmss-100-2to1-4thread-first	arm64	Neoverse-V2	281.95 ms	279.44 ms	-1.03% (-1.46% to +0.16%)
↳ build-circuit			10.76 ms	10.82 ms	+1.14% (-1.39% to +3.52%)
↳ prove-recursion			257.80 ms	255.06 ms	-1.23% (-1.59% to -0.08%)
↳ reduce-claims			5.24 ms	5.27 ms	+0.52% (-1.15% to +1.98%)
↳ reduce-in-rows			1.42 ms	1.40 ms	-1.87% (-4.81% to +5.16%)
↳ verify			6.83 ms	6.83 ms	+0.42% (-0.42% to +1.90%)
aggregate-leanxmss-100-2to1-4thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	189.65 ms	190.63 ms	-0.07% (-1.05% to +5.68%)
↳ build-circuit			14.62 ms	14.47 ms	-1.00% (-3.00% to +0.36%)
↳ prove-recursion			158.82 ms	159.38 ms	+0.49% (-1.02% to +6.25%)
↳ reduce-claims			6.72 ms	7.32 ms	+5.54% (-17.21% to +18.09%)
↳ reduce-in-rows			1.73 ms	1.75 ms	+0.38% (-1.61% to +4.67%)
↳ verify			7.76 ms	7.79 ms	+0.28% (-0.97% to +1.05%)
aggregate-leanxmss-100-2to1-4thread-node	arm64	Neoverse-V2	296.01 ms	291.77 ms	-1.42% (-3.46% to -0.75%)
↳ build-circuit			11.60 ms	11.43 ms	-2.39% (-2.83% to +1.39%)
↳ prove-recursion			266.79 ms	262.48 ms	-1.62% (-3.52% to -0.91%)
↳ reduce-claims			10.31 ms	10.22 ms	-0.81% (-3.84% to +0.06%)
↳ reduce-in-rows			1.45 ms	1.42 ms	-2.18% (-3.55% to +4.03%)
aggregate-leanxmss-100-2to1-4thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	199.02 ms	199.73 ms	+0.34% (-0.48% to +0.57%)
↳ build-circuit			15.41 ms	15.44 ms	-0.82% (-3.20% to +4.44%)
↳ prove-recursion			162.38 ms	162.91 ms	+0.33% (-0.28% to +0.57%)
↳ reduce-claims			12.04 ms	12.00 ms	-0.14% (-1.64% to +5.83%)
↳ reduce-in-rows			1.80 ms	1.79 ms	+0.12% (-10.03% to +2.54%)
aggregate-leanxmss-100-2to1-8thread-first	arm64	Neoverse-V2	165.58 ms	163.76 ms	-1.11% (-1.58% to -0.86%)
↳ build-circuit			10.50 ms	10.51 ms	-0.11% (-1.13% to +0.79%)
↳ prove-recursion			142.73 ms	140.68 ms	-1.50% (-1.67% to -0.98%)
↳ reduce-claims			4.10 ms	4.13 ms	+0.70% (-1.46% to +8.19%)
↳ reduce-in-rows			1.40 ms	1.41 ms	+0.87% (-0.06% to +2.31%)
↳ verify			6.99 ms	6.93 ms	-0.61% (-2.12% to +1.30%)
aggregate-leanxmss-100-2to1-8thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	120.15 ms	119.14 ms	-0.95% (-1.52% to +1.92%)
↳ build-circuit			13.33 ms	13.78 ms	+3.12% (-2.72% to +6.06%)
↳ prove-recursion			90.69 ms	90.47 ms	-0.22% (-0.51% to +0.98%)
↳ reduce-claims			6.12 ms	5.82 ms	-4.11% (-31.94% to +24.26%)
↳ reduce-in-rows			1.56 ms	1.64 ms	+5.16% (-3.51% to +12.22%)
↳ verify			7.83 ms	7.85 ms	-0.16% (-1.37% to +1.38%)
aggregate-leanxmss-100-2to1-8thread-node	arm64	Neoverse-V2	174.73 ms	171.81 ms	-1.38% (-1.86% to -1.13%)
↳ build-circuit			11.42 ms	11.41 ms	-0.05% (-3.36% to +1.49%)
↳ prove-recursion			148.09 ms	145.46 ms	-1.81% (-1.90% to -1.30%)
↳ reduce-claims			7.66 ms	7.72 ms	+0.81% (-1.49% to +2.25%)
↳ reduce-in-rows			1.44 ms	1.44 ms	-0.48% (-1.71% to +1.32%)
aggregate-leanxmss-100-2to1-8thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	125.51 ms	126.18 ms	+0.33% (-0.83% to +17.71%)
↳ build-circuit			14.44 ms	14.66 ms	+0.53% (-4.55% to +3.29%)
↳ prove-recursion			92.56 ms	92.55 ms	+0.01% (-0.49% to +22.34%)
↳ reduce-claims			9.95 ms	10.49 ms	+2.81% (-6.02% to +18.77%)
↳ reduce-in-rows			1.71 ms	1.72 ms	+1.15% (-2.63% to +6.85%)
aggregate-leanxmss-100-2to1-first	arm64	Neoverse-V2	107.62 ms	106.76 ms	-0.80% (-1.64% to +3.36%)
↳ build-circuit			10.17 ms	10.21 ms	+0.07% (-0.14% to +2.05%)
↳ prove-recursion			84.64 ms	83.92 ms	-0.79% (-1.75% to +3.94%)
↳ reduce-claims			4.03 ms	3.82 ms	-5.64% (-11.47% to +6.34%)
↳ reduce-in-rows			1.38 ms	1.40 ms	-0.03% (-1.48% to +2.46%)
↳ verify			7.22 ms	7.21 ms	-0.08% (-0.74% to +1.05%)
aggregate-leanxmss-100-2to1-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	86.35 ms	85.31 ms	-1.82% (-2.57% to +3.49%)
↳ build-circuit			13.53 ms	13.46 ms	+0.64% (-3.52% to +10.02%)
↳ prove-recursion			55.32 ms	55.21 ms	-0.52% (-1.68% to +1.25%)
↳ reduce-claims			7.22 ms	6.60 ms	-11.15% (-23.54% to +13.16%)
↳ reduce-in-rows			1.61 ms	1.62 ms	-0.77% (-3.20% to +1.96%)
↳ verify			8.58 ms	8.54 ms	-0.88% (-1.38% to +1.06%)
aggregate-leanxmss-100-2to1-node	arm64	Neoverse-V2	112.26 ms	110.91 ms	-0.96% (-2.93% to +0.00%)
↳ build-circuit			11.01 ms	11.21 ms	+0.38% (-0.57% to +2.34%)
↳ prove-recursion			87.15 ms	86.08 ms	-1.11% (-3.45% to -0.50%)
↳ reduce-claims			6.67 ms	6.54 ms	-1.85% (-3.26% to +6.46%)
↳ reduce-in-rows			1.45 ms	1.43 ms	+0.55% (-2.38% to +2.48%)
aggregate-leanxmss-100-2to1-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	89.97 ms	88.31 ms	-1.84% (-2.65% to -0.61%)
↳ build-circuit			14.46 ms	14.13 ms	-1.51% (-5.16% to -0.03%)
↳ prove-recursion			56.71 ms	56.00 ms	-1.24% (-1.26% to +0.51%)
↳ reduce-claims			10.74 ms	9.56 ms	-9.80% (-14.86% to +2.77%)
↳ reduce-in-rows			1.73 ms	1.74 ms	+1.10% (-2.65% to +2.48%)
aggregate-leanxmss-100-4to1-1thread-first	arm64	Neoverse-V2	2.00 s	1.97 s	-1.68% (-1.85% to -0.11%)
↳ build-circuit			25.23 ms	25.20 ms	-0.12% (-1.12% to +5.68%)
↳ prove-recursion			1.94 s	1.91 s	-1.73% (-1.92% to -0.21%)
↳ reduce-claims			21.26 ms	21.44 ms	+0.71% (-0.17% to +1.47%)
↳ reduce-in-rows			2.73 ms	2.72 ms	-0.47% (-1.82% to +5.07%)
↳ verify			13.67 ms	13.66 ms	-0.10% (-0.45% to +0.79%)
aggregate-leanxmss-100-4to1-1thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	1.11 s	1.12 s	-0.28% (-0.90% to +3.33%)
↳ build-circuit			31.66 ms	33.00 ms	+0.08% (-2.21% to +5.93%)
↳ prove-recursion			1.03 s	1.04 s	-0.70% (-0.92% to +3.51%)
↳ reduce-claims			25.09 ms	25.36 ms	+1.05% (-3.60% to +7.58%)
↳ reduce-in-rows			3.35 ms	3.40 ms	+2.86% (-0.53% to +7.69%)
↳ verify			14.35 ms	14.25 ms	-0.93% (-4.55% to +2.14%)
aggregate-leanxmss-100-4to1-1thread-node	arm64	Neoverse-V2	2.18 s	2.14 s	-1.88% (-2.99% to -1.46%)
↳ build-circuit			29.03 ms	29.06 ms	+0.48% (-0.66% to +2.60%)
↳ prove-recursion			2.08 s	2.04 s	-2.01% (-3.17% to -1.56%)
↳ reduce-claims			55.04 ms	55.34 ms	+0.31% (-1.59% to +0.96%)
↳ reduce-in-rows			2.92 ms	2.91 ms	+1.18% (-3.32% to +4.35%)
aggregate-leanxmss-100-4to1-1thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	1.22 s	1.22 s	+0.37% (-2.03% to +2.27%)
↳ build-circuit			40.16 ms	39.93 ms	-0.42% (-5.54% to +6.15%)
↳ prove-recursion			1.10 s	1.09 s	-0.02% (-2.18% to +2.10%)
↳ reduce-claims			60.64 ms	61.46 ms	+2.31% (-1.94% to +7.13%)
↳ reduce-in-rows			3.46 ms	3.57 ms	+4.12% (-6.26% to +5.04%)
aggregate-leanxmss-100-4to1-4thread-first	arm64	Neoverse-V2	555.87 ms	547.78 ms	-1.43% (-2.33% to -0.74%)
↳ build-circuit			21.82 ms	21.39 ms	-0.35% (-5.55% to +0.06%)
↳ prove-recursion			509.77 ms	502.55 ms	-1.50% (-2.53% to -0.44%)
↳ reduce-claims			7.08 ms	6.92 ms	-2.34% (-5.58% to +1.30%)
↳ reduce-in-rows			2.74 ms	2.67 ms	-1.86% (-3.69% to +0.41%)
↳ verify			13.52 ms	13.28 ms	-0.78% (-2.88% to +0.01%)
aggregate-leanxmss-100-4to1-4thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	350.65 ms	352.08 ms	+0.41% (-0.22% to +1.25%)
↳ build-circuit			31.70 ms	32.44 ms	+1.87% (-2.93% to +13.59%)
↳ prove-recursion			291.29 ms	291.71 ms	+0.02% (-0.52% to +0.39%)
↳ reduce-claims			8.85 ms	9.46 ms	+2.02% (-1.64% to +9.73%)
↳ reduce-in-rows			3.61 ms	3.67 ms	+1.78% (-3.97% to +2.93%)
↳ verify			14.85 ms	15.04 ms	+1.37% (-0.01% to +4.50%)
aggregate-leanxmss-100-4to1-4thread-node	arm64	Neoverse-V2	601.68 ms	595.74 ms	-1.25% (-2.17% to -0.22%)
↳ build-circuit			25.72 ms	25.47 ms	-0.53% (-1.57% to +3.49%)
↳ prove-recursion			543.10 ms	536.38 ms	-1.24% (-2.36% to -1.10%)
↳ reduce-claims			17.04 ms	17.01 ms	-0.15% (-1.80% to +0.87%)
↳ reduce-in-rows			2.92 ms	2.92 ms	-0.96% (-3.12% to +4.70%)
aggregate-leanxmss-100-4to1-4thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	383.54 ms	382.96 ms	-0.11% (-7.52% to +1.99%)
↳ build-circuit			36.08 ms	36.79 ms	+1.94% (-3.46% to +12.73%)
↳ prove-recursion			307.40 ms	306.35 ms	-0.27% (-8.31% to -0.07%)
↳ reduce-claims			21.31 ms	21.32 ms	+0.26% (-19.95% to +16.22%)
↳ reduce-in-rows			3.89 ms	3.82 ms	+0.03% (-3.03% to +4.69%)
aggregate-leanxmss-100-4to1-8thread-first	arm64	Neoverse-V2	319.72 ms	315.55 ms	-1.03% (-1.30% to -0.45%)
↳ build-circuit			20.86 ms	21.09 ms	+1.09% (-3.21% to +5.61%)
↳ prove-recursion			276.39 ms	272.81 ms	-1.25% (-1.99% to -0.16%)
↳ reduce-claims			5.37 ms	5.23 ms	+0.12% (-6.71% to +11.00%)
↳ reduce-in-rows			2.68 ms	2.71 ms	+1.33% (-2.46% to +5.39%)
↳ verify			13.72 ms	13.61 ms	+0.38% (-2.32% to +3.26%)
aggregate-leanxmss-100-4to1-8thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	236.63 ms	238.95 ms	+1.09% (-1.58% to +6.38%)
↳ build-circuit			33.45 ms	33.61 ms	-1.25% (-8.17% to +3.52%)
↳ prove-recursion			176.07 ms	177.03 ms	+0.79% (-0.02% to +7.22%)
↳ reduce-claims			7.79 ms	7.57 ms	+3.20% (-16.01% to +14.77%)
↳ reduce-in-rows			3.85 ms	3.84 ms	-0.69% (-3.51% to +2.63%)
↳ verify			15.92 ms	15.83 ms	-0.54% (-6.90% to +2.20%)
aggregate-leanxmss-100-4to1-8thread-node	arm64	Neoverse-V2	348.22 ms	341.98 ms	-1.65% (-2.54% to -0.85%)
↳ build-circuit			24.86 ms	24.77 ms	-0.81% (-9.99% to +2.30%)
↳ prove-recursion			294.63 ms	289.39 ms	-1.77% (-2.52% to -1.22%)
↳ reduce-claims			11.87 ms	11.87 ms	+0.72% (-0.30% to +1.40%)
↳ reduce-in-rows			2.92 ms	2.89 ms	-1.44% (-18.31% to +1.95%)
aggregate-leanxmss-100-4to1-8thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	257.49 ms	258.35 ms	+0.25% (-0.27% to +4.96%)
↳ build-circuit			38.44 ms	37.38 ms	-1.87% (-5.10% to +0.32%)
↳ prove-recursion			184.09 ms	186.76 ms	+0.70% (-1.72% to +8.00%)
↳ reduce-claims			15.19 ms	15.51 ms	-0.66% (-2.68% to +16.75%)
↳ reduce-in-rows			4.21 ms	4.20 ms	-0.22% (-5.00% to +2.01%)
aggregate-leanxmss-100-4to1-first	arm64	Neoverse-V2	202.83 ms	200.94 ms	-1.06% (-1.25% to -0.81%)
↳ build-circuit			21.12 ms	21.09 ms	-0.17% (-0.98% to +0.87%)
↳ prove-recursion			160.09 ms	158.00 ms	-1.36% (-1.52% to -1.10%)
↳ reduce-claims			4.47 ms	4.64 ms	+4.61% (-4.93% to +7.46%)
↳ reduce-in-rows			2.74 ms	2.72 ms	+0.04% (-2.49% to +0.68%)
↳ verify			14.41 ms	14.37 ms	-0.26% (-0.51% to +0.29%)
aggregate-leanxmss-100-4to1-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	151.49 ms	151.78 ms	-0.26% (-0.43% to +0.64%)
↳ build-circuit			31.11 ms	31.19 ms	+0.26% (-0.45% to +0.62%)
↳ prove-recursion			91.74 ms	92.03 ms	+0.40% (-0.87% to +1.08%)
↳ reduce-claims			8.17 ms	7.84 ms	-4.60% (-13.70% to +10.26%)
↳ reduce-in-rows			3.66 ms	3.69 ms	+0.32% (-0.70% to +3.84%)
↳ verify			16.49 ms	16.49 ms	+0.23% (-1.12% to +0.52%)
aggregate-leanxmss-100-4to1-node	arm64	Neoverse-V2	221.19 ms	219.31 ms	-1.10% (-1.35% to -0.49%)
↳ build-circuit			24.62 ms	24.78 ms	+0.18% (-0.33% to +1.61%)
↳ prove-recursion			171.05 ms	169.16 ms	-1.37% (-1.76% to -0.51%)
↳ reduce-claims			9.29 ms	9.41 ms	-0.02% (-1.65% to +1.83%)
↳ reduce-in-rows			2.91 ms	2.91 ms	+0.49% (-1.00% to +0.97%)
aggregate-leanxmss-100-4to1-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	162.77 ms	162.19 ms	-0.01% (-2.53% to +0.75%)
↳ build-circuit			35.96 ms	35.61 ms	-0.96% (-1.53% to +3.02%)
↳ prove-recursion			95.42 ms	95.73 ms	+0.43% (-0.32% to +1.15%)
↳ reduce-claims			12.57 ms	12.58 ms	-3.69% (-22.56% to +7.44%)
↳ reduce-in-rows			4.01 ms	4.01 ms	+0.78% (-1.33% to +2.17%)
falcon-7	arm64	Neoverse-V2	577.27 ms	567.87 ms	-1.43% (-1.99% to -0.76%)
↳ build-witness			24.99 ms	25.13 ms	+0.04% (-2.27% to +5.12%)
↳ commit			84.36 ms	84.05 ms	-0.25% (-0.83% to +1.52%)
↳ execute-program			31.30 ms	31.40 ms	-0.34% (-5.03% to +12.58%)
↳ flock-reductions			188.28 ms	178.57 ms	-5.06% (-5.79% to -3.85%)
↳ pcs-open			205.09 ms	206.19 ms	+0.51% (-0.29% to +1.37%)
↳ prove-bus			33.42 ms	33.12 ms	-0.40% (-1.98% to +2.88%)
↳ prove-constraints			7.72 ms	7.77 ms	+1.42% (-1.70% to +4.65%)
falcon-7	x86-64	INTEL(R) XEON(R) GOLD 6548N	321.08 ms	321.01 ms	-0.30% (-2.59% to +0.18%)
↳ build-witness			21.61 ms	21.28 ms	-1.04% (-2.20% to +1.54%)
↳ commit			43.25 ms	43.13 ms	-0.63% (-0.77% to +0.17%)
↳ execute-program			40.35 ms	40.62 ms	+0.37% (-0.90% to +2.98%)
↳ flock-reductions			83.29 ms	83.28 ms	-0.45% (-4.22% to +0.41%)
↳ pcs-open			99.81 ms	99.64 ms	-0.40% (-8.22% to +0.41%)
↳ prove-bus			24.42 ms	24.31 ms	-0.73% (-1.52% to +0.87%)
↳ prove-constraints			8.55 ms	8.61 ms	+0.57% (-0.61% to +1.79%)
fibonacci-asm-2000000	arm64	Neoverse-V2	883.68 ms	865.39 ms	-1.89% (-2.07% to -1.52%)
↳ build-witness			35.96 ms	35.96 ms	-0.53% (-8.20% to +0.34%)
↳ commit			128.31 ms	128.34 ms	-0.07% (-1.14% to +0.15%)
↳ execute-program			63.94 ms	65.00 ms	+2.44% (-1.39% to +3.65%)
↳ flock-reductions			309.08 ms	291.98 ms	-5.63% (-6.18% to -5.52%)
↳ pcs-open			280.41 ms	280.71 ms	+0.21% (-0.68% to +1.11%)
↳ prove-bus			53.25 ms	53.29 ms	+0.08% (-0.46% to +10.16%)
↳ prove-constraints			10.77 ms	10.77 ms	+0.01% (-1.05% to +8.66%)
fibonacci-asm-2000000	x86-64	INTEL(R) XEON(R) GOLD 6548N	468.43 ms	469.13 ms	+0.55% (-0.50% to +2.38%)
↳ build-witness			34.85 ms	34.87 ms	+0.09% (-0.36% to +1.55%)
↳ commit			62.93 ms	62.63 ms	-0.39% (-1.50% to +0.14%)
↳ execute-program			74.49 ms	73.60 ms	-1.26% (-2.69% to +2.83%)
↳ flock-reductions			118.22 ms	118.43 ms	+0.29% (+0.15% to +7.88%)
↳ pcs-open			131.92 ms	132.20 ms	+0.22% (+0.12% to +8.66%)
↳ prove-bus			33.88 ms	33.88 ms	-0.33% (-0.73% to +0.24%)
↳ prove-constraints			12.24 ms	12.27 ms	-0.45% (-1.22% to +1.95%)
hash-50000	arm64	Neoverse-V2	548.79 ms	541.92 ms	-1.23% (-1.36% to -0.67%)
↳ build-witness			24.51 ms	24.51 ms	-0.03% (-0.37% to +0.09%)
↳ commit			83.66 ms	83.60 ms	-0.03% (-0.61% to +0.18%)
↳ execute-program			23.28 ms	23.15 ms	-0.61% (-1.11% to +1.24%)
↳ flock-reductions			176.14 ms	169.80 ms	-3.59% (-3.89% to -3.19%)
↳ pcs-open			205.61 ms	205.96 ms	+0.22% (-0.07% to +1.13%)
↳ prove-bus			27.81 ms	27.80 ms	-0.38% (-0.70% to +1.41%)
↳ prove-constraints			7.16 ms	7.11 ms	-1.51% (-1.95% to +4.06%)
hash-50000	x86-64	INTEL(R) XEON(R) GOLD 6548N	311.38 ms	310.99 ms	-0.13% (-1.43% to +4.96%)
↳ build-witness			24.28 ms	23.44 ms	-2.12% (-5.67% to +2.24%)
↳ commit			41.71 ms	41.15 ms	-1.37% (-1.93% to +0.47%)
↳ execute-program			30.85 ms	31.48 ms	+1.99% (-1.88% to +32.41%)
↳ flock-reductions			86.27 ms	85.90 ms	+0.04% (-1.01% to +1.95%)
↳ pcs-open			95.82 ms	96.74 ms	+0.27% (-0.87% to +3.60%)
↳ prove-bus			21.75 ms	22.24 ms	+1.06% (-9.50% to +3.23%)
↳ prove-constraints			9.22 ms	9.29 ms	+2.13% (-3.43% to +4.24%)
leansphincs-26	arm64	Neoverse-V2	193.09 ms	190.51 ms	-1.41% (-1.76% to +1.17%)
↳ build-witness			7.10 ms	7.06 ms	-0.54% (-1.58% to -0.08%)
↳ commit			24.28 ms	24.22 ms	-0.14% (-1.34% to +0.70%)
↳ execute-program			11.56 ms	11.54 ms	-0.32% (-1.66% to +2.20%)
↳ flock-reductions			65.73 ms	63.00 ms	-4.32% (-4.95% to -3.11%)
↳ pcs-open			65.54 ms	65.81 ms	+0.44% (-0.05% to +5.88%)
↳ prove-bus			14.31 ms	14.26 ms	-0.04% (-0.88% to +3.73%)
↳ prove-constraints			4.22 ms	4.27 ms	+1.33% (-1.90% to +2.44%)
leansphincs-26	x86-64	INTEL(R) XEON(R) GOLD 6548N	127.66 ms	127.82 ms	-0.14% (-0.57% to +1.87%)
↳ build-witness			7.82 ms	7.90 ms	+1.50% (-7.41% to +2.82%)
↳ commit			13.25 ms	13.31 ms	+0.52% (-0.39% to +2.95%)
↳ execute-program			15.81 ms	15.83 ms	+0.30% (-2.20% to +2.35%)
↳ flock-reductions			36.83 ms	37.04 ms	+0.58% (-3.01% to +2.95%)
↳ pcs-open			35.01 ms	34.96 ms	-0.15% (-1.41% to +1.63%)
↳ prove-bus			13.53 ms	13.55 ms	+0.16% (-1.42% to +1.39%)
↳ prove-constraints			5.32 ms	5.36 ms	+0.38% (-0.00% to +4.20%)
leanxmss-100	arm64	Neoverse-V2	173.72 ms	171.85 ms	-1.11% (-1.79% to +1.71%)
↳ build-witness			6.03 ms	6.09 ms	+0.66% (-0.19% to +22.89%)
↳ commit			20.84 ms	20.89 ms	+0.42% (-0.37% to +4.57%)
↳ execute-program			10.20 ms	10.16 ms	-0.21% (-0.52% to +0.68%)
↳ flock-reductions			60.94 ms	58.43 ms	-4.08% (-6.06% to -3.76%)
↳ pcs-open			59.21 ms	59.94 ms	+1.58% (-0.44% to +6.37%)
↳ prove-bus			12.60 ms	12.55 ms	-0.21% (-3.49% to -0.13%)
↳ prove-constraints			3.82 ms	3.79 ms	-1.51% (-12.00% to +0.44%)
leanxmss-100	x86-64	INTEL(R) XEON(R) GOLD 6548N	111.58 ms	112.05 ms	+0.37% (-0.03% to +0.90%)
↳ build-witness			6.32 ms	6.32 ms	+0.39% (-1.24% to +2.64%)
↳ commit			11.65 ms	11.71 ms	+0.50% (+0.33% to +1.54%)
↳ execute-program			13.20 ms	13.25 ms	+0.33% (-2.15% to +3.01%)
↳ flock-reductions			33.62 ms	33.88 ms	+1.02% (-0.51% to +1.38%)
↳ pcs-open			30.49 ms	30.80 ms	+1.13% (-0.95% to +1.46%)
↳ prove-bus			11.35 ms	11.18 ms	-1.36% (-1.65% to -0.43%)
↳ prove-constraints			4.64 ms	4.64 ms	+0.22% (-3.04% to +0.77%)
leanxmss-100-1thread	arm64	Neoverse-V2	1.94 s	1.91 s	-2.00% (-2.42% to -1.89%)
↳ build-witness			66.14 ms	66.46 ms	+0.23% (+0.01% to +0.68%)
↳ commit			303.51 ms	304.21 ms	+0.70% (-0.95% to +0.74%)
↳ execute-program			10.32 ms	10.29 ms	+0.68% (-0.71% to +16.13%)
↳ flock-reductions			700.47 ms	657.94 ms	-6.05% (-7.25% to -5.92%)
↳ pcs-open			731.01 ms	732.54 ms	+0.51% (-0.05% to +0.97%)
↳ prove-bus			118.46 ms	118.56 ms	+0.18% (-3.17% to +0.54%)
↳ prove-constraints			14.89 ms	14.95 ms	-0.48% (-1.03% to +2.46%)
leanxmss-100-1thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	1.04 s	1.04 s	-0.72% (-1.58% to +0.90%)
↳ build-witness			75.29 ms	75.40 ms	-0.71% (-1.62% to +0.93%)
↳ commit			189.03 ms	188.79 ms	-0.99% (-1.72% to +0.80%)
↳ execute-program			13.28 ms	13.11 ms	+0.72% (-2.25% to +4.60%)
↳ flock-reductions			304.37 ms	302.07 ms	-0.34% (-1.33% to -0.08%)
↳ pcs-open			358.26 ms	357.33 ms	-0.44% (-2.23% to +0.99%)
↳ prove-bus			85.28 ms	85.35 ms	-0.50% (-4.60% to +4.88%)
↳ prove-constraints			18.48 ms	18.84 ms	+1.48% (-0.00% to +2.21%)
leanxmss-100-4thread	arm64	Neoverse-V2	513.46 ms	502.76 ms	-1.92% (-2.65% to -1.40%)
↳ build-witness			17.76 ms	17.71 ms	-0.50% (-0.75% to +0.55%)
↳ commit			76.17 ms	76.61 ms	+0.49% (+0.10% to +0.90%)
↳ execute-program			10.20 ms	10.18 ms	-0.09% (-2.90% to +0.45%)
↳ flock-reductions			184.48 ms	173.34 ms	-6.10% (-6.58% to -5.32%)
↳ pcs-open			186.24 ms	187.01 ms	-0.09% (-1.12% to +1.77%)
↳ prove-bus			32.39 ms	32.41 ms	+0.25% (-0.28% to +0.64%)
↳ prove-constraints			5.50 ms	5.49 ms	-0.18% (-0.70% to +0.18%)
leanxmss-100-4thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	308.96 ms	307.00 ms	-0.03% (-0.86% to +0.20%)
↳ build-witness			21.89 ms	21.76 ms	+0.02% (-1.56% to +2.44%)
↳ commit			50.23 ms	50.28 ms	-0.27% (-1.36% to +1.37%)
↳ execute-program			13.97 ms	13.86 ms	+0.44% (-2.50% to +2.23%)
↳ flock-reductions			88.17 ms	87.83 ms	-0.52% (-0.89% to +0.25%)
↳ pcs-open			99.04 ms	98.59 ms	+0.30% (-1.34% to +0.83%)
↳ prove-bus			27.65 ms	27.60 ms	-0.46% (-2.24% to +0.95%)
↳ prove-constraints			7.62 ms	7.59 ms	-0.23% (-2.72% to +0.76%)
leanxmss-100-8thread	arm64	Neoverse-V2	286.72 ms	281.70 ms	-1.75% (-2.43% to -1.21%)
↳ build-witness			9.90 ms	9.90 ms	-0.13% (-1.54% to +0.36%)
↳ commit			39.28 ms	39.47 ms	-0.00% (-0.40% to +2.22%)
↳ execute-program			10.18 ms	10.22 ms	-0.54% (-1.98% to +2.24%)
↳ flock-reductions			101.51 ms	96.23 ms	-5.17% (-5.36% to -5.13%)
↳ pcs-open			102.41 ms	102.40 ms	+0.29% (-1.65% to +1.39%)
↳ prove-bus			19.08 ms	19.12 ms	-0.40% (-1.05% to +1.13%)
↳ prove-constraints			4.16 ms	4.21 ms	+1.45% (-1.07% to +3.04%)
leanxmss-100-8thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	179.15 ms	178.96 ms	-0.13% (-6.46% to +0.51%)
↳ build-witness			12.15 ms	12.16 ms	+0.07% (-8.55% to +1.37%)
↳ commit			24.83 ms	24.87 ms	-0.08% (-10.82% to +0.60%)
↳ execute-program			13.36 ms	13.19 ms	+0.22% (-3.37% to +3.19%)
↳ flock-reductions			48.96 ms	49.02 ms	+0.12% (-9.78% to +0.36%)
↳ pcs-open			57.85 ms	57.73 ms	-0.16% (-2.43% to +0.31%)
↳ prove-bus			16.56 ms	16.50 ms	+0.44% (-7.05% to +0.86%)
↳ prove-constraints			5.35 ms	5.35 ms	+0.93% (-4.37% to +2.78%)
shielded-258	arm64	Neoverse-V2	776.63 ms	764.61 ms	-1.52% (-1.77% to -1.39%)
↳ build-witness			31.37 ms	31.34 ms	-0.09% (-2.88% to +0.48%)
↳ commit			115.35 ms	115.31 ms	-0.04% (-0.42% to +0.24%)
↳ execute-program			56.72 ms	56.67 ms	-0.04% (-0.80% to +0.24%)
↳ flock-reductions			249.57 ms	237.58 ms	-4.87% (-4.99% to -4.62%)
↳ pcs-open			258.72 ms	258.91 ms	+0.20% (-0.46% to +0.39%)
↳ prove-bus			54.31 ms	54.34 ms	-0.01% (-0.50% to +1.08%)
↳ prove-constraints			9.91 ms	10.08 ms	+0.97% (+0.01% to +2.37%)
shielded-258	x86-64	INTEL(R) XEON(R) GOLD 6548N	435.86 ms	442.94 ms	+1.50% (-0.51% to +2.68%)
↳ build-witness			32.85 ms	32.64 ms	+0.20% (-0.87% to +1.68%)
↳ commit			56.99 ms	57.17 ms	+0.78% (-2.46% to +2.34%)
↳ execute-program			66.10 ms	67.77 ms	+2.48% (+1.89% to +6.82%)
↳ flock-reductions			109.32 ms	110.04 ms	+0.44% (-0.56% to +1.75%)
↳ pcs-open			122.26 ms	124.14 ms	+1.64% (-1.63% to +2.92%)
↳ prove-bus			36.84 ms	37.90 ms	+2.17% (+0.45% to +14.55%)
↳ prove-constraints			11.87 ms	12.32 ms	+3.61% (+1.22% to +4.64%)
stateproof-5	arm64	Neoverse-V2	1.28 s	1.27 s	-1.20% (-2.46% to -0.52%)
↳ build-witness			51.98 ms	52.33 ms	+0.24% (-1.96% to +3.01%)
↳ commit			192.27 ms	192.31 ms	+0.35% (-2.01% to +1.21%)
↳ execute-program			76.73 ms	76.07 ms	+0.88% (-13.54% to +10.21%)
↳ flock-reductions			389.21 ms	367.98 ms	-4.45% (-5.75% to -2.67%)
↳ pcs-open			468.65 ms	471.18 ms	+0.11% (-2.16% to +3.09%)
↳ prove-bus			84.18 ms	83.99 ms	-0.11% (-1.67% to +2.30%)
↳ prove-constraints			15.94 ms	15.58 ms	-0.95% (-3.89% to +1.25%)
stateproof-5	x86-64	INTEL(R) XEON(R) GOLD 6548N	708.73 ms	708.96 ms	-0.84% (-1.51% to +0.43%)
↳ build-witness			51.71 ms	51.55 ms	+0.27% (-3.35% to +1.08%)
↳ commit			97.78 ms	97.02 ms	-1.24% (-2.15% to +0.38%)
↳ execute-program			99.11 ms	97.31 ms	-1.56% (-5.37% to +2.27%)
↳ flock-reductions			170.33 ms	168.65 ms	-0.64% (-3.60% to +0.07%)
↳ pcs-open			216.25 ms	217.16 ms	-0.02% (-3.53% to +3.12%)
↳ prove-bus			55.37 ms	55.29 ms	+0.30% (-2.17% to +3.05%)
↳ prove-constraints			17.76 ms	17.72 ms	+0.37% (-0.91% to +2.27%)

Verifying time

benchmark	runner	CPU	base	this PR	change
aggregate-leanxmss-100-2to1-1thread-first	arm64	Neoverse-V2	4.64 ms	4.62 ms	-0.29% (-18.23% to +0.18%)
aggregate-leanxmss-100-2to1-1thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.46 ms	4.47 ms	+0.24% (-1.11% to +0.94%)
aggregate-leanxmss-100-2to1-1thread-node	arm64	Neoverse-V2	7.27 ms	7.22 ms	-2.54% (-23.80% to -0.66%)
aggregate-leanxmss-100-2to1-1thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	6.99 ms	7.04 ms	-0.12% (-5.39% to +5.93%)
aggregate-leanxmss-100-2to1-4thread-first	arm64	Neoverse-V2	3.61 ms	3.64 ms	+0.93% (-3.14% to +2.28%)
aggregate-leanxmss-100-2to1-4thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	3.75 ms	3.83 ms	+2.38% (-0.59% to +6.78%)
aggregate-leanxmss-100-2to1-4thread-node	arm64	Neoverse-V2	4.35 ms	4.28 ms	-1.54% (-2.82% to +0.94%)
aggregate-leanxmss-100-2to1-4thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.78 ms	4.75 ms	-0.28% (-1.85% to +0.14%)
aggregate-leanxmss-100-2to1-8thread-first	arm64	Neoverse-V2	3.55 ms	3.52 ms	-0.62% (-1.33% to +0.21%)
aggregate-leanxmss-100-2to1-8thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	3.69 ms	3.74 ms	+0.97% (-4.12% to +6.17%)
aggregate-leanxmss-100-2to1-8thread-node	arm64	Neoverse-V2	3.93 ms	3.93 ms	-0.07% (-1.23% to +0.98%)
aggregate-leanxmss-100-2to1-8thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.08 ms	4.08 ms	-0.23% (-1.59% to +4.23%)
aggregate-leanxmss-100-2to1-first	arm64	Neoverse-V2	3.43 ms	3.45 ms	+0.51% (-1.07% to +2.03%)
aggregate-leanxmss-100-2to1-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.17 ms	4.20 ms	+0.90% (-5.74% to +2.91%)
aggregate-leanxmss-100-2to1-node	arm64	Neoverse-V2	3.69 ms	3.70 ms	+0.39% (-0.48% to +1.32%)
aggregate-leanxmss-100-2to1-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.10 ms	4.12 ms	+0.41% (-2.67% to +2.40%)
aggregate-leanxmss-100-4to1-1thread-first	arm64	Neoverse-V2	5.52 ms	5.50 ms	-0.19% (-0.52% to +3.61%)
aggregate-leanxmss-100-4to1-1thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	5.63 ms	5.68 ms	+2.26% (-4.06% to +7.66%)
aggregate-leanxmss-100-4to1-1thread-node	arm64	Neoverse-V2	10.83 ms	10.76 ms	-0.61% (-4.56% to +2.09%)
aggregate-leanxmss-100-4to1-1thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	11.94 ms	11.69 ms	-0.95% (-4.85% to +4.35%)
aggregate-leanxmss-100-4to1-4thread-first	arm64	Neoverse-V2	3.95 ms	3.96 ms	-0.32% (-0.54% to +1.72%)
aggregate-leanxmss-100-4to1-4thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.27 ms	4.12 ms	-3.62% (-5.56% to +2.97%)
aggregate-leanxmss-100-4to1-4thread-node	arm64	Neoverse-V2	5.35 ms	5.49 ms	+1.68% (-5.78% to +2.99%)
aggregate-leanxmss-100-4to1-4thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	6.12 ms	6.10 ms	-0.52% (-3.23% to +3.15%)
aggregate-leanxmss-100-4to1-8thread-first	arm64	Neoverse-V2	3.71 ms	3.81 ms	+1.59% (-0.26% to +5.25%)
aggregate-leanxmss-100-4to1-8thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.09 ms	4.14 ms	+1.10% (-3.17% to +3.82%)
aggregate-leanxmss-100-4to1-8thread-node	arm64	Neoverse-V2	4.56 ms	4.56 ms	-1.07% (-2.64% to +1.57%)
aggregate-leanxmss-100-4to1-8thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	5.24 ms	5.24 ms	+0.37% (-1.47% to +4.39%)
aggregate-leanxmss-100-4to1-first	arm64	Neoverse-V2	3.75 ms	3.72 ms	-0.41% (-1.10% to -0.06%)
aggregate-leanxmss-100-4to1-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.42 ms	4.40 ms	+0.02% (-2.27% to +1.94%)
aggregate-leanxmss-100-4to1-node	arm64	Neoverse-V2	4.25 ms	4.25 ms	-0.47% (-0.83% to +0.47%)
aggregate-leanxmss-100-4to1-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.63 ms	4.67 ms	+0.64% (-2.09% to +1.41%)
falcon-7	arm64	Neoverse-V2	4.30 ms	4.30 ms	-0.91% (-2.85% to +4.64%)
falcon-7	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.72 ms	4.76 ms	+0.92% (-0.22% to +1.92%)
fibonacci-asm-2000000	arm64	Neoverse-V2	4.02 ms	4.02 ms	-0.25% (-0.55% to +0.76%)
fibonacci-asm-2000000	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.54 ms	4.53 ms	-0.08% (-0.75% to +0.94%)
hash-50000	arm64	Neoverse-V2	4.04 ms	4.02 ms	-0.31% (-0.87% to +1.04%)
hash-50000	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.70 ms	4.76 ms	+1.34% (-0.96% to +3.20%)
leansphincs-26	arm64	Neoverse-V2	4.01 ms	4.00 ms	-0.06% (-0.66% to +0.16%)
leansphincs-26	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.70 ms	4.71 ms	-0.48% (-2.01% to +2.31%)
leanxmss-100	arm64	Neoverse-V2	4.00 ms	4.00 ms	+0.01% (-4.45% to +2.10%)
leanxmss-100	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.47 ms	4.45 ms	-0.50% (-1.20% to +1.17%)
leanxmss-100-1thread	arm64	Neoverse-V2	3.98 ms	4.01 ms	-0.14% (-2.41% to +1.12%)
leanxmss-100-1thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.12 ms	4.10 ms	-0.18% (-2.26% to +1.02%)
leanxmss-100-4thread	arm64	Neoverse-V2	3.76 ms	3.75 ms	-0.06% (-0.54% to +0.40%)
leanxmss-100-4thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.36 ms	4.29 ms	-1.55% (-4.06% to +0.44%)
leanxmss-100-8thread	arm64	Neoverse-V2	3.91 ms	3.88 ms	-1.00% (-4.04% to +4.02%)
leanxmss-100-8thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.44 ms	4.37 ms	-0.46% (-3.59% to +0.50%)
shielded-258	arm64	Neoverse-V2	4.17 ms	4.17 ms	-0.21% (-0.44% to +0.71%)
shielded-258	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.61 ms	4.64 ms	+1.29% (-0.43% to +1.58%)
stateproof-5	arm64	Neoverse-V2	4.59 ms	4.54 ms	-0.43% (-2.93% to +0.96%)
stateproof-5	x86-64	INTEL(R) XEON(R) GOLD 6548N	5.10 ms	5.62 ms	+4.96% (+0.32% to +14.65%)

Time per operation

benchmark	runner	CPU	base	this PR	change
blake2s-batch	arm64	Neoverse-V2	2.92 ns	2.92 ns	+0.00% (-0.31% to +0.27%)
blake2s-batch	x86-64	INTEL(R) XEON(R) GOLD 6548N	0.80 ns	0.80 ns	-0.13% (-3.41% to +1.51%)
f192-inv	arm64	Neoverse-V2	244.96 ns	244.97 ns	+0.00% (-0.06% to +0.02%)
f192-inv	x86-64	INTEL(R) XEON(R) GOLD 6548N	266.14 ns	266.05 ns	-0.13% (-0.45% to +0.81%)
f192-mul-latency	arm64	Neoverse-V2	8.03 ns	8.03 ns	-0.21% (-0.68% to +0.54%)
f192-mul-latency	x86-64	INTEL(R) XEON(R) GOLD 6548N	7.58 ns	7.56 ns	-0.18% (-0.89% to +0.29%)
f192-mul-throughput	arm64	Neoverse-V2	2.94 ns	2.94 ns	+0.00% (-0.03% to +0.00%)
f192-mul-throughput	x86-64	INTEL(R) XEON(R) GOLD 6548N	4.46 ns	4.45 ns	-0.13% (-0.45% to +0.51%)
f192-mul4-latency	arm64	Neoverse-V2	12.74 ns	13.08 ns	+0.90% (-2.84% to +3.07%)
f192-mul4-latency	x86-64	INTEL(R) XEON(R) GOLD 6548N	10.12 ns	10.11 ns	-0.20% (-0.39% to +0.60%)
f192-mul4-throughput	arm64	Neoverse-V2	10.77 ns	10.76 ns	-0.10% (-0.27% to +0.11%)
f192-mul4-throughput	x86-64	INTEL(R) XEON(R) GOLD 6548N	15.66 ns	15.65 ns	+0.28% (-4.94% to +1.20%)
f192-square	arm64	Neoverse-V2	1.70 ns	1.70 ns	-0.06% (-0.18% to +0.00%)
f192-square	x86-64	INTEL(R) XEON(R) GOLD 6548N	3.28 ns	3.29 ns	-0.09% (-7.33% to +1.25%)
f64-inv	arm64	Neoverse-V2	211.56 ns	211.54 ns	+0.03% (-0.07% to +0.15%)
f64-inv	x86-64	INTEL(R) XEON(R) GOLD 6548N	237.96 ns	237.57 ns	-0.12% (-0.21% to +0.21%)
f64-mul	arm64	Neoverse-V2	0.56 ns	0.56 ns	+0.00% (-0.36% to +0.00%)
f64-mul	x86-64	INTEL(R) XEON(R) GOLD 6548N	0.87 ns	0.86 ns	-0.12% (-0.46% to +0.12%)
f64-square	arm64	Neoverse-V2	0.55 ns	0.55 ns	-0.18% (-1.09% to +0.00%)
f64-square	x86-64	INTEL(R) XEON(R) GOLD 6548N	1.07 ns	1.07 ns	-0.09% (-1.47% to +0.28%)
ntt-forward-22	arm64	Neoverse-V2	3.70 ms	3.68 ms	-0.62% (-2.56% to +0.38%)
ntt-forward-22	x86-64	INTEL(R) XEON(R) GOLD 6548N	2.21 ms	2.19 ms	-0.95% (-2.27% to +12.89%)
transpose-64bytes	arm64	Neoverse-V2	12.00 ns	12.00 ns	-0.02% (-0.04% to +0.03%)
transpose-64bytes	x86-64	INTEL(R) XEON(R) GOLD 6548N	0.84 ns	0.85 ns	+0.71% (-0.24% to +2.74%)
transpose-64x64	arm64	Neoverse-V2	96.44 ns	96.44 ns	+0.06% (-0.15% to +0.10%)
transpose-64x64	x86-64	INTEL(R) XEON(R) GOLD 6548N	73.32 ns	73.36 ns	-0.14% (-0.23% to +0.25%)

Proof size

benchmark	runner	CPU	base	this PR	change
aggregate-leanxmss-100-2to1-1thread-first	arm64	Neoverse-V2	260,427 B	260,427 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-1thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	260,427 B	260,427 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-1thread-node	arm64	Neoverse-V2	259,659 B	259,659 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-1thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	259,659 B	259,659 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-4thread-first	arm64	Neoverse-V2	260,427 B	260,427 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-4thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	260,427 B	260,427 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-4thread-node	arm64	Neoverse-V2	259,659 B	259,659 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-4thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	259,659 B	259,659 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-8thread-first	arm64	Neoverse-V2	260,427 B	260,427 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-8thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	260,427 B	260,427 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-8thread-node	arm64	Neoverse-V2	259,659 B	259,659 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-8thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	259,659 B	259,659 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-first	arm64	Neoverse-V2	260,427 B	260,427 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	260,427 B	260,427 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-node	arm64	Neoverse-V2	259,659 B	259,659 B	0 B (+0.00%)
aggregate-leanxmss-100-2to1-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	259,659 B	259,659 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-1thread-first	arm64	Neoverse-V2	283,331 B	283,331 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-1thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	283,331 B	283,331 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-1thread-node	arm64	Neoverse-V2	285,027 B	285,027 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-1thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	285,027 B	285,027 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-4thread-first	arm64	Neoverse-V2	283,331 B	283,331 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-4thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	283,331 B	283,331 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-4thread-node	arm64	Neoverse-V2	285,027 B	285,027 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-4thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	285,027 B	285,027 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-8thread-first	arm64	Neoverse-V2	283,331 B	283,331 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-8thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	283,331 B	283,331 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-8thread-node	arm64	Neoverse-V2	285,027 B	285,027 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-8thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	285,027 B	285,027 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-first	arm64	Neoverse-V2	283,331 B	283,331 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	283,331 B	283,331 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-node	arm64	Neoverse-V2	285,027 B	285,027 B	0 B (+0.00%)
aggregate-leanxmss-100-4to1-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	285,027 B	285,027 B	0 B (+0.00%)
falcon-7	arm64	Neoverse-V2	326,382 B	326,382 B	0 B (+0.00%)
falcon-7	x86-64	INTEL(R) XEON(R) GOLD 6548N	326,382 B	326,382 B	0 B (+0.00%)
fibonacci-asm-2000000	arm64	Neoverse-V2	366,742 B	366,742 B	0 B (+0.00%)
fibonacci-asm-2000000	x86-64	INTEL(R) XEON(R) GOLD 6548N	366,742 B	366,742 B	0 B (+0.00%)
hash-50000	arm64	Neoverse-V2	326,742 B	326,742 B	0 B (+0.00%)
hash-50000	x86-64	INTEL(R) XEON(R) GOLD 6548N	326,742 B	326,742 B	0 B (+0.00%)
leansphincs-26	arm64	Neoverse-V2	319,734 B	319,734 B	0 B (+0.00%)
leansphincs-26	x86-64	INTEL(R) XEON(R) GOLD 6548N	319,734 B	319,734 B	0 B (+0.00%)
leanxmss-100	arm64	Neoverse-V2	307,670 B	307,670 B	0 B (+0.00%)
leanxmss-100	x86-64	INTEL(R) XEON(R) GOLD 6548N	307,670 B	307,670 B	0 B (+0.00%)
leanxmss-100-1thread	arm64	Neoverse-V2	307,670 B	307,670 B	0 B (+0.00%)
leanxmss-100-1thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	307,670 B	307,670 B	0 B (+0.00%)
leanxmss-100-4thread	arm64	Neoverse-V2	307,670 B	307,670 B	0 B (+0.00%)
leanxmss-100-4thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	307,670 B	307,670 B	0 B (+0.00%)
leanxmss-100-8thread	arm64	Neoverse-V2	307,670 B	307,670 B	0 B (+0.00%)
leanxmss-100-8thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	307,670 B	307,670 B	0 B (+0.00%)
shielded-258	arm64	Neoverse-V2	357,614 B	357,614 B	0 B (+0.00%)
shielded-258	x86-64	INTEL(R) XEON(R) GOLD 6548N	357,614 B	357,614 B	0 B (+0.00%)
stateproof-5	arm64	Neoverse-V2	350,614 B	350,614 B	0 B (+0.00%)
stateproof-5	x86-64	INTEL(R) XEON(R) GOLD 6548N	350,614 B	350,614 B	0 B (+0.00%)

Peak memory

benchmark	runner	CPU	base	this PR	change
aggregate-leanxmss-100-2to1-1thread-first	arm64	Neoverse-V2	1780.58 MiB	1780.60 MiB	-0.00% (-0.00% to +0.00%)
aggregate-leanxmss-100-2to1-1thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	1176.34 MiB	1174.59 MiB	-0.17% (-0.17% to +0.17%)
aggregate-leanxmss-100-2to1-1thread-node	arm64	Neoverse-V2	1780.83 MiB	1780.85 MiB	-0.00% (-0.00% to +0.00%)
aggregate-leanxmss-100-2to1-1thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	1176.35 MiB	1176.59 MiB	+0.00% (-0.34% to +0.17%)
aggregate-leanxmss-100-2to1-4thread-first	arm64	Neoverse-V2	1831.31 MiB	1832.32 MiB	+0.02% (-0.17% to +0.17%)
aggregate-leanxmss-100-2to1-4thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	1205.02 MiB	1203.88 MiB	-0.12% (-0.21% to +0.19%)
aggregate-leanxmss-100-2to1-4thread-node	arm64	Neoverse-V2	1832.44 MiB	1834.21 MiB	+0.05% (-0.11% to +0.19%)
aggregate-leanxmss-100-2to1-4thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	1210.02 MiB	1211.58 MiB	+0.13% (-0.32% to +0.47%)
aggregate-leanxmss-100-2to1-8thread-first	arm64	Neoverse-V2	1871.82 MiB	1871.05 MiB	-0.05% (-0.16% to +0.35%)
aggregate-leanxmss-100-2to1-8thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	1187.99 MiB	1185.97 MiB	-0.23% (-0.52% to +0.43%)
aggregate-leanxmss-100-2to1-8thread-node	arm64	Neoverse-V2	1875.45 MiB	1875.26 MiB	+0.08% (-0.21% to +0.28%)
aggregate-leanxmss-100-2to1-8thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	1200.86 MiB	1206.35 MiB	+0.29% (-0.30% to +0.68%)
aggregate-leanxmss-100-2to1-first	arm64	Neoverse-V2	1953.71 MiB	1955.91 MiB	+0.25% (-0.41% to +0.56%)
aggregate-leanxmss-100-2to1-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	1294.56 MiB	1295.50 MiB	+0.22% (-0.68% to +0.56%)
aggregate-leanxmss-100-2to1-node	arm64	Neoverse-V2	1980.79 MiB	1981.75 MiB	+0.14% (-0.14% to +0.41%)
aggregate-leanxmss-100-2to1-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	1368.79 MiB	1375.48 MiB	+0.47% (-0.40% to +1.07%)
aggregate-leanxmss-100-4to1-1thread-first	arm64	Neoverse-V2	2266.89 MiB	2266.71 MiB	-0.01% (-0.09% to +0.00%)
aggregate-leanxmss-100-4to1-1thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	1452.62 MiB	1454.62 MiB	-0.02% (-0.14% to +0.29%)
aggregate-leanxmss-100-4to1-1thread-node	arm64	Neoverse-V2	2267.20 MiB	2267.21 MiB	+0.01% (-0.00% to +0.18%)
aggregate-leanxmss-100-4to1-1thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	1458.37 MiB	1460.62 MiB	+0.14% (-0.15% to +0.27%)
aggregate-leanxmss-100-4to1-4thread-first	arm64	Neoverse-V2	2319.74 MiB	2318.86 MiB	-0.04% (-0.19% to +0.12%)
aggregate-leanxmss-100-4to1-4thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	1490.56 MiB	1490.98 MiB	+0.00% (-0.43% to +0.42%)
aggregate-leanxmss-100-4to1-4thread-node	arm64	Neoverse-V2	2320.68 MiB	2320.48 MiB	-0.01% (-0.17% to +0.09%)
aggregate-leanxmss-100-4to1-4thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	1503.35 MiB	1504.93 MiB	+0.16% (-0.31% to +0.30%)
aggregate-leanxmss-100-4to1-8thread-first	arm64	Neoverse-V2	2361.78 MiB	2362.85 MiB	+0.04% (-0.13% to +0.39%)
aggregate-leanxmss-100-4to1-8thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	1192.30 MiB	1195.43 MiB	+0.13% (-0.15% to +0.38%)
aggregate-leanxmss-100-4to1-8thread-node	arm64	Neoverse-V2	2363.84 MiB	2367.36 MiB	+0.15% (-0.06% to +0.53%)
aggregate-leanxmss-100-4to1-8thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	1212.50 MiB	1214.09 MiB	-0.09% (-0.28% to +0.30%)
aggregate-leanxmss-100-4to1-first	arm64	Neoverse-V2	2446.07 MiB	2450.17 MiB	+0.01% (-0.07% to +0.33%)
aggregate-leanxmss-100-4to1-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	1314.55 MiB	1320.09 MiB	+0.15% (-0.58% to +0.42%)
aggregate-leanxmss-100-4to1-node	arm64	Neoverse-V2	2468.88 MiB	2466.69 MiB	-0.05% (-0.24% to +0.17%)
aggregate-leanxmss-100-4to1-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	1398.50 MiB	1396.75 MiB	-0.12% (-1.00% to +0.37%)
falcon-7	arm64	Neoverse-V2	5751.30 MiB	5750.24 MiB	-0.02% (-0.20% to +0.07%)
falcon-7	x86-64	INTEL(R) XEON(R) GOLD 6548N	3209.74 MiB	3206.29 MiB	-0.13% (-0.23% to +0.12%)
fibonacci-asm-2000000	arm64	Neoverse-V2	9071.46 MiB	9072.76 MiB	+0.05% (-0.14% to +0.10%)
fibonacci-asm-2000000	x86-64	INTEL(R) XEON(R) GOLD 6548N	4232.85 MiB	4231.35 MiB	-0.04% (-0.10% to +0.03%)
hash-50000	arm64	Neoverse-V2	5767.00 MiB	5768.74 MiB	+0.05% (-0.18% to +0.12%)
hash-50000	x86-64	INTEL(R) XEON(R) GOLD 6548N	3043.16 MiB	3044.40 MiB	+0.01% (-0.09% to +0.08%)
leansphincs-26	arm64	Neoverse-V2	1983.58 MiB	1981.50 MiB	-0.02% (-0.37% to +0.41%)
leansphincs-26	x86-64	INTEL(R) XEON(R) GOLD 6548N	1021.31 MiB	1021.58 MiB	+0.21% (-0.17% to +0.39%)
leanxmss-100	arm64	Neoverse-V2	1747.89 MiB	1749.93 MiB	+0.07% (-0.63% to +0.38%)
leanxmss-100	x86-64	INTEL(R) XEON(R) GOLD 6548N	877.30 MiB	877.23 MiB	-0.00% (-0.44% to +0.20%)
leanxmss-100-1thread	arm64	Neoverse-V2	1598.61 MiB	1592.55 MiB	-0.13% (-0.61% to +0.63%)
leanxmss-100-1thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	990.36 MiB	990.37 MiB	+0.00% (-0.20% to +0.20%)
leanxmss-100-4thread	arm64	Neoverse-V2	1652.85 MiB	1652.85 MiB	+0.01% (-0.22% to +0.09%)
leanxmss-100-4thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	998.50 MiB	998.47 MiB	-0.02% (-0.45% to +0.40%)
leanxmss-100-8thread	arm64	Neoverse-V2	1689.60 MiB	1689.42 MiB	-0.03% (-0.14% to +0.27%)
leanxmss-100-8thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	778.66 MiB	780.43 MiB	+0.01% (-0.49% to +0.52%)
shielded-258	arm64	Neoverse-V2	8320.09 MiB	8314.39 MiB	-0.02% (-0.09% to +0.08%)
shielded-258	x86-64	INTEL(R) XEON(R) GOLD 6548N	4268.77 MiB	4260.63 MiB	-0.12% (-0.36% to -0.01%)
stateproof-5	arm64	Neoverse-V2	12494.12 MiB	12492.38 MiB	-0.02% (-0.09% to +0.06%)
stateproof-5	x86-64	INTEL(R) XEON(R) GOLD 6548N	6175.10 MiB	6175.23 MiB	+0.00% (-0.00% to +0.06%)

allocations

benchmark	runner	CPU	base	this PR	change
aggregate-leanxmss-100-2to1-1thread-first	arm64	Neoverse-V2	23.34 µs	23.34 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-1thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	23.39 µs	23.39 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-1thread-node	arm64	Neoverse-V2	27.23 µs	27.23 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-1thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	27.27 µs	27.27 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-4thread-first	arm64	Neoverse-V2	23.44 µs	23.44 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-4thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	23.48 µs	23.48 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-4thread-node	arm64	Neoverse-V2	27.33 µs	27.33 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-4thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	27.37 µs	27.37 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-8thread-first	arm64	Neoverse-V2	23.55 µs	23.55 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-8thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	23.60 µs	23.60 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-8thread-node	arm64	Neoverse-V2	27.44 µs	27.44 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-8thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	27.49 µs	27.49 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-first	arm64	Neoverse-V2	23.79 µs	23.79 µs	+0.00% (-0.06% to +0.00%)
aggregate-leanxmss-100-2to1-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	24.25 µs	24.25 µs	+0.00% (-0.00% to +0.01%)
aggregate-leanxmss-100-2to1-node	arm64	Neoverse-V2	27.68 µs	27.68 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	28.14 µs	28.14 µs	+0.00% (+0.00% to +0.01%)
aggregate-leanxmss-100-4to1-1thread-first	arm64	Neoverse-V2	41.83 µs	41.83 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-1thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	41.88 µs	41.88 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-1thread-node	arm64	Neoverse-V2	49.74 µs	49.74 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-1thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	49.78 µs	49.78 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-4thread-first	arm64	Neoverse-V2	41.95 µs	41.95 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-4thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	41.99 µs	41.99 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-4thread-node	arm64	Neoverse-V2	49.85 µs	49.85 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-4thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	49.90 µs	49.90 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-8thread-first	arm64	Neoverse-V2	42.09 µs	42.09 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-8thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	42.14 µs	42.14 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-8thread-node	arm64	Neoverse-V2	50.00 µs	50.00 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-8thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	50.04 µs	50.04 µs	+0.00% (-0.00% to +0.00%)
aggregate-leanxmss-100-4to1-first	arm64	Neoverse-V2	42.36 µs	42.36 µs	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	42.93 µs	42.93 µs	+0.00% (-0.01% to +0.00%)
aggregate-leanxmss-100-4to1-node	arm64	Neoverse-V2	50.26 µs	50.26 µs	+0.00% (-0.04% to +0.00%)
aggregate-leanxmss-100-4to1-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	50.83 µs	50.84 µs	+0.00% (+0.00% to +0.01%)
falcon-7	arm64	Neoverse-V2	15.83 µs	15.83 µs	+0.00% (-0.01% to +0.03%)
falcon-7	x86-64	INTEL(R) XEON(R) GOLD 6548N	17.50 µs	17.50 µs	+0.00% (-0.06% to +0.02%)
fibonacci-asm-2000000	arm64	Neoverse-V2	11.45 µs	11.45 µs	+0.00% (-0.06% to +0.03%)
fibonacci-asm-2000000	x86-64	INTEL(R) XEON(R) GOLD 6548N	12.30 µs	12.30 µs	+0.00% (-0.01% to +0.00%)
hash-50000	arm64	Neoverse-V2	16.81 µs	16.81 µs	+0.00% (-0.27% to +0.00%)
hash-50000	x86-64	INTEL(R) XEON(R) GOLD 6548N	18.66 µs	18.67 µs	+0.05% (+0.03% to +0.16%)
leansphincs-26	arm64	Neoverse-V2	25.51 µs	25.51 µs	+0.00% (-0.14% to +0.13%)
leansphincs-26	x86-64	INTEL(R) XEON(R) GOLD 6548N	27.05 µs	27.05 µs	+0.03% (-0.05% to +0.05%)
leanxmss-100	arm64	Neoverse-V2	26.14 µs	26.14 µs	+0.00% (-0.09% to +0.00%)
leanxmss-100	x86-64	INTEL(R) XEON(R) GOLD 6548N	27.64 µs	27.64 µs	-0.00% (-0.01% to +0.00%)
leanxmss-100-1thread	arm64	Neoverse-V2	24.97 µs	24.97 µs	+0.00% (+0.00% to +0.00%)
leanxmss-100-1thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	25.42 µs	25.42 µs	+0.00% (+0.00% to +0.00%)
leanxmss-100-4thread	arm64	Neoverse-V2	25.25 µs	25.25 µs	+0.00% (+0.00% to +0.00%)
leanxmss-100-4thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	25.70 µs	25.70 µs	-0.00% (-0.00% to +0.00%)
leanxmss-100-8thread	arm64	Neoverse-V2	25.55 µs	25.55 µs	+0.00% (+0.00% to +0.00%)
leanxmss-100-8thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	26.00 µs	26.00 µs	+0.00% (-0.00% to +0.00%)
shielded-258	arm64	Neoverse-V2	45.20 µs	45.20 µs	+0.00% (+0.00% to +0.01%)
shielded-258	x86-64	INTEL(R) XEON(R) GOLD 6548N	47.23 µs	47.21 µs	-0.03% (-0.29% to +0.01%)
stateproof-5	arm64	Neoverse-V2	19.27 µs	19.28 µs	+0.01% (-0.12% to +0.05%)
stateproof-5	x86-64	INTEL(R) XEON(R) GOLD 6548N	21.20 µs	21.20 µs	-0.03% (-0.07% to +0.10%)

heap-peak

benchmark	runner	CPU	base	this PR	change
aggregate-leanxmss-100-2to1-1thread-first	arm64	Neoverse-V2	738.67 ms	738.67 ms	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-1thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	373.29 ms	373.29 ms	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-1thread-node	arm64	Neoverse-V2	738.32 ms	738.32 ms	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-1thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	372.94 ms	372.94 ms	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-2to1-4thread-first	arm64	Neoverse-V2	738.64 ms	738.65 ms	+0.01% (-0.02% to +0.01%)
aggregate-leanxmss-100-2to1-4thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	373.33 ms	373.30 ms	+0.00% (-0.02% to +0.02%)
aggregate-leanxmss-100-2to1-4thread-node	arm64	Neoverse-V2	738.30 ms	738.31 ms	+0.00% (+0.00% to +0.01%)
aggregate-leanxmss-100-2to1-4thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	373.03 ms	373.03 ms	-0.00% (-0.02% to +0.00%)
aggregate-leanxmss-100-2to1-8thread-first	arm64	Neoverse-V2	738.53 ms	738.51 ms	+0.00% (-0.00% to +0.00%)
aggregate-leanxmss-100-2to1-8thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	373.35 ms	373.35 ms	-0.00% (-0.01% to +0.02%)
aggregate-leanxmss-100-2to1-8thread-node	arm64	Neoverse-V2	738.25 ms	738.25 ms	-0.00% (-0.01% to +0.02%)
aggregate-leanxmss-100-2to1-8thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	373.06 ms	373.02 ms	-0.01% (-0.02% to +0.00%)
aggregate-leanxmss-100-2to1-first	arm64	Neoverse-V2	738.82 ms	738.82 ms	+0.00% (-0.01% to +0.00%)
aggregate-leanxmss-100-2to1-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	373.94 ms	373.91 ms	-0.01% (-0.03% to +0.02%)
aggregate-leanxmss-100-2to1-node	arm64	Neoverse-V2	738.02 ms	738.10 ms	+0.01% (-0.07% to +0.06%)
aggregate-leanxmss-100-2to1-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	373.60 ms	373.63 ms	+0.00% (-0.01% to +0.04%)
aggregate-leanxmss-100-4to1-1thread-first	arm64	Neoverse-V2	1.48 s	1.48 s	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-1thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	746.21 ms	746.21 ms	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-1thread-node	arm64	Neoverse-V2	1.48 s	1.48 s	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-1thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	745.63 ms	745.63 ms	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-4thread-first	arm64	Neoverse-V2	1.48 s	1.48 s	+0.00% (+0.00% to +0.00%)
aggregate-leanxmss-100-4to1-4thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	746.31 ms	746.30 ms	-0.00% (-0.01% to +0.01%)
aggregate-leanxmss-100-4to1-4thread-node	arm64	Neoverse-V2	1.48 s	1.48 s	-0.00% (-0.00% to +0.00%)
aggregate-leanxmss-100-4to1-4thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	745.71 ms	745.72 ms	+0.00% (-0.01% to +0.01%)
aggregate-leanxmss-100-4to1-8thread-first	arm64	Neoverse-V2	1.48 s	1.48 s	-0.00% (-0.00% to +0.01%)
aggregate-leanxmss-100-4to1-8thread-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	746.39 ms	746.40 ms	+0.00% (-0.01% to +0.00%)
aggregate-leanxmss-100-4to1-8thread-node	arm64	Neoverse-V2	1.48 s	1.48 s	-0.00% (-0.00% to +0.01%)
aggregate-leanxmss-100-4to1-8thread-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	745.81 ms	745.81 ms	+0.00% (-0.01% to +0.01%)
aggregate-leanxmss-100-4to1-first	arm64	Neoverse-V2	1.48 s	1.48 s	+0.01% (-0.01% to +0.01%)
aggregate-leanxmss-100-4to1-first	x86-64	INTEL(R) XEON(R) GOLD 6548N	747.08 ms	747.01 ms	-0.01% (-0.03% to -0.00%)
aggregate-leanxmss-100-4to1-node	arm64	Neoverse-V2	1.48 s	1.48 s	+0.00% (-0.00% to +0.01%)
aggregate-leanxmss-100-4to1-node	x86-64	INTEL(R) XEON(R) GOLD 6548N	746.46 ms	746.46 ms	-0.01% (-0.02% to +0.02%)
falcon-7	arm64	Neoverse-V2	5.43 s	5.43 s	-0.00% (-0.00% to +0.00%)
falcon-7	x86-64	INTEL(R) XEON(R) GOLD 6548N	2.42 s	2.42 s	+0.00% (-0.00% to +0.00%)
fibonacci-asm-2000000	arm64	Neoverse-V2	8.70 s	8.70 s	-0.00% (-0.00% to +0.00%)
fibonacci-asm-2000000	x86-64	INTEL(R) XEON(R) GOLD 6548N	3.95 s	3.95 s	+0.00% (-0.00% to +0.01%)
hash-50000	arm64	Neoverse-V2	5.45 s	5.45 s	-0.00% (-0.00% to +0.00%)
hash-50000	x86-64	INTEL(R) XEON(R) GOLD 6548N	2.30 s	2.30 s	-0.00% (-0.00% to +0.00%)
leansphincs-26	arm64	Neoverse-V2	1.72 s	1.72 s	-0.00% (-0.01% to +0.00%)
leansphincs-26	x86-64	INTEL(R) XEON(R) GOLD 6548N	776.33 ms	776.34 ms	+0.00% (-0.00% to +0.01%)
leanxmss-100	arm64	Neoverse-V2	1.49 s	1.49 s	-0.00% (-0.00% to +0.00%)
leanxmss-100	x86-64	INTEL(R) XEON(R) GOLD 6548N	666.94 ms	666.94 ms	-0.00% (-0.01% to +0.00%)
leanxmss-100-1thread	arm64	Neoverse-V2	1.49 s	1.49 s	+0.00% (+0.00% to +0.00%)
leanxmss-100-1thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	667.32 ms	667.32 ms	+0.00% (+0.00% to +0.00%)
leanxmss-100-4thread	arm64	Neoverse-V2	1.49 s	1.49 s	-0.00% (-0.00% to +0.01%)
leanxmss-100-4thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	667.28 ms	667.29 ms	+0.00% (-0.01% to +0.01%)
leanxmss-100-8thread	arm64	Neoverse-V2	1.49 s	1.49 s	+0.00% (-0.00% to +0.01%)
leanxmss-100-8thread	x86-64	INTEL(R) XEON(R) GOLD 6548N	667.31 ms	667.32 ms	+0.00% (-0.00% to +0.01%)
shielded-258	arm64	Neoverse-V2	7.89 s	7.89 s	-0.00% (-0.00% to +0.00%)
shielded-258	x86-64	INTEL(R) XEON(R) GOLD 6548N	3.66 s	3.66 s	+0.00% (-0.01% to +0.00%)
stateproof-5	arm64	Neoverse-V2	12.44 s	12.44 s	+0.00% (-0.00% to +0.00%)
stateproof-5	x86-64	INTEL(R) XEON(R) GOLD 6548N	5.82 s	5.82 s	+0.00% (-0.00% to +0.00%)