EXP-067NEGATIVE
Canonical GPU engineering / SHA-256d

Message-schedule and consumer-round fusion

Moving W18…W61 assignments directly before their first consuming rounds reduces live ranges, registers or instructions.

Claim statusNO END-TO-END MINING ADVANTAGE DEMONSTRATED

Diagnostic work, exact algebra and local capabilities are not treated as end-to-end mining advantage.

Date2026-08-14
HardwareNVIDIA GeForce RTX 5080 · exact SHA-256d verifier
Run scopePreregistered campaign with sealed result and audit
ReproducibilityGenerate the same moves deterministically, checksum written bytes, compare disassembly and rerun balanced timing.
01 / Question & hypothesis

What was tested?

Moving W18…W61 assignments directly before their first consuming rounds reduces live ranges, registers or instructions.

02 / Scientific basis

Why the test is meaningful

Manual scheduling is useful only if the compiler does not already perform equivalent dependency and lifetime management.

W_t generated immediately before round thash order and equations unchangedbenefit requires lower SASS/registers and faster timing
Message-schedule and consumer-round fusionVisual reading of the published metrics and gates for EXP-067; it summarizes the registered result, not a mining advantage.EXP-067 / MESSAGE-SCHEDULE AND CONSUMER-ROUND FUSIONKERNEL RATIO0.995602SASS DIFFERENCE0REGISTER DIFFERENCE0
FIGURE / RESULT READINGVisual reading of the published metrics and gates for EXP-067; it summarizes the registered result, not a mining advantage.
03 / Method

How it was tested

Use a deterministic source generator for 44 moves and evaluate four fresh headers in 32 balanced runs.

04 / Observed result

What happened

Kernel ratio0.995602
SASS difference0
Register difference0

Baseline and candidate emitted identical SASS: 2,507 instructions, 1,176 SHF, 629 LOP3, 408 IADD3 and 44 registers. Candidate lost 4/4 with kernel ratio 0.995602.

05 / Validation

Exactness and statistical controls

A checksum line-ending issue invalidated the first attempt; checksums were then taken from written bytes and the whole frozen campaign repeated.

06 / Interpretation

What the result means

ptxas already manages schedule lifetimes; manual source fusion is closed.

Limitations

  • The result concerns this compiler and source body.
  • Other schedule algorithms are separate mechanisms.
  • Identical SASS makes the source rewrite non-causal.
07 / Reproduction

Evidence trail

Generate the same moves deterministically, checksum written bytes, compare disassembly and rerun balanced timing.

Canonical variants

CANONICAL-EXP-143PREREGISTERED-CAMPAIGNSEALED-AUDIT

Source: internally audited canonical reports. Local filesystem structure, private headers and operational identifiers are excluded from publication.