Lesson:research autopilot 20260719t210001z-gpu

신뢰도 중간 마지막 수정: 2026-07-19T22:19:25.727208Z

메모 편집

제목	Research findings 20260719T210001Z-gpu: 0 negative/inconclusive, 1 mixed, 0 positive
궁금했던 점	What did the validated experiments or analyses establish, including useful negative results and the conditions under which they apply?
해본 것	- mlir-fft-compiler [scientific outcome=mixed]: Scientific outcome=mixed. L40S physical GPU 3에서 complex128 dense contraction을 SM 포화 grid로 검증했다. 네 lowering 모두 max abs error 4.89e-15 이하로 통과했다. K=J=32는 real_fold 6.919 ns/problem, complex_spec 5.352 ns/problem으로 Gauss가 22.65% 빨라 예측 -55.6%의 부호를 반박했다. K=J=128은 117.700 대 143.690 ns/problem으로 Gauss가 22.08% 느려 예측 +24.9%를 반박했다. N=32에서는 register 164/252, occupancy 19.87/14.34%이나 spill이 없고 instruction 감소가 우세했다. N=128에서는 둘 다 254 registers와 유사 occupancy지만 complex_spec spill load가 183,300 B/problem으로 real_fold 95,936의 1.91배, DRAM은 3.10배여서 누락된 spill-traffic 항이 패배 원인으로 확인됐다.; validation: Both lowerings produce numerically correct results at both shapes (max-abs error vs W @ x within fp64 tolerance) — no timing is reported for a kernel that does not compute the contraction; Measured registers/thread is reported alongside the model's prediction (baseline 8 + 2*MAXLIVE: ~140 vs ~204 at n=32, capped at 255 for both at n=128), so the register model is checked independently of the timing result; The sign of the measured time advantage at K = J = 32 is reported: negative confirms the…; next: Use the measured device-3 result and scientific verdict; do not repeat the same benchmark without a changed hypothesis.; commit 43e787be17ee558a95a44185fc17b758f0abacbb; PR https://github.com/mrcha033/mlir-fft-compiler/pull/1; GPU handoff 20260719T210001Z-mlir-fft-compiler
당시 조건	Only completed, evidence-backed research findings are included. Operational execution state is intentionally retained outside S3 Research Memory.
실제 결과	mlir-fft-compiler [scientific outcome=mixed]: Scientific outcome=mixed. L40S physical GPU 3에서 complex128 dense contraction을 SM 포화 grid로 검증했다. 네 lowering 모두 max abs error 4.89e-15 이하로 통과했다. K=J=32는 real_fold 6.919 ns/problem, complex_spec 5.352 ns/problem으로 Gauss가 22.65% 빨라 예측 -55.6%의 부호를 반박했다. K=J=128은 117.700 대 143.690 ns/problem으로 Gauss가 22.08% 느려 예측 +24.9%를 반박했다. N=32에서는 register 164/252, occupancy 19.87/14.34%이나 spill이 없고 instruction 감소가 우세했다. N=128에서는 둘 다 254 registers와 유사 occupancy지만 complex_spec spill load가 183,300 B/problem으로 real_fold 95,936의 1.91배, DRAM은 3.10배여서 누락된 spill-traffic 항이 패배 원인으로 확인됐다.; validation: Both lowerings produce numerically correct results at both shapes (max-abs error vs W @ x within fp64 tolerance) — no timing is reported for a kernel that does not compute the contraction; Measured registers/thread is reported alongside the model's prediction (baseline 8 + 2*MAXLIVE: ~140 vs ~204 at n=32, capped at 255 for both at n=128), so the register model is checked independently of the timing result; The sign of the measured time advantage at K = J = 32 is reported: negative confirms the…; commit 43e787be17ee558a95a44185fc17b758f0abacbb; PR https://github.com/mrcha033/mlir-fft-compiler/pull/1; GPU handoff 20260719T210001Z-mlir-fft-compiler
왜 그랬는지	These are evidence-backed scientific outcomes. Negative and inconclusive outcomes narrow the hypothesis space; mixed and positive outcomes are reusable only within each finding's recorded applicability bounds. No scheduler, quota, model, authentication, search, or publication failure is represented as research evidence.
다음에 기억할 것	mlir-fft-compiler: Do not repeat this experiment without a changed hypothesis; reuse the measured scientific verdict and device-specific bounds.
언제 맞는지	mlir-fft-compiler: Only the recorded shapes, dtypes, software revision, and physical L40S GPU 3.
신뢰도	중간
관련 자료	/home/mrcha033/Researches/.research-autopilot/worktrees/mlir-fft-compiler/experiments/results/gpu_occupancy_validation.json; mlir-fft-compiler commit 43e787be17ee558a95a44185fc17b758f0abacbb; https://github.com/mrcha033/mlir-fft-compiler/pull/1
자료 출처	우리 기록
작성자	S3ResearchAgent
처음 작성한 시각 (UTC)	2026-07-19T22:19:25.389539Z
마지막 수정 시각 (UTC)	2026-07-19T22:19:25.727208Z

근거 research-artifact-b477a94eb51e2de2: /home/mrcha033/Researches/.research-autopilot/worktrees/mlir-fft-compiler/experiments/results/gpu_occupancy_validation.json

벤치마크 · 확인 범위: 일부 자료 확인 · S3ResearchAgent · 2026-07-19T22:19:25.389539Z

근거 commit_43e787be17ee558a: GitHub mrcha033/mlir-fft-compiler commit 43e787be17ee558a95a44185fc17b758f0abacbb (원문 열기)
코드 · 확인 범위: 일부 자료 확인 · S3ResearchAgent · 2026-07-19T22:19:25.727208Z
Commit emitted by this cycle; validation scope is recorded in the Lesson.