perf/row-major trace LDE for GPU path - #715

Merged

MauroToscano merged 51 commits into

mainfrom

perf/cpu-lde-rework_gpu

Jun 29, 2026

Commits on Jun 8, 2026

add row major batched lde fft primitives
jotabulacios
committed
cc48f43
Make LDETraceTable row-major
jotabulacios
committed
46c0080
Wire prover to row-major batched LDE
jotabulacios
committed
a8db4ee
read trace row major
jotabulacios
committed
c55dc05
Move the batched-FFT and row-major-LDE unit tests into corresponding file
jotabulacios
committed
59f7c0d
resolve conflicts
jotabulacios
committed
4bd73d4
Merge branch 'main' into perf/cpu-lde-rework
jotabulacios
authored
ab786f5

Commits on Jun 9, 2026

fix disk-spill EmptyCommitment in row-major LDE
jotabulacios
committed
08eaa8b

Commits on Jun 10, 2026

Commits on Jun 16, 2026

solve conflicts
jotabulacios
committed
26a2bcd

Commits on Jun 17, 2026

Remove dead FFT module and gate legacy twiddles
jotabulacios
committed
665d82a

Commits on Jun 18, 2026

Harden parallel row-major bit-reverse permute
jotabulacios
committed
b69edd3
Guard columns_to_row_major; clarify hasher doc
jotabulacios
committed
47ded68
solve conflicts
jotabulacios
committed
14bac6b
Merge branch 'main' into perf/cpu-lde-rework
diegokingston
authored
18d1058
Merge branch 'main' into perf/cpu-lde-rework
diegokingston
authored
56fe834
Merge branch 'main' into perf/cpu-lde-rework
MauroToscano
authored
3770047

Commits on Jun 24, 2026

Deduplicate commit_rows_bit_reversed and bit_reverse_vec
jotabulacios
committed
2191f65
add gpu tests
jotabulacios
committed
94b3f29
Add GPU/CPU Merkle root parity test
jotabulacios
committed
8dc297f
Add GPU/CPU Merkle root parity tests for base and ext3 aux trace
jotabulacios
committed
b3d915d
Add GPU/CPU barycentric OOD parity tests
jotabulacios
committed
fd5da8b
Fix ext3 pre-strided layout in barycentric parity test
jotabulacios
committed
c795edc
Fix instruments double-billing GPU fused pipeline in R1
jotabulacios
committed
b3d7366

Commits on Jun 25, 2026

Commits on Jun 26, 2026

Fix gpu_lde_threshold OnceLock: re-read env var in test builds
jotabulacios
committed
ff71a65
Fix cross-stream race: synchronize after transpose before returning handle
jotabulacios
committed
09bc2dc
Add parity tests for new row-major GPU pipeline
jotabulacios
committed
53af5d9
Remove dead batched-keep GPU LDE functions
jotabulacios
committed
2613718
fix lint
jotabulacios
committed
e5d5952
fix conflicts
jotabulacios
committed
0e1c694
fix conflicts
jotabulacios
committed
8c87f74

Commits on Jun 29, 2026