| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| ... | |
| * | Module format: bits and bytes (design 027's stage 1): I64_BITS (and, or, xor,... | eunhhu | 8 days | 4 | -5/+210 |
| * | Design 027: the computation core inside 026's contract: an app's state one ty... | eunhhu | 8 days | 1 | -0/+189 |
| * | Research 140: a suite of eleven specs from different fields (DNS, SHA-256, in... | eunhhu | 8 days | 14 | -0/+205 |
| * | Design 026: the app contract written as mathematics, the part to freeze at 1.... | eunhhu | 8 days | 1 | -0/+185 |
| * | Batch executor: the round's fewest-left marks at the end of Batch: put after ... | eunhhu | 9 days | 4 | -1/+77 |
| * | Batch executor: a nested count loop's rounds sweep their lanes and masks back... | eunhhu | 9 days | 1 | -5/+42 |
| * | chore: add Apache 2.0 license | eunhhu | 10 days | 1 | -0/+202 |
| * | Research 139: what a program made at load per batch loop would take (the deci... | eunhhu | 11 days | 1 | -2/+49 |
| * | 043 programs (research 139, in progress): a batch loop's body as the program ... | eunhhu | 11 days | 3 | -0/+457 |
| * | Research 138: the batch executor's stage 5, design 025 made: sums into cells ... | eunhhu | 11 days | 34 | -5/+489 |
| * | Makefile: loops aligned in the batch executor alone (BATCH_OPTIMIZE): the lis... | eunhhu | 11 days | 1 | -6/+8 |
| * | Makefile: the lists' and the batch executor's loops each start at a 64-byte l... | eunhhu | 11 days | 1 | -1/+4 |
| * | Batch executor, stage 5 of design 025: decide finds a kept round's bit once (... | eunhhu | 11 days | 1 | -18/+15 |
| * | Batch executor, stage 5 of design 025: the sums into cells at once from round... | eunhhu | 11 days | 2 | -2/+145 |
| * | Design 025: stage 5's bars measured again on b16: BFS's candidate tail about ... | eunhhu | 11 days | 1 | -2/+8 |
| * | Design 024: its header says stages 1-4 of 025 are made (134-137) | eunhhu | 12 days | 1 | -1/+1 |
| * | Research 137: the batch executor's stage 4 (columns read in place, comparison... | eunhhu | 12 days | 31 | -2/+333 |
| * | Batch executor, stage 4 of design 025: a four-byte comparison counts its 1s i... | eunhhu | 12 days | 1 | -14/+28 |
| * | Design 025: stage 4's knapsack bar by arithmetic (8.6 ms: two 4 KB copies and... | eunhhu | 12 days | 1 | -2/+5 |
| * | Batch executor, stage 4 of design 025: a column read at affine rows in memory... | eunhhu | 12 days | 2 | -4/+86 |
| * | Research 136: the batch executor's stage 3 (four-byte MUL, reductions with no... | eunhhu | 12 days | 31 | -2/+272 |
| * | Batch executor: a list's cells once a batch (cells_of, the first four lists i... | eunhhu | 12 days | 1 | -11/+27 |
| * | check.py: REDUCE_SOURCE, reductions that start with no lanes (a count, unifor... | eunhhu | 12 days | 1 | -0/+68 |
| * | 041 runs of b12 = d371271 (four-byte MUL) and b13 = 38f54ac (reductions with ... | eunhhu | 12 days | 2 | -0/+120 |
| * | Batch executor, stage 3 of design 025: a reduction starts with no lanes: whil... | eunhhu | 12 days | 1 | -8/+57 |
| * | Design 025: stage 3's matmul bar against what the four-byte MUL gave (26.57 m... | eunhhu | 12 days | 1 | -2/+3 |
| * | Batch executor: select and deselect named select_arm and deselect_arm (POSIX'... | eunhhu | 12 days | 1 | -6/+6 |
| * | Batch executor, stage 3 of design 025: four-byte MUL (words_op): a product be... | eunhhu | 12 days | 1 | -1/+6 |
| * | Research 135: batch plans' stage 2 measured (b11 = 7959885 against b10, inter... | eunhhu | 12 days | 5 | -2/+172 |
| * | Batch plans: stage 2's check let go (segments run as made); with it gone the ... | eunhhu | 12 days | 1 | -38/+5 |
| * | Batch plans, stage 2 of design 025: segments, runs of a batch loop's instruct... | eunhhu | 12 days | 1 | -5/+232 |
| * | Design 025: stage 2 reshaped after stage 1 (134): scalar segments planned at ... | eunhhu | 12 days | 1 | -10/+22 |
| * | Research 134: batch plans' stage 1 measured (b10 = a6fabf0 against b9, interl... | eunhhu | 12 days | 7 | -3/+227 |
| * | Batch plans: the check (PURE_BATCH_CHECK) let go, stage 1 made; with it gone ... | eunhhu | 12 days | 1 | -122/+0 |
| * | Batch plans: the recurrences read once at load too (design 025, stage 1): the... | eunhhu | 12 days | 1 | -98/+227 |
| * | Batch plans, stage 1 of design 025: each batch loop's block read once at load... | eunhhu | 12 days | 4 | -30/+150 |
| * | Research 132 and 133 corrected: the per-batch cost was icount's three rounds ... | eunhhu | 12 days | 2 | -10/+14 |
| * | Design 025: batch programs (132's A), the block's facts once at load and only... | eunhhu | 12 days | 5 | -3/+271 |
| * | 042 tiles: knapsack's two passes with the lanes numbered the batch executor's... | eunhhu | 12 days | 2 | -3/+32 |
| * | PURE_REUSE_KEPT a define: as an enum constant UINT32_MAX is past int, which c... | eunhhu | 12 days | 1 | -1/+1 |
| * | Research 133: register tiles measured (132's deciding experiment): past the b... | eunhhu | 12 days | 4 | -3/+91 |
| * | 042 tiles (research 133): a register-tile interpreter in two forms, computed ... | eunhhu | 12 days | 3 | -0/+794 |
| * | Research 132 (no code): where the batch executor's cost sits (an element's pa... | eunhhu | 12 days | 3 | -0/+194 |
| * | Research 131: the batch executor's general cuts, by the user's choice of the ... | eunhhu | 12 days | 3 | -2/+97 |
| * | 041 runs of b8 = d0558a3 and b9 = 33e9908 (the suite three ways and counts: b... | eunhhu | 12 days | 4 | -0/+618 |
| * | Batch: a selection's rows given back affine over the lanes around when they a... | eunhhu | 12 days | 3 | -6/+27 |
| * | Batch: general cuts (docs/research/131). A value computed by the same code as... | eunhhu | 12 days | 6 | -79/+636 |
| * | Research 130: the batch executor's frontiers (a count's guard, walks in a bat... | eunhhu | 12 days | 3 | -2/+116 |
| * | 041 runs of b7 = 1f8faac (the suite three ways and counts: batches BFS 41.0 m... | eunhhu | 12 days | 7 | -0/+549 |
| * | Batch: filters, if !L.has(k) { a line appending } in a batch's block or in a ... | eunhhu | 2026-10-06 | 6 | -20/+517 |
