sweep+plot: add primitive hand-tiled (16×16×16) variant to Case-6 decode composite study
Wires a fourth variant into the Cube-SP × PE-SP long-context decode composite command-form study to surface the worst-case PE_CPU dispatch inflation that the coarse composite forms delegate to PE_SCHEDULER (ADR-0065): per-block DMA of Q/K/V slices, tl.dot per 16³ block, deferred K-inner sum outside the K loop. Renamed _tiled.py → _hand_tiled_16x16x16.py; coarse-primitive kernel retained for the 4-cases bench, paper scripts, and the golden byte-equal regression guard. New breakdown bar chart at S_kv=128K shows engine (~460 μs) dominates all three variants; hand-tiled adds ~68 μs PE_CPU dispatch on top — composite forms sit essentially on the memory-bound floor. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
@@ -45,11 +45,12 @@ _N_RANKS = 64 # C·P for the Case-6 64-way split.
|
||||
|
||||
# variant key → (display label, colour, marker)
|
||||
_VARIANT_STYLE = {
|
||||
"primitive": ("primitive (tl.dot, hand-tiled)", "#c0504d", "o"),
|
||||
"composite": ("composite GEMM", "#3b6ea5", "s"),
|
||||
"composite_extended": ("composite + softmax_merge", "#4f8a4f", "^"),
|
||||
"primitive_tiled": ("primitive hand-tiled (16×16×16)", "#8b5a2b", "D"),
|
||||
"primitive": ("primitive (tl.dot, hand-tiled)", "#c0504d", "o"),
|
||||
"composite": ("composite GEMM", "#3b6ea5", "s"),
|
||||
"composite_extended": ("composite + softmax_merge", "#4f8a4f", "^"),
|
||||
}
|
||||
_ORDER = ("primitive", "composite", "composite_extended")
|
||||
_ORDER = ("primitive_tiled", "composite", "composite_extended")
|
||||
|
||||
|
||||
def _load() -> dict:
|
||||
|
||||
Reference in New Issue
Block a user