aboutsummaryrefslogtreecommitdiffstats
path: root/benchmarks/sample.c
diff options
context:
space:
mode:
authoreunhhu <[email protected]>2026-10-08 17:48:03 +0100
committereunhhu <[email protected]>2026-10-08 17:48:03 +0100
commit59510b75aaf6781fe7dfa10f617c06d5272bd746 (patch)
tree0633e7104d5326b0e1e9649595edb60c45a31270 /benchmarks/sample.c
parentc38f9f11145445b4d4e3d284fd1256cc891bb3db (diff)
downloadpure-59510b75aaf6781fe7dfa10f617c06d5272bd746.tar.gz
pure-59510b75aaf6781fe7dfa10f617c06d5272bd746.zip
Research 137: the batch executor's stage 4 (columns read in place, comparisons counting their lanes): knapsack 10.07 to 8.25 ms (bar 8.6; b9 11.46, the compiled code 7.9), its instructions 12.2% below b9's; what is left of knapsack's gap is the per-batch interpretation (about 3 ms), not the kernels (ADD 1.7, GT 1.7, the masked set 0.55 ms); the others the same by instructions; make check's server_view_workers timed out once (the known flake), passed 3/3 alone and in a full rerun; bank queries' instructions within 0.1% of b9's, TPC-C -0.3%; the two large gaps left (matmul's nested rounds, BFS's candidate tail) the user's to choose
Diffstat (limited to 'benchmarks/sample.c')
0 files changed, 0 insertions, 0 deletions