Skip to content

Track tensor4all-rs a515738, record allocations, and add a four-thread scaling sweep #6

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
shinaoka merged 4 commits into from
Oct 10, 2026
Merged

Track tensor4all-rs a515738, record allocations, and add a four-thread scaling sweep #6

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
d7e84e8
Record the timed operation's allocation counters in every arm
Oct 10, 2026
b8fcf48
Track tensor4all-rs a515738 and migrate the TreeTN patch cutoff
Oct 10, 2026
05ebd78
Add an optional multithreaded scaling sweep and fix the input cache r…
Oct 10, 2026
5db10f1
Record the a515738 sweep on the Ryzen AI 9 HX 470, one and four threads
Oct 10, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 7 additions & 1 deletion AGENTS.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,13 @@ ACI residual tolerance is not an L2 tolerance. Keep its internal metric explicit

Time only the benchmarked elementwise or contraction operation. Input construction, cache I/O, compression, conversion, patch preparation, output conversion, and accuracy evaluation stay outside timed regions.

Run release builds with one pinned CPU core and single-threaded Rayon and BLAS. Warm up any JIT-compiled external implementation before timing it.
Run release builds with one pinned CPU core and single-threaded Rayon and BLAS. That single-core sweep is the source of truth, and its profile name is the machine alone. Warm up any JIT-compiled external implementation before timing it.

A machine may carry one additional scaling profile named `<machine>-<n>t`,
produced with `BENCH_THREADS=n`, which pins a contiguous core range and lets the
same three cases run on `n` Rayon and BLAS threads. Its `run.yaml` records the
thread count and the core range. Treat it as scaling evidence only: the
single-core profile stays the reference, and the two profiles are never merged.

## Results

Expand Down
Loading
Loading