Choose number of replications based on data
Open
@mdboom is already working on this.
Since Apr 9, 2024.
3.13
interpreter-core
performance
- Dominant language
- Python
- Stars
- 77.2k
- Forks
- 35.9k
- PR merge metrics
- PR metrics pending
Description
Currently we replicate the LOAD_FAST and STORE_FAST instructions 8 times (for opargs 0-7). The _INIT_CALL_PY_EXACT_ARGS micro-op gets replicated 5 times.
These numbers were chosen arbitrarily as initial values to test replication.
We should choose the number of replications based on actual data.
We should also look into replicating COPY and SWAP, as it is much simpler to do stack caching on instructions with a fixed stack effect.
E.g. SWAP indexes the stack using oparg, but SWAP_2 is simply SWAP_2(left, right -- right, left).
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.