Instruction coersion in QRE
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 1k
- Forks
- 212
- Avg merge
- 3d 8h
- Merged PRs (30d)
- 65
Description
Instructions in QRE are a primitive, and carry a natural the structure of a type system (all arrow types). As such there is a natural notion of subtypes and supertypes. An already implemented example is H_XZ (ID 0x10) is a subtype of ONE_QUBIT_CLIFFORD (ID 0x50), which in turn is a subtype of GENERIC (ID 0xFFFF). While generally converting one instruction to another requires synthesis (and hence belongs to ISATransform), doing this for simple coersions such as the above is very heavyweight.
A natural approach to coersion would be to add a static database (presumably along with the predefined instruction IDs) that implements valid coersions. The database would need to provide functionality for user added IDs. Querying for valid coersion could be added to the Instruction class. Generators for iterating over superclasses could also be added to the Instruction class, best as a wrapper around such an iterator directly implemented in the database.
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.
Research direction
Start by locating the Instruction implementation, the predefined instruction IDs, and the existing ISATransform path. Review how H_XZ, ONE_QUBIT_CLIFFORD, and GENERIC are represented, then determine how a coercion database could support both predefined and user-added IDs. Done means valid superclass queries and coercion checks are available through Instruction without requiring synthesis.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- compilers, quantum-computing
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 30/100