-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
1 changed file
with
18 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,18 @@ | ||
# Assembly | ||
|
||
`u1` needs to have an assembly language so that it is possible to test optimizations in instruction size. | ||
|
||
The assembly format should take highly generalized instructions. For instance, we do not want to include an `increment` instruction initially, even though it is likely we might need it. This is so that we make no assumptions. The backend can decide to lower `1 add` to become `increment` at the ISA-level. These sorts of small changes to the instructions should happen in the backend and not the assembly until they are finalized. | ||
|
||
## Emulator | ||
|
||
To test the correctness of written assembly using unit tests, we need to create an emulator. It does not need to run the actual ISA, just the assembly files, since it is only being used to test the consistency of the assembly based on the architecture spec. | ||
|
||
## Benchmark assembly | ||
|
||
To test various backends, several benchmarks will be created. Each of these benchmarks should tell us something about the performance of an actual real-world program. | ||
|
||
- Basic N-body (from benchmark game) | ||
- Fannkuch-redux (from benchmark game) | ||
- k-nucleotide (from benchmark game) | ||
- CSV parsing |