-
Notifications
You must be signed in to change notification settings - Fork 451
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.
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
Implement unsigned operators #27
Conversation
For consistency and clear definitions, why not make all int ops work using big_int (of size 32 bits/64 bits, respectively)? I feel like that might be a good move for the proto. |
Looks generally good to me. This made me realize that, without a single-precision ieee754 type, we'll have a hard time representing the precise semantics of |
Forgot to git add the test file earlier. @kg, hm, not sure if that would be clearer. You would enlarge the int32/64 value domain with illegal values, and you'd have to implement the wrap-around masking manually. Currently, the hacks are at least localised in a few operators. |
Simplify build-test
* Remove float atomics * Use 0xfe as prefix byte * Add sign-extending conversion operators * Remove sign-extending atomic-operators (encoding proposal 3)
- replaced exceptions with tags - removed let blocks - removed now unnecessary (unreachable)s The output of all examples is as before.
Raise JS memory limit to 100
…bAssembly#27) This patch brings the description of the proposed instruction set extension in sync with the reference implementation. It also fixes a few other bugs such as a dead link in the TOC and the syntax in the examples. The document still warrants a refactoring as some things could do with better explanations, some TODOs left to complete, and some of the examples still make use of deprecated features from the function-references proposal.
@lukewagner, @kg, it's rather ugly, going through big_int, but seems to work.