feat(interface-types) New numbers and strings instructions #1329
Conversation
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
Basically the `x-to-y` instructions have been renamed `y.from_x`. This patch updates the instruction. The binary representation isn't specified yet, so it's just arbitrary values.
Hywan
added
🎉 enhancement
…ructions. The `string.lift_memory` instruction replaces `memory-to-string`, and `string.lower_memory` replaces `string-to-memory`.
This method allows to peek the last item on the stack (if any) by reference.
Hywan
changed the title
10 tasks
MarkMcCaskey
approved these changes
Mar 24, 2020
| ) | ||
| })?; | ||
|
|
||
| let pointer = to_native::<i32>(&inputs[0], instruction)? as usize; |
There was a problem hiding this comment.
Does the fact this is an i32 cause a problem? It seems that we can't access the upper 2gb of memory by casting from i32 to usize because -1i32 as usize is 18446744073709551615; -1i32 as u32 as usize is 4294967295 which seems closer to what we'd want here.
If this is a mistake, perhaps we should add some tests using the upper 2gb of memory to make sure everything is working as expected as this issue will likely be in a lot of places!
There was a problem hiding this comment.
Hmm. The pointer is of kind i32 in the spec. It is regular a Wasm core type i32. Our API needs a usize, hence the casting.
You're right we need to check it is positive. I'm opening an issue to focus on that.
|
bors r+ |
bors bot
added a commit
that referenced
this pull request
Mar 26, 2020
1329: feat(interface-types) New numbers and strings instructions r=Hywan a=Hywan This PR does 3 things: * Transforms the `x-to-y` number lowerings and liftings instructions to the new set of instructions `y.from_x`. It's just the opposite. It's more logical, but the diff is dense due to that. The `lowering_lifting` module is renamed `numbers`. * Renames `memory-to-string` to `string.lift_memory`, and `string-to-memory` to `string.lower_memory`. It's also more logical, but the diff is dense due to that too. The `memory_to_string` and `string_to_module` modules are merged into the `strings` module. * Introduce the new `string.size` instruction (in the `strings` module). I think commits are clean and atomic enough, so it might be better to review it like that. For binary encoding/decoding, the binary representation is absolutely entirely arbitrary (since the working notes define nothing, obviously). Cf https://github.com/WebAssembly/interface-types/blob/master/proposals/interface-types/working-notes/Instructions.md Co-authored-by: Ivan Enderlin <ivan.enderlin@hoa-project.net> Co-authored-by: Ivan Enderlin <ivan.enderlin@wanadoo.fr>
Timed out |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR does 3 things:
x-to-ynumber lowerings and liftings instructions to the new set of instructionsy.from_x. It's just the opposite. It's more logical, but the diff is dense due to that. Thelowering_liftingmodule is renamednumbers.memory-to-stringtostring.lift_memory, andstring-to-memorytostring.lower_memory. It's also more logical, but the diff is dense due to that too. Thememory_to_stringandstring_to_modulemodules are merged into thestringsmodule.string.sizeinstruction (in thestringsmodule).I think commits are clean and atomic enough, so it might be better to review it like that.
For binary encoding/decoding, the binary representation is absolutely entirely arbitrary (since the working notes define nothing, obviously).
Cf https://github.com/WebAssembly/interface-types/blob/master/proposals/interface-types/working-notes/Instructions.md