[WIP] - #129
Closed
bbyalcinkaya wants to merge 1 commit into
Closed
[WIP]#129bbyalcinkaya wants to merge 1 commit into
bbyalcinkaya wants to merge 1 commit into
Conversation
…t object tags Three defects in the I256 support added by #127, each checked against the Soroban host rather than against the surrounding rules. - The i256 arithmetic host functions report `(Object, ArithDomain)`, not `(Value, ArithDomain)`. The 6-argument arm of `impl_bignum_host_fns!` builds its error with `ScErrorType::Object` (soroban-env-host/src/host/num.rs:39-45), and `i256_add`, `_sub`, `_mul`, `_div` and `_rem_euclid` all expand through that arm (soroban-env-host/src/host.rs:1568-1578). - `i256_rem_euclid` accepted `i256::MIN` by `-1` and returned `0`. `I256` is `ethnum::I256` (soroban-env-common/src/num.rs:10), whose `checked_rem_euclid` rejects `rhs == 0 || (self == MIN && rhs == -1)` (ethnum-1.5.2/src/int/api.rs:635-641) -- the same two pairs as `checked_div`, even though the mathematical remainder is representable. This was a silent wrong answer, not a stuck term. The comment above the rule was also wrong about `modInt`: K's `modInt` is e-division and always lands in `[0, absInt(B))` (domains.md:1262), so `absInt(B)` is redundant rather than load-bearing. Left the expression alone, corrected the claim. - `getTagWithFlag(true, _)` was missing both 256-bit types. Under `alwaysAllocate`, `addObject` tags the handle with `getTagWithFlag(AA, SCV)`; with no entry, a value that fits the small encoding fell through to `owise` and took `getTag`, so an object stored in `<hostObjects>` was tagged `U256Small` (12) / `I256Small` (13). `isObject` is tag 64..77, so such a handle then failed `isObject` and `loadObject` took the `-small` branch, making `fromSmall` decode the object index as the value. U64, I64, U128, I128 and Symbol all carry this line for that reason; U256 and I256 were the only two small-capable types missing it. Tag numbers confirmed against `Tag` in soroban-env-common/src/val.rs (`U256Small = 12`, `U256Object = 70`, `I256Small = 13`, `I256Object = 71`). i256.wast pins the first two: the six existing error assertions move to `ErrObject`, and `i256::MIN rem_euclid -1` is added. The tag rules are only reachable with `alwaysAllocate` set, which the .wast harness does not do; they were checked by building with the cell forced to `true`, under which u256.wast, double_u256.wast and i256.wast all fail before the change and pass after.
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
No description provided.