Skip to content

fix typos in math.rs #1816

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.

Sign up for GitHub

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

Open
leopardracer wants to merge 2 commits into
base: main
Choose a base branch
from
Open

fix typos in math.rs #1816

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
297c804
Update math.rs
leopardracer Jul 3, 2025
c199d2e
Merge branch 'main' into main
seanyoung Jul 21, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions src/emit/math.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -201,7 +201,7 @@ fn signed_ovf_detect<'a, T: TargetRuntime<'a> + ?Sized>(
.build_or(left_is_zero, right_is_zero, "")
.unwrap();

// Will resolve to one if signs are differnet
// Will resolve to one if signs are different
let different_signs = bin
.builder
.build_xor(left_sign_bit, right_sign_bit, "")
Expand Down Expand Up @@ -327,7 +327,7 @@ pub(super) fn multiply<'a, T: TargetRuntime<'a> + ?Sized>(
}
// LLVM-IR can handle multiplication of sizes up to 64 bits. If the size is larger, we need to implement our own multiplication function.
// We divide the operands into sizes of 32 bits (check __mul32 in stdlib/bigint.c documentation).
// If the size is not divisble by 32, we extend it to the next 32 bits. For example, int72 will be extended to int96.
// If the size is not divisible by 32, we extend it to the next 32 bits. For example, int72 will be extended to int96.
// Here, we zext the operands to the nearest 32 bits. zext is called instead of sext because we need to do unsigned multiplication by default.
// It will not matter in terms of mul without overflow, because we always truncate the result to the bit size of the operands.
// In mul with overflow however, it is needed so that overflow can be detected if the most significant bits of the result are not zeros.
Expand All @@ -352,7 +352,7 @@ pub(super) fn multiply<'a, T: TargetRuntime<'a> + ?Sized>(

// Unsigned overflow detection Approach:
// If the size is a multiple of 32, we call __mul32_with_builtin_ovf and it returns an overflow flag (check __mul32_with_builtin_ovf in stdlib/bigint.c documentation)
// If that is not the case, some extra work has to be done. We have to check the extended bits for any set bits. If there is any, an overflow occured.
// If that is not the case, some extra work has to be done. We have to check the extended bits for any set bits. If there is any, an overflow occurred.
// For example, if we have uint72, it will be extended to uint96. __mul32 with ovf will raise an ovf flag if the result overflows 96 bits, not 72.
// We account for that by checking the extended leftmost bits. In the example mentioned, they will be 96-72=24 bits.
let return_val = bin
Expand Down Expand Up @@ -405,7 +405,7 @@ pub(super) fn multiply<'a, T: TargetRuntime<'a> + ?Sized>(
// Until this point, we only checked the extended bits for ovf. But mul ovf can take place any where from bit size to double bit size.
// For example: If we have uint72, it will be extended to uint96. We only checked the most significant 24 bits for overflow, which can happen up to 72*2=144 bits.
// bool __mul32_with_builtin_ovf takes care of overflowing bits beyond 96.
// What is left now is to or these two ovf flags, and check if any one of them is set. If so, an overflow occured.
// What is left now is to or these two ovf flags, and check if any one of them is set. If so, an overflow occurred.
let lowbit = bin
.builder
.build_int_truncate(
Expand Down
Loading