@@ -8,6 +8,37 @@ and this project adheres to
8
8
9
9
## [ Unreleased]
10
10
11
+ ## [ 0.2.16] ( https://github.com/rust-lang/compiler-builtins/compare/libm-v0.2.15...libm-v0.2.16 ) - 2025-09-22
12
+
13
+ ### Other
14
+
15
+ - Remove usage of the to-be-deprecated ` core::f32 ` , ` core::f64 ` items
16
+ - define and implement ` trait NarrowingDiv ` for unsigned integer division
17
+ - Fix some typos
18
+ - Use ` CARGO_CFG_*_{F16,F128} ` rather than invoking rustc
19
+ - Simplify the configuration for no-panic
20
+ - Implement ` floor ` and ` ceil ` in assembly on ` i586 `
21
+ - Avoid inlining ` floor ` into ` rem_pio2 `
22
+ - Update for new warn-by-default clippy lints
23
+ - Use ` x86_no_sse ` configuration in more places
24
+ - Enable tests that were skipped on PowerPC
25
+ - Allow a new lint failure in nightly
26
+ - Test building custom targets and resolve an issue probing ` rustc `
27
+ - Use the compiler to determine whether or not to enable ` f16 ` and ` f128 `
28
+ - Improved integer utilities, implement shifts and bug fixes for i256 and u256
29
+ - Clean up and sort manifest keys
30
+ - Fix incorrect result and add tests
31
+ - Fix unintentional skips in ` binop_common `
32
+ - Use ` x.biteq(y) ` rather than ` x.to_bits() == y.to_bits() `
33
+ - Fix new ` dead_code ` warnings from recent nightlies
34
+ - Reuse ` libm ` 's ` Caat ` and ` CastFrom ` in ` compiler-builtins `
35
+ - Reuse ` MinInt ` and ` Int ` from ` libm ` in ` compiler-builtins `
36
+ - Add a note saying why we use ` frintx ` rather than ` frintn `
37
+ - Typo in README.md
38
+ - Update ` CmpResult ` to use a pointer-sized return type
39
+ - Clean up unused files
40
+ - Fix ` i256::MAX `
41
+
11
42
## [ 0.2.15] ( https://github.com/rust-lang/compiler-builtins/compare/libm-v0.2.14...libm-v0.2.15 ) - 2025-05-06
12
43
13
44
### Other
0 commit comments