From 0d6012c673e7d5c7a20c3cc9e653ec08a246766d Mon Sep 17 00:00:00 2001 From: titzer Date: Fri, 23 Oct 2015 14:35:41 -0700 Subject: Replace `operation` with `operator` --- FeatureTest.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'FeatureTest.md') diff --git a/FeatureTest.md b/FeatureTest.md index 0010b1e..22bd780 100644 --- a/FeatureTest.md +++ b/FeatureTest.md @@ -33,7 +33,7 @@ it can be constant-folded by WebAssembly engines. To illustrate, consider 4 examples: -* [`i32.min_s`](FutureFeatures.md#additional-integer-operations) - Strategy 2 +* [`i32.min_s`](FutureFeatures.md#additional-integer-operators) - Strategy 2 could be used to translate `(i32.min_s lhs rhs)` into an equivalent expression that stores `lhs` and `rhs` in locals then uses `i32.lt_s` and `conditional`. * [Threads](PostMVP.md#threads) - If an application uses `#ifdef` extensively @@ -51,7 +51,7 @@ To illustrate, consider 4 examples: the application to test `(has_feature "mprotect")` to avoid calling the `abort()`. The `has_feature` query could be exposed to C++ code via the existing `__builtin_cpu_supports`. -* [SIMD](PostMVP.md#fixed-width-simd) - When SIMD operations have a good-enough +* [SIMD](PostMVP.md#fixed-width-simd) - When SIMD operators have a good-enough polyfill, e.g., `f32x4.fma` via `f32x4.mul`/`add`, Strategy 2 could be used (similar to the `i32.min_s` example above). However, when a SIMD feature has no efficient polyfill (e.g., `f64x2`, which introduces both operators *and* -- cgit v1.2.3