user0/rust
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Rollup merge of #71269 - Mark-Simulacrum:sat-float-casts, r=nikic

Browse source 
Define UB in float-to-int casts to saturate

This closes #10184 by defining the behavior there to saturate infinities and values exceeding the integral range (on the lower or upper end). `NaN` is sent to zero.
This commit is contained in:
Dylan DPC 2020-05-06 16:58:50 +02:00 • committed by GitHub
commit 14d608f1d8
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
6 changed files with 420 additions and 52 deletions
Download patch file Download diff file Expand all files Collapse all files

12
src/test/codegen/unchecked-float-casts.rs
View file

@ -1,7 +1,7 @@
// compile-flags: -C no-prepopulate-passes
// This file tests that we don't generate any code for saturation when using the
// unchecked intrinsics.
// This file tests that we don't generate any code for saturation if
// -Z saturating-float-casts is not enabled.
// compile-flags: -C opt-level=3
#![crate_type = "lib"]
@ -12,7 +12,7 @@ pub fn f32_to_u32(x: f32) -> u32 {
// CHECK-NOT: fcmp
// CHECK-NOT: icmp
// CHECK-NOT: select
x as u32
unsafe { x.to_int_unchecked() }
}
// CHECK-LABEL: @f32_to_i32
@ -22,7 +22,7 @@ pub fn f32_to_i32(x: f32) -> i32 {
// CHECK-NOT: fcmp
// CHECK-NOT: icmp
// CHECK-NOT: select
x as i32
unsafe { x.to_int_unchecked() }
}
#[no_mangle]
@ -31,5 +31,5 @@ pub fn f64_to_u16(x: f64) -> u16 {
// CHECK-NOT: fcmp
// CHECK-NOT: icmp
// CHECK-NOT: select
x as u16
unsafe { x.to_int_unchecked() }
}