Skip to content

Replace _mm_movemask_pi8 with the fallback_impl #290

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.

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

Merged
merged 2 commits into from
Sep 23, 2020
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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: 1 addition & 7 deletions src/codegen/reductions/mask/x86.rs
Original file line number Diff line number Diff line change
Expand Up @@ -19,13 +19,7 @@ mod avx2;
/// x86 64-bit m8x8 implementation
macro_rules! x86_m8x8_impl {
($id:ident) => {
cfg_if! {
if #[cfg(all(target_arch = "x86_64", target_feature = "sse"))] {
x86_m8x8_sse_impl!($id);
} else {
fallback_impl!($id);
}
}
fallback_impl!($id);
};
}

Expand Down
32 changes: 0 additions & 32 deletions src/codegen/reductions/mask/x86/sse.rs
Original file line number Diff line number Diff line change
Expand Up @@ -34,35 +34,3 @@ macro_rules! x86_m32x4_sse_impl {
}
};
}

macro_rules! x86_m8x8_sse_impl {
($id:ident) => {
impl All for $id {
#[inline]
#[target_feature(enable = "sse")]
unsafe fn all(self) -> bool {
#[cfg(target_arch = "x86")]
use crate::arch::x86::_mm_movemask_pi8;
#[cfg(target_arch = "x86_64")]
use crate::arch::x86_64::_mm_movemask_pi8;
// _mm_movemask_pi8(a) creates an 8bit mask containing the most
// significant bit of each byte of `a`. If all bits are set,
// then all 8 lanes of the mask are true.
_mm_movemask_pi8(crate::mem::transmute(self))
== u8::max_value() as i32
}
}
impl Any for $id {
#[inline]
#[target_feature(enable = "sse")]
unsafe fn any(self) -> bool {
#[cfg(target_arch = "x86")]
use crate::arch::x86::_mm_movemask_pi8;
#[cfg(target_arch = "x86_64")]
use crate::arch::x86_64::_mm_movemask_pi8;

_mm_movemask_pi8(crate::mem::transmute(self)) != 0
}
}
};
}
1 change: 0 additions & 1 deletion src/lib.rs
Original file line number Diff line number Diff line change
Expand Up @@ -209,7 +209,6 @@
link_llvm_intrinsics,
core_intrinsics,
stmt_expr_attributes,
mmx_target_feature,
crate_visibility_modifier,
custom_inner_attributes
)]
Expand Down