Skip to content

Revert "Change implementation of vld1_dup" #2112

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

Merged
folkertdev merged 1 commit into from
May 10, 2026
+9 −9
Merged

Revert "Change implementation of vld1_dup" #2112

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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 crates/core_arch/src/arm_shared/neon/generated.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14614,8 +14614,8 @@ pub fn vhsubq_u32(a: uint32x4_t, b: uint32x4_t) -> uint32x4_t {
#[unstable(feature = "stdarch_neon_f16", issue = "136306")]
#[cfg(not(target_arch = "arm64ec"))]
pub unsafe fn vld1_dup_f16(ptr: *const f16) -> float16x4_t {
let x = vld1_lane_f16::<0>(ptr, transmute(f16x4::splat(0.0)));
vdup_lane_f16::<0>(x)
let x: float16x4_t = vld1_lane_f16::<0>(ptr, transmute(f16x4::splat(0.0)));
simd_shuffle!(x, x, [0, 0, 0, 0])
}
#[doc = "Load one single-element structure and replicate to all lanes of one register"]
#[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vld1q_dup_f16)"]
Expand All @@ -14633,8 +14633,8 @@ pub unsafe fn vld1_dup_f16(ptr: *const f16) -> float16x4_t {
#[unstable(feature = "stdarch_neon_f16", issue = "136306")]
#[cfg(not(target_arch = "arm64ec"))]
pub unsafe fn vld1q_dup_f16(ptr: *const f16) -> float16x8_t {
let x = vld1q_lane_f16::<0>(ptr, transmute(f16x8::splat(0.0)));
vdupq_laneq_f16::<0>(x)
let x: float16x8_t = vld1q_lane_f16::<0>(ptr, transmute(f16x8::splat(0.0)));
simd_shuffle!(x, x, [0, 0, 0, 0, 0, 0, 0, 0])
}
#[doc = "Load one single-element structure and Replicate to all lanes (of one register)."]
#[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vld1_dup_f32)"]
Expand Down
10 changes: 5 additions & 5 deletions crates/stdarch-gen-arm/spec/neon/arm_shared.spec.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2835,7 +2835,7 @@ intrinsics:
- FnCall: [static_assert_uimm_bits!, [LANE, '{type[3]}']]
- FnCall: [simd_insert!, [src, "LANE as u32", "*ptr"]]

- name: "vld1{neon_type[1].dup_nox}"
- name: "vld1{type[2]}_{neon_type[1]}"
doc: "Load one single-element structure and replicate to all lanes of one register"
arguments: ["ptr: {type[0]}"]
return_type: "{neon_type[1]}"
Expand All @@ -2849,11 +2849,11 @@ intrinsics:
safety:
unsafe: [neon]
types:
- ["*const f16", float16x4_t, '_lane', 'f16x4']
- ["*const f16", float16x8_t, 'q_laneq', 'f16x8']
- ["*const f16", float16x4_t, '_dup', 'f16x4', "[0, 0, 0, 0]"]
- ["*const f16", float16x8_t, 'q_dup', 'f16x8', "[0, 0, 0, 0, 0, 0, 0, 0]"]
compose:
- Let: [x, {FnCall: ["vld1{neon_type[1].lane_nox}", [ptr, {FnCall: [transmute, ["{type[3]}::splat(0.0)"]]}], [0]]}]
- FnCall: ['vdup{type[2]}_{neon_type[1]}', [x], [0]]
- Let: [x, "{neon_type[1]}", "vld1{neon_type[1].lane_nox}::<0>(ptr, transmute({type[3]}::splat(0.0)))"]
- FnCall: [simd_shuffle!, [x, x, "{type[4]}"]]


- name: "vld2{neon_type[1].nox}"
Expand Down
Loading