Rollup merge of #151346 - folkertdev:simd-splat, r=workingjubilee

add `simd_splat` intrinsic

Add `simd_splat` which lowers to the LLVM canonical splat sequence.

```llvm
insertelement <N x elem> poison, elem %x, i32 0
shufflevector <N x elem> v0, <N x elem> poison, <N x i32> zeroinitializer
```

Right now we try to fake it using one of

```rust
fn splat(x: u32) -> u32x8 {
    u32x8::from_array([x; 8])
}
```

or (in `stdarch`)

```rust
fn splat(value: $elem_type) -> $name {
    #[derive(Copy, Clone)]
    #[repr(simd)]
    struct JustOne([$elem_type; 1]);
    let one = JustOne([value]);
    // SAFETY: 0 is always in-bounds because we're shuffling
    // a simd type with exactly one element.
    unsafe { simd_shuffle!(one, one, [0; $len]) }
}
```

Both of these can confuse the LLVM optimizer, producing sub-par code. Some examples:

- https://github.com/rust-lang/rust/issues/60637
- https://github.com/rust-lang/rust/issues/137407
- https://github.com/rust-lang/rust/issues/122623
- https://github.com/rust-lang/rust/issues/97804

---

As far as I can tell there is no way to provide a fallback implementation for this intrinsic, because there is no `const` way of evaluating the number of elements (there might be issues beyond that, too). So, I added implementations for all 4 backends.

Both GCC and const-eval appear to have some issues with simd vectors containing pointers. I have a workaround for GCC, but haven't yet been able to make const-eval work. See the comments below.

Currently this just adds the intrinsic, it does not actually use it anywhere yet.

src/intrinsics/simd.rs

@@ -348,6 +348,31 @@ pub(super) fn codegen_simd_intrinsic_call<'tcx>(
             ret.write_cvalue(fx, ret_lane);
         }
 
+        sym::simd_splat => {
+            intrinsic_args!(fx, args => (value); intrinsic);
+
+            if !ret.layout().ty.is_simd() {
+                report_simd_type_validation_error(fx, intrinsic, span, ret.layout().ty);
+                return;
+            }
+            let (lane_count, lane_ty) = ret.layout().ty.simd_size_and_type(fx.tcx);
+
+            if value.layout().ty != lane_ty {
+                fx.tcx.dcx().span_fatal(
+                    span,
+                    format!(
+                        "[simd_splat] expected element type {lane_ty:?}, got {got:?}",
+                        got = value.layout().ty
+                    ),
+                );
+            }
+
+            for i in 0..lane_count {
+                let ret_lane = ret.place_lane(fx, i.into());
+                ret_lane.write_cvalue(fx, value);
+            }
+        }
+
         sym::simd_neg
         | sym::simd_bswap
         | sym::simd_bitreverse