Rollup merge of #142078 - sayantn:more-intrinsics, r=workingjubilee

Add SIMD funnel shift and round-to-even intrinsics

This PR adds 3 new SIMD intrinsics

 - `simd_funnel_shl` - funnel shift left
 - `simd_funnel_shr` - funnel shift right
 - `simd_round_ties_even` (vector version of `round_ties_even_fN`)

TODO (future PR): implement `simd_fsh{l,r}` in miri, cg_gcc and cg_clif (it is surprisingly hard to implement without branches, the common tricks that rotate uses doesn't work because we have 2 elements now. e.g, the `-n&31` trick used by cg_gcc to implement rotate doesn't work with this because then `fshl(a, b, 0)` will be `a | b`)

[#t-compiler > More SIMD intrinsics](https://rust-lang.zulipchat.com/#narrow/channel/131828-t-compiler/topic/More.20SIMD.20intrinsics/with/522130286)

`@rustbot` label T-compiler T-libs A-intrinsics F-core_intrinsics
r? `@workingjubilee`

Author: GuillaumeGomez

src/intrinsics/simd.rs

@@ -36,6 +36,7 @@ pub trait EvalContextExt<'tcx>: crate::MiriInterpCxExt<'tcx> {
             | "ceil"
             | "floor"
             | "round"
+            | "round_ties_even"
             | "trunc"
             | "fsqrt"
             | "fsin"
@@ -71,6 +72,7 @@ pub trait EvalContextExt<'tcx>: crate::MiriInterpCxExt<'tcx> {
                     "ceil" => Op::Round(rustc_apfloat::Round::TowardPositive),
                     "floor" => Op::Round(rustc_apfloat::Round::TowardNegative),
                     "round" => Op::Round(rustc_apfloat::Round::NearestTiesToAway),
+                    "round_ties_even" => Op::Round(rustc_apfloat::Round::NearestTiesToEven),
                     "trunc" => Op::Round(rustc_apfloat::Round::TowardZero),
                     "ctlz" => Op::Numeric(sym::ctlz),
                     "ctpop" => Op::Numeric(sym::ctpop),

tests/pass/intrinsics/portable-simd.rs

@@ -569,6 +569,10 @@ fn simd_round() {
         f32x4::from_array([0.9, 1.001, 2.0, -4.5]).round(),
         f32x4::from_array([1.0, 1.0, 2.0, -5.0])
     );
+    assert_eq!(
+        unsafe { intrinsics::simd_round_ties_even(f32x4::from_array([0.9, 1.001, 2.0, -4.5])) },
+        f32x4::from_array([1.0, 1.0, 2.0, -4.0])
+    );
     assert_eq!(
         f32x4::from_array([0.9, 1.001, 2.0, -4.5]).trunc(),
         f32x4::from_array([0.0, 1.0, 2.0, -4.0])
@@ -586,6 +590,10 @@ fn simd_round() {
         f64x4::from_array([0.9, 1.001, 2.0, -4.5]).round(),
         f64x4::from_array([1.0, 1.0, 2.0, -5.0])
     );
+    assert_eq!(
+        unsafe { intrinsics::simd_round_ties_even(f64x4::from_array([0.9, 1.001, 2.0, -4.5])) },
+        f64x4::from_array([1.0, 1.0, 2.0, -4.0])
+    );
     assert_eq!(
         f64x4::from_array([0.9, 1.001, 2.0, -4.5]).trunc(),
         f64x4::from_array([0.0, 1.0, 2.0, -4.0])