x64: Add support for phadd{w,d} instructions (#5896)

This commit adds support for the bare lowering of the `iadd_pairwise`
instruction with `i16x8` and `i32x4` types on the x64 backend. These
lowerings are achieved with the `phaddw` and `phaddd` instructions,
respectively. Additionally AVX encodings of these instructions are added
too.

The motivation for these new lowerings comes from the relaxed-simd
proposal which will use them in the deterministic lowering of some
instructions on the x64 backend.

cranelift/filetests/filetests/isa/x64/iadd-pairwise-avx.clif

@@ -0,0 +1,54 @@
+test compile precise-output
+set enable_simd
+target x86_64 has_avx
+
+function %iadd_pairwise_i16x8(i16x8, i16x8) -> i16x8 {
+block0(v0: i16x8, v1: i16x8):
+  v2 = iadd_pairwise v0, v1
+  return v2
+}
+
+; VCode:
+;   pushq   %rbp
+;   movq    %rsp, %rbp
+; block0:
+;   vphaddw %xmm0, %xmm1, %xmm0
+;   movq    %rbp, %rsp
+;   popq    %rbp
+;   ret
+; 
+; Disassembled:
+; block0: ; offset 0x0
+;   pushq %rbp
+;   movq %rsp, %rbp
+; block1: ; offset 0x4
+;   vphaddw %xmm1, %xmm0, %xmm0
+;   movq %rbp, %rsp
+;   popq %rbp
+;   retq
+
+function %iadd_pairwise_i32x4(i32x4, i32x4) -> i32x4 {
+block0(v0: i32x4, v1: i32x4):
+  v2 = iadd_pairwise v0, v1
+  return v2
+}
+
+; VCode:
+;   pushq   %rbp
+;   movq    %rsp, %rbp
+; block0:
+;   vphaddd %xmm0, %xmm1, %xmm0
+;   movq    %rbp, %rsp
+;   popq    %rbp
+;   ret
+; 
+; Disassembled:
+; block0: ; offset 0x0
+;   pushq %rbp
+;   movq %rsp, %rbp
+; block1: ; offset 0x4
+;   vphaddd %xmm1, %xmm0, %xmm0
+;   movq %rbp, %rsp
+;   popq %rbp
+;   retq
+

cranelift/filetests/filetests/isa/x64/iadd-pairwise.clif

@@ -0,0 +1,54 @@
+test compile precise-output
+set enable_simd
+target x86_64
+
+function %iadd_pairwise_i16x8(i16x8, i16x8) -> i16x8 {
+block0(v0: i16x8, v1: i16x8):
+  v2 = iadd_pairwise v0, v1
+  return v2
+}
+
+; VCode:
+;   pushq   %rbp
+;   movq    %rsp, %rbp
+; block0:
+;   phaddw  %xmm0, %xmm1, %xmm0
+;   movq    %rbp, %rsp
+;   popq    %rbp
+;   ret
+; 
+; Disassembled:
+; block0: ; offset 0x0
+;   pushq %rbp
+;   movq %rsp, %rbp
+; block1: ; offset 0x4
+;   phaddw %xmm1, %xmm0
+;   movq %rbp, %rsp
+;   popq %rbp
+;   retq
+
+function %iadd_pairwise_i32x4(i32x4, i32x4) -> i32x4 {
+block0(v0: i32x4, v1: i32x4):
+  v2 = iadd_pairwise v0, v1
+  return v2
+}
+
+; VCode:
+;   pushq   %rbp
+;   movq    %rsp, %rbp
+; block0:
+;   phaddd  %xmm0, %xmm1, %xmm0
+;   movq    %rbp, %rsp
+;   popq    %rbp
+;   ret
+; 
+; Disassembled:
+; block0: ; offset 0x0
+;   pushq %rbp
+;   movq %rsp, %rbp
+; block1: ; offset 0x4
+;   phaddd %xmm1, %xmm0
+;   movq %rbp, %rsp
+;   popq %rbp
+;   retq
+