[AArch64] Remove redundant FMOV for zero-extended i32/i16 loads to f64

Previously, a load from i32 or i16, followed by zero-extension to i64 and a
bitcast to f64, would emit separate FMOV instructions.

This patch introduces new corresponding TableGen patterns to avoid the unnecessary FMOV.

Tests added:
  - load_u64_from_u32.ll
  - load_u64_from_u16.ll

Author: Amichaxx

llvm/lib/Target/AArch64/AArch64InstrInfo.td

@@ -3913,6 +3913,14 @@ defm LDRSW  : LoadUI<0b10, 0, 0b10, GPR64, uimm12s4, "ldrsw",
 def : Pat<(i64 (zextloadi32 (am_indexed32 GPR64sp:$Rn, uimm12s4:$offset))),
       (SUBREG_TO_REG (i64 0), (LDRWui GPR64sp:$Rn, uimm12s4:$offset), sub_32)>;
 
+// load zero-extended i32, bitcast to f64
+def : Pat <(f64 (bitconvert (i64 (zextloadi32 (am_indexed32 GPR64sp:$Rn, uimm12s4:$offset))))),
+       (SUBREG_TO_REG (i64 0), (LDRSui GPR64sp:$Rn, uimm12s4:$offset), ssub)>;
+
+// load zero-extended i16, bitcast to f64
+def : Pat <(f64 (bitconvert (i64 (zextloadi16 (am_indexed32 GPR64sp:$Rn, uimm12s2:$offset))))),
+           (SUBREG_TO_REG (i64 0), (LDRHui GPR64sp:$Rn, uimm12s2:$offset), hsub)>;
+    
 // Pre-fetch.
 def PRFMui : PrefetchUI<0b11, 0, 0b10, "prfm",
                         [(AArch64Prefetch timm:$Rt,
@@ -10986,4 +10994,4 @@ defm FMMLA : SIMDThreeSameVectorFP8MatrixMul<"fmmla">;
 include "AArch64InstrAtomics.td"
 include "AArch64SVEInstrInfo.td"
 include "AArch64SMEInstrInfo.td"
-include "AArch64InstrGISel.td"
+include "AArch64InstrGISel.td"

llvm/test/CodeGen/AArch64/load_u64_from_u16.ll

@@ -0,0 +1,15 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+; RUN: llc -mtriple=aarch64-linux-gnu -o - %s | FileCheck %s
+
+define double @_Z9load_u64_from_u16_testPj(ptr %n){
+; CHECK-LABEL: _Z9load_u64_from_u16_testPj:
+; CHECK:       // %bb.0: // %entry
+; CHECK-NEXT:    ldr h0, [x0]
+; CHECK-NEXT:    ret
+entry:
+  %0 = load i16, ptr %n, align 2
+  %conv = zext i16 %0 to i64
+  %1 = bitcast i64 %conv to double
+  ret double %1
+}
+

llvm/test/CodeGen/AArch64/load_u64_from_u32.ll

@@ -0,0 +1,14 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+; RUN: llc -mtriple=aarch64-linux-gnu -o - %s | FileCheck %s
+
+define double @_Z9load_u64_from_u32_testPj(ptr %n) {
+; CHECK-LABEL: _Z9load_u64_from_u32_testPj:
+; CHECK:       // %bb.0: // %entry
+; CHECK-NEXT:    ldr s0, [x0]
+; CHECK-NEXT:    ret
+entry:
+  %0 = load i32, ptr %n, align 4
+  %conv = zext i32 %0 to i64
+  %1 = bitcast i64 %conv to double
+  ret double %1
+}