RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-23 11:13:28 +01:00
Code Issues Projects Releases Wiki Activity

[AArch64][FIX] f16 indexed patterns encoding restrictions.

Browse Source
This commit is contained in:
Pavel Iliin 2020-04-22 12:34:36 +01:00
parent 869cef2c82
commit 3003b7b6ab
1 changed files with 7 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
lib/Target/AArch64/AArch64InstrFormats.td
View File

@ -8068,29 +8068,29 @@ multiclass SIMDFPIndexedTiedPatterns<string INST, SDPatternOperator OpNode> {
let Predicates = [HasNEON, HasFullFP16] in {
// Patterns for f16: DUPLANE, DUP scalar and vector_extract.
def : Pat<(v8f16 (OpNode (v8f16 V128:$Rd), (v8f16 V128:$Rn),
(AArch64duplane16 (v8f16 V128:$Rm),
(AArch64duplane16 (v8f16 V128_lo:$Rm),
VectorIndexH:$idx))),
(!cast<Instruction>(INST # "v8i16_indexed")
V128:$Rd, V128:$Rn, V128:$Rm, VectorIndexH:$idx)>;
V128:$Rd, V128:$Rn, V128_lo:$Rm, VectorIndexH:$idx)>;
def : Pat<(v8f16 (OpNode (v8f16 V128:$Rd), (v8f16 V128:$Rn),
(AArch64dup (f16 FPR16Op:$Rm)))),
(!cast<Instruction>(INST # "v8i16_indexed") V128:$Rd, V128:$Rn,
(SUBREG_TO_REG (i32 0), FPR16Op:$Rm, hsub), (i64 0))>;
def : Pat<(v4f16 (OpNode (v4f16 V64:$Rd), (v4f16 V64:$Rn),
(AArch64duplane16 (v8f16 V128:$Rm),
VectorIndexS:$idx))),
(AArch64duplane16 (v8f16 V128_lo:$Rm),
VectorIndexH:$idx))),
(!cast<Instruction>(INST # "v4i16_indexed")
V64:$Rd, V64:$Rn, V128:$Rm, VectorIndexS:$idx)>;
V64:$Rd, V64:$Rn, V128_lo:$Rm, VectorIndexH:$idx)>;
def : Pat<(v4f16 (OpNode (v4f16 V64:$Rd), (v4f16 V64:$Rn),
(AArch64dup (f16 FPR16Op:$Rm)))),
(!cast<Instruction>(INST # "v4i16_indexed") V64:$Rd, V64:$Rn,
(SUBREG_TO_REG (i32 0), FPR16Op:$Rm, hsub), (i64 0))>;
def : Pat<(f16 (OpNode (f16 FPR16:$Rd), (f16 FPR16:$Rn),
(vector_extract (v8f16 V128:$Rm), VectorIndexH:$idx))),
(vector_extract (v8f16 V128_lo:$Rm), VectorIndexH:$idx))),
(!cast<Instruction>(INST # "v1i16_indexed") FPR16:$Rd, FPR16:$Rn,
V128:$Rm, VectorIndexH:$idx)>;
V128_lo:$Rm, VectorIndexH:$idx)>;
} // Predicates = [HasNEON, HasFullFP16]
// 2 variants for the .2s version: DUPLANE from 128-bit and DUP scalar.