aboutsummaryrefslogtreecommitdiffstats
path: root/lib/Target/ARM/ARMInstrVFP.td
diff options
context:
space:
mode:
Diffstat (limited to 'lib/Target/ARM/ARMInstrVFP.td')
-rw-r--r--lib/Target/ARM/ARMInstrVFP.td107
1 files changed, 89 insertions, 18 deletions
diff --git a/lib/Target/ARM/ARMInstrVFP.td b/lib/Target/ARM/ARMInstrVFP.td
index e0a9314..afff016 100644
--- a/lib/Target/ARM/ARMInstrVFP.td
+++ b/lib/Target/ARM/ARMInstrVFP.td
@@ -11,16 +11,10 @@
//
//===----------------------------------------------------------------------===//
-def SDT_FTOI : SDTypeProfile<1, 1, [SDTCisVT<0, f32>, SDTCisFP<1>]>;
-def SDT_ITOF : SDTypeProfile<1, 1, [SDTCisFP<0>, SDTCisVT<1, f32>]>;
def SDT_CMPFP0 : SDTypeProfile<0, 1, [SDTCisFP<0>]>;
def SDT_VMOVDRR : SDTypeProfile<1, 2, [SDTCisVT<0, f64>, SDTCisVT<1, i32>,
SDTCisSameAs<1, 2>]>;
-def arm_ftoui : SDNode<"ARMISD::FTOUI", SDT_FTOI>;
-def arm_ftosi : SDNode<"ARMISD::FTOSI", SDT_FTOI>;
-def arm_sitof : SDNode<"ARMISD::SITOF", SDT_ITOF>;
-def arm_uitof : SDNode<"ARMISD::UITOF", SDT_ITOF>;
def arm_fmstat : SDNode<"ARMISD::FMSTAT", SDTNone, [SDNPInGlue, SDNPOutGlue]>;
def arm_cmpfp : SDNode<"ARMISD::CMPFP", SDT_ARMCmp, [SDNPOutGlue]>;
def arm_cmpfp0 : SDNode<"ARMISD::CMPFPw0", SDT_CMPFP0, [SDNPOutGlue]>;
@@ -633,7 +627,7 @@ multiclass vcvt_inst<string opc, bits<2> rm,
def SS : ASuInp<0b11101, 0b11, 0b1100, 0b11, 0,
(outs SPR:$Sd), (ins SPR:$Sm),
NoItinerary, !strconcat("vcvt", opc, ".s32.f32\t$Sd, $Sm"),
- [(set SPR:$Sd, (arm_ftosi (node SPR:$Sm)))]>,
+ []>,
Requires<[HasFPARMv8]> {
let Inst{17-16} = rm;
}
@@ -641,7 +635,7 @@ multiclass vcvt_inst<string opc, bits<2> rm,
def US : ASuInp<0b11101, 0b11, 0b1100, 0b01, 0,
(outs SPR:$Sd), (ins SPR:$Sm),
NoItinerary, !strconcat("vcvt", opc, ".u32.f32\t$Sd, $Sm"),
- [(set SPR:$Sd, (arm_ftoui (node SPR:$Sm)))]>,
+ []>,
Requires<[HasFPARMv8]> {
let Inst{17-16} = rm;
}
@@ -649,7 +643,7 @@ multiclass vcvt_inst<string opc, bits<2> rm,
def SD : ASuInp<0b11101, 0b11, 0b1100, 0b11, 0,
(outs SPR:$Sd), (ins DPR:$Dm),
NoItinerary, !strconcat("vcvt", opc, ".s32.f64\t$Sd, $Dm"),
- [(set SPR:$Sd, (arm_ftosi (f64 (node (f64 DPR:$Dm)))))]>,
+ []>,
Requires<[HasFPARMv8, HasDPVFP]> {
bits<5> Dm;
@@ -664,7 +658,7 @@ multiclass vcvt_inst<string opc, bits<2> rm,
def UD : ASuInp<0b11101, 0b11, 0b1100, 0b01, 0,
(outs SPR:$Sd), (ins DPR:$Dm),
NoItinerary, !strconcat("vcvt", opc, ".u32.f64\t$Sd, $Dm"),
- [(set SPR:$Sd, (arm_ftoui (f64 (node (f64 DPR:$Dm)))))]>,
+ []>,
Requires<[HasFPARMv8, HasDPVFP]> {
bits<5> Dm;
@@ -676,6 +670,27 @@ multiclass vcvt_inst<string opc, bits<2> rm,
let Inst{8} = 1;
}
}
+
+ let Predicates = [HasFPARMv8] in {
+ def : Pat<(i32 (fp_to_sint (node SPR:$a))),
+ (COPY_TO_REGCLASS
+ (!cast<Instruction>(NAME#"SS") SPR:$a),
+ GPR)>;
+ def : Pat<(i32 (fp_to_uint (node SPR:$a))),
+ (COPY_TO_REGCLASS
+ (!cast<Instruction>(NAME#"US") SPR:$a),
+ GPR)>;
+ }
+ let Predicates = [HasFPARMv8, HasDPVFP] in {
+ def : Pat<(i32 (fp_to_sint (node (f64 DPR:$a)))),
+ (COPY_TO_REGCLASS
+ (!cast<Instruction>(NAME#"SD") DPR:$a),
+ GPR)>;
+ def : Pat<(i32 (fp_to_uint (node (f64 DPR:$a)))),
+ (COPY_TO_REGCLASS
+ (!cast<Instruction>(NAME#"UD") DPR:$a),
+ GPR)>;
+ }
}
defm VCVTA : vcvt_inst<"a", 0b00, frnd>;
@@ -980,14 +995,22 @@ class AVConv1InSs_Encode<bits<5> opcod1, bits<2> opcod2, bits<4> opcod3,
def VSITOD : AVConv1IDs_Encode<0b11101, 0b11, 0b1000, 0b1011,
(outs DPR:$Dd), (ins SPR:$Sm),
IIC_fpCVTID, "vcvt", ".f64.s32\t$Dd, $Sm",
- [(set DPR:$Dd, (f64 (arm_sitof SPR:$Sm)))]> {
+ []> {
let Inst{7} = 1; // s32
}
+let Predicates=[HasVFP2, HasDPVFP] in {
+ def : VFPPat<(f64 (sint_to_fp GPR:$a)),
+ (VSITOD (COPY_TO_REGCLASS GPR:$a, SPR))>;
+
+ def : VFPPat<(f64 (sint_to_fp (i32 (load addrmode5:$a)))),
+ (VSITOD (VLDRS addrmode5:$a))>;
+}
+
def VSITOS : AVConv1InSs_Encode<0b11101, 0b11, 0b1000, 0b1010,
(outs SPR:$Sd),(ins SPR:$Sm),
IIC_fpCVTIS, "vcvt", ".f32.s32\t$Sd, $Sm",
- [(set SPR:$Sd, (arm_sitof SPR:$Sm))]> {
+ []> {
let Inst{7} = 1; // s32
// Some single precision VFP instructions may be executed on both NEON and
@@ -995,17 +1018,31 @@ def VSITOS : AVConv1InSs_Encode<0b11101, 0b11, 0b1000, 0b1010,
let D = VFPNeonA8Domain;
}
+def : VFPNoNEONPat<(f32 (sint_to_fp GPR:$a)),
+ (VSITOS (COPY_TO_REGCLASS GPR:$a, SPR))>;
+
+def : VFPNoNEONPat<(f32 (sint_to_fp (i32 (load addrmode5:$a)))),
+ (VSITOS (VLDRS addrmode5:$a))>;
+
def VUITOD : AVConv1IDs_Encode<0b11101, 0b11, 0b1000, 0b1011,
(outs DPR:$Dd), (ins SPR:$Sm),
IIC_fpCVTID, "vcvt", ".f64.u32\t$Dd, $Sm",
- [(set DPR:$Dd, (f64 (arm_uitof SPR:$Sm)))]> {
+ []> {
let Inst{7} = 0; // u32
}
+let Predicates=[HasVFP2, HasDPVFP] in {
+ def : VFPPat<(f64 (uint_to_fp GPR:$a)),
+ (VUITOD (COPY_TO_REGCLASS GPR:$a, SPR))>;
+
+ def : VFPPat<(f64 (uint_to_fp (i32 (load addrmode5:$a)))),
+ (VUITOD (VLDRS addrmode5:$a))>;
+}
+
def VUITOS : AVConv1InSs_Encode<0b11101, 0b11, 0b1000, 0b1010,
(outs SPR:$Sd), (ins SPR:$Sm),
IIC_fpCVTIS, "vcvt", ".f32.u32\t$Sd, $Sm",
- [(set SPR:$Sd, (arm_uitof SPR:$Sm))]> {
+ []> {
let Inst{7} = 0; // u32
// Some single precision VFP instructions may be executed on both NEON and
@@ -1013,6 +1050,12 @@ def VUITOS : AVConv1InSs_Encode<0b11101, 0b11, 0b1000, 0b1010,
let D = VFPNeonA8Domain;
}
+def : VFPNoNEONPat<(f32 (uint_to_fp GPR:$a)),
+ (VUITOS (COPY_TO_REGCLASS GPR:$a, SPR))>;
+
+def : VFPNoNEONPat<(f32 (uint_to_fp (i32 (load addrmode5:$a)))),
+ (VUITOS (VLDRS addrmode5:$a))>;
+
// FP -> Int:
class AVConv1IsD_Encode<bits<5> opcod1, bits<2> opcod2, bits<4> opcod3,
@@ -1055,14 +1098,22 @@ class AVConv1InsS_Encode<bits<5> opcod1, bits<2> opcod2, bits<4> opcod3,
def VTOSIZD : AVConv1IsD_Encode<0b11101, 0b11, 0b1101, 0b1011,
(outs SPR:$Sd), (ins DPR:$Dm),
IIC_fpCVTDI, "vcvt", ".s32.f64\t$Sd, $Dm",
- [(set SPR:$Sd, (arm_ftosi (f64 DPR:$Dm)))]> {
+ []> {
let Inst{7} = 1; // Z bit
}
+let Predicates=[HasVFP2, HasDPVFP] in {
+ def : VFPPat<(i32 (fp_to_sint (f64 DPR:$a))),
+ (COPY_TO_REGCLASS (VTOSIZD DPR:$a), GPR)>;
+
+ def : VFPPat<(store (i32 (fp_to_sint (f64 DPR:$a))), addrmode5:$ptr),
+ (VSTRS (VTOSIZD DPR:$a), addrmode5:$ptr)>;
+}
+
def VTOSIZS : AVConv1InsS_Encode<0b11101, 0b11, 0b1101, 0b1010,
(outs SPR:$Sd), (ins SPR:$Sm),
IIC_fpCVTSI, "vcvt", ".s32.f32\t$Sd, $Sm",
- [(set SPR:$Sd, (arm_ftosi SPR:$Sm))]> {
+ []> {
let Inst{7} = 1; // Z bit
// Some single precision VFP instructions may be executed on both NEON and
@@ -1070,17 +1121,31 @@ def VTOSIZS : AVConv1InsS_Encode<0b11101, 0b11, 0b1101, 0b1010,
let D = VFPNeonA8Domain;
}
+def : VFPNoNEONPat<(i32 (fp_to_sint SPR:$a)),
+ (COPY_TO_REGCLASS (VTOSIZS SPR:$a), GPR)>;
+
+def : VFPNoNEONPat<(store (i32 (fp_to_sint (f32 SPR:$a))), addrmode5:$ptr),
+ (VSTRS (VTOSIZS SPR:$a), addrmode5:$ptr)>;
+
def VTOUIZD : AVConv1IsD_Encode<0b11101, 0b11, 0b1100, 0b1011,
(outs SPR:$Sd), (ins DPR:$Dm),
IIC_fpCVTDI, "vcvt", ".u32.f64\t$Sd, $Dm",
- [(set SPR:$Sd, (arm_ftoui (f64 DPR:$Dm)))]> {
+ []> {
let Inst{7} = 1; // Z bit
}
+let Predicates=[HasVFP2, HasDPVFP] in {
+ def : VFPPat<(i32 (fp_to_uint (f64 DPR:$a))),
+ (COPY_TO_REGCLASS (VTOUIZD DPR:$a), GPR)>;
+
+ def : VFPPat<(store (i32 (fp_to_uint (f64 DPR:$a))), addrmode5:$ptr),
+ (VSTRS (VTOUIZD DPR:$a), addrmode5:$ptr)>;
+}
+
def VTOUIZS : AVConv1InsS_Encode<0b11101, 0b11, 0b1100, 0b1010,
(outs SPR:$Sd), (ins SPR:$Sm),
IIC_fpCVTSI, "vcvt", ".u32.f32\t$Sd, $Sm",
- [(set SPR:$Sd, (arm_ftoui SPR:$Sm))]> {
+ []> {
let Inst{7} = 1; // Z bit
// Some single precision VFP instructions may be executed on both NEON and
@@ -1088,6 +1153,12 @@ def VTOUIZS : AVConv1InsS_Encode<0b11101, 0b11, 0b1100, 0b1010,
let D = VFPNeonA8Domain;
}
+def : VFPNoNEONPat<(i32 (fp_to_uint SPR:$a)),
+ (COPY_TO_REGCLASS (VTOUIZS SPR:$a), GPR)>;
+
+def : VFPNoNEONPat<(store (i32 (fp_to_uint (f32 SPR:$a))), addrmode5:$ptr),
+ (VSTRS (VTOUIZS SPR:$a), addrmode5:$ptr)>;
+
// And the Z bit '0' variants, i.e. use the rounding mode specified by FPSCR.
let Uses = [FPSCR] in {
// FIXME: Verify encoding after integrated assembler is working.
generated by cgit v1.1 at 2024-12-22 10:02:37 +0000