Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

JIT ARM64-SVE: Add Sve.LoadVector*ZeroExtendTo*() #101291

Merged
merged 7 commits into from
Apr 25, 2024
Merged
Show file tree
Hide file tree
Changes from 2 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions src/coreclr/jit/emitarm64sve.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -4349,8 +4349,11 @@ void emitter::emitInsSve_R_R_R(instruction ins,
break;

case INS_sve_ld1b:
case INS_sve_ld1sb:
case INS_sve_ld1h:
case INS_sve_ld1sh:
case INS_sve_ld1w:
case INS_sve_ld1sw:
case INS_sve_ld1d:
return emitIns_R_R_R_I(ins, size, reg1, reg2, reg3, 0, opt);

Expand Down
24 changes: 24 additions & 0 deletions src/coreclr/jit/gentree.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -26398,6 +26398,30 @@ bool GenTreeHWIntrinsic::OperIsMemoryLoad(GenTree** pAddr) const
break;

case NI_Sve_LoadVector:
case NI_Sve_LoadVectorByteZeroExtendToInt16:
case NI_Sve_LoadVectorByteZeroExtendToInt32:
case NI_Sve_LoadVectorByteZeroExtendToInt64:
case NI_Sve_LoadVectorByteZeroExtendToUInt16:
case NI_Sve_LoadVectorByteZeroExtendToUInt32:
case NI_Sve_LoadVectorByteZeroExtendToUInt64:
case NI_Sve_LoadVectorInt16SignExtendToInt32:
case NI_Sve_LoadVectorInt16SignExtendToInt64:
case NI_Sve_LoadVectorInt16SignExtendToUInt32:
case NI_Sve_LoadVectorInt16SignExtendToUInt64:
case NI_Sve_LoadVectorInt32SignExtendToInt64:
case NI_Sve_LoadVectorInt32SignExtendToUInt64:
case NI_Sve_LoadVectorSByteSignExtendToInt16:
case NI_Sve_LoadVectorSByteSignExtendToInt32:
case NI_Sve_LoadVectorSByteSignExtendToInt64:
case NI_Sve_LoadVectorSByteSignExtendToUInt16:
case NI_Sve_LoadVectorSByteSignExtendToUInt32:
case NI_Sve_LoadVectorSByteSignExtendToUInt64:
case NI_Sve_LoadVectorUInt16ZeroExtendToInt32:
case NI_Sve_LoadVectorUInt16ZeroExtendToInt64:
case NI_Sve_LoadVectorUInt16ZeroExtendToUInt32:
case NI_Sve_LoadVectorUInt16ZeroExtendToUInt64:
case NI_Sve_LoadVectorUInt32ZeroExtendToInt64:
case NI_Sve_LoadVectorUInt32ZeroExtendToUInt64:
addr = Op(2);
break;
#endif // TARGET_ARM64
Expand Down
24 changes: 24 additions & 0 deletions src/coreclr/jit/hwintrinsiclistarm64sve.h
Original file line number Diff line number Diff line change
Expand Up @@ -29,6 +29,30 @@ HARDWARE_INTRINSIC(Sve, CreateTrueMaskUInt32,
HARDWARE_INTRINSIC(Sve, CreateTrueMaskUInt64, -1, 1, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ptrue, INS_invalid, INS_invalid}, HW_Category_EnumPattern, HW_Flag_Scalable|HW_Flag_HasImmediateOperand|HW_Flag_ReturnsPerElementMask)

HARDWARE_INTRINSIC(Sve, LoadVector, -1, 2, true, {INS_sve_ld1b, INS_sve_ld1b, INS_sve_ld1h, INS_sve_ld1h, INS_sve_ld1w, INS_sve_ld1w, INS_sve_ld1d, INS_sve_ld1d, INS_sve_ld1w, INS_sve_ld1d}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorByteZeroExtendToInt16, -1, 2, false, {INS_invalid, INS_invalid, INS_sve_ld1b, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorByteZeroExtendToInt32, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1b, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorByteZeroExtendToInt64, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1b, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorByteZeroExtendToUInt16, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1b, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorByteZeroExtendToUInt32, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1b, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorByteZeroExtendToUInt64, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1b, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorInt16SignExtendToInt32, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1sh, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorInt16SignExtendToInt64, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1sh, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorInt16SignExtendToUInt32, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1sh, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorInt16SignExtendToUInt64, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1sh, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorInt32SignExtendToInt64, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1sw, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorInt32SignExtendToUInt64, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1sw, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorSByteSignExtendToInt16, -1, 2, false, {INS_invalid, INS_invalid, INS_sve_ld1sb, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorSByteSignExtendToInt32, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1sb, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorSByteSignExtendToInt64, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1sb, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorSByteSignExtendToUInt16, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1sb, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorSByteSignExtendToUInt32, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1sb, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorSByteSignExtendToUInt64, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1sb, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorUInt16ZeroExtendToInt32, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1h, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorUInt16ZeroExtendToInt64, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1h, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorUInt16ZeroExtendToUInt32, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1h, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorUInt16ZeroExtendToUInt64, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1h, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorUInt32ZeroExtendToInt64, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1w, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorUInt32ZeroExtendToUInt64, -1, 2, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ld1w, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_LowMaskedOperation)



Expand Down
Loading
Loading