Skip to content

JIT: Added SVE LoadVector*NonFaultingZeroExtendTo* APIs #102860

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

Merged
merged 5 commits into from
May 31, 2024
Merged
Show file tree
Hide file tree
Changes from all 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
5 changes: 4 additions & 1 deletion src/coreclr/jit/emitarm64sve.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -4383,9 +4383,12 @@ void emitter::emitInsSve_R_R_R(instruction ins,
case INS_sve_ld1w:
case INS_sve_ld1sw:
case INS_sve_ld1d:
case INS_sve_ldnf1b:
case INS_sve_ldnf1sb:
case INS_sve_ldnf1h:
case INS_sve_ldnf1sh:
case INS_sve_ldnf1w:
case INS_sve_ldnf1sw:
case INS_sve_ldnf1sb:
return emitIns_R_R_R_I(ins, size, reg1, reg2, reg3, 0, opt);

default:
Expand Down
12 changes: 12 additions & 0 deletions src/coreclr/jit/hwintrinsiclistarm64sve.h
Original file line number Diff line number Diff line change
Expand Up @@ -71,6 +71,12 @@ HARDWARE_INTRINSIC(Sve, FusedMultiplySubtractNegated,
HARDWARE_INTRINSIC(Sve, LeadingSignCount, -1, -1, false, {INS_sve_cls, INS_invalid, INS_sve_cls, INS_invalid, INS_sve_cls, INS_invalid, INS_sve_cls, INS_invalid, INS_invalid, INS_invalid}, HW_Category_SIMD, HW_Flag_Scalable|HW_Flag_BaseTypeFromFirstArg|HW_Flag_EmbeddedMaskedOperation|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LeadingZeroCount, -1, -1, false, {INS_sve_clz, INS_sve_clz, INS_sve_clz, INS_sve_clz, INS_sve_clz, INS_sve_clz, INS_sve_clz, INS_sve_clz, INS_invalid, INS_invalid}, HW_Category_SIMD, HW_Flag_Scalable|HW_Flag_BaseTypeFromFirstArg|HW_Flag_EmbeddedMaskedOperation|HW_Flag_LowMaskedOperation)
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_ExplicitMaskedOperation|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorByteNonFaultingZeroExtendToInt16, -1, 1, false, {INS_invalid, INS_invalid, INS_sve_ldnf1b, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_EmbeddedMaskedOperation|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorByteNonFaultingZeroExtendToInt32, -1, 1, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ldnf1b, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_EmbeddedMaskedOperation|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorByteNonFaultingZeroExtendToInt64, -1, 1, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ldnf1b, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_EmbeddedMaskedOperation|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorByteNonFaultingZeroExtendToUInt16, -1, 1, false, {INS_invalid, INS_invalid, INS_invalid, INS_sve_ldnf1b, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_EmbeddedMaskedOperation|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorByteNonFaultingZeroExtendToUInt32, -1, 1, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ldnf1b, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_EmbeddedMaskedOperation|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorByteNonFaultingZeroExtendToUInt64, -1, 1, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ldnf1b, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_EmbeddedMaskedOperation|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_ExplicitMaskedOperation|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_ExplicitMaskedOperation|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_ExplicitMaskedOperation|HW_Flag_LowMaskedOperation)
Expand Down Expand Up @@ -101,10 +107,16 @@ HARDWARE_INTRINSIC(Sve, LoadVectorSByteSignExtendToInt64,
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_ExplicitMaskedOperation|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_ExplicitMaskedOperation|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_ExplicitMaskedOperation|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorUInt16NonFaultingZeroExtendToInt32, -1, 1, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ldnf1h, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_EmbeddedMaskedOperation|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorUInt16NonFaultingZeroExtendToInt64, -1, 1, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ldnf1h, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_EmbeddedMaskedOperation|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorUInt16NonFaultingZeroExtendToUInt32, -1, 1, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ldnf1h, INS_invalid, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_EmbeddedMaskedOperation|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorUInt16NonFaultingZeroExtendToUInt64, -1, 1, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ldnf1h, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_EmbeddedMaskedOperation|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_ExplicitMaskedOperation|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_ExplicitMaskedOperation|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_ExplicitMaskedOperation|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_ExplicitMaskedOperation|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorUInt32NonFaultingZeroExtendToInt64, -1, 1, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ldnf1w, INS_invalid, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_EmbeddedMaskedOperation|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, LoadVectorUInt32NonFaultingZeroExtendToUInt64, -1, 1, false, {INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_invalid, INS_sve_ldnf1w, INS_invalid, INS_invalid}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_EmbeddedMaskedOperation|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_ExplicitMaskedOperation|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_ExplicitMaskedOperation|HW_Flag_LowMaskedOperation)
HARDWARE_INTRINSIC(Sve, Load2xVectorAndUnzip, -1, 2, true, {INS_sve_ld2b, INS_sve_ld2b, INS_sve_ld2h, INS_sve_ld2h, INS_sve_ld2w, INS_sve_ld2w, INS_sve_ld2d, INS_sve_ld2d, INS_sve_ld2w, INS_sve_ld2d}, HW_Category_MemoryLoad, HW_Flag_Scalable|HW_Flag_SpecialImport|HW_Flag_SpecialCodeGen|HW_Flag_MultiReg|HW_Flag_ExplicitMaskedOperation|HW_Flag_LowMaskedOperation|HW_Flag_NeedsConsecutiveRegisters|HW_Flag_BaseTypeFromFirstArg)
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -1710,6 +1710,78 @@ internal Arm64() { }
/// </summary>
public static unsafe Vector<ulong> LoadVectorUInt32ZeroExtendToUInt64(Vector<ulong> mask, uint* address) { throw new PlatformNotSupportedException(); }

/// <summary>
/// svint16_t svldnf1ub_s16(svbool_t pg, const uint8_t *base)
/// LDNF1B Zresult.H, Pg/Z, [Xbase, #0, MUL VL]
/// </summary>
public static unsafe Vector<short> LoadVectorByteNonFaultingZeroExtendToInt16(byte* address) { throw new PlatformNotSupportedException(); }

/// <summary>
/// svint32_t svldnf1ub_s32(svbool_t pg, const uint8_t *base)
/// LDNF1B Zresult.S, Pg/Z, [Xbase, #0, MUL VL]
/// </summary>
public static unsafe Vector<int> LoadVectorByteNonFaultingZeroExtendToInt32(byte* address) { throw new PlatformNotSupportedException(); }

/// <summary>
/// svint64_t svldnf1ub_s64(svbool_t pg, const uint8_t *base)
/// LDNF1B Zresult.D, Pg/Z, [Xbase, #0, MUL VL]
/// </summary>
public static unsafe Vector<long> LoadVectorByteNonFaultingZeroExtendToInt64(byte* address) { throw new PlatformNotSupportedException(); }

/// <summary>
/// svuint16_t svldnf1ub_u16(svbool_t pg, const uint8_t *base)
/// LDNF1B Zresult.H, Pg/Z, [Xbase, #0, MUL VL]
/// </summary>
public static unsafe Vector<ushort> LoadVectorByteNonFaultingZeroExtendToUInt16(byte* address) { throw new PlatformNotSupportedException(); }

/// <summary>
/// svuint32_t svldnf1ub_u32(svbool_t pg, const uint8_t *base)
/// LDNF1B Zresult.S, Pg/Z, [Xbase, #0, MUL VL]
/// </summary>
public static unsafe Vector<uint> LoadVectorByteNonFaultingZeroExtendToUInt32(byte* address) { throw new PlatformNotSupportedException(); }

/// <summary>
/// svuint64_t svldnf1ub_u64(svbool_t pg, const uint8_t *base)
/// LDNF1B Zresult.D, Pg/Z, [Xbase, #0, MUL VL]
/// </summary>
public static unsafe Vector<ulong> LoadVectorByteNonFaultingZeroExtendToUInt64(byte* address) { throw new PlatformNotSupportedException(); }

/// <summary>
/// svint32_t svldnf1uh_s32(svbool_t pg, const uint16_t *base)
/// LDNF1H Zresult.S, Pg/Z, [Xbase, #0, MUL VL]
/// </summary>
public static unsafe Vector<int> LoadVectorUInt16NonFaultingZeroExtendToInt32(ushort* address) { throw new PlatformNotSupportedException(); }

/// <summary>
/// svint64_t svldnf1uh_s64(svbool_t pg, const uint16_t *base)
/// LDNF1H Zresult.D, Pg/Z, [Xbase, #0, MUL VL]
/// </summary>
public static unsafe Vector<long> LoadVectorUInt16NonFaultingZeroExtendToInt64(ushort* address) { throw new PlatformNotSupportedException(); }

/// <summary>
/// svuint32_t svldnf1uh_u32(svbool_t pg, const uint16_t *base)
/// LDNF1H Zresult.S, Pg/Z, [Xbase, #0, MUL VL]
/// </summary>
public static unsafe Vector<uint> LoadVectorUInt16NonFaultingZeroExtendToUInt32(ushort* address) { throw new PlatformNotSupportedException(); }

/// <summary>
/// svuint64_t svldnf1uh_u64(svbool_t pg, const uint16_t *base)
/// LDNF1H Zresult.D, Pg/Z, [Xbase, #0, MUL VL]
/// </summary>
public static unsafe Vector<ulong> LoadVectorUInt16NonFaultingZeroExtendToUInt64(ushort* address) { throw new PlatformNotSupportedException(); }

/// <summary>
/// svint64_t svldnf1uw_s64(svbool_t pg, const uint32_t *base)
/// LDNF1W Zresult.D, Pg/Z, [Xbase, #0, MUL VL]
/// </summary>
public static unsafe Vector<long> LoadVectorUInt32NonFaultingZeroExtendToInt64(uint* address) { throw new PlatformNotSupportedException(); }

/// <summary>
/// svuint64_t svldnf1uw_u64(svbool_t pg, const uint32_t *base)
/// LDNF1W Zresult.D, Pg/Z, [Xbase, #0, MUL VL]
/// </summary>
public static unsafe Vector<ulong> LoadVectorUInt32NonFaultingZeroExtendToUInt64(uint* address) { throw new PlatformNotSupportedException(); }

/// <summary>
/// svuint8x2_t svld2[_u8](svbool_t pg, const uint8_t *base)
/// LD2B {Zresult0.B, Zresult1.B}, Pg/Z, [Xbase, #0, MUL VL]
Expand Down
Loading
Loading