1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2025-01-31 12:41:49 +01:00

[AArch64][SVE] Add remaining SVE2 mla indexed intrinsics.

Summary:
Add remaining SVE2 mla indexed intrinsics:
- sqdmlalb, sqdmlalt, sqdmlslb, sqdmlslt

Add suffix _lanes and switch immediate types to i32 for all mla indexed intrinsics to align with ACLE builtin definitions.

Reviewers: efriedma, sdesmalen, cameron.mcinally, c-rhodes, rengolin, kmclaughlin

Subscribers: tschuett, kristof.beyls, hiraditya, rkruppe, arphaman, psnobl, llvm-commits, amehsan

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D73633
This commit is contained in:
Danilo Carvalho Grael 2020-01-29 11:28:13 -05:00
parent c96a29c643
commit 95d6d6606f
4 changed files with 401 additions and 169 deletions

View File

@ -1100,7 +1100,7 @@ let TargetPrefix = "aarch64" in { // All intrinsics start with "llvm.aarch64.".
[LLVMMatchType<0>,
LLVMSubdivide2VectorType<0>,
LLVMSubdivide2VectorType<0>,
llvm_i64_ty],
llvm_i32_ty],
[IntrNoMem, ImmArg<3>]>;
// NOTE: There is no relationship between these intrinsics beyond an attempt
@ -1791,13 +1791,17 @@ def int_aarch64_sve_sqshrunt : SVE2_2VectorArg_Imm_Narrowing_Intrinsic;
def int_aarch64_sve_sqrshrunb : SVE2_1VectorArg_Imm_Narrowing_Intrinsic;
def int_aarch64_sve_sqrshrunt : SVE2_2VectorArg_Imm_Narrowing_Intrinsic;
def int_aarch64_sve_smlalb : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_smlalt : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_umlalb : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_umlalt : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_smlslb : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_smlslt : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_umlslb : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_umlslt : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_smlalb_lane : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_smlalt_lane : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_umlalb_lane : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_umlalt_lane : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_smlslb_lane : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_smlslt_lane : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_umlslb_lane : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_umlslt_lane : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_sqdmlalb_lane : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_sqdmlalt_lane : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_sqdmlslb_lane : SVE2_3VectorArg_Indexed_Intrinsic;
def int_aarch64_sve_sqdmlslt_lane : SVE2_3VectorArg_Indexed_Intrinsic;
}

View File

@ -1467,14 +1467,14 @@ let Predicates = [HasSVE2] in {
defm SQDMULLT_ZZZI : sve2_int_mul_long_by_indexed_elem<0b101, "sqdmullt">;
// SVE2 integer multiply-add long (indexed)
defm SMLALB_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1000, "smlalb", int_aarch64_sve_smlalb>;
defm SMLALT_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1001, "smlalt", int_aarch64_sve_smlalt>;
defm UMLALB_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1010, "umlalb", int_aarch64_sve_umlalb>;
defm UMLALT_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1011, "umlalt", int_aarch64_sve_umlalt>;
defm SMLSLB_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1100, "smlslb", int_aarch64_sve_smlslb>;
defm SMLSLT_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1101, "smlslt", int_aarch64_sve_smlslt>;
defm UMLSLB_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1110, "umlslb", int_aarch64_sve_umlslb>;
defm UMLSLT_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1111, "umlslt", int_aarch64_sve_umlslt>;
defm SMLALB_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1000, "smlalb", int_aarch64_sve_smlalb_lane>;
defm SMLALT_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1001, "smlalt", int_aarch64_sve_smlalt_lane>;
defm UMLALB_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1010, "umlalb", int_aarch64_sve_umlalb_lane>;
defm UMLALT_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1011, "umlalt", int_aarch64_sve_umlalt_lane>;
defm SMLSLB_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1100, "smlslb", int_aarch64_sve_smlslb_lane>;
defm SMLSLT_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1101, "smlslt", int_aarch64_sve_smlslt_lane>;
defm UMLSLB_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1110, "umlslb", int_aarch64_sve_umlslb_lane>;
defm UMLSLT_ZZZI : sve2_int_mla_long_by_indexed_elem<0b1111, "umlslt", int_aarch64_sve_umlslt_lane>;
// SVE2 integer multiply-add long (vectors, unpredicated)
defm SMLALB_ZZZ : sve2_int_mla_long<0b10000, "smlalb">;
@ -1487,10 +1487,10 @@ let Predicates = [HasSVE2] in {
defm UMLSLT_ZZZ : sve2_int_mla_long<0b10111, "umlslt">;
// SVE2 saturating multiply-add long (indexed)
defm SQDMLALB_ZZZI : sve2_int_mla_long_by_indexed_elem<0b0100, "sqdmlalb", null_frag>;
defm SQDMLALT_ZZZI : sve2_int_mla_long_by_indexed_elem<0b0101, "sqdmlalt", null_frag>;
defm SQDMLSLB_ZZZI : sve2_int_mla_long_by_indexed_elem<0b0110, "sqdmlslb", null_frag>;
defm SQDMLSLT_ZZZI : sve2_int_mla_long_by_indexed_elem<0b0111, "sqdmlslt", null_frag>;
defm SQDMLALB_ZZZI : sve2_int_mla_long_by_indexed_elem<0b0100, "sqdmlalb", int_aarch64_sve_sqdmlalb_lane>;
defm SQDMLALT_ZZZI : sve2_int_mla_long_by_indexed_elem<0b0101, "sqdmlalt", int_aarch64_sve_sqdmlalt_lane>;
defm SQDMLSLB_ZZZI : sve2_int_mla_long_by_indexed_elem<0b0110, "sqdmlslb", int_aarch64_sve_sqdmlslb_lane>;
defm SQDMLSLT_ZZZI : sve2_int_mla_long_by_indexed_elem<0b0111, "sqdmlslt", int_aarch64_sve_sqdmlslt_lane>;
// SVE2 saturating multiply-add long (vectors, unpredicated)
defm SQDMLALB_ZZZ : sve2_int_mla_long<0b11000, "sqdmlalb">;

View File

@ -2414,7 +2414,7 @@ multiclass sve2_int_mla_by_indexed_elem<bits<2> opc, bit S, string asm,
multiclass sve2_int_mla_long_by_indexed_elem<bits<4> opc, string asm, SDPatternOperator op> {
def _S : sve2_int_mla_by_indexed_elem<0b10, { opc{3}, 0b0, opc{2-1}, ?, opc{0} },
asm, ZPR32, ZPR16, ZPR3b16, VectorIndexH> {
asm, ZPR32, ZPR16, ZPR3b16, VectorIndexH32b> {
bits<3> Zm;
bits<3> iop;
let Inst{20-19} = iop{2-1};
@ -2422,7 +2422,7 @@ multiclass sve2_int_mla_long_by_indexed_elem<bits<4> opc, string asm, SDPatternO
let Inst{11} = iop{0};
}
def _D : sve2_int_mla_by_indexed_elem<0b11, { opc{3}, 0b0, opc{2-1}, ?, opc{0} },
asm, ZPR64, ZPR32, ZPR4b32, VectorIndexS> {
asm, ZPR64, ZPR32, ZPR4b32, VectorIndexS32b> {
bits<4> Zm;
bits<2> iop;
let Inst{20} = iop{1};
@ -2430,8 +2430,8 @@ multiclass sve2_int_mla_long_by_indexed_elem<bits<4> opc, string asm, SDPatternO
let Inst{11} = iop{0};
}
def : SVE_4_Op_Imm_Pat<nxv4i32, op, nxv4i32, nxv8i16, nxv8i16, i64, VectorIndexH_timm, !cast<Instruction>(NAME # _S)>;
def : SVE_4_Op_Imm_Pat<nxv2i64, op, nxv2i64, nxv4i32, nxv4i32, i64, VectorIndexS_timm, !cast<Instruction>(NAME # _D)>;
def : SVE_4_Op_Imm_Pat<nxv4i32, op, nxv4i32, nxv8i16, nxv8i16, i32, VectorIndexH32b_timm, !cast<Instruction>(NAME # _S)>;
def : SVE_4_Op_Imm_Pat<nxv2i64, op, nxv2i64, nxv4i32, nxv4i32, i32, VectorIndexS32b_timm, !cast<Instruction>(NAME # _D)>;
}
//===----------------------------------------------------------------------===//

View File

@ -9,10 +9,10 @@ define <vscale x 4 x i32> @smlalb_i32(<vscale x 4 x i32> %a,
; CHECK-LABEL: smlalb_i32
; CHECK: smlalb z0.s, z1.h, z2.h[1]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlalb.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 1)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlalb.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 1)
ret <vscale x 4 x i32> %res
}
@ -22,10 +22,10 @@ define <vscale x 4 x i32> @smlalb_i32_2(<vscale x 4 x i32> %a,
; CHECK-LABEL: smlalb_i32_2
; CHECK: smlalb z0.s, z1.h, z2.h[7]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlalb.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 7)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlalb.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 7)
ret <vscale x 4 x i32> %res
}
@ -35,10 +35,10 @@ define <vscale x 2 x i64> @smlalb_i64(<vscale x 2 x i64> %a,
; CHECK-LABEL: smlalb_i64
; CHECK: smlalb z0.d, z1.s, z2.s[0]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlalb.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 0)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlalb.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 0)
ret <vscale x 2 x i64> %res
}
@ -48,10 +48,10 @@ define <vscale x 2 x i64> @smlalb_i64_2(<vscale x 2 x i64> %a,
; CHECK-LABEL: smlalb_i64_2
; CHECK: smlalb z0.d, z1.s, z2.s[3]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlalb.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 3)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlalb.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 3)
ret <vscale x 2 x i64> %res
}
@ -64,10 +64,10 @@ define <vscale x 4 x i32> @smlalt_i32(<vscale x 4 x i32> %a,
; CHECK-LABEL: smlalt_i32
; CHECK: smlalt z0.s, z1.h, z2.h[1]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlalt.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 1)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlalt.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 1)
ret <vscale x 4 x i32> %res
}
@ -77,10 +77,10 @@ define <vscale x 4 x i32> @smlalt_i32_2(<vscale x 4 x i32> %a,
; CHECK-LABEL: smlalt_i32_2
; CHECK: smlalt z0.s, z1.h, z2.h[7]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlalt.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 7)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlalt.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 7)
ret <vscale x 4 x i32> %res
}
@ -90,10 +90,10 @@ define <vscale x 2 x i64> @smlalt_i64(<vscale x 2 x i64> %a,
; CHECK-LABEL: smlalt_i64
; CHECK: smlalt z0.d, z1.s, z2.s[0]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlalt.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 0)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlalt.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 0)
ret <vscale x 2 x i64> %res
}
@ -103,10 +103,10 @@ define <vscale x 2 x i64> @smlalt_i64_2(<vscale x 2 x i64> %a,
; CHECK-LABEL: smlalt_i64_2
; CHECK: smlalt z0.d, z1.s, z2.s[3]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlalt.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 3)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlalt.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 3)
ret <vscale x 2 x i64> %res
}
@ -119,10 +119,10 @@ define <vscale x 4 x i32> @umlalb_i32(<vscale x 4 x i32> %a,
; CHECK-LABEL: umlalb_i32
; CHECK: umlalb z0.s, z1.h, z2.h[1]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlalb.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 1)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlalb.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 1)
ret <vscale x 4 x i32> %res
}
@ -132,10 +132,10 @@ define <vscale x 4 x i32> @umlalb_i32_2(<vscale x 4 x i32> %a,
; CHECK-LABEL: umlalb_i32_2
; CHECK: umlalb z0.s, z1.h, z2.h[7]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlalb.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 7)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlalb.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 7)
ret <vscale x 4 x i32> %res
}
@ -145,10 +145,10 @@ define <vscale x 2 x i64> @umlalb_i64(<vscale x 2 x i64> %a,
; CHECK-LABEL: umlalb_i64
; CHECK: umlalb z0.d, z1.s, z2.s[0]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlalb.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 0)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlalb.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 0)
ret <vscale x 2 x i64> %res
}
@ -158,10 +158,10 @@ define <vscale x 2 x i64> @umlalb_i64_2(<vscale x 2 x i64> %a,
; CHECK-LABEL: umlalb_i64_2
; CHECK: umlalb z0.d, z1.s, z2.s[3]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlalb.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 3)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlalb.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 3)
ret <vscale x 2 x i64> %res
}
@ -174,10 +174,10 @@ define <vscale x 4 x i32> @umlalt_i32(<vscale x 4 x i32> %a,
; CHECK-LABEL: umlalt_i32
; CHECK: umlalt z0.s, z1.h, z2.h[1]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlalt.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 1)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlalt.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 1)
ret <vscale x 4 x i32> %res
}
@ -187,10 +187,10 @@ define <vscale x 4 x i32> @umlalt_i32_2(<vscale x 4 x i32> %a,
; CHECK-LABEL: umlalt_i32_2
; CHECK: umlalt z0.s, z1.h, z2.h[7]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlalt.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 7)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlalt.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 7)
ret <vscale x 4 x i32> %res
}
@ -200,10 +200,10 @@ define <vscale x 2 x i64> @umlalt_i64(<vscale x 2 x i64> %a,
; CHECK-LABEL: umlalt_i64
; CHECK: umlalt z0.d, z1.s, z2.s[0]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlalt.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 0)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlalt.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 0)
ret <vscale x 2 x i64> %res
}
@ -213,10 +213,10 @@ define <vscale x 2 x i64> @umlalt_i64_2(<vscale x 2 x i64> %a,
; CHECK-LABEL: umlalt_i64_2
; CHECK: umlalt z0.d, z1.s, z2.s[3]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlalt.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 3)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlalt.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 3)
ret <vscale x 2 x i64> %res
}
@ -229,10 +229,10 @@ define <vscale x 4 x i32> @smlslb_i32(<vscale x 4 x i32> %a,
; CHECK-LABEL: smlslb_i32
; CHECK: smlslb z0.s, z1.h, z2.h[1]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlslb.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 1)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlslb.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 1)
ret <vscale x 4 x i32> %res
}
@ -242,10 +242,10 @@ define <vscale x 4 x i32> @smlslb_i32_2(<vscale x 4 x i32> %a,
; CHECK-LABEL: smlslb_i32_2
; CHECK: smlslb z0.s, z1.h, z2.h[7]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlslb.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 7)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlslb.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 7)
ret <vscale x 4 x i32> %res
}
@ -255,10 +255,10 @@ define <vscale x 2 x i64> @smlslb_i64(<vscale x 2 x i64> %a,
; CHECK-LABEL: smlslb_i64
; CHECK: smlslb z0.d, z1.s, z2.s[0]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlslb.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 0)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlslb.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 0)
ret <vscale x 2 x i64> %res
}
@ -268,10 +268,10 @@ define <vscale x 2 x i64> @smlslb_i64_2(<vscale x 2 x i64> %a,
; CHECK-LABEL: smlslb_i64_2
; CHECK: smlslb z0.d, z1.s, z2.s[3]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlslb.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 3)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlslb.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 3)
ret <vscale x 2 x i64> %res
}
@ -284,10 +284,10 @@ define <vscale x 4 x i32> @smlslt_i32(<vscale x 4 x i32> %a,
; CHECK-LABEL: smlslt_i32
; CHECK: smlslt z0.s, z1.h, z2.h[1]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlslt.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 1)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlslt.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 1)
ret <vscale x 4 x i32> %res
}
@ -297,10 +297,10 @@ define <vscale x 4 x i32> @smlslt_i32_2(<vscale x 4 x i32> %a,
; CHECK-LABEL: smlslt_i32_2
; CHECK: smlslt z0.s, z1.h, z2.h[7]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlslt.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 7)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.smlslt.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 7)
ret <vscale x 4 x i32> %res
}
@ -310,10 +310,10 @@ define <vscale x 2 x i64> @smlslt_i64(<vscale x 2 x i64> %a,
; CHECK-LABEL: smlslt_i64
; CHECK: smlslt z0.d, z1.s, z2.s[0]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlslt.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 0)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlslt.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 0)
ret <vscale x 2 x i64> %res
}
@ -323,10 +323,10 @@ define <vscale x 2 x i64> @smlslt_i64_2(<vscale x 2 x i64> %a,
; CHECK-LABEL: smlslt_i64_2
; CHECK: smlslt z0.d, z1.s, z2.s[3]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlslt.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 3)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.smlslt.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 3)
ret <vscale x 2 x i64> %res
}
@ -339,10 +339,10 @@ define <vscale x 4 x i32> @umlslb_i32(<vscale x 4 x i32> %a,
; CHECK-LABEL: umlslb_i32
; CHECK: umlslb z0.s, z1.h, z2.h[1]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlslb.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 1)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlslb.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 1)
ret <vscale x 4 x i32> %res
}
@ -352,10 +352,10 @@ define <vscale x 4 x i32> @umlslb_i32_2(<vscale x 4 x i32> %a,
; CHECK-LABEL: umlslb_i32_2
; CHECK: umlslb z0.s, z1.h, z2.h[7]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlslb.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 7)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlslb.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 7)
ret <vscale x 4 x i32> %res
}
@ -365,10 +365,10 @@ define <vscale x 2 x i64> @umlslb_i64(<vscale x 2 x i64> %a,
; CHECK-LABEL: umlslb_i64
; CHECK: umlslb z0.d, z1.s, z2.s[0]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlslb.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 0)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlslb.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 0)
ret <vscale x 2 x i64> %res
}
@ -378,10 +378,10 @@ define <vscale x 2 x i64> @umlslb_i64_2(<vscale x 2 x i64> %a,
; CHECK-LABEL: umlslb_i64_2
; CHECK: umlslb z0.d, z1.s, z2.s[3]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlslb.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 3)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlslb.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 3)
ret <vscale x 2 x i64> %res
}
@ -394,10 +394,10 @@ define <vscale x 4 x i32> @umlslt_i32(<vscale x 4 x i32> %a,
; CHECK-LABEL: umlslt_i32
; CHECK: umlslt z0.s, z1.h, z2.h[1]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlslt.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 1)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlslt.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 1)
ret <vscale x 4 x i32> %res
}
@ -407,10 +407,10 @@ define <vscale x 4 x i32> @umlslt_i32_2(<vscale x 4 x i32> %a,
; CHECK-LABEL: umlslt_i32_2
; CHECK: umlslt z0.s, z1.h, z2.h[7]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlslt.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i64 7)
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.umlslt.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 7)
ret <vscale x 4 x i32> %res
}
@ -420,10 +420,10 @@ define <vscale x 2 x i64> @umlslt_i64(<vscale x 2 x i64> %a,
; CHECK-LABEL: umlslt_i64
; CHECK: umlslt z0.d, z1.s, z2.s[0]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlslt.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 0)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlslt.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 0)
ret <vscale x 2 x i64> %res
}
@ -433,26 +433,254 @@ define <vscale x 2 x i64> @umlslt_i64_2(<vscale x 2 x i64> %a,
; CHECK-LABEL: umlslt_i64_2
; CHECK: umlslt z0.d, z1.s, z2.s[3]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlslt.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i64 3)
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.umlslt.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 3)
ret <vscale x 2 x i64> %res
}
declare <vscale x 4 x i32> @llvm.aarch64.sve.smlalb.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i64)
declare <vscale x 2 x i64> @llvm.aarch64.sve.smlalb.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i64)
declare <vscale x 4 x i32> @llvm.aarch64.sve.smlalt.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i64)
declare <vscale x 2 x i64> @llvm.aarch64.sve.smlalt.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i64)
declare <vscale x 4 x i32> @llvm.aarch64.sve.umlalb.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i64)
declare <vscale x 2 x i64> @llvm.aarch64.sve.umlalb.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i64)
declare <vscale x 4 x i32> @llvm.aarch64.sve.umlalt.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i64)
declare <vscale x 2 x i64> @llvm.aarch64.sve.umlalt.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i64)
declare <vscale x 4 x i32> @llvm.aarch64.sve.smlslb.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i64)
declare <vscale x 2 x i64> @llvm.aarch64.sve.smlslb.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i64)
declare <vscale x 4 x i32> @llvm.aarch64.sve.smlslt.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i64)
declare <vscale x 2 x i64> @llvm.aarch64.sve.smlslt.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i64)
declare <vscale x 4 x i32> @llvm.aarch64.sve.umlslb.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i64)
declare <vscale x 2 x i64> @llvm.aarch64.sve.umlslb.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i64)
declare <vscale x 4 x i32> @llvm.aarch64.sve.umlslt.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i64)
declare <vscale x 2 x i64> @llvm.aarch64.sve.umlslt.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i64)
;
; SQDMLALB
;
define <vscale x 4 x i32> @sqdmlalb_i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c) {
; CHECK-LABEL: sqdmlalb_i32
; CHECK: sqdmlalb z0.s, z1.h, z2.h[1]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.sqdmlalb.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 1)
ret <vscale x 4 x i32> %res
}
define <vscale x 4 x i32> @sqdmlalb_i32_2(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c) {
; CHECK-LABEL: sqdmlalb_i32_2
; CHECK: sqdmlalb z0.s, z1.h, z2.h[7]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.sqdmlalb.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 7)
ret <vscale x 4 x i32> %res
}
define <vscale x 2 x i64> @sqdmlalb_i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c) {
; CHECK-LABEL: sqdmlalb_i64
; CHECK: sqdmlalb z0.d, z1.s, z2.s[0]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.sqdmlalb.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 0)
ret <vscale x 2 x i64> %res
}
define <vscale x 2 x i64> @sqdmlalb_i64_2(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c) {
; CHECK-LABEL: sqdmlalb_i64_2
; CHECK: sqdmlalb z0.d, z1.s, z2.s[3]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.sqdmlalb.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 3)
ret <vscale x 2 x i64> %res
}
;
; SQDMLALT
;
define <vscale x 4 x i32> @sqdmlalt_i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c) {
; CHECK-LABEL: sqdmlalt_i32
; CHECK: sqdmlalt z0.s, z1.h, z2.h[1]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.sqdmlalt.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 1)
ret <vscale x 4 x i32> %res
}
define <vscale x 4 x i32> @sqdmlalt_i32_2(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c) {
; CHECK-LABEL: sqdmlalt_i32_2
; CHECK: sqdmlalt z0.s, z1.h, z2.h[7]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.sqdmlalt.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 7)
ret <vscale x 4 x i32> %res
}
define <vscale x 2 x i64> @sqdmlalt_i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c) {
; CHECK-LABEL: sqdmlalt_i64
; CHECK: sqdmlalt z0.d, z1.s, z2.s[0]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.sqdmlalt.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 0)
ret <vscale x 2 x i64> %res
}
define <vscale x 2 x i64> @sqdmlalt_i64_2(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c) {
; CHECK-LABEL: sqdmlalt_i64_2
; CHECK: sqdmlalt z0.d, z1.s, z2.s[3]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.sqdmlalt.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 3)
ret <vscale x 2 x i64> %res
}
;
; SQDMLSLB
;
define <vscale x 4 x i32> @sqdmlslb_i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c) {
; CHECK-LABEL: sqdmlslb_i32
; CHECK: sqdmlslb z0.s, z1.h, z2.h[1]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.sqdmlslb.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 1)
ret <vscale x 4 x i32> %res
}
define <vscale x 4 x i32> @sqdmlslb_i32_2(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c) {
; CHECK-LABEL: sqdmlslb_i32_2
; CHECK: sqdmlslb z0.s, z1.h, z2.h[7]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.sqdmlslb.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 7)
ret <vscale x 4 x i32> %res
}
define <vscale x 2 x i64> @sqdmlslb_i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c) {
; CHECK-LABEL: sqdmlslb_i64
; CHECK: sqdmlslb z0.d, z1.s, z2.s[0]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.sqdmlslb.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 0)
ret <vscale x 2 x i64> %res
}
define <vscale x 2 x i64> @sqdmlslb_i64_2(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c) {
; CHECK-LABEL: sqdmlslb_i64_2
; CHECK: sqdmlslb z0.d, z1.s, z2.s[3]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.sqdmlslb.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 3)
ret <vscale x 2 x i64> %res
}
;
; SQDMLSLT
;
define <vscale x 4 x i32> @sqdmlslt_i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c) {
; CHECK-LABEL: sqdmlslt_i32
; CHECK: sqdmlslt z0.s, z1.h, z2.h[1]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.sqdmlslt.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 1)
ret <vscale x 4 x i32> %res
}
define <vscale x 4 x i32> @sqdmlslt_i32_2(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c) {
; CHECK-LABEL: sqdmlslt_i32_2
; CHECK: sqdmlslt z0.s, z1.h, z2.h[7]
; CHECK-NEXT: ret
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.sqdmlslt.lane.nxv4i32(<vscale x 4 x i32> %a,
<vscale x 8 x i16> %b,
<vscale x 8 x i16> %c,
i32 7)
ret <vscale x 4 x i32> %res
}
define <vscale x 2 x i64> @sqdmlslt_i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c) {
; CHECK-LABEL: sqdmlslt_i64
; CHECK: sqdmlslt z0.d, z1.s, z2.s[0]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.sqdmlslt.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 0)
ret <vscale x 2 x i64> %res
}
define <vscale x 2 x i64> @sqdmlslt_i64_2(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c) {
; CHECK-LABEL: sqdmlslt_i64_2
; CHECK: sqdmlslt z0.d, z1.s, z2.s[3]
; CHECK-NEXT: ret
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.sqdmlslt.lane.nxv2i64(<vscale x 2 x i64> %a,
<vscale x 4 x i32> %b,
<vscale x 4 x i32> %c,
i32 3)
ret <vscale x 2 x i64> %res
}
declare <vscale x 4 x i32> @llvm.aarch64.sve.smlalb.lane.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i32)
declare <vscale x 2 x i64> @llvm.aarch64.sve.smlalb.lane.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i32)
declare <vscale x 4 x i32> @llvm.aarch64.sve.smlalt.lane.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i32)
declare <vscale x 2 x i64> @llvm.aarch64.sve.smlalt.lane.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i32)
declare <vscale x 4 x i32> @llvm.aarch64.sve.umlalb.lane.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i32)
declare <vscale x 2 x i64> @llvm.aarch64.sve.umlalb.lane.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i32)
declare <vscale x 4 x i32> @llvm.aarch64.sve.umlalt.lane.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i32)
declare <vscale x 2 x i64> @llvm.aarch64.sve.umlalt.lane.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i32)
declare <vscale x 4 x i32> @llvm.aarch64.sve.smlslb.lane.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i32)
declare <vscale x 2 x i64> @llvm.aarch64.sve.smlslb.lane.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i32)
declare <vscale x 4 x i32> @llvm.aarch64.sve.smlslt.lane.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i32)
declare <vscale x 2 x i64> @llvm.aarch64.sve.smlslt.lane.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i32)
declare <vscale x 4 x i32> @llvm.aarch64.sve.umlslb.lane.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i32)
declare <vscale x 2 x i64> @llvm.aarch64.sve.umlslb.lane.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i32)
declare <vscale x 4 x i32> @llvm.aarch64.sve.umlslt.lane.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i32)
declare <vscale x 2 x i64> @llvm.aarch64.sve.umlslt.lane.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i32)
declare <vscale x 4 x i32> @llvm.aarch64.sve.sqdmlalb.lane.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i32)
declare <vscale x 2 x i64> @llvm.aarch64.sve.sqdmlalb.lane.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i32)
declare <vscale x 4 x i32> @llvm.aarch64.sve.sqdmlalt.lane.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i32)
declare <vscale x 2 x i64> @llvm.aarch64.sve.sqdmlalt.lane.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i32)
declare <vscale x 4 x i32> @llvm.aarch64.sve.sqdmlslb.lane.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i32)
declare <vscale x 2 x i64> @llvm.aarch64.sve.sqdmlslb.lane.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i32)
declare <vscale x 4 x i32> @llvm.aarch64.sve.sqdmlslt.lane.nxv4i32(<vscale x 4 x i32>, <vscale x 8 x i16>, <vscale x 8 x i16>, i32)
declare <vscale x 2 x i64> @llvm.aarch64.sve.sqdmlslt.lane.nxv2i64(<vscale x 2 x i64>, <vscale x 4 x i32>, <vscale x 4 x i32>, i32)