From d9c49c4ce1fae15a1f787c8ac8420c2845240197 Mon Sep 17 00:00:00 2001 From: lioncash Date: Thu, 29 Dec 2022 16:11:01 +0000 Subject: [PATCH 1/4] OpcodeDispatcher: Handle VPUNPCKLDQ --- .../Interface/Core/OpcodeDispatcher.cpp | 1 + .../Interface/Core/X86Tables/VEXTables.cpp | 2 +- unittests/ASM/VEX/vpunpckldq.asm | 39 +++++++++++++++++++ 3 files changed, 41 insertions(+), 1 deletion(-) create mode 100644 unittests/ASM/VEX/vpunpckldq.asm diff --git a/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp b/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp index 5b440dab8..9a84d3a3b 100644 --- a/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp +++ b/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp @@ -5904,6 +5904,7 @@ void OpDispatchBuilder::InstallHostSpecificOpcodeHandlers() { {OPD(1, 0b10, 0x5F), 1, &OpDispatchBuilder::AVXVectorScalarALUOp}, {OPD(1, 0b11, 0x5F), 1, &OpDispatchBuilder::AVXVectorScalarALUOp}, + {OPD(1, 0b01, 0x62), 1, &OpDispatchBuilder::VPUNPCKLOp<4>}, {OPD(1, 0b01, 0x63), 1, &OpDispatchBuilder::VPACKSSOp<2>}, {OPD(1, 0b01, 0x64), 1, &OpDispatchBuilder::AVXVectorALUOp}, {OPD(1, 0b01, 0x65), 1, &OpDispatchBuilder::AVXVectorALUOp}, diff --git a/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp b/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp index 40896e983..8512c5dcd 100644 --- a/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp +++ b/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp @@ -76,7 +76,7 @@ void InitializeVEXTables() { {OPD(1, 0b01, 0x60), 1, X86InstInfo{"VPUNPCKLBW", TYPE_UNDEC, FLAGS_NONE, 0, nullptr}}, {OPD(1, 0b01, 0x61), 1, X86InstInfo{"VPUNPCKLWD", TYPE_UNDEC, FLAGS_NONE, 0, nullptr}}, - {OPD(1, 0b01, 0x62), 1, X86InstInfo{"VPUNPCKLDQ", TYPE_UNDEC, FLAGS_NONE, 0, nullptr}}, + {OPD(1, 0b01, 0x62), 1, X86InstInfo{"VPUNPCKLDQ", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, {OPD(1, 0b01, 0x63), 1, X86InstInfo{"VPACKSSWB", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, {OPD(1, 0b01, 0x64), 1, X86InstInfo{"VPCMPGTB", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, {OPD(1, 0b01, 0x65), 1, X86InstInfo{"VPCMPGTW", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, diff --git a/unittests/ASM/VEX/vpunpckldq.asm b/unittests/ASM/VEX/vpunpckldq.asm new file mode 100644 index 000000000..a138a20ea --- /dev/null +++ b/unittests/ASM/VEX/vpunpckldq.asm @@ -0,0 +1,39 @@ +%ifdef CONFIG +{ + "HostFeatures": ["AVX"], + "RegData": { + "XMM2": ["0x6566676845464748", "0x6162636441424344", "0x0000000000000000", "0x0000000000000000"], + "XMM3": ["0x6566676845464748", "0x6162636441424344", "0x0000000000000000", "0x0000000000000000"], + "XMM4": ["0x6566676845464748", "0x6162636441424344", "0xAAAAAAAAEEEEEEEE", "0xBBBBBBBBFFFFFFFF"], + "XMM5": ["0x6566676845464748", "0x6162636441424344", "0xAAAAAAAAEEEEEEEE", "0xBBBBBBBBFFFFFFFF"] + }, + "MemoryRegions": { + "0x100000000": "4096" + } +} +%endif + +lea rdx, [rel .data] + +vmovapd ymm0, [rdx] +vmovapd ymm1, [rdx + 32] + +vpunpckldq xmm2, xmm0, [rdx + 32] +vpunpckldq xmm3, xmm0, xmm1 + +vpunpckldq ymm4, ymm0, [rdx + 32] +vpunpckldq ymm5, ymm0, ymm1 + +hlt + +align 32 +.data: +dq 0x4142434445464748 +dq 0x5152535455565758 +dq 0xFFFFFFFFEEEEEEEE +dq 0xDDDDDDDDCCCCCCCC + +dq 0x6162636465666768 +dq 0x7172737475767778 +dq 0xBBBBBBBBAAAAAAAA +dq 0x9999999988888888 From aaf8e3757d2591beadc5018c6fc1226a34162acd Mon Sep 17 00:00:00 2001 From: lioncash Date: Thu, 29 Dec 2022 16:19:26 +0000 Subject: [PATCH 2/4] OpcodeDispatcher: Handle VPUNPCKLQDQ --- .../Interface/Core/OpcodeDispatcher.cpp | 1 + .../Interface/Core/X86Tables/VEXTables.cpp | 2 +- unittests/ASM/VEX/vpunpcklqdq.asm | 39 +++++++++++++++++++ 3 files changed, 41 insertions(+), 1 deletion(-) create mode 100644 unittests/ASM/VEX/vpunpcklqdq.asm diff --git a/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp b/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp index 9a84d3a3b..509994e3f 100644 --- a/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp +++ b/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp @@ -5911,6 +5911,7 @@ void OpDispatchBuilder::InstallHostSpecificOpcodeHandlers() { {OPD(1, 0b01, 0x66), 1, &OpDispatchBuilder::AVXVectorALUOp}, {OPD(1, 0b01, 0x67), 1, &OpDispatchBuilder::VPACKUSOp<2>}, {OPD(1, 0b01, 0x6B), 1, &OpDispatchBuilder::VPACKSSOp<4>}, + {OPD(1, 0b01, 0x6C), 1, &OpDispatchBuilder::VPUNPCKLOp<8>}, {OPD(1, 0b01, 0x6E), 1, &OpDispatchBuilder::MOVBetweenGPR_FPR}, {OPD(1, 0b01, 0x6F), 1, &OpDispatchBuilder::VMOVAPS_VMOVAPD_Op}, diff --git a/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp b/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp index 8512c5dcd..4cceec7d6 100644 --- a/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp +++ b/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp @@ -175,7 +175,7 @@ void InitializeVEXTables() { {OPD(1, 0b01, 0x69), 1, X86InstInfo{"VPUNPCKHWD", TYPE_UNDEC, FLAGS_NONE, 0, nullptr}}, {OPD(1, 0b01, 0x6A), 1, X86InstInfo{"VPUNPCKHDQ", TYPE_UNDEC, FLAGS_NONE, 0, nullptr}}, {OPD(1, 0b01, 0x6B), 1, X86InstInfo{"VPACKSSDW", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, - {OPD(1, 0b01, 0x6C), 1, X86InstInfo{"VPUNPCKLQDQ", TYPE_UNDEC, FLAGS_NONE, 0, nullptr}}, + {OPD(1, 0b01, 0x6C), 1, X86InstInfo{"VPUNPCKLQDQ", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, {OPD(1, 0b01, 0x6D), 1, X86InstInfo{"VPUNPCKHQDQ", TYPE_UNDEC, FLAGS_NONE, 0, nullptr}}, {OPD(1, 0b01, 0x6E), 1, X86InstInfo{"VMOV*", TYPE_INST, GenFlagsDstSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_XMM_FLAGS | FLAGS_SF_SRC_GPR, 0, nullptr}}, diff --git a/unittests/ASM/VEX/vpunpcklqdq.asm b/unittests/ASM/VEX/vpunpcklqdq.asm new file mode 100644 index 000000000..3f607cd64 --- /dev/null +++ b/unittests/ASM/VEX/vpunpcklqdq.asm @@ -0,0 +1,39 @@ +%ifdef CONFIG +{ + "HostFeatures": ["AVX"], + "RegData": { + "XMM2": ["0x4142434445464748", "0x6162636465666768", "0x0000000000000000", "0x0000000000000000"], + "XMM3": ["0x4142434445464748", "0x6162636465666768", "0x0000000000000000", "0x0000000000000000"], + "XMM4": ["0x4142434445464748", "0x6162636465666768", "0xFFFFFFFFFFFFFFFF", "0xBBBBBBBBBBBBBBBB"], + "XMM5": ["0x4142434445464748", "0x6162636465666768", "0xFFFFFFFFFFFFFFFF", "0xBBBBBBBBBBBBBBBB"] + }, + "MemoryRegions": { + "0x100000000": "4096" + } +} +%endif + +lea rdx, [rel .data] + +vmovapd ymm0, [rdx] +vmovapd ymm1, [rdx + 32] + +vpunpcklqdq xmm2, xmm0, [rdx + 32] +vpunpcklqdq xmm3, xmm0, xmm1 + +vpunpcklqdq ymm4, ymm0, [rdx + 32] +vpunpcklqdq ymm5, ymm0, ymm1 + +hlt + +align 32 +.data: +dq 0x4142434445464748 +dq 0x5152535455565758 +dq 0xFFFFFFFFFFFFFFFF +dq 0xEEEEEEEEEEEEEEEE + +dq 0x6162636465666768 +dq 0x7172737475767778 +dq 0xBBBBBBBBBBBBBBBB +dq 0xCCCCCCCCCCCCCCCC From 7b30a241c05a3139b1e9c1a155af6b1622005209 Mon Sep 17 00:00:00 2001 From: lioncash Date: Thu, 29 Dec 2022 16:24:33 +0000 Subject: [PATCH 3/4] OpcodeDispatcher: Handle VPUNPCKLWD --- .../Interface/Core/OpcodeDispatcher.cpp | 1 + .../Core/OpcodeDispatcher/Vector.cpp | 2 + .../Interface/Core/X86Tables/VEXTables.cpp | 2 +- unittests/ASM/VEX/vpunpcklwd.asm | 39 +++++++++++++++++++ 4 files changed, 43 insertions(+), 1 deletion(-) create mode 100644 unittests/ASM/VEX/vpunpcklwd.asm diff --git a/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp b/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp index 509994e3f..6986d445b 100644 --- a/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp +++ b/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp @@ -5904,6 +5904,7 @@ void OpDispatchBuilder::InstallHostSpecificOpcodeHandlers() { {OPD(1, 0b10, 0x5F), 1, &OpDispatchBuilder::AVXVectorScalarALUOp}, {OPD(1, 0b11, 0x5F), 1, &OpDispatchBuilder::AVXVectorScalarALUOp}, + {OPD(1, 0b01, 0x61), 1, &OpDispatchBuilder::VPUNPCKLOp<2>}, {OPD(1, 0b01, 0x62), 1, &OpDispatchBuilder::VPUNPCKLOp<4>}, {OPD(1, 0b01, 0x63), 1, &OpDispatchBuilder::VPACKSSOp<2>}, {OPD(1, 0b01, 0x64), 1, &OpDispatchBuilder::AVXVectorALUOp}, diff --git a/External/FEXCore/Source/Interface/Core/OpcodeDispatcher/Vector.cpp b/External/FEXCore/Source/Interface/Core/OpcodeDispatcher/Vector.cpp index 8933e1149..d13418da6 100644 --- a/External/FEXCore/Source/Interface/Core/OpcodeDispatcher/Vector.cpp +++ b/External/FEXCore/Source/Interface/Core/OpcodeDispatcher/Vector.cpp @@ -905,6 +905,8 @@ void OpDispatchBuilder::VPUNPCKLOp(OpcodeArgs) { StoreResult(FPRClass, Op, Result, -1); } +template +void OpDispatchBuilder::VPUNPCKLOp<2>(OpcodeArgs); template void OpDispatchBuilder::VPUNPCKLOp<4>(OpcodeArgs); template diff --git a/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp b/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp index 4cceec7d6..12d78e7c2 100644 --- a/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp +++ b/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp @@ -75,7 +75,7 @@ void InitializeVEXTables() { {OPD(1, 0b01, 0x57), 1, X86InstInfo{"VXORPD", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, {OPD(1, 0b01, 0x60), 1, X86InstInfo{"VPUNPCKLBW", TYPE_UNDEC, FLAGS_NONE, 0, nullptr}}, - {OPD(1, 0b01, 0x61), 1, X86InstInfo{"VPUNPCKLWD", TYPE_UNDEC, FLAGS_NONE, 0, nullptr}}, + {OPD(1, 0b01, 0x61), 1, X86InstInfo{"VPUNPCKLWD", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, {OPD(1, 0b01, 0x62), 1, X86InstInfo{"VPUNPCKLDQ", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, {OPD(1, 0b01, 0x63), 1, X86InstInfo{"VPACKSSWB", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, {OPD(1, 0b01, 0x64), 1, X86InstInfo{"VPCMPGTB", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, diff --git a/unittests/ASM/VEX/vpunpcklwd.asm b/unittests/ASM/VEX/vpunpcklwd.asm new file mode 100644 index 000000000..22bd6329b --- /dev/null +++ b/unittests/ASM/VEX/vpunpcklwd.asm @@ -0,0 +1,39 @@ +%ifdef CONFIG +{ + "HostFeatures": ["AVX"], + "RegData": { + "XMM2": ["0x6566454667684748", "0x6162414263644344", "0x0000000000000000", "0x0000000000000000"], + "XMM3": ["0x6566454667684748", "0x6162414263644344", "0x0000000000000000", "0x0000000000000000"], + "XMM4": ["0x6566454667684748", "0x6162414263644344", "0x5555DDDD4444CCCC", "0x7777FFFF6666EEEE"], + "XMM5": ["0x6566454667684748", "0x6162414263644344", "0x5555DDDD4444CCCC", "0x7777FFFF6666EEEE"] + }, + "MemoryRegions": { + "0x100000000": "4096" + } +} +%endif + +lea rdx, [rel .data] + +vmovapd ymm0, [rdx] +vmovapd ymm1, [rdx + 32] + +vpunpcklwd xmm2, xmm0, [rdx + 32] +vpunpcklwd xmm3, xmm0, xmm1 + +vpunpcklwd ymm4, ymm0, [rdx + 32] +vpunpcklwd ymm5, ymm0, ymm1 + +hlt + +align 32 +.data: +dq 0x4142434445464748 +dq 0x5152535455565758 +dq 0xFFFFEEEEDDDDCCCC +dq 0xBBBBAAAA99998888 + +dq 0x6162636465666768 +dq 0x7172737475767778 +dq 0x7777666655554444 +dq 0x3333222211110000 From d91f2ed6b07bd319d1bb2fb9aa959eb78638dce4 Mon Sep 17 00:00:00 2001 From: lioncash Date: Thu, 29 Dec 2022 16:31:57 +0000 Subject: [PATCH 4/4] OpcodeDispatcher: Handle VPUNPCKLBW --- .../Interface/Core/OpcodeDispatcher.cpp | 1 + .../Core/OpcodeDispatcher/Vector.cpp | 2 + .../Interface/Core/X86Tables/VEXTables.cpp | 2 +- unittests/ASM/VEX/vpunpcklbw.asm | 39 +++++++++++++++++++ 4 files changed, 43 insertions(+), 1 deletion(-) create mode 100644 unittests/ASM/VEX/vpunpcklbw.asm diff --git a/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp b/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp index 6986d445b..e82faeb75 100644 --- a/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp +++ b/External/FEXCore/Source/Interface/Core/OpcodeDispatcher.cpp @@ -5904,6 +5904,7 @@ void OpDispatchBuilder::InstallHostSpecificOpcodeHandlers() { {OPD(1, 0b10, 0x5F), 1, &OpDispatchBuilder::AVXVectorScalarALUOp}, {OPD(1, 0b11, 0x5F), 1, &OpDispatchBuilder::AVXVectorScalarALUOp}, + {OPD(1, 0b01, 0x60), 1, &OpDispatchBuilder::VPUNPCKLOp<1>}, {OPD(1, 0b01, 0x61), 1, &OpDispatchBuilder::VPUNPCKLOp<2>}, {OPD(1, 0b01, 0x62), 1, &OpDispatchBuilder::VPUNPCKLOp<4>}, {OPD(1, 0b01, 0x63), 1, &OpDispatchBuilder::VPACKSSOp<2>}, diff --git a/External/FEXCore/Source/Interface/Core/OpcodeDispatcher/Vector.cpp b/External/FEXCore/Source/Interface/Core/OpcodeDispatcher/Vector.cpp index d13418da6..ba426bfa8 100644 --- a/External/FEXCore/Source/Interface/Core/OpcodeDispatcher/Vector.cpp +++ b/External/FEXCore/Source/Interface/Core/OpcodeDispatcher/Vector.cpp @@ -905,6 +905,8 @@ void OpDispatchBuilder::VPUNPCKLOp(OpcodeArgs) { StoreResult(FPRClass, Op, Result, -1); } +template +void OpDispatchBuilder::VPUNPCKLOp<1>(OpcodeArgs); template void OpDispatchBuilder::VPUNPCKLOp<2>(OpcodeArgs); template diff --git a/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp b/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp index 12d78e7c2..142cdabb8 100644 --- a/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp +++ b/External/FEXCore/Source/Interface/Core/X86Tables/VEXTables.cpp @@ -74,7 +74,7 @@ void InitializeVEXTables() { {OPD(1, 0b00, 0x57), 1, X86InstInfo{"VXORPS", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, {OPD(1, 0b01, 0x57), 1, X86InstInfo{"VXORPD", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, - {OPD(1, 0b01, 0x60), 1, X86InstInfo{"VPUNPCKLBW", TYPE_UNDEC, FLAGS_NONE, 0, nullptr}}, + {OPD(1, 0b01, 0x60), 1, X86InstInfo{"VPUNPCKLBW", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, {OPD(1, 0b01, 0x61), 1, X86InstInfo{"VPUNPCKLWD", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, {OPD(1, 0b01, 0x62), 1, X86InstInfo{"VPUNPCKLDQ", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, {OPD(1, 0b01, 0x63), 1, X86InstInfo{"VPACKSSWB", TYPE_INST, GenFlagsSameSize(SIZE_128BIT) | FLAGS_MODRM | FLAGS_VEX_1ST_SRC | FLAGS_XMM_FLAGS, 0, nullptr}}, diff --git a/unittests/ASM/VEX/vpunpcklbw.asm b/unittests/ASM/VEX/vpunpcklbw.asm new file mode 100644 index 000000000..3acfab363 --- /dev/null +++ b/unittests/ASM/VEX/vpunpcklbw.asm @@ -0,0 +1,39 @@ +%ifdef CONFIG +{ + "HostFeatures": ["AVX"], + "RegData": { + "XMM2": ["0x6545664667476848", "0x6141624263436444", "0x0000000000000000", "0x0000000000000000"], + "XMM3": ["0x6545664667476848", "0x6141624263436444", "0x0000000000000000", "0x0000000000000000"], + "XMM4": ["0x6545664667476848", "0x6141624263436444", "0x55DD55DD44CC44CC", "0x77FF77FF66EE66EE"], + "XMM5": ["0x6545664667476848", "0x6141624263436444", "0x55DD55DD44CC44CC", "0x77FF77FF66EE66EE"] + }, + "MemoryRegions": { + "0x100000000": "4096" + } +} +%endif + +lea rdx, [rel .data] + +vmovapd ymm0, [rdx] +vmovapd ymm1, [rdx + 32] + +vpunpcklbw xmm2, xmm0, [rdx + 32] +vpunpcklbw xmm3, xmm0, xmm1 + +vpunpcklbw ymm4, ymm0, [rdx + 32] +vpunpcklbw ymm5, ymm0, ymm1 + +hlt + +align 32 +.data: +dq 0x4142434445464748 +dq 0x5152535455565758 +dq 0xFFFFEEEEDDDDCCCC +dq 0xBBBBAAAA99998888 + +dq 0x6162636465666768 +dq 0x7172737475767778 +dq 0x7777666655554444 +dq 0x3333222211110000