Copyright (C) Intel Corporation. All Rights Reserved. # Notices and Disclaimers Intel technologies may require enabled hardware, software or service activation. No product or component can be absolutely secure. Your costs and results may vary. All product plans and roadmaps are subject to change without notice. The products described may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request. Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade. Code names are used by Intel to identify products, technologies, or services that are in development and not publicly available. These are not “commercial” names and not intended to function as trademarks. No license (express or implied, by estoppel or otherwise) to any intellectual property rights is granted by this document, with the sole exception that a) you may publish an unmodified copy and b) code, identified as Sample Code in this document is licensed subject to the Zero-Clause BSD open source license (0BSD), [https://opensource.org/licenses/0BSD](https://opensource.org/licenses/0BSD). You may create software implementations based on this document and in compliance with the foregoing that are intended to execute on the Intel product(s) referenced in this document. No rights are granted to create modifications or derivatives of this document. You may not use or facilitate the use of this document in connection with any infringement or other legal analysis concerning Intel products described herein. You agree to grant Intel a non-exclusive, royalty-free license to any patent claim thereafter drafted which includes subject matter disclosed herein. If you give Intel any comments or suggestions related to this document or the information contained in it, Intel can use them in any way and disclose them to anyone, without payment or other obligations to you. You represent and warrant that you own, or have sufficient rights from the owner of, any such comments or suggestions, and the intellectual property rights in them, to grant the above permission. © Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries. Other names and brands may be claimed as the property of others. The code in this file is not 'Sample Code'. Std::Boolean::Strict_Or Std::Integer::Add Std::Integer::Ne Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Effective_Address Handle_RIP_Relative_Address Logical_Mem_Read Mem_Fetch Report_Invalid_Opcode Report_Not_Available_Exception Write_XMM Instr_SHUF_lanes False CR4 CR0 Memory_Read Explicitly_Aligned Next_IP let register_size := 128; let element_size := 32; let disp8n := 1; let (ea_offset, segment, next_ip0, is_rip_relative) := Effective_Address?(address_size, context, mod, rm, disp8n, ip); let imm8 := Mem_Fetch?(next_ip0, 8); let effective_address := Handle_RIP_Relative_Address(address_size, is_rip_relative, ea_offset, next_ip0 + 1); LockPrefix #UD REP_Not0 #UD CR4_OSFXSR_0 #UD CR0_EM_1 #UD CR0_TS_1 #NM let src1 := Logical_Mem_Read?(segment, effective_address, register_size, alignment_type=>Explicitly_Aligned); Next_IP := next_ip0 + 1; let result := Instr_SHUF_lanes(element_size, register_size, src1, imm8); Write_XMM(context.rex_r4 ++ context.rex_r3 ++ reg, result, (context.vex_prefix_present or context.evex_prefix_present)); xmm1 m1282 imm8 SSE2 result ModRM:reg src1 ModRM:r/m Std::Boolean::Strict_Or Std::Integer::Add Std::Integer::Ne Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Calculate_RM Mem_Fetch Read_XMM Report_Invalid_Opcode Report_Not_Available_Exception Write_XMM Instr_SHUF_lanes CR4 CR0 Next_IP let register_size := 128; let element_size := 32; let src1 := Read_XMM(Calculate_RM(mod, rm, context), register_size); let imm8 := Mem_Fetch?(ip, 8); LockPrefix #UD REP_Not0 #UD CR4_OSFXSR_0 #UD CR0_EM_1 #UD CR0_TS_1 #NM Next_IP := ip + 1; let result := Instr_SHUF_lanes(element_size, register_size, src1, imm8); Write_XMM(context.rex_r4 ++ context.rex_r3 ++ reg, result, (context.vex_prefix_present or context.evex_prefix_present)); xmm1 xmm2 imm8 SSE2 result ModRM:reg src1 ModRM:r/m let lanes := register_size / 128; let elements := 128 / element_size; let index_size := Log2(elements); var result := Zero(register_size); for lane := 0 to lanes-1 do let s1 := src1[lane *: 128]; var lane_result := Zero(128); for i := 0 to elements-1 do let index := Integer::Align_Down(i, elements) + Unsigned(imm8[i % elements *: index_size]); let r := s1[index *: element_size]; lane_result[i *: element_size] := r; endfor; result[lane *: 128] := lane_result; endfor; Std::Integer::Add Std::Integer::Align_Down Std::Integer::Le Std::Integer::Lt Std::Integer::Multiply Std::Integer::Positive_Divide Std::Integer::Positive_Remainder Std::Integer::Subtract Std::Bits::Unsigned Std::Bits::Zero Std::Integer::Log2 Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Current_Mode Effective_Address Handle_RIP_Relative_Address Logical_Mem_Read Mem_Fetch Report_Invalid_Opcode Report_Not_Available_Exception Write_XMM Instr_SHUF_lanes False Real_Address_Mode Virtual_8086_Mode CR4 CR0 Memory_Read Normal_Alignment XCR0 Next_IP let register_size := 128; let element_size := 32; let disp8n := 1; let (ea_offset, segment, next_ip0, is_rip_relative) := Effective_Address?(address_size, context, mod, rm, disp8n, ip); let imm8 := Mem_Fetch?(next_ip0, 8); let effective_address := Handle_RIP_Relative_Address(address_size, is_rip_relative, ea_offset, next_ip0 + 1); if not ((not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM let src1 := Logical_Mem_Read?(segment, effective_address, register_size); Next_IP := next_ip0 + 1; let result := Instr_SHUF_lanes(element_size, register_size, src1, imm8); Write_XMM(context.rex_r4 ++ context.rex_r3 ++ reg, result, (context.vex_prefix_present or context.evex_prefix_present)); xmm1 m1282 imm8 AVX result ModRM:reg src1 ModRM:r/m Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Calculate_RM Current_Mode Mem_Fetch Read_XMM Report_Invalid_Opcode Report_Not_Available_Exception Write_XMM Instr_SHUF_lanes Real_Address_Mode Virtual_8086_Mode CR4 CR0 XCR0 Next_IP let register_size := 128; let element_size := 32; let src1 := Read_XMM(Calculate_RM(mod, rm, context), register_size); let imm8 := Mem_Fetch?(ip, 8); if not ((not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM Next_IP := ip + 1; let result := Instr_SHUF_lanes(element_size, register_size, src1, imm8); Write_XMM(context.rex_r4 ++ context.rex_r3 ++ reg, result, (context.vex_prefix_present or context.evex_prefix_present)); xmm1 xmm2 imm8 AVX result ModRM:reg src1 ModRM:r/m Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Current_Mode Effective_Address Handle_RIP_Relative_Address Logical_Mem_Read Mem_Fetch Report_Invalid_Opcode Report_Not_Available_Exception Write_YMM Instr_SHUF_lanes False Real_Address_Mode Virtual_8086_Mode CR4 CR0 Memory_Read Normal_Alignment XCR0 Next_IP let register_size := 256; let element_size := 32; let disp8n := 1; let (ea_offset, segment, next_ip0, is_rip_relative) := Effective_Address?(address_size, context, mod, rm, disp8n, ip); let imm8 := Mem_Fetch?(next_ip0, 8); let effective_address := Handle_RIP_Relative_Address(address_size, is_rip_relative, ea_offset, next_ip0 + 1); if not ((not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM let src1 := Logical_Mem_Read?(segment, effective_address, register_size); Next_IP := next_ip0 + 1; let result := Instr_SHUF_lanes(element_size, register_size, src1, imm8); Write_YMM(context.rex_r4 ++ context.rex_r3 ++ reg, result); ymm1 m2562 imm8 AVX2 result ModRM:reg src1 ModRM:r/m Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Calculate_RM Current_Mode Mem_Fetch Read_YMM Report_Invalid_Opcode Report_Not_Available_Exception Write_YMM Instr_SHUF_lanes Real_Address_Mode Virtual_8086_Mode CR4 CR0 XCR0 Next_IP let register_size := 256; let element_size := 32; let src1 := Read_YMM(Calculate_RM(mod, rm, context)); let imm8 := Mem_Fetch?(ip, 8); if not ((not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM Next_IP := ip + 1; let result := Instr_SHUF_lanes(element_size, register_size, src1, imm8); Write_YMM(context.rex_r4 ++ context.rex_r3 ++ reg, result); ymm1 ymm2 imm8 AVX2 result ModRM:reg src1 ModRM:r/m Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Integer::Le Std::Integer::Positive_Divide Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Current_Mode Effective_Address Handle_RIP_Relative_Address Mem_Fetch Read_Mask Read_XMM Report_Invalid_Opcode Report_Not_Available_Exception Vector_Mem_Read Write_XMM Instr_SHUF_lanes_m Real_Address_Mode Virtual_8086_Mode CR4 CR0 XCR0 Next_IP let register_size := 128; let element_size := 32; let disp8n := if context.evex_b == 0b1 then element_size / 8 else 16; let (ea_offset, segment, next_ip0, is_rip_relative) := Effective_Address?(address_size, context, mod, rm, disp8n, ip); let old := Read_XMM(context.rex_r4 ++ context.rex_r3 ++ reg, register_size); let k := Read_Mask(context.evex_aaa); let imm8 := Mem_Fetch?(next_ip0, 8); let effective_address := Handle_RIP_Relative_Address(address_size, is_rip_relative, ea_offset, next_ip0 + 1); if not ((context.rex_w == 0b0) and (not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111) and (context.evex_v4 == 0b0)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD VEX_W_1 #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM let src1 := Vector_Mem_Read?(register_size, element_size, context.evex_b == 0b1, segment, effective_address); Next_IP := next_ip0 + 1; let result := Instr_SHUF_lanes_m(element_size, register_size, old, k, src1, imm8, context, mod); Write_XMM(context.rex_r4 ++ context.rex_r3 ++ reg, result, (context.vex_prefix_present or context.evex_prefix_present)); xmm1 {k1}{z} m1282 imm8 AVX512FAVX512VL AVX10.1 result ModRM:reg src1 ModRM:r/m Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Calculate_RM Current_Mode Mem_Fetch Read_Mask Read_XMM Report_Invalid_Opcode Report_Not_Available_Exception Write_XMM Instr_SHUF_lanes_m Real_Address_Mode Virtual_8086_Mode CR4 CR0 XCR0 Next_IP let register_size := 128; let element_size := 32; let old := Read_XMM(context.rex_r4 ++ context.rex_r3 ++ reg, register_size); let k := Read_Mask(context.evex_aaa); let src1 := Read_XMM(Calculate_RM(mod, rm, context), register_size); let imm8 := Mem_Fetch?(ip, 8); if not ((context.evex_b == 0b0) and (context.rex_w == 0b0) and (not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111) and (context.evex_v4 == 0b0)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD VEX_W_1 #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM Next_IP := ip + 1; let result := Instr_SHUF_lanes_m(element_size, register_size, old, k, src1, imm8, context, mod); Write_XMM(context.rex_r4 ++ context.rex_r3 ++ reg, result, (context.vex_prefix_present or context.evex_prefix_present)); xmm1 {k1}{z} xmm2 imm8 AVX512FAVX512VL AVX10.1 result ModRM:reg src1 ModRM:r/m Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Integer::Le Std::Integer::Positive_Divide Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Current_Mode Effective_Address Handle_RIP_Relative_Address Mem_Fetch Read_Mask Report_Invalid_Opcode Report_Not_Available_Exception Vector_Mem_Read Write_XMM Instr_SHUF_lanes_z Real_Address_Mode Virtual_8086_Mode CR4 CR0 XCR0 Next_IP let register_size := 128; let element_size := 32; let disp8n := if context.evex_b == 0b1 then element_size / 8 else 16; let (ea_offset, segment, next_ip0, is_rip_relative) := Effective_Address?(address_size, context, mod, rm, disp8n, ip); let k := Read_Mask(context.evex_aaa); let imm8 := Mem_Fetch?(next_ip0, 8); let effective_address := Handle_RIP_Relative_Address(address_size, is_rip_relative, ea_offset, next_ip0 + 1); if not ((context.rex_w == 0b0) and (not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111) and (context.evex_v4 == 0b0)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD VEX_W_1 #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM let src1 := Vector_Mem_Read?(register_size, element_size, context.evex_b == 0b1, segment, effective_address); Next_IP := next_ip0 + 1; let result := Instr_SHUF_lanes_z(element_size, register_size, k, src1, imm8, context, mod); Write_XMM(context.rex_r4 ++ context.rex_r3 ++ reg, result, (context.vex_prefix_present or context.evex_prefix_present)); xmm1 {k1}{z} m1282 imm8 AVX512FAVX512VL AVX10.1 result ModRM:reg src1 ModRM:r/m Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Calculate_RM Current_Mode Mem_Fetch Read_Mask Read_XMM Report_Invalid_Opcode Report_Not_Available_Exception Write_XMM Instr_SHUF_lanes_z Real_Address_Mode Virtual_8086_Mode CR4 CR0 XCR0 Next_IP let register_size := 128; let element_size := 32; let k := Read_Mask(context.evex_aaa); let src1 := Read_XMM(Calculate_RM(mod, rm, context), register_size); let imm8 := Mem_Fetch?(ip, 8); if not ((context.evex_b == 0b0) and (context.rex_w == 0b0) and (not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111) and (context.evex_v4 == 0b0)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD VEX_W_1 #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM Next_IP := ip + 1; let result := Instr_SHUF_lanes_z(element_size, register_size, k, src1, imm8, context, mod); Write_XMM(context.rex_r4 ++ context.rex_r3 ++ reg, result, (context.vex_prefix_present or context.evex_prefix_present)); xmm1 {k1}{z} xmm2 imm8 AVX512FAVX512VL AVX10.1 result ModRM:reg src1 ModRM:r/m let lanes := register_size / 128; let elements := 128 / element_size; let index_size := Log2(elements); var result := Zero(register_size); for lane := 0 to lanes-1 do let lane_old := old[lane *: 128]; let s1 := src1[lane *: 128]; var lane_result := Zero(128); for i := 0 to elements-1 do if k[lane * elements + i] == 0b1 then let index := Integer::Align_Down(i, elements) + Unsigned(imm8[i % elements *: index_size]); let r := s1[index *: element_size]; lane_result[i *: element_size] := r; else lane_result[i *: element_size] := lane_old[i *: element_size]; endif; endfor; result[lane *: 128] := lane_result; endfor; Std::Integer::Add Std::Integer::Align_Down Std::Integer::Le Std::Integer::Lt Std::Integer::Multiply Std::Integer::Positive_Divide Std::Integer::Positive_Remainder Std::Integer::Subtract Std::Bits::Unsigned Std::Bits::Eq Std::Bits::Zero Std::Integer::Log2 let lanes := register_size / 128; let elements := 128 / element_size; let index_size := Log2(elements); var result := Zero(register_size); for lane := 0 to lanes-1 do let s1 := src1[lane *: 128]; var lane_result := Zero(128); for i := 0 to elements-1 do if k[lane * elements + i] == 0b1 then let index := Integer::Align_Down(i, elements) + Unsigned(imm8[i % elements *: index_size]); let r := s1[index *: element_size]; lane_result[i *: element_size] := r; else lane_result[i *: element_size] := Zero(element_size); endif; endfor; result[lane *: 128] := lane_result; endfor; Std::Integer::Add Std::Integer::Align_Down Std::Integer::Le Std::Integer::Lt Std::Integer::Multiply Std::Integer::Positive_Divide Std::Integer::Positive_Remainder Std::Integer::Subtract Std::Bits::Unsigned Std::Bits::Eq Std::Bits::Zero Std::Integer::Log2 Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Integer::Le Std::Integer::Positive_Divide Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Current_Mode Effective_Address Handle_RIP_Relative_Address Mem_Fetch Read_Mask Read_YMM Report_Invalid_Opcode Report_Not_Available_Exception Vector_Mem_Read Write_YMM Instr_SHUF_lanes_m Real_Address_Mode Virtual_8086_Mode CR4 CR0 XCR0 Next_IP let register_size := 256; let element_size := 32; let disp8n := if context.evex_b == 0b1 then element_size / 8 else 32; let (ea_offset, segment, next_ip0, is_rip_relative) := Effective_Address?(address_size, context, mod, rm, disp8n, ip); let old := Read_YMM(context.rex_r4 ++ context.rex_r3 ++ reg); let k := Read_Mask(context.evex_aaa); let imm8 := Mem_Fetch?(next_ip0, 8); let effective_address := Handle_RIP_Relative_Address(address_size, is_rip_relative, ea_offset, next_ip0 + 1); if not ((context.rex_w == 0b0) and (not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111) and (context.evex_v4 == 0b0)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD VEX_W_1 #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM let src1 := Vector_Mem_Read?(register_size, element_size, context.evex_b == 0b1, segment, effective_address); Next_IP := next_ip0 + 1; let result := Instr_SHUF_lanes_m(element_size, register_size, old, k, src1, imm8, context, mod); Write_YMM(context.rex_r4 ++ context.rex_r3 ++ reg, result); ymm1 {k1}{z} m2562 imm8 AVX512FAVX512VL AVX10.1 result ModRM:reg src1 ModRM:r/m Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Calculate_RM Current_Mode Mem_Fetch Read_Mask Read_YMM Report_Invalid_Opcode Report_Not_Available_Exception Write_YMM Instr_SHUF_lanes_m Real_Address_Mode Virtual_8086_Mode CR4 CR0 XCR0 Next_IP let register_size := 256; let element_size := 32; let old := Read_YMM(context.rex_r4 ++ context.rex_r3 ++ reg); let k := Read_Mask(context.evex_aaa); let src1 := Read_YMM(Calculate_RM(mod, rm, context)); let imm8 := Mem_Fetch?(ip, 8); if not ((context.evex_b == 0b0) and (context.rex_w == 0b0) and (not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111) and (context.evex_v4 == 0b0)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD VEX_W_1 #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM Next_IP := ip + 1; let result := Instr_SHUF_lanes_m(element_size, register_size, old, k, src1, imm8, context, mod); Write_YMM(context.rex_r4 ++ context.rex_r3 ++ reg, result); ymm1 {k1}{z} ymm2 imm8 AVX512FAVX512VL AVX10.1 result ModRM:reg src1 ModRM:r/m Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Integer::Le Std::Integer::Positive_Divide Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Current_Mode Effective_Address Handle_RIP_Relative_Address Mem_Fetch Read_Mask Report_Invalid_Opcode Report_Not_Available_Exception Vector_Mem_Read Write_YMM Instr_SHUF_lanes_z Real_Address_Mode Virtual_8086_Mode CR4 CR0 XCR0 Next_IP let register_size := 256; let element_size := 32; let disp8n := if context.evex_b == 0b1 then element_size / 8 else 32; let (ea_offset, segment, next_ip0, is_rip_relative) := Effective_Address?(address_size, context, mod, rm, disp8n, ip); let k := Read_Mask(context.evex_aaa); let imm8 := Mem_Fetch?(next_ip0, 8); let effective_address := Handle_RIP_Relative_Address(address_size, is_rip_relative, ea_offset, next_ip0 + 1); if not ((context.rex_w == 0b0) and (not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111) and (context.evex_v4 == 0b0)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD VEX_W_1 #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM let src1 := Vector_Mem_Read?(register_size, element_size, context.evex_b == 0b1, segment, effective_address); Next_IP := next_ip0 + 1; let result := Instr_SHUF_lanes_z(element_size, register_size, k, src1, imm8, context, mod); Write_YMM(context.rex_r4 ++ context.rex_r3 ++ reg, result); ymm1 {k1}{z} m2562 imm8 AVX512FAVX512VL AVX10.1 result ModRM:reg src1 ModRM:r/m Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Calculate_RM Current_Mode Mem_Fetch Read_Mask Read_YMM Report_Invalid_Opcode Report_Not_Available_Exception Write_YMM Instr_SHUF_lanes_z Real_Address_Mode Virtual_8086_Mode CR4 CR0 XCR0 Next_IP let register_size := 256; let element_size := 32; let k := Read_Mask(context.evex_aaa); let src1 := Read_YMM(Calculate_RM(mod, rm, context)); let imm8 := Mem_Fetch?(ip, 8); if not ((context.evex_b == 0b0) and (context.rex_w == 0b0) and (not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111) and (context.evex_v4 == 0b0)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD VEX_W_1 #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM Next_IP := ip + 1; let result := Instr_SHUF_lanes_z(element_size, register_size, k, src1, imm8, context, mod); Write_YMM(context.rex_r4 ++ context.rex_r3 ++ reg, result); ymm1 {k1}{z} ymm2 imm8 AVX512FAVX512VL AVX10.1 result ModRM:reg src1 ModRM:r/m Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Integer::Le Std::Integer::Positive_Divide Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Current_Mode Effective_Address Handle_RIP_Relative_Address Mem_Fetch Read_Mask Read_ZMM Report_Invalid_Opcode Report_Not_Available_Exception Vector_Mem_Read Write_ZMM Instr_SHUF_lanes_m Real_Address_Mode Virtual_8086_Mode CR4 CR0 XCR0 Next_IP let register_size := 512; let element_size := 32; let disp8n := if context.evex_b == 0b1 then element_size / 8 else 64; let (ea_offset, segment, next_ip0, is_rip_relative) := Effective_Address?(address_size, context, mod, rm, disp8n, ip); let old := Read_ZMM(context.rex_r4 ++ context.rex_r3 ++ reg); let k := Read_Mask(context.evex_aaa); let imm8 := Mem_Fetch?(next_ip0, 8); let effective_address := Handle_RIP_Relative_Address(address_size, is_rip_relative, ea_offset, next_ip0 + 1); if not ((context.rex_w == 0b0) and (not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111) and (context.evex_v4 == 0b0)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD VEX_W_1 #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM let src1 := Vector_Mem_Read?(register_size, element_size, context.evex_b == 0b1, segment, effective_address); Next_IP := next_ip0 + 1; let result := Instr_SHUF_lanes_m(element_size, register_size, old, k, src1, imm8, context, mod); Write_ZMM(context.rex_r4 ++ context.rex_r3 ++ reg, result); zmm1 {k1}{z} m5122 imm8 AVX512F AVX10.1 result ModRM:reg src1 ModRM:r/m Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Calculate_RM Current_Mode Mem_Fetch Read_Mask Read_ZMM Report_Invalid_Opcode Report_Not_Available_Exception Write_ZMM Instr_SHUF_lanes_m Real_Address_Mode Virtual_8086_Mode CR4 CR0 XCR0 Next_IP let register_size := 512; let element_size := 32; let old := Read_ZMM(context.rex_r4 ++ context.rex_r3 ++ reg); let k := Read_Mask(context.evex_aaa); let src1 := Read_ZMM(Calculate_RM(mod, rm, context)); let imm8 := Mem_Fetch?(ip, 8); if not ((context.evex_b == 0b0) and (context.rex_w == 0b0) and (not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111) and (context.evex_v4 == 0b0)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD VEX_W_1 #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM Next_IP := ip + 1; let result := Instr_SHUF_lanes_m(element_size, register_size, old, k, src1, imm8, context, mod); Write_ZMM(context.rex_r4 ++ context.rex_r3 ++ reg, result); zmm1 {k1}{z} zmm2 imm8 AVX512F AVX10.1 result ModRM:reg src1 ModRM:r/m Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Integer::Le Std::Integer::Positive_Divide Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Current_Mode Effective_Address Handle_RIP_Relative_Address Mem_Fetch Read_Mask Report_Invalid_Opcode Report_Not_Available_Exception Vector_Mem_Read Write_ZMM Instr_SHUF_lanes_z Real_Address_Mode Virtual_8086_Mode CR4 CR0 XCR0 Next_IP let register_size := 512; let element_size := 32; let disp8n := if context.evex_b == 0b1 then element_size / 8 else 64; let (ea_offset, segment, next_ip0, is_rip_relative) := Effective_Address?(address_size, context, mod, rm, disp8n, ip); let k := Read_Mask(context.evex_aaa); let imm8 := Mem_Fetch?(next_ip0, 8); let effective_address := Handle_RIP_Relative_Address(address_size, is_rip_relative, ea_offset, next_ip0 + 1); if not ((context.rex_w == 0b0) and (not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111) and (context.evex_v4 == 0b0)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD VEX_W_1 #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM let src1 := Vector_Mem_Read?(register_size, element_size, context.evex_b == 0b1, segment, effective_address); Next_IP := next_ip0 + 1; let result := Instr_SHUF_lanes_z(element_size, register_size, k, src1, imm8, context, mod); Write_ZMM(context.rex_r4 ++ context.rex_r3 ++ reg, result); zmm1 {k1}{z} m5122 imm8 AVX512F AVX10.1 result ModRM:reg src1 ModRM:r/m Std::Boolean::Not Std::Boolean::Strict_And Std::Boolean::Strict_Or Std::Integer::Add Std::Bits::Add_int Std::Bits::Append Std::Bits::Eq Std::Bits::Not Calculate_RM Current_Mode Mem_Fetch Read_Mask Read_ZMM Report_Invalid_Opcode Report_Not_Available_Exception Write_ZMM Instr_SHUF_lanes_z Real_Address_Mode Virtual_8086_Mode CR4 CR0 XCR0 Next_IP let register_size := 512; let element_size := 32; let k := Read_Mask(context.evex_aaa); let src1 := Read_ZMM(Calculate_RM(mod, rm, context)); let imm8 := Mem_Fetch?(ip, 8); if not ((context.evex_b == 0b0) and (context.rex_w == 0b0) and (not context.vvvv[3] == 0b1) and (not context.vvvv[2:0] == 0b111) and (context.evex_v4 == 0b0)) then Report_Invalid_Opcode!(); endif; LockPrefix #UD ModeRealOrVirtual #UD VEX_W_1 #UD CR4_OSXSAVE_0 #UD XCR0_AVX_0 #UD CR0_TS_1 #NM Next_IP := ip + 1; let result := Instr_SHUF_lanes_z(element_size, register_size, k, src1, imm8, context, mod); Write_ZMM(context.rex_r4 ++ context.rex_r3 ++ reg, result); zmm1 {k1}{z} zmm2 imm8 AVX512F AVX10.1 result ModRM:reg src1 ModRM:r/m vector/permute An explicitly-aligned memory access is performed. If EVEX.b == 0b1, a single element is read from memory and broadcast across the vector. If EVEX.z == 0b0, the previous value of the result is read into `old` and the masking variant of the operation is used; if EVEX.z == 0b1, the zeroing variant of the operation is used.