CPYPWTN, CPYMWTN, CPYEWTN Memory copy, writes unprivileged, reads and writes non-temporal These instructions copy a requested number of bytes in memory from a source address to a destination address. The prologue, main, and epilogue instructions are expected to be run in succession and to appear consecutively in memory: CPYPWTN, then CPYMWTN, and then CPYEWTN. CPYPWTN performs some preconditioning of the arguments suitable for using the CPYMWTN instruction, and copies an IMPLEMENTATION DEFINED portion of the requested number of bytes. CPYMWTN copies a further IMPLEMENTATION DEFINED portion of the remaining bytes. CPYEWTN copies any final remaining bytes. The ability to copy an IMPLEMENTATION DEFINED number of bytes allows an implementation to optimize how the bytes being copied are divided between the different instructions. For more information on exceptions specific to memory copy instructions, see Memory Copy and Memory Set exceptions. The architecture supports two algorithms for the memory copy: option A and option B. Which algorithm is used is IMPLEMENTATION DEFINED. Portable software should not assume that the choice of algorithm is constant. For CPYPWTN: If Xn[63:55] != '000000000', the copy size Xn is saturated to 0x007FFFFFFFFFFFFF. After saturation is performed, the direction of the memory copy is based on the following: If (Xs[55:0] > Xd[55:0]) and (Xd[55:0] + saturated copy size) > Xs[55:0], then the direction is forward. If (Xs[55:0] < Xd[55:0]) and (Xs[55:0] + saturated copy size) > Xd[55:0], then the direction is backward. Otherwise, the direction is an IMPLEMENTATION DEFINED choice between forward and backward. On completion of CPYPWTN, option A: PSTATE.{N,Z,C,V} are set to {0,0,0,0}. If the copy is in the forward direction, then: Xn holds -1 times the number of bytes in the saturated copy size remaining to be copied. Xs holds the original Xs + saturated copy size. Xd holds the original Xd + saturated copy size. If the copy is in the backward direction, then: Xn holds the number of bytes in the saturated copy size remaining to be copied. Xs and Xd are unchanged. On completion of CPYPWTN, option B: If the copy is in the forward direction, then: Xn holds the number of bytes in the saturated copy size remaining to be copied. Xs holds the lowest address that has not been copied from. Xd holds the lowest address that has not been copied to. PSTATE.{N,Z,C,V} are set to {0,0,1,0}. If the copy is in the backward direction, then: Xn holds the number of bytes in the saturated copy size remaining to be copied. Xs holds the highest address that has not been copied from + 1. Xd holds the highest address that has not been copied to + 1. PSTATE.{N,Z,C,V} are set to {1,0,1,0}. For CPYMWTN, option A, when PSTATE.C = 0: Xn holds a signed 64-bit integer. If the copy is in the forward direction (Xn holds a negative number), then: Xn holds -1 times the number of bytes remaining to be copied. Xs holds the lowest address to be copied from - Xn. Xd holds the lowest address to be copied to - Xn. On completion of the instruction, Xn holds -1 times the number of bytes remaining to be copied. If the copy is in the backward direction (Xn holds a positive number), then: Xn holds the number of bytes remaining to be copied. Xs holds the highest address to be copied from + 1 - Xn. Xd holds the highest address to be copied to + 1 - Xn. On completion of the instruction, Xn holds the number of bytes remaining to be copied. For CPYMWTN, option B, when PSTATE.C = 1: Xn holds the number of bytes remaining to be copied. If the copy is in the forward direction (PSTATE.N == 0), then: Xs holds the lowest address to be copied from. Xd holds the lowest address to be copied to. On completion of the instruction: Xn holds the number of bytes remaining to be copied. Xs holds the lowest address that has not been copied from. Xd holds the lowest address that has not been copied to. If the copy is in the backward direction (PSTATE.N == 1), then: Xs holds the highest address to be copied from + 1. Xd holds the highest address to be copied to + 1. On completion of the instruction: Xn holds the number of bytes remaining to be copied. Xs holds the highest address that has not been copied from + 1. Xd holds the highest address that has not been copied to + 1. For CPYEWTN, option A, when PSTATE.C = 0: Xn holds a signed 64-bit integer. If the copy is in the forward direction (Xn holds a negative number), then: Xn holds -1 times the number of bytes remaining to be copied. Xs holds the lowest address to be copied from - Xn. Xd holds the lowest address to be copied to - Xn. On completion of the instruction, Xn holds 0. If the copy is in the backward direction (Xn holds a positive number), then: Xn holds the number of bytes remaining to be copied. Xs holds the highest address to be copied from + 1 - Xn. Xd holds the highest address to be copied to + 1 - Xn. On completion of the instruction, Xn holds 0. For CPYEWTN, option B, when PSTATE.C = 1: Xn holds the number of bytes remaining to be copied. If the copy is in the forward direction (PSTATE.N == 0), then: Xs holds the lowest address to be copied from. Xd holds the lowest address to be copied to. On completion of the instruction: Xn holds 0. Xs holds the lowest address that has not been copied from. Xd holds the lowest address that has not been copied to. If the copy is in the backward direction (PSTATE.N == 1), then: Xs holds the highest address to be copied from + 1. Xd holds the highest address to be copied to + 1. On completion of the instruction: Xn holds 0. Xs holds the highest address that has not been copied from + 1. Xd holds the highest address that has not been copied to + 1. Explicit Memory Write effects produced by the instruction behave as if the instruction was executed at EL0 if the Effective value of PSTATE.UAO is 0 and either: The instruction is executed at EL1. The instruction is executed at EL2 when the Effective value of HCR_EL2().{E2H, TGE} is '11'. Otherwise, the Explicit Memory Write effects operate with the restrictions determined by the Exception level at which the instruction is executed. For information about the CONSTRAINED UNPREDICTABLE behavior of this instruction, see Architectural Constraints on UNPREDICTABLE behaviors, and particularly Memory Copy and Memory Set CPY* and Crossing a page boundary with different memory types or Shareability attributes. 0 1 1 1 0 1 0 1 1 0 1 0 1 0 0 CPYPWTN [<Xd>]!, [<Xs>]!, <Xn>! 0 1 CPYMWTN [<Xd>]!, [<Xs>]!, <Xn>! 1 0 CPYEWTN [<Xd>]!, [<Xs>]!, <Xn>! if !IsFeatureImplemented(FEAT_MOPS) || sz != '00' then EndOfDecode(Decode_UNDEF); end; var memcpy : CPYParams; memcpy.d = UInt(Rd); memcpy.s = UInt(Rs); memcpy.n = UInt(Rn); let options : bits(4) = op2; let rnontemporal : boolean = options[3] == '1'; let wnontemporal : boolean = options[2] == '1'; case op1 of when '00' => memcpy.stage = MOPSStage_Prologue; when '01' => memcpy.stage = MOPSStage_Main; when '10' => memcpy.stage = MOPSStage_Epilogue; end; <Xd> For the "Prologue" variant: is the 64-bit name of the general-purpose register that holds the destination address and is updated by the instruction, encoded in the "Rd" field. <Xd> For the "Epilogue" and "Main" variants: is the 64-bit name of the general-purpose register that holds an encoding of the destination address, encoded in the "Rd" field. <Xs> For the "Prologue" variant: is the 64-bit name of the general-purpose register that holds the source address and is updated by the instruction, encoded in the "Rs" field. <Xs> For the "Epilogue" and "Main" variants: is the 64-bit name of the general-purpose register that holds an encoding of the source address, encoded in the "Rs" field. <Xn> For the "Prologue" variant: is the 64-bit name of the general-purpose register that holds the number of bytes to be transferred and is updated by the instruction to encode the remaining size and destination, encoded in the "Rn" field. <Xn> For the "Main" variant: is the 64-bit name of the general-purpose register that holds an encoding of the number of bytes to be transferred, encoded in the "Rn" field. <Xn> For the "Epilogue" variant: is the 64-bit name of the general-purpose register that holds an encoding of the number of bytes to be transferred and is set to zero on completion of the instruction, encoded in the "Rn" field. CheckMOPSEnabled(); CheckCPYConstrainedUnpredictable(memcpy.n, memcpy.d, memcpy.s); memcpy.nzcv = PSTATE.[N,Z,C,V]; memcpy.toaddress = X{64}(memcpy.d); memcpy.fromaddress = X{64}(memcpy.s); if memcpy.stage == MOPSStage_Prologue then memcpy.cpysize = UInt(X{64}(memcpy.n)); else memcpy.cpysize = SInt(X{64}(memcpy.n)); end; memcpy.implements_option_a = CPYOptionA(); let rprivileged : boolean = (if options[1] == '1' then AArch64_IsUnprivAccessPriv() else PSTATE.EL != EL0); let wprivileged : boolean = (if options[0] == '1' then AArch64_IsUnprivAccessPriv() else PSTATE.EL != EL0); let raccdesc : AccessDescriptor = CreateAccDescMOPS(MemOp_LOAD, rprivileged, rnontemporal); let waccdesc : AccessDescriptor = CreateAccDescMOPS(MemOp_STORE, wprivileged, wnontemporal); if memcpy.stage == MOPSStage_Prologue then if memcpy.cpysize > ArchMaxMOPSCPYSize then memcpy.cpysize = ArchMaxMOPSCPYSize; end; memcpy.forward = IsMemCpyForward(memcpy); if memcpy.implements_option_a then memcpy.nzcv = '0000'; if memcpy.forward then // Copy in the forward direction offsets the arguments. memcpy.toaddress = memcpy.toaddress + memcpy.cpysize; memcpy.fromaddress = memcpy.fromaddress + memcpy.cpysize; memcpy.cpysize = 0 - memcpy.cpysize; end; else if !memcpy.forward then // Copy in the reverse direction offsets the arguments. memcpy.toaddress = memcpy.toaddress + memcpy.cpysize; memcpy.fromaddress = memcpy.fromaddress + memcpy.cpysize; memcpy.nzcv = '1010'; else memcpy.nzcv = '0010'; end; end; end; memcpy.stagecpysize = MemCpyStageSize(memcpy); if memcpy.stage != MOPSStage_Prologue then memcpy.forward = memcpy.cpysize < 0 || (!memcpy.implements_option_a && memcpy.nzcv[3] == '0'); CheckMemCpyParams(memcpy, options); end; var copied : integer; var iswrite : boolean; var memaddrdesc : AddressDescriptor; var memstatus : PhysMemRetStatus; var fault : boolean = FALSE; var B : MOPSBlockSize = 0; if memcpy.implements_option_a then while memcpy.stagecpysize != 0 && !fault looplimit ArchMaxMOPSCPYSize do // IMP DEF selection of the block size that is worked on. While many // implementations might make this constant, that is not assumed. B = CPYSizeChoice(memcpy); if memcpy.forward then assert B <= -1 * memcpy.stagecpysize; (copied, iswrite, memaddrdesc, memstatus) = MemCpyBytes( memcpy.toaddress + memcpy.cpysize, memcpy.fromaddress + memcpy.cpysize, memcpy.forward, B, raccdesc, waccdesc); if copied != B then fault = TRUE; else memcpy.cpysize = memcpy.cpysize + B; memcpy.stagecpysize = memcpy.stagecpysize + B; end; else assert B <= memcpy.stagecpysize; memcpy.cpysize = memcpy.cpysize - B; memcpy.stagecpysize = memcpy.stagecpysize - B; (copied, iswrite, memaddrdesc, memstatus) = MemCpyBytes( memcpy.toaddress + memcpy.cpysize, memcpy.fromaddress + memcpy.cpysize, memcpy.forward, B, raccdesc, waccdesc); if copied != B then fault = TRUE; memcpy.cpysize = memcpy.cpysize + B; memcpy.stagecpysize = memcpy.stagecpysize + B; end; end; end; else while memcpy.stagecpysize > 0 && !fault looplimit ArchMaxMOPSCPYSize do // IMP DEF selection of the block size that is worked on. While many // implementations might make this constant, that is not assumed. B = CPYSizeChoice(memcpy); assert B <= memcpy.stagecpysize; if memcpy.forward then (copied, iswrite, memaddrdesc, memstatus) = MemCpyBytes(memcpy.toaddress, memcpy.fromaddress, memcpy.forward, B, raccdesc, waccdesc); if copied != B then fault = TRUE; else memcpy.fromaddress = memcpy.fromaddress + B; memcpy.toaddress = memcpy.toaddress + B; end; else (copied, iswrite, memaddrdesc, memstatus) = MemCpyBytes(memcpy.toaddress - B, memcpy.fromaddress - B, memcpy.forward, B, raccdesc, waccdesc); if copied != B then fault = TRUE; else memcpy.fromaddress = memcpy.fromaddress - B; memcpy.toaddress = memcpy.toaddress - B; end; end; if !fault then memcpy.cpysize = memcpy.cpysize - B; memcpy.stagecpysize = memcpy.stagecpysize - B; end; end; end; UpdateCpyRegisters(memcpy, fault, copied); if fault then if IsFault(memaddrdesc) then AArch64_Abort(memaddrdesc.fault); else let accdesc : AccessDescriptor = if iswrite then waccdesc else raccdesc; HandleExternalAbort(memstatus, iswrite, memaddrdesc, B, accdesc); end; end; if memcpy.stage == MOPSStage_Prologue then PSTATE.[N,Z,C,V] = memcpy.nzcv; end; 2026-03-12 12:23:09 2025-09_rel_asl1