gem5/src/arch/alpha/isa/mem.isa

738 lines
21 KiB
Plaintext
Raw Normal View History

// -*- mode:c++ -*-
// Copyright (c) 2003-2005 The Regents of The University of Michigan
// All rights reserved.
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are
// met: redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer;
// redistributions in binary form must reproduce the above copyright
// notice, this list of conditions and the following disclaimer in the
// documentation and/or other materials provided with the distribution;
// neither the name of the copyright holders nor the names of its
// contributors may be used to endorse or promote products derived from
// this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
//
// Authors: Steve Reinhardt
// Kevin Lim
////////////////////////////////////////////////////////////////////
//
// Memory-format instructions: LoadAddress, Load, Store
//
output header {{
/**
* Base class for general Alpha memory-format instructions.
*/
class Memory : public AlphaStaticInst
{
protected:
/// Memory request flags. See mem_req_base.hh.
unsigned memAccessFlags;
/// Pointer to EAComp object.
Changes to untemplate StaticInst and StaticInstPtr, change the isa to a namespace instead of a class, an improvement to the architecture specific header file selection system, and fixed up a few include paths. arch/alpha/alpha_linux_process.cc: Added using directive for AlphaISA namespace arch/alpha/alpha_memory.hh: arch/alpha/isa/branch.isa: cpu/pc_event.hh: Added typedefs for Addr arch/alpha/alpha_tru64_process.cc: arch/alpha/arguments.cc: Added using directive for AlphaISA arch/alpha/ev5.hh: Added an include of arch/alpha/isa_traits.hh, and a using directive for the AlphaISA namespace. arch/alpha/faults.hh: Added a typedef for the Addr type, and changed the formatting of the faults slightly. arch/alpha/isa/main.isa: Untemplatized StaticInst, added a using for namespace AlphaISA to show up in decoder.cc and the exec.ccs, relocated makeNop to decoder.hh arch/alpha/isa/mem.isa: Untemplatized StaticInst and StaticInstPtr arch/alpha/isa/pal.isa: cpu/base_dyn_inst.cc: Untemplatized StaticInstPtr arch/alpha/isa_traits.hh: Changed variables to be externs instead of static since they are part of a namespace and not a class. arch/alpha/stacktrace.cc: Untemplatized StaticInstPtr, and added a using directive for AlphaISA. arch/alpha/stacktrace.hh: Added some typedefs for Addr and MachInst, and untemplatized StaticInstPtr arch/alpha/vtophys.cc: Added a using directive for AlphaISA arch/alpha/vtophys.hh: Added the AlphaISA namespace specifier where needed arch/isa_parser.py: Changed the placement of the definition of the decodeInst function to be outside the namespaceInst namespace. base/loader/object_file.hh: cpu/o3/bpred_unit.hh: Added a typedef for Addr base/loader/symtab.hh: Added a typedef for Addr, and added a TheISA to Addr in another typedef base/remote_gdb.cc: Added a using namespace TheISA, and untemplatized StaticInstPtr base/remote_gdb.hh: Added typedefs for Addr and MachInst cpu/base.cc: Added TheISA specifier to some variables exported from the isa. cpu/base.hh: Added a typedef for Addr, and TheISA to some variables from the ISA cpu/base_dyn_inst.hh: Untemplatized StaticInstPtr, and added TheISA specifier to some variables from the ISA. cpu/exec_context.hh: Added some typedefs for types from the isa, and added TheISA specifier to some variables from the isa cpu/exetrace.hh: Added typedefs for some types from the ISA, and untemplatized StaticInstPtr cpu/memtest/memtest.cc: cpu/o3/btb.cc: dev/baddev.cc: dev/ide_ctrl.cc: dev/ide_disk.cc: dev/isa_fake.cc: dev/ns_gige.cc: dev/pciconfigall.cc: dev/platform.cc: dev/sinic.cc: dev/uart8250.cc: kern/freebsd/freebsd_system.cc: kern/linux/linux_system.cc: kern/system_events.cc: kern/tru64/dump_mbuf.cc: kern/tru64/tru64_events.cc: sim/process.cc: sim/pseudo_inst.cc: sim/system.cc: Added using namespace TheISA cpu/memtest/memtest.hh: cpu/trace/opt_cpu.hh: cpu/trace/reader/itx_reader.hh: dev/ide_disk.hh: dev/pcidev.hh: dev/platform.hh: dev/tsunami.hh: sim/system.hh: sim/vptr.hh: Added typedef for Addr cpu/o3/2bit_local_pred.hh: Changed the include to use arch/isa_traits.hh instead of arch/alpha/isa_traits.hh. Added typedef for Addr cpu/o3/alpha_cpu.hh: Added typedefs for Addr and IntReg cpu/o3/alpha_cpu_impl.hh: Added this-> to setNextPC to fix a problem since it didn't depend on template parameters any more. Removed "typename" where it was no longer needed. cpu/o3/alpha_dyn_inst.hh: Cleaned up some typedefs, and untemplatized StaticInst cpu/o3/alpha_dyn_inst_impl.hh: untemplatized StaticInstPtr cpu/o3/alpha_impl.hh: Fixed up a typedef of MachInst cpu/o3/bpred_unit_impl.hh: Added a using TheISA::MachInst to a function cpu/o3/btb.hh: Changed an include from arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef for Addr cpu/o3/commit.hh: Removed a typedef of Impl::ISA as ISA, since TheISA takes care of this now. cpu/o3/cpu.cc: Cleaned up namespace issues cpu/o3/cpu.hh: Cleaned up namespace usage cpu/o3/decode.hh: Removed typedef of ISA, and changed it to TheISA cpu/o3/fetch.hh: Fized up typedefs, and changed ISA to TheISA cpu/o3/free_list.hh: Changed include of arch/alpha/isa_traits.hh to arch/isa_traits.hh cpu/o3/iew.hh: Removed typedef of ISA cpu/o3/iew_impl.hh: Added TheISA namespace specifier to MachInst cpu/o3/ras.hh: Changed include from arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef for Addr. cpu/o3/regfile.hh: Changed ISA to TheISA, and added some typedefs for Addr, IntReg, FloatReg, and MiscRegFile cpu/o3/rename.hh: Changed ISA to TheISA, and added a typedef for RegIndex cpu/o3/rename_map.hh: Added an include for arch/isa_traits.hh, and a typedef for RegIndex cpu/o3/rob.hh: Added a typedef for RegIndex cpu/o3/store_set.hh: cpu/o3/tournament_pred.hh: Changed an include of arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef of Addr cpu/ozone/cpu.hh: Changed ISA into TheISA, and untemplatized StaticInst cpu/pc_event.cc: Added namespace specifier TheISA to Addr types cpu/profile.hh: kern/kernel_stats.hh: Added typedef for Addr, and untemplatized StaticInstPtr cpu/simple/cpu.cc: Changed using directive from LittleEndianGuest to AlphaISA, which will contain both namespaces. Added TheISA where needed, and untemplatized StaticInst cpu/simple/cpu.hh: Added a typedef for MachInst, and untemplatized StaticInst cpu/static_inst.cc: Untemplatized StaticInst cpu/static_inst.hh: Untemplatized StaticInst by using the TheISA namespace dev/alpha_console.cc: Added using namespace AlphaISA dev/simple_disk.hh: Added typedef for Addr and fixed up some formatting dev/sinicreg.hh: Added TheISA namespace specifier where needed dev/tsunami.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Added using namespace TheISA. It might be better for it to be AlphaISA dev/tsunami_cchip.cc: Added typedef for TheISA. It might be better for it to be AlphaISA kern/linux/aligned.hh: sim/pseudo_inst.hh: Added TheISA namespace specifier to Addr kern/linux/linux_threadinfo.hh: Added typedef for Addr, and TheISA namespace specifier to StackPointerReg kern/tru64/mbuf.hh: Added TheISA to Addr type in structs sim/process.hh: Added typedefs of Addr, RegFile, and MachInst sim/syscall_emul.cc: Added using namespace TheISA, and a cast of VMPageSize to the int type sim/syscall_emul.hh: Added typecast for Addr, and TheISA namespace specifier for where needed --HG-- extra : convert_revision : 91d4f6ca33a73b21c1f1771d74bfdea3b80eff45
2006-02-19 08:34:37 +01:00
const StaticInstPtr eaCompPtr;
/// Pointer to MemAcc object.
Changes to untemplate StaticInst and StaticInstPtr, change the isa to a namespace instead of a class, an improvement to the architecture specific header file selection system, and fixed up a few include paths. arch/alpha/alpha_linux_process.cc: Added using directive for AlphaISA namespace arch/alpha/alpha_memory.hh: arch/alpha/isa/branch.isa: cpu/pc_event.hh: Added typedefs for Addr arch/alpha/alpha_tru64_process.cc: arch/alpha/arguments.cc: Added using directive for AlphaISA arch/alpha/ev5.hh: Added an include of arch/alpha/isa_traits.hh, and a using directive for the AlphaISA namespace. arch/alpha/faults.hh: Added a typedef for the Addr type, and changed the formatting of the faults slightly. arch/alpha/isa/main.isa: Untemplatized StaticInst, added a using for namespace AlphaISA to show up in decoder.cc and the exec.ccs, relocated makeNop to decoder.hh arch/alpha/isa/mem.isa: Untemplatized StaticInst and StaticInstPtr arch/alpha/isa/pal.isa: cpu/base_dyn_inst.cc: Untemplatized StaticInstPtr arch/alpha/isa_traits.hh: Changed variables to be externs instead of static since they are part of a namespace and not a class. arch/alpha/stacktrace.cc: Untemplatized StaticInstPtr, and added a using directive for AlphaISA. arch/alpha/stacktrace.hh: Added some typedefs for Addr and MachInst, and untemplatized StaticInstPtr arch/alpha/vtophys.cc: Added a using directive for AlphaISA arch/alpha/vtophys.hh: Added the AlphaISA namespace specifier where needed arch/isa_parser.py: Changed the placement of the definition of the decodeInst function to be outside the namespaceInst namespace. base/loader/object_file.hh: cpu/o3/bpred_unit.hh: Added a typedef for Addr base/loader/symtab.hh: Added a typedef for Addr, and added a TheISA to Addr in another typedef base/remote_gdb.cc: Added a using namespace TheISA, and untemplatized StaticInstPtr base/remote_gdb.hh: Added typedefs for Addr and MachInst cpu/base.cc: Added TheISA specifier to some variables exported from the isa. cpu/base.hh: Added a typedef for Addr, and TheISA to some variables from the ISA cpu/base_dyn_inst.hh: Untemplatized StaticInstPtr, and added TheISA specifier to some variables from the ISA. cpu/exec_context.hh: Added some typedefs for types from the isa, and added TheISA specifier to some variables from the isa cpu/exetrace.hh: Added typedefs for some types from the ISA, and untemplatized StaticInstPtr cpu/memtest/memtest.cc: cpu/o3/btb.cc: dev/baddev.cc: dev/ide_ctrl.cc: dev/ide_disk.cc: dev/isa_fake.cc: dev/ns_gige.cc: dev/pciconfigall.cc: dev/platform.cc: dev/sinic.cc: dev/uart8250.cc: kern/freebsd/freebsd_system.cc: kern/linux/linux_system.cc: kern/system_events.cc: kern/tru64/dump_mbuf.cc: kern/tru64/tru64_events.cc: sim/process.cc: sim/pseudo_inst.cc: sim/system.cc: Added using namespace TheISA cpu/memtest/memtest.hh: cpu/trace/opt_cpu.hh: cpu/trace/reader/itx_reader.hh: dev/ide_disk.hh: dev/pcidev.hh: dev/platform.hh: dev/tsunami.hh: sim/system.hh: sim/vptr.hh: Added typedef for Addr cpu/o3/2bit_local_pred.hh: Changed the include to use arch/isa_traits.hh instead of arch/alpha/isa_traits.hh. Added typedef for Addr cpu/o3/alpha_cpu.hh: Added typedefs for Addr and IntReg cpu/o3/alpha_cpu_impl.hh: Added this-> to setNextPC to fix a problem since it didn't depend on template parameters any more. Removed "typename" where it was no longer needed. cpu/o3/alpha_dyn_inst.hh: Cleaned up some typedefs, and untemplatized StaticInst cpu/o3/alpha_dyn_inst_impl.hh: untemplatized StaticInstPtr cpu/o3/alpha_impl.hh: Fixed up a typedef of MachInst cpu/o3/bpred_unit_impl.hh: Added a using TheISA::MachInst to a function cpu/o3/btb.hh: Changed an include from arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef for Addr cpu/o3/commit.hh: Removed a typedef of Impl::ISA as ISA, since TheISA takes care of this now. cpu/o3/cpu.cc: Cleaned up namespace issues cpu/o3/cpu.hh: Cleaned up namespace usage cpu/o3/decode.hh: Removed typedef of ISA, and changed it to TheISA cpu/o3/fetch.hh: Fized up typedefs, and changed ISA to TheISA cpu/o3/free_list.hh: Changed include of arch/alpha/isa_traits.hh to arch/isa_traits.hh cpu/o3/iew.hh: Removed typedef of ISA cpu/o3/iew_impl.hh: Added TheISA namespace specifier to MachInst cpu/o3/ras.hh: Changed include from arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef for Addr. cpu/o3/regfile.hh: Changed ISA to TheISA, and added some typedefs for Addr, IntReg, FloatReg, and MiscRegFile cpu/o3/rename.hh: Changed ISA to TheISA, and added a typedef for RegIndex cpu/o3/rename_map.hh: Added an include for arch/isa_traits.hh, and a typedef for RegIndex cpu/o3/rob.hh: Added a typedef for RegIndex cpu/o3/store_set.hh: cpu/o3/tournament_pred.hh: Changed an include of arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef of Addr cpu/ozone/cpu.hh: Changed ISA into TheISA, and untemplatized StaticInst cpu/pc_event.cc: Added namespace specifier TheISA to Addr types cpu/profile.hh: kern/kernel_stats.hh: Added typedef for Addr, and untemplatized StaticInstPtr cpu/simple/cpu.cc: Changed using directive from LittleEndianGuest to AlphaISA, which will contain both namespaces. Added TheISA where needed, and untemplatized StaticInst cpu/simple/cpu.hh: Added a typedef for MachInst, and untemplatized StaticInst cpu/static_inst.cc: Untemplatized StaticInst cpu/static_inst.hh: Untemplatized StaticInst by using the TheISA namespace dev/alpha_console.cc: Added using namespace AlphaISA dev/simple_disk.hh: Added typedef for Addr and fixed up some formatting dev/sinicreg.hh: Added TheISA namespace specifier where needed dev/tsunami.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Added using namespace TheISA. It might be better for it to be AlphaISA dev/tsunami_cchip.cc: Added typedef for TheISA. It might be better for it to be AlphaISA kern/linux/aligned.hh: sim/pseudo_inst.hh: Added TheISA namespace specifier to Addr kern/linux/linux_threadinfo.hh: Added typedef for Addr, and TheISA namespace specifier to StackPointerReg kern/tru64/mbuf.hh: Added TheISA to Addr type in structs sim/process.hh: Added typedefs of Addr, RegFile, and MachInst sim/syscall_emul.cc: Added using namespace TheISA, and a cast of VMPageSize to the int type sim/syscall_emul.hh: Added typecast for Addr, and TheISA namespace specifier for where needed --HG-- extra : convert_revision : 91d4f6ca33a73b21c1f1771d74bfdea3b80eff45
2006-02-19 08:34:37 +01:00
const StaticInstPtr memAccPtr;
/// Constructor
Changes to support automatic renaming of the shadow registers at decode time. This requires using an ExtMachInst (uint64_t) instead of the normal MachInst; the ExtMachInst is packed with extra decode context information. In the case of Alpha, the PAL mode is included. The shadow registers are folded into the normal integer registers to ease renaming indexing. Include the removed Opcdec class of instructions for faulting when a pal mode only instruction is decoded in non-pal mode. arch/alpha/ev5.cc: Changes to automatically map the shadow registers if the instruction is in PAL mode. arch/alpha/isa/branch.isa: arch/alpha/isa/decoder.isa: arch/alpha/isa/fp.isa: arch/alpha/isa/int.isa: arch/alpha/isa/mem.isa: arch/alpha/isa/pal.isa: arch/alpha/isa/unimp.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. arch/alpha/isa/main.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. The decoder (for Alpha) uses the 32nd bit in order to determine if the machine is in PAL mode. If it is, then it refers to the reg_redir table to determine the true index of the register it is using. Also include the opcdec instruction definition. arch/alpha/isa_traits.hh: Define ExtMachInst type that is used by the static inst in order to decode the instruction, given the context of being in pal mode or not. Redefine the number of Int registers, splitting it into NumIntArchRegs (32) and NumIntRegs (32 + 8 shadow registers). Change the dependence tags to reflect the integer registers include the 8 shadow registers. Define function to make an ExtMachInst. Currently it is somewhat specific to Alpha; in the future it must be decided to make this more generic and possibly slower, or leave it specific to each architecture and ifdef it within the CPU. arch/isa_parser.py: Have static insts decode on the ExtMachInst. base/remote_gdb.cc: Support the automatic remapping of shadow registers. Remote GDB must now look at the PC being read in order to tell if it should use the normal register indices or the shadow register indices. cpu/o3/regfile.hh: Comment out the pal registers; they are now a part of the integer registers. cpu/simple/cpu.cc: Create an ExtMachInst to decode on, based on the normal MachInst and the PC of the instructoin. cpu/static_inst.hh: Change from MachInst to ExtMachInst to support shadow register renaming. --HG-- extra : convert_revision : 1d23eabf735e297068e1917445a6348e9f8c88d5
2006-03-03 21:28:25 +01:00
Memory(const char *mnem, ExtMachInst _machInst, OpClass __opClass,
Changes to untemplate StaticInst and StaticInstPtr, change the isa to a namespace instead of a class, an improvement to the architecture specific header file selection system, and fixed up a few include paths. arch/alpha/alpha_linux_process.cc: Added using directive for AlphaISA namespace arch/alpha/alpha_memory.hh: arch/alpha/isa/branch.isa: cpu/pc_event.hh: Added typedefs for Addr arch/alpha/alpha_tru64_process.cc: arch/alpha/arguments.cc: Added using directive for AlphaISA arch/alpha/ev5.hh: Added an include of arch/alpha/isa_traits.hh, and a using directive for the AlphaISA namespace. arch/alpha/faults.hh: Added a typedef for the Addr type, and changed the formatting of the faults slightly. arch/alpha/isa/main.isa: Untemplatized StaticInst, added a using for namespace AlphaISA to show up in decoder.cc and the exec.ccs, relocated makeNop to decoder.hh arch/alpha/isa/mem.isa: Untemplatized StaticInst and StaticInstPtr arch/alpha/isa/pal.isa: cpu/base_dyn_inst.cc: Untemplatized StaticInstPtr arch/alpha/isa_traits.hh: Changed variables to be externs instead of static since they are part of a namespace and not a class. arch/alpha/stacktrace.cc: Untemplatized StaticInstPtr, and added a using directive for AlphaISA. arch/alpha/stacktrace.hh: Added some typedefs for Addr and MachInst, and untemplatized StaticInstPtr arch/alpha/vtophys.cc: Added a using directive for AlphaISA arch/alpha/vtophys.hh: Added the AlphaISA namespace specifier where needed arch/isa_parser.py: Changed the placement of the definition of the decodeInst function to be outside the namespaceInst namespace. base/loader/object_file.hh: cpu/o3/bpred_unit.hh: Added a typedef for Addr base/loader/symtab.hh: Added a typedef for Addr, and added a TheISA to Addr in another typedef base/remote_gdb.cc: Added a using namespace TheISA, and untemplatized StaticInstPtr base/remote_gdb.hh: Added typedefs for Addr and MachInst cpu/base.cc: Added TheISA specifier to some variables exported from the isa. cpu/base.hh: Added a typedef for Addr, and TheISA to some variables from the ISA cpu/base_dyn_inst.hh: Untemplatized StaticInstPtr, and added TheISA specifier to some variables from the ISA. cpu/exec_context.hh: Added some typedefs for types from the isa, and added TheISA specifier to some variables from the isa cpu/exetrace.hh: Added typedefs for some types from the ISA, and untemplatized StaticInstPtr cpu/memtest/memtest.cc: cpu/o3/btb.cc: dev/baddev.cc: dev/ide_ctrl.cc: dev/ide_disk.cc: dev/isa_fake.cc: dev/ns_gige.cc: dev/pciconfigall.cc: dev/platform.cc: dev/sinic.cc: dev/uart8250.cc: kern/freebsd/freebsd_system.cc: kern/linux/linux_system.cc: kern/system_events.cc: kern/tru64/dump_mbuf.cc: kern/tru64/tru64_events.cc: sim/process.cc: sim/pseudo_inst.cc: sim/system.cc: Added using namespace TheISA cpu/memtest/memtest.hh: cpu/trace/opt_cpu.hh: cpu/trace/reader/itx_reader.hh: dev/ide_disk.hh: dev/pcidev.hh: dev/platform.hh: dev/tsunami.hh: sim/system.hh: sim/vptr.hh: Added typedef for Addr cpu/o3/2bit_local_pred.hh: Changed the include to use arch/isa_traits.hh instead of arch/alpha/isa_traits.hh. Added typedef for Addr cpu/o3/alpha_cpu.hh: Added typedefs for Addr and IntReg cpu/o3/alpha_cpu_impl.hh: Added this-> to setNextPC to fix a problem since it didn't depend on template parameters any more. Removed "typename" where it was no longer needed. cpu/o3/alpha_dyn_inst.hh: Cleaned up some typedefs, and untemplatized StaticInst cpu/o3/alpha_dyn_inst_impl.hh: untemplatized StaticInstPtr cpu/o3/alpha_impl.hh: Fixed up a typedef of MachInst cpu/o3/bpred_unit_impl.hh: Added a using TheISA::MachInst to a function cpu/o3/btb.hh: Changed an include from arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef for Addr cpu/o3/commit.hh: Removed a typedef of Impl::ISA as ISA, since TheISA takes care of this now. cpu/o3/cpu.cc: Cleaned up namespace issues cpu/o3/cpu.hh: Cleaned up namespace usage cpu/o3/decode.hh: Removed typedef of ISA, and changed it to TheISA cpu/o3/fetch.hh: Fized up typedefs, and changed ISA to TheISA cpu/o3/free_list.hh: Changed include of arch/alpha/isa_traits.hh to arch/isa_traits.hh cpu/o3/iew.hh: Removed typedef of ISA cpu/o3/iew_impl.hh: Added TheISA namespace specifier to MachInst cpu/o3/ras.hh: Changed include from arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef for Addr. cpu/o3/regfile.hh: Changed ISA to TheISA, and added some typedefs for Addr, IntReg, FloatReg, and MiscRegFile cpu/o3/rename.hh: Changed ISA to TheISA, and added a typedef for RegIndex cpu/o3/rename_map.hh: Added an include for arch/isa_traits.hh, and a typedef for RegIndex cpu/o3/rob.hh: Added a typedef for RegIndex cpu/o3/store_set.hh: cpu/o3/tournament_pred.hh: Changed an include of arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef of Addr cpu/ozone/cpu.hh: Changed ISA into TheISA, and untemplatized StaticInst cpu/pc_event.cc: Added namespace specifier TheISA to Addr types cpu/profile.hh: kern/kernel_stats.hh: Added typedef for Addr, and untemplatized StaticInstPtr cpu/simple/cpu.cc: Changed using directive from LittleEndianGuest to AlphaISA, which will contain both namespaces. Added TheISA where needed, and untemplatized StaticInst cpu/simple/cpu.hh: Added a typedef for MachInst, and untemplatized StaticInst cpu/static_inst.cc: Untemplatized StaticInst cpu/static_inst.hh: Untemplatized StaticInst by using the TheISA namespace dev/alpha_console.cc: Added using namespace AlphaISA dev/simple_disk.hh: Added typedef for Addr and fixed up some formatting dev/sinicreg.hh: Added TheISA namespace specifier where needed dev/tsunami.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Added using namespace TheISA. It might be better for it to be AlphaISA dev/tsunami_cchip.cc: Added typedef for TheISA. It might be better for it to be AlphaISA kern/linux/aligned.hh: sim/pseudo_inst.hh: Added TheISA namespace specifier to Addr kern/linux/linux_threadinfo.hh: Added typedef for Addr, and TheISA namespace specifier to StackPointerReg kern/tru64/mbuf.hh: Added TheISA to Addr type in structs sim/process.hh: Added typedefs of Addr, RegFile, and MachInst sim/syscall_emul.cc: Added using namespace TheISA, and a cast of VMPageSize to the int type sim/syscall_emul.hh: Added typecast for Addr, and TheISA namespace specifier for where needed --HG-- extra : convert_revision : 91d4f6ca33a73b21c1f1771d74bfdea3b80eff45
2006-02-19 08:34:37 +01:00
StaticInstPtr _eaCompPtr = nullStaticInstPtr,
StaticInstPtr _memAccPtr = nullStaticInstPtr)
: AlphaStaticInst(mnem, _machInst, __opClass),
memAccessFlags(0), eaCompPtr(_eaCompPtr), memAccPtr(_memAccPtr)
{
}
std::string
generateDisassembly(Addr pc, const SymbolTable *symtab) const;
public:
Changes to untemplate StaticInst and StaticInstPtr, change the isa to a namespace instead of a class, an improvement to the architecture specific header file selection system, and fixed up a few include paths. arch/alpha/alpha_linux_process.cc: Added using directive for AlphaISA namespace arch/alpha/alpha_memory.hh: arch/alpha/isa/branch.isa: cpu/pc_event.hh: Added typedefs for Addr arch/alpha/alpha_tru64_process.cc: arch/alpha/arguments.cc: Added using directive for AlphaISA arch/alpha/ev5.hh: Added an include of arch/alpha/isa_traits.hh, and a using directive for the AlphaISA namespace. arch/alpha/faults.hh: Added a typedef for the Addr type, and changed the formatting of the faults slightly. arch/alpha/isa/main.isa: Untemplatized StaticInst, added a using for namespace AlphaISA to show up in decoder.cc and the exec.ccs, relocated makeNop to decoder.hh arch/alpha/isa/mem.isa: Untemplatized StaticInst and StaticInstPtr arch/alpha/isa/pal.isa: cpu/base_dyn_inst.cc: Untemplatized StaticInstPtr arch/alpha/isa_traits.hh: Changed variables to be externs instead of static since they are part of a namespace and not a class. arch/alpha/stacktrace.cc: Untemplatized StaticInstPtr, and added a using directive for AlphaISA. arch/alpha/stacktrace.hh: Added some typedefs for Addr and MachInst, and untemplatized StaticInstPtr arch/alpha/vtophys.cc: Added a using directive for AlphaISA arch/alpha/vtophys.hh: Added the AlphaISA namespace specifier where needed arch/isa_parser.py: Changed the placement of the definition of the decodeInst function to be outside the namespaceInst namespace. base/loader/object_file.hh: cpu/o3/bpred_unit.hh: Added a typedef for Addr base/loader/symtab.hh: Added a typedef for Addr, and added a TheISA to Addr in another typedef base/remote_gdb.cc: Added a using namespace TheISA, and untemplatized StaticInstPtr base/remote_gdb.hh: Added typedefs for Addr and MachInst cpu/base.cc: Added TheISA specifier to some variables exported from the isa. cpu/base.hh: Added a typedef for Addr, and TheISA to some variables from the ISA cpu/base_dyn_inst.hh: Untemplatized StaticInstPtr, and added TheISA specifier to some variables from the ISA. cpu/exec_context.hh: Added some typedefs for types from the isa, and added TheISA specifier to some variables from the isa cpu/exetrace.hh: Added typedefs for some types from the ISA, and untemplatized StaticInstPtr cpu/memtest/memtest.cc: cpu/o3/btb.cc: dev/baddev.cc: dev/ide_ctrl.cc: dev/ide_disk.cc: dev/isa_fake.cc: dev/ns_gige.cc: dev/pciconfigall.cc: dev/platform.cc: dev/sinic.cc: dev/uart8250.cc: kern/freebsd/freebsd_system.cc: kern/linux/linux_system.cc: kern/system_events.cc: kern/tru64/dump_mbuf.cc: kern/tru64/tru64_events.cc: sim/process.cc: sim/pseudo_inst.cc: sim/system.cc: Added using namespace TheISA cpu/memtest/memtest.hh: cpu/trace/opt_cpu.hh: cpu/trace/reader/itx_reader.hh: dev/ide_disk.hh: dev/pcidev.hh: dev/platform.hh: dev/tsunami.hh: sim/system.hh: sim/vptr.hh: Added typedef for Addr cpu/o3/2bit_local_pred.hh: Changed the include to use arch/isa_traits.hh instead of arch/alpha/isa_traits.hh. Added typedef for Addr cpu/o3/alpha_cpu.hh: Added typedefs for Addr and IntReg cpu/o3/alpha_cpu_impl.hh: Added this-> to setNextPC to fix a problem since it didn't depend on template parameters any more. Removed "typename" where it was no longer needed. cpu/o3/alpha_dyn_inst.hh: Cleaned up some typedefs, and untemplatized StaticInst cpu/o3/alpha_dyn_inst_impl.hh: untemplatized StaticInstPtr cpu/o3/alpha_impl.hh: Fixed up a typedef of MachInst cpu/o3/bpred_unit_impl.hh: Added a using TheISA::MachInst to a function cpu/o3/btb.hh: Changed an include from arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef for Addr cpu/o3/commit.hh: Removed a typedef of Impl::ISA as ISA, since TheISA takes care of this now. cpu/o3/cpu.cc: Cleaned up namespace issues cpu/o3/cpu.hh: Cleaned up namespace usage cpu/o3/decode.hh: Removed typedef of ISA, and changed it to TheISA cpu/o3/fetch.hh: Fized up typedefs, and changed ISA to TheISA cpu/o3/free_list.hh: Changed include of arch/alpha/isa_traits.hh to arch/isa_traits.hh cpu/o3/iew.hh: Removed typedef of ISA cpu/o3/iew_impl.hh: Added TheISA namespace specifier to MachInst cpu/o3/ras.hh: Changed include from arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef for Addr. cpu/o3/regfile.hh: Changed ISA to TheISA, and added some typedefs for Addr, IntReg, FloatReg, and MiscRegFile cpu/o3/rename.hh: Changed ISA to TheISA, and added a typedef for RegIndex cpu/o3/rename_map.hh: Added an include for arch/isa_traits.hh, and a typedef for RegIndex cpu/o3/rob.hh: Added a typedef for RegIndex cpu/o3/store_set.hh: cpu/o3/tournament_pred.hh: Changed an include of arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef of Addr cpu/ozone/cpu.hh: Changed ISA into TheISA, and untemplatized StaticInst cpu/pc_event.cc: Added namespace specifier TheISA to Addr types cpu/profile.hh: kern/kernel_stats.hh: Added typedef for Addr, and untemplatized StaticInstPtr cpu/simple/cpu.cc: Changed using directive from LittleEndianGuest to AlphaISA, which will contain both namespaces. Added TheISA where needed, and untemplatized StaticInst cpu/simple/cpu.hh: Added a typedef for MachInst, and untemplatized StaticInst cpu/static_inst.cc: Untemplatized StaticInst cpu/static_inst.hh: Untemplatized StaticInst by using the TheISA namespace dev/alpha_console.cc: Added using namespace AlphaISA dev/simple_disk.hh: Added typedef for Addr and fixed up some formatting dev/sinicreg.hh: Added TheISA namespace specifier where needed dev/tsunami.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Added using namespace TheISA. It might be better for it to be AlphaISA dev/tsunami_cchip.cc: Added typedef for TheISA. It might be better for it to be AlphaISA kern/linux/aligned.hh: sim/pseudo_inst.hh: Added TheISA namespace specifier to Addr kern/linux/linux_threadinfo.hh: Added typedef for Addr, and TheISA namespace specifier to StackPointerReg kern/tru64/mbuf.hh: Added TheISA to Addr type in structs sim/process.hh: Added typedefs of Addr, RegFile, and MachInst sim/syscall_emul.cc: Added using namespace TheISA, and a cast of VMPageSize to the int type sim/syscall_emul.hh: Added typecast for Addr, and TheISA namespace specifier for where needed --HG-- extra : convert_revision : 91d4f6ca33a73b21c1f1771d74bfdea3b80eff45
2006-02-19 08:34:37 +01:00
const StaticInstPtr &eaCompInst() const { return eaCompPtr; }
const StaticInstPtr &memAccInst() const { return memAccPtr; }
};
/**
* Base class for memory-format instructions using a 32-bit
* displacement (i.e. most of them).
*/
class MemoryDisp32 : public Memory
{
protected:
/// Displacement for EA calculation (signed).
int32_t disp;
/// Constructor.
Changes to support automatic renaming of the shadow registers at decode time. This requires using an ExtMachInst (uint64_t) instead of the normal MachInst; the ExtMachInst is packed with extra decode context information. In the case of Alpha, the PAL mode is included. The shadow registers are folded into the normal integer registers to ease renaming indexing. Include the removed Opcdec class of instructions for faulting when a pal mode only instruction is decoded in non-pal mode. arch/alpha/ev5.cc: Changes to automatically map the shadow registers if the instruction is in PAL mode. arch/alpha/isa/branch.isa: arch/alpha/isa/decoder.isa: arch/alpha/isa/fp.isa: arch/alpha/isa/int.isa: arch/alpha/isa/mem.isa: arch/alpha/isa/pal.isa: arch/alpha/isa/unimp.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. arch/alpha/isa/main.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. The decoder (for Alpha) uses the 32nd bit in order to determine if the machine is in PAL mode. If it is, then it refers to the reg_redir table to determine the true index of the register it is using. Also include the opcdec instruction definition. arch/alpha/isa_traits.hh: Define ExtMachInst type that is used by the static inst in order to decode the instruction, given the context of being in pal mode or not. Redefine the number of Int registers, splitting it into NumIntArchRegs (32) and NumIntRegs (32 + 8 shadow registers). Change the dependence tags to reflect the integer registers include the 8 shadow registers. Define function to make an ExtMachInst. Currently it is somewhat specific to Alpha; in the future it must be decided to make this more generic and possibly slower, or leave it specific to each architecture and ifdef it within the CPU. arch/isa_parser.py: Have static insts decode on the ExtMachInst. base/remote_gdb.cc: Support the automatic remapping of shadow registers. Remote GDB must now look at the PC being read in order to tell if it should use the normal register indices or the shadow register indices. cpu/o3/regfile.hh: Comment out the pal registers; they are now a part of the integer registers. cpu/simple/cpu.cc: Create an ExtMachInst to decode on, based on the normal MachInst and the PC of the instructoin. cpu/static_inst.hh: Change from MachInst to ExtMachInst to support shadow register renaming. --HG-- extra : convert_revision : 1d23eabf735e297068e1917445a6348e9f8c88d5
2006-03-03 21:28:25 +01:00
MemoryDisp32(const char *mnem, ExtMachInst _machInst, OpClass __opClass,
Changes to untemplate StaticInst and StaticInstPtr, change the isa to a namespace instead of a class, an improvement to the architecture specific header file selection system, and fixed up a few include paths. arch/alpha/alpha_linux_process.cc: Added using directive for AlphaISA namespace arch/alpha/alpha_memory.hh: arch/alpha/isa/branch.isa: cpu/pc_event.hh: Added typedefs for Addr arch/alpha/alpha_tru64_process.cc: arch/alpha/arguments.cc: Added using directive for AlphaISA arch/alpha/ev5.hh: Added an include of arch/alpha/isa_traits.hh, and a using directive for the AlphaISA namespace. arch/alpha/faults.hh: Added a typedef for the Addr type, and changed the formatting of the faults slightly. arch/alpha/isa/main.isa: Untemplatized StaticInst, added a using for namespace AlphaISA to show up in decoder.cc and the exec.ccs, relocated makeNop to decoder.hh arch/alpha/isa/mem.isa: Untemplatized StaticInst and StaticInstPtr arch/alpha/isa/pal.isa: cpu/base_dyn_inst.cc: Untemplatized StaticInstPtr arch/alpha/isa_traits.hh: Changed variables to be externs instead of static since they are part of a namespace and not a class. arch/alpha/stacktrace.cc: Untemplatized StaticInstPtr, and added a using directive for AlphaISA. arch/alpha/stacktrace.hh: Added some typedefs for Addr and MachInst, and untemplatized StaticInstPtr arch/alpha/vtophys.cc: Added a using directive for AlphaISA arch/alpha/vtophys.hh: Added the AlphaISA namespace specifier where needed arch/isa_parser.py: Changed the placement of the definition of the decodeInst function to be outside the namespaceInst namespace. base/loader/object_file.hh: cpu/o3/bpred_unit.hh: Added a typedef for Addr base/loader/symtab.hh: Added a typedef for Addr, and added a TheISA to Addr in another typedef base/remote_gdb.cc: Added a using namespace TheISA, and untemplatized StaticInstPtr base/remote_gdb.hh: Added typedefs for Addr and MachInst cpu/base.cc: Added TheISA specifier to some variables exported from the isa. cpu/base.hh: Added a typedef for Addr, and TheISA to some variables from the ISA cpu/base_dyn_inst.hh: Untemplatized StaticInstPtr, and added TheISA specifier to some variables from the ISA. cpu/exec_context.hh: Added some typedefs for types from the isa, and added TheISA specifier to some variables from the isa cpu/exetrace.hh: Added typedefs for some types from the ISA, and untemplatized StaticInstPtr cpu/memtest/memtest.cc: cpu/o3/btb.cc: dev/baddev.cc: dev/ide_ctrl.cc: dev/ide_disk.cc: dev/isa_fake.cc: dev/ns_gige.cc: dev/pciconfigall.cc: dev/platform.cc: dev/sinic.cc: dev/uart8250.cc: kern/freebsd/freebsd_system.cc: kern/linux/linux_system.cc: kern/system_events.cc: kern/tru64/dump_mbuf.cc: kern/tru64/tru64_events.cc: sim/process.cc: sim/pseudo_inst.cc: sim/system.cc: Added using namespace TheISA cpu/memtest/memtest.hh: cpu/trace/opt_cpu.hh: cpu/trace/reader/itx_reader.hh: dev/ide_disk.hh: dev/pcidev.hh: dev/platform.hh: dev/tsunami.hh: sim/system.hh: sim/vptr.hh: Added typedef for Addr cpu/o3/2bit_local_pred.hh: Changed the include to use arch/isa_traits.hh instead of arch/alpha/isa_traits.hh. Added typedef for Addr cpu/o3/alpha_cpu.hh: Added typedefs for Addr and IntReg cpu/o3/alpha_cpu_impl.hh: Added this-> to setNextPC to fix a problem since it didn't depend on template parameters any more. Removed "typename" where it was no longer needed. cpu/o3/alpha_dyn_inst.hh: Cleaned up some typedefs, and untemplatized StaticInst cpu/o3/alpha_dyn_inst_impl.hh: untemplatized StaticInstPtr cpu/o3/alpha_impl.hh: Fixed up a typedef of MachInst cpu/o3/bpred_unit_impl.hh: Added a using TheISA::MachInst to a function cpu/o3/btb.hh: Changed an include from arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef for Addr cpu/o3/commit.hh: Removed a typedef of Impl::ISA as ISA, since TheISA takes care of this now. cpu/o3/cpu.cc: Cleaned up namespace issues cpu/o3/cpu.hh: Cleaned up namespace usage cpu/o3/decode.hh: Removed typedef of ISA, and changed it to TheISA cpu/o3/fetch.hh: Fized up typedefs, and changed ISA to TheISA cpu/o3/free_list.hh: Changed include of arch/alpha/isa_traits.hh to arch/isa_traits.hh cpu/o3/iew.hh: Removed typedef of ISA cpu/o3/iew_impl.hh: Added TheISA namespace specifier to MachInst cpu/o3/ras.hh: Changed include from arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef for Addr. cpu/o3/regfile.hh: Changed ISA to TheISA, and added some typedefs for Addr, IntReg, FloatReg, and MiscRegFile cpu/o3/rename.hh: Changed ISA to TheISA, and added a typedef for RegIndex cpu/o3/rename_map.hh: Added an include for arch/isa_traits.hh, and a typedef for RegIndex cpu/o3/rob.hh: Added a typedef for RegIndex cpu/o3/store_set.hh: cpu/o3/tournament_pred.hh: Changed an include of arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef of Addr cpu/ozone/cpu.hh: Changed ISA into TheISA, and untemplatized StaticInst cpu/pc_event.cc: Added namespace specifier TheISA to Addr types cpu/profile.hh: kern/kernel_stats.hh: Added typedef for Addr, and untemplatized StaticInstPtr cpu/simple/cpu.cc: Changed using directive from LittleEndianGuest to AlphaISA, which will contain both namespaces. Added TheISA where needed, and untemplatized StaticInst cpu/simple/cpu.hh: Added a typedef for MachInst, and untemplatized StaticInst cpu/static_inst.cc: Untemplatized StaticInst cpu/static_inst.hh: Untemplatized StaticInst by using the TheISA namespace dev/alpha_console.cc: Added using namespace AlphaISA dev/simple_disk.hh: Added typedef for Addr and fixed up some formatting dev/sinicreg.hh: Added TheISA namespace specifier where needed dev/tsunami.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Added using namespace TheISA. It might be better for it to be AlphaISA dev/tsunami_cchip.cc: Added typedef for TheISA. It might be better for it to be AlphaISA kern/linux/aligned.hh: sim/pseudo_inst.hh: Added TheISA namespace specifier to Addr kern/linux/linux_threadinfo.hh: Added typedef for Addr, and TheISA namespace specifier to StackPointerReg kern/tru64/mbuf.hh: Added TheISA to Addr type in structs sim/process.hh: Added typedefs of Addr, RegFile, and MachInst sim/syscall_emul.cc: Added using namespace TheISA, and a cast of VMPageSize to the int type sim/syscall_emul.hh: Added typecast for Addr, and TheISA namespace specifier for where needed --HG-- extra : convert_revision : 91d4f6ca33a73b21c1f1771d74bfdea3b80eff45
2006-02-19 08:34:37 +01:00
StaticInstPtr _eaCompPtr = nullStaticInstPtr,
StaticInstPtr _memAccPtr = nullStaticInstPtr)
: Memory(mnem, _machInst, __opClass, _eaCompPtr, _memAccPtr),
disp(MEMDISP)
{
}
};
/**
* Base class for a few miscellaneous memory-format insts
* that don't interpret the disp field: wh64, fetch, fetch_m, ecb.
* None of these instructions has a destination register either.
*/
class MemoryNoDisp : public Memory
{
protected:
/// Constructor
Changes to support automatic renaming of the shadow registers at decode time. This requires using an ExtMachInst (uint64_t) instead of the normal MachInst; the ExtMachInst is packed with extra decode context information. In the case of Alpha, the PAL mode is included. The shadow registers are folded into the normal integer registers to ease renaming indexing. Include the removed Opcdec class of instructions for faulting when a pal mode only instruction is decoded in non-pal mode. arch/alpha/ev5.cc: Changes to automatically map the shadow registers if the instruction is in PAL mode. arch/alpha/isa/branch.isa: arch/alpha/isa/decoder.isa: arch/alpha/isa/fp.isa: arch/alpha/isa/int.isa: arch/alpha/isa/mem.isa: arch/alpha/isa/pal.isa: arch/alpha/isa/unimp.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. arch/alpha/isa/main.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. The decoder (for Alpha) uses the 32nd bit in order to determine if the machine is in PAL mode. If it is, then it refers to the reg_redir table to determine the true index of the register it is using. Also include the opcdec instruction definition. arch/alpha/isa_traits.hh: Define ExtMachInst type that is used by the static inst in order to decode the instruction, given the context of being in pal mode or not. Redefine the number of Int registers, splitting it into NumIntArchRegs (32) and NumIntRegs (32 + 8 shadow registers). Change the dependence tags to reflect the integer registers include the 8 shadow registers. Define function to make an ExtMachInst. Currently it is somewhat specific to Alpha; in the future it must be decided to make this more generic and possibly slower, or leave it specific to each architecture and ifdef it within the CPU. arch/isa_parser.py: Have static insts decode on the ExtMachInst. base/remote_gdb.cc: Support the automatic remapping of shadow registers. Remote GDB must now look at the PC being read in order to tell if it should use the normal register indices or the shadow register indices. cpu/o3/regfile.hh: Comment out the pal registers; they are now a part of the integer registers. cpu/simple/cpu.cc: Create an ExtMachInst to decode on, based on the normal MachInst and the PC of the instructoin. cpu/static_inst.hh: Change from MachInst to ExtMachInst to support shadow register renaming. --HG-- extra : convert_revision : 1d23eabf735e297068e1917445a6348e9f8c88d5
2006-03-03 21:28:25 +01:00
MemoryNoDisp(const char *mnem, ExtMachInst _machInst, OpClass __opClass,
Changes to untemplate StaticInst and StaticInstPtr, change the isa to a namespace instead of a class, an improvement to the architecture specific header file selection system, and fixed up a few include paths. arch/alpha/alpha_linux_process.cc: Added using directive for AlphaISA namespace arch/alpha/alpha_memory.hh: arch/alpha/isa/branch.isa: cpu/pc_event.hh: Added typedefs for Addr arch/alpha/alpha_tru64_process.cc: arch/alpha/arguments.cc: Added using directive for AlphaISA arch/alpha/ev5.hh: Added an include of arch/alpha/isa_traits.hh, and a using directive for the AlphaISA namespace. arch/alpha/faults.hh: Added a typedef for the Addr type, and changed the formatting of the faults slightly. arch/alpha/isa/main.isa: Untemplatized StaticInst, added a using for namespace AlphaISA to show up in decoder.cc and the exec.ccs, relocated makeNop to decoder.hh arch/alpha/isa/mem.isa: Untemplatized StaticInst and StaticInstPtr arch/alpha/isa/pal.isa: cpu/base_dyn_inst.cc: Untemplatized StaticInstPtr arch/alpha/isa_traits.hh: Changed variables to be externs instead of static since they are part of a namespace and not a class. arch/alpha/stacktrace.cc: Untemplatized StaticInstPtr, and added a using directive for AlphaISA. arch/alpha/stacktrace.hh: Added some typedefs for Addr and MachInst, and untemplatized StaticInstPtr arch/alpha/vtophys.cc: Added a using directive for AlphaISA arch/alpha/vtophys.hh: Added the AlphaISA namespace specifier where needed arch/isa_parser.py: Changed the placement of the definition of the decodeInst function to be outside the namespaceInst namespace. base/loader/object_file.hh: cpu/o3/bpred_unit.hh: Added a typedef for Addr base/loader/symtab.hh: Added a typedef for Addr, and added a TheISA to Addr in another typedef base/remote_gdb.cc: Added a using namespace TheISA, and untemplatized StaticInstPtr base/remote_gdb.hh: Added typedefs for Addr and MachInst cpu/base.cc: Added TheISA specifier to some variables exported from the isa. cpu/base.hh: Added a typedef for Addr, and TheISA to some variables from the ISA cpu/base_dyn_inst.hh: Untemplatized StaticInstPtr, and added TheISA specifier to some variables from the ISA. cpu/exec_context.hh: Added some typedefs for types from the isa, and added TheISA specifier to some variables from the isa cpu/exetrace.hh: Added typedefs for some types from the ISA, and untemplatized StaticInstPtr cpu/memtest/memtest.cc: cpu/o3/btb.cc: dev/baddev.cc: dev/ide_ctrl.cc: dev/ide_disk.cc: dev/isa_fake.cc: dev/ns_gige.cc: dev/pciconfigall.cc: dev/platform.cc: dev/sinic.cc: dev/uart8250.cc: kern/freebsd/freebsd_system.cc: kern/linux/linux_system.cc: kern/system_events.cc: kern/tru64/dump_mbuf.cc: kern/tru64/tru64_events.cc: sim/process.cc: sim/pseudo_inst.cc: sim/system.cc: Added using namespace TheISA cpu/memtest/memtest.hh: cpu/trace/opt_cpu.hh: cpu/trace/reader/itx_reader.hh: dev/ide_disk.hh: dev/pcidev.hh: dev/platform.hh: dev/tsunami.hh: sim/system.hh: sim/vptr.hh: Added typedef for Addr cpu/o3/2bit_local_pred.hh: Changed the include to use arch/isa_traits.hh instead of arch/alpha/isa_traits.hh. Added typedef for Addr cpu/o3/alpha_cpu.hh: Added typedefs for Addr and IntReg cpu/o3/alpha_cpu_impl.hh: Added this-> to setNextPC to fix a problem since it didn't depend on template parameters any more. Removed "typename" where it was no longer needed. cpu/o3/alpha_dyn_inst.hh: Cleaned up some typedefs, and untemplatized StaticInst cpu/o3/alpha_dyn_inst_impl.hh: untemplatized StaticInstPtr cpu/o3/alpha_impl.hh: Fixed up a typedef of MachInst cpu/o3/bpred_unit_impl.hh: Added a using TheISA::MachInst to a function cpu/o3/btb.hh: Changed an include from arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef for Addr cpu/o3/commit.hh: Removed a typedef of Impl::ISA as ISA, since TheISA takes care of this now. cpu/o3/cpu.cc: Cleaned up namespace issues cpu/o3/cpu.hh: Cleaned up namespace usage cpu/o3/decode.hh: Removed typedef of ISA, and changed it to TheISA cpu/o3/fetch.hh: Fized up typedefs, and changed ISA to TheISA cpu/o3/free_list.hh: Changed include of arch/alpha/isa_traits.hh to arch/isa_traits.hh cpu/o3/iew.hh: Removed typedef of ISA cpu/o3/iew_impl.hh: Added TheISA namespace specifier to MachInst cpu/o3/ras.hh: Changed include from arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef for Addr. cpu/o3/regfile.hh: Changed ISA to TheISA, and added some typedefs for Addr, IntReg, FloatReg, and MiscRegFile cpu/o3/rename.hh: Changed ISA to TheISA, and added a typedef for RegIndex cpu/o3/rename_map.hh: Added an include for arch/isa_traits.hh, and a typedef for RegIndex cpu/o3/rob.hh: Added a typedef for RegIndex cpu/o3/store_set.hh: cpu/o3/tournament_pred.hh: Changed an include of arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef of Addr cpu/ozone/cpu.hh: Changed ISA into TheISA, and untemplatized StaticInst cpu/pc_event.cc: Added namespace specifier TheISA to Addr types cpu/profile.hh: kern/kernel_stats.hh: Added typedef for Addr, and untemplatized StaticInstPtr cpu/simple/cpu.cc: Changed using directive from LittleEndianGuest to AlphaISA, which will contain both namespaces. Added TheISA where needed, and untemplatized StaticInst cpu/simple/cpu.hh: Added a typedef for MachInst, and untemplatized StaticInst cpu/static_inst.cc: Untemplatized StaticInst cpu/static_inst.hh: Untemplatized StaticInst by using the TheISA namespace dev/alpha_console.cc: Added using namespace AlphaISA dev/simple_disk.hh: Added typedef for Addr and fixed up some formatting dev/sinicreg.hh: Added TheISA namespace specifier where needed dev/tsunami.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Added using namespace TheISA. It might be better for it to be AlphaISA dev/tsunami_cchip.cc: Added typedef for TheISA. It might be better for it to be AlphaISA kern/linux/aligned.hh: sim/pseudo_inst.hh: Added TheISA namespace specifier to Addr kern/linux/linux_threadinfo.hh: Added typedef for Addr, and TheISA namespace specifier to StackPointerReg kern/tru64/mbuf.hh: Added TheISA to Addr type in structs sim/process.hh: Added typedefs of Addr, RegFile, and MachInst sim/syscall_emul.cc: Added using namespace TheISA, and a cast of VMPageSize to the int type sim/syscall_emul.hh: Added typecast for Addr, and TheISA namespace specifier for where needed --HG-- extra : convert_revision : 91d4f6ca33a73b21c1f1771d74bfdea3b80eff45
2006-02-19 08:34:37 +01:00
StaticInstPtr _eaCompPtr = nullStaticInstPtr,
StaticInstPtr _memAccPtr = nullStaticInstPtr)
: Memory(mnem, _machInst, __opClass, _eaCompPtr, _memAccPtr)
{
}
std::string
generateDisassembly(Addr pc, const SymbolTable *symtab) const;
};
}};
output decoder {{
std::string
Memory::generateDisassembly(Addr pc, const SymbolTable *symtab) const
{
return csprintf("%-10s %c%d,%d(r%d)", mnemonic,
flags[IsFloating] ? 'f' : 'r', RA, MEMDISP, RB);
}
std::string
MemoryNoDisp::generateDisassembly(Addr pc, const SymbolTable *symtab) const
{
return csprintf("%-10s (r%d)", mnemonic, RB);
}
}};
def format LoadAddress(code) {{
iop = InstObjParams(name, Name, 'MemoryDisp32', CodeBlock(code))
header_output = BasicDeclare.subst(iop)
decoder_output = BasicConstructor.subst(iop)
decode_block = BasicDecode.subst(iop)
exec_output = BasicExecute.subst(iop)
}};
def template LoadStoreDeclare {{
/**
* Static instruction class for "%(mnemonic)s".
*/
class %(class_name)s : public %(base_class)s
{
protected:
/**
* "Fake" effective address computation class for "%(mnemonic)s".
*/
class EAComp : public %(base_class)s
{
public:
/// Constructor
Changes to support automatic renaming of the shadow registers at decode time. This requires using an ExtMachInst (uint64_t) instead of the normal MachInst; the ExtMachInst is packed with extra decode context information. In the case of Alpha, the PAL mode is included. The shadow registers are folded into the normal integer registers to ease renaming indexing. Include the removed Opcdec class of instructions for faulting when a pal mode only instruction is decoded in non-pal mode. arch/alpha/ev5.cc: Changes to automatically map the shadow registers if the instruction is in PAL mode. arch/alpha/isa/branch.isa: arch/alpha/isa/decoder.isa: arch/alpha/isa/fp.isa: arch/alpha/isa/int.isa: arch/alpha/isa/mem.isa: arch/alpha/isa/pal.isa: arch/alpha/isa/unimp.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. arch/alpha/isa/main.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. The decoder (for Alpha) uses the 32nd bit in order to determine if the machine is in PAL mode. If it is, then it refers to the reg_redir table to determine the true index of the register it is using. Also include the opcdec instruction definition. arch/alpha/isa_traits.hh: Define ExtMachInst type that is used by the static inst in order to decode the instruction, given the context of being in pal mode or not. Redefine the number of Int registers, splitting it into NumIntArchRegs (32) and NumIntRegs (32 + 8 shadow registers). Change the dependence tags to reflect the integer registers include the 8 shadow registers. Define function to make an ExtMachInst. Currently it is somewhat specific to Alpha; in the future it must be decided to make this more generic and possibly slower, or leave it specific to each architecture and ifdef it within the CPU. arch/isa_parser.py: Have static insts decode on the ExtMachInst. base/remote_gdb.cc: Support the automatic remapping of shadow registers. Remote GDB must now look at the PC being read in order to tell if it should use the normal register indices or the shadow register indices. cpu/o3/regfile.hh: Comment out the pal registers; they are now a part of the integer registers. cpu/simple/cpu.cc: Create an ExtMachInst to decode on, based on the normal MachInst and the PC of the instructoin. cpu/static_inst.hh: Change from MachInst to ExtMachInst to support shadow register renaming. --HG-- extra : convert_revision : 1d23eabf735e297068e1917445a6348e9f8c88d5
2006-03-03 21:28:25 +01:00
EAComp(ExtMachInst machInst);
%(BasicExecDeclare)s
};
/**
* "Fake" memory access instruction class for "%(mnemonic)s".
*/
class MemAcc : public %(base_class)s
{
public:
/// Constructor
Changes to support automatic renaming of the shadow registers at decode time. This requires using an ExtMachInst (uint64_t) instead of the normal MachInst; the ExtMachInst is packed with extra decode context information. In the case of Alpha, the PAL mode is included. The shadow registers are folded into the normal integer registers to ease renaming indexing. Include the removed Opcdec class of instructions for faulting when a pal mode only instruction is decoded in non-pal mode. arch/alpha/ev5.cc: Changes to automatically map the shadow registers if the instruction is in PAL mode. arch/alpha/isa/branch.isa: arch/alpha/isa/decoder.isa: arch/alpha/isa/fp.isa: arch/alpha/isa/int.isa: arch/alpha/isa/mem.isa: arch/alpha/isa/pal.isa: arch/alpha/isa/unimp.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. arch/alpha/isa/main.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. The decoder (for Alpha) uses the 32nd bit in order to determine if the machine is in PAL mode. If it is, then it refers to the reg_redir table to determine the true index of the register it is using. Also include the opcdec instruction definition. arch/alpha/isa_traits.hh: Define ExtMachInst type that is used by the static inst in order to decode the instruction, given the context of being in pal mode or not. Redefine the number of Int registers, splitting it into NumIntArchRegs (32) and NumIntRegs (32 + 8 shadow registers). Change the dependence tags to reflect the integer registers include the 8 shadow registers. Define function to make an ExtMachInst. Currently it is somewhat specific to Alpha; in the future it must be decided to make this more generic and possibly slower, or leave it specific to each architecture and ifdef it within the CPU. arch/isa_parser.py: Have static insts decode on the ExtMachInst. base/remote_gdb.cc: Support the automatic remapping of shadow registers. Remote GDB must now look at the PC being read in order to tell if it should use the normal register indices or the shadow register indices. cpu/o3/regfile.hh: Comment out the pal registers; they are now a part of the integer registers. cpu/simple/cpu.cc: Create an ExtMachInst to decode on, based on the normal MachInst and the PC of the instructoin. cpu/static_inst.hh: Change from MachInst to ExtMachInst to support shadow register renaming. --HG-- extra : convert_revision : 1d23eabf735e297068e1917445a6348e9f8c88d5
2006-03-03 21:28:25 +01:00
MemAcc(ExtMachInst machInst);
%(BasicExecDeclare)s
};
public:
/// Constructor.
Changes to support automatic renaming of the shadow registers at decode time. This requires using an ExtMachInst (uint64_t) instead of the normal MachInst; the ExtMachInst is packed with extra decode context information. In the case of Alpha, the PAL mode is included. The shadow registers are folded into the normal integer registers to ease renaming indexing. Include the removed Opcdec class of instructions for faulting when a pal mode only instruction is decoded in non-pal mode. arch/alpha/ev5.cc: Changes to automatically map the shadow registers if the instruction is in PAL mode. arch/alpha/isa/branch.isa: arch/alpha/isa/decoder.isa: arch/alpha/isa/fp.isa: arch/alpha/isa/int.isa: arch/alpha/isa/mem.isa: arch/alpha/isa/pal.isa: arch/alpha/isa/unimp.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. arch/alpha/isa/main.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. The decoder (for Alpha) uses the 32nd bit in order to determine if the machine is in PAL mode. If it is, then it refers to the reg_redir table to determine the true index of the register it is using. Also include the opcdec instruction definition. arch/alpha/isa_traits.hh: Define ExtMachInst type that is used by the static inst in order to decode the instruction, given the context of being in pal mode or not. Redefine the number of Int registers, splitting it into NumIntArchRegs (32) and NumIntRegs (32 + 8 shadow registers). Change the dependence tags to reflect the integer registers include the 8 shadow registers. Define function to make an ExtMachInst. Currently it is somewhat specific to Alpha; in the future it must be decided to make this more generic and possibly slower, or leave it specific to each architecture and ifdef it within the CPU. arch/isa_parser.py: Have static insts decode on the ExtMachInst. base/remote_gdb.cc: Support the automatic remapping of shadow registers. Remote GDB must now look at the PC being read in order to tell if it should use the normal register indices or the shadow register indices. cpu/o3/regfile.hh: Comment out the pal registers; they are now a part of the integer registers. cpu/simple/cpu.cc: Create an ExtMachInst to decode on, based on the normal MachInst and the PC of the instructoin. cpu/static_inst.hh: Change from MachInst to ExtMachInst to support shadow register renaming. --HG-- extra : convert_revision : 1d23eabf735e297068e1917445a6348e9f8c88d5
2006-03-03 21:28:25 +01:00
%(class_name)s(ExtMachInst machInst);
%(BasicExecDeclare)s
%(InitiateAccDeclare)s
%(CompleteAccDeclare)s
};
}};
def template InitiateAccDeclare {{
Fault initiateAcc(%(CPU_exec_context)s *, Trace::InstRecord *) const;
}};
def template CompleteAccDeclare {{
Split SimpleCPU into two different models, AtomicSimpleCPU and TimingSimpleCPU, which use atomic and timing memory accesses respectively. Common code is factored into the BaseSimpleCPU class. AtomicSimpleCPU includes an option (simulate_stalls) to add delays based on the estimated latency reported by the atomic accesses. Plain old "SimpleCPU" is gone; I have not updated all the config files (just test/test.py). Also fixes to get timing accesses working in new memory model and to get split-phase memory instruction definitions working with new memory model as well. arch/alpha/isa/main.isa: Need to include packet_impl.h for functions that use Packet objects. arch/alpha/isa/mem.isa: Change completeAcc() methods to take Packet object pointers. Also split out StoreCond template for completeAcc(), since that's the only one that needs write_result and we get an unused variable warning if we always have it in there. build/SConstruct: Update list of recognized CPU model names. configs/test/test.py: Change SimpleCPU to AtomicSimpleCPU. cpu/SConscript: Define sources for new CPU models. Add split memory access methods to CPU model signatures. cpu/cpu_models.py: cpu/static_inst.hh: Define new CPU models. cpu/simple/base.cc: cpu/simple/base.hh: Factor out pieces specific to Atomic or Timing models. mem/bus.cc: Bus needs to be able to route timing packets based on explicit dest so responses can get back to requester. Set dest to Packet::Broadcast to indicate that dest should be derived from address. Also set packet src field based on port from which packet is sent. mem/bus.hh: Set packet src field based on port from which packet is sent. mem/packet.hh: Define Broadcast destination address to indicate that packet should be routed based on address. mem/physical.cc: Set packet dest on response so packet is routed back to requester properly. mem/port.cc: Flag blob packets as Broadcast. python/m5/objects/PhysicalMemory.py: Change default latency to be 1 cycle. --HG-- rename : cpu/simple/cpu.cc => cpu/simple/base.cc rename : cpu/simple/cpu.hh => cpu/simple/base.hh extra : convert_revision : e9646af6406a20c8c605087936dc4683375c2132
2006-05-16 23:36:50 +02:00
Fault completeAcc(Packet *, %(CPU_exec_context)s *,
Trace::InstRecord *) const;
}};
def template LoadStoreConstructor {{
/** TODO: change op_class to AddrGenOp or something (requires
* creating new member of OpClass enum in op_class.hh, updating
* config files, etc.). */
Changes to support automatic renaming of the shadow registers at decode time. This requires using an ExtMachInst (uint64_t) instead of the normal MachInst; the ExtMachInst is packed with extra decode context information. In the case of Alpha, the PAL mode is included. The shadow registers are folded into the normal integer registers to ease renaming indexing. Include the removed Opcdec class of instructions for faulting when a pal mode only instruction is decoded in non-pal mode. arch/alpha/ev5.cc: Changes to automatically map the shadow registers if the instruction is in PAL mode. arch/alpha/isa/branch.isa: arch/alpha/isa/decoder.isa: arch/alpha/isa/fp.isa: arch/alpha/isa/int.isa: arch/alpha/isa/mem.isa: arch/alpha/isa/pal.isa: arch/alpha/isa/unimp.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. arch/alpha/isa/main.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. The decoder (for Alpha) uses the 32nd bit in order to determine if the machine is in PAL mode. If it is, then it refers to the reg_redir table to determine the true index of the register it is using. Also include the opcdec instruction definition. arch/alpha/isa_traits.hh: Define ExtMachInst type that is used by the static inst in order to decode the instruction, given the context of being in pal mode or not. Redefine the number of Int registers, splitting it into NumIntArchRegs (32) and NumIntRegs (32 + 8 shadow registers). Change the dependence tags to reflect the integer registers include the 8 shadow registers. Define function to make an ExtMachInst. Currently it is somewhat specific to Alpha; in the future it must be decided to make this more generic and possibly slower, or leave it specific to each architecture and ifdef it within the CPU. arch/isa_parser.py: Have static insts decode on the ExtMachInst. base/remote_gdb.cc: Support the automatic remapping of shadow registers. Remote GDB must now look at the PC being read in order to tell if it should use the normal register indices or the shadow register indices. cpu/o3/regfile.hh: Comment out the pal registers; they are now a part of the integer registers. cpu/simple/cpu.cc: Create an ExtMachInst to decode on, based on the normal MachInst and the PC of the instructoin. cpu/static_inst.hh: Change from MachInst to ExtMachInst to support shadow register renaming. --HG-- extra : convert_revision : 1d23eabf735e297068e1917445a6348e9f8c88d5
2006-03-03 21:28:25 +01:00
inline %(class_name)s::EAComp::EAComp(ExtMachInst machInst)
: %(base_class)s("%(mnemonic)s (EAComp)", machInst, IntAluOp)
{
%(ea_constructor)s;
}
Changes to support automatic renaming of the shadow registers at decode time. This requires using an ExtMachInst (uint64_t) instead of the normal MachInst; the ExtMachInst is packed with extra decode context information. In the case of Alpha, the PAL mode is included. The shadow registers are folded into the normal integer registers to ease renaming indexing. Include the removed Opcdec class of instructions for faulting when a pal mode only instruction is decoded in non-pal mode. arch/alpha/ev5.cc: Changes to automatically map the shadow registers if the instruction is in PAL mode. arch/alpha/isa/branch.isa: arch/alpha/isa/decoder.isa: arch/alpha/isa/fp.isa: arch/alpha/isa/int.isa: arch/alpha/isa/mem.isa: arch/alpha/isa/pal.isa: arch/alpha/isa/unimp.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. arch/alpha/isa/main.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. The decoder (for Alpha) uses the 32nd bit in order to determine if the machine is in PAL mode. If it is, then it refers to the reg_redir table to determine the true index of the register it is using. Also include the opcdec instruction definition. arch/alpha/isa_traits.hh: Define ExtMachInst type that is used by the static inst in order to decode the instruction, given the context of being in pal mode or not. Redefine the number of Int registers, splitting it into NumIntArchRegs (32) and NumIntRegs (32 + 8 shadow registers). Change the dependence tags to reflect the integer registers include the 8 shadow registers. Define function to make an ExtMachInst. Currently it is somewhat specific to Alpha; in the future it must be decided to make this more generic and possibly slower, or leave it specific to each architecture and ifdef it within the CPU. arch/isa_parser.py: Have static insts decode on the ExtMachInst. base/remote_gdb.cc: Support the automatic remapping of shadow registers. Remote GDB must now look at the PC being read in order to tell if it should use the normal register indices or the shadow register indices. cpu/o3/regfile.hh: Comment out the pal registers; they are now a part of the integer registers. cpu/simple/cpu.cc: Create an ExtMachInst to decode on, based on the normal MachInst and the PC of the instructoin. cpu/static_inst.hh: Change from MachInst to ExtMachInst to support shadow register renaming. --HG-- extra : convert_revision : 1d23eabf735e297068e1917445a6348e9f8c88d5
2006-03-03 21:28:25 +01:00
inline %(class_name)s::MemAcc::MemAcc(ExtMachInst machInst)
: %(base_class)s("%(mnemonic)s (MemAcc)", machInst, %(op_class)s)
{
%(memacc_constructor)s;
}
Changes to support automatic renaming of the shadow registers at decode time. This requires using an ExtMachInst (uint64_t) instead of the normal MachInst; the ExtMachInst is packed with extra decode context information. In the case of Alpha, the PAL mode is included. The shadow registers are folded into the normal integer registers to ease renaming indexing. Include the removed Opcdec class of instructions for faulting when a pal mode only instruction is decoded in non-pal mode. arch/alpha/ev5.cc: Changes to automatically map the shadow registers if the instruction is in PAL mode. arch/alpha/isa/branch.isa: arch/alpha/isa/decoder.isa: arch/alpha/isa/fp.isa: arch/alpha/isa/int.isa: arch/alpha/isa/mem.isa: arch/alpha/isa/pal.isa: arch/alpha/isa/unimp.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. arch/alpha/isa/main.isa: Changes for automatically using the shadow registers. Now instructions must decode based on an ExtMachInst, which is a MachInst with any decode context information concatenated onto the higher order bits. The decoder (for Alpha) uses the 32nd bit in order to determine if the machine is in PAL mode. If it is, then it refers to the reg_redir table to determine the true index of the register it is using. Also include the opcdec instruction definition. arch/alpha/isa_traits.hh: Define ExtMachInst type that is used by the static inst in order to decode the instruction, given the context of being in pal mode or not. Redefine the number of Int registers, splitting it into NumIntArchRegs (32) and NumIntRegs (32 + 8 shadow registers). Change the dependence tags to reflect the integer registers include the 8 shadow registers. Define function to make an ExtMachInst. Currently it is somewhat specific to Alpha; in the future it must be decided to make this more generic and possibly slower, or leave it specific to each architecture and ifdef it within the CPU. arch/isa_parser.py: Have static insts decode on the ExtMachInst. base/remote_gdb.cc: Support the automatic remapping of shadow registers. Remote GDB must now look at the PC being read in order to tell if it should use the normal register indices or the shadow register indices. cpu/o3/regfile.hh: Comment out the pal registers; they are now a part of the integer registers. cpu/simple/cpu.cc: Create an ExtMachInst to decode on, based on the normal MachInst and the PC of the instructoin. cpu/static_inst.hh: Change from MachInst to ExtMachInst to support shadow register renaming. --HG-- extra : convert_revision : 1d23eabf735e297068e1917445a6348e9f8c88d5
2006-03-03 21:28:25 +01:00
inline %(class_name)s::%(class_name)s(ExtMachInst machInst)
: %(base_class)s("%(mnemonic)s", machInst, %(op_class)s,
new EAComp(machInst), new MemAcc(machInst))
{
%(constructor)s;
}
}};
def template EACompExecute {{
Fault
%(class_name)s::EAComp::execute(%(CPU_exec_context)s *xc,
Trace::InstRecord *traceData) const
{
Addr EA;
Fault fault = NoFault;
%(fp_enable_check)s;
%(op_decl)s;
%(op_rd)s;
%(code)s;
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
%(op_wb)s;
xc->setEA(EA);
}
return fault;
}
}};
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
def template LoadMemAccExecute {{
Fault
%(class_name)s::MemAcc::execute(%(CPU_exec_context)s *xc,
Trace::InstRecord *traceData) const
{
Addr EA;
Fault fault = NoFault;
%(fp_enable_check)s;
%(op_decl)s;
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
%(op_rd)s;
EA = xc->getEA();
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
fault = xc->read(EA, (uint%(mem_acc_size)d_t&)Mem, memAccessFlags);
%(code)s;
}
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
%(op_wb)s;
}
return fault;
}
}};
def template LoadExecute {{
Fault %(class_name)s::execute(%(CPU_exec_context)s *xc,
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
Trace::InstRecord *traceData) const
{
Addr EA;
Fault fault = NoFault;
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
%(fp_enable_check)s;
%(op_decl)s;
%(op_rd)s;
%(ea_code)s;
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
fault = xc->read(EA, (uint%(mem_acc_size)d_t&)Mem, memAccessFlags);
%(memacc_code)s;
}
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
%(op_wb)s;
}
return fault;
}
}};
def template LoadInitiateAcc {{
Fault %(class_name)s::initiateAcc(%(CPU_exec_context)s *xc,
Trace::InstRecord *traceData) const
{
Addr EA;
Fault fault = NoFault;
%(fp_enable_check)s;
%(op_src_decl)s;
%(op_rd)s;
%(ea_code)s;
if (fault == NoFault) {
fault = xc->read(EA, (uint%(mem_acc_size)d_t &)Mem, memAccessFlags);
}
return fault;
}
}};
def template LoadCompleteAcc {{
Split SimpleCPU into two different models, AtomicSimpleCPU and TimingSimpleCPU, which use atomic and timing memory accesses respectively. Common code is factored into the BaseSimpleCPU class. AtomicSimpleCPU includes an option (simulate_stalls) to add delays based on the estimated latency reported by the atomic accesses. Plain old "SimpleCPU" is gone; I have not updated all the config files (just test/test.py). Also fixes to get timing accesses working in new memory model and to get split-phase memory instruction definitions working with new memory model as well. arch/alpha/isa/main.isa: Need to include packet_impl.h for functions that use Packet objects. arch/alpha/isa/mem.isa: Change completeAcc() methods to take Packet object pointers. Also split out StoreCond template for completeAcc(), since that's the only one that needs write_result and we get an unused variable warning if we always have it in there. build/SConstruct: Update list of recognized CPU model names. configs/test/test.py: Change SimpleCPU to AtomicSimpleCPU. cpu/SConscript: Define sources for new CPU models. Add split memory access methods to CPU model signatures. cpu/cpu_models.py: cpu/static_inst.hh: Define new CPU models. cpu/simple/base.cc: cpu/simple/base.hh: Factor out pieces specific to Atomic or Timing models. mem/bus.cc: Bus needs to be able to route timing packets based on explicit dest so responses can get back to requester. Set dest to Packet::Broadcast to indicate that dest should be derived from address. Also set packet src field based on port from which packet is sent. mem/bus.hh: Set packet src field based on port from which packet is sent. mem/packet.hh: Define Broadcast destination address to indicate that packet should be routed based on address. mem/physical.cc: Set packet dest on response so packet is routed back to requester properly. mem/port.cc: Flag blob packets as Broadcast. python/m5/objects/PhysicalMemory.py: Change default latency to be 1 cycle. --HG-- rename : cpu/simple/cpu.cc => cpu/simple/base.cc rename : cpu/simple/cpu.hh => cpu/simple/base.hh extra : convert_revision : e9646af6406a20c8c605087936dc4683375c2132
2006-05-16 23:36:50 +02:00
Fault %(class_name)s::completeAcc(Packet *pkt,
%(CPU_exec_context)s *xc,
Trace::InstRecord *traceData) const
{
Fault fault = NoFault;
%(fp_enable_check)s;
%(op_decl)s;
Split SimpleCPU into two different models, AtomicSimpleCPU and TimingSimpleCPU, which use atomic and timing memory accesses respectively. Common code is factored into the BaseSimpleCPU class. AtomicSimpleCPU includes an option (simulate_stalls) to add delays based on the estimated latency reported by the atomic accesses. Plain old "SimpleCPU" is gone; I have not updated all the config files (just test/test.py). Also fixes to get timing accesses working in new memory model and to get split-phase memory instruction definitions working with new memory model as well. arch/alpha/isa/main.isa: Need to include packet_impl.h for functions that use Packet objects. arch/alpha/isa/mem.isa: Change completeAcc() methods to take Packet object pointers. Also split out StoreCond template for completeAcc(), since that's the only one that needs write_result and we get an unused variable warning if we always have it in there. build/SConstruct: Update list of recognized CPU model names. configs/test/test.py: Change SimpleCPU to AtomicSimpleCPU. cpu/SConscript: Define sources for new CPU models. Add split memory access methods to CPU model signatures. cpu/cpu_models.py: cpu/static_inst.hh: Define new CPU models. cpu/simple/base.cc: cpu/simple/base.hh: Factor out pieces specific to Atomic or Timing models. mem/bus.cc: Bus needs to be able to route timing packets based on explicit dest so responses can get back to requester. Set dest to Packet::Broadcast to indicate that dest should be derived from address. Also set packet src field based on port from which packet is sent. mem/bus.hh: Set packet src field based on port from which packet is sent. mem/packet.hh: Define Broadcast destination address to indicate that packet should be routed based on address. mem/physical.cc: Set packet dest on response so packet is routed back to requester properly. mem/port.cc: Flag blob packets as Broadcast. python/m5/objects/PhysicalMemory.py: Change default latency to be 1 cycle. --HG-- rename : cpu/simple/cpu.cc => cpu/simple/base.cc rename : cpu/simple/cpu.hh => cpu/simple/base.hh extra : convert_revision : e9646af6406a20c8c605087936dc4683375c2132
2006-05-16 23:36:50 +02:00
Mem = pkt->get<typeof(Mem)>();
if (fault == NoFault) {
%(memacc_code)s;
}
if (fault == NoFault) {
%(op_wb)s;
}
return fault;
}
}};
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
def template StoreMemAccExecute {{
Fault
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
%(class_name)s::MemAcc::execute(%(CPU_exec_context)s *xc,
Trace::InstRecord *traceData) const
{
Addr EA;
Fault fault = NoFault;
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
uint64_t write_result = 0;
%(fp_enable_check)s;
%(op_decl)s;
%(op_rd)s;
EA = xc->getEA();
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
%(code)s;
}
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
fault = xc->write((uint%(mem_acc_size)d_t&)Mem, EA,
memAccessFlags, &write_result);
if (traceData) { traceData->setData(Mem); }
}
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
%(postacc_code)s;
}
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
%(op_wb)s;
}
return fault;
}
}};
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
def template StoreExecute {{
Fault %(class_name)s::execute(%(CPU_exec_context)s *xc,
Trace::InstRecord *traceData) const
{
Addr EA;
Fault fault = NoFault;
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
uint64_t write_result = 0;
%(fp_enable_check)s;
%(op_decl)s;
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
%(op_rd)s;
%(ea_code)s;
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
%(memacc_code)s;
}
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
fault = xc->write((uint%(mem_acc_size)d_t&)Mem, EA,
memAccessFlags, &write_result);
if (traceData) { traceData->setData(Mem); }
}
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
%(postacc_code)s;
}
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
%(op_wb)s;
}
return fault;
}
}};
def template StoreInitiateAcc {{
Fault %(class_name)s::initiateAcc(%(CPU_exec_context)s *xc,
Trace::InstRecord *traceData) const
{
Addr EA;
Fault fault = NoFault;
%(fp_enable_check)s;
%(op_decl)s;
%(op_rd)s;
%(ea_code)s;
if (fault == NoFault) {
%(memacc_code)s;
}
if (fault == NoFault) {
fault = xc->write((uint%(mem_acc_size)d_t&)Mem, EA,
Split SimpleCPU into two different models, AtomicSimpleCPU and TimingSimpleCPU, which use atomic and timing memory accesses respectively. Common code is factored into the BaseSimpleCPU class. AtomicSimpleCPU includes an option (simulate_stalls) to add delays based on the estimated latency reported by the atomic accesses. Plain old "SimpleCPU" is gone; I have not updated all the config files (just test/test.py). Also fixes to get timing accesses working in new memory model and to get split-phase memory instruction definitions working with new memory model as well. arch/alpha/isa/main.isa: Need to include packet_impl.h for functions that use Packet objects. arch/alpha/isa/mem.isa: Change completeAcc() methods to take Packet object pointers. Also split out StoreCond template for completeAcc(), since that's the only one that needs write_result and we get an unused variable warning if we always have it in there. build/SConstruct: Update list of recognized CPU model names. configs/test/test.py: Change SimpleCPU to AtomicSimpleCPU. cpu/SConscript: Define sources for new CPU models. Add split memory access methods to CPU model signatures. cpu/cpu_models.py: cpu/static_inst.hh: Define new CPU models. cpu/simple/base.cc: cpu/simple/base.hh: Factor out pieces specific to Atomic or Timing models. mem/bus.cc: Bus needs to be able to route timing packets based on explicit dest so responses can get back to requester. Set dest to Packet::Broadcast to indicate that dest should be derived from address. Also set packet src field based on port from which packet is sent. mem/bus.hh: Set packet src field based on port from which packet is sent. mem/packet.hh: Define Broadcast destination address to indicate that packet should be routed based on address. mem/physical.cc: Set packet dest on response so packet is routed back to requester properly. mem/port.cc: Flag blob packets as Broadcast. python/m5/objects/PhysicalMemory.py: Change default latency to be 1 cycle. --HG-- rename : cpu/simple/cpu.cc => cpu/simple/base.cc rename : cpu/simple/cpu.hh => cpu/simple/base.hh extra : convert_revision : e9646af6406a20c8c605087936dc4683375c2132
2006-05-16 23:36:50 +02:00
memAccessFlags, NULL);
if (traceData) { traceData->setData(Mem); }
}
return fault;
}
}};
def template StoreCompleteAcc {{
Split SimpleCPU into two different models, AtomicSimpleCPU and TimingSimpleCPU, which use atomic and timing memory accesses respectively. Common code is factored into the BaseSimpleCPU class. AtomicSimpleCPU includes an option (simulate_stalls) to add delays based on the estimated latency reported by the atomic accesses. Plain old "SimpleCPU" is gone; I have not updated all the config files (just test/test.py). Also fixes to get timing accesses working in new memory model and to get split-phase memory instruction definitions working with new memory model as well. arch/alpha/isa/main.isa: Need to include packet_impl.h for functions that use Packet objects. arch/alpha/isa/mem.isa: Change completeAcc() methods to take Packet object pointers. Also split out StoreCond template for completeAcc(), since that's the only one that needs write_result and we get an unused variable warning if we always have it in there. build/SConstruct: Update list of recognized CPU model names. configs/test/test.py: Change SimpleCPU to AtomicSimpleCPU. cpu/SConscript: Define sources for new CPU models. Add split memory access methods to CPU model signatures. cpu/cpu_models.py: cpu/static_inst.hh: Define new CPU models. cpu/simple/base.cc: cpu/simple/base.hh: Factor out pieces specific to Atomic or Timing models. mem/bus.cc: Bus needs to be able to route timing packets based on explicit dest so responses can get back to requester. Set dest to Packet::Broadcast to indicate that dest should be derived from address. Also set packet src field based on port from which packet is sent. mem/bus.hh: Set packet src field based on port from which packet is sent. mem/packet.hh: Define Broadcast destination address to indicate that packet should be routed based on address. mem/physical.cc: Set packet dest on response so packet is routed back to requester properly. mem/port.cc: Flag blob packets as Broadcast. python/m5/objects/PhysicalMemory.py: Change default latency to be 1 cycle. --HG-- rename : cpu/simple/cpu.cc => cpu/simple/base.cc rename : cpu/simple/cpu.hh => cpu/simple/base.hh extra : convert_revision : e9646af6406a20c8c605087936dc4683375c2132
2006-05-16 23:36:50 +02:00
Fault %(class_name)s::completeAcc(Packet *pkt,
%(CPU_exec_context)s *xc,
Trace::InstRecord *traceData) const
{
Fault fault = NoFault;
%(fp_enable_check)s;
%(op_dest_decl)s;
Split SimpleCPU into two different models, AtomicSimpleCPU and TimingSimpleCPU, which use atomic and timing memory accesses respectively. Common code is factored into the BaseSimpleCPU class. AtomicSimpleCPU includes an option (simulate_stalls) to add delays based on the estimated latency reported by the atomic accesses. Plain old "SimpleCPU" is gone; I have not updated all the config files (just test/test.py). Also fixes to get timing accesses working in new memory model and to get split-phase memory instruction definitions working with new memory model as well. arch/alpha/isa/main.isa: Need to include packet_impl.h for functions that use Packet objects. arch/alpha/isa/mem.isa: Change completeAcc() methods to take Packet object pointers. Also split out StoreCond template for completeAcc(), since that's the only one that needs write_result and we get an unused variable warning if we always have it in there. build/SConstruct: Update list of recognized CPU model names. configs/test/test.py: Change SimpleCPU to AtomicSimpleCPU. cpu/SConscript: Define sources for new CPU models. Add split memory access methods to CPU model signatures. cpu/cpu_models.py: cpu/static_inst.hh: Define new CPU models. cpu/simple/base.cc: cpu/simple/base.hh: Factor out pieces specific to Atomic or Timing models. mem/bus.cc: Bus needs to be able to route timing packets based on explicit dest so responses can get back to requester. Set dest to Packet::Broadcast to indicate that dest should be derived from address. Also set packet src field based on port from which packet is sent. mem/bus.hh: Set packet src field based on port from which packet is sent. mem/packet.hh: Define Broadcast destination address to indicate that packet should be routed based on address. mem/physical.cc: Set packet dest on response so packet is routed back to requester properly. mem/port.cc: Flag blob packets as Broadcast. python/m5/objects/PhysicalMemory.py: Change default latency to be 1 cycle. --HG-- rename : cpu/simple/cpu.cc => cpu/simple/base.cc rename : cpu/simple/cpu.hh => cpu/simple/base.hh extra : convert_revision : e9646af6406a20c8c605087936dc4683375c2132
2006-05-16 23:36:50 +02:00
if (fault == NoFault) {
%(postacc_code)s;
}
if (fault == NoFault) {
%(op_wb)s;
}
return fault;
}
}};
def template StoreCondCompleteAcc {{
Fault %(class_name)s::completeAcc(Packet *pkt,
%(CPU_exec_context)s *xc,
Trace::InstRecord *traceData) const
{
Fault fault = NoFault;
%(fp_enable_check)s;
%(op_dest_decl)s;
uint64_t write_result = pkt->req->getScResult();
if (fault == NoFault) {
%(postacc_code)s;
}
if (fault == NoFault) {
%(op_wb)s;
}
return fault;
}
}};
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
def template MiscMemAccExecute {{
Fault %(class_name)s::MemAcc::execute(%(CPU_exec_context)s *xc,
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
Trace::InstRecord *traceData) const
{
Addr EA;
Fault fault = NoFault;
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
%(fp_enable_check)s;
%(op_decl)s;
%(op_rd)s;
EA = xc->getEA();
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
%(code)s;
}
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
return NoFault;
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
}
}};
def template MiscExecute {{
Fault %(class_name)s::execute(%(CPU_exec_context)s *xc,
Trace::InstRecord *traceData) const
{
Addr EA;
Fault fault = NoFault;
%(fp_enable_check)s;
%(op_decl)s;
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
%(op_rd)s;
%(ea_code)s;
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
if (fault == NoFault) {
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
%(memacc_code)s;
}
Changed the fault enum into a class, and fixed everything up to work with it. Next, the faults need to be pulled out of all the other code so that they are only used to communicate between the CPU and the ISA. SConscript: The new faults.cc file in sim allocates the system wide faults. When these faults are generated through a function interface in the ISA, this file may go away. arch/alpha/alpha_memory.cc: Changed Fault to Fault * and took the underscores out of fault names. arch/alpha/alpha_memory.hh: Changed Fault to Fault *. Also, added an include for the alpha faults. arch/alpha/ev5.cc: Changed the fault_addr array into a fault_addr function. Once all of the faults can be expected to have the same type, fault_addr can go away completely and the info it provided will come from the fault itself. Also, Fault was changed to Fault *, and underscores were taken out of fault names. arch/alpha/isa/decoder.isa: Changed Fault to Fault * and took the underscores out fault names. arch/alpha/isa/fp.isa: Changed Fault to Fault *, and took the underscores out of fault names. arch/alpha/isa/main.isa: Changed Fault to Fault *, removed underscores from fault names, and made an include of the alpha faults show up in all the generated files. arch/alpha/isa/mem.isa: Changed Fault to Fault * and removed underscores from fault names. arch/alpha/isa/unimp.isa: arch/alpha/isa/unknown.isa: cpu/exec_context.hh: cpu/ozone/cpu.hh: cpu/simple/cpu.cc: dev/alpha_console.cc: dev/ide_ctrl.cc: dev/isa_fake.cc: dev/pciconfigall.cc: dev/pcidev.cc: dev/pcidev.hh: dev/tsunami_cchip.cc: dev/tsunami_io.cc: dev/tsunami_pchip.cc: Changed Fault to Fault *, and removed underscores from fault names. arch/alpha/isa_traits.hh: Changed the include of arch/alpha/faults.hh to sim/faults.hh, since the alpha faults weren't needed. cpu/base_dyn_inst.cc: Changed Fault to Fault *, and removed underscores from fault names. This file probably shouldn't use the Unimplemented Opcode fault. cpu/base_dyn_inst.hh: Changed Fault to Fault * and took the underscores out of the fault names. cpu/exec_context.cc: cpu/o3/alpha_dyn_inst.hh: cpu/o3/alpha_dyn_inst_impl.hh: cpu/o3/fetch.hh: dev/alpha_console.hh: dev/baddev.hh: dev/ide_ctrl.hh: dev/isa_fake.hh: dev/ns_gige.hh: dev/pciconfigall.hh: dev/sinic.hh: dev/tsunami_cchip.hh: dev/tsunami_io.hh: dev/tsunami_pchip.hh: dev/uart.hh: dev/uart8250.hh: Changed Fault to Fault *. cpu/o3/alpha_cpu.hh: Changed Fault to Fault *, removed underscores from fault names. cpu/o3/alpha_cpu_impl.hh: Changed Fault to Fault *, removed underscores from fault names, and changed the fault_addr array to the fault_addr function. Once all faults are from the ISA, this function will probably go away. cpu/o3/commit_impl.hh: cpu/o3/fetch_impl.hh: dev/baddev.cc: Changed Fault to Fault *, and removed underscores from the fault names. cpu/o3/regfile.hh: Added an include for the alpha specific faults which will hopefully go away once the ipr stuff is moved, changed Fault to Fault *, and removed the underscores from fault names. cpu/simple/cpu.hh: Changed Fault to Fault * dev/ns_gige.cc: Changed Fault to Fault *, and removdd underscores from fault names. dev/sinic.cc: Changed Fault to Fault *, and removed the underscores from fault names. dev/uart8250.cc: Chanted Fault to Fault *, and removed underscores from fault names. kern/kernel_stats.cc: Removed underscores from fault names, and from NumFaults. kern/kernel_stats.hh: Changed the predeclaration of Fault from an enum to a class, and changd the "fault" function to work with the classes instead of the enum. Once there are no system wide faults anymore, this code will simplify back to something like it was originally. sim/faults.cc: This allocates the system wide faults. sim/faults.hh: This declares the system wide faults. sim/syscall_emul.cc: sim/syscall_emul.hh: Removed the underscores from fault names. --HG-- rename : arch/alpha/faults.cc => sim/faults.cc rename : arch/alpha/faults.hh => sim/faults.hh extra : convert_revision : 253d39258237333ae8ec4d8047367cb3ea68569d
2006-02-16 07:22:51 +01:00
return NoFault;
}
}};
def template MiscInitiateAcc {{
Fault %(class_name)s::initiateAcc(%(CPU_exec_context)s *xc,
Trace::InstRecord *traceData) const
{
panic("Misc instruction does not support split access method!");
return NoFault;
}
}};
def template MiscCompleteAcc {{
Split SimpleCPU into two different models, AtomicSimpleCPU and TimingSimpleCPU, which use atomic and timing memory accesses respectively. Common code is factored into the BaseSimpleCPU class. AtomicSimpleCPU includes an option (simulate_stalls) to add delays based on the estimated latency reported by the atomic accesses. Plain old "SimpleCPU" is gone; I have not updated all the config files (just test/test.py). Also fixes to get timing accesses working in new memory model and to get split-phase memory instruction definitions working with new memory model as well. arch/alpha/isa/main.isa: Need to include packet_impl.h for functions that use Packet objects. arch/alpha/isa/mem.isa: Change completeAcc() methods to take Packet object pointers. Also split out StoreCond template for completeAcc(), since that's the only one that needs write_result and we get an unused variable warning if we always have it in there. build/SConstruct: Update list of recognized CPU model names. configs/test/test.py: Change SimpleCPU to AtomicSimpleCPU. cpu/SConscript: Define sources for new CPU models. Add split memory access methods to CPU model signatures. cpu/cpu_models.py: cpu/static_inst.hh: Define new CPU models. cpu/simple/base.cc: cpu/simple/base.hh: Factor out pieces specific to Atomic or Timing models. mem/bus.cc: Bus needs to be able to route timing packets based on explicit dest so responses can get back to requester. Set dest to Packet::Broadcast to indicate that dest should be derived from address. Also set packet src field based on port from which packet is sent. mem/bus.hh: Set packet src field based on port from which packet is sent. mem/packet.hh: Define Broadcast destination address to indicate that packet should be routed based on address. mem/physical.cc: Set packet dest on response so packet is routed back to requester properly. mem/port.cc: Flag blob packets as Broadcast. python/m5/objects/PhysicalMemory.py: Change default latency to be 1 cycle. --HG-- rename : cpu/simple/cpu.cc => cpu/simple/base.cc rename : cpu/simple/cpu.hh => cpu/simple/base.hh extra : convert_revision : e9646af6406a20c8c605087936dc4683375c2132
2006-05-16 23:36:50 +02:00
Fault %(class_name)s::completeAcc(Packet *pkt,
%(CPU_exec_context)s *xc,
Trace::InstRecord *traceData) const
{
panic("Misc instruction does not support split access method!");
return NoFault;
}
}};
// load instructions use Ra as dest, so check for
// Ra == 31 to detect nops
def template LoadNopCheckDecode {{
{
AlphaStaticInst *i = new %(class_name)s(machInst);
if (RA == 31) {
i = makeNop(i);
}
return i;
}
}};
// for some load instructions, Ra == 31 indicates a prefetch (not a nop)
def template LoadPrefetchCheckDecode {{
{
if (RA != 31) {
return new %(class_name)s(machInst);
}
else {
return new %(class_name)sPrefetch(machInst);
}
}
}};
let {{
def LoadStoreBase(name, Name, ea_code, memacc_code, mem_flags, inst_flags,
postacc_code = '', base_class = 'MemoryDisp32',
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
decode_template = BasicDecode, exec_template_base = ''):
# Make sure flags are in lists (convert to lists if not).
mem_flags = makeList(mem_flags)
inst_flags = makeList(inst_flags)
# add hook to get effective addresses into execution trace output.
ea_code += '\nif (traceData) { traceData->setAddr(EA); }\n'
# generate code block objects
ea_cblk = CodeBlock(ea_code)
memacc_cblk = CodeBlock(memacc_code)
postacc_cblk = CodeBlock(postacc_code)
# Some CPU models execute the memory operation as an atomic unit,
# while others want to separate them into an effective address
# computation and a memory access operation. As a result, we need
# to generate three StaticInst objects. Note that the latter two
# are nested inside the larger "atomic" one.
# generate InstObjParams for EAComp object
ea_iop = InstObjParams(name, Name, base_class, ea_cblk, inst_flags)
# generate InstObjParams for MemAcc object
memacc_iop = InstObjParams(name, Name, base_class, memacc_cblk, inst_flags)
# in the split execution model, the MemAcc portion is responsible
# for the post-access code.
memacc_iop.postacc_code = postacc_cblk.code
# generate InstObjParams for InitiateAcc, CompleteAcc object
# The code used depends on the template being used
if (exec_template_base == 'Load'):
initiateacc_cblk = CodeBlock(ea_code + memacc_code)
completeacc_cblk = CodeBlock(memacc_code + postacc_code)
Split SimpleCPU into two different models, AtomicSimpleCPU and TimingSimpleCPU, which use atomic and timing memory accesses respectively. Common code is factored into the BaseSimpleCPU class. AtomicSimpleCPU includes an option (simulate_stalls) to add delays based on the estimated latency reported by the atomic accesses. Plain old "SimpleCPU" is gone; I have not updated all the config files (just test/test.py). Also fixes to get timing accesses working in new memory model and to get split-phase memory instruction definitions working with new memory model as well. arch/alpha/isa/main.isa: Need to include packet_impl.h for functions that use Packet objects. arch/alpha/isa/mem.isa: Change completeAcc() methods to take Packet object pointers. Also split out StoreCond template for completeAcc(), since that's the only one that needs write_result and we get an unused variable warning if we always have it in there. build/SConstruct: Update list of recognized CPU model names. configs/test/test.py: Change SimpleCPU to AtomicSimpleCPU. cpu/SConscript: Define sources for new CPU models. Add split memory access methods to CPU model signatures. cpu/cpu_models.py: cpu/static_inst.hh: Define new CPU models. cpu/simple/base.cc: cpu/simple/base.hh: Factor out pieces specific to Atomic or Timing models. mem/bus.cc: Bus needs to be able to route timing packets based on explicit dest so responses can get back to requester. Set dest to Packet::Broadcast to indicate that dest should be derived from address. Also set packet src field based on port from which packet is sent. mem/bus.hh: Set packet src field based on port from which packet is sent. mem/packet.hh: Define Broadcast destination address to indicate that packet should be routed based on address. mem/physical.cc: Set packet dest on response so packet is routed back to requester properly. mem/port.cc: Flag blob packets as Broadcast. python/m5/objects/PhysicalMemory.py: Change default latency to be 1 cycle. --HG-- rename : cpu/simple/cpu.cc => cpu/simple/base.cc rename : cpu/simple/cpu.hh => cpu/simple/base.hh extra : convert_revision : e9646af6406a20c8c605087936dc4683375c2132
2006-05-16 23:36:50 +02:00
elif (exec_template_base.startswith('Store')):
initiateacc_cblk = CodeBlock(ea_code + memacc_code)
completeacc_cblk = CodeBlock(postacc_code)
else:
initiateacc_cblk = ''
completeacc_cblk = ''
initiateacc_iop = InstObjParams(name, Name, base_class, initiateacc_cblk,
inst_flags)
completeacc_iop = InstObjParams(name, Name, base_class, completeacc_cblk,
inst_flags)
if (exec_template_base == 'Load'):
initiateacc_iop.ea_code = ea_cblk.code
initiateacc_iop.memacc_code = memacc_cblk.code
completeacc_iop.memacc_code = memacc_cblk.code
completeacc_iop.postacc_code = postacc_cblk.code
Split SimpleCPU into two different models, AtomicSimpleCPU and TimingSimpleCPU, which use atomic and timing memory accesses respectively. Common code is factored into the BaseSimpleCPU class. AtomicSimpleCPU includes an option (simulate_stalls) to add delays based on the estimated latency reported by the atomic accesses. Plain old "SimpleCPU" is gone; I have not updated all the config files (just test/test.py). Also fixes to get timing accesses working in new memory model and to get split-phase memory instruction definitions working with new memory model as well. arch/alpha/isa/main.isa: Need to include packet_impl.h for functions that use Packet objects. arch/alpha/isa/mem.isa: Change completeAcc() methods to take Packet object pointers. Also split out StoreCond template for completeAcc(), since that's the only one that needs write_result and we get an unused variable warning if we always have it in there. build/SConstruct: Update list of recognized CPU model names. configs/test/test.py: Change SimpleCPU to AtomicSimpleCPU. cpu/SConscript: Define sources for new CPU models. Add split memory access methods to CPU model signatures. cpu/cpu_models.py: cpu/static_inst.hh: Define new CPU models. cpu/simple/base.cc: cpu/simple/base.hh: Factor out pieces specific to Atomic or Timing models. mem/bus.cc: Bus needs to be able to route timing packets based on explicit dest so responses can get back to requester. Set dest to Packet::Broadcast to indicate that dest should be derived from address. Also set packet src field based on port from which packet is sent. mem/bus.hh: Set packet src field based on port from which packet is sent. mem/packet.hh: Define Broadcast destination address to indicate that packet should be routed based on address. mem/physical.cc: Set packet dest on response so packet is routed back to requester properly. mem/port.cc: Flag blob packets as Broadcast. python/m5/objects/PhysicalMemory.py: Change default latency to be 1 cycle. --HG-- rename : cpu/simple/cpu.cc => cpu/simple/base.cc rename : cpu/simple/cpu.hh => cpu/simple/base.hh extra : convert_revision : e9646af6406a20c8c605087936dc4683375c2132
2006-05-16 23:36:50 +02:00
elif (exec_template_base.startswith('Store')):
initiateacc_iop.ea_code = ea_cblk.code
initiateacc_iop.memacc_code = memacc_cblk.code
completeacc_iop.postacc_code = postacc_cblk.code
# generate InstObjParams for unified execution
cblk = CodeBlock(ea_code + memacc_code + postacc_code)
iop = InstObjParams(name, Name, base_class, cblk, inst_flags)
iop.ea_constructor = ea_cblk.constructor
iop.ea_code = ea_cblk.code
iop.memacc_constructor = memacc_cblk.constructor
iop.memacc_code = memacc_cblk.code
iop.postacc_code = postacc_cblk.code
if mem_flags:
s = '\n\tmemAccessFlags = ' + string.join(mem_flags, '|') + ';'
iop.constructor += s
memacc_iop.constructor += s
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
# select templates
Split SimpleCPU into two different models, AtomicSimpleCPU and TimingSimpleCPU, which use atomic and timing memory accesses respectively. Common code is factored into the BaseSimpleCPU class. AtomicSimpleCPU includes an option (simulate_stalls) to add delays based on the estimated latency reported by the atomic accesses. Plain old "SimpleCPU" is gone; I have not updated all the config files (just test/test.py). Also fixes to get timing accesses working in new memory model and to get split-phase memory instruction definitions working with new memory model as well. arch/alpha/isa/main.isa: Need to include packet_impl.h for functions that use Packet objects. arch/alpha/isa/mem.isa: Change completeAcc() methods to take Packet object pointers. Also split out StoreCond template for completeAcc(), since that's the only one that needs write_result and we get an unused variable warning if we always have it in there. build/SConstruct: Update list of recognized CPU model names. configs/test/test.py: Change SimpleCPU to AtomicSimpleCPU. cpu/SConscript: Define sources for new CPU models. Add split memory access methods to CPU model signatures. cpu/cpu_models.py: cpu/static_inst.hh: Define new CPU models. cpu/simple/base.cc: cpu/simple/base.hh: Factor out pieces specific to Atomic or Timing models. mem/bus.cc: Bus needs to be able to route timing packets based on explicit dest so responses can get back to requester. Set dest to Packet::Broadcast to indicate that dest should be derived from address. Also set packet src field based on port from which packet is sent. mem/bus.hh: Set packet src field based on port from which packet is sent. mem/packet.hh: Define Broadcast destination address to indicate that packet should be routed based on address. mem/physical.cc: Set packet dest on response so packet is routed back to requester properly. mem/port.cc: Flag blob packets as Broadcast. python/m5/objects/PhysicalMemory.py: Change default latency to be 1 cycle. --HG-- rename : cpu/simple/cpu.cc => cpu/simple/base.cc rename : cpu/simple/cpu.hh => cpu/simple/base.hh extra : convert_revision : e9646af6406a20c8c605087936dc4683375c2132
2006-05-16 23:36:50 +02:00
# define aliases... most StoreCond templates are the same as the
# corresponding Store templates (only CompleteAcc is different).
StoreCondMemAccExecute = StoreMemAccExecute
StoreCondExecute = StoreExecute
StoreCondInitiateAcc = StoreInitiateAcc
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
memAccExecTemplate = eval(exec_template_base + 'MemAccExecute')
fullExecTemplate = eval(exec_template_base + 'Execute')
initiateAccTemplate = eval(exec_template_base + 'InitiateAcc')
completeAccTemplate = eval(exec_template_base + 'CompleteAcc')
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
# (header_output, decoder_output, decode_block, exec_output)
return (LoadStoreDeclare.subst(iop), LoadStoreConstructor.subst(iop),
decode_template.subst(iop),
EACompExecute.subst(ea_iop)
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
+ memAccExecTemplate.subst(memacc_iop)
+ fullExecTemplate.subst(iop)
+ initiateAccTemplate.subst(initiateacc_iop)
+ completeAccTemplate.subst(completeacc_iop))
}};
def format LoadOrNop(memacc_code, ea_code = {{ EA = Rb + disp; }},
mem_flags = [], inst_flags = []) {{
(header_output, decoder_output, decode_block, exec_output) = \
LoadStoreBase(name, Name, ea_code, memacc_code, mem_flags, inst_flags,
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
decode_template = LoadNopCheckDecode,
exec_template_base = 'Load')
}};
// Note that the flags passed in apply only to the prefetch version
def format LoadOrPrefetch(memacc_code, ea_code = {{ EA = Rb + disp; }},
mem_flags = [], pf_flags = [], inst_flags = []) {{
# declare the load instruction object and generate the decode block
(header_output, decoder_output, decode_block, exec_output) = \
LoadStoreBase(name, Name, ea_code, memacc_code, mem_flags, inst_flags,
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
decode_template = LoadPrefetchCheckDecode,
exec_template_base = 'Load')
# Declare the prefetch instruction object.
# Make sure flag args are lists so we can mess with them.
mem_flags = makeList(mem_flags)
pf_flags = makeList(pf_flags)
inst_flags = makeList(inst_flags)
pf_mem_flags = mem_flags + pf_flags + ['NO_FAULT']
pf_inst_flags = inst_flags + ['IsMemRef', 'IsLoad',
'IsDataPrefetch', 'MemReadOp']
(pf_header_output, pf_decoder_output, _, pf_exec_output) = \
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
LoadStoreBase(name, Name + 'Prefetch', ea_code,
'xc->prefetch(EA, memAccessFlags);',
pf_mem_flags, pf_inst_flags, exec_template_base = 'Misc')
header_output += pf_header_output
decoder_output += pf_decoder_output
exec_output += pf_exec_output
}};
def format Store(memacc_code, ea_code = {{ EA = Rb + disp; }},
mem_flags = [], inst_flags = []) {{
(header_output, decoder_output, decode_block, exec_output) = \
LoadStoreBase(name, Name, ea_code, memacc_code, mem_flags, inst_flags,
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
exec_template_base = 'Store')
}};
def format StoreCond(memacc_code, postacc_code,
ea_code = {{ EA = Rb + disp; }},
mem_flags = [], inst_flags = []) {{
(header_output, decoder_output, decode_block, exec_output) = \
LoadStoreBase(name, Name, ea_code, memacc_code, mem_flags, inst_flags,
Split SimpleCPU into two different models, AtomicSimpleCPU and TimingSimpleCPU, which use atomic and timing memory accesses respectively. Common code is factored into the BaseSimpleCPU class. AtomicSimpleCPU includes an option (simulate_stalls) to add delays based on the estimated latency reported by the atomic accesses. Plain old "SimpleCPU" is gone; I have not updated all the config files (just test/test.py). Also fixes to get timing accesses working in new memory model and to get split-phase memory instruction definitions working with new memory model as well. arch/alpha/isa/main.isa: Need to include packet_impl.h for functions that use Packet objects. arch/alpha/isa/mem.isa: Change completeAcc() methods to take Packet object pointers. Also split out StoreCond template for completeAcc(), since that's the only one that needs write_result and we get an unused variable warning if we always have it in there. build/SConstruct: Update list of recognized CPU model names. configs/test/test.py: Change SimpleCPU to AtomicSimpleCPU. cpu/SConscript: Define sources for new CPU models. Add split memory access methods to CPU model signatures. cpu/cpu_models.py: cpu/static_inst.hh: Define new CPU models. cpu/simple/base.cc: cpu/simple/base.hh: Factor out pieces specific to Atomic or Timing models. mem/bus.cc: Bus needs to be able to route timing packets based on explicit dest so responses can get back to requester. Set dest to Packet::Broadcast to indicate that dest should be derived from address. Also set packet src field based on port from which packet is sent. mem/bus.hh: Set packet src field based on port from which packet is sent. mem/packet.hh: Define Broadcast destination address to indicate that packet should be routed based on address. mem/physical.cc: Set packet dest on response so packet is routed back to requester properly. mem/port.cc: Flag blob packets as Broadcast. python/m5/objects/PhysicalMemory.py: Change default latency to be 1 cycle. --HG-- rename : cpu/simple/cpu.cc => cpu/simple/base.cc rename : cpu/simple/cpu.hh => cpu/simple/base.hh extra : convert_revision : e9646af6406a20c8c605087936dc4683375c2132
2006-05-16 23:36:50 +02:00
postacc_code, exec_template_base = 'StoreCond')
}};
// Use 'MemoryNoDisp' as base: for wh64, fetch, ecb
def format MiscPrefetch(ea_code, memacc_code,
mem_flags = [], inst_flags = []) {{
(header_output, decoder_output, decode_block, exec_output) = \
LoadStoreBase(name, Name, ea_code, memacc_code, mem_flags, inst_flags,
Change how memory operands are handled in ISA descriptions. Should enable implementation of split-phase timing loads with new memory model. May create slight timing differences under FullCPU, as I believe we were not handling software prefetches correctly before when the split MemAcc/Exec model was used. I haven't looked into this in any detail though. arch/alpha/isa/decoder.isa: HwLoadStore format split into separate HwLoad and HwStore formats. Copy instructions now fall under MiscPrefetch format. Mem_write_result is now just write_result in store conditionals. arch/alpha/isa/mem.isa: Split MemAccExecute and LoadStoreExecute templates into separate templates for loads and stores; now that memory operands are handled differently from registers, it's impossible to have a single template serve both. Also unified the handling of "regular" prefetches (loads to r31) and "misc" prefetches (e.g., wh64) under the new scheme. It looks like SW prefetches were not handled correctly in FullCPU up til now, since we generated an execute() method for the outer instruction but didn't generate a proper method for MemAcc::execute() (instead getting a default no-op method for that). arch/alpha/isa/pal.isa: Split HwLoadStore into separate HwLoad and HwStore formats to select proper template (see change to mem.isa in this changeset). arch/isa_parser.py: Stop trying to treat memory operands like register operands, since we never used them in a uniform way anyway, and it made it impossible to do split-phase loads as needed for the new CPU model. Now there's no more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does register operands, and the template code is responsible for formulating the call to the memory system. Right now the only thing exported by InstObjParams is a new attribute 'mem_acc_size' which gives the memory access size in bits, though more attributes can be added if needed. Also moved code in findOperands() method to OperandDescriptorList.__init__(), which is where it belongs. --HG-- extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
2006-02-10 15:12:55 +01:00
base_class = 'MemoryNoDisp', exec_template_base = 'Misc')
}};