----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2354/ -----------------------------------------------------------
Review request for Default. Repository: gem5 Description ------- Changeset 10316:523ac80e06ae --------------------------- arch, cpu: Factor out the ExecContext into a proper base class We currently generate and compile one version of the ISA code per CPU model. This is obviously wasting a lot of resources at compile time. This changeset factors out the interface into a separate ExecContext class, which also serves as documentation for the interface between CPUs and the ISA code. While doing so, this changeset also fixes up interface inconsistencies between the different CPU models. The main argument for using one set of ISA code per CPU model has always been performance as this avoid indirect branches in the generated code. However, this argument does not hold water. Booting Linux on a simulated ARM system running in atomic mode (opt/10.linux-boot/realview-simple-atomic) is actually 2% faster (compiled using clang 3.4) after applying this patch. Additionally, compilation time is decreased by 35%. Diffs ----- SConstruct 79fde1c67ed8 src/arch/SConscript 79fde1c67ed8 src/arch/arm/isa/includes.isa 79fde1c67ed8 src/arch/isa_parser.py 79fde1c67ed8 src/cpu/SConscript 79fde1c67ed8 src/cpu/base_dyn_inst.hh 79fde1c67ed8 src/cpu/checker/SConsopts 79fde1c67ed8 src/cpu/checker/cpu.hh 79fde1c67ed8 src/cpu/exec_context.hh 79fde1c67ed8 src/cpu/exec_context.cc PRE-CREATION src/cpu/inorder/SConsopts 79fde1c67ed8 src/cpu/inorder/inorder_dyn_inst.hh 79fde1c67ed8 src/cpu/inorder/inorder_dyn_inst.cc 79fde1c67ed8 src/cpu/minor/SConsopts 79fde1c67ed8 src/cpu/minor/exec_context.hh 79fde1c67ed8 src/cpu/nocpu/SConsopts 79fde1c67ed8 src/cpu/o3/SConsopts 79fde1c67ed8 src/cpu/o3/dyn_inst.hh 79fde1c67ed8 src/cpu/ozone/SConsopts 79fde1c67ed8 src/cpu/simple/SConsopts 79fde1c67ed8 src/cpu/simple/base.hh 79fde1c67ed8 src/cpu/simple_thread.cc 79fde1c67ed8 src/cpu/static_inst.hh 79fde1c67ed8 Diff: http://reviews.gem5.org/r/2354/diff/ Testing ------- All regressions pass Thanks, Andreas Hansson _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
