> On May 24, 2013, 10:07 p.m., Nathan Binkert wrote: > > src/sim/clocked_object.hh, line 194 > > <http://reviews.gem5.org/r/1883/diff/1/?file=35579#file35579line194> > > > > You're making a virtual function call every single time you want to > > calculate a clock edge? That's crazy, no? Shouldn't you cache this value > > locally and have the clock domains propagate updates whenever they happen? > > (which should be comparatively rare) > >
Crazy or not, the impact on the total regression run time seems to be very minor. We can definitely implement the caching. If it's fine with you, I'd suggest to keep that for a separate patch. - Andreas ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/1883/#review4351 ----------------------------------------------------------- On May 24, 2013, 3:33 a.m., Andreas Hansson wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem5.org/r/1883/ > ----------------------------------------------------------- > > (Updated May 24, 2013, 3:33 a.m.) > > > Review request for Default. > > > Description > ------- > > Changeset 9740:6bbd5dbeb97e > --------------------------- > sim: Add the notion of clock domains to all ClockedObjects > > This patch adds the notion of source- and derived-clock domains to the > ClockedObjects. As such, all clock information is moved to the clock > domain, and the ClockedObjects are grouped into domains. > > The clock domains are either source domains, with a specific clock > period, or derived domains that have a parent domain and a divider > (potentially chained). For piece of logic that runs at a derived clock > (a ratio of the clock its parent is running at) the necessary derived > clock domain is created from its corresponding parent clock > domain. For now, the derived clock domain only supports a divider, > thus ensuring a lower speed compared to its parent. Multiplier > functionality implies a PLL logic that has not been modelled yet > (create a separate clock instead). > > The clock domains should be used as a mechanism to provide a > controllable clock source that affects clock for every clocked object > lying beneath it. The clock of the domain can (in a future patch) be > controlled by a handler responsible for dynamic frequency scaling of > the respective clock domains. > > All the config scripts have been retro-fitted with clock domains. For > the System a default SrcClockDomain is created. For CPUs that run at a > different speed than the system, there is a seperate clock domain > created. This domain incorporates the CPU and the associated > caches. As before, Ruby runs under its own clock domain. > > > Diffs > ----- > > configs/common/CacheConfig.py 782b7284de21 > configs/common/Simulation.py 782b7284de21 > configs/example/fs.py 782b7284de21 > configs/example/memtest.py 782b7284de21 > configs/example/ruby_direct_test.py 782b7284de21 > configs/example/ruby_fs.py 782b7284de21 > configs/example/ruby_mem_test.py 782b7284de21 > configs/example/ruby_network_test.py 782b7284de21 > configs/example/ruby_random_test.py 782b7284de21 > configs/example/se.py 782b7284de21 > configs/ruby/MESI_CMP_directory.py 782b7284de21 > configs/ruby/MI_example.py 782b7284de21 > configs/ruby/MOESI_CMP_directory.py 782b7284de21 > configs/ruby/MOESI_CMP_token.py 782b7284de21 > configs/ruby/MOESI_hammer.py 782b7284de21 > configs/ruby/Network_test.py 782b7284de21 > configs/ruby/Ruby.py 782b7284de21 > src/arch/alpha/AlphaSystem.py 782b7284de21 > src/arch/mips/MipsSystem.py 782b7284de21 > src/cpu/BaseCPU.py 782b7284de21 > src/cpu/dummy_checker.cc 782b7284de21 > src/cpu/o3/checker.cc 782b7284de21 > src/dev/Ethernet.py 782b7284de21 > src/mem/ruby/system/RubyMemoryControl.py 782b7284de21 > src/sim/ClockDomain.py PRE-CREATION > src/sim/ClockedObject.py 782b7284de21 > src/sim/SConscript 782b7284de21 > src/sim/clock_domain.hh PRE-CREATION > src/sim/clock_domain.cc PRE-CREATION > src/sim/clocked_object.hh 782b7284de21 > tests/configs/base_config.py 782b7284de21 > tests/configs/inorder-timing.py 782b7284de21 > tests/configs/memtest-ruby.py 782b7284de21 > tests/configs/memtest.py 782b7284de21 > tests/configs/o3-timing-checker.py 782b7284de21 > tests/configs/o3-timing-mp-ruby.py 782b7284de21 > tests/configs/o3-timing-mp.py 782b7284de21 > tests/configs/o3-timing-ruby.py 782b7284de21 > tests/configs/o3-timing.py 782b7284de21 > tests/configs/pc-simple-timing-ruby.py 782b7284de21 > tests/configs/rubytest-ruby.py 782b7284de21 > tests/configs/simple-atomic-dummychecker.py 782b7284de21 > tests/configs/simple-atomic-mp-ruby.py 782b7284de21 > tests/configs/simple-atomic-mp.py 782b7284de21 > tests/configs/simple-atomic.py 782b7284de21 > tests/configs/simple-timing-mp-ruby.py 782b7284de21 > tests/configs/simple-timing-mp.py 782b7284de21 > tests/configs/simple-timing-ruby.py 782b7284de21 > tests/configs/simple-timing.py 782b7284de21 > tests/configs/tgen-simple-dram.py 782b7284de21 > tests/configs/tgen-simple-mem.py 782b7284de21 > tests/configs/twosys-tsunami-simple-atomic.py 782b7284de21 > > Diff: http://reviews.gem5.org/r/1883/diff/ > > > Testing > ------- > > All regressions pass > > > Thanks, > > Andreas Hansson > > _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
