> On Sept. 20, 2013, 7:46 a.m., Andreas Hansson wrote: > > I do not really see the point here. Could you be more clear around what > > this "integration" would involve? In any case, I would vote not to include > > the McPat source in gem5, and if it really needs to live in the source > > tree, get the users that want to have it to clone/checkout in ext/mcpat. > > Nathan Binkert wrote: > I agree. Shouldn't McPAT be maintained in its own repository? Shouldn't > it just use EXTRAS? > > Brad Beckmann wrote: > Thanks Tony for posting this initial patch. I know it has been a few > weeks, but want to restart this discussion. We would like to include this > version of McPAT directly into gem5 so that we can keep it "in sync" with the > gem5 output. We fear that if we move it to a separate source tree, it will > become stale with the constantly evolving gem5 statistics and configurations. > There are also secondary benefits from AMD's perspective in keeping it the > same respository that I'd rather not get into. > > So what is the benefit of having it in a separate repository? It isn't > that much code and there is already many other external tools/scripts in the > gem5 ext directory. > > Yasuko Eckert wrote: > I wanted to restart this discussion. I know there are many demands from > different people to add McPAT to gem5, so it would be nice to check in this > original McPAT patch as well as three improvement patches from AMD (#2151, > 2117, and 2118) very soon. Could we shoot for the end of the week?
Hi Yasuko, Thanks again for all the effort. I would still like to see if it is really necessary to add all the sources to the gem5 tree, as opposed to simply cloning/downloading mcpat in a ext/ folder like we do for e.g. DRAMSim2. I'd imagine #2151 and #2118 could easily be added next to a stock mcpat. The big question is #2117. As far as I understand, the patch mainly improves the run-time and memory requirements of mcpat. How does that compare to the stock mcpat? Also, how does it compare to mcpat 1.0? I'm personally not familiar with the improvements/differences between the two versions. If #2117 is really critical, perhaps we could consider having the gem5 build process apply it to the downloaded mcpat (I have proposed this in the past). - Andreas ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2021/#review4721 ----------------------------------------------------------- On Dec. 9, 2013, 10:49 p.m., Anthony Gutierrez wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem5.org/r/2021/ > ----------------------------------------------------------- > > (Updated Dec. 9, 2013, 10:49 p.m.) > > > Review request for Default. > > > Repository: gem5 > > > Description > ------- > > Changeset 9994:23fa89b810a1 > --------------------------- > ext: add McPAT source > > this adds the source for mcpat, a power, area, and timing modeling framework. > this will allow for the future integration of mcpat into gem5. > > > Diffs > ----- > > ext/mcpat/ARM_A9.xml PRE-CREATION > ext/mcpat/ARM_A9_2000.xml PRE-CREATION > ext/mcpat/ARM_A9_800.xml PRE-CREATION > ext/mcpat/Alpha21364.xml PRE-CREATION > ext/mcpat/Niagara1.xml PRE-CREATION > ext/mcpat/Niagara1_sharing.xml PRE-CREATION > ext/mcpat/Niagara1_sharing_DC.xml PRE-CREATION > ext/mcpat/Niagara1_sharing_SBT.xml PRE-CREATION > ext/mcpat/Niagara1_sharing_ST.xml PRE-CREATION > ext/mcpat/Niagara2.xml PRE-CREATION > ext/mcpat/Penryn.xml PRE-CREATION > ext/mcpat/README PRE-CREATION > ext/mcpat/XML_Parse.h PRE-CREATION > ext/mcpat/XML_Parse.cc PRE-CREATION > ext/mcpat/Xeon.xml PRE-CREATION > ext/mcpat/arch_const.h PRE-CREATION > ext/mcpat/array.h PRE-CREATION > ext/mcpat/array.cc PRE-CREATION > ext/mcpat/basic_components.h PRE-CREATION > ext/mcpat/basic_components.cc PRE-CREATION > ext/mcpat/cacti/README PRE-CREATION > ext/mcpat/cacti/Ucache.h PRE-CREATION > ext/mcpat/cacti/Ucache.cc PRE-CREATION > ext/mcpat/cacti/arbiter.h PRE-CREATION > ext/mcpat/cacti/arbiter.cc PRE-CREATION > ext/mcpat/cacti/area.h PRE-CREATION > ext/mcpat/cacti/area.cc PRE-CREATION > ext/mcpat/cacti/bank.h PRE-CREATION > ext/mcpat/cacti/bank.cc PRE-CREATION > ext/mcpat/cacti/basic_circuit.h PRE-CREATION > ext/mcpat/cacti/basic_circuit.cc PRE-CREATION > ext/mcpat/cacti/batch_tests PRE-CREATION > ext/mcpat/cacti/cache.cfg PRE-CREATION > ext/mcpat/cacti/cacti.i PRE-CREATION > ext/mcpat/cacti/cacti.mk PRE-CREATION > ext/mcpat/cacti/cacti_interface.h PRE-CREATION > ext/mcpat/cacti/cacti_interface.cc PRE-CREATION > ext/mcpat/cacti/component.h PRE-CREATION > ext/mcpat/cacti/component.cc PRE-CREATION > ext/mcpat/cacti/const.h PRE-CREATION > ext/mcpat/cacti/contention.dat PRE-CREATION > ext/mcpat/cacti/crossbar.h PRE-CREATION > ext/mcpat/cacti/crossbar.cc PRE-CREATION > ext/mcpat/cacti/decoder.h PRE-CREATION > ext/mcpat/cacti/decoder.cc PRE-CREATION > ext/mcpat/cacti/htree2.h PRE-CREATION > ext/mcpat/cacti/htree2.cc PRE-CREATION > ext/mcpat/cacti/io.h PRE-CREATION > ext/mcpat/cacti/io.cc PRE-CREATION > ext/mcpat/cacti/main.cc PRE-CREATION > ext/mcpat/cacti/makefile PRE-CREATION > ext/mcpat/cacti/mat.h PRE-CREATION > ext/mcpat/cacti/mat.cc PRE-CREATION > ext/mcpat/cacti/nuca.h PRE-CREATION > ext/mcpat/cacti/nuca.cc PRE-CREATION > ext/mcpat/cacti/parameter.h PRE-CREATION > ext/mcpat/cacti/parameter.cc PRE-CREATION > ext/mcpat/cacti/router.h PRE-CREATION > ext/mcpat/cacti/router.cc PRE-CREATION > ext/mcpat/cacti/subarray.h PRE-CREATION > ext/mcpat/cacti/subarray.cc PRE-CREATION > ext/mcpat/cacti/technology.cc PRE-CREATION > ext/mcpat/cacti/uca.h PRE-CREATION > ext/mcpat/cacti/uca.cc PRE-CREATION > ext/mcpat/cacti/wire.h PRE-CREATION > ext/mcpat/cacti/wire.cc PRE-CREATION > ext/mcpat/core.h PRE-CREATION > ext/mcpat/core.cc PRE-CREATION > ext/mcpat/globalvar.h PRE-CREATION > ext/mcpat/interconnect.h PRE-CREATION > ext/mcpat/interconnect.cc PRE-CREATION > ext/mcpat/iocontrollers.h PRE-CREATION > ext/mcpat/iocontrollers.cc PRE-CREATION > ext/mcpat/logic.h PRE-CREATION > ext/mcpat/logic.cc PRE-CREATION > ext/mcpat/main.cc PRE-CREATION > ext/mcpat/makefile PRE-CREATION > ext/mcpat/mcpat.mk PRE-CREATION > ext/mcpat/mcpatXeonCore.mk PRE-CREATION > ext/mcpat/memoryctrl.h PRE-CREATION > ext/mcpat/memoryctrl.cc PRE-CREATION > ext/mcpat/noc.h PRE-CREATION > ext/mcpat/noc.cc PRE-CREATION > ext/mcpat/processor.h PRE-CREATION > ext/mcpat/processor.cc PRE-CREATION > ext/mcpat/results/A9_2000 PRE-CREATION > ext/mcpat/results/A9_2000_withIOC PRE-CREATION > ext/mcpat/results/A9_800 PRE-CREATION > ext/mcpat/results/Alpha21364 PRE-CREATION > ext/mcpat/results/Alpha21364_90nm PRE-CREATION > ext/mcpat/results/Penryn PRE-CREATION > ext/mcpat/results/T1 PRE-CREATION > ext/mcpat/results/T1_DC_64 PRE-CREATION > ext/mcpat/results/T1_SBT_64 PRE-CREATION > ext/mcpat/results/T1_ST_64 PRE-CREATION > ext/mcpat/results/T2 PRE-CREATION > ext/mcpat/results/Xeon_core PRE-CREATION > ext/mcpat/results/Xeon_uncore PRE-CREATION > ext/mcpat/sharedcache.h PRE-CREATION > ext/mcpat/sharedcache.cc PRE-CREATION > ext/mcpat/technology_xeon_core.cc PRE-CREATION > ext/mcpat/version.h PRE-CREATION > ext/mcpat/xmlParser.h PRE-CREATION > ext/mcpat/xmlParser.cc PRE-CREATION > > Diff: http://reviews.gem5.org/r/2021/diff/ > > > Testing > ------- > > > Thanks, > > Anthony Gutierrez > > _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
