I suspect that something is wrong with your fs.py and it doesn't boot the second CPU. But I can't be sure. I will show you my diff between my dvfs_enabled fs.py and the original one. Please let me know if this works for you.
121,123c121,127 < test_sys.cpu_clk_domain = SrcClockDomain(clock = options.cpu_clock, < voltage_domain = < test_sys.cpu_voltage_domain) --- > test_sys.cpu_clk_domain = [SrcClockDomain(clock = options.cpu_clock, > voltage_domain = > test_sys.cpu_voltage_domain, > domain_id=i) > for i in xrange(np)] > > test_sys.dvfs_handler.domains = test_sys.cpu_clk_domain > test_sys.dvfs_handler.enable = 1 140c144,146 < test_sys.cpu = [TestCPUClass(clk_domain=test_sys.cpu_clk_domain, cpu_id=i) --- > test_sys.cpu = [TestCPUClass(clk_domain=test_sys.cpu_clk_domain[i], > cpu_id=i) > for i in xrange(np)] > test_sys.cpu = [TestCPUClass(socket_id=i) 2015-05-29 10:54 GMT+03:00 冰兲轌地 <[email protected]>: > Hi,Giorgos > When I was into the m5term 3456 to change the options of the dvfs,I found > this directory /sys/devices/system/cpu/cpu0/cpufreq/ does exist,but > /sys/devices/system/cpu/cpu1/cpufreq/ does not exist.I creat 2 cpus , but > only have one can change the options of the dvfs. Ideally, there should be a > "cpufreq" for every core which provides the interface to change the > frequency of the core. What should I do to fix this problem? > > My command line is > ./build/ARM/gem5.opt --debug-flags=DVFS,EnergyCtrl > --debug-file=dfvs_debug.log --stats-file=ValMemLat8MBn2.5percore-stats.txt > configs/example/fs.py --cpu-type=AtomicSimpleCPU -n 2 > --machine-type=VExpress_EMM > --kernel=/home/jinsong/gem5/linux-linaro-tracking-gem5/vmlinux > --dtb-filename=/home/jinsong/gem5/linux-linaro-tracking-gem5/arch/arm/boot/dts/vexpress-v2p-ca15-tc1-gem5_dvfs_per_core_2cpus.dtb > --disk-image=../disks/arm-ubuntu-natty-headless.img '--cpu-clock=[1 GHz,750 > MHz,500 MHz]' > > > ------------------ 原始邮件 ------------------ > 发件人: "Giorgo Kopanas";<[email protected]>; > 发送时间: 2015年5月28日(星期四) 晚上8:09 > 收件人: "gem5 users mailing list"<[email protected]>; > 主题: Re: [gem5-users]回复： 回复： 回复： about DVFS Futher Experiments in Per-core > DVFS > > if you are sure that everything is loading up correctly and your fs.py > is correct you should try to experiment under this folder > /sys/devices/system/cpu/cpu0/cpufreq/ by changing the options of the > dvfs and more. > > ~Giorgos > > 2015-05-28 3:28 GMT-04:00 冰兲轌地 <[email protected]>: >> thank you very much for your reply! >> I changed the fs.py script like this : >> >> test_sys.cpu = [TestCPUClass(socket_id=i) >> for i in xrange(np)] >> test_sys.cpu_clk_domain = [SrcClockDomain(clock = options.cpu_clock, >> voltage_domain = >> test_sys.cpu_voltage_domain, >> domain_id = i) >> for i in xrange(np)] >> >> >> test_sys.dvfs_handler.domains = test_sys.cpu_clk_domain >> test_sys.dvfs_handler.enable = 1 >> >> if options.kernel is not None: >> test_sys.kernel = binary(options.kernel) >> >> if options.script is not None: >> test_sys.readfile = options.script >> >> if options.lpae: >> test_sys.have_lpae = True >> >> if options.virtualisation: >> test_sys.have_virtualization = True >> >> test_sys.init_param = options.init_param >> >> # For now, assign all the CPUs to the same clock domain >> test_sys.cpu = [TestCPUClass(clk_domain=test_sys.cpu_clk_domain[i], >> cpu_id=i) >> for i in xrange(np)] >> if is_kvm_cpu(TestCPUClass) or is_kvm_cpu(FutureClass): >> test_sys.vm = KvmVM() >> >> But I find every core also in the same clock (and voltage) domains. >> Could you tell me where it is wrong?what should I do next? >> >> Bing Liang >> >> >> ------------------ 原始邮件 ------------------ >> 发件人: "Giorgos Kopanas";<[email protected]>; >> 发送时间: 2015年5月26日(星期二) 晚上7:19 >> 收件人: "gem5 users mailing list"<[email protected]>; >> 主题: Re: [gem5-users]回复： 回复： about DVFS Futher Experiments in Per-core DVFS >> >> https://www.mail-archive.com/[email protected]/msg11068.html >> If I remember correctly you have to think how to patch them correctly >> in the fs.py. >> >> You are welcome, >> ~Giorgos >> >> 2015-05-26 7:03 GMT-04:00 冰兲轌地 <[email protected]>: >>> Thanks for your help! I can't find this old mail.Could you give me the >>> link >>> of the mail? >>> >>> Bing Liang >>> >>> >>> ------------------ 原始邮件 ------------------ >>> 发件人: "Giorgos Kopanas";<[email protected]>; >>> 发送时间: 2015年5月26日(星期二) 下午5:31 >>> 收件人: "gem5 users mailing list"<[email protected]>; >>> 主题: Re: [gem5-users]回复： about DVFS Futher Experiments in Per-core DVFS >>> >>> Check in the history of the mailing list there is an old mail saying how >>> to >>> do this. When i was trying it helped me. If you can not find it please >>> tell >>> so i can search for it and link it to you. >>> >>> Giorgos >>> >>> On May 26, 2015 5:29 AM, "Giorgos Kopanas" <[email protected]> >>> wrote: >>>> >>>> On May 26, 2015 4:02 AM, "冰兲轌地" <[email protected]> wrote: >>>>> >>>>> HI Andreas： >>>>> thanks for your reply,Could you tell me the path of BaseCPU object in >>>>> gem5? >>>>> >>>>> Bing Liang >>>>> >>>>> >>>>> ------------------ 原始邮件 ------------------ >>>>> 发件人: "Andreas Hansson";<[email protected]>; >>>>> 发送时间: 2015年5月26日(星期二) 下午3:11 >>>>> 收件人: "gem5 users mailing list"<[email protected]>; >>>>> 主题: Re: [gem5-users] about DVFS Futher Experiments in Per-core DVFS >>>>> >>>>> Hi Bing Liang, >>>>> >>>>> The socket_id is a parameter on the BaseCPU object, and you set it like >>>>> any other parameter, e.g. system.cpu[1].socket_id = 1 >>>>> >>>>> Andreas >>>>> >>>>> From: 冰兲轌地 <[email protected]> >>>>> Reply-To: gem5 users mailing list <[email protected]> >>>>> Date: Tuesday, 26 May 2015 05:59 >>>>> To: gem5-users <[email protected]> >>>>> Subject: [gem5-users] about DVFS Futher Experiments in Per-core DVFS >>>>> >>>>> Hi，everyone: >>>>> when I was test the DVFS experiments in per-core DVFS,the page is >>>>> http://www.m5sim.org/Running_gem5#Experimenting_with_DVFS ,there is a >>>>> sentence I don't know how to do. It is >>>>> >>>>> Change the socket_id to have a separate socket per CPU core >>>>> >>>>> Does anyone konws which file should I modify to change the socket_id >>>>> ，and >>>>> how to change the socket_id？ >>>>> thanks for any help! >>>>> Bing Liang >>>>> >>>>> -- IMPORTANT NOTICE: The contents of this email and any attachments are >>>>> confidential and may also be privileged. If you are not the intended >>>>> recipient, please notify the sender immediately and do not disclose the >>>>> contents to any other person, use it for any purpose, or store or copy >>>>> the >>>>> information in any medium. Thank you. >>>>> >>>>> ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, >>>>> Registered in England & Wales, Company No: 2557590 >>>>> ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1 >>>>> 9NJ, >>>>> Registered in England & Wales, Company No: 2548782 >>>>> >>>>> _______________________________________________ >>>>> gem5-users mailing list >>>>> [email protected] >>>>> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >>> >>> >>> _______________________________________________ >>> gem5-users mailing list >>> [email protected] >>> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >> _______________________________________________ >> gem5-users mailing list >> [email protected] >> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >> >> _______________________________________________ >> gem5-users mailing list >> [email protected] >> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > _______________________________________________ > gem5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > > _______________________________________________ > gem5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users _______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
