Re: [PATCH 0/6] cpufreq: Add 'load_table' debugfs file to show colleced CPUs load
On 9 July 2013 13:27, Chanwoo Choi wrote: > So, I'd like you to review patch1,patch2, patch6. If you with that I resend > patch1/2/6, > I will resend new patchset incluing in patch1/2/6. Yeah, I will do that only. Don't resend anything for now. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 0/6] cpufreq: Add 'load_table' debugfs file to show colleced CPUs load
On 07/09/2013 03:50 PM, Viresh Kumar wrote: > On 5 July 2013 14:16, Chanwoo Choi wrote: >> Second, previous performance/powersave governor haven't calculated CPUs load >> becuase these governor didn't change CPU frequency according to CPUs load. >> But, >> load_table debugfs file always should indicate the collected CPUs data >> regardless >> of the kind of cpufreq governor. So, the patch3/4/5 implement that >> performance/ >> powersave governor will check periodically CPUs load by calling >> dbs_check_cpu() >> with timer. > > I raised a query on how can we call dbs_check_cpu() from > performance/powersave? Also, calling this routine will degrade > performance without any sense. So, I vote not for doing it. You're right. The performance/powersave don't usually need calling operation of dbs_check_cpu(). Only, this patch aims at checking CPUs load on load_table debugfs file. I'm going to consider more efficient way than this patchset. For example, But, following patctes haven't the dependency about upper description about performance/powersave. If user changes cpufreq governor from ondemand/conservative to performance/powersave, patch2 did reset all of the data for load_table. cpufreq: Add debugfs directory for cpufreq cpufreq: stats: Add 'load_table' debugfs file to show accumulated data of CPUs Documentation: cpufreq: load_table: Update load_table debugfs file documentation So, I'd like you to review patch1,patch2, patch6. If you with that I resend patch1/2/6, I will resend new patchset incluing in patch1/2/6. Thanks, Best Regards, Chanwoo Choi -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 0/6] cpufreq: Add 'load_table' debugfs file to show colleced CPUs load
On 5 July 2013 14:16, Chanwoo Choi wrote: > Second, previous performance/powersave governor haven't calculated CPUs load > becuase these governor didn't change CPU frequency according to CPUs load. > But, > load_table debugfs file always should indicate the collected CPUs data > regardless > of the kind of cpufreq governor. So, the patch3/4/5 implement that > performance/ > powersave governor will check periodically CPUs load by calling > dbs_check_cpu() > with timer. I raised a query on how can we call dbs_check_cpu() from performance/powersave? Also, calling this routine will degrade performance without any sense. So, I vote not for doing it. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 0/6] cpufreq: Add 'load_table' debugfs file to show colleced CPUs load
On 5 July 2013 14:16, Chanwoo Choi cw00.c...@samsung.com wrote: Second, previous performance/powersave governor haven't calculated CPUs load becuase these governor didn't change CPU frequency according to CPUs load. But, load_table debugfs file always should indicate the collected CPUs data regardless of the kind of cpufreq governor. So, the patch3/4/5 implement that performance/ powersave governor will check periodically CPUs load by calling dbs_check_cpu() with timer. I raised a query on how can we call dbs_check_cpu() from performance/powersave? Also, calling this routine will degrade performance without any sense. So, I vote not for doing it. -- To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 0/6] cpufreq: Add 'load_table' debugfs file to show colleced CPUs load
On 07/09/2013 03:50 PM, Viresh Kumar wrote: On 5 July 2013 14:16, Chanwoo Choi cw00.c...@samsung.com wrote: Second, previous performance/powersave governor haven't calculated CPUs load becuase these governor didn't change CPU frequency according to CPUs load. But, load_table debugfs file always should indicate the collected CPUs data regardless of the kind of cpufreq governor. So, the patch3/4/5 implement that performance/ powersave governor will check periodically CPUs load by calling dbs_check_cpu() with timer. I raised a query on how can we call dbs_check_cpu() from performance/powersave? Also, calling this routine will degrade performance without any sense. So, I vote not for doing it. You're right. The performance/powersave don't usually need calling operation of dbs_check_cpu(). Only, this patch aims at checking CPUs load on load_table debugfs file. I'm going to consider more efficient way than this patchset. For example, But, following patctes haven't the dependency about upper description about performance/powersave. If user changes cpufreq governor from ondemand/conservative to performance/powersave, patch2 did reset all of the data for load_table. cpufreq: Add debugfs directory for cpufreq cpufreq: stats: Add 'load_table' debugfs file to show accumulated data of CPUs Documentation: cpufreq: load_table: Update load_table debugfs file documentation So, I'd like you to review patch1,patch2, patch6. If you with that I resend patch1/2/6, I will resend new patchset incluing in patch1/2/6. Thanks, Best Regards, Chanwoo Choi -- To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 0/6] cpufreq: Add 'load_table' debugfs file to show colleced CPUs load
On 9 July 2013 13:27, Chanwoo Choi cw00.c...@samsung.com wrote: So, I'd like you to review patch1,patch2, patch6. If you with that I resend patch1/2/6, I will resend new patchset incluing in patch1/2/6. Yeah, I will do that only. Don't resend anything for now. -- To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
[PATCH 0/6] cpufreq: Add 'load_table' debugfs file to show colleced CPUs load
This patchset add 'load_table' debugfs file to provide collected CPUs data. The load_table debugfs file gives below CPU datas. - measured time - old CPU frequency - new CPU frequency - each CPU load These data will mean the change of CPU frequency according to CPUs load at specific measured time. Also, the user can determine the storage size of colleced CPUs data. The range is from 10 to 1000. Second, previous performance/powersave governor haven't calculated CPUs load becuase these governor didn't change CPU frequency according to CPUs load. But, load_table debugfs file always should indicate the collected CPUs data regardless of the kind of cpufreq governor. So, the patch3/4/5 implement that performance/ powersave governor will check periodically CPUs load by calling dbs_check_cpu() with timer. Finally, the patch 6 explain the detailed description of load_table debugfs file. Thanks, Chanwoo Choi [Test Result] - the kind of SoC : Samsung EXYNOS4412 - a range of frequency : 200 ~ 1400MHz - Ondemand governor and the number of online CPU is 4 Time(ms) Old Freq(Hz) New Freq(Hz) CPU0 CPU1 CPU2 CPU3 23820 50 50 53 86 273 23920 50 40 66 40 042 24020 40 40 71 71 10 52 24120 40 30 33 27 45 65 24220 30 30 437 71 34 24320 30 30 185 38 16 24420 30 20 641 15 51 24520 20 20 12 62 151 24620 20 20 951 058 24720 20 20 32 32 11 27 - Performance governor and the number of online CPU is 4 Time(ms) Old Freq(Hz) New Freq(Hz) CPU0 CPU1 CPU2 CPU3 3425930140 140 0000 3425945140 140 0000 3425960140 140 0000 3427105140 140 0000 3427109140 140 0033 100 3428135140 140 0000 3428425140 140 0000 3429385140 140 0000 3429400140 140 0000 3429415140 140 0000 - Powersave governor and the number of online CPU is 4 Time(ms) Old Freq(Hz) New Freq(Hz) CPU0 CPU1 CPU2 CPU3 345194520 20 3002 345393020 20 0101 345394520 20 2200 345396020 20 1010 345506520 20 0000 345507520 20 1510 345557020 20 0000 345560520 20 0003 345635020 20 0000 345636020 20 41 239 43 - Powersave governor and the number of online CPU is 2 (CPU[1-2] is offline) Time(ms) Old Freq(Hz) New Freq(Hz) CPU0 CPU3 350193020 20 00 350204020 20 00 350210020 20 75 16 350257520 20 00 350302520 20 11 2 350310020 20 62 14 350345520 20 10 350393020 20 39 350400020 20 65 15 350444020 20 10 Chanwoo Choi (6): cpufreq: Add debugfs directory for cpufreq cpufreq: stats: Add 'load_table' debugfs file to show accumulated data of CPUs cpufreq: Update governor core to support all governors cpufreq: performance: Add support to collect CPUs load periodically cpufreq: powersave: Add support to collect CPUs load periodically Documentation: cpufreq: load_table: Update load_table debugfs file documentation Documentation/cpu-freq/cpufreq-stats.txt | 39 - drivers/cpufreq/Kconfig | 6 + drivers/cpufreq/cpufreq.c| 61 drivers/cpufreq/cpufreq_governor.c | 37 - drivers/cpufreq/cpufreq_governor.h | 11 ++ drivers/cpufreq/cpufreq_performance.c| 156 ++- drivers/cpufreq/cpufreq_powersave.c | 158 ++- drivers/cpufreq/cpufreq_stats.c | 256 --- include/linux/cpufreq.h | 7 + 9 files changed, 692 insertions(+), 39 deletions(-) -- 1.8.0 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
[PATCH 0/6] cpufreq: Add 'load_table' debugfs file to show colleced CPUs load
This patchset add 'load_table' debugfs file to provide collected CPUs data. The load_table debugfs file gives below CPU datas. - measured time - old CPU frequency - new CPU frequency - each CPU load These data will mean the change of CPU frequency according to CPUs load at specific measured time. Also, the user can determine the storage size of colleced CPUs data. The range is from 10 to 1000. Second, previous performance/powersave governor haven't calculated CPUs load becuase these governor didn't change CPU frequency according to CPUs load. But, load_table debugfs file always should indicate the collected CPUs data regardless of the kind of cpufreq governor. So, the patch3/4/5 implement that performance/ powersave governor will check periodically CPUs load by calling dbs_check_cpu() with timer. Finally, the patch 6 explain the detailed description of load_table debugfs file. Thanks, Chanwoo Choi [Test Result] - the kind of SoC : Samsung EXYNOS4412 - a range of frequency : 200 ~ 1400MHz - Ondemand governor and the number of online CPU is 4 Time(ms) Old Freq(Hz) New Freq(Hz) CPU0 CPU1 CPU2 CPU3 23820 50 50 53 86 273 23920 50 40 66 40 042 24020 40 40 71 71 10 52 24120 40 30 33 27 45 65 24220 30 30 437 71 34 24320 30 30 185 38 16 24420 30 20 641 15 51 24520 20 20 12 62 151 24620 20 20 951 058 24720 20 20 32 32 11 27 - Performance governor and the number of online CPU is 4 Time(ms) Old Freq(Hz) New Freq(Hz) CPU0 CPU1 CPU2 CPU3 3425930140 140 0000 3425945140 140 0000 3425960140 140 0000 3427105140 140 0000 3427109140 140 0033 100 3428135140 140 0000 3428425140 140 0000 3429385140 140 0000 3429400140 140 0000 3429415140 140 0000 - Powersave governor and the number of online CPU is 4 Time(ms) Old Freq(Hz) New Freq(Hz) CPU0 CPU1 CPU2 CPU3 345194520 20 3002 345393020 20 0101 345394520 20 2200 345396020 20 1010 345506520 20 0000 345507520 20 1510 345557020 20 0000 345560520 20 0003 345635020 20 0000 345636020 20 41 239 43 - Powersave governor and the number of online CPU is 2 (CPU[1-2] is offline) Time(ms) Old Freq(Hz) New Freq(Hz) CPU0 CPU3 350193020 20 00 350204020 20 00 350210020 20 75 16 350257520 20 00 350302520 20 11 2 350310020 20 62 14 350345520 20 10 350393020 20 39 350400020 20 65 15 350444020 20 10 Chanwoo Choi (6): cpufreq: Add debugfs directory for cpufreq cpufreq: stats: Add 'load_table' debugfs file to show accumulated data of CPUs cpufreq: Update governor core to support all governors cpufreq: performance: Add support to collect CPUs load periodically cpufreq: powersave: Add support to collect CPUs load periodically Documentation: cpufreq: load_table: Update load_table debugfs file documentation Documentation/cpu-freq/cpufreq-stats.txt | 39 - drivers/cpufreq/Kconfig | 6 + drivers/cpufreq/cpufreq.c| 61 drivers/cpufreq/cpufreq_governor.c | 37 - drivers/cpufreq/cpufreq_governor.h | 11 ++ drivers/cpufreq/cpufreq_performance.c| 156 ++- drivers/cpufreq/cpufreq_powersave.c | 158 ++- drivers/cpufreq/cpufreq_stats.c | 256 --- include/linux/cpufreq.h | 7 + 9 files changed, 692 insertions(+), 39 deletions(-) -- 1.8.0 -- To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/