tag:blogger.com,1999:blog-6469704299235308349.post2144460983458828154..comments2024-02-09T16:24:46.087+11:00Comments on -ck hacking: BFS 0.441, 3.11-ck1ckhttp://www.blogger.com/profile/02904761195451530213noreply@blogger.comBlogger123125tag:blogger.com,1999:blog-6469704299235308349.post-29870216224840710892019-01-12T02:03:23.540+11:002019-01-12T02:03:23.540+11:00ThanksThanksthomasmappbehttps://www.blogger.com/profile/12644974527387354272noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-78813323959468100442014-01-10T17:46:01.118+11:002014-01-10T17:46:01.118+11:00@ graysky: I really hope to see an updated version...@ graysky: I really hope to see an updated version of your report (http://repo-ck.com/bench/cpu_schedulers_compared.pdf) when bfs for 3.13 comes out :-)Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-4827359670261725112013-11-19T00:06:58.093+11:002013-11-19T00:06:58.093+11:00@Ulrich,
thx, "overread" it, because th...@Ulrich,<br /><br />thx, "overread" it, because there is no dead lock boot warning here and googled for the wrong sentence. Thx.<br /><br />CU sysitosMikehttps://www.blogger.com/profile/12391045215046883684noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-91683877358413907032013-11-18T05:38:11.925+11:002013-11-18T05:38:11.925+11:00Hi, Mike,
the INTEL_PSTATE appears to be not suppo...Hi, Mike,<br />the INTEL_PSTATE appears to be not supported for my "Penryn" Core2Duo P8400, neither by hardware nor by the kernel pstate driver? Kernel doesn't tell anything even if booting with commandline added "intel_pstate=enable". But I only did a short, i.e. too long, web search that didn't provide results stating at which generation of Intel cpus this feature was implemented. I assume going from Core i3 to newer.<br /><br />Btw., lowering the dirty_background_* and dirty_* in the vm would follow CK's approach until 3.7-ck1, setting both *_ratio to 1 in 'mm-decrease_default_dirty_ratio-1.patch', what he dropped since 3.8. (For my old unicore system this "=1" setup was unusable at that former times.)<br /><br />I, personally, will wait some weeks for openSUSE 13.1 and the repos to mature. I had too many bad experiences in the past when upgrading openSUSEs shortly after release date.<br /><br />Best regards, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-28506286128288816482013-11-18T04:07:09.464+11:002013-11-18T04:07:09.464+11:00@Mike ,
the answer Chen gave me on 2013-10-10 on ...@Mike , <br />the answer Chen gave me on 2013-10-10 on this page not good enough?<br />UlrichAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-49937254498897274442013-11-17T10:07:32.989+11:002013-11-17T10:07:32.989+11:00Hi Manuel,
a "ls /sys/devices/system/cpu/int...Hi Manuel,<br /><br />a "ls /sys/devices/system/cpu/intel_pstate" should output something like that: "max_perf_pct min_perf_pct no_turbo". Than PSTATE is active.<br /><br />For my vm settings I even must lower it, because writing on an USB thumb drive with ntfs3g still blocks the system.<br />So my /etc/sysctl.conf contains now:<br />vm.dirty_bytes = 209715200<br />vm.dirty_background_bytes = 104857600<br />(Btw. you can use either bytes or ratio values there)<br /><br />cu sysitos<br /><br />PS: Using OpenSuse 13.1 now ;)Mikehttps://www.blogger.com/profile/12391045215046883684noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-75561683585928698602013-11-17T09:28:18.211+11:002013-11-17T09:28:18.211+11:00Mmh, I'm not sure about the X86_INTEL_PSTATE, ...Mmh, I'm not sure about the X86_INTEL_PSTATE, it is activated in my kernel .config since months. But how can I be sure, that it's active or see the difference, if not? The active driver is acpi-cpufreq, the actual governor is ondemand.<br /><br />Regarding the consensus about the vm tunables, I have to say that I had ported my best-effort settings from -unicore- to my new system -dualcore-: Now seeing that Mike's (sysitos') settings are quite better for multicore systems.<br />4 > /proc/sys/vm/dirty_ratio (default openSUSE 12.3 @ 20)<br />2 > /proc/sys/vm/dirty_background_ratio (default -"- @ 10)<br />These seem to eliminate most of the A/V hickups I had with heavy swapping.<br />I still keep my opinion, that it'll remain a per system, per user job to adjust these for his/her own, what we don't want for linux future.<br /><br />Best regards, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-45862241405098751012013-11-17T02:25:31.673+11:002013-11-17T02:25:31.673+11:00Hi Chen,
maybe I miss something, but could you pl...Hi Chen,<br /><br />maybe I miss something, but could you please describe the advantages of your circle dead-lock patch and why it should be used.<br />As I see in the patch, you set the mutex_lock one time for the whole systems and not like Con for every CPU (which does mean for my i7 8x lock and unlock).<br /><br />Thanks.<br /><br />Btw, maybe you can contribute your patch to the ZEN Kernel team.<br /><br />cu sysitosMikehttps://www.blogger.com/profile/12391045215046883684noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-34519002635858123932013-11-16T18:13:51.803+11:002013-11-16T18:13:51.803+11:00Unfortunately, this BFS port breaks resuming from ...Unfortunately, this BFS port breaks resuming from hibernation for me :(. Shows nothing but blank screen and hangs.Oleksandr Natalenkohttps://www.blogger.com/profile/12098091624630953604noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-19172481657367686692013-11-16T12:21:37.967+11:002013-11-16T12:21:37.967+11:00Hi Manuel,
as I already wrote, if you are on Open...Hi Manuel,<br /><br />as I already wrote, if you are on OpenSuse (like me) or any other non Ubuntu distro, you don't need the ondemand governor patches. Ubuntu only forgot to reenable the X86_INTEL_PSTATE, which you should use for your Intel Core 2 Duo. It is much better than the other governors and handles the CPU clock speed for its own. But maybe Con could say more, if and how BFS would influence there.<br /><br />PS: As you can read in Phoronix comments to the benchmark test of the better ondemand code in Linux 3.12, these is only in Ubuntu distros without X86_INTEL_PSTATE.<br /><br />CU sysitos<br />Mikehttps://www.blogger.com/profile/12391045215046883684noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-44900869548692276262013-11-16T10:31:46.920+11:002013-11-16T10:31:46.920+11:00Thanks, will try!Thanks, will try!Oleksandr Natalenkohttps://www.blogger.com/profile/12098091624630953604noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-5094367322295654362013-11-16T00:26:41.263+11:002013-11-16T00:26:41.263+11:00It's a pity -- the results don't show rell...It's a pity -- the results don't show relly relevant performance differences in PTS complex system test (with only KDE up in a konsole) on here:<br /><br />-fdefer-pop / -fno-defer-pop<br />159 / 161 Tps - PostMark<br />2810.11 / 2810.07 MB/s - RAMspeed integer<br />3095.08 / 3077.63 MB/s - RAMspeed floating point<br />106.47 / 106.38 s - C-Ray<br />6027.83 / 6003.55 Rps - Apache<br /><br />Best regards, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-82809362029128456472013-11-15T20:32:56.921+11:002013-11-15T20:32:56.921+11:00For those who waiting bfs on 3.12, I have ported b...For those who waiting bfs on 3.12, I have ported bfs-0441 to 3.12, there is 3 conflicts, but seems that are minor ones. After resolved the conflicts and build the kernel, it runs on my core2 machine.<br /><br />Before ck release new version of bfs on 3.12, you can try this out.<br /><br />bfs patch at<br />https://bitbucket.org/alfredchen/linux-gc/commits/b2912adfc8af58528e5a9d846e4873c1caa67331/raw/<br /><br />And also my patch to fix the circle dead-lock<br />https://bitbucket.org/alfredchen/linux-gc/commits/6266b7678235c85575cfeeb380fc65ae94b0fd67/raw/ <br /><br /><br />All credit goes to ck. :)Alfred Chenhttps://www.blogger.com/profile/03164306846702841944noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-10301350658810998062013-11-15T04:46:33.765+11:002013-11-15T04:46:33.765+11:00sorry, something goes wrong with my firefoxsorry, something goes wrong with my firefoxAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-44202587647184955902013-11-15T04:42:08.298+11:002013-11-15T04:42:08.298+11:00under KDE as a normal user - i´m too lazy for some...under KDE as a normal user - i´m too lazy for something elseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-6050014397744116942013-11-15T03:15:41.756+11:002013-11-15T03:15:41.756+11:00OMG! Sorry! It obviously was to late last night an...OMG! Sorry! It obviously was to late last night and I was too blind to find that "/usr/bin/phoronix-test-suite interactive" would lead to the selections menu. Some follow-up questions: Do you run it with X desktop up and in a console window? As root? Or booted into single user session without X?<br /><br />Thank you, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-27559730043134217692013-11-14T08:45:51.410+11:002013-11-14T08:45:51.410+11:00#!/usr/bin/env xdg-open
[Desktop Entry]
Name=Phoro...#!/usr/bin/env xdg-open<br />[Desktop Entry]<br />Name=Phoronix Test Suite<br />GenericName=Benchmarking Utility<br />GenericName[de]=Benchmarks<br />Comment=An Automated, Open-Source Testing Framework<br />Comment[fr]=Outil pour étalonner les performances de votre ordinateur<br />Exec=phoronix-test-suite interactive<br />StartupNotify=true<br />Icon=phoronix-test-suite<br />Terminal=true<br />Type=Application<br />Categories=System;Monitor;<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-31517048204618408792013-11-14T07:40:14.763+11:002013-11-14T07:40:14.763+11:00Sorry, guys'n gals, I'm quite new with lin...Sorry, guys'n gals, I'm quite new with linux benchmarking.<br />I want to deliver the above questioned performance-diff in the way of the previously posted ones and installed the most recent PTS and miss a GUI or a value in the console version to select "point 3 'complex system test'" as mentioned in "Anonymous 19 October 2013 06:33" in this thread.<br /><br />Thank you for any help,<br />Manuel Krause<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-84142095545463583492013-11-13T21:58:59.819+11:002013-11-13T21:58:59.819+11:00How much does -fno-defer-pop persistent affect per...How much does -fno-defer-pop persistent affect performance? Also, I saw you and Mike suggesting different tweaks and was wondering if there is any consensus about which values to use for the vm tunables on a desktop (activities like web browsing, watching movies, compiling...)Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-35513676875454189832013-11-12T05:06:17.397+11:002013-11-12T05:06:17.397+11:00Thank you, Anonymous @10 November 2013 12:25, that...Thank you, Anonymous @10 November 2013 12:25, that made the -fno-defer-pop persistent! -- And it seems to have a positive effect.<br /><br />I have to say, that I already had applied the ondemand-governor patches when they came up in this blog's thread. <br />Setting /sys/devices/system/cpu/cpufreq/ondemand/sampling_down_factor to sth. > 1 doesn't help. <br />Regarding the vm tunables, I've now:<br />8 > /proc/sys/vm/dirty_ratio (default openSUSE 12.3 @ 20)<br />3 > /proc/sys/vm/dirty_background_ratio (default -"- @ 10)<br />70 > /proc/sys/vm/swappiness (default -"- @ 60)<br />Going lower with these values I'd get more stuttering in A/V playback during swapping. But it's very hard to subjectively distinguish interactivity with each variation of these three tunables.<br />Additionally, setting the BFQ runtime tunable "low_latency" to 0 (i.e. off instead of 1 == on == default) results in worse interactivity. Just for the records.<br />And... then I've set CONFIG_RCU_BOOST_PRIO=94 (results in -95 in htop) letting me to manually adjust the prios of kswapd, kwin, Xorg and pulseaudio via schedtool higher or lower than RCU_BOOST_PRIO. But at the latest from then I'm currently flying blindly, so far.<br />Adjusting rr_interval slightly higher / lower than 6 (BFS default) both decreased interactivity, on here.<br /><br />Best regards, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-43540068376551351302013-11-10T12:25:04.351+11:002013-11-10T12:25:04.351+11:00./Makefile
KBUILD_CFLAGS := -Wall -Wundef -Wstr..../Makefile<br /><br />KBUILD_CFLAGS := -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs \<br /> -fno-strict-aliasing -fno-common \<br /> -Werror-implicit-function-declaration \<br /> -Wno-format-security \<br /> -fno-delete-null-pointer-checks -fno-defer-pop<br />KBUILD_AFLAGS_KERNEL :=<br /><br />I believe...Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-92009225449350699912013-11-10T01:01:28.380+11:002013-11-10T01:01:28.380+11:00Thank you very much for these hints, fo far! Needs...Thank you very much for these hints, fo far! Needs some time for testing. :-)<br /><br />Regarding the cflags & cxxflags: I'm not sure where to add the -fno-defer-pop. I've added them at the end of HOSTCFLAGS & HOSTCXXFLAGS, but reading /usr/src/linux/kernel/bounds.s this is not persistent, -fdefer-pop remains active.<br /><br />Any ideas? Thank you in advance,<br />ManuelAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-28994689179606352222013-11-09T22:08:12.332+11:002013-11-09T22:08:12.332+11:00Hi Manuel,
if you use a Intel Core2Duo, than I wo...Hi Manuel,<br /><br />if you use a Intel Core2Duo, than I would prefere X86_INTEL_PSTATE and switch from ondemand (at least on 3.11, 3.12 should be better with ondemand) to powersave (Laptop) or performance (PC) governor. I had some trouble with it, already described on this blog, with some detailed links.<br /><br />cu sysitosMikehttps://www.blogger.com/profile/12391045215046883684noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-68965875855510063182013-11-09T21:49:19.060+11:002013-11-09T21:49:19.060+11:00Hi Oliver,
I use for the make job the toolsched f...Hi Oliver,<br /><br />I use for the make job the toolsched from Con, you can find it on this blog, but 1 year or so in the past. It does automatically use the schedtool for make (and other user defined apps). But if there are big compile jobs with heavy io, it could be a problem with "vm.dirty_ratio", as described by me below. If the cache is to big, the write back of data to the disk take to long and blocks the other apps.<br /><br />Btw. Con, I do have a question for toolsched. If I use 'sudo make install' than I got a error, that make is not found. Ok, I can use 'su -c "make install" ', but thats not sudo. Any suggestions?<br />Thanks.<br /><br />cu sysitosMikehttps://www.blogger.com/profile/12391045215046883684noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-8439400862348506002013-11-09T05:20:04.536+11:002013-11-09T05:20:04.536+11:00@ Jan Killius - can you give me your config, pleas...@ Jan Killius - can you give me your config, please!?<br /><br />for clarification:<br />YYY=<br />XXX+TICK_CPU_ACCOUNTING instead of XXX+IRQ_TIME_ACCOUNTINGAnonymousnoreply@blogger.com