tag:blogger.com,1999:blog-6469704299235308349.post2731657541011461383..comments2024-03-28T15:50:13.644+11:00Comments on -ck hacking: linux-4.8-ck6, MuQSS version 0.135ckhttp://www.blogger.com/profile/02904761195451530213noreply@blogger.comBlogger60125tag:blogger.com,1999:blog-6469704299235308349.post-15185155601103522642016-12-06T06:43:52.565+11:002016-12-06T06:43:52.565+11:00Try building it without cgroups.Try building it without cgroups.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-85571335966949014242016-12-06T01:42:16.921+11:002016-12-06T01:42:16.921+11:00To run it in muqss, is it: schedtool -I -e virtual...To run it in muqss, is it: schedtool -I -e virtualbox ?<br />VirtualBox worked with bfs for years. Have been following all the blogs since this one to current, and read up on stubs but have found scant information on creating stubs. Have read several wiki's on cgroups (incl Arch & Redhat), and scheduling and looking at the example for amarok, is best I can comprehend of it.<br />I'm currently arch 4.8.11-1 stock since vbox quit working in ck.<br />Thank you.<br /><br />-billAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-43846287255292886732016-11-12T06:07:56.589+11:002016-11-12T06:07:56.589+11:00Thank you for reply ck - I figured it did but cou...Thank you for reply ck - I figured it did but could only find cgroups reference to a *.slice file on pc regarding vbox. Setting cgroups to work with muqss way over my head for now, but again thanks.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-27696602820181892752016-11-12T01:38:41.205+11:002016-11-12T01:38:41.205+11:00Turn out the problem with osu! only appear if you ...Turn out the problem with osu! only appear if you set the FPS limiter to Unlimited. I'm unable to reproduce the problem with FPS limiter set to 120fps or 240fps.<br /><br />You may need a complex map such as this one: https://osu.ppy.sh/s/157896<br /><br />Play the map on Xtra difficulty with Auto mod (press F2 and select Auto).<br /><br />I'm currently using MuQSS 1769b2d on Linux 4.8.<br />Intel HD Graphics 5500.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-60415871481211154352016-11-12T00:50:27.328+11:002016-11-12T00:50:27.328+11:00Updated results here:
http://openbenchmarking.org/...Updated results here:<br />http://openbenchmarking.org/result/1611094-LO-CFSVSMUQS89<br /><br />PedroAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-1521426138865933062016-11-11T20:25:30.154+11:002016-11-11T20:25:30.154+11:00BFS has turned into muqss. There is no more BFS.BFS has turned into muqss. There is no more BFS.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-68753972629720292932016-11-11T20:24:40.826+11:002016-11-11T20:24:40.826+11:00Yes it is a container, not a virtual machine, whic...Yes it is a container, not a virtual machine, which means it relies on cgroups entirely which do nothing on muqss.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-60980180624762481872016-11-11T19:38:11.577+11:002016-11-11T19:38:11.577+11:00Thanks for looking into it. During the night i hav...Thanks for looking into it. During the night i have tried the same build with patches 0001,0011,0012,0013 into the laptop with Atom Z520 and the situation is even worst since it boots only 1/10 with panics about "update_process_times.. tick_periodic.. tick_handle_periodic.. local_apic_timer.. smp_apic_timer... EIP: scheduler_tick"<br /><br />Boot again with 4.7.6 + bfs works still great without any issue.<br />I know that look back seems a bad way to go ahead, but... do you plan prepare a set of bfs patches also for kernel 4.8.x?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-89343279108773525892016-11-11T17:00:20.263+11:002016-11-11T17:00:20.263+11:00Does VirtualBox require cgroups or Isochronous sch...Does VirtualBox require cgroups or Isochronous scheduling? <br />https://lkml.org/lkml/2016/10/29/4<br />Windows 7, Ubuntu, and Arch guests on Arch host extremely sluggish to load and run, audio is broken and lags rendering guests unuseable, WinXP guest blue screens on launch. Arch Linux 4.8.6-2-ck (piledriver) since MuQSS.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-86018515654376687322016-11-11T12:12:03.851+11:002016-11-11T12:12:03.851+11:00Unusual error. There's a change I've commi...Unusual error. There's a change I've committed to git which hopefully will help with this. I'll probably have to wrap up changes into a new release to coincide with 4.8.7 as well.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-39350033158284127122016-11-11T11:51:16.922+11:002016-11-11T11:51:16.922+11:00Thank you for your help, i have just prepared a bu...Thank you for your help, i have just prepared a build with your patches 0001,0011,0012,0013 (since i need also BFQ).<br />Trying it in a 64bit test box that has also Virtualbox installed, i obtain:<br /> <br /><br />Nov 10 16:25:07 localhost kernel: usercopy: kernel memory overwrite attempt detected to ffff94c316442dc8 (kmalloc-8) (128 bytes)<br />Nov 10 16:25:07 localhost kernel: ------------[ cut here ]------------<br />Nov 10 16:25:07 localhost kernel: kernel BUG at mm/usercopy.c:75!<br />Nov 10 16:25:07 localhost kernel: invalid opcode: 0000 [#1] SMP<br />Nov 10 16:25:07 localhost kernel: Modules linked in: vboxpci(OE) vboxnetadp(OE) vboxnetflt(OE) vboxdrv(OE) bfq_iosched fuse intel_rapl x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel intel_cstate intel_uncore ppdev intel_rapl_perf snd_hda_codec_hdmi snd_hda_codec_realtek snd_hda_codec_generic snd_hda_intel snd_hda_codec snd_hda_core snd_hwdep iTCO_wdt snd_seq snd_seq_device iTCO_vendor_support snd_pcm snd_timer nuvoton_cir parport_pc snd mei_me i2c_i801 rc_core mei parport soundcore lpc_ich shpchp i2c_smbus tpm_tis tpm_tis_core tpm binfmt_misc i915 i2c_algo_bit drm_kms_helper drm crc32c_intel r8169 serio_raw ata_generic pata_acpi mii fjes video<br />Nov 10 16:25:07 localhost kernel: CPU: 0 PID: 6400 Comm: wineserver Tainted: G OE 4.8.6-201.muqss.bfq.fc24.x86_64 #1<br />Nov 10 16:25:07 localhost kernel: Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./H67M, BIOS P2.10 04/27/2012<br />Nov 10 16:25:07 localhost kernel: task: ffff94c283840000 task.stack: ffff94c283980000<br />Nov 10 16:25:07 localhost kernel: RIP: 0010:[] [] __check_object_size+0x77/0x1dc<br />Nov 10 16:25:07 localhost kernel: RSP: 0018:ffff94c283983ee0 EFLAGS: 00010282<br />Nov 10 16:25:07 localhost kernel: RAX: 000000000000005e RBX: ffff94c316442dc8 RCX: 0000000000000000<br />Nov 10 16:25:07 localhost kernel: RDX: 0000000000000000 RSI: ffff94c31f20df88 RDI: ffff94c31f20df88<br />Nov 10 16:25:07 localhost kernel: RBP: ffff94c283983f00 R08: 00000000000af2bf R09: 0000000000000005<br />Nov 10 16:25:07 localhost kernel: R10: 0000000000000008 R11: 000000000000033b R12: 0000000000000080<br />Nov 10 16:25:07 localhost kernel: R13: 0000000000000000 R14: ffff94c316442e48 R15: ffff94c316442dc8<br />Nov 10 16:25:07 localhost kernel: FS: 00007f16b4130700(0000) GS:ffff94c31f200000(0000) knlGS:0000000000000000<br />Nov 10 16:25:07 localhost kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033<br />Nov 10 16:25:07 localhost kernel: CR2: 00007faf2afef870 CR3: 0000000092fc3000 CR4: 00000000000406f0<br />Nov 10 16:25:07 localhost kernel: Stack:<br />Nov 10 16:25:07 localhost kernel: 0000000000000080 0000000000000080 00007fff7d684db0 0000000000001903<br />Nov 10 16:25:07 localhost kernel: ffff94c283983f48 ffffffffaa0ccdb8 ffff94c316442dc8 000000003ec2e714<br />Nov 10 16:25:07 localhost kernel: 000000000000000f 00000000013de7a0 00000000013e2410 000000000000001f<br />Nov 10 16:25:07 localhost kernel: Call Trace:<br />Nov 10 16:25:07 localhost kernel: [] SyS_sched_setaffinity+0x68/0x100<br />Nov 10 16:25:07 localhost kernel: [] entry_SYSCALL_64_fastpath+0x1a/0xa4<br />Nov 10 16:25:07 localhost kernel: Code: 48 0f 44 d1 48 c7 c6 8e f0 a4 aa 48 c7 c1 02 36 a4 aa 48 0f 44 f1 4d 89 e1 49 89 c0 48 89 d9 48 c7 c7 88 b6 a4 aa e8 04 bb f6 ff <0f> 0b e8 12 af fb ff 85 c0 75 78 48 89 df e8 06 76 e3 ff 84 c0<br />Nov 10 16:25:07 localhost kernel: RIP [] __check_object_size+0x77/0x1dc<br />Nov 10 16:25:07 localhost kernel: RSP <br />Nov 10 16:25:07 localhost kernel: ---[ end trace fcfa1d973eabc43d ]---Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-21669181284342050672016-11-11T09:01:15.200+11:002016-11-11T09:01:15.200+11:00These warnings should now be fixed in git.These warnings should now be fixed in git.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-87277331477546605772016-11-10T20:50:33.126+11:002016-11-10T20:50:33.126+11:00Yes, and that's the same thing I linked you.Yes, and that's the same thing I linked you.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-35599644163182300912016-11-10T20:23:20.591+11:002016-11-10T20:23:20.591+11:00Well, i'm in fact using the pf6 post-factum pa...Well, i'm in fact using the pf6 post-factum patch due to the easy integration into Fedora kernels to deploy the same kernel into all the low performance boxes i have to manage.<br /><br />Would you suggest to use the patch: <br />http://ck.kolivas.org/patches/4.0/4.8/4.8-ck6/patches/0001-The-MuQSS-Multiple-Queue-Skiplist-Scheduler-pronounc.patch<br /><br />"Alone" with nothing else?<br /><br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-59558958856881818842016-11-10T19:27:03.037+11:002016-11-10T19:27:03.037+11:00Thanks. Probably the result of the aggressive time...Thanks. Probably the result of the aggressive timer changes. Try muqss by itself from the -ck split out patches: <a href="http://ck.kolivas.org/patches/4.0/4.8/4.8-ck6/patches/0001-The-MuQSS-Multiple-Queue-Skiplist-Scheduler-pronounc.patch" rel="nofollow">4.8-muqss135</a>ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-55621654793428857342016-11-10T19:24:33.145+11:002016-11-10T19:24:33.145+11:00I'm using your patches since a lot of time to ...I'm using your patches since a lot of time to give some life to an old atom Z520 with great satisfaction.<br /><br />Since MuQSS (now i'm using version 0.135 with HZ 100) i'm experiencing random panics at boot.<br />When the boot process doesn't hangs, the system seems to work very well (except hybernation - but i'm not sure it is related)<br /><br />The same kernel without MuQSS works and boots with no problems.<br /><br />Here attached a part of the trace:<br />[] __hrtimer_run_queues+0xcb/0/x2a0<br />[] ? perf_trace_sched_switch+0x180/0x180<br />[] hrtimer_interrupt+0x8a/0x180<br />[] local_apic_timer_interrupt+0x32/0x60<br />[] smp_apic_timer_interrupt+0x34/0x3c<br />[] apic_timer_interrupt+0x34/0x3cAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-88584994134583904522016-11-10T07:24:03.444+11:002016-11-10T07:24:03.444+11:00If you don't care at all about power savings t...If you don't care at all about power savings then sure periodic timer tick is fine compared to tickless idle. You should also set your cpu frequency governor to performance as well then since no matter how good the governor is, it will never work as well as simply setting it to performance. The only reason to set it to anything else is to save power, reduce heat, and decrease fan noise. They do still use less power when idle with the performance governor, but not as much as with cpu frequency scaling enabled.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-40060112584262392562016-11-10T05:49:04.024+11:002016-11-10T05:49:04.024+11:00But an old desktop system on a power cable does no...But an old desktop system on a power cable does not need to spare little bit of electric power but wants to have a little more throughput: <br />Having an old y2009 Core2Duo I think of<br />timeout>1ms <br />rr_interval>6<br />none CONFIG_NO_HZ_IDLE<br />???<br /><br />Thanx for your work, RalphAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-20393800996953531022016-11-09T22:15:43.595+11:002016-11-09T22:15:43.595+11:00You might be right, but fact is that my Core2 Duo ...You might be right, but fact is that my Core2 Duo machine consumes less CPU (about 10%!) whith periodic timers when using vapoursynth video filters. I compared this with different kernels (your ck and liquorix) - it's always the same!Floriannoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-3575848144810738322016-11-09T20:45:19.555+11:002016-11-09T20:45:19.555+11:00I also ran the same tests you did with the Phoroni...I also ran the same tests you did with the Phoronix test suite.<br />The results are here:<br />http://openbenchmarking.org/result/1611092-LO-CFSVSMUQS19<br /><br />The results shows the same things as the ones you got with MuQSS112.<br />I'll think I'll redo the pgbench test with 'in buffer' scaling instead of the default 'on disk' because it seems the SSD is a bottleneck in the last test.<br /><br />PedroAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-63360818249232795252016-11-09T20:37:17.192+11:002016-11-09T20:37:17.192+11:00I had a quick look at the interbench results. I ha...I had a quick look at the interbench results. I have the impression that overall:<br />- CFS@1000Hz is not that bad<br />- the difference between Interactive values is rather small<br />But you understand theses results better than I do. Am I correct ?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-40384352836201734632016-11-09T09:01:49.461+11:002016-11-09T09:01:49.461+11:00Periodic timer ticks use more power than tickless ...Periodic timer ticks use more power than tickless idle. You can save more power than 100Hz alone by going 100Hz with tickless idle.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-38752959916985976722016-11-09T07:54:18.384+11:002016-11-09T07:54:18.384+11:00I can confirm that with my old hardware periodic t...I can confirm that with my old hardware periodic timer ticks result in a lower latency with 100 HZ. Performance is better. I can't confirm that periodic timer ticks are only for reducing power consume.Floriannoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-56843303636748820722016-11-09T06:16:33.250+11:002016-11-09T06:16:33.250+11:00Thanks Pedro. Those results look more like it and ...Thanks Pedro. Those results look more like it and are consistent in which categories are better on muqss. You shouldn't need to set the load as it will detect the number of logical CPUs automatically.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-50240297276288218522016-11-09T00:12:04.085+11:002016-11-09T00:12:04.085+11:00Thanks, I forgot about that.
I've re-done the ...Thanks, I forgot about that.<br />I've re-done the tests. The first time I used performance governor with CFS@300Hz, and then for every other test powersave (the default).<br />The numbers of loops per ms goes from ~800,000 to ~4,000,000, that's better.<br />The results are more like the ones expected. I can't look at them in detail right now, but they are on the spreadsheet.<br /><br />PedroAnonymousnoreply@blogger.com