tag:blogger.com,1999:blog-6469704299235308349.post3751778680828839256..comments2024-03-28T15:50:13.644+11:00Comments on -ck hacking: linux-4.8-ck4, MuQSS CPU scheduler v0.116ckhttp://www.blogger.com/profile/02904761195451530213noreply@blogger.comBlogger52125tag:blogger.com,1999:blog-6469704299235308349.post-90934352259417966232016-10-29T08:09:45.888+11:002016-10-29T08:09:45.888+11:00New muqss and -ck version coming out soon so if yo...New muqss and -ck version coming out soon so if you were about to build a new kernel for 4.8.5, hold out a few hours :)ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-90421500155231956612016-10-29T00:00:35.250+11:002016-10-29T00:00:35.250+11:00Added a comparison with SMT_NICE enabled or disabl...Added a comparison with SMT_NICE enabled or disabled. The overhead of SMT_NICE enabled is about 1%.<br /><br />PedroAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-58686117084900992692016-10-28T09:00:27.157+11:002016-10-28T09:00:27.157+11:00I've put new results of the runqlat tests with...I've put new results of the runqlat tests with MuQSS 116 in both interactive and non-interactive mode.<br /><br />For the curious, I also tested throughput with different frequency governor on CFS. They make significant differences with make j1, but at high load they seem roughly equivalent. Unfortunately I have no means to see the differences in power usage.<br /><br />PedroAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-71737684619023251052016-10-28T08:32:23.820+11:002016-10-28T08:32:23.820+11:00Yes that is correct.Yes that is correct.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-88439887583648584032016-10-28T08:30:41.029+11:002016-10-28T08:30:41.029+11:00https://www.kernel.org/pub/linux/kernel/people/and...https://www.kernel.org/pub/linux/kernel/people/andrea/misc/31-44-100-1000/31-44-100-1000.html<br /><br />Didn't realize there was so much of a performance hit,<br /><br />so you say with the new changes it should be fine even for e.g. FPS games that require low latency and low-latency audio ?<br /><br />Good news :)kernelOfTruthnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-59593816690056723972016-10-28T07:21:34.569+11:002016-10-28T07:21:34.569+11:00Yes. The main reason for those changes was to impr...Yes. The main reason for those changes was to improve latency on low Hz builds. Latest git muqss has low latency on 100Hz which I'm going to start recommending for the next version.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-43329630110896384002016-10-28T07:08:56.519+11:002016-10-28T07:08:56.519+11:00This warning here remained harmless, means: no neg...This warning here remained harmless, means: no negative effect. You mean, the RCU_NOCB* ones are useless?<br />BR, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-18701530187329006302016-10-28T06:51:40.673+11:002016-10-28T06:51:40.673+11:00Thanks. Those 2 extra features are of no use. They...Thanks. Those 2 extra features are of no use. They just weren't supported by muqss until recently. Fortunately that warning is harmless on resume but I should silence it.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-39690260001216597412016-10-28T01:52:00.164+11:002016-10-28T01:52:00.164+11:00@ck:
Out of curiosity I've also added the two ...@ck:<br />Out of curiosity I've also added the two newest commits upon v0.116 (all six) and enabled the new possible settings: <br />CONFIG_TICK_CPU_ACCOUNTING=y<br />CONFIG_RCU_NOCB_CPU=y<br />CONFIG_RCU_NOCB_CPU_ALL=y<br />(Being in doubt, whether the latter two appear useful.)<br /><br />At resume I get the following warning, you may want to have a look at: http://pastebin.com/FPAbxAv0<br /><br />BR, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-3824377191687450692016-10-27T07:49:30.714+11:002016-10-27T07:49:30.714+11:00@Ck,
my results you have to download them to see:...@Ck,<br /><br />my results you have to download them to see:<br />https://www.dropbox.com/sh/dybposdl9t52u7o/AACT64VoaLQUoV3Yh6iAsS9ea?dl=0<br /><br />you can see that ck kernel with ondemand gives me around 200 less points in unigine-heaven compared to archlinux stock kernel but using performance governour works the same as using no cool and quiet and gives both kernels equal performance again.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-23595297419982285202016-10-27T06:21:18.774+11:002016-10-27T06:21:18.774+11:00@Ck,
switched back to ck kernel without the 4 pat...@Ck,<br /><br />switched back to ck kernel without the 4 patches and again back with the patches. And even though I said they don't change sth they make a little difference especially when I have higher fps in ark it works good and the stutters don't appear. So the only problem I have now is the relativ huge performance loss with ondemand. I will also try some unigine benchmarks if disabling cool n quiet, switching kernel or using performance governour makes any difference.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-53073177384050759072016-10-27T00:44:02.338+11:002016-10-27T00:44:02.338+11:00Would just like to report that latest 4 patches (a...Would just like to report that latest 4 patches (after 0.116 -ck4) fixed problems with games for me, mainly Dota 2.<br />(Jitter, mini-freezes every 10 seconds and FPS fluctuations)<br />I am using latest legacy Nvidia drivers 340.98.<br />So thank you.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-87948564349465709832016-10-26T12:02:11.147+11:002016-10-26T12:02:11.147+11:00@CK:
I commented success too early.
Got the follow...@CK:<br />I commented success too early.<br />Got the following after third resume attempt:<br />http://pastebin.com/dReCcQGm<br /><br />Maybe you'd understand it.<br />BR, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-11031171151585811352016-10-26T10:20:56.009+11:002016-10-26T10:20:56.009+11:00@CK:
Also after making heavy use of /dev/shm backe...@CK:<br />Also after making heavy use of /dev/shm backed by swapping-to-second-disk, in the presence of flash-streamed TV in firefox, the recovery of video and other KDE windows is almost immediate.<br />This is a really remarkable advantage, as we may all know about the swap bottleneck.<br /><br />BR, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-90580526356417366972016-10-26T08:06:32.841+11:002016-10-26T08:06:32.841+11:00@Ck,
im not such a linux expert but as far as I u...@Ck,<br /><br />im not such a linux expert but as far as I understand Cool n Quiet is a bios function like intel eist. It underclocks the cpu when not in use. So im already using ondemand. I will try tommorow some other games because ark is a bad game to test I think I saw sth similar on Saints Row 4. and yes I tried to disable Cool n Quiet and it worked but maybe it's just the extra fps boost that did the trick.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-3787260891424250442016-10-26T07:58:39.711+11:002016-10-26T07:58:39.711+11:00@Anonymous, Ck,
I think I have an answer on this....@Anonymous, Ck,<br /><br />I think I have an answer on this.<br />Cool n quiet for AMD means dynamic frequency/voltage/etc. scaling. I just tried disabling it and frequency stays at max speed at all times, which means it's about the same as running with "performance" governor.<br />In addition, when I run my gaming tests, mainline kernel is always better with "ondemand" governor by a large margin, BFS/MUX using "ondemand" are comparably slow.<br />Check https://docs.google.com/spreadsheets/d/1EayezAsGlJdXjZbS3b9m7YtvtRF-DJ3xrT3hYCvfymQ/edit?usp=sharing , page "Perf. (DRI2), OND GOV", You'll see BFS/MUX reaches about 30 FPS while standard Ubuntu kernel reaches 42 FPS.<br />This explains gains and losses with Anonymous tests.<br /><br />I can't explain this (and won't speculate), but Con can for sure :)<br /><br />Br, EduardoAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-23667557419409786602016-10-26T07:45:44.502+11:002016-10-26T07:45:44.502+11:00I cannot comment on those dmesg lines, as I don...I cannot comment on those dmesg lines, as I don't get them. (Possible reasons: Still stuck with 4.7.10, using HZ_1000 all the time and no shiny modern governors available for my old 2core cpu.)<br /><br />But using the mentioned 4 commits makes a positive difference on here anyways.<br /><br />BR, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-16574797385304235892016-10-26T07:38:30.286+11:002016-10-26T07:38:30.286+11:00Okay, running it right now,
thanks :)Okay, running it right now,<br /><br />thanks :)kernelOfTruthnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-1563179510053577302016-10-26T07:30:37.481+11:002016-10-26T07:30:37.481+11:00@Ck,
Sorry for noise about VRQ, it just seemed ve...@Ck,<br /><br />Sorry for noise about VRQ, it just seemed very similar.<br />No issues with mux 116 + diablo 3 + cool n quiet.<br /><br />Br, EduardoAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-69325665663749518692016-10-26T07:27:22.049+11:002016-10-26T07:27:22.049+11:00I just tried the latest commits and no change.I tr...I just tried the latest commits and no change.I tried ark again and using stock kernel gives me 5 - 10 more fps maybe thats why it doesn't stutter but why is it faster with stock?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-36863355308451888112016-10-26T07:26:27.704+11:002016-10-26T07:26:27.704+11:00As Manuel suggested in the comments below, this ma...As Manuel suggested in the comments below, this may be fixed in git with one of the extra patches post 116. There are some minor bugfixes in there; one or more of them may be related.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-27316044052589864842016-10-26T07:21:32.905+11:002016-10-26T07:21:32.905+11:00Yes, I'd like to second that proposal for thos...Yes, I'd like to second that proposal for those people experiencing lags with plain v0.116: <br />On my system these atm. 4 added commits cure a lazy mouse pointer after periods of being unmoved and graphical window content update delays at switching to top within KDE (also in the presence of video playback, what formerly stuttered at that very moment).<br />Using these addon commits is IMO worth a try.<br /><br />BR, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-36511746608753153102016-10-26T07:20:52.305+11:002016-10-26T07:20:52.305+11:00Con,
for me they tend to manifest under higher lo...Con,<br /><br />for me they tend to manifest under higher load for the CPU<br /><br /><br />just had a kernel compilation and wanted to do some editing on PDF file (via WINE + PDF XChange Editor),<br /><br />out of the sudden the win program (PDF XChange Editor) crashed - which isn't something out of the ordinary,<br /><br />but then everything got VERY sluggish.<br /><br />The mouse pointer hardly moved for lots of seconds then moved a tiny bit (almost slurred), then nothing and several seconds later it moved again.<br /><br />Compiz was running with sched_yield NONE, aking a look at the "cursor" of gnome-terminal<br /><br />it blinked for 2 seconds (2 blinks), then several seconds nothing, then again 2 blinks and long time nothing ...<br /><br /><br />Afaik the state of MuQSS was at a7c0de9472e18df1ab604e24b82b5fc6e6b97d43 (0.116)<br /><br />Kernel compilation load didn't seem to be accounted in this state (switching to vt2 after several minutes) and checking via htop [load was 4-4.5],<br /><br />there also was some delay but terminal input more or less worked<br /><br /><br />after 12+ minutes of kernel compilation and still no finishing (usually takes around 8 minutes)<br /><br />The only thing I could get out of dmesg was the following (not really sure if it helps):<br /><br />[ 1100.731633] NOHZ: local_softirq_pending 02<br />[ 1168.311173] NOHZ: local_softirq_pending 202<br />[ 1168.311235] NOHZ: local_softirq_pending 202<br />[ 1168.312015] NOHZ: local_softirq_pending 202<br />[ 1168.312029] NOHZ: local_softirq_pending 202<br />[ 1168.312208] NOHZ: local_softirq_pending 202<br />[ 1168.312412] NOHZ: local_softirq_pending 202<br />[ 1168.455853] NOHZ: local_softirq_pending 202<br />[ 1168.456871] NOHZ: local_softirq_pending 202<br />[ 1168.456957] NOHZ: local_softirq_pending 202<br />[ 1908.524277] BUG: workqueue lockup - pool cpus=0-7 flags=0x4 nice=0 stuck for 36s!<br />[ 1908.524282] Showing busy workqueues and worker pools:<br />[ 1908.524286] workqueue events_freezable_power_: flags=0x84<br />[ 1908.524287] pwq 0: cpus=0 node=0 flags=0x0 nice=0 active=1/256<br />[ 1908.524290] pending: disk_events_workfn<br />[ 1908.524297] workqueue writeback: flags=0x4e<br />[ 1908.524298] pwq 16: cpus=0-7 flags=0x4 nice=0 active=1/256<br />[ 1908.524299] pending: wb_workfn<br />[ 1908.524322] workqueue kcryptd: flags=0x2a<br />[ 1908.524322] pwq 16: cpus=0-7 flags=0x4 nice=0 active=4/8<br />[ 1908.524324] pending: kcryptd_crypt, kcryptd_crypt, kcryptd_crypt, kcryptd_crypt<br />[ 1908.524331] workqueue edac-poller: flags=0x6000a<br />[ 1908.524332] pwq 16: cpus=0-7 flags=0x4 nice=0 active=1/1<br />[ 1908.524333] pending: edac_mc_workq_function [edac_core]kernelOfTruthnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-53475967627290072482016-10-26T07:20:35.123+11:002016-10-26T07:20:35.123+11:00Please do not report issues with VRQ here. It has ...Please do not report issues with VRQ here. It has nothing to do with my development and only confuses the discussion.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-15614275278497986982016-10-26T07:19:17.724+11:002016-10-26T07:19:17.724+11:00@Ck,
Forgot to add that I play games using "...@Ck,<br /><br />Forgot to add that I play games using "performance" governor, eg. max freq all the time. I have enabled Cool'n Quiet since ages and no other kernels show the issue, except VRQ (will test latest mux soon).<br />Latest tests show that "ondemand" behaviour, using VRQ, does not show the problem. But I'll test again w/o C&Q + VRQ/MUX and post results.<br /><br />Br, EduardoAnonymousnoreply@blogger.com