tag:blogger.com,1999:blog-6469704299235308349.post6311629886111821692..comments2024-03-28T15:50:13.644+11:00Comments on -ck hacking: linux-4.8-ck2, MuQSS version 0.114ckhttp://www.blogger.com/profile/02904761195451530213noreply@blogger.comBlogger17125tag:blogger.com,1999:blog-6469704299235308349.post-76323247034643760722016-10-22T10:47:12.490+11:002016-10-22T10:47:12.490+11:00You're assuming cfs is accurate to subtick res...You're assuming cfs is accurate to subtick resolution which it is NOT. If cpu load goes up to 1% then it has been hovering at below this for a few ticks in a row and then has accumulated enough to finally register at the resolution and frequency of 'top'.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-2894178697539232312016-10-22T10:45:32.547+11:002016-10-22T10:45:32.547+11:00The nanosecond accounting indeed seems to be a bit...The nanosecond accounting indeed seems to be a bit too fine-grained. For kicks I switched to CFS and an idle machine is indeed idle; neither user space apps nor kthreads show up as fluttery 1%-ers.Holger Hoffstättehttps://www.blogger.com/profile/05303916065597906848noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-23626291421239187772016-10-22T09:58:15.094+11:002016-10-22T09:58:15.094+11:00You can try git 4.8-muqss branch again to see if i...You can try git 4.8-muqss branch again to see if it's fixed now.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-5797443604801015832016-10-22T07:15:36.553+11:002016-10-22T07:15:36.553+11:00No problems on Haswell here (E-1245 V3) with 4.8.3...No problems on Haswell here (E-1245 V3) with 4.8.3<br /><br />cpu load is around 0.02 with one instance of audacious playing MP3 music<br /><br />ThankskernelOfTruthnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-14261316606136593272016-10-22T07:10:12.349+11:002016-10-22T07:10:12.349+11:00Thanks Manuel, that actually helps a lot.Thanks Manuel, that actually helps a lot.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-49982834933447992142016-10-22T07:03:42.176+11:002016-10-22T07:03:42.176+11:00@ck:
There really seems to be an issue with 114, o...@ck:<br />There really seems to be an issue with 114, on my 4.7.9 "full combo" shown as not wanting to hibernate via TOI any more.<br />From my last known good until 114, that was 112+X, I've cherry-picked one commit to revert: 2460840c... and system is working fine again. <br />Please, have a look on it again and re-think the reversion.<br />(Of course I got a one-time dmesg warning at boot after removal of that one commit (newer commits rely on the latter)). For completeness, I've put the warning here: http://pastebin.com/NiUS98CQ<br /><br />Looking forward to a more serious improvement, and<br />BR, as always, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-43162822023591180372016-10-22T00:22:11.430+11:002016-10-22T00:22:11.430+11:00No problems with muqss-114 on either 4.8.3 or 4.8....No problems with muqss-114 on either 4.8.3 or 4.8.4-rc1 (running as we speak). I'm using intel_pstate everywhere though, so maybe that plays a role (vs. schedutil and all the weird bugs still in mainline).<br />The super-fine accounting certainly seems to have an effect on top utilization; ffmpeg now reaches 800% (which seems a tad much, compared to ~790% previously) and a completely idle system was previously measured as 0%, whereas now I have a frequent 1% utilization, probably because the snmpd measures itself and rounds up the now accurately measured values. Or something went wrong in the conversion to µs/jiffies.Holger Hoffstättehttps://www.blogger.com/profile/05303916065597906848noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-29627970275634906522016-10-21T21:45:49.560+11:002016-10-21T21:45:49.560+11:004.8.3-ck2 with BFS is fine, with MuQSS isn't, ...4.8.3-ck2 with BFS is fine, with MuQSS isn't, so probably. Sarenhttps://www.blogger.com/profile/01376294704785882565noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-85791974279418044232016-10-21T21:23:39.988+11:002016-10-21T21:23:39.988+11:00So your problem is 4.8, not muqss? Muqss on 4.7 is...So your problem is 4.8, not muqss? Muqss on 4.7 is only trivially different to 4.8.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-87177403579232528442016-10-21T20:16:01.103+11:002016-10-21T20:16:01.103+11:00I can't even get systemd loading, so no error ...I can't even get systemd loading, so no error messages can be retrived, no tty. It just freezes at "Loading initial ramdisk" .Sarenhttps://www.blogger.com/profile/01376294704785882565noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-5483876285750290142016-10-21T20:14:08.799+11:002016-10-21T20:14:08.799+11:00It is stuck at "Loading initial ramdisk"...It is stuck at "Loading initial ramdisk" and nothing more.<br /><br />BTW 4.7.9 with muqss 114 however is fine too.Sarenhttps://www.blogger.com/profile/01376294704785882565noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-16765656323532560782016-10-21T20:11:02.150+11:002016-10-21T20:11:02.150+11:00Trying to run mpd locks up 4.8.2 and 4.8.3 for me ...Trying to run mpd locks up 4.8.2 and 4.8.3 for me on my Haswell machine. On 4.8.2 it's an instant lock up, no tty, no route to the ssh host, nothing is usable. On 4.8.3 it gets weirder, my shell does not start in new terminal windows, in already open terminal windows commands inputted post-mpd act like they hang indefinitely(unlaunched/dead shell like the others?), trying to log in via ssh seems to hang indefinitely, and finally everything is frozen. The attempted ssh connection still hangs afterwards, and no means of getting into a tty either.DerVerruckteFuchshttps://www.blogger.com/profile/09352176250733116474noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-4267671617327087672016-10-21T19:15:05.263+11:002016-10-21T19:15:05.263+11:00Define "does not work well."Define "does not work well."ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-15839410997774380052016-10-21T19:01:10.580+11:002016-10-21T19:01:10.580+11:00Hi, I am using a system with haswell cpu.
4.8+ ck/...Hi, I am using a system with haswell cpu.<br />4.8+ ck/ck2 does not work well with muqss 112, 114, bfs is ok.<br />4.7.8 with muqss 112 however is fine.Sarenhttps://www.blogger.com/profile/01376294704785882565noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-58496875723630526762016-10-21T15:44:00.574+11:002016-10-21T15:44:00.574+11:00Thanks Con! I have 4.8.3 MuQSS v114 built and run...Thanks Con! I have 4.8.3 MuQSS v114 built and running for both x86 SMP and i686 noSMP with BFQ v8r4; no problems so far. Finally used localmodconfig so my build times are down to a half hour (from 2.5; package size is also ~1/5).jwh7https://www.blogger.com/profile/09659185315567537391noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-43682065834541087712016-10-21T12:48:50.799+11:002016-10-21T12:48:50.799+11:00Multiple packages rather than multicore machines.Multiple packages rather than multicore machines.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-70116257337000528802016-10-21T12:48:16.583+11:002016-10-21T12:48:16.583+11:00By multi-cpu machines do you mean multi-core with ...By multi-cpu machines do you mean multi-core with 8 cores and 16 hyperthreads, or machines with more than one die installed? zarhttps://www.blogger.com/profile/11802628129636689039noreply@blogger.com