tag:blogger.com,1999:blog-2963790426029213933.post6777279049392643166..comments2024-02-29T00:33:07.382-08:00Comments on Alfred Chen's Blog: PDS 0.98m releaseAlfred Chenhttp://www.blogger.com/profile/03164306846702841944noreply@blogger.comBlogger23125tag:blogger.com,1999:blog-2963790426029213933.post-67570218150484939012018-04-13T00:37:24.819-07:002018-04-13T00:37:24.819-07:00Hi artafinde,
If I knew about your package before...Hi artafinde,<br /><br />If I knew about your package before I wouldn't have bothered with mine. I see you have some changes from -ck, like disabling NUMA. <br /><br />I like the target CPU auto detection of my patch because I usually build a new kernel on 3+ machines, each with a different CPU -- is there a way you can incorporate this into your patch? Though, since it renames the package I'm not sure if it breaks AUR managers in some way or not... <br /><br /><br />Regards,<br />Mitja Horvat<br /><br /><br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-84885360143763595592018-04-12T13:51:27.046-07:002018-04-12T13:51:27.046-07:00You can use my published to AUR package
https://au...You can use my published to AUR package<br />https://aur.archlinux.org/packages/linux-gc/<br />artafindehttps://aur.archlinux.org/account/artafindenoreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-47493194163820204352018-04-12T06:05:06.647-07:002018-04-12T06:05:06.647-07:00Thank you, I will try to use your PKGBUILD (or imp...Thank you, I will try to use your PKGBUILD (or improve mine), it looks neatly done. I have been making my own, uggly version for Manjaro.<br /><br />DzonAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-22820278647115952592018-04-12T02:00:33.488-07:002018-04-12T02:00:33.488-07:00Hi All,
for my own convenience I've created a...Hi All,<br /><br />for my own convenience I've created an AUR package in the same spirit as linux-ck, but with the PDS scheduler instead.<br /><br />If anyone is interested, you can check it at http://github.com/pinkfluid/linux-pds<br /><br />By default it builds a version optimized for the current CPU. If you want to build a generic kernel (or for a different CPU architecture), use the FLAVOR=generic (or FLAVOR=skylake ...etc).<br /><br />Quick build instructions:<br /><br /># git clone http://github.com/pinkfluid/linux-pds<br /># cd linux-pds<br /># FLAVOR=generic makepkg -ci<br /><br />As I don't have much free time lately, please feel free to clone and improve it as I will have very little to no time to support it.<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-69172678972933082192018-04-08T09:10:22.460-07:002018-04-08T09:10:22.460-07:00Just feel free to offer experimental CacheHot stuf...Just feel free to offer experimental CacheHot stuff to test -- even if you don't consider it perfect.<br /><br />Your last formula together with your hints really wasn't bad at all for my sytem. With 4.15.xy I also tested a version with SCHED_CACHE_HOT_SWITCHES_TH == 12, where you had suggested 13 for me, and at least one "subsystem" didn't like it, namely TOI, whereas with 13 I only saw improvements (subjectively).<br /><br />And, yes, the current PDS for 4.16 is also still running fine. :-)<br /><br />BR, thank you for and keep up your good work and brilliant ideas coming,<br />Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-71293610570123078182018-04-07T00:15:01.917-07:002018-04-07T00:15:01.917-07:00There is no choice, I have working on this branch....There is no choice, I have working on this branch. :)Andy Lavrhttps://www.blogger.com/profile/06797718949697091110noreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-47584306178889605542018-04-06T21:48:19.795-07:002018-04-06T21:48:19.795-07:00Current status of the cachehot patch is, it is har...Current status of the cachehot patch is, it is hard to tune the parameters to get benifit as the overhead already been reduced in 098m. I will see how it goes in the incoming weeks.Alfred Chenhttps://www.blogger.com/profile/03164306846702841944noreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-46042518831043723232018-04-06T21:38:38.586-07:002018-04-06T21:38:38.586-07:00It's still too early to port to 4.17. :)It's still too early to port to 4.17. :)Alfred Chenhttps://www.blogger.com/profile/03164306846702841944noreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-78856890695549414322018-04-05T08:03:50.101-07:002018-04-05T08:03:50.101-07:00Or e.g. "PDSlim" ;-)
OMG, together with ...Or e.g. "PDSlim" ;-)<br />OMG, together with a new Cache HOT patch... what naming would be still conforming to conventions ^^<br />BR, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-17234295124161479712018-04-05T07:39:14.615-07:002018-04-05T07:39:14.615-07:00@Anonymous:
Also, Pedro shows up on here quite oft...@Anonymous:<br />Also, Pedro shows up on here quite often. Thanks for that!<br />His most recent benchmark is: https://docs.google.com/spreadsheets/d/163U3H-gnVeGopMrHiJLeEY1b7XlvND2yoceKbOvQRm4/edit#gid=857299933<br /><br />In it's info tab you can see how he composes and describes his benchmarks.<br /><br />BR, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-25036766846119616092018-04-05T07:15:37.464-07:002018-04-05T07:15:37.464-07:00@Anonymous:
Oh, I just remebered it: "interbe...@Anonymous:<br />Oh, I just remebered it: "interbench". You can search Con's Blog for it, e.g. http://ck-hack.blogspot.de/search?q=benchmark&max-results=20&by-date=true<br />or some search like that.<br /><br />But I have to add, that I have no experience with it and it's usefulness at all.<br /><br />BR, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-1244945091799144492018-04-05T07:05:15.917-07:002018-04-05T07:05:15.917-07:00@Anonymous:
Don't be too sceptic -- just trust...@Anonymous:<br />Don't be too sceptic -- just trust your feelings... ;-)<br /><br />I don't know what you want to compare so that we don't end up with apples and oranges.<br />For an overall throughput oriented testing the phoronix test suite may be worth a try.<br />Also, Alfred published his sanity test scripts he used some time ago. Don't know if it's up-to-date any more, it's from 2015 [https://bitbucket.org/alfredchen/linux-gc/downloads/sanity AND https://bitbucket.org/alfredchen/linux-gc/downloads/compile_throughput].<br /><br />And then, going to interactivity/ low-latency comparisons it'll become more and more difficult. IIRC, Con Kolivas had provided something related, years ago. ATM I'm unable to recall it.<br /><br />BR, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-74968557531421944662018-04-05T05:34:46.543-07:002018-04-05T05:34:46.543-07:00My system also looks snappier but I'm very sce...My system also looks snappier but I'm very sceptic. I would like to have a benchmark that give me numbers to compare with stock kernel.<br />What tools can be used to measure the improvements by PDS? I haven't found a reproducible way of measuring it... Any ideas?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-85524728120480479482018-04-05T03:35:47.235-07:002018-04-05T03:35:47.235-07:00Thanks Alfred, I fixed and tested on mainline (4.1...Thanks Alfred, I fixed and tested on mainline (4.17). It works well.<br />The patch i sent you to the email.Andy Lavrhttps://www.blogger.com/profile/06797718949697091110noreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-1500749451663273842018-04-04T22:08:08.780-07:002018-04-04T22:08:08.780-07:00@Alfred,
compiled & running both on i7 and on...@Alfred,<br /><br />compiled & running both on i7 and on Ryzen, no problems so far. System seems snappier than before, at least that's the feeling.<br /><br />Thanks for Your efforts ;)<br /><br />BR, EduardoAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-76800935164570396852018-04-04T18:06:05.811-07:002018-04-04T18:06:05.811-07:00Both built and running; no problems.Both built and running; no problems.jwh7https://www.blogger.com/profile/09659185315567537391noreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-59075128346756668222018-04-04T12:40:29.177-07:002018-04-04T12:40:29.177-07:00> Another obvious change is in the patch size, ...> Another obvious change is in the patch size, -5KB!(No one cares, right?)<br />Maybe you should have named this release "PDS 0.98lite" Or "Diet PDS" :-P<br />Thanks Alfred; will build x64 and x86-UP tonight.jwh7https://www.blogger.com/profile/09659185315567537391noreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-12851716728793324752018-04-04T12:32:57.022-07:002018-04-04T12:32:57.022-07:00Compiles/boots OK for me, thank you.Compiles/boots OK for me, thank you.Oleksandr Natalenkohttps://www.blogger.com/profile/12098091624630953604noreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-68948612264388608232018-04-04T07:21:00.334-07:002018-04-04T07:21:00.334-07:00Regarding your host migration plans:
I've als...Regarding your host migration plans:<br /><br />I've also seen the warning messages from your bitbucket about the 2G limitation, but the've gone away some weeks/ months ago. Anyway, unless there is a specific reason to maintain two separate repos, it'll primarily save you time to only have the github. The missing patch folder now is also in place there, so there's no visible reason to me to not go that way.<br /><br />BR, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-65840544345523123672018-04-04T07:06:35.391-07:002018-04-04T07:06:35.391-07:00@Alfred:
Also on here, the 0.98m is running well (...@Alfred:<br />Also on here, the 0.98m is running well (after I managed to remove the obstacle).<br />Fortunately the TOI patch for 4.15 also works for 4.16 without editing (unfortunately all it's limitations remain).<br /><br />Good work from yours, thank you! Now I'm curious for your next hot CachHot patch. :-))<br /><br />BR, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-61829554852016324442018-04-04T04:04:56.768-07:002018-04-04T04:04:56.768-07:00None, cachehot will create another new branch in t...None, cachehot will create another new branch in the code path, but these commits mainly improve existed code path by cutting overhead.Alfred Chenhttps://www.blogger.com/profile/03164306846702841944noreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-52911800179827275142018-04-04T04:03:52.135-07:002018-04-04T04:03:52.135-07:00Up and running... All good...
Good work.Up and running... All good...<br />Good work.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2963790426029213933.post-68749338498235490362018-04-04T03:48:00.134-07:002018-04-04T03:48:00.134-07:00Building...
Is the cachehot patch work included in...Building...<br />Is the cachehot patch work included in this patch?<br />Anonymousnoreply@blogger.com