tag:blogger.com,1999:blog-6469704299235308349.post2950164654518499899..comments2024-02-09T16:24:46.087+11:00Comments on -ck hacking: 3.1.0-ck1, BFS 0.414ckhttp://www.blogger.com/profile/02904761195451530213noreply@blogger.comBlogger20125tag:blogger.com,1999:blog-6469704299235308349.post-56529586357971458812011-11-09T03:17:59.967+11:002011-11-09T03:17:59.967+11:00[Ralph Ulrich]
@graysky, I take your benchmarkings...[Ralph Ulrich]<br />@graysky, I take your benchmarkings as a reassurance there ire no regressions when applied to a new linux release! Also I know BFS is meant to reduce latency and you measure kind of throughput... Keep up benchmarking!<br /><br />Recently I saw the verb "latency" near profiling during "make menuconfig". Is there this cabability to measure in the kernel? I need to go into that further ....Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-28079942666357873722011-11-08T19:41:43.381+11:002011-11-08T19:41:43.381+11:00@GG- dude, ck's blog is not the right place fo...@GG- dude, ck's blog is not the right place for this question. Start a thread on the forums. Also, bfs is not an I/O scheduler. You are confusing it with BFQ which isn't yet released for 3.1.<br /><br />My advice, build the linux-ck package from AUR and sort out your hibernation issue. If it is related to ck's patches, then post here.grayskyhttp://repo-ck.comnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-30874534104100126522011-11-08T18:18:18.893+11:002011-11-08T18:18:18.893+11:00I have some issues on atom-ck kernel from archlinu...I have some issues on atom-ck kernel from archlinux repo with uswsusp. Hibernation hangs just before the compression starts sometimes.<br />Also if I choose bfs as elevator, sometimes I have to press some buttons (like shift,ctrl etc) to make system response to me while opening an application. :\Gürkan Gürhttps://www.blogger.com/profile/10030775512879400506noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-67010252267513984902011-11-08T07:07:42.423+11:002011-11-08T07:07:42.423+11:00@anon - I'm no statistician. I'd be glad ...@anon - I'm no statistician. I'd be glad to provide the dataset I collected to you if you're game for an analysis.grayskyhttp://www.repo-ck.comnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-43621021842203893362011-11-08T02:44:44.341+11:002011-11-08T02:44:44.341+11:00@graysky
Statistical significance adds to credibi...@graysky<br /><br />Statistical significance adds to credibility. It is not, however, a measure of improvement! Always start from reporting deltas or relative speedup. <br /><br />But since you've started talking about statistics, you should notice that the distributions are clearly skewed and simple ANOVA is a bad idea. Do ranksum or U-test instead and, maybe, add AUC under ROC.<br /><br />The tails and the skewedness is actually very important here, because those tails result in random but noticeable stalls.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-20740767342686568472011-11-05T23:20:50.175+11:002011-11-05T23:20:50.175+11:00[Ralph Ulrich]
Also here runs better then vanilla...[Ralph Ulrich]<br /><br />Also here runs better then vanilla linux-3.1, also patched with 178 patches of Greg KH queued stable release:<br />Mac-Mini of 2009 core2, 64bit Gentoo unstable, using gcc-4.6.2Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-39706661083477775202011-11-05T22:57:30.310+11:002011-11-05T22:57:30.310+11:00The usual make benchmark of 3.1 vs. 3.1 + ck1 (bfs...The usual make benchmark of 3.1 vs. 3.1 + ck1 (bfs v0.414).<br /><br />http://s7.postimage.org/689295789/3_1.png<br /><br />Statistically significant increase in make performance. The above was on a dual quad machine with 16 threads running "make -j16 bzImage" of linux v3.1grayskyhttp://repo-ck.comnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-30800550907701729172011-11-04T22:34:38.915+11:002011-11-04T22:34:38.915+11:00Im looking forward to do this.
Is it OK when I mak...Im looking forward to do this.<br />Is it OK when I make a picture of the screen and send it to you via email.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-67469741318104630302011-11-04T21:37:35.709+11:002011-11-04T21:37:35.709+11:00Thanks!
If you could reproduce it with the full d...Thanks!<br /><br />If you could reproduce it with the full debugging options that would be far more useful.<br /><br />General setup ---> <br />[ ] Configure standard kernel features (expert users) ---><br />-*- Load all symbols for debugging/ksymoops <br />-*- Include all symbols in kallsyms <br /><br />Kernel hacking ---> <br />[*] Kernel debugging<br />[*] Compile the kernel with debug info<br />[ ] Reduce debugging information <br />-*- Compile the kernel with frame pointersckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-89014181014555394442011-11-04T21:31:16.271+11:002011-11-04T21:31:16.271+11:00I could reproduce the problem.
general protection...I could reproduce the problem.<br /><br />general protection fault: 0000 [#1] SMP<br />CPU4<br /><br />Process fsck.ext4 (pid 688,...<br /><br />Call Trace:<br />resched_closest_idle.isra.64+0x5b/0x70<br />schedule+0x9f8/0xca0<br />? vfs_read+0x10d/0x180<br />? sys_read+0x4a/0x90<br />retint_careful+0x14/0x32<br />RIP resched_best_mask+0x30/0x150<br /><br />Tested with Ubuntu 64 bit 10.10 on a Intel core i7 2600KAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-7752861679237261092011-11-04T08:15:39.848+11:002011-11-04T08:15:39.848+11:00Thanks ck!
$ dmesg | grep BFS
BFS CPU scheduler v...Thanks ck!<br /><br />$ dmesg | grep BFS<br />BFS CPU scheduler v0.414 by Con Kolivas.grayskyhttp://repo-ck.comnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-25221953346954240072011-11-04T07:19:42.617+11:002011-11-04T07:19:42.617+11:00Thanks for the links.
@Anonymous: The prerequisit...Thanks for the links.<br /><br />@Anonymous: The prerequisites for a task being flagged as rq->preempt_task are exactly the same as those for being selected in earliest_deadline_task, in the preemption function. Picking it before all other tasks in edt() is done precisely to avoid going through the same tests for no reason.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-28241911054406402282011-11-04T03:22:18.910+11:002011-11-04T03:22:18.910+11:00When I started the kernel with the BFS patch for t...When I started the kernel with the BFS patch for the first time, it directly went into panic. I wasn't able to catch exactly the output. But I can say it is related to the scheduler, because in the call-trace, there was resched_closet_idle and other scheduling related functions.<br />I tryed to reproduce the problem, but now everything is working fine.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-52942143600078409202011-11-04T02:30:54.562+11:002011-11-04T02:30:54.562+11:00http://bo.mirror.garr.it/linux/kernel/people/ck/http://bo.mirror.garr.it/linux/kernel/people/ck/Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-64679385340670045692011-11-04T02:27:54.264+11:002011-11-04T02:27:54.264+11:00@ck ufpr.br have rsync mirror too
rsync://kernel....@ck ufpr.br have rsync mirror too<br /><br />rsync://kernel.c3sl.ufpr.br/pub/linux/kernel/people/ckKazuonoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-15514930518326188222011-11-04T02:06:06.595+11:002011-11-04T02:06:06.595+11:00In the function earliest_deadline_task you take th...In the function earliest_deadline_task you take the rq->preempt task before any other task in the global runqueue, even before realtime tasks, like the ones with SCHED_IDLEPRIO. Wouldn’t it make better latencies if it would be looked up if there is a realtime task and then look at the rq->preempt.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-42757078801201578782011-11-04T01:54:46.733+11:002011-11-04T01:54:46.733+11:00@ck: There is a mirror of your patches in http://k...@ck: There is a mirror of your patches in http://kernel.c3sl.ufpr.br/pub/linux/kernel/people/ck/Anonymoushttps://www.blogger.com/profile/06717250053155630558noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-23285098928395373172011-11-04T01:22:28.343+11:002011-11-04T01:22:28.343+11:00Tested and build fine.
install on test server and ...Tested and build fine.<br />install on test server and work fine.Micronhttps://www.blogger.com/profile/00612104081287508741noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-16464495398128159702011-11-03T20:40:53.193+11:002011-11-03T20:40:53.193+11:00Ubuntu packages are welcome :)Ubuntu packages are welcome :)Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-69609743813341372852011-11-03T17:45:50.564+11:002011-11-03T17:45:50.564+11:00tested is aproved :)tested is aproved :)Anonymousnoreply@blogger.com