tag:blogger.com,1999:blog-6469704299235308349.post5138539290714337026..comments2024-03-28T15:50:13.644+11:00Comments on -ck hacking: BFS 450, 3.16-ck1ckhttp://www.blogger.com/profile/02904761195451530213noreply@blogger.comBlogger51125tag:blogger.com,1999:blog-6469704299235308349.post-51231061064626044902014-08-24T12:01:52.439+10:002014-08-24T12:01:52.439+10:00I'm watching this thread. I will update my -gc...I'm watching this thread. I will update my -gc branch by re-basing 0450 and sync with 3.16.2 from mainline, hopefully next week. As ck said debug is not finished, I will not include these 3 patches so you can apply updated ones if you affected by similar issues.Alfred Chenhttps://www.blogger.com/profile/03164306846702841944noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-2850046041858409322014-08-23T19:04:25.758+10:002014-08-23T19:04:25.758+10:00ck > "Let's try it for a day or two an...ck > "Let's try it for a day or two and then I can formalise these changes as a new BFS if nothing shows up."<br /><br />I hope you will release: new fixed BFS releases, or as a cumulative patch, for the previous kernel versions, like 3.15 and also for the current lts 3.14<br /><br />thanks,<br />bye, NicCoAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-75606590122722379042014-08-23T17:19:22.672+10:002014-08-23T17:19:22.672+10:00Got that, thanks.Got that, thanks.Oleksandr Natalenkohttps://www.blogger.com/profile/12098091624630953604noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-48627176913506708532014-08-23T11:14:56.962+10:002014-08-23T11:14:56.962+10:00The NEW patch set also works well for a non affect...The NEW patch set also works well for a non affected system with a 3.16.y-gc patched kernel,<br /> applied on top on here:<br />bfs450-resched-scap.patch<br />bfs450-sched-ipi.patch<br />bfs450-add-preempt-resched.patch (No.1 with fuzz o.k, No.8 failed, as already removed o.k)<br /><br />I hope Alfred Chen does consider this safe... ^^<br /><br />Thank you all, and best regards, <br />Manuel Krause<br /><br />BTW, I knew the "behavioural issues" are meant regarding my system, that's why I found it so funny as it can have double meaning for real life..Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-85808291041015840612014-08-23T08:24:52.960+10:002014-08-23T08:24:52.960+10:00@PF: From linux 3.13 setting just the "tif ne...@PF: From linux 3.13 setting just the "tif needs resched" flag alone was not enough to trigger a descheduling from certain places in the code, it needed the "preempt needs resched" tagged to trigger a different type of descheduling to hand over to another process or kick it off a cpu where it should no longer be.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-79595182274274321062014-08-23T00:37:21.488+10:002014-08-23T00:37:21.488+10:00I can confirm that the last set of patches fix the...I can confirm that the last set of patches fix the ath9k issue.<br />I'll report back if I encounter any issues.<br /><br />kudos to pf an ck for solving this issue, and big thanks!ooonoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-83256808749145793492014-08-22T23:15:16.563+10:002014-08-22T23:15:16.563+10:00@ck: also with the last set of patches we do not n...@ck: also with the last set of patches we do not need KVM workaround as well. I've reverted it too.Oleksandr Natalenkohttps://www.blogger.com/profile/12098091624630953604noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-11357357738043832902014-08-22T23:02:03.571+10:002014-08-22T23:02:03.571+10:00@ck: if that is correct fix, we kindly ask you to ...@ck: if that is correct fix, we kindly ask you to explain what's happened :).Oleksandr Natalenkohttps://www.blogger.com/profile/12098091624630953604noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-35178759035501071762014-08-22T22:30:06.360+10:002014-08-22T22:30:06.360+10:00Aha! Now we're talking! This last set of patch...Aha! Now we're talking! This last set of patches is the correct fix (unlike the tifcheck patch). Let's try it for a day or two and then I can formalise these changes as a new BFS if nothing shows up. Thanks for testing!ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-74716179226850937122014-08-22T22:26:50.437+10:002014-08-22T22:26:50.437+10:00@ck: also the last combination of patches works OK...@ck: also the last combination of patches works OK (both ath9k and i8k) with preemption enabled.Oleksandr Natalenkohttps://www.blogger.com/profile/12098091624630953604noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-3069865699453560112014-08-22T21:53:09.525+10:002014-08-22T21:53:09.525+10:00@ck: applying sched-ipi patch only and disabling p...@ck: applying sched-ipi patch only and disabling preemption completely does the trick — ath9k seems to work OK, but i8k doesn't work.Oleksandr Natalenkohttps://www.blogger.com/profile/12098091624630953604noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-75083073278529148952014-08-22T21:50:37.821+10:002014-08-22T21:50:37.821+10:00This comment has been removed by the author.Oleksandr Natalenkohttps://www.blogger.com/profile/12098091624630953604noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-27265678213337655082014-08-22T20:33:28.738+10:002014-08-22T20:33:28.738+10:00@PF: And after that try this combination:
bfs450-r...@PF: And after that try this combination:<br />bfs450-resched-scap.patch <br />bfs450-sched-ipi.patch <br />+<br /><a href="http://ck.kolivas.org/patches/bfs/3.0/3.16/test/bfs450-add-preempt-resched.patch" rel="nofollow">bfs450-add-preempt-resched.patch</a>ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-80034383314793332332014-08-22T19:14:35.217+10:002014-08-22T19:14:35.217+10:00With "behavioural issues" he means the w...With "behavioural issues" he means the way your system behaves.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-22863320984273043822014-08-22T16:14:09.243+10:002014-08-22T16:14:09.243+10:00@PF: Here's a crazy thing. Try with only the s...@PF: Here's a crazy thing. Try with only the sched-ipi patch and disable all preempt completely in your config and see what that does please.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-62457868436149616762014-08-22T15:45:02.945+10:002014-08-22T15:45:02.945+10:00Yes they apply equally there too except for bfs450...Yes they apply equally there too except for bfs450-resched-scap.patchckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-78756600646776490082014-08-22T15:39:15.172+10:002014-08-22T15:39:15.172+10:00The same 3 bugfix patches can be applied to Kernel...The same 3 bugfix patches can be applied to Kernel 3.15.x?<br /><br />thanks<br />NicCoAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-69911241997704921662014-08-22T12:21:06.579+10:002014-08-22T12:21:06.579+10:00Hehe, Con, you're getting funny... My "be...Hehe, Con, you're getting funny... My "behavioural issues"...? ;-) Besides tracking and applying your patches...? ^^ Keep up this sense of humor!<br />No, really funny, and self-ironical for me!<br /><br />I still want to read Alfred Chen on this, too, as I currently run his "old" 3.16.y-gc patches, so far, and he hasn't published an updated one for your, Con's, 3.16 release yet.<br /><br />Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-88732962556578713662014-08-22T09:11:03.579+10:002014-08-22T09:11:03.579+10:00All 3 are bugfixes for everyone, but the patches h...All 3 are bugfixes for everyone, but the patches have not been finalised. If you have no behavioural issues you are unlikely to see anything by applying them.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-559283970639486752014-08-22T09:03:45.817+10:002014-08-22T09:03:45.817+10:00So, if I read correctly, the patches:
(1) http://c...So, if I read correctly, the patches:<br />(1) http://ck.kolivas.org/patches/bfs/3.0/3.16/test/bfs450-sched-ipi.patch<br />(2) http://ck.kolivas.org/patches/bfs/3.0/3.16/test/bfs450-tifcheck_in_cond_resched.patch<br /> are of benefit?<br />And this one is not needed?:<br />(3) http://ck.kolivas.org/patches/bfs/3.0/3.16/test/bfs450-resched-scap.patch<br /><br />Do these patches "only" (but thankfully!) heal the issues post-factum and others reported, or are they considered as bug-fixes for BFS?<br /><br />@ Alfred Chen: Would you recommend the patches 1-2 to be applied onto your 3.16.y-gc patched kernel, too?<br /><br />Best regards, Manuel KrauseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-24238876815724080942014-08-21T20:26:08.721+10:002014-08-21T20:26:08.721+10:00OK, if there's necessity to test more patches,...OK, if there's necessity to test more patches, feel free to send them to me.Oleksandr Natalenkohttps://www.blogger.com/profile/12098091624630953604noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-65415758881223314162014-08-21T19:40:48.242+10:002014-08-21T19:40:48.242+10:00Thanks PF. I'd spent the last couple of days a...Thanks PF. I'd spent the last couple of days auditing code to see what might be responsible and that was the only solution I could come up with. The behaviour with this patch is definitely correct, but it's a bit disappointing because it means there's something fundamentally different in BFS handling the resched flag compared to mainline and I didn't intend to start diverting from mainline in this way. I'll keep auditing the code to see if there's an obvious trigger to act on this flag in a different place that I've missed but it's fair to say this is a sane solution for the time being and if I can't come up with anything, I'll just run with it.ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-57848124314806253782014-08-21T18:44:31.881+10:002014-08-21T18:44:31.881+10:00@ck it works :).
So, I guess, the working solutio...@ck it works :).<br /><br />So, I guess, the working solution on top of bare -ck1 is:<br /><br />1. https://github.com/pfactum/pf-kernel/commit/44b3e870e656a11aa7116c236b7e00591141a68a — brings back scheduler_ipi()<br /><br />2. https://github.com/pfactum/pf-kernel/commit/6a180442f154c5a624ee377dacfcc0b8631eb1e0 — uses tif_need_resched() in cond_resched()<br /><br />Also I've reverted KVM workarounds here:<br /><br />3. https://github.com/pfactum/pf-kernel/commit/ad4d566baf9a825f41240ce1785096028fdacd45<br /><br />KVM+QEMU works OK.<br /><br />Also, we don't need special i8k workaround. I've reverted it, and i8k seems to work as well.<br /><br />I'm going to test it more, but now everything seems to be OK. Thanks!Oleksandr Natalenkohttps://www.blogger.com/profile/12098091624630953604noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-39506135788932660842014-08-21T17:12:59.531+10:002014-08-21T17:12:59.531+10:00PF thanks for your tireless testing.
Can you give...PF thanks for your tireless testing.<br /><br />Can you give this crazy patch a try on top of sched ipi please?<br /><a href="http://ck.kolivas.org/patches/bfs/3.0/3.16/test/bfs450-tifcheck_in_cond_resched.patch" rel="nofollow">bfs450-tifcheck_in_cond_resched.patch</a>ckhttps://www.blogger.com/profile/02904761195451530213noreply@blogger.comtag:blogger.com,1999:blog-6469704299235308349.post-60374345958034734362014-08-21T13:20:55.567+10:002014-08-21T13:20:55.567+10:00@pf, thanks for testing anyway.@pf, thanks for testing anyway.Alfred Chenhttps://www.blogger.com/profile/03164306846702841944noreply@blogger.com