LWN: Comments on "Kernel topics on the radar" https://lwn.net/Articles/864603/ This is a special feed containing comments posted to the individual LWN article titled "Kernel topics on the radar". en-us Tue, 21 Oct 2025 12:30:29 +0000 Tue, 21 Oct 2025 12:30:29 +0000 https://www.rssboard.org/rss-specification lwn@lwn.net Kernel topics on the radar https://lwn.net/Articles/865698/ https://lwn.net/Articles/865698/ ClumsyApe <div class="FormattedComment"> Sticking my head out to give another shiny +1 :-) A kernel newbie like me sees such content as distilled gold !<br> </div> Mon, 09 Aug 2021 15:48:43 +0000 Bring back the old kernel page ... https://lwn.net/Articles/865380/ https://lwn.net/Articles/865380/ Wol <div class="FormattedComment"> It might - hopefully would - encourage more people to write more articles, though :-)<br> <p> Cheers,<br> Wol<br> </div> Thu, 05 Aug 2021 19:46:28 +0000 Bring back the old kernel page ... https://lwn.net/Articles/865251/ https://lwn.net/Articles/865251/ corbet We can consider that, but ... everything that was on the kernel page before is still to be found in the weekly edition; from the kernel point of view no content has been lost at all. So bringing back that page wouldn't necessarily help much. Wed, 04 Aug 2021 21:12:34 +0000 Bring back the old kernel page ... https://lwn.net/Articles/865231/ https://lwn.net/Articles/865231/ Wol <div class="FormattedComment"> Yes I know you&#x27;ll need a page editor and people to write the articles, but it would be nice to see it back.<br> <p> Cheers,<br> Wol<br> </div> Wed, 04 Aug 2021 19:12:27 +0000 +1 to making this "a semi-regular LWN feature" https://lwn.net/Articles/865222/ https://lwn.net/Articles/865222/ Nikratio <div class="FormattedComment"> Yes, please do this regularly.<br> </div> Wed, 04 Aug 2021 18:32:51 +0000 Kernel topics on the radar https://lwn.net/Articles/865138/ https://lwn.net/Articles/865138/ khim <p>The whole server infrastructure of Google is built around fibers nowadays thus one can be reasonably sure they are not going anywhere.</p> <p>This being said the chances are high that what was actually presented is what Kubernetes was in relation to <a href="https://research.google/pubs/pub43438/">Borg</a>: reimplementation of the same, tried and tested, idea but with an entirely different code and API.</p> <p>Thus obviously API discussion is very much needed.</p> Wed, 04 Aug 2021 08:00:58 +0000 Kernel topics on the radar https://lwn.net/Articles/865100/ https://lwn.net/Articles/865100/ eacb <div class="FormattedComment"> I definitely join this whishers list, and note that would beautifuly line up with lwn&#x27;s moto.<br> </div> Tue, 03 Aug 2021 21:39:38 +0000 Kernel topics on the radar https://lwn.net/Articles/865070/ https://lwn.net/Articles/865070/ null_byte <div class="FormattedComment"> Very glad to see this. It&#x27;s so hard to keep up with all of the emails, so I especially appreciate the overview.<br> </div> Tue, 03 Aug 2021 17:34:24 +0000 Kernel topics on the radar https://lwn.net/Articles/865058/ https://lwn.net/Articles/865058/ jakogut <div class="FormattedComment"> I resubscribed just for this post. Please make it a regular thing.<br> </div> Tue, 03 Aug 2021 16:52:29 +0000 Kernel topics on the radar https://lwn.net/Articles/864991/ https://lwn.net/Articles/864991/ hak8or <div class="FormattedComment"> And yet another person chiming in here to say they are a huge fan of a &quot;view of the lands&quot; posts like this. It helps us who work with the kernel but not lurk the mailing lists keep a finger on the pulse of upstream.<br> <p> While the Google Fibers patch seems interesting, I have serious qualms on anything from Google being subject to the kernels requirement of never breaking userspace, and therefore requiring that API be supported forever. Google is known to throw stuff at the wall and quickly abandon it if it falls out of grace, which is an extremely incompatible perspective on what the Linux kernel does. And to submit it without any example code is just, well, to me seems like a bad faith submission and makes me feel it should be looked with even more scrutiy than other such feature additions (in similar scope).<br> </div> Tue, 03 Aug 2021 13:32:00 +0000 Kernel topics on the radar https://lwn.net/Articles/864983/ https://lwn.net/Articles/864983/ cwhitecrowdstrike <div class="FormattedComment"> Agreed! +1 for this as a regular feature.<br> </div> Tue, 03 Aug 2021 11:51:15 +0000 +1 to making this "a semi-regular LWN feature" https://lwn.net/Articles/864973/ https://lwn.net/Articles/864973/ or <div class="FormattedComment"> Yes, it would be very nice to get an overwiew like this more often or even regularly! <br> </div> Tue, 03 Aug 2021 09:46:18 +0000 Kernel topics on the radar https://lwn.net/Articles/864964/ https://lwn.net/Articles/864964/ unixbhaskar <div class="FormattedComment"> Thanks,Jon. This will certainly helpful for the quick view of what is churnning in the oven(aka kernel). I mean, those who do not follow the lkml rigorously can get hugely benefitted by this. <br> </div> Tue, 03 Aug 2021 07:13:26 +0000 Kernel topics on the radar https://lwn.net/Articles/864962/ https://lwn.net/Articles/864962/ itsmycpu <div class="FormattedComment"> So for task isolation with version 2 of the patch set, if I want to remove as much OS noise as possible, I would do the following on a nohz CPU:<br> <p> assert( (prctl(PR_ISOL_FEAT, 0, 0, 0, 0) &amp; ISOL_F_QUIESCE) != 0);<br> int max_silence = prctl(PR_ISOL_FEAT, ISOL_F_QUIESCE, 0, 0, 0);<br> assert( prctl(PR_ISOL_SET, ISOL_F_QUIESCE, max_silence, 0, 0) &gt;= 0);<br> assert( prctl(PR_ISOL_CTRL_SET, ISOL_F_QUIESCE, 0, 0, 0) &gt;= 0);<br> <p> Correct? (Assuming I&#x27;m not afraid of regressions..what regressions?)<br> </div> Tue, 03 Aug 2021 05:40:17 +0000 Kernel topics on the radar https://lwn.net/Articles/864951/ https://lwn.net/Articles/864951/ itsmycpu <div class="FormattedComment"> Another +1 for making this a (semi-)regular feature.<br> <p> A +1 also for additional general and direct support for (more) complete task isolation.<br> </div> Mon, 02 Aug 2021 23:12:36 +0000 +1 to making this "a semi-regular LWN feature" https://lwn.net/Articles/864934/ https://lwn.net/Articles/864934/ linuxjacques <div class="FormattedComment"> I too like this feature.<br> <p> </div> Mon, 02 Aug 2021 19:41:41 +0000 +1 to making this "a semi-regular LWN feature" https://lwn.net/Articles/864933/ https://lwn.net/Articles/864933/ mliuzzi <div class="FormattedComment"> yes please!<br> </div> Mon, 02 Aug 2021 19:36:24 +0000 +1 to making this "a semi-regular LWN feature" https://lwn.net/Articles/864930/ https://lwn.net/Articles/864930/ knurd <div class="FormattedComment"> Yeah, please make this a &quot;a semi-regular LWN feature&quot;, otherwise that field is left to websites that don&#x27;t get even close to the stellar quality reporting here at LWN.net – which is not in the interest of the Linux world, as those websites sometimes inadvertently get things wrong and thus spread misconceptions with their reporting.<br> </div> Mon, 02 Aug 2021 18:19:27 +0000