LWN: Comments on "The 5.12 kernel has been released" https://lwn.net/Articles/853289/ This is a special feed containing comments posted to the individual LWN article titled "The 5.12 kernel has been released". en-us Fri, 19 Sep 2025 22:46:02 +0000 Fri, 19 Sep 2025 22:46:02 +0000 https://www.rssboard.org/rss-specification lwn@lwn.net The 5.12 kernel has been released https://lwn.net/Articles/857273/ https://lwn.net/Articles/857273/ flussence <div class="FormattedComment"> Some closure: I&#x27;ve now successfully convinced an LTO kernel to build and boot by turning the modules mentioned in the compilation errors into built-ins. The bzImage + /lib/modules/ is a few KB smaller in total than without LTO, but it seems like noise.<br> <p> Overall, I feel like there&#x27;s much bigger kernel size wins to be made simply by breaking up monolithic GPU drivers into more granular model ranges...<br> </div> Wed, 26 May 2021 12:02:08 +0000 The 5.12 kernel has been released https://lwn.net/Articles/854710/ https://lwn.net/Articles/854710/ flussence <div class="FormattedComment"> I&#x27;ve been playing around with the LTO stuff, unsuccessfully. Here&#x27;s what I&#x27;ve found:<br> <p> It&#x27;s next to impossible to discover unless you&#x27;re already aware of the exact git commit message with the magic incantation. I couldn&#x27;t figure out what was locking out the settings in nconfig, even with the symbol search&#x27;s usually useful metadata (it&#x27;s empty for HAVE_LTO_CLANG), and had to resort to grepping makefiles. To save others the time, the answer&#x27;s `make LLVM=1 LLVM_IAS=1` — both vars are needed on every config and compile command.<br> <p> Having TRIM_UNUSED_KSYMS enabled caused the build to explode with undefined module symbols in modpost, though it looks like the kernel image itself built successfully. The symbols are all internal-sounding ones: __x86_indirect_thunk_r11, __SCT__preempt_schedule{,_notrace}, and __{get,put}_user_{1,4,8}.<br> <p> The help text for that option explicitly suggests this is a thing that&#x27;s supposed to work with LTO. It doesn&#x27;t seem to be caused solely by that option, as I can&#x27;t reproduce the error from a fresh allnoconfig just by flipping it on. I have no idea how to efficiently narrow down the source and I gave up after hours of trying the hard way. Having it disabled succeeds, but the resulting kernel is massive.<br> <p> A plain kernel built with clang and no LTO has no problems though, I&#x27;ve got two machines running on it now.<br> </div> Wed, 28 Apr 2021 12:57:26 +0000 The 5.12 kernel has been released https://lwn.net/Articles/854440/ https://lwn.net/Articles/854440/ flussence <div class="FormattedComment"> Okay, preemption model switching at *runtime*? That&#x27;s a really cool surprise. (I know you&#x27;re not meant to do it at runtime, but still)<br> </div> Mon, 26 Apr 2021 11:15:34 +0000