Posts

Occasionally, compiling the Linux kernel with a new compiler flag will result in a boot failure. If you are lucky, there will be some output to the serial console but that may not happen if the issue happens in early boot code before the serial driver has loaded. When this happens, it usually requires building part of the kernel without the compiler flag (or the “negative” version of it) to try and figure out the exact translation unit and function that causes the problem. I’ll go over this process at a high level to help others who might encounter this same issue. ...

As a Linux kernel developer, I will often need to build and boot new kernels to hunt down issues or test new functionality for regressions. While it is possible to manually install these kernels on machines, it is easiest to use the distribution’s package manager, as the kernel does not need to be built on the machine it is being installed on. With .deb and .rpm-based systems, it is easy to build a kernel package within the kernel source itself, using the bindeb-pkg and binrpm-pkg targets respectively. However, for Arch Linux, my distribution of choice, that is not so simple. Furthermore, when doing certain types of development, such as bisecting an issue, it is more convenient to do all the building in an actual source tree, rather than one that is managed by the Arch Build System. The following process might not be the most efficient or optimal way to do this process (the Arch wiki has a whole article about doing a git bisect with a PKGBUILD) but it works for me :) ...

Occasionally, I will forget to link something from the mailing list in this post. To see my full mailing list activity (patches, reviews, and reports), you can view it on lore.kernel.org. Linux kernel patches Boot fixes: These patches fix boot failures that we have seen in testing. In this case, it is a User Mode Linux fix, which we are looking to enable in our continuous integration setup, as it is useful for quick testing. ...

Occasionally, I will forget to link something from the mailing list in this post. To see my full mailing list activity (patches, reviews, and reports), you can view it on lore.kernel.org. Linux kernel patches Miscellaneous improvements: These are series or patches that do not really fit into any of the other categories I typically use with my reports. Allow CONFIG_DEBUG_INFO_DWARF5=y + CONFIG_DEBUG_INFO_BTF=y (v1) tools/resolve_btfids: Do not print any commands when building silently (v1) -Wenum-conversion: This warning occurs when one enumerated type is implicitly converted to another enumerated type, which is typically a bug, although it might not actually be a problem in practice if the used enum has the same value as the intended enum (which is the case here). It is still good to get it corrected, as the enums might not stay in sync. ...

Occasionally, I will forget to link something from the mailing list in this post. To see my full mailing list activity (patches, reviews, and reports), you can view it on lore.kernel.org. Linux kernel patches Stable patches: The stable trees are the trees that most users consume so keeping them building and as warning and issue free as reasonable is important. [PATCH 5.4] Input: touchscreen - Fix backport of a02dcde595f7cbd240ccd64de96034ad91cffc40 (v1) [PATCH RFC 4.9 0/5] Fix booting arm64 big endian with QEMU 5.0.0+ (v1) [PATCH 4.4,4.9] power: reset: ltc2952: Fix use of floating point literals (v1) Patches for clang and CONFIG_WERROR (arm64/x86_64) -Wpointer-bool-conversion: This warning is usually harmless, as it is typically just a developer checking if an array in the middle of a structure is NULL, which is not possible if the structure is not NULL, which would likely cause issues further up a call chain. In this one case, the fix was just to remove the check. ...

Occasionally, I will forget to link something from the mailing list in this post. To see my full mailing list activity (patches, reviews, and reports), you can view it on lore.kernel.org. Linux kernel patches Android patches: Android is one of the biggest downstream consumers of our work. Making sure that it stays working is incredibly important so that it can reach real users. Revert "ANDROID: disable INFINIBAND_QIB from allmodconfig build" ANDROID: allmodconfig: Re-enable TEST_KMOD ...

Occasionally, I will forget to link something from the mailing list in this post. To see my full mailing list activity (patches, reviews, and reports), you can view it on lore.kernel.org. Linux kernel patches Build errors: These are hard errors that appear for one reason or another. The first two are from a recent change in LLVM to avoid crashing the compiler (see the commit message for more details, I learned floating point literals exist in C). The third series is a resend that fixes ARCH=hexagon allmodconfig, which is an important target for us because ARCH=hexagon requires LLVM. Lastly, the final series fixes a generic issue uncovered by the kernel test robot with an ARCH=hexagon randconfig. ...

My co-maintainer Nick Desaulniers wrote a great post about taking a several thousand line C file that exposed a compiler bug down to 12 lines with creduce. I thought I would do the same thing with a bug that only happens with link time optimization (LTO) in the Linux kernel, which is a bit of a different beast. Hopefully this post can help others reduce their own bugs and think about the best way to triage a bug. ...

This was a bit of a shorter month for me, as I took some vacation at the beginning of the month to recouperate and meet half of my girlfriend’s extended family and friends for the first time. Thankfully, the rest of the ClangBuiltLinux team was able to keep everything churning along in my absence and I was able to be super productive the rest of the month once I returned. ...

Linux kernel patches tools: compiler-gcc.h: Guard error attribute use with __has_attribute (v1): The minimum version of GCC was raised to 5.1, which helped simplify a lot of code but caused issues for older versions of clang. This helps resolve it. Harden clang against unknown flag options (v2): A follow up. ptp: ocp: Avoid operator precedence warning in ptp_ocp_summary_show() (v1, v2): An interesting warning around operator precedence, which will not be an issue in practice but now that Linus has decreed that warnings are unacceptable, it is important to clean them up! ...