Re: [Linaro-TCWG-CI] gcc-15-3256-g1c4b9826bd0: Failure on aarch64
On Thu 2024-09-05 13:59:00, Maxim Kuvyrkov wrote: > Hi Filip, > > FYI, your patch below causes linux kernel build failure on arm64. Could you > take a look, please? > Hi Maxim, I'm taking a look now. It seems that the __builtin_popcount I insert in the switch conversion pass doesn't expand to either bitmagic or an internal function call and ends up as a call to libgcc in the binary. Currently I'm trying to reproduce this and figure out why does it happen. Thanks, Filip ___ linaro-toolchain mailing list -- linaro-toolchain@lists.linaro.org To unsubscribe send an email to linaro-toolchain-le...@lists.linaro.org
[automated CI mails] Need help with a regression Linaro CI notified me of
Hi all. I got an automated mail about this GCC regression related to one of my commits. https://linaro.atlassian.net/browse/GNU-1578 The mail itself mentioned only a testsuite regression. I fixed that. However, this was also linked in the mail: https://ci.linaro.org/job/tcwg_bootstrap_build--master-arm-bootstrap_profiled-build/817/artifact/artifacts/notify/mail-body.txt/*view*/ That looks like a different regression. As far as I can tell, `make profiledbootstrap` is failing on arm. I have only very limited access to arm machines. I tried to reproduce this on an Aarch64 machine and didn't succeed. Could you check that this regression is still present on current trunk and/or help me reproduce it? Btw the "Instructions to reproduce the build" link doesn't work for me. I get "Request is not authorized". https://git-us.linaro.org/toolchain/ci/interesting-commits.git/plain/gcc/sha1/1381a5114788a2e9234ff54e0cd7a3c810f0d02d/tcwg_gcc_check/master-aarch64/reproduction_instructions.txt Thanks, Filip Kastl ___ linaro-toolchain mailing list -- linaro-toolchain@lists.linaro.org To unsubscribe send an email to linaro-toolchain-le...@lists.linaro.org
