Hi Jeff, On 4/23/26 03:22, Jeffrey Law wrote: > > On 4/22/2026 4:39 AM, Kito Cheng wrote: >> I guess you need to remove the RISC-V from the title since this is not >> RISC-V specific, and RISC-V maintainer actually can't approve this >> (Jeff is global maintainer, so he is not count on this ) > So why isn't this a problem on the other big endian targets GCC supports > (there's several)? > I'm not sure about other big endian targets, but I did manage to confirm that this is an issue on RISC-V. However, I was not able to reproduce this on a normal build and had to manually disable atomic_compare_exchange_n (for example) in config.status.
Since the bug seems impossible to reproduce on a normal build, I've decided to drop this patch for now. As for the failing tests, the first patch in the series fixes the issue in the general case. Best regards, Aleksa
