On Fri, Jan 14, 2022 at 08:25:31PM -0800, Randy Dunlap wrote: > Hi Luc, Hi Randy, Apologies for this reply delayed for much too long. > It seems that $subject version adds builtins for ATOMIC_ACQUIRE and > ATOMIC_RELEASE, but GCC has those too, so there are a few thousand > (OK, I didn't count them - I killed it quickly) of these: (e.g.) Yes indeed, it was one of the things added in this release. > .. CHECK ../init/do_mounts_initrd.c > .command-line: note: in included file: > builtin:1:9: sparse: warning: preprocessor token __ATOMIC_ACQUIRE redefined > builtin:0:0: sparse: this was the original definition > builtin:1:9: sparse: warning: preprocessor token __ATOMIC_SEQ_CST redefined > builtin:0:0: sparse: this was the original definition > builtin:1:9: sparse: warning: preprocessor token __ATOMIC_ACQ_REL redefined > builtin:0:0: sparse: this was the original definition > builtin:1:9: sparse: warning: preprocessor token __ATOMIC_RELEASE redefined > builtin:0:0: sparse: this was the original definition > > Any suggestions for how to avoid these warnings? Do you still see these? If yes, can you tell a bit about your setup and the exact command line used? I'm asking because I never saw a problem with this, same for the test bots. It should be exactly the same as for any other predefined value. However, as predefined, they should all be reported from 'builtin:0:0:' and the 'builtin:1:9:' look as if those are from a '-include <somefile.h>' on the command line. Best regards, -- Luc
