On Thu, Dec 17, 2020 at 10:00:37AM -0500, Jeff Layton wrote: > Overlayfs's volatile mounts want to be able to sample an error for their > own purposes, without preventing a later opener from potentially seeing > the error. umm ... can't they just copy the errseq_t they're interested in, followed by calling errseq_check() later? actually, isn't errseq_check() buggy in the face of multiple watchers? consider this: worker.es starts at 0 t2.es = errseq_sample(&worker.es) errseq_set(&worker.es, -EIO) t1.es = errseq_sample(&worker.es) t2.err = errseq_check_and_advance(&es, t2.es) ** this sets ERRSEQ_SEEN ** t1.err = errseq_check(&worker.es, t1.es) ** reports an error, even though the only change is that ERRSEQ_SEEN moved **. i think errseq_check() should be: if (likely(cur | ERRSEQ_SEEN) == (since | ERRSEQ_SEEN)) return 0; i'm not yet convinced other changes are needed to errseq. but i am having great trouble understanding exactly what overlayfs is trying to do.
