On Tue, 20 Jul 2021 at 17:19, Miklos Szeredi <miklos@xxxxxxxxxx> wrote: > So on one instance a file on lower gets executed and on another > instance sharing the lower layer the file is truncated. The truncate > is currently denied due to the negative i_writecount on the lower > file. Also behavior is inconsistent between open(path, O_TRUNC) and > truncate(path) even though the two should be equivalent. > > Applied with the following description: > [...] Also adding the following documentation in the "Non-standard behavior" section: c) If a file residing on a lower layer is being executed, then opening that file for write or truncating the file will not be denied with ETXTBSY. Looked at the POSIX standard and it only documents ETXTBUSY for O_RDWR and O_WRONLY and not for truncate(2) or O_TRUNC. So strictly speaking this patch doesn't even change the POSIX correctness. Thanks, Miklos
