On Mon, Mar 5, 2012 at 2:48 PM, Andy Whitcroft <apw@xxxxxxxxxxxxx> wrote: > I was not trying to suggest that code could look at whether > ->permission() was NULL from above and use that to rely on the values > *shudder*. I was mearly expressing the feeling that it seemed that > there was a distinct possibily that those fields were really only > there for the ops and below to use. From your description here it > seems we should be able to assume these fields are populated, which > makes these Yama checks valid and so I need to look to have them > populated. I'd argue that overlayfs is also special (anything that opens a file will only be able to access the underlying inode) and Yama should do a vfs_getattr() to access the file attributes. Thanks, Miklos -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
