On Wed, Feb 27, 2019 at 11:44:14PM -0600, Steve French wrote: > Won't that change behavior and cause the success case (server returned > 0, and does support POSIX and &inode pointer is ok) to now return > -EINVAL. > You're right. I'm really sorry. I don't know why I misread the code... Also looking at it now, this was a false positive in Smatch. I have been working on that code in the past couple days and I don't have the warning in my latest build. regards, dan carpenter
