On Mon, Feb 18, 2019 at 02:34:15PM +1300, Thomas Munro wrote:
> You mentioned that you were on RHEL 7.5, which ships with a 3.10
> kernel. When you eventually move to RHEL 8 (4.x) or even install an
> alternative newer kernel for 7.x, the problem will hopefully go away
> by itself. Considering that, the lack of other complaints, and the
> availability of a workaround (--no-sync followed by "sync"), I'm not
> so sure it's worth committing this.
If we get to ignore EINVAL, then we won't know that fsync has failed
even in cases where the caller has done an incorrect thing by using a
special file, which sounds like a bad idea for the core code as much
as any plugins calling that?
--
Michael