On Mon, Jan 16, 2017 at 4:02 PM, Miklos Szeredi <miklos@xxxxxxxxxx> wrote: > On Sun, Jan 15, 2017 at 2:57 PM, Amir Goldstein <amir73il@xxxxxxxxx> wrote: >> This is needed for choosing between concurrent copyup >> using O_TMPFILE and legacy copyup using workdir+rename. > > I'm really wondering if we should constrain upper fs to those that: > > - can do RENAME_EXCHANGE and RENAME_WHITEOUT > - can do ->tmpfile() which is currently a superset of the above > - can do xattr, again a superset Makes sense to me. Let me know if you want me to add the rename flag test. > > The question is whether anybody actually using it with an fs that > doesn't have all of the above. Because if so, we need to keep > supporting them. Perhaps we should add warnings about deprecation and > if nobody complains we can remove support for non-conformant fs. > But how exactly do we "support" those fs right now? Any attempt to use them would result in -EINVAL, because we will bw requesting RENAME_EXCHANGE and RENAME_WHITEOUT -- 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
