Re: [PATCH v1 2/2] fs/writeback: skip inodes with potential writeback hang in wait_sb_inodes()

On Tue, 25 Nov 2025 at 02:10, Joanne Koong joannelkoong@gmail.com wrote:

Prior to the changes added in commit 0c58a97f919c ("fuse: remove tmp folio for writebacks and internal rb tree"), fuse didn't ensure that data was written back for sync. The folio was marked as not under writeback anymore, even if it was still under writeback.

This is the main point. Fuse has existed for 20 years without data integrity guarantees. Reverting to that behavior is *not* a regression, it's simply a decades old bug. And solving that bug is darn hard, which is why it's not an option at this point.

One idea to limit the scope of this revert is to constrain it to suspend. I'm not sure how easy or ugly that might be.

Thanks, Miklos