On Mon, Aug 31, 2020 at 4:22 PM cgxu <cgxu519@xxxxxxxxxxxx> wrote: > > On 5/6/20 5:53 PM, Chengguang Xu wrote: > > Current syncfs(2) syscall on overlayfs just calls sync_filesystem() > > on upper_sb to synchronize whole dirty inodes in upper filesystem > > regardless of the overlay ownership of the inode. In the use case of > > container, when multiple containers using the same underlying upper > > filesystem, it has some shortcomings as below. > > > > (1) Performance > > Synchronization is probably heavy because it actually syncs unnecessary > > inodes for target overlayfs. > > > > (2) Interference > > Unplanned synchronization will probably impact IO performance of > > unrelated container processes on the other overlayfs. > > Hi Miklos, Jack, Amir and folks > > Recently I got another idea to mitigate the syncfs interferes between > instances, I would like to talk with you guys first before I post > full patch series and hope to get some comments about it. Isn't stacked mmap supposed to fix that one as well? I.e. overlay inode will contain correct dirty state and VFS can correctly sync only those inodes which were dirtied by this instance. Thanks, Miklos
