Hi everyone,
As I discussed with Thomas on IRC, I'd still like to try to get Deepa's series merged to have 64-bit inode timestamps in VFS, as a prerequisite to fixing the individual file systems for y2038 as the next step.
The orignal tree that I've been testing with is in https://github.com/deepa-hub/vfs vfs_timespec64 (see description at [1] Unfortunately this was never in linux-next so far, so it's definitely against our normal rules, and if there are any objections (or new issues getting found), then we should defer it another merge window.
Until very recently, the branch worked fine with both linux-next and 4.17-rc, but with some last minute changes to the NFS and Overlayfs trees, it gained some nasty conflicts.
I've merged Trond's NFS branch and Miklos' Overlayfs branches into it, fixed up those conflicts and now uploaded it to my tree at git://git.kernel.org/pub/scm/linux/kernel/git/arnd/playground.git Obviously this will break if either of them rebases their trees once more, in which case I assume Stephen will drop my branch from linux-next.
Once those two are merged into mainline and nothing unexpected has come up from linux-next testing, we could either send a pull request to Linus for the merged branch, or regenerate the flag-day patch in it using coccinelle, which should result in identical contents.
Arnd
On Tue, 5 Jun 2018, Arnd Bergmann wrote:
Hi everyone,
As I discussed with Thomas on IRC, I'd still like to try to get Deepa's series merged to have 64-bit inode timestamps in VFS, as a prerequisite to fixing the individual file systems for y2038 as the next step.
The orignal tree that I've been testing with is in https://github.com/deepa-hub/vfs vfs_timespec64 (see description at [1] Unfortunately this was never in linux-next so far, so it's definitely against our normal rules, and if there are any objections (or new issues getting found), then we should defer it another merge window.
Until very recently, the branch worked fine with both linux-next and 4.17-rc, but with some last minute changes to the NFS and Overlayfs trees, it gained some nasty conflicts.
I've merged Trond's NFS branch and Miklos' Overlayfs branches into it, fixed up those conflicts and now uploaded it to my tree at git://git.kernel.org/pub/scm/linux/kernel/git/arnd/playground.git Obviously this will break if either of them rebases their trees once more, in which case I assume Stephen will drop my branch from linux-next.
Once those two are merged into mainline and nothing unexpected has come up from linux-next testing, we could either send a pull request to Linus for the merged branch, or regenerate the flag-day patch in it using coccinelle, which should result in identical contents.
I think there is no point to drag that out for the next merge window. The whole thing needs to be done in one go for the core changes which means that you're going to play that catchup game forever. Let's get over with it towards the end of the merge window.
Thanks,
tglx
-
Arnd Bergmann -
Thomas Gleixner