Xiubo Li <xiubli@xxxxxxxxxx> writes: > Hi Luis, > > On 27/02/2023 17:27, Luís Henriques wrote: >> xiubli@xxxxxxxxxx writes: >> >>> From: Xiubo Li <xiubli@xxxxxxxxxx> >>> >>> This patch series is based on Jeff Layton's previous great work and effort >>> on this. >>> >>> Since v15 we have added the ceph qa teuthology test cases for this [1][2], >>> which will test both the file name and contents encryption features and at >>> the same time they will also test the IO benchmarks. >>> >>> To support the fscrypt we also have some other work in ceph [3][4][5][6][7][8]: >>> >>> [1] https://github.com/ceph/ceph/pull/48628 >>> [2] https://github.com/ceph/ceph/pull/49934 >>> [3] https://github.com/ceph/ceph/pull/43588 >>> [4] https://github.com/ceph/ceph/pull/37297 >>> [5] https://github.com/ceph/ceph/pull/45192 >>> [6] https://github.com/ceph/ceph/pull/45312 >>> [7] https://github.com/ceph/ceph/pull/40828 >>> [8] https://github.com/ceph/ceph/pull/45224 >>> [9] https://github.com/ceph/ceph/pull/45073 >>> >>> The [8] and [9] are still undering testing and will soon be merged after >>> that. All the others had been merged long time ago. >> Thanks a lot for your work on this, Xiubo (and Jeff!). I assume this set >> is what's on the 'testing' branch. I've done some testing on that branch >> recently, but I'll start having another look at v16 and run some more >> tests. > > Yeah, this full fscrypt patch series has been in the 'testing' branch 2 months > ago as we discussed in the IRC. After that I just appended the following two > commits: > > > libceph: defer removing the req from osdc just after req->r_callback > ceph: drop the messages from MDS when unmounting > > > And also 2 small other minor fixes in the commit comments and folded the > following double free fixing patch reported by Ilya: > > https://patchwork.kernel.org/project/ceph-devel/patch/20230111011403.570964-1-xiubli@xxxxxxxxxx/ Awesome, thanks for confirming. Just wanted to make sure there wasn't anything new in this set. >> Am I right assuming that this v16 patchset means we are NOT getting this >> merged into 6.3? > > Yeah, not get merged yet. Because we were never able to put this to test since > we had lots of infra issues with teuthology in last 2 months. > > Since the teuthology Labs are back to work now we are running the tests and > after that the patches will be ready. Yeah, I saw these infrastructure issues and all the trouble it has been causing. Hopefully everything will be fine for the next merge window. Again, thanks for the update! Cheers -- Luís
