On Sat, May 07, 2022 at 02:03:09PM +0200, Christian Brauner wrote: > On Sun, May 01, 2022 at 07:46:43PM +0800, Zorro Lang wrote: > > On Thu, Apr 28, 2022 at 05:15:48PM +0200, Christian Brauner wrote: > > > From: "Christian Brauner (Microsoft)" <brauner@xxxxxxxxxx> > > > > > > Hey everyone, > > > > > > As announced multiple times already we need to rename and split the > > > idmapped mount testsuite into separate source files and also give it a > > > better name to reflect the fact that it covers a lot more than just > > > idmapped mounts. > > > > > > I have decided against compiling different binaries for now. Instead we > > > compile a single vfstest binary that can be called with various command > > > line switches to run the various test suites. This is not different than > > > what we did for the idmapped-mounts binary. Of course, nothing prevents > > > us from using multiple binaries in the future. > > > > > > Note that this will have a conflict with Jan's series at: > > > https://lore.kernel.org/fstests/20220425131809.qzjrah7cw67mzzcw@zlang-mailbox/T/#m47d8da68ef1aff250918398e8d2228729a6acf97 > > > > Hi Christian, > > > > Thanks for your patches! But even without above conflict patches, I still can't > > merge your patchset, failures as below [1]. > > Hi Zorro! > (Resending, since my last message had a messed-up To: line.) > > The fstests list has a limit on how large a single patch can be in terms > of number of lines (10.000, I believe). Since this patchset contains > patches that change about 10.000 lines the fstests list drops those > patches. I ran into this issue before when we added the testsuite in the > first place but it has never been fixed. > > The consequence of this is that b4 is missing patches since they are on > no list. This becomes more obvious if you use b4's quilt option which > creates a folder with individual patches: > > 0001_src_rename_idmapped_mounts_folder.patch > 0002_src_vfs_rename_idmapped_mounts_c_file.patch > 0003_vfstest_rename_struct_t_idmapped_mounts.patch > 0004_utils_add_missing_global_h_include.patch > 0006_utils_move_helpers_into_utils.patch > 0007_missing_move_sys_execveat_to_missing_h.patch > 0008_utils_add_struct_test_suite.patch > 0011_vfstest_split_out_remaining_idmapped_mount_tests.patch > series > > so this is missing patches 9 and 10... Oh, I didn't notice that. Let me check ... > > > May you rebase your patches on latest fstests for-next branch, and send out > > again? As Christoph Hellwig has Acked your patchset, we'd better to make it > > I think we will have the same problem, i.e., patches will not be on the > list since they are too large and so b4 won't work. To work around this > I'll give you a signed tag to pull from in the cover letter. Hmm...I just used another method(avoid b4) to merge your 11 patches, I think it works: # git log --oneline 14714a1b (HEAD -> for-brauner) vfstest: split out remaining idmapped mount tests ed027078 vfstest: split out btrfs idmapped mounts test fd026c1a vfstests: split idmapped mount tests into separate suite 1cec19ec utils: add struct test_suite 3dadfe72 missing: move sys_execveat() to missing.h ba0370e4 utils: move helpers into utils 781f86e1 utils: add struct vfstest_info e1ffed6b utils: add missing global.h include 4f68af21 vfstest: rename struct t_idmapped_mounts 8fcd0f72 src/vfs: rename idmapped-mounts.c file 0b03254e src: rename idmapped-mounts folder fbc6486b (redhat/master, origin/master, origin/HEAD, rh-master, master) generic: test that renaming into a directory fails with EDQUOT ... ... And yes, patch 9 and 10 are really huge: $ git show ed027078 | wc -l 8223 $ git show fd026c1a | wc -l 15641 But I still can't merge it into for-next branch, so you have to do once rebase at least. As we have 3 more idmapped patches to from XuYang to be merged, you'd better to wait the fstests release this week, then do once rebase on for-next branch. Hope that can save a little of your time. Thanks, Zorro > > Christian >
