Hi all, This is a RFC-quality pass at making xfstests perform more rigorous testing of the btrfs/xfs file clone, reflink, and dedupe ioctls. There are now tests of the basic functionality of the three ioctls; tests to ensure that the filesystem exhibits the expected copy on write semantics; tests to try to suss out race conditions in the new write paths; tests to ensure that the ioctls peform basic disk accounting correctly; tests of the interaction between reflink and the various fallocate verbs (allocate, punch, collapse, insert zeroes); and some attempts to test the upper limits of reflinking. The first patch in the series adds fuzz testing to ext4 and XFS; aside from being first in line, it isn't tied to any of the reflink functionality. To run these tests, you'll have to patch xfsprogs to have reflink and dedupe support[1]; the patch ought to apply fairly cleanly against the upstream git. They should more or less work with the btrfs that appears in 4.1, though if you want to test the XFS implementation, you're going to have to apply a lot of patches to the kernel and xfsprogs. See the cover letters[2][3] for those patchsets for more information. Known issues: * I think the race checks for dedupe could be a little sharper at finding mistakes. * I started the numbering really high to prevent the tests from colliding with whatever new tests might arrive; this will require some intervention to fix. * When ext4 gains reflink support, it shouldn't be difficult to make these tests run on it. The patch set is based on the current xfstest master on kernel.org. * If the copy_file_range syscall ever comes around, we'll have to adapt xfs_io to use that in addition to the btrfs ioctls. Comments and questions are, as always, welcome. --D [1] http://djwong.org/docs/03-xfs_io-reflink-and-dedupe.patch [2] See thread "[RFC 00/15] xfsprogs: support the reflink btree" dated today. [3] http://oss.sgi.com/archives/xfs/2015-06/msg00407.html -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
