Hi, This is the v3 patchset for testing the XFS umount hang problem caused by the pending dquota log item in AIL. The problem has not been fixed in XFS yet, and Carlos Maiolino is working on it [1]. The bug that the test case can not reproduced on a 15GB scratch device has been fixed, and many thanks to Eryu Guan. I had tested it on 4.14-rc8, and the problem can be reproduced reliably now. Comments and questions are welcome. Regards, Tao --- Changes since v2: v3: * dmflakey: support error_writes feature for dm-flakey * check the availability of error_writes by trial instead of version number * xfs: test for umount hang caused by the pending dquota log item in AIL * calculate log_ofs correctly by using convert cmd in xfs_db * common/rc: factor out _scratch_xfs_[get|set]_sb_field * reimplement _scratch_xfs_[get|set]_sb_field based on _scratch_xfs_[get|set]_metadata_field. v2: * http://www.spinics.net/lists/fstests/msg07792.html v1: * https://www.spinics.net/lists/fstests/msg07622.html [1]: https://www.spinics.net/lists/linux-xfs/msg11614.html Hou Tao (4): dmflakey: support multiple dm targets for a dm-flakey device dmflakey: support error_writes feature for dm-flakey xfs: test for umount hang caused by the pending dquota log item in AIL common/rc: factor out _scratch_xfs_[get|set]_sb_field common/dmflakey | 29 +++++++++- common/fuzzy | 33 ----------- common/xfs | 47 ++++++++++++++++ tests/xfs/007 | 6 +- tests/xfs/098 | 4 +- tests/xfs/186 | 3 +- tests/xfs/199 | 13 ++--- tests/xfs/307 | 11 +--- tests/xfs/308 | 11 +--- tests/xfs/339 | 6 +- tests/xfs/340 | 2 +- tests/xfs/999 | 163 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ tests/xfs/999.out | 2 + tests/xfs/group | 1 + 14 files changed, 257 insertions(+), 74 deletions(-) create mode 100755 tests/xfs/999 create mode 100644 tests/xfs/999.out -- 2.9.5 -- To unsubscribe from this list: send the line "unsubscribe fstests" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
