As it turns out we can't remove the forced initialization of the block bitmap when initializing the inode bitmap; e2fsck will complain because it explicitly checks for this condition. Thanks to Rakish Iyer at Google for finding this through his testing effort (debugging sucks; testing rocks!) and Curt Wohlgemuth for his analysis of the test failure and finding its cause. I've removed the first patch that was in the previous version of this patch series for now, and will be patching e2fsprogs to eliminate this check. Since we can't guarantee that people will be running a sufficiently new version of e2fsprogs, we'll probably have to keep the block bitmap initialization in the upstream kernel for at least a year or two. Theodore Ts'o (2): ext4: fold ext4_claim_inode into ext4_new_inode ext4: fix race when setting bitmap_uptodate flag fs/ext4/balloc.c | 59 +++++++++---- fs/ext4/ext4.h | 11 ++- fs/ext4/ialloc.c | 238 ++++++++++++++++++++++------------------------------- fs/ext4/mballoc.c | 79 ++++-------------- 4 files changed, 161 insertions(+), 226 deletions(-) -- 1.7.9.107.g8e04a -- 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
