Hi folks, This is v4 of the following patchset https://lore.kernel.org/r/20201124155130.40848-1-hsiangkao@xxxxxxxxxx , which tends to simplify xfs_dialloc() logic. This version includes Dave's original patch https://lore.kernel.org/r/20201124221623.GC2842436@xxxxxxxxxxxxxxxxxxx to avoid the original double call of xfs_dialloc() and confusing ialloc_context with some split in order for better review and minor modification (e.g. ino isn't passed in xfs_ialloc()). I'm not quite sure what's messy ENOSPC mentioned in the original patch since return 0 and *ipp = NULL in xfs_dir_ialloc() would cause NULL-dereference in its callers, so I leave this part alone, although I think noroom has no use at all (it can be cleaned up as well with a following patch) (at a glance, the final shape looks almost ok...) I dropped [PATCH v1 3/3] since xfs_ialloc_select_ag() already looks simple enough (comments about this are welcome... I can re-add this if needed.) I'm re-runing xfstests -g auto with the whole series for now since a bit left behind this time, hopefully no extra issues here. I've dropped all RVBs from "[PATCH v4 3/6]" since some more modification here... Thanks for your time. changes since v3 (Christoph): - (2/6) get rid of a brace; - (2/6) keeping xfs_buf_relse() in the callers; - (2/6) get rid of "XXX:" comment; - (3/6) rename xfs_ialloc() to xfs_init_new_inode(); - (3/6) rename pino to parent_ino; - (3/6) convert the return type from int to struct xfs_inode * (so I dropped all RVBs of this patch, since more modification here....) - (4/6) refine a comment; - (5/6) add a found_ag lebel to have a central place for the sucessful return, since there are somewhat too many labels here, I didn't add another error lebel to make *IO_agbp = NULL and return error; - (6/6) drop the else after the break. changes since v2: - use struct xfs_dquot_acct * instead of void * on dqinfo (Darrick); - rename xfs_ialloc() to xfs_dir_ialloc_init() (Dave); - fix a temporary state compile error due to (ialloc_context -> &ialloc_context) on [PATCH v2 3/6]; - collect more RVBs from the reply of v2; - Cc Eric to confirm dqinfo; changes since v1: - add Dave's patch with spilt and minor update; - update comments above xfs_ialloc_ag_alloc() suggested by Darrick; - collect RVBs to "xfs: convert noroom, okalloc in xfs_dialloc() to bool" "xfs: kill ialloced in xfs_dialloc()" since no real logic changes ("(!error)" to "(error==0)" suggested by Darrick has been updated). Thanks, Gao Xiang Dave Chinner (4): xfs: introduce xfs_dialloc_roll() xfs: move on-disk inode allocation out of xfs_ialloc() xfs: move xfs_dialloc_roll() into xfs_dialloc() xfs: spilt xfs_dialloc() into 2 functions Gao Xiang (2): xfs: convert noroom, okalloc in xfs_dialloc() to bool xfs: kill ialloced in xfs_dialloc() fs/xfs/libxfs/xfs_ialloc.c | 174 +++++++++++++------------ fs/xfs/libxfs/xfs_ialloc.h | 36 +++--- fs/xfs/xfs_inode.c | 259 +++++++++---------------------------- fs/xfs/xfs_inode.h | 6 +- fs/xfs/xfs_qm.c | 27 ++-- fs/xfs/xfs_symlink.c | 8 +- 6 files changed, 197 insertions(+), 313 deletions(-) -- 2.18.4
