ext4: retry allocations conservatively
authorTheodore Ts'o <tytso@mit.edu>
Sun, 1 Oct 2017 21:59:54 +0000 (17:59 -0400)
committerTheodore Ts'o <tytso@mit.edu>
Sun, 1 Oct 2017 21:59:54 +0000 (17:59 -0400)
commit68fd97504ad2f70850c47ce45caa110a0ca843d2
tree7cee951697a59a7c4f3dfe939de6339e3869d772
parent545052e9e35a34af95d2e870ac3fe2894376e6e9
ext4: retry allocations conservatively

Now that we no longer try to reserve metadata blocks for delayed
allocations (which tended to overestimate the required number of
blocks significantly), we really don't need retry allocations when the
disk is very full as aggressively any more.

The only time when it makes sense to retry an allocation is if we have
freshly deleted blocks that will only become available after a
transaction commit.  And if we lose that race, it's not worth it to
try more than once.

Signed-off-by: Theodore Ts'o <tytso@mit.edu>
fs/ext4/balloc.c