ext4: re-enable extent zeroout optimization on encrypted files

For encrypted files, commit 36086d43f657 ("ext4 crypto: fix bugs in
ext4_encrypted_zeroout()") disabled the optimization where when a write
occurs to the middle of an unwritten extent, the head and/or tail of the
extent (when they aren't too large) are zeroed out, turned into an
initialized extent, and merged with the part being written to.  This
optimization helps prevent fragmentation of the extent tree.

However, disabling this optimization also made fscrypt_zeroout_range()
nearly impossible to test, as now it's only reachable via the very rare
case in ext4_split_extent_at() where allocating a new extent tree block
fails due to ENOSPC.  'gce-xfstests -c ext4/encrypt -g auto' doesn't
even hit this at all.

It's preferable to avoid really rare cases that are hard to test.

That commit also cited data corruption in xfstest generic/127 as a
reason to disable the extent zeroout optimization, but that's no longer
reproducible anymore.  It also cited fscrypt_zeroout_range() having poor
performance, but I've written a patch to fix that.

Therefore, re-enable the extent zeroout optimization on encrypted files.

Signed-off-by: Eric Biggers <ebiggers@google.com>
Link: https://lore.kernel.org/r/20191226161114.53606-1-ebiggers@kernel.org
Signed-off-by: Theodore Ts'o <tytso@mit.edu>

diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 4ba8215..ed55ca7 100644 (file)
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -3719,9 +3719,6 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
                max_zeroout = sbi->s_extent_max_zeroout_kb >>
                        (inode->i_sb->s_blocksize_bits - 10);
 
-       if (IS_ENCRYPTED(inode))
-               max_zeroout = 0;
-
        /*
         * five cases:
         * 1. split the extent into three extents.