| commit | 82e6536e7b2b52f2ff08844c256b5c056271c977 | |
| author | Dan Carpenter <dan.carpenter@oracle.com> | |
| Tue, 8 Jan 2019 12:54:38 +0000 (8 15:54 +0300) | ||
| committer | Dan Carpenter <dan.carpenter@oracle.com> | |
| Tue, 8 Jan 2019 12:54:38 +0000 (8 15:54 +0300) | ||
| tree | 8cec837088898196e87688a7709fa146ccbf3769 | treesnapshot (tar.gz zip) |
| parent | 13a3ac197fa8a079387188c954e7753e1fa03c2c | commitdiff |
slist: handle the fast path in __merge_stree()
The normal thing when we're merging strees is for almost
all the states to stay the same. It's not totally obvious
that merge_sm_states() and add_possible() are basically no-ops
in this case but this patch makes that more clear.
I would have thought that this patch would be a slightly speed
up but my benchmark says it is a small slow down. Oh well. At
least the code is more obvious.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
The normal thing when we're merging strees is for almost
all the states to stay the same. It's not totally obvious
that merge_sm_states() and add_possible() are basically no-ops
in this case but this patch makes that more clear.
I would have thought that this patch would be a slightly speed
up but my benchmark says it is a small slow down. Oh well. At
least the code is more obvious.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
| smatch_slist.c | diffblobblamehistory |