Fix long-standing excessive file merges
Since switching to the multihead approach, we've been creating
excessive file-level merges where files are marked as merged with
their ancestors.
This explicitly checks at commit time whether the two parent versions
are linearly related, and if so, reduces the file check-in to a
non-merge. Then the file is compared against the remaining parent,
and, if equal, skips check-in of that file (as it's not changed).
Since we're not checking in all files that were different between
versions, we no longer need to mark so many files for merge. This
removes most of the 'm' state marking as well.
Finally, it is possible to do a tree-level merge with no file-level
changes. This will happen if one user changes file A and another
changes file B. Thus, if we have have two parents, we allow commit to
proceed even if there are no file-level changes.
Mercurial git BK (*)
storage revlog delta compressed revisions SCCS weave
storage naming by filename by revision hash by filename
merge file DAGs changeset DAG file DAGs?
consistency SHA1 SHA1 CRC
signable? yes yes no
retrieve file tip O(1) O(1) O(revs)
add rev O(1) O(1) O(revs)
find prev file rev O(1) O(changesets) O(revs)
annotate file O(revs) O(changesets) O(revs)
find file changeset O(1) O(changesets) ?
checkout O(files) O(files) O(revs)?
commit O(changes) O(changes) ?
6 patches/s 6 patches/s slow
diff working dir O(changes) O(changes) ?
< 1s < 1s ?
tree diff revs O(changes) O(changes) ?
< 1s < 1s ?
hardlink clone O(files) O(revisions) O(files)
find remote csets O(log new) rsync: O(revisions) ?
git-http: O(changesets)
pull remote csets O(patch) O(modified files) O(patch)
repo growth O(patch) O(revisions) O(patch)
kernel history 300M 3.5G? 250M?
lines of code 2500 6500 (+ cogito) ??
* I've never used BK so this is just guesses