Mercurial > hg
diff mercurial/utils/storageutil.py @ 40445:634b45317459 stable
changegroup: restore default node ordering (issue6001)
Changeset db5501d9 changed the default node ordering from "storage" to
"linearize".
While the new API is more explicit and cleaner, the "linearize" order is
problematic on certain repositories like netbeans where it makes bundling
slower the more nodes we bundle.
Pushing and pulling 100 changesets was ~20% slower and pushing and pulling
1000 changesets was ~600% slower.
A very quick analysis of profile traces showed that the pull operation was
taking more time creating the delta.
Putting back the old default order seems to be the safe option. With more time
during the next cycle, we can understand better the impact of sorting with the
DAG order by default, the source of the regression and how to mitigate it.
/!\ We are still waiting for the full performance impact but with this patch,
bundling and pulling locally (not on the performance workstation) 1000
changesets on the netbeans repository is as fast as before the regression.
Differential Revision: https://phab.mercurial-scm.org/D5196
author | Boris Feld <boris.feld@octobus.net> |
---|---|
date | Wed, 31 Oct 2018 12:08:37 -0700 |
parents | b0fbd1792e2d |
children | 4fe63b573791 |
line wrap: on
line diff
--- a/mercurial/utils/storageutil.py Mon Oct 29 17:26:25 2018 +0100 +++ b/mercurial/utils/storageutil.py Wed Oct 31 12:08:37 2018 -0700 @@ -337,11 +337,11 @@ if nodesorder == 'nodes': revs = [frev(n) for n in nodes] - elif nodesorder == 'storage': - revs = sorted(frev(n) for n in nodes) - else: + elif nodesorder == 'linear': revs = set(frev(n) for n in nodes) revs = dagop.linearize(revs, store.parentrevs) + else: # storage and default + revs = sorted(frev(n) for n in nodes) prevrev = None