sg/commit-graph-cleanups (was Re: What's cooking in git.git (Jun 2020, #03; Wed, 17))
- From
Derrick Stolee <stolee@gmail.com>
- Date
- Jun 18, 2020, 13:32 UTC
- Message-ID
- <76f63479-c142-86f2-c045-f802f7794a7d@gmail.com>
- In-Reply-To
- <xmqqpn9wq3tl.fsf@gitster.c.googlers.com>
On 6/18/2020 2:35 AM, Junio C Hamano wrote:
> * sg/commit-graph-cleanups (2020-06-17) 18 commits
You have this topic marked with "Code cleanup" which is definitely true of the first submission, but not the second.
> - commit-graph: persist existence of changed-paths > - commit-graph: change test to die on parse, not load
These two add an important behavior change to persist the changed-paths after a user opts-in to using them.
> - bloom: enforce a minimum size of 8 bytes > - commit-graph: check all leading directories in changed path Bloom filters
These two have meaningful performance gains.
> - commit-graph: check chunk sizes after writing > - commit-graph: simplify chunk writes into loop > - commit-graph: unify the signatures of all write_graph_chunk_*() functions > - commit-graph: place bloom_settings in context
These four (from the second series) _are_ cleanups, but were harder to apply than the others from the first series.
Show 10 quoted lines
> - commit-graph: simplify write_commit_graph_file() #2 > - commit-graph: simplify write_commit_graph_file() #1 > - commit-graph: simplify parse_commit_graph() #2 > - commit-graph: simplify parse_commit_graph() #1 > - commit-graph: clean up #includes > - diff.h: drop diff_tree_oid() & friends' return value > - commit-slab: add a function to deep free entries on the slab > - commit-graph-format.txt: all multi-byte numbers are in network byte order > - commit-graph: fix parsing the Chunk Lookup table > - tree-walk.c: don't match submodule entries for 'submod/anything'
These are all cleanups from the first series.
In summary, if you would like to keep these together as a single topic, then perhaps this summary might be good:
The changed-path Bloom filter is improved using ideas from an independent implementation.
Thanks, -Stolee