Mercurial > public > mercurial-scm > hg
annotate rust/hg-core/src/revlog/nodemap_docket.rs @ 48178:f12a19d03d2c
fix: reduce number of tool executions
By grouping together (path, ctx) pairs according to the inputs they would
provide to fixer tools, we can deduplicate executions of fixer tools to
significantly reduce the amount of time spent running slow tools.
This change does not handle clean files in the working copy, which could still
be deduplicated against the files in the checked out commit. It's a little
harder to do that because the filerev is not available in the workingfilectx
(and it doesn't exist for added files).
Anecdotally, this change makes some real uses cases at Google 10x faster. I
think we were originally hesitant to do this because the benefits weren't
obvious, and implementing it efficiently is kind of tricky. If we simply
memoized the formatter execution function, we would be keeping tons of file
content in memory.
Also included is a regression test for a corner case that I broke with my first
attempt at optimizing this code.
Differential Revision: https://phab.mercurial-scm.org/D11280
author | Danny Hooper <hooper@google.com> |
---|---|
date | Thu, 02 Sep 2021 14:08:45 -0700 |
parents | 001d747c2baf |
children | 07d8d144c222 |
rev | line source |
---|---|
46443
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
1 use crate::errors::{HgError, HgResultExt}; |
46630
842f2372ced6
rhg: Don?t attempt to read persistent nodemap without .hg/requires opt-in
Simon Sapin <simon.sapin@octobus.net>
parents:
46443
diff
changeset
|
2 use crate::requirements; |
46390
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
3 use bytes_cast::{unaligned, BytesCast}; |
47955
e834b79def74
rust: Switch to the memmap2-rs crate
Simon Sapin <simon.sapin@octobus.net>
parents:
46630
diff
changeset
|
4 use memmap2::Mmap; |
46090
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
5 use std::path::{Path, PathBuf}; |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
6 |
46167
8a4914397d02
rust: introduce Repo and Vfs types for filesystem abstraction
Simon Sapin <simon.sapin@octobus.net>
parents:
46090
diff
changeset
|
7 use crate::repo::Repo; |
46090
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
8 use crate::utils::strip_suffix; |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
9 |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
10 const ONDISK_VERSION: u8 = 1; |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
11 |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
12 pub(super) struct NodeMapDocket { |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
13 pub data_length: usize, |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
14 // TODO: keep here more of the data from `parse()` when we need it |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
15 } |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
16 |
46390
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
17 #[derive(BytesCast)] |
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
18 #[repr(C)] |
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
19 struct DocketHeader { |
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
20 uid_size: u8, |
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
21 _tip_rev: unaligned::U64Be, |
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
22 data_length: unaligned::U64Be, |
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
23 _data_unused: unaligned::U64Be, |
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
24 tip_node_size: unaligned::U64Be, |
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
25 } |
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
26 |
46090
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
27 impl NodeMapDocket { |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
28 /// Return `Ok(None)` when the caller should proceed without a persistent |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
29 /// nodemap: |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
30 /// |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
31 /// * This revlog does not have a `.n` docket file (it is not generated for |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
32 /// small revlogs), or |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
33 /// * The docket has an unsupported version number (repositories created by |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
34 /// later hg, maybe that should be a requirement instead?), or |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
35 /// * The docket file points to a missing (likely deleted) data file (this |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
36 /// can happen in a rare race condition). |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
37 pub fn read_from_file( |
46167
8a4914397d02
rust: introduce Repo and Vfs types for filesystem abstraction
Simon Sapin <simon.sapin@octobus.net>
parents:
46090
diff
changeset
|
38 repo: &Repo, |
46090
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
39 index_path: &Path, |
47963
001d747c2baf
rust: Return HgError instead of RevlogError in revlog constructors
Simon Sapin <simon.sapin@octobus.net>
parents:
47955
diff
changeset
|
40 ) -> Result<Option<(Self, Mmap)>, HgError> { |
46630
842f2372ced6
rhg: Don?t attempt to read persistent nodemap without .hg/requires opt-in
Simon Sapin <simon.sapin@octobus.net>
parents:
46443
diff
changeset
|
41 if !repo |
842f2372ced6
rhg: Don?t attempt to read persistent nodemap without .hg/requires opt-in
Simon Sapin <simon.sapin@octobus.net>
parents:
46443
diff
changeset
|
42 .requirements() |
842f2372ced6
rhg: Don?t attempt to read persistent nodemap without .hg/requires opt-in
Simon Sapin <simon.sapin@octobus.net>
parents:
46443
diff
changeset
|
43 .contains(requirements::NODEMAP_REQUIREMENT) |
842f2372ced6
rhg: Don?t attempt to read persistent nodemap without .hg/requires opt-in
Simon Sapin <simon.sapin@octobus.net>
parents:
46443
diff
changeset
|
44 { |
842f2372ced6
rhg: Don?t attempt to read persistent nodemap without .hg/requires opt-in
Simon Sapin <simon.sapin@octobus.net>
parents:
46443
diff
changeset
|
45 // If .hg/requires does not opt it, don’t try to open a nodemap |
842f2372ced6
rhg: Don?t attempt to read persistent nodemap without .hg/requires opt-in
Simon Sapin <simon.sapin@octobus.net>
parents:
46443
diff
changeset
|
46 return Ok(None); |
842f2372ced6
rhg: Don?t attempt to read persistent nodemap without .hg/requires opt-in
Simon Sapin <simon.sapin@octobus.net>
parents:
46443
diff
changeset
|
47 } |
842f2372ced6
rhg: Don?t attempt to read persistent nodemap without .hg/requires opt-in
Simon Sapin <simon.sapin@octobus.net>
parents:
46443
diff
changeset
|
48 |
46090
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
49 let docket_path = index_path.with_extension("n"); |
46443
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
50 let docket_bytes = if let Some(bytes) = |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
51 repo.store_vfs().read(&docket_path).io_not_found_as_none()? |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
52 { |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
53 bytes |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
54 } else { |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
55 return Ok(None); |
46090
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
56 }; |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
57 |
46390
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
58 let input = if let Some((&ONDISK_VERSION, rest)) = |
46090
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
59 docket_bytes.split_first() |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
60 { |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
61 rest |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
62 } else { |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
63 return Ok(None); |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
64 }; |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
65 |
46443
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
66 /// Treat any error as a parse error |
47963
001d747c2baf
rust: Return HgError instead of RevlogError in revlog constructors
Simon Sapin <simon.sapin@octobus.net>
parents:
47955
diff
changeset
|
67 fn parse<T, E>(result: Result<T, E>) -> Result<T, HgError> { |
001d747c2baf
rust: Return HgError instead of RevlogError in revlog constructors
Simon Sapin <simon.sapin@octobus.net>
parents:
47955
diff
changeset
|
68 result |
001d747c2baf
rust: Return HgError instead of RevlogError in revlog constructors
Simon Sapin <simon.sapin@octobus.net>
parents:
47955
diff
changeset
|
69 .map_err(|_| HgError::corrupted("nodemap docket parse error")) |
46443
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
70 } |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
71 |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
72 let (header, rest) = parse(DocketHeader::from_bytes(input))?; |
46390
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
73 let uid_size = header.uid_size as usize; |
46090
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
74 // TODO: do we care about overflow for 4 GB+ nodemap files on 32-bit |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
75 // systems? |
46390
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
76 let tip_node_size = header.tip_node_size.get() as usize; |
0800aa42bb4c
rust: use the bytes-cast crate to parse persistent nodemaps
Simon Sapin <simon.sapin@octobus.net>
parents:
46167
diff
changeset
|
77 let data_length = header.data_length.get() as usize; |
46443
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
78 let (uid, rest) = parse(u8::slice_from_bytes(rest, uid_size))?; |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
79 let (_tip_node, _rest) = |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
80 parse(u8::slice_from_bytes(rest, tip_node_size))?; |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
81 let uid = parse(std::str::from_utf8(uid))?; |
46090
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
82 let docket = NodeMapDocket { data_length }; |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
83 |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
84 let data_path = rawdata_path(&docket_path, uid); |
46443
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
85 // TODO: use `vfs.read()` here when the `persistent-nodemap.mmap` |
46090
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
86 // config is false? |
46443
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
87 if let Some(mmap) = repo |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
88 .store_vfs() |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
89 .mmap_open(&data_path) |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
90 .io_not_found_as_none()? |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
91 { |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
92 if mmap.len() >= data_length { |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
93 Ok(Some((docket, mmap))) |
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
94 } else { |
47963
001d747c2baf
rust: Return HgError instead of RevlogError in revlog constructors
Simon Sapin <simon.sapin@octobus.net>
parents:
47955
diff
changeset
|
95 Err(HgError::corrupted("persistent nodemap too short")) |
46090
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
96 } |
46443
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
97 } else { |
46630
842f2372ced6
rhg: Don?t attempt to read persistent nodemap without .hg/requires opt-in
Simon Sapin <simon.sapin@octobus.net>
parents:
46443
diff
changeset
|
98 // Even if .hg/requires opted in, some revlogs are deemed small |
842f2372ced6
rhg: Don?t attempt to read persistent nodemap without .hg/requires opt-in
Simon Sapin <simon.sapin@octobus.net>
parents:
46443
diff
changeset
|
99 // enough to not need a persistent nodemap. |
46443
43d63979a75e
rust: use HgError in RevlogError and Vfs
Simon Sapin <simon.sapin@octobus.net>
parents:
46390
diff
changeset
|
100 Ok(None) |
46090
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
101 } |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
102 } |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
103 } |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
104 |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
105 fn rawdata_path(docket_path: &Path, uid: &str) -> PathBuf { |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
106 let docket_name = docket_path |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
107 .file_name() |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
108 .expect("expected a base name") |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
109 .to_str() |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
110 .expect("expected an ASCII file name in the store"); |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
111 let prefix = strip_suffix(docket_name, ".n.a") |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
112 .or_else(|| strip_suffix(docket_name, ".n")) |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
113 .expect("expected docket path in .n or .n.a"); |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
114 let name = format!("{}-{}.nd", prefix, uid); |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
115 docket_path |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
116 .parent() |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
117 .expect("expected a non-root path") |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
118 .join(name) |
9eb07ab3f2d4
rhg: use persistent nodemap when available
Simon Sapin <simon-commits@exyr.org>
parents:
diff
changeset
|
119 } |