Btreemap iter intertwined #71510

ssomers · 2020-04-24T13:31:37Z

3 commits:

Introduced benchmarks for BTreeMap::iter(). Benchmarks named iter_20 were of the whole iteration process, so I renamed them. Also the benchmarks of range that I wrote earlier weren't very good. I included an (awkwardly named) one that compares iter() to range(..) on the same set, because the contrast is surprising:

 name                                           ns/iter
 btree::map::range_unbounded_unbounded          28,176
 btree::map::range_unbounded_vs_iter            89,369

Both dig up the same pair of leaf edges. range(..) also checks that some keys are correctly ordered, the only thing iter() does more is to copy the map's length.

Slightly refactoring the code to what I find more readable (not in chronological order of discovery), boosts performance:

>cargo-benchcmp.exe benchcmp a1 a2 --threshold 5
 name                                   a1 ns/iter  a2 ns/iter  diff ns/iter   diff %  speedup
 btree::map::find_rand_100              18          17                    -1   -5.56%   x 1.06
 btree::map::first_and_last_10k         64          71                     7   10.94%   x 0.90
 btree::map::iter_0                     2,939       2,209               -730  -24.84%   x 1.33
 btree::map::iter_1                     6,845       2,696             -4,149  -60.61%   x 2.54
 btree::map::iter_100                   8,556       3,672             -4,884  -57.08%   x 2.33
 btree::map::iter_10k                   9,292       5,884             -3,408  -36.68%   x 1.58
 btree::map::iter_1m                    10,268      6,510             -3,758  -36.60%   x 1.58
 btree::map::iteration_mut_100000       478,575     453,050          -25,525   -5.33%   x 1.06
 btree::map::range_unbounded_unbounded  28,176      36,169             7,993   28.37%   x 0.78
 btree::map::range_unbounded_vs_iter    89,369      38,290           -51,079  -57.16%   x 2.33
 btree::set::clone_100_and_remove_all   4,801       4,245               -556  -11.58%   x 1.13
 btree::set::clone_10k_and_remove_all   529,450     496,030          -33,420   -6.31%   x 1.07

But you can tell from the range_unbounded_* lines that, despite an unwarranted, vengeful attack on the range_unbounded_unbounded benchmark, this change still doesn't allow iter() to catch up with range(..).

I guess that range(..) copes so well because it intertwines the leftmost and rightmost descend towards leaf edges, doing the two root node accesses close together, perhaps exploiting a CPU's internal pipelining? So the third commit distils a version of range_search (which we can't use directly because of the Ord bound), and we get another boost:

cargo-benchcmp.exe benchcmp a2 a3 --threshold 5
 name                                   a2 ns/iter  a3 ns/iter  diff ns/iter   diff %  speedup
 btree::map::first_and_last_100         40          43                     3    7.50%   x 0.93
 btree::map::first_and_last_10k         71          64                    -7   -9.86%   x 1.11
 btree::map::iter_0                     2,209       1,719               -490  -22.18%   x 1.29
 btree::map::iter_1                     2,696       2,205               -491  -18.21%   x 1.22
 btree::map::iter_100                   3,672       2,943               -729  -19.85%   x 1.25
 btree::map::iter_10k                   5,884       3,929             -1,955  -33.23%   x 1.50
 btree::map::iter_1m                    6,510       5,532               -978  -15.02%   x 1.18
 btree::map::iteration_mut_100000       453,050     476,667           23,617    5.21%   x 0.95
 btree::map::range_included_excluded    405,075     371,297          -33,778   -8.34%   x 1.09
 btree::map::range_included_included    427,577     397,440          -30,137   -7.05%   x 1.08
 btree::map::range_unbounded_unbounded  36,169      28,175            -7,994  -22.10%   x 1.28
 btree::map::range_unbounded_vs_iter    38,290      30,838            -7,452  -19.46%   x 1.24

But I think this is just fake news from the microbenchmarking media. iter() is still trying to catch up with range(..). And we can sure do without another function. So I would skip this 3rd commit.

r? @Mark-Simulacrum

Mark-Simulacrum

This looks good to me -- while looking at this I also thought #71523 would be a nice cleanup, so posted that separately.

Mark-Simulacrum · 2020-04-24T17:31:44Z

Not sure if the draft status is intended to be indicative that you're planning more work here. I'll hold off on approving until you ping me confirming you're happy with this.

ssomers · 2020-04-24T19:27:26Z

I made it draft partly because I was wondering whether the CI would succeed, since x.py test fails for me (in the latest masters, something about doc). But also, I don't like the 3rd commit adding a bunch of code that already exists in two other flavours and I don't think it actually boosts performance in real code. Though I do like having a function shared by the 3 types of iteration and similar to the 2 types of range.

…f edges

ssomers · 2020-04-24T22:55:39Z

Okay, I adapted the 2nd and 3rd commit to #71523 and squashed them, and it looks neat enough to me now. Performance now says:

 name                                   c1 ns/iter  c2 ns/iter  diff ns/iter   diff %  speedup
 btree::map::find_rand_100              18          17                    -1   -5.56%   x 1.06
 btree::map::first_and_last_100         40          43                     3    7.50%   x 0.93
 btree::map::iter_0                     3,012       1,774             -1,238  -41.10%   x 1.70
 btree::map::iter_1                     7,016       2,255             -4,761  -67.86%   x 3.11
 btree::map::iter_100                   8,770       3,510             -5,260  -59.98%   x 2.50
 btree::map::iter_10k                   9,523       4,017             -5,506  -57.82%   x 2.37
 btree::map::iter_1m                    10,525      5,612             -4,913  -46.68%   x 1.88
 btree::map::iteration_20               86          81                    -5   -5.81%   x 1.06
 btree::map::iteration_mut_100000       490,270     464,320          -25,950   -5.29%   x 1.06
 btree::map::range_included_excluded    408,180     379,105          -29,075   -7.12%   x 1.08
 btree::map::range_included_included    431,697     407,905          -23,792   -5.51%   x 1.06
 btree::map::range_included_unbounded   116,462     105,607          -10,855   -9.32%   x 1.10
 btree::map::range_unbounded_unbounded  28,880      37,183             8,303   28.75%   x 0.78
 btree::map::range_unbounded_vs_iter    91,573      33,270           -58,303  -63.67%   x 2.75
 btree::set::clone_100_and_remove_all   4,900       4,237               -663  -13.53%   x 1.16
 btree::set::clone_10k_and_remove_all   543,730     506,790          -36,940   -6.79%   x 1.07

so iter() wins from range(..) this time.

Mark-Simulacrum · 2020-05-05T12:18:04Z

@bors r+

bors · 2020-05-05T12:18:06Z

📌 Commit 8730227 has been approved by Mark-Simulacrum

@Mark-Simulacrum

…r=Mark-Simulacrum Btreemap iter intertwined 3 commits: 1. Introduced benchmarks for `BTreeMap::iter()`. Benchmarks named `iter_20` were of the whole iteration process, so I renamed them. Also the benchmarks of `range` that I wrote earlier weren't very good. I included an (awkwardly named) one that compares `iter()` to `range(..)` on the same set, because the contrast is surprising: ``` name ns/iter btree::map::range_unbounded_unbounded 28,176 btree::map::range_unbounded_vs_iter 89,369 ``` Both dig up the same pair of leaf edges. `range(..)` also checks that some keys are correctly ordered, the only thing `iter()` does more is to copy the map's length. 2. Slightly refactoring the code to what I find more readable (not in chronological order of discovery), boosts performance: ``` >cargo-benchcmp.exe benchcmp a1 a2 --threshold 5 name a1 ns/iter a2 ns/iter diff ns/iter diff % speedup btree::map::find_rand_100 18 17 -1 -5.56% x 1.06 btree::map::first_and_last_10k 64 71 7 10.94% x 0.90 btree::map::iter_0 2,939 2,209 -730 -24.84% x 1.33 btree::map::iter_1 6,845 2,696 -4,149 -60.61% x 2.54 btree::map::iter_100 8,556 3,672 -4,884 -57.08% x 2.33 btree::map::iter_10k 9,292 5,884 -3,408 -36.68% x 1.58 btree::map::iter_1m 10,268 6,510 -3,758 -36.60% x 1.58 btree::map::iteration_mut_100000 478,575 453,050 -25,525 -5.33% x 1.06 btree::map::range_unbounded_unbounded 28,176 36,169 7,993 28.37% x 0.78 btree::map::range_unbounded_vs_iter 89,369 38,290 -51,079 -57.16% x 2.33 btree::set::clone_100_and_remove_all 4,801 4,245 -556 -11.58% x 1.13 btree::set::clone_10k_and_remove_all 529,450 496,030 -33,420 -6.31% x 1.07 ``` But you can tell from the `range_unbounded_*` lines that, despite an unwarranted, vengeful attack on the range_unbounded_unbounded benchmark, this change still doesn't allow `iter()` to catch up with `range(..)`. 3. I guess that `range(..)` copes so well because it intertwines the leftmost and rightmost descend towards leaf edges, doing the two root node accesses close together, perhaps exploiting a CPU's internal pipelining? So the third commit distils a version of `range_search` (which we can't use directly because of the `Ord` bound), and we get another boost: ``` cargo-benchcmp.exe benchcmp a2 a3 --threshold 5 name a2 ns/iter a3 ns/iter diff ns/iter diff % speedup btree::map::first_and_last_100 40 43 3 7.50% x 0.93 btree::map::first_and_last_10k 71 64 -7 -9.86% x 1.11 btree::map::iter_0 2,209 1,719 -490 -22.18% x 1.29 btree::map::iter_1 2,696 2,205 -491 -18.21% x 1.22 btree::map::iter_100 3,672 2,943 -729 -19.85% x 1.25 btree::map::iter_10k 5,884 3,929 -1,955 -33.23% x 1.50 btree::map::iter_1m 6,510 5,532 -978 -15.02% x 1.18 btree::map::iteration_mut_100000 453,050 476,667 23,617 5.21% x 0.95 btree::map::range_included_excluded 405,075 371,297 -33,778 -8.34% x 1.09 btree::map::range_included_included 427,577 397,440 -30,137 -7.05% x 1.08 btree::map::range_unbounded_unbounded 36,169 28,175 -7,994 -22.10% x 1.28 btree::map::range_unbounded_vs_iter 38,290 30,838 -7,452 -19.46% x 1.24 ``` But I think this is just fake news from the microbenchmarking media. `iter()` is still trying to catch up with `range(..)`. And we can sure do without another function. So I would skip this 3rd commit. r? @Mark-Simulacrum

@Mark-Simulacrum

…r=Mark-Simulacrum Btreemap iter intertwined 3 commits: 1. Introduced benchmarks for `BTreeMap::iter()`. Benchmarks named `iter_20` were of the whole iteration process, so I renamed them. Also the benchmarks of `range` that I wrote earlier weren't very good. I included an (awkwardly named) one that compares `iter()` to `range(..)` on the same set, because the contrast is surprising: ``` name ns/iter btree::map::range_unbounded_unbounded 28,176 btree::map::range_unbounded_vs_iter 89,369 ``` Both dig up the same pair of leaf edges. `range(..)` also checks that some keys are correctly ordered, the only thing `iter()` does more is to copy the map's length. 2. Slightly refactoring the code to what I find more readable (not in chronological order of discovery), boosts performance: ``` >cargo-benchcmp.exe benchcmp a1 a2 --threshold 5 name a1 ns/iter a2 ns/iter diff ns/iter diff % speedup btree::map::find_rand_100 18 17 -1 -5.56% x 1.06 btree::map::first_and_last_10k 64 71 7 10.94% x 0.90 btree::map::iter_0 2,939 2,209 -730 -24.84% x 1.33 btree::map::iter_1 6,845 2,696 -4,149 -60.61% x 2.54 btree::map::iter_100 8,556 3,672 -4,884 -57.08% x 2.33 btree::map::iter_10k 9,292 5,884 -3,408 -36.68% x 1.58 btree::map::iter_1m 10,268 6,510 -3,758 -36.60% x 1.58 btree::map::iteration_mut_100000 478,575 453,050 -25,525 -5.33% x 1.06 btree::map::range_unbounded_unbounded 28,176 36,169 7,993 28.37% x 0.78 btree::map::range_unbounded_vs_iter 89,369 38,290 -51,079 -57.16% x 2.33 btree::set::clone_100_and_remove_all 4,801 4,245 -556 -11.58% x 1.13 btree::set::clone_10k_and_remove_all 529,450 496,030 -33,420 -6.31% x 1.07 ``` But you can tell from the `range_unbounded_*` lines that, despite an unwarranted, vengeful attack on the range_unbounded_unbounded benchmark, this change still doesn't allow `iter()` to catch up with `range(..)`. 3. I guess that `range(..)` copes so well because it intertwines the leftmost and rightmost descend towards leaf edges, doing the two root node accesses close together, perhaps exploiting a CPU's internal pipelining? So the third commit distils a version of `range_search` (which we can't use directly because of the `Ord` bound), and we get another boost: ``` cargo-benchcmp.exe benchcmp a2 a3 --threshold 5 name a2 ns/iter a3 ns/iter diff ns/iter diff % speedup btree::map::first_and_last_100 40 43 3 7.50% x 0.93 btree::map::first_and_last_10k 71 64 -7 -9.86% x 1.11 btree::map::iter_0 2,209 1,719 -490 -22.18% x 1.29 btree::map::iter_1 2,696 2,205 -491 -18.21% x 1.22 btree::map::iter_100 3,672 2,943 -729 -19.85% x 1.25 btree::map::iter_10k 5,884 3,929 -1,955 -33.23% x 1.50 btree::map::iter_1m 6,510 5,532 -978 -15.02% x 1.18 btree::map::iteration_mut_100000 453,050 476,667 23,617 5.21% x 0.95 btree::map::range_included_excluded 405,075 371,297 -33,778 -8.34% x 1.09 btree::map::range_included_included 427,577 397,440 -30,137 -7.05% x 1.08 btree::map::range_unbounded_unbounded 36,169 28,175 -7,994 -22.10% x 1.28 btree::map::range_unbounded_vs_iter 38,290 30,838 -7,452 -19.46% x 1.24 ``` But I think this is just fake news from the microbenchmarking media. `iter()` is still trying to catch up with `range(..)`. And we can sure do without another function. So I would skip this 3rd commit. r? @Mark-Simulacrum

@Mark-Simulacrum

…r=Mark-Simulacrum Btreemap iter intertwined 3 commits: 1. Introduced benchmarks for `BTreeMap::iter()`. Benchmarks named `iter_20` were of the whole iteration process, so I renamed them. Also the benchmarks of `range` that I wrote earlier weren't very good. I included an (awkwardly named) one that compares `iter()` to `range(..)` on the same set, because the contrast is surprising: ``` name ns/iter btree::map::range_unbounded_unbounded 28,176 btree::map::range_unbounded_vs_iter 89,369 ``` Both dig up the same pair of leaf edges. `range(..)` also checks that some keys are correctly ordered, the only thing `iter()` does more is to copy the map's length. 2. Slightly refactoring the code to what I find more readable (not in chronological order of discovery), boosts performance: ``` >cargo-benchcmp.exe benchcmp a1 a2 --threshold 5 name a1 ns/iter a2 ns/iter diff ns/iter diff % speedup btree::map::find_rand_100 18 17 -1 -5.56% x 1.06 btree::map::first_and_last_10k 64 71 7 10.94% x 0.90 btree::map::iter_0 2,939 2,209 -730 -24.84% x 1.33 btree::map::iter_1 6,845 2,696 -4,149 -60.61% x 2.54 btree::map::iter_100 8,556 3,672 -4,884 -57.08% x 2.33 btree::map::iter_10k 9,292 5,884 -3,408 -36.68% x 1.58 btree::map::iter_1m 10,268 6,510 -3,758 -36.60% x 1.58 btree::map::iteration_mut_100000 478,575 453,050 -25,525 -5.33% x 1.06 btree::map::range_unbounded_unbounded 28,176 36,169 7,993 28.37% x 0.78 btree::map::range_unbounded_vs_iter 89,369 38,290 -51,079 -57.16% x 2.33 btree::set::clone_100_and_remove_all 4,801 4,245 -556 -11.58% x 1.13 btree::set::clone_10k_and_remove_all 529,450 496,030 -33,420 -6.31% x 1.07 ``` But you can tell from the `range_unbounded_*` lines that, despite an unwarranted, vengeful attack on the range_unbounded_unbounded benchmark, this change still doesn't allow `iter()` to catch up with `range(..)`. 3. I guess that `range(..)` copes so well because it intertwines the leftmost and rightmost descend towards leaf edges, doing the two root node accesses close together, perhaps exploiting a CPU's internal pipelining? So the third commit distils a version of `range_search` (which we can't use directly because of the `Ord` bound), and we get another boost: ``` cargo-benchcmp.exe benchcmp a2 a3 --threshold 5 name a2 ns/iter a3 ns/iter diff ns/iter diff % speedup btree::map::first_and_last_100 40 43 3 7.50% x 0.93 btree::map::first_and_last_10k 71 64 -7 -9.86% x 1.11 btree::map::iter_0 2,209 1,719 -490 -22.18% x 1.29 btree::map::iter_1 2,696 2,205 -491 -18.21% x 1.22 btree::map::iter_100 3,672 2,943 -729 -19.85% x 1.25 btree::map::iter_10k 5,884 3,929 -1,955 -33.23% x 1.50 btree::map::iter_1m 6,510 5,532 -978 -15.02% x 1.18 btree::map::iteration_mut_100000 453,050 476,667 23,617 5.21% x 0.95 btree::map::range_included_excluded 405,075 371,297 -33,778 -8.34% x 1.09 btree::map::range_included_included 427,577 397,440 -30,137 -7.05% x 1.08 btree::map::range_unbounded_unbounded 36,169 28,175 -7,994 -22.10% x 1.28 btree::map::range_unbounded_vs_iter 38,290 30,838 -7,452 -19.46% x 1.24 ``` But I think this is just fake news from the microbenchmarking media. `iter()` is still trying to catch up with `range(..)`. And we can sure do without another function. So I would skip this 3rd commit. r? @Mark-Simulacrum

@Mark-Simulacrum

…r=Mark-Simulacrum Btreemap iter intertwined 3 commits: 1. Introduced benchmarks for `BTreeMap::iter()`. Benchmarks named `iter_20` were of the whole iteration process, so I renamed them. Also the benchmarks of `range` that I wrote earlier weren't very good. I included an (awkwardly named) one that compares `iter()` to `range(..)` on the same set, because the contrast is surprising: ``` name ns/iter btree::map::range_unbounded_unbounded 28,176 btree::map::range_unbounded_vs_iter 89,369 ``` Both dig up the same pair of leaf edges. `range(..)` also checks that some keys are correctly ordered, the only thing `iter()` does more is to copy the map's length. 2. Slightly refactoring the code to what I find more readable (not in chronological order of discovery), boosts performance: ``` >cargo-benchcmp.exe benchcmp a1 a2 --threshold 5 name a1 ns/iter a2 ns/iter diff ns/iter diff % speedup btree::map::find_rand_100 18 17 -1 -5.56% x 1.06 btree::map::first_and_last_10k 64 71 7 10.94% x 0.90 btree::map::iter_0 2,939 2,209 -730 -24.84% x 1.33 btree::map::iter_1 6,845 2,696 -4,149 -60.61% x 2.54 btree::map::iter_100 8,556 3,672 -4,884 -57.08% x 2.33 btree::map::iter_10k 9,292 5,884 -3,408 -36.68% x 1.58 btree::map::iter_1m 10,268 6,510 -3,758 -36.60% x 1.58 btree::map::iteration_mut_100000 478,575 453,050 -25,525 -5.33% x 1.06 btree::map::range_unbounded_unbounded 28,176 36,169 7,993 28.37% x 0.78 btree::map::range_unbounded_vs_iter 89,369 38,290 -51,079 -57.16% x 2.33 btree::set::clone_100_and_remove_all 4,801 4,245 -556 -11.58% x 1.13 btree::set::clone_10k_and_remove_all 529,450 496,030 -33,420 -6.31% x 1.07 ``` But you can tell from the `range_unbounded_*` lines that, despite an unwarranted, vengeful attack on the range_unbounded_unbounded benchmark, this change still doesn't allow `iter()` to catch up with `range(..)`. 3. I guess that `range(..)` copes so well because it intertwines the leftmost and rightmost descend towards leaf edges, doing the two root node accesses close together, perhaps exploiting a CPU's internal pipelining? So the third commit distils a version of `range_search` (which we can't use directly because of the `Ord` bound), and we get another boost: ``` cargo-benchcmp.exe benchcmp a2 a3 --threshold 5 name a2 ns/iter a3 ns/iter diff ns/iter diff % speedup btree::map::first_and_last_100 40 43 3 7.50% x 0.93 btree::map::first_and_last_10k 71 64 -7 -9.86% x 1.11 btree::map::iter_0 2,209 1,719 -490 -22.18% x 1.29 btree::map::iter_1 2,696 2,205 -491 -18.21% x 1.22 btree::map::iter_100 3,672 2,943 -729 -19.85% x 1.25 btree::map::iter_10k 5,884 3,929 -1,955 -33.23% x 1.50 btree::map::iter_1m 6,510 5,532 -978 -15.02% x 1.18 btree::map::iteration_mut_100000 453,050 476,667 23,617 5.21% x 0.95 btree::map::range_included_excluded 405,075 371,297 -33,778 -8.34% x 1.09 btree::map::range_included_included 427,577 397,440 -30,137 -7.05% x 1.08 btree::map::range_unbounded_unbounded 36,169 28,175 -7,994 -22.10% x 1.28 btree::map::range_unbounded_vs_iter 38,290 30,838 -7,452 -19.46% x 1.24 ``` But I think this is just fake news from the microbenchmarking media. `iter()` is still trying to catch up with `range(..)`. And we can sure do without another function. So I would skip this 3rd commit. r? @Mark-Simulacrum

@ghost

Rollup of 6 pull requests Successful merges: - rust-lang#71510 (Btreemap iter intertwined) - rust-lang#71727 (SipHasher with keys initialized to 0 should just use new()) - rust-lang#71889 (Explain our RwLock implementation) - rust-lang#71905 (Add command aliases from Cargo to x.py commands) - rust-lang#71914 (Backport 1.43.1 release notes to master) - rust-lang#71921 (explain the types used in the open64 call) Failed merges: r? @ghost

rust-highfive assigned Mark-Simulacrum Apr 24, 2020

rust-highfive added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Apr 24, 2020

Mark-Simulacrum approved these changes Apr 24, 2020

View reviewed changes

ssomers added 2 commits April 25, 2020 00:05

Introduce BTreeMap benches of iter itself

8f5c66c

Speed up BTreeMap iteration by intertwined descend to the initial lea…

8730227

…f edges

ssomers force-pushed the btreemap_iter_intertwined branch from 92d2969 to 8730227 Compare April 24, 2020 22:51

ssomers marked this pull request as ready for review April 24, 2020 22:55

joelpalmer added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels May 5, 2020

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels May 5, 2020

Dylan-DPC-zz mentioned this pull request May 5, 2020

Rollup of 7 pull requests #71920

Closed

Dylan-DPC-zz mentioned this pull request May 5, 2020

Rollup of 6 pull requests #71927

Closed

Dylan-DPC-zz mentioned this pull request May 5, 2020

Rollup of 8 pull requests #71933

Closed

Dylan-DPC-zz mentioned this pull request May 5, 2020

Rollup of 6 pull requests #71934

Closed

Dylan-DPC-zz mentioned this pull request May 6, 2020

Rollup of 6 pull requests #71949

Merged

bors merged commit 78a25cb into rust-lang:master May 6, 2020

ssomers deleted the btreemap_iter_intertwined branch May 6, 2020 16:15

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Btreemap iter intertwined #71510

Btreemap iter intertwined #71510

ssomers commented Apr 24, 2020 •

edited

Loading

Mark-Simulacrum left a comment

Mark-Simulacrum commented Apr 24, 2020

ssomers commented Apr 24, 2020

ssomers commented Apr 24, 2020 •

edited

Loading

Mark-Simulacrum commented May 5, 2020

bors commented May 5, 2020

Btreemap iter intertwined #71510

Btreemap iter intertwined #71510

Conversation

ssomers commented Apr 24, 2020 • edited Loading

Mark-Simulacrum left a comment

Choose a reason for hiding this comment

Mark-Simulacrum commented Apr 24, 2020

ssomers commented Apr 24, 2020

ssomers commented Apr 24, 2020 • edited Loading

Mark-Simulacrum commented May 5, 2020

bors commented May 5, 2020

ssomers commented Apr 24, 2020 •

edited

Loading

ssomers commented Apr 24, 2020 •

edited

Loading