Auto merge of #88075 - Xuanwo:vec_deque_retain, r=dtolnay

Optimize unnecessary check in VecDeque::retain

This pr is highly inspired by https://github.com/rust-lang/rust/pull/88060 which shared the same idea: we can split the `for` loop into stages so that we can remove unnecessary checks like `del > 0`.

## Benchmarks

Before

```rust
test collections::vec_deque::tests::bench_retain_half_10000  ... bench:     290,125 ns/iter (+/- 8,717)
test collections::vec_deque::tests::bench_retain_odd_10000   ... bench:     291,588 ns/iter (+/- 9,621)
test collections::vec_deque::tests::bench_retain_whole_10000 ... bench:     287,426 ns/iter (+/- 9,009)
```

After

```rust
test collections::vec_deque::tests::bench_retain_half_10000  ... bench:     243,940 ns/iter (+/- 8,563)
test collections::vec_deque::tests::bench_retain_odd_10000   ... bench:     242,768 ns/iter (+/- 3,903)
test collections::vec_deque::tests::bench_retain_whole_10000 ... bench:     202,926 ns/iter (+/- 6,332)
```

Based on the current benchmark, this PR will improve the perf of `VecDeque::retain` by around 16%. For special cases, the improvement will be up to 30%.
Signed-off-by: NXuanwo <github@xuanwo.io>

library/alloc/src/collections/vec_deque/mod.rs

@@ -2129,16 +2129,32 @@ pub fn retain<F>(&mut self, mut f: F)
         F: FnMut(&T) -> bool,
     {
         let len = self.len();
-        let mut del = 0;
-        for i in 0..len {
-            if !f(&self[i]) {
-                del += 1;
-            } else if del > 0 {
-                self.swap(i - del, i);
+        let mut idx = 0;
+        let mut cur = 0;
+
+        // Stage 1: All values are retained.
+        while cur < len {
+            if !f(&self[cur]) {
+                cur += 1;
+                break;
             }
+            cur += 1;
+            idx += 1;
         }
-        if del > 0 {
-            self.truncate(len - del);
+        // Stage 2: Swap retained value into current idx.
+        while cur < len {
+            if !f(&self[cur]) {
+                cur += 1;
+                continue;
+            }
+
+            self.swap(idx, cur);
+            cur += 1;
+            idx += 1;
+        }
+        // Stage 3: Trancate all values after idx.
+        if cur != idx {
+            self.truncate(idx);
         }
     }
 

library/alloc/src/collections/vec_deque/tests.rs

@@ -40,6 +40,39 @@ fn bench_pop_back_100(b: &mut test::Bencher) {
     })
 }
 
+#[bench]
+#[cfg_attr(miri, ignore)] // isolated Miri does not support benchmarks
+fn bench_retain_whole_10000(b: &mut test::Bencher) {
+    let v = (1..100000).collect::<VecDeque<u32>>();
+
+    b.iter(|| {
+        let mut v = v.clone();
+        v.retain(|x| *x > 0)
+    })
+}
+
+#[bench]
+#[cfg_attr(miri, ignore)] // isolated Miri does not support benchmarks
+fn bench_retain_odd_10000(b: &mut test::Bencher) {
+    let v = (1..100000).collect::<VecDeque<u32>>();
+
+    b.iter(|| {
+        let mut v = v.clone();
+        v.retain(|x| x & 1 == 0)
+    })
+}
+
+#[bench]
+#[cfg_attr(miri, ignore)] // isolated Miri does not support benchmarks
+fn bench_retain_half_10000(b: &mut test::Bencher) {
+    let v = (1..100000).collect::<VecDeque<u32>>();
+
+    b.iter(|| {
+        let mut v = v.clone();
+        v.retain(|x| *x > 50000)
+    })
+}
+
 #[bench]
 #[cfg_attr(miri, ignore)] // isolated Miri does not support benchmarks
 fn bench_pop_front_100(b: &mut test::Bencher) {