runtime: add per-p page allocation cache

This change adds a per-p free page cache which the page allocator may
allocate out of without a lock. The change also introduces a completely
lockless page allocator fast path.

Although the cache contains at most 64 pages (and usually less), the
vast majority (85%+) of page allocations are exactly 1 page in size.

Updates #35112.

Change-Id: I170bf0a9375873e7e3230845eb1df7e5cf741b78
Reviewed-on: https://go-review.googlesource.com/c/go/+/195701
Run-TryBot: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Austin Clements <austin@google.com>

src/runtime/export_test.go

@@ -7,6 +7,7 @@
 package runtime
 
 import (
+	"math/bits"
 	"runtime/internal/atomic"
 	"runtime/internal/sys"
 	"unsafe"
@@ -358,6 +359,10 @@ func ReadMemStatsSlow() (base, slow MemStats) {
 			pg := mheap_.pages.chunks[i].scavenged.popcntRange(0, pallocChunkPages)
 			slow.HeapReleased += uint64(pg) * pageSize
 		}
+		for _, p := range allp {
+			pg := bits.OnesCount64(p.pcache.scav)
+			slow.HeapReleased += uint64(pg) * pageSize
+		}
 
 		// Unused space in the current arena also counts as released space.
 		slow.HeapReleased += uint64(mheap_.curArena.end - mheap_.curArena.base)
@@ -879,3 +884,20 @@ func CheckScavengedBitsCleared(mismatches []BitsMismatch) (n int, ok bool) {
 	})
 	return
 }
+
+func PageCachePagesLeaked() (leaked uintptr) {
+	stopTheWorld("PageCachePagesLeaked")
+
+	// Walk over destroyed Ps and look for unflushed caches.
+	deadp := allp[len(allp):cap(allp)]
+	for _, p := range deadp {
+		// Since we're going past len(allp) we may see nil Ps.
+		// Just ignore them.
+		if p != nil {
+			leaked += uintptr(bits.OnesCount64(p.pcache.cache))
+		}
+	}
+
+	startTheWorld()
+	return
+}