btree: Implement faster binary search algorithm

This implements a binary search algorithm for B-Trees that reduces
branching to the absolute minimum necessary for a binary search
algorithm. It also enables the compiler to inline the comparator to
ensure that the only slowdown when doing binary search is from waiting
for memory accesses. Additionally, it instructs the compiler to unroll
the loop, which gives an additional 40% improve with Clang and 8%
improvement with GCC.

Consumers must opt into using the faster algorithm. At present, only
B-Trees used inside kernel code have been modified to use the faster
algorithm.

Micro-benchmarks suggest that this can improve binary search performance
by up to 3.5 times when compiling with Clang 16 and up to 1.9 times when
compiling with GCC 12.2.

Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Richard Yao <richard.yao@alumni.stonybrook.edu>
Closes #14866

module/zfs/dsl_scan.c

@@ -4877,6 +4877,7 @@ scan_exec_io(dsl_pool_t *dp, const blkptr_t *bp, int zio_flags,
  * with single operation.  Plus it makes scrubs more sequential and reduces
  * chances that minor extent change move it within the B-tree.
  */
+__attribute__((always_inline)) inline
 static int
 ext_size_compare(const void *x, const void *y)
 {
@@ -4885,13 +4886,17 @@ ext_size_compare(const void *x, const void *y)
 	return (TREE_CMP(*a, *b));
 }
 
+ZFS_BTREE_FIND_IN_BUF_FUNC(ext_size_find_in_buf, uint64_t,
+    ext_size_compare)
+
 static void
 ext_size_create(range_tree_t *rt, void *arg)
 {
 	(void) rt;
 	zfs_btree_t *size_tree = arg;
 
-	zfs_btree_create(size_tree, ext_size_compare, sizeof (uint64_t));
+	zfs_btree_create(size_tree, ext_size_compare, ext_size_find_in_buf,
+	    sizeof (uint64_t));
 }
 
 static void