From: Lai Jiangshan <laijs@cn.fujitsu.com>
Date: Mon, 17 Oct 2011 14:19:10 +0000 (-0400)
Subject: rculfhash: Simplify lookup_bucket()
X-Git-Tag: v0.7.0~43^2~90
X-Git-Url: https://git.lttng.org./?a=commitdiff_plain;h=a4ea2223fa701bd34b09fb3eede037908dbdd0c8;p=userspace-rcu.git

rculfhash: Simplify lookup_bucket()

They are the same, but I don't think the compiler can optimize it.

And it also helps for understanding the following code.

[ Edit by Mathieu Desnoyers:
  - Add a comment that describes the equivalence between get_count_order
    and fls for this lookup of index + 1. ]

Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
---

diff --git a/rculfhash.c b/rculfhash.c
index d733d6b..189f8c8 100644
--- a/rculfhash.c
+++ b/rculfhash.c
@@ -724,7 +724,12 @@ struct _cds_lfht_node *lookup_bucket(struct cds_lfht *ht, unsigned long size,
 
 	assert(size > 0);
 	index = hash & (size - 1);
-	order = get_count_order_ulong(index + 1);
+	/*
+	 * equivalent to get_count_order_ulong(index + 1), but optimizes
+	 * away the non-existing 0 special-case for
+	 * get_count_order_ulong.
+	 */
+	order = fls_ulong(index);
 
 	dbg_printf("lookup hash %lu index %lu order %lu aridx %lu\n",
 		   hash, index, order, index & (!order ? 0 : ((1UL << (order - 1)) - 1)));