From: Chunsong Feng <fengchunsong@huawei.com>
Date: Tue, 29 Jun 2021 03:26:07 +0000 (+0000)
Subject: common: Use double instead of long double to improve performance
X-Git-Tag: v17.1.0~1523^2
X-Git-Url: http://git-server-git.apps.pok.os.sepia.ceph.com/?a=commitdiff_plain;h=58753cf50e89cc3d7169447aa83d13cd3a8c0347;p=ceph.git

common: Use double instead of long double to improve performance

To convert namoseconds to seconds, the precision needs to be 10,
and the precision of double is 15, which is enough to use.
On aarch64, double division uses the div instruction, while long
double uses the gcc buildin _divtf3, which has poor performance.
Therefore, use double instead of long double for better performance.

Signed-off-by: Chunsong Feng <fengchunsong@huawei.com>
---

diff --git a/src/include/utime.h b/src/include/utime.h
index 512149db03e..fad66af7939 100644
--- a/src/include/utime.h
+++ b/src/include/utime.h
@@ -224,7 +224,7 @@ public:
 
   // cast to double
   operator double() const {
-    return (double)sec() + ((double)nsec() / 1000000000.0L);
+    return (double)sec() + ((double)nsec() / 1000000000.0f);
   }
   operator ceph_timespec() const {
     ceph_timespec ts;