Fix description of mutex acquisition.

Reviewed by:	kib
X-MFC with:	r272070
Sponsored by:	Nginx, Inc.
This commit is contained in:
Sergey Kandaurov 2014-09-26 04:33:27 +00:00
parent c1cd488caf
commit 663222b9f6
Notes: svn2git 2020-12-20 02:59:44 +00:00
svn path=/head/; revision=272153

View File

@ -29,7 +29,7 @@
.\"
.\" $FreeBSD$
.\"
.Dd September 20, 2014
.Dd September 26, 2014
.Dt LIBTHR 3
.Os
.Sh NAME
@ -85,8 +85,14 @@ owning the lock.
.Nm
performs a contested mutex acquisition in three stages, each of which
is more resource-consuming than the previous.
The first two stages are only applied for a mutex of
.Dv PTHREAD_MUTEX_ADAPTIVE_NP
type and
.Dv PTHREAD_PRIO_NONE
protocol (see
.Xr pthread_mutexattr 3 ) .
.Pp
First, a spin loop
First, on SMP systems, a spin loop
is performed, where the library attempts to acquire the lock by
.Xr atomic 9
operations.