diff --git a/share/man/man9/LOCK_PROFILING.9 b/share/man/man9/LOCK_PROFILING.9 index c8a3c419652a..23aba541d3ff 100644 --- a/share/man/man9/LOCK_PROFILING.9 +++ b/share/man/man9/LOCK_PROFILING.9 @@ -67,7 +67,8 @@ The total number of times another thread tried to acquire the lock while it was held after having been acquired at this point. .El .Pp -In addition, the average hold time and average wait time are derived from the total hold time +In addition, the average hold time and average wait time are derived +from the total hold time and total wait time respectively and the number of acquisitions. .Pp The @@ -143,18 +144,18 @@ and line number, followed by the name of the lock in parentheses. Mutex profiling support appeared in .Fx 5.0 . Generalized lock profiling support appeared in -.Fx 5.0 . +.Fx 7.0 . .Sh AUTHORS .An -nosplit The -.Nm MUTEX_PROFILING +.Nm MUTEX_PROFILING code was written by .An Eivind Eklund Aq eivind@FreeBSD.org , .An Dag-Erling Sm\(/orgrav Aq des@FreeBSD.org and -.An Robert Watson Aq rwatson@FreeBSD.org . +.An Robert Watson Aq rwatson@FreeBSD.org . The -.Nm LOCK_PROFILING +.Nm code was written by .An Kip Macy Aq kmacy@FreeBSD.org . This manual page was written by @@ -170,7 +171,8 @@ without it. The .Dv LOCK_PROFILING option also prevents inlining of the mutex code, which can result in a -fairly severe performance penalty. This is, however, not always the case. +fairly severe performance penalty. +This is, however, not always the case. .Dv LOCK_PROFILING can introduce a substantial performance overhead that is easily monitorable using other profiling tools, so combining profiling tools