This is the mail archive of the binutils@sourceware.org mailing list for the binutils project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: gprof reports wrong execution times

From: Nick Clifton <nickc at redhat dot com>
To: Anton Shepelev <anton dot txt at gmail dot com>, binutils at sourceware dot org
Date: Thu, 15 Sep 2016 11:06:22 +0100
Subject: Re: gprof reports wrong execution times
Authentication-results: sourceware.org; auth=none
References: <20160914210849.46dc656fb1a8eb60a0ebc8ea@gmail.com> <20160915125741.452f9fdb95e2cb982f0b1e76@gmail.com>

Hi Anton,

> This  is  the correct and expected behavior, because
> the man page says:
> 
>    The granularity of the sampling is  shown,  but  re-
>    mains  statistical at best.  We assume that the time
>    for each execution of a function can be expressed by
>    the  total time for the function divided by the num-
>    ber of times the function is called.  Thus the  time
>    propagated  along  the  call graph arcs to the func-
>    tion's parents is directly proportional to the  num-
>    ber of times that arc is traversed.

Right.  See also the "Inaccuracy of gprof Output" chapter in the gprof documentation.

Cheers
  Nick

References:
- gprof reports wrong execution times
  - From: Anton Shepelev
- Re: gprof reports wrong execution times
  - From: Anton Shepelev

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]