diff options
Diffstat (limited to 'Doc')
| -rw-r--r-- | Doc/lib/libprofile.tex | 84 |
1 files changed, 32 insertions, 52 deletions
diff --git a/Doc/lib/libprofile.tex b/Doc/lib/libprofile.tex index 9e4f91b..a1634b9 100644 --- a/Doc/lib/libprofile.tex +++ b/Doc/lib/libprofile.tex @@ -553,75 +553,55 @@ calibration. \section{Calibration \label{profile-calibration}} -The profiler class has a hard coded constant that is added to each +The profiler subtracts a constant from each event handling time to compensate for the overhead of calling the time -function, and socking away the results. The following procedure can -be used to obtain this constant for a given platform (see discussion +function, and socking away the results. By default, the constant is 0. +The following procedure can +be used to obtain a better constant for a given platform (see discussion in section Limitations above). \begin{verbatim} import profile pr = profile.Profile() -print pr.calibrate(100) -print pr.calibrate(100) -print pr.calibrate(100) +for i in range(5): + print pr.calibrate(10000) \end{verbatim} -The argument to \method{calibrate()} is the number of times to try to -do the sample calls to get the CPU times. If your computer is -\emph{very} fast, you might have to do: - -\begin{verbatim} -pr.calibrate(1000) -\end{verbatim} - -or even: - -\begin{verbatim} -pr.calibrate(10000) -\end{verbatim} +The method executes the number of Python calls given by the argument, +directly and again under the profiler, measuring the time for both. +It then computes the hidden overhead per profiler event, and returns +that as a float. For example, on an 800 MHz Pentium running +Windows 2000, and using Python's time.clock() as the timer, +the magical number is about 12.5e-6. The object of this exercise is to get a fairly consistent result. -When you have a consistent answer, you are ready to use that number in -the source code. For a Sun Sparcstation 1000 running Solaris 2.3, the -magical number is about .00053. If you have a choice, you are better -off with a smaller constant, and your results will ``less often'' show -up as negative in profile statistics. +If your computer is \emph{very} fast, or your timer function has poor +resolution, you might have to pass 100000, or even 1000000, to get +consistent results. -The following shows how the trace_dispatch() method in the Profile -class should be modified to install the calibration constant on a Sun -Sparcstation 1000: +When you have a consistent answer, +there are three ways you can use it:\footnote{Prior to Python 2.2, it + was necessary to edit the profiler source code to embed the bias as + a literal number. You still can, but that method is no longer + described, because no longer needed.} \begin{verbatim} -def trace_dispatch(self, frame, event, arg): - t = self.timer() - t = t[0] + t[1] - self.t - .00053 # Calibration constant - - if self.dispatch[event](frame,t): - t = self.timer() - self.t = t[0] + t[1] - else: - r = self.timer() - self.t = r[0] + r[1] - t # put back unrecorded delta - return -\end{verbatim} +import profile -Note that if there is no calibration constant, then the line -containing the callibration constant should simply say: +# 1. Apply computed bias to all Profile instances created hereafter. +profile.Profile.bias = -\begin{verbatim} -t = t[0] + t[1] - self.t # no calibration constant +# 2. Apply computed bias to a specific Profile instance. +pr = profile.Profile() +pr.bias = your_computed_bias + +# 3. Specify computed bias in instance constructor. +pr = profile.Profile(bias=your_computed_bias) \end{verbatim} -You can also achieve the same results using a derived class (and the -profiler will actually run equally fast!!), but the above method is -the simplest to use. I could have made the profiler ``self -calibrating,'' but it would have made the initialization of the -profiler class slower, and would have required some \emph{very} fancy -coding, or else the use of a variable where the constant \samp{.00053} -was placed in the code shown. This is a \strong{VERY} critical -performance section, and there is no reason to use a variable lookup -at this point, when a constant can be used. +If you have a choice, you are better off choosing a smaller constant, and +then your results will ``less often'' show up as negative in profile +statistics. \section{Extensions --- Deriving Better Profilers} |