Re: [boost] Re: Re: Re: Re: Profiling Library suggestion

From: Martin Slater (mslater_at_[hidden])
Date: 2005-02-12 10:53:35

>>
>>> On Fri, 11 Feb 2005 19:30:36 -0500, christopher diggins wrote:
>>> [...]
>>> std::clock() and I assume QueryPerformanceCounter() give you an 
>>> amount of
>>> clock ticks.  This number is only incremented when the CPU works on 
>>> your
>>> task.
>>
>> AFAIK your assumption is wrong.
>
>
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/winui/winui/windowsuserinterface/windowing/timers/timerreference/timerfunctions/queryperformancecounter.asp

watch the line wrap. QueryPerformanceCounter uses the CPU clock and is 
not per task but system wide.

> Can you point me to documentation which supports your assertion?
>
>> The only problem with QueryPerformanceCounter() is that it can "jump"
>> on some chipsets.
>
>
> I believe this problem is perhaps not restricted to 
> QueryPerformanceCounter() but to any counter counting mechanism, as 
> the problem is at the chipset level not the API level.

No, there is a very particualar problem relating to 
QueryPerformanceCounter() on some (old?) chipsets the can cause it go 
backwards in time and wrap after approximately 20 minutes. I reported 
this to microsoft some years ago and it was confirmed an issue. (at the 
time only happened to me on one particualar celeron / motherboard 
configuration). By backwards in time I mean you could call the counter 
in a loop and have a reported time earlier than one got the previous 
iteration. This bit me on commercial code and I resorted to
    timeBeginPeriod(1);
    timeGetTime();
instead with no issues.

Martin