Ray
Tue Oct 12 15:21:21 CDT 2004
Depends on your definition of "lie". If you're trying to get accurate
absolute time, then yes, it's a lie (but then it never even vaguely
pretended to be that). If you're trying to measure *performance*, then
"KeQueryPerformanceCounter" isn't lying at all.
Pavel A. wrote:
> Maxim noted an interesting detail...
> If KeQueryPerformanceCounter is based on rdtsc, will it lie on
> CPU with variable clock rate (Pentium Mobile etc), or will it correct the
> time ?
>
> --PA
>
> "Don Burn" wrote:
>
>
>>NO, NO, NO, NOOOOOOOOOOO!!!!
>>
>>Don't use assembler in the kernel, it won't work with 64-bit and in general
>>is a VERY STUPID THING TO DO!. Use KeQueryPerformanceCounter, this will use
>>the RDTSC if available and it is portable.
>>
>>
>>--
>>Don Burn (MVP, Windows DDK)
>>Windows 2k/XP/2k3 Filesystem and Driver Consulting
>>Remove StopSpam from the email to reply
>>
>>
>>"Maxim S. Shatskih" <maxim@storagecraft.com> wrote in message
>>news:u2nXUVEsEHA.2732@TK2MSFTNGP09.phx.gbl...
>>
>>> Try RDTSC, and let's hope that it will not lie due to CPU power
>>
>>management
>>
>>>schemes. At least the APIC timer lies grossly due to this.
>>>
>>>--
>>>Maxim Shatskih, Windows DDK MVP
>>>StorageCraft Corporation
>>>maxim@storagecraft.com
>>>
http://www.storagecraft.com
>>>
>>>"Michael Andreas" <dachrinne@t-online.de> wrote in message
>>>news:ckfvs4$2sj$01$1@news.t-online.com...
>>>
>>>>Hi,
>>>>
>>>>I am seeking for a driver with a
>>>>high resolution timestamp (nanosecond).
>>>>Who can tell me an example or links?
>>>>
>>>>Thanks michael
>>>>
>>>>
>>>
>>>
>>
>>
--
../ray\..
Please remove ".spamblock" from my email address if you need to contact
me outside the newsgroup.