Re: Is the x86 clock stoppable?

Anthony J. Albert wrote On 09/23/05 14:42,:
> On 23 Sep 2005 11:45:05 -0700, andrewspencers@xxxxxxxxx wrote:
>
>
>>CJT wrote:
>>
>>>Homework?
>>
>>I haven't had homework for a very long time.
>>I vaguely remember being annoyed once upon a time about an unstoppable
>>6502. (Maybe because it used dram instead of sram for its registers, or
>>was there some other reason?) At the time I was told that CMOS
>>processors were stoppable, and that therefore maybe I'd want a 65C02.
>>Since AMD's and Intel's modern x86 processors are CMOS, and surely use
>>only sram registers, I assume that they're stoppable. But I wanted to
>>be sure.
>>I'm simply thinking about the possibility of a "pause" button on the
>>front of a computer, to be used instead of the power or reset button as
>>the emergency button of last resort to stop the operation of malicious
>>software, so that potentially valuable forensic evidence wouldn't be
>>wiped from the processor or from main memory.
>>That same idea was also the reason for my other posting in this
>>newsgroup at the same time (with subject "PC reset button resets
>>memory?").
>
>
> Despite the advantages of speed of static memory cells, for the
> registers, I wouldn't be surprised if there were at least some on-chip
> DRAM, especially in on-chip cache, because of the smaller size per
> cell.
>
> If that's the case, then I'd bet that there's a certain minimum speed
> below which you couldn't trust the processor.
>
> Anthony Albert

They can also have dynamic logic on the chip (horrors!), just because
it's CMOS does not mean no dynamic logic.

If you want a stop clock x86, then get one. Low power/embedded x86
CPUs will have specific methods to stop the clock, perhaps even a gating
line.

As another poster suggested the minimum clock frequency will be found
in the hardware manual for the CPU in question.

.

Loading