Re: Cyrix M2

bofh@snoopy.virtual.net.au
Fri, 25 Apr 97 15:26:01 +1100


>> > I remember a big discussion about MMX that happened not
>> > too long ago. The conclusion was that the penalty to switch
>> > between FP & MMX was too great for a true multitasking OS.
>[SNIPPED]
>>
>> (Cyrix had to use FP registers for MMX to be at all compatible
>> with Intel. I don't think they liked having to do it :-). Intel
>> did it to save the cost of engineering their silicon to hold a
>> few more registers.)
>>
>[SNIPPED]
>It isn't a cost issue! Sources at Cyrix say they had to have a way for an
>unknown operating system to be able to save and restore the new registers
>without having to rewrite all the operating systems that use
>Cytix/Pentiums/ix86/etc.

>It was presumed that, during a context switch FSAVE and FRSTOR will be
>executed to save/restore the state of the FP Unit. If executed, these
>instructions should automatically save and restore the MMX register(s)
>also. I don't know if they have to extend the buffer length for this
>operation. If so, the point is moot because the operating systems will
>have to be rewritten to provide this additional storage.

It wouldn't be difficult to add a feature to a new CPU which had to be
explicitely enabled from ring 0 before it could be used. Then if the OS
had allocated enough space for saving the registers it could enable the
feature. I expect that most OSs would quickly get patched to support a
feature such as MMX if it was added in that way.

Russell Coker