tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Silly question - a further one : TLB flush

> Simplicity in the design of the chip (and so in silicium surface)

This, IMHO, has never been an Intel goal (Motorola always had better orthogonal design, starting with the 68000).

> You have to have a 64bit tlb anyway for support of the 64bit mode;
> so in 32bit mode it's much simpler, from a electronic design POW, to
> just set the 32 upper bits to 0 on data path entry of the TBL than
> to have internals of the TLB work in a different way (basically requiring
> every TLB registers to be dual-mode, as well as the lookup and eviction
> logics). This would be a quite high cost on the silicium, maybe near of

Not really. This is only a matter of multiplexing a32-63 according to a mode bit. Example for the comparator :

TLB (Low) ------ CMP ---> result(1)
A0 - A31  --------+
A32- A63  ------>MUX <---- 32/64 BIT-MODE
TLB (High) ----->CMP ---> result (2)

> doubing the on-die size of the TLB. This would also be a cost in timing, as
> you're adding extra gates in the critical path.

True, there is one extra stage, but I don't think it really matters much.

Well, Manuel, on peut poursuivre par mail, or on the regional list, because I don't think this is kernel relevant anyhow ;)


Home | Main Index | Thread Index | Old Index