Re: [PATCH 1/1] x86: replace cpu_up hard-coded mdelay with variable

[Ingo Molnar]

* Brown, Len <len.brown@xxxxxxxxx> wrote:

> > What's the cutoff for 'modern hardware' - which CPUs stopped requiring
> > the delay?
> 
> This is the topic of ongoing research, and I'm not ready to send
> the patch setting a new default until I've heard back from a few more HW people.
> 
> Every system I've tested appears to work with delay 0.
> Were I to guess, I'd venture that every
> system that runs an X86_64 kernel might count as "modern" -- even
> the 2005 AMD Turion laptop I've got in the bone pile.

Could we use the apic version as a cutoff perhaps?

It would be nice to 'automatically' include modern 32-bit x86 systems 
as well.

Any failure here would be relatively easy to bisect to, so we might as 
well guess a bit and refine the quirk condition if needed?

Thanks,

	Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/