Re: [PATCH] x86, AMD: Correct F15h IC aliasing issue

[Andrew Morton]

On Sun, 24 Jul 2011 15:40:46 +0200 Borislav Petkov <bp@xxxxxxxxx> wrote:

> > > +	unalias_va_addr
> > > +			[X86-64] Unalias VA address by clearing slice [14:12]
> > > +			1: only on 32-bit
> > > +			2: only on 64-bit
> > > +			off: disabled on both 32 and 64-bit

How much performance difference does this actually make?

> > This says nothing about why it's cleared, what it does and why one 
> > would want to handle it differently on 32-bit and 64-bit.
> 
> This is there in case users care more about the 3bits ASLR granularity
> than invalidations amount, for example.

To make this decision the users will want to know how much their
computers will slow down.  We should tell them.
--
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/