RE: [PATCH v2 1/3] arch/arm64: Introduce a capability to tell whether 16-bit VMID is available

[Shameerali Kolothum Thodi]

> -----Original Message-----
> From: Will Deacon [mailto:will@xxxxxxxxxx]
> Sent: 21 July 2021 16:23
> To: Shameerali Kolothum Thodi <shameerali.kolothum.thodi@xxxxxxxxxx>
> Cc: linux-arm-kernel@xxxxxxxxxxxxxxxxxxx; kvmarm@xxxxxxxxxxxxxxxxxxxxx;
> linux-kernel@xxxxxxxxxxxxxxx; maz@xxxxxxxxxx; catalin.marinas@xxxxxxx;
> james.morse@xxxxxxx; julien.thierry.kdev@xxxxxxxxx;
> suzuki.poulose@xxxxxxx; jean-philippe@xxxxxxxxxx;
> Alexandru.Elisei@xxxxxxx; Linuxarm <linuxarm@xxxxxxxxxx>
> Subject: Re: [PATCH v2 1/3] arch/arm64: Introduce a capability to tell whether
> 16-bit VMID is available
> 
> On Wed, Jun 16, 2021 at 04:56:04PM +0100, Shameer Kolothum wrote:
> > From: Julien Grall <julien.grall@xxxxxxx>
> >
> > At the moment, the function kvm_get_vmid_bits() is looking up for the
> > sanitized value of ID_AA64MMFR1_EL1 and extract the information
> > regarding the number of VMID bits supported.
> >
> > This is fine as the function is mainly used during VMID roll-over. New
> > use in a follow-up patch will require the function to be called a every
> > context switch so we want the function to be more efficient.
> >
> > A new capability is introduced to tell whether 16-bit VMID is
> > available.
> 
> I don't really buy this rationale. The VMID allocator introduced later on
> caches this value in the static 'vmid_bits' variable, and that gets used
> on vCPU enter via vmid_gen_match() in the kvm_arm_update_vmid() fastpath.
> 
> So I would prefer that we just expose an accessor for that than introduce
> a static key and new cpufeature just for kvm_get_vttbr().

Ok. Will change it to an accessor.

Thanks,
Shameer