On Wed, May 30, 2018 at 01:47:01PM +0100, Marc Zyngier wrote:
> Up to ARMv8.3, the combinaison of Stage-1 and Stage-2 attributes
> results in the strongest attribute of the two stages. This means
> that the hypervisor has to perform quite a lot of cache maintenance
> just in case the guest has some non-cacheable mappings around.
>
> ARMv8.4 solves this problem by offering a different mode (FWB) where
> Stage-2 has total control over the memory attribute (this is limited
> to systems where both I/O and instruction caches are coherent with
s/caches/fetches/ -- the I-caches themselves aren't coherent with the
D-caches (or we could omit I-cache maintenance).
i.e. this implies IDC, but not DIC.
> the dcache). This is achieved by having a different set of memory
> attributes in the page tables, and a new bit set in HCR_EL2.
>
> On such a system, we can then safely sidestep any form of dcache
> management.
>
> Acked-by: Catalin Marinas <catalin.marinas@xxxxxxx>
> Signed-off-by: Marc Zyngier <marc.zyngier@xxxxxxx>
> ---
> static inline bool vcpu_has_cache_enabled(struct kvm_vcpu *vcpu)
> {
> @@ -268,7 +269,10 @@ static inline void __clean_dcache_guest_page(kvm_pfn_t pfn, unsigned long size)
> {
> void *va = page_address(pfn_to_page(pfn));
>
> - kvm_flush_dcache_to_poc(va, size);
> + if (!cpus_have_const_cap(ARM64_HAS_STAGE2_FWB))
> + kvm_flush_dcache_to_poc(va, size);
> + else
> + kvm_flush_dcache_to_pou(va, size);
> }
Te commit message said instruction fetches were coherent, and that no
D-cache maintenance was necessary, so why do we need maintenance to the
PoU?
> +static void cpu_has_fwb(const struct arm64_cpu_capabilities *__unused)
> +{
> + u64 val = read_sysreg_s(SYS_CLIDR_EL1);
> +
> + /* Check that CLIDR_EL1.LOU{U,IS} are both 0 */
> + WARN_ON(val & (7 << 27 | 7 << 21));
> +}
What about CTR_EL0.IDC?
Thanks,
Mark.
