On 08/07/2012 05:57 AM, Alexander Graf wrote:
> When we map a page that wasn't icache cleared before, do so when first
> mapping it in KVM using the same information bits as the Linux mapping
> logic. That way we are 100% sure that any page we map does not have stale
> entries in the icache.
>
> What we really would need is a method to flush the icache only when we
> actually map a page executable for the guest. But with the current
> abstraction that is hard to do, and this way we don't have a huge performance
> penalty, so better be safe now and micro optimize things later.
>
> Signed-off-by: Alexander Graf <ag...@suse.de>
> ---
> arch/powerpc/include/asm/kvm_host.h | 10 ++++++++++
> arch/powerpc/mm/mem.c | 1 +
> 2 files changed, 11 insertions(+), 0 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/kvm_host.h
> b/arch/powerpc/include/asm/kvm_host.h
> index ed75bc9..c0a2fc1 100644
> --- a/arch/powerpc/include/asm/kvm_host.h
> +++ b/arch/powerpc/include/asm/kvm_host.h
> @@ -33,6 +33,7 @@
> #include <asm/kvm_asm.h>
> #include <asm/processor.h>
> #include <asm/page.h>
> +#include <asm/cacheflush.h>
>
> #define KVM_MAX_VCPUS NR_CPUS
> #define KVM_MAX_VCORES NR_CPUS
> @@ -550,5 +551,14 @@ struct kvm_vcpu_arch {
> #define KVM_MMIO_REG_FQPR 0x0060
>
> #define __KVM_HAVE_ARCH_WQP
> +#define __KVM_HAVE_ARCH_MAP_PAGE
> +static inline void kvm_arch_map_page(struct page *page)
> +{
> + /* Need to invalidate the icache for new pages */
> + if (!test_bit(PG_arch_1, &page->flags)) {
> + flush_dcache_icache_page(page);
> + set_bit(PG_arch_1, &page->flags);
> + }
> +}
Shouldn't this test CPU_FTR_COHERENT_ICACHE?
-Scott
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
