lkml.org 
[lkml]   [2023]   [Apr]   [3]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [PATCH v2 3/7] KVM: arm64: Implement __kvm_tlb_flush_range_vmid_ipa()
On Wed, Mar 29, 2023 at 5:59 PM Oliver Upton <oliver.upton@linux.dev> wrote:
>
> On Mon, Feb 06, 2023 at 05:23:36PM +0000, Raghavendra Rao Ananta wrote:
> > Define __kvm_tlb_flush_range_vmid_ipa() (for VHE and nVHE)
>
> bikeshed: Personally, I find that range implies it takes an address as an
> argument already. Maybe just call it __kvm_tlb_flush_vmid_range()
>
Hmm, since TLBI instructions takes-in a variety of ranges, VA or IPA,
I just thought of extending the '_ipa' to make things clear. Moreover
it aligns with the existing __kvm_tlb_flush_vmid_ipa(). WDYT?

Thank you.
Raghavendra

> > to flush a range of stage-2 page-tables using IPA in one go.
> > If the system supports FEAT_TLBIRANGE, the following patches
> > would conviniently replace global TLBI such as vmalls12e1is
> > in the map, unmap, and dirty-logging paths with ripas2e1is
> > instead.
> >
> > Signed-off-by: Raghavendra Rao Ananta <rananta@google.com>
> > ---
> > arch/arm64/include/asm/kvm_asm.h | 3 +++
> > arch/arm64/kvm/hyp/nvhe/hyp-main.c | 12 ++++++++++++
> > arch/arm64/kvm/hyp/nvhe/tlb.c | 28 ++++++++++++++++++++++++++++
> > arch/arm64/kvm/hyp/vhe/tlb.c | 24 ++++++++++++++++++++++++
> > 4 files changed, 67 insertions(+)
> >
> > diff --git a/arch/arm64/include/asm/kvm_asm.h b/arch/arm64/include/asm/kvm_asm.h
> > index 995ff048e8851..80a8ea85e84f8 100644
> > --- a/arch/arm64/include/asm/kvm_asm.h
> > +++ b/arch/arm64/include/asm/kvm_asm.h
> > @@ -79,6 +79,7 @@ enum __kvm_host_smccc_func {
> > __KVM_HOST_SMCCC_FUNC___pkvm_init_vm,
> > __KVM_HOST_SMCCC_FUNC___pkvm_init_vcpu,
> > __KVM_HOST_SMCCC_FUNC___pkvm_teardown_vm,
> > + __KVM_HOST_SMCCC_FUNC___kvm_tlb_flush_range_vmid_ipa,
> > };
> >
> > #define DECLARE_KVM_VHE_SYM(sym) extern char sym[]
> > @@ -243,6 +244,8 @@ extern void __kvm_flush_vm_context(void);
> > extern void __kvm_flush_cpu_context(struct kvm_s2_mmu *mmu);
> > extern void __kvm_tlb_flush_vmid_ipa(struct kvm_s2_mmu *mmu, phys_addr_t ipa,
> > int level);
> > +extern void __kvm_tlb_flush_range_vmid_ipa(struct kvm_s2_mmu *mmu, phys_addr_t start,
> > + phys_addr_t end, int level, int tlb_level);
> > extern void __kvm_tlb_flush_vmid(struct kvm_s2_mmu *mmu);
> >
> > extern void __kvm_timer_set_cntvoff(u64 cntvoff);
> > diff --git a/arch/arm64/kvm/hyp/nvhe/hyp-main.c b/arch/arm64/kvm/hyp/nvhe/hyp-main.c
> > index 728e01d4536b0..5787eee4c9fe4 100644
> > --- a/arch/arm64/kvm/hyp/nvhe/hyp-main.c
> > +++ b/arch/arm64/kvm/hyp/nvhe/hyp-main.c
> > @@ -125,6 +125,17 @@ static void handle___kvm_tlb_flush_vmid_ipa(struct kvm_cpu_context *host_ctxt)
> > __kvm_tlb_flush_vmid_ipa(kern_hyp_va(mmu), ipa, level);
> > }
> >
> > +static void handle___kvm_tlb_flush_range_vmid_ipa(struct kvm_cpu_context *host_ctxt)
> > +{
> > + DECLARE_REG(struct kvm_s2_mmu *, mmu, host_ctxt, 1);
> > + DECLARE_REG(phys_addr_t, start, host_ctxt, 2);
> > + DECLARE_REG(phys_addr_t, end, host_ctxt, 3);
> > + DECLARE_REG(int, level, host_ctxt, 4);
> > + DECLARE_REG(int, tlb_level, host_ctxt, 5);
> > +
> > + __kvm_tlb_flush_range_vmid_ipa(kern_hyp_va(mmu), start, end, level, tlb_level);
> > +}
> > +
> > static void handle___kvm_tlb_flush_vmid(struct kvm_cpu_context *host_ctxt)
> > {
> > DECLARE_REG(struct kvm_s2_mmu *, mmu, host_ctxt, 1);
> > @@ -315,6 +326,7 @@ static const hcall_t host_hcall[] = {
> > HANDLE_FUNC(__kvm_vcpu_run),
> > HANDLE_FUNC(__kvm_flush_vm_context),
> > HANDLE_FUNC(__kvm_tlb_flush_vmid_ipa),
> > + HANDLE_FUNC(__kvm_tlb_flush_range_vmid_ipa),
> > HANDLE_FUNC(__kvm_tlb_flush_vmid),
> > HANDLE_FUNC(__kvm_flush_cpu_context),
> > HANDLE_FUNC(__kvm_timer_set_cntvoff),
> > diff --git a/arch/arm64/kvm/hyp/nvhe/tlb.c b/arch/arm64/kvm/hyp/nvhe/tlb.c
> > index d296d617f5896..7398dd00445e7 100644
> > --- a/arch/arm64/kvm/hyp/nvhe/tlb.c
> > +++ b/arch/arm64/kvm/hyp/nvhe/tlb.c
> > @@ -109,6 +109,34 @@ void __kvm_tlb_flush_vmid_ipa(struct kvm_s2_mmu *mmu,
> > __tlb_switch_to_host(&cxt);
> > }
> >
> > +void __kvm_tlb_flush_range_vmid_ipa(struct kvm_s2_mmu *mmu, phys_addr_t start,
> > + phys_addr_t end, int level, int tlb_level)
> > +{
> > + struct tlb_inv_context cxt;
> > +
> > + dsb(ishst);
> > +
> > + /* Switch to requested VMID */
> > + __tlb_switch_to_guest(mmu, &cxt);
> > +
> > + __kvm_tlb_flush_range(ipas2e1is, mmu, start, end, level, tlb_level);
> > +
> > + /*
> > + * Range-based ipas2e1is flushes only Stage-2 entries, and since the
> > + * VA isn't available for Stage-1 entries, flush the entire stage-1.
> > + */
>
> nit: if we are going to preserve some of the commentary over in
> __kvm_tlb_flush_vmid_ipa(), I would prefer just an exact copy/paste.
> But, FWIW, I think you can just elide the clarifying comments altogether
> since the relationship between stage-1 and stage-2 invalidations is
> already documented.
>
> > + dsb(ish);
> > + __tlbi(vmalle1is);
> > + dsb(ish);
> > + isb();
> > +
> > + /* See the comment below in __kvm_tlb_flush_vmid_ipa() */
>
> Same comment as above.
>
> --
> Thanks,
> Oliver

\
 
 \ /
  Last update: 2023-04-03 23:10    [W:0.293 / U:0.112 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site