* [PATCH v4 00/46] KVM: PPC: Book3S: C-ify the P9 entry/exit code
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
I think enough changes and fixes have gone in since last round
to repost.
I put a git tree here to make things easier to get.
https://github.com/npiggin/linux/tree/kvm-in-c-2
Main changes since v3:
- Hopefully fixed LPCR sanising [from Fabiano review]
- Added MSR[HV] clearing of guest MSR (like MSR[ME] setting).
- Restored a lost comment about PPR saving, improved comments a bit
[Daniel review]
- Added isyncs, removed XXX isync comment.
- Fixed cede comment [Fabiano]
- In stop handling real mode handlers patch, fixed problem of nested
hcall handler calling cede and xics hcalls and consuming them
[Fabiano]
- Filter xics hcalls from being consumed if !kvmppc_xics_enabled
[Fabiano, Alexey noticed and Cedric provided the right recipe and
comment]
- Removed some more of radix, nested, and virt mode paths from the
P7/8 code [noticed by Alexey].
- Rebased on 5.12-rc4
- Hopefully I responded to all reviews and collected all reviewed-bys.
Thanks,
Nick
Nicholas Piggin (46):
KVM: PPC: Book3S HV: Nested move LPCR sanitising to sanitise_hv_regs
KVM: PPC: Book3S HV: Add a function to filter guest LPCR bits
KVM: PPC: Book3S HV: Disallow LPCR[AIL] to be set to 1 or 2
KVM: PPC: Book3S HV: Prevent radix guests from setting LPCR[TC]
KVM: PPC: Book3S HV: Remove redundant mtspr PSPB
KVM: PPC: Book3S HV: remove unused kvmppc_h_protect argument
KVM: PPC: Book3S HV: Fix CONFIG_SPAPR_TCE_IOMMU=n default hcalls
powerpc/64s: Remove KVM handler support from CBE_RAS interrupts
powerpc/64s: remove KVM SKIP test from instruction breakpoint handler
KVM: PPC: Book3S HV: Ensure MSR[ME] is always set in guest MSR
KVM: PPC: Book3S HV: Ensure MSR[HV] is always clear in guest MSR
KVM: PPC: Book3S 64: move KVM interrupt entry to a common entry point
KVM: PPC: Book3S 64: Move GUEST_MODE_SKIP test into KVM
KVM: PPC: Book3S 64: add hcall interrupt handler
KVM: PPC: Book3S 64: Move hcall early register setup to KVM
KVM: PPC: Book3S 64: Move interrupt early register setup to KVM
KVM: PPC: Book3S 64: move bad_host_intr check to HV handler
KVM: PPC: Book3S 64: Minimise hcall handler calling convention
differences
KVM: PPC: Book3S HV P9: Move radix MMU switching instructions together
KVM: PPC: Book3S HV P9: implement kvmppc_xive_pull_vcpu in C
KVM: PPC: Book3S HV P9: Move xive vcpu context management into
kvmhv_p9_guest_entry
KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9
path
KVM: PPC: Book3S HV P9: Move setting HDEC after switching to guest
LPCR
KVM: PPC: Book3S HV P9: Use large decrementer for HDEC
KVM: PPC: Book3S HV P9: Use host timer accounting to avoid decrementer
read
KVM: PPC: Book3S HV P9: Reduce mftb per guest entry/exit
powerpc: add set_dec_or_work API for safely updating decrementer
KVM: PPC: Book3S HV P9: Reduce irq_work vs guest decrementer races
KVM: PPC: Book3S HV P9: Implement the rest of the P9 path in C
KVM: PPC: Book3S HV P9: inline kvmhv_load_hv_regs_and_go into
__kvmhv_vcpu_entry_p9
KVM: PPC: Book3S HV P9: Read machine check registers while MSR[RI] is
0
KVM: PPC: Book3S HV P9: Improve exit timing accounting coverage
KVM: PPC: Book3S HV P9: Move SPR loading after expiry time check
KVM: PPC: Book3S HV P9: Add helpers for OS SPR handling
KVM: PPC: Book3S HV P9: Switch to guest MMU context as late as
possible
KVM: PPC: Book3S HV: Implement radix prefetch workaround by disabling
MMU
KVM: PPC: Book3S HV: Remove support for dependent threads mode on P9
KVM: PPC: Book3S HV: Remove radix guest support from P7/8 path
KVM: PPC: Book3S HV: Remove virt mode checks from real mode handlers
KVM: PPC: Book3S HV: Remove unused nested HV tests in XICS emulation
KVM: PPC: Book3S HV P9: Allow all P9 processors to enable nested HV
KVM: PPC: Book3S HV: small pseries_do_hcall cleanup
KVM: PPC: Book3S HV: add virtual mode handlers for HPT hcalls and page
faults
KVM: PPC: Book3S HV P9: implement hash guest support
KVM: PPC: Book3S HV P9: implement hash host / hash guest support
KVM: PPC: Book3S HV: remove ISA v3.0 and v3.1 support from P7/8 path
arch/powerpc/include/asm/asm-prototypes.h | 3 +-
arch/powerpc/include/asm/exception-64s.h | 13 +
arch/powerpc/include/asm/kvm_asm.h | 3 +-
arch/powerpc/include/asm/kvm_book3s.h | 2 +
arch/powerpc/include/asm/kvm_book3s_64.h | 8 +
arch/powerpc/include/asm/kvm_host.h | 1 -
arch/powerpc/include/asm/kvm_ppc.h | 20 +-
arch/powerpc/include/asm/mmu_context.h | 6 -
arch/powerpc/include/asm/time.h | 16 +
arch/powerpc/kernel/exceptions-64s.S | 257 ++------
arch/powerpc/kernel/security.c | 5 +-
arch/powerpc/kernel/time.c | 21 +-
arch/powerpc/kvm/Makefile | 6 +
arch/powerpc/kvm/book3s.c | 11 +-
arch/powerpc/kvm/book3s_64_entry.S | 394 ++++++++++++
arch/powerpc/kvm/book3s_64_vio_hv.c | 12 -
arch/powerpc/kvm/book3s_hv.c | 730 ++++++++++++----------
arch/powerpc/kvm/book3s_hv_builtin.c | 138 +---
arch/powerpc/kvm/book3s_hv_interrupt.c | 528 ++++++++++++++++
arch/powerpc/kvm/book3s_hv_interrupts.S | 9 +-
arch/powerpc/kvm/book3s_hv_nested.c | 22 +-
arch/powerpc/kvm/book3s_hv_ras.c | 2 +
arch/powerpc/kvm/book3s_hv_rm_mmu.c | 15 +-
arch/powerpc/kvm/book3s_hv_rm_xics.c | 15 -
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 641 +------------------
arch/powerpc/kvm/book3s_segment.S | 7 +
arch/powerpc/kvm/book3s_xive.c | 119 +++-
arch/powerpc/kvm/book3s_xive.h | 7 -
arch/powerpc/kvm/book3s_xive_native.c | 10 -
arch/powerpc/mm/book3s64/radix_pgtable.c | 27 +-
arch/powerpc/mm/book3s64/radix_tlb.c | 46 --
arch/powerpc/mm/mmu_context.c | 4 +-
arch/powerpc/platforms/powernv/idle.c | 52 +-
33 files changed, 1632 insertions(+), 1518 deletions(-)
create mode 100644 arch/powerpc/kvm/book3s_64_entry.S
create mode 100644 arch/powerpc/kvm/book3s_hv_interrupt.c
--
2.23.0
^ permalink raw reply [flat|nested] 188+ messages in thread* [PATCH v4 01/46] KVM: PPC: Book3S HV: Nested move LPCR sanitising to sanitise_hv_regs
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
This will get a bit more complicated in future patches. Move it
into the helper function.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_nested.c | 18 ++++++++++++------
1 file changed, 12 insertions(+), 6 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
index 0cd0e7aad588..2fe1fea4c934 100644
--- a/arch/powerpc/kvm/book3s_hv_nested.c
+++ b/arch/powerpc/kvm/book3s_hv_nested.c
@@ -134,6 +134,16 @@ static void save_hv_return_state(struct kvm_vcpu *vcpu, int trap,
static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
{
+ struct kvmppc_vcore *vc = vcpu->arch.vcore;
+ u64 mask;
+
+ /*
+ * Don't let L1 change LPCR bits for the L2 except these:
+ */
+ mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
+ LPCR_LPES | LPCR_MER;
+ hr->lpcr = (vc->lpcr & ~mask) | (hr->lpcr & mask);
+
/*
* Don't let L1 enable features for L2 which we've disabled for L1,
* but preserve the interrupt cause field.
@@ -271,8 +281,6 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
u64 hv_ptr, regs_ptr;
u64 hdec_exp;
s64 delta_purr, delta_spurr, delta_ic, delta_vtb;
- u64 mask;
- unsigned long lpcr;
if (vcpu->kvm->arch.l1_ptcr == 0)
return H_NOT_AVAILABLE;
@@ -321,9 +329,7 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
vcpu->arch.nested_vcpu_id = l2_hv.vcpu_token;
vcpu->arch.regs = l2_regs;
vcpu->arch.shregs.msr = vcpu->arch.regs.msr;
- mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
- LPCR_LPES | LPCR_MER;
- lpcr = (vc->lpcr & ~mask) | (l2_hv.lpcr & mask);
+
sanitise_hv_regs(vcpu, &l2_hv);
restore_hv_regs(vcpu, &l2_hv);
@@ -335,7 +341,7 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
r = RESUME_HOST;
break;
}
- r = kvmhv_run_single_vcpu(vcpu, hdec_exp, lpcr);
+ r = kvmhv_run_single_vcpu(vcpu, hdec_exp, l2_hv.lpcr);
} while (is_kvmppc_resume_guest(r));
/* save L2 state for return */
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 01/46] KVM: PPC: Book3S HV: Nested move LPCR sanitising to sanitise_hv_regs
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
This will get a bit more complicated in future patches. Move it
into the helper function.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_nested.c | 18 ++++++++++++------
1 file changed, 12 insertions(+), 6 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
index 0cd0e7aad588..2fe1fea4c934 100644
--- a/arch/powerpc/kvm/book3s_hv_nested.c
+++ b/arch/powerpc/kvm/book3s_hv_nested.c
@@ -134,6 +134,16 @@ static void save_hv_return_state(struct kvm_vcpu *vcpu, int trap,
static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
{
+ struct kvmppc_vcore *vc = vcpu->arch.vcore;
+ u64 mask;
+
+ /*
+ * Don't let L1 change LPCR bits for the L2 except these:
+ */
+ mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
+ LPCR_LPES | LPCR_MER;
+ hr->lpcr = (vc->lpcr & ~mask) | (hr->lpcr & mask);
+
/*
* Don't let L1 enable features for L2 which we've disabled for L1,
* but preserve the interrupt cause field.
@@ -271,8 +281,6 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
u64 hv_ptr, regs_ptr;
u64 hdec_exp;
s64 delta_purr, delta_spurr, delta_ic, delta_vtb;
- u64 mask;
- unsigned long lpcr;
if (vcpu->kvm->arch.l1_ptcr = 0)
return H_NOT_AVAILABLE;
@@ -321,9 +329,7 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
vcpu->arch.nested_vcpu_id = l2_hv.vcpu_token;
vcpu->arch.regs = l2_regs;
vcpu->arch.shregs.msr = vcpu->arch.regs.msr;
- mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
- LPCR_LPES | LPCR_MER;
- lpcr = (vc->lpcr & ~mask) | (l2_hv.lpcr & mask);
+
sanitise_hv_regs(vcpu, &l2_hv);
restore_hv_regs(vcpu, &l2_hv);
@@ -335,7 +341,7 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
r = RESUME_HOST;
break;
}
- r = kvmhv_run_single_vcpu(vcpu, hdec_exp, lpcr);
+ r = kvmhv_run_single_vcpu(vcpu, hdec_exp, l2_hv.lpcr);
} while (is_kvmppc_resume_guest(r));
/* save L2 state for return */
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* Re: [PATCH v4 01/46] KVM: PPC: Book3S HV: Nested move LPCR sanitising to sanitise_hv_regs
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 18:08 ` Fabiano Rosas
-1 siblings, 0 replies; 188+ messages in thread
From: Fabiano Rosas @ 2021-03-23 18:08 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Nicholas Piggin <npiggin@gmail.com> writes:
> This will get a bit more complicated in future patches. Move it
> into the helper function.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
> ---
> arch/powerpc/kvm/book3s_hv_nested.c | 18 ++++++++++++------
> 1 file changed, 12 insertions(+), 6 deletions(-)
>
> diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
> index 0cd0e7aad588..2fe1fea4c934 100644
> --- a/arch/powerpc/kvm/book3s_hv_nested.c
> +++ b/arch/powerpc/kvm/book3s_hv_nested.c
> @@ -134,6 +134,16 @@ static void save_hv_return_state(struct kvm_vcpu *vcpu, int trap,
>
> static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
> {
> + struct kvmppc_vcore *vc = vcpu->arch.vcore;
> + u64 mask;
> +
> + /*
> + * Don't let L1 change LPCR bits for the L2 except these:
> + */
> + mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
> + LPCR_LPES | LPCR_MER;
> + hr->lpcr = (vc->lpcr & ~mask) | (hr->lpcr & mask);
> +
> /*
> * Don't let L1 enable features for L2 which we've disabled for L1,
> * but preserve the interrupt cause field.
> @@ -271,8 +281,6 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
> u64 hv_ptr, regs_ptr;
> u64 hdec_exp;
> s64 delta_purr, delta_spurr, delta_ic, delta_vtb;
> - u64 mask;
> - unsigned long lpcr;
>
> if (vcpu->kvm->arch.l1_ptcr == 0)
> return H_NOT_AVAILABLE;
> @@ -321,9 +329,7 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
> vcpu->arch.nested_vcpu_id = l2_hv.vcpu_token;
> vcpu->arch.regs = l2_regs;
> vcpu->arch.shregs.msr = vcpu->arch.regs.msr;
> - mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
> - LPCR_LPES | LPCR_MER;
> - lpcr = (vc->lpcr & ~mask) | (l2_hv.lpcr & mask);
> +
> sanitise_hv_regs(vcpu, &l2_hv);
> restore_hv_regs(vcpu, &l2_hv);
>
> @@ -335,7 +341,7 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
> r = RESUME_HOST;
> break;
> }
> - r = kvmhv_run_single_vcpu(vcpu, hdec_exp, lpcr);
> + r = kvmhv_run_single_vcpu(vcpu, hdec_exp, l2_hv.lpcr);
> } while (is_kvmppc_resume_guest(r));
>
> /* save L2 state for return */
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 01/46] KVM: PPC: Book3S HV: Nested move LPCR sanitising to sanitise_hv_regs
@ 2021-03-23 18:08 ` Fabiano Rosas
0 siblings, 0 replies; 188+ messages in thread
From: Fabiano Rosas @ 2021-03-23 18:08 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Nicholas Piggin <npiggin@gmail.com> writes:
> This will get a bit more complicated in future patches. Move it
> into the helper function.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
> ---
> arch/powerpc/kvm/book3s_hv_nested.c | 18 ++++++++++++------
> 1 file changed, 12 insertions(+), 6 deletions(-)
>
> diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
> index 0cd0e7aad588..2fe1fea4c934 100644
> --- a/arch/powerpc/kvm/book3s_hv_nested.c
> +++ b/arch/powerpc/kvm/book3s_hv_nested.c
> @@ -134,6 +134,16 @@ static void save_hv_return_state(struct kvm_vcpu *vcpu, int trap,
>
> static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
> {
> + struct kvmppc_vcore *vc = vcpu->arch.vcore;
> + u64 mask;
> +
> + /*
> + * Don't let L1 change LPCR bits for the L2 except these:
> + */
> + mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
> + LPCR_LPES | LPCR_MER;
> + hr->lpcr = (vc->lpcr & ~mask) | (hr->lpcr & mask);
> +
> /*
> * Don't let L1 enable features for L2 which we've disabled for L1,
> * but preserve the interrupt cause field.
> @@ -271,8 +281,6 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
> u64 hv_ptr, regs_ptr;
> u64 hdec_exp;
> s64 delta_purr, delta_spurr, delta_ic, delta_vtb;
> - u64 mask;
> - unsigned long lpcr;
>
> if (vcpu->kvm->arch.l1_ptcr = 0)
> return H_NOT_AVAILABLE;
> @@ -321,9 +329,7 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
> vcpu->arch.nested_vcpu_id = l2_hv.vcpu_token;
> vcpu->arch.regs = l2_regs;
> vcpu->arch.shregs.msr = vcpu->arch.regs.msr;
> - mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
> - LPCR_LPES | LPCR_MER;
> - lpcr = (vc->lpcr & ~mask) | (l2_hv.lpcr & mask);
> +
> sanitise_hv_regs(vcpu, &l2_hv);
> restore_hv_regs(vcpu, &l2_hv);
>
> @@ -335,7 +341,7 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
> r = RESUME_HOST;
> break;
> }
> - r = kvmhv_run_single_vcpu(vcpu, hdec_exp, lpcr);
> + r = kvmhv_run_single_vcpu(vcpu, hdec_exp, l2_hv.lpcr);
> } while (is_kvmppc_resume_guest(r));
>
> /* save L2 state for return */
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 01/46] KVM: PPC: Book3S HV: Nested move LPCR sanitising to sanitise_hv_regs
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-31 2:47 ` Paul Mackerras
-1 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 2:47 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc
On Tue, Mar 23, 2021 at 11:02:20AM +1000, Nicholas Piggin wrote:
> This will get a bit more complicated in future patches. Move it
> into the helper function.
This does change L1-visible behaviour, because now the L1 hypervisor
can see the LPCR bits that L0 is using, whereas previously it couldn't
(and that was deliberate). I can't point to a specific scenario where
that is a real problem, but nevertheless it worries me. And the
behaviour change should have been mentioned in the commit message at
least.
Paul.
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 01/46] KVM: PPC: Book3S HV: Nested move LPCR sanitising to sanitise_hv_regs
@ 2021-03-31 2:47 ` Paul Mackerras
0 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 2:47 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc
On Tue, Mar 23, 2021 at 11:02:20AM +1000, Nicholas Piggin wrote:
> This will get a bit more complicated in future patches. Move it
> into the helper function.
This does change L1-visible behaviour, because now the L1 hypervisor
can see the LPCR bits that L0 is using, whereas previously it couldn't
(and that was deliberate). I can't point to a specific scenario where
that is a real problem, but nevertheless it worries me. And the
behaviour change should have been mentioned in the commit message at
least.
Paul.
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 02/46] KVM: PPC: Book3S HV: Add a function to filter guest LPCR bits
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Guest LPCR depends on hardware type, and future changes will add
restrictions based on errata and guest MMU mode. Move this logic
to a common function and use it for the cases where the guest
wants to update its LPCR (or the LPCR of a nested guest).
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_book3s.h | 2 +
arch/powerpc/kvm/book3s_hv.c | 60 ++++++++++++++++++---------
arch/powerpc/kvm/book3s_hv_nested.c | 3 +-
3 files changed, 45 insertions(+), 20 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_book3s.h b/arch/powerpc/include/asm/kvm_book3s.h
index 2f5f919f6cd3..3eec3ef6f083 100644
--- a/arch/powerpc/include/asm/kvm_book3s.h
+++ b/arch/powerpc/include/asm/kvm_book3s.h
@@ -258,6 +258,8 @@ extern long kvmppc_hv_get_dirty_log_hpt(struct kvm *kvm,
extern void kvmppc_harvest_vpa_dirty(struct kvmppc_vpa *vpa,
struct kvm_memory_slot *memslot,
unsigned long *map);
+extern unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc,
+ unsigned long lpcr);
extern void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr,
unsigned long mask);
extern void kvmppc_set_fscr(struct kvm_vcpu *vcpu, u64 fscr);
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 13bad6bf4c95..c4539c38c639 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -1635,6 +1635,27 @@ static int kvm_arch_vcpu_ioctl_set_sregs_hv(struct kvm_vcpu *vcpu,
return 0;
}
+/*
+ * Enforce limits on guest LPCR values based on hardware availability,
+ * guest configuration, and possibly hypervisor support and security
+ * concerns.
+ */
+unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
+{
+ /* On POWER8 and above, userspace can modify AIL */
+ if (!cpu_has_feature(CPU_FTR_ARCH_207S))
+ lpcr &= ~LPCR_AIL;
+
+ /*
+ * On POWER9, allow userspace to enable large decrementer for the
+ * guest, whether or not the host has it enabled.
+ */
+ if (!cpu_has_feature(CPU_FTR_ARCH_300))
+ lpcr &= ~LPCR_LD;
+
+ return lpcr;
+}
+
static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
bool preserve_top32)
{
@@ -1643,6 +1664,23 @@ static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
u64 mask;
spin_lock(&vc->lock);
+
+ /*
+ * Userspace can only modify
+ * DPFD (default prefetch depth), ILE (interrupt little-endian),
+ * TC (translation control), AIL (alternate interrupt location),
+ * LD (large decrementer).
+ * These are subject to restrictions from kvmppc_filter_lcpr_hv().
+ */
+ mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD;
+
+ /* Broken 32-bit version of LPCR must not clear top bits */
+ if (preserve_top32)
+ mask &= 0xFFFFFFFF;
+
+ new_lpcr = kvmppc_filter_lpcr_hv(vc,
+ (vc->lpcr & ~mask) | (new_lpcr & mask));
+
/*
* If ILE (interrupt little-endian) has changed, update the
* MSR_LE bit in the intr_msr for each vcpu in this vcore.
@@ -1661,25 +1699,8 @@ static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
}
}
- /*
- * Userspace can only modify DPFD (default prefetch depth),
- * ILE (interrupt little-endian) and TC (translation control).
- * On POWER8 and POWER9 userspace can also modify AIL (alt. interrupt loc.).
- */
- mask = LPCR_DPFD | LPCR_ILE | LPCR_TC;
- if (cpu_has_feature(CPU_FTR_ARCH_207S))
- mask |= LPCR_AIL;
- /*
- * On POWER9, allow userspace to enable large decrementer for the
- * guest, whether or not the host has it enabled.
- */
- if (cpu_has_feature(CPU_FTR_ARCH_300))
- mask |= LPCR_LD;
+ vc->lpcr = new_lpcr;
- /* Broken 32-bit version of LPCR must not clear top bits */
- if (preserve_top32)
- mask &= 0xFFFFFFFF;
- vc->lpcr = (vc->lpcr & ~mask) | (new_lpcr & mask);
spin_unlock(&vc->lock);
}
@@ -4641,8 +4662,9 @@ void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr, unsigned long mask)
struct kvmppc_vcore *vc = kvm->arch.vcores[i];
if (!vc)
continue;
+
spin_lock(&vc->lock);
- vc->lpcr = (vc->lpcr & ~mask) | lpcr;
+ vc->lpcr = kvmppc_filter_lpcr_hv(vc, (vc->lpcr & ~mask) | lpcr);
spin_unlock(&vc->lock);
if (++cores_done >= kvm->arch.online_vcores)
break;
diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
index 2fe1fea4c934..f7b441b3eb17 100644
--- a/arch/powerpc/kvm/book3s_hv_nested.c
+++ b/arch/powerpc/kvm/book3s_hv_nested.c
@@ -142,7 +142,8 @@ static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
*/
mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
LPCR_LPES | LPCR_MER;
- hr->lpcr = (vc->lpcr & ~mask) | (hr->lpcr & mask);
+ hr->lpcr = kvmppc_filter_lpcr_hv(vc,
+ (vc->lpcr & ~mask) | (hr->lpcr & mask));
/*
* Don't let L1 enable features for L2 which we've disabled for L1,
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 02/46] KVM: PPC: Book3S HV: Add a function to filter guest LPCR bits
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Guest LPCR depends on hardware type, and future changes will add
restrictions based on errata and guest MMU mode. Move this logic
to a common function and use it for the cases where the guest
wants to update its LPCR (or the LPCR of a nested guest).
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_book3s.h | 2 +
arch/powerpc/kvm/book3s_hv.c | 60 ++++++++++++++++++---------
arch/powerpc/kvm/book3s_hv_nested.c | 3 +-
3 files changed, 45 insertions(+), 20 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_book3s.h b/arch/powerpc/include/asm/kvm_book3s.h
index 2f5f919f6cd3..3eec3ef6f083 100644
--- a/arch/powerpc/include/asm/kvm_book3s.h
+++ b/arch/powerpc/include/asm/kvm_book3s.h
@@ -258,6 +258,8 @@ extern long kvmppc_hv_get_dirty_log_hpt(struct kvm *kvm,
extern void kvmppc_harvest_vpa_dirty(struct kvmppc_vpa *vpa,
struct kvm_memory_slot *memslot,
unsigned long *map);
+extern unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc,
+ unsigned long lpcr);
extern void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr,
unsigned long mask);
extern void kvmppc_set_fscr(struct kvm_vcpu *vcpu, u64 fscr);
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 13bad6bf4c95..c4539c38c639 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -1635,6 +1635,27 @@ static int kvm_arch_vcpu_ioctl_set_sregs_hv(struct kvm_vcpu *vcpu,
return 0;
}
+/*
+ * Enforce limits on guest LPCR values based on hardware availability,
+ * guest configuration, and possibly hypervisor support and security
+ * concerns.
+ */
+unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
+{
+ /* On POWER8 and above, userspace can modify AIL */
+ if (!cpu_has_feature(CPU_FTR_ARCH_207S))
+ lpcr &= ~LPCR_AIL;
+
+ /*
+ * On POWER9, allow userspace to enable large decrementer for the
+ * guest, whether or not the host has it enabled.
+ */
+ if (!cpu_has_feature(CPU_FTR_ARCH_300))
+ lpcr &= ~LPCR_LD;
+
+ return lpcr;
+}
+
static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
bool preserve_top32)
{
@@ -1643,6 +1664,23 @@ static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
u64 mask;
spin_lock(&vc->lock);
+
+ /*
+ * Userspace can only modify
+ * DPFD (default prefetch depth), ILE (interrupt little-endian),
+ * TC (translation control), AIL (alternate interrupt location),
+ * LD (large decrementer).
+ * These are subject to restrictions from kvmppc_filter_lcpr_hv().
+ */
+ mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD;
+
+ /* Broken 32-bit version of LPCR must not clear top bits */
+ if (preserve_top32)
+ mask &= 0xFFFFFFFF;
+
+ new_lpcr = kvmppc_filter_lpcr_hv(vc,
+ (vc->lpcr & ~mask) | (new_lpcr & mask));
+
/*
* If ILE (interrupt little-endian) has changed, update the
* MSR_LE bit in the intr_msr for each vcpu in this vcore.
@@ -1661,25 +1699,8 @@ static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
}
}
- /*
- * Userspace can only modify DPFD (default prefetch depth),
- * ILE (interrupt little-endian) and TC (translation control).
- * On POWER8 and POWER9 userspace can also modify AIL (alt. interrupt loc.).
- */
- mask = LPCR_DPFD | LPCR_ILE | LPCR_TC;
- if (cpu_has_feature(CPU_FTR_ARCH_207S))
- mask |= LPCR_AIL;
- /*
- * On POWER9, allow userspace to enable large decrementer for the
- * guest, whether or not the host has it enabled.
- */
- if (cpu_has_feature(CPU_FTR_ARCH_300))
- mask |= LPCR_LD;
+ vc->lpcr = new_lpcr;
- /* Broken 32-bit version of LPCR must not clear top bits */
- if (preserve_top32)
- mask &= 0xFFFFFFFF;
- vc->lpcr = (vc->lpcr & ~mask) | (new_lpcr & mask);
spin_unlock(&vc->lock);
}
@@ -4641,8 +4662,9 @@ void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr, unsigned long mask)
struct kvmppc_vcore *vc = kvm->arch.vcores[i];
if (!vc)
continue;
+
spin_lock(&vc->lock);
- vc->lpcr = (vc->lpcr & ~mask) | lpcr;
+ vc->lpcr = kvmppc_filter_lpcr_hv(vc, (vc->lpcr & ~mask) | lpcr);
spin_unlock(&vc->lock);
if (++cores_done >= kvm->arch.online_vcores)
break;
diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
index 2fe1fea4c934..f7b441b3eb17 100644
--- a/arch/powerpc/kvm/book3s_hv_nested.c
+++ b/arch/powerpc/kvm/book3s_hv_nested.c
@@ -142,7 +142,8 @@ static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
*/
mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
LPCR_LPES | LPCR_MER;
- hr->lpcr = (vc->lpcr & ~mask) | (hr->lpcr & mask);
+ hr->lpcr = kvmppc_filter_lpcr_hv(vc,
+ (vc->lpcr & ~mask) | (hr->lpcr & mask));
/*
* Don't let L1 enable features for L2 which we've disabled for L1,
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* Re: [PATCH v4 02/46] KVM: PPC: Book3S HV: Add a function to filter guest LPCR bits
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 18:17 ` Fabiano Rosas
-1 siblings, 0 replies; 188+ messages in thread
From: Fabiano Rosas @ 2021-03-23 18:17 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Nicholas Piggin <npiggin@gmail.com> writes:
> Guest LPCR depends on hardware type, and future changes will add
> restrictions based on errata and guest MMU mode. Move this logic
> to a common function and use it for the cases where the guest
> wants to update its LPCR (or the LPCR of a nested guest).
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
> ---
> arch/powerpc/include/asm/kvm_book3s.h | 2 +
> arch/powerpc/kvm/book3s_hv.c | 60 ++++++++++++++++++---------
> arch/powerpc/kvm/book3s_hv_nested.c | 3 +-
> 3 files changed, 45 insertions(+), 20 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/kvm_book3s.h b/arch/powerpc/include/asm/kvm_book3s.h
> index 2f5f919f6cd3..3eec3ef6f083 100644
> --- a/arch/powerpc/include/asm/kvm_book3s.h
> +++ b/arch/powerpc/include/asm/kvm_book3s.h
> @@ -258,6 +258,8 @@ extern long kvmppc_hv_get_dirty_log_hpt(struct kvm *kvm,
> extern void kvmppc_harvest_vpa_dirty(struct kvmppc_vpa *vpa,
> struct kvm_memory_slot *memslot,
> unsigned long *map);
> +extern unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc,
> + unsigned long lpcr);
> extern void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr,
> unsigned long mask);
> extern void kvmppc_set_fscr(struct kvm_vcpu *vcpu, u64 fscr);
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index 13bad6bf4c95..c4539c38c639 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1635,6 +1635,27 @@ static int kvm_arch_vcpu_ioctl_set_sregs_hv(struct kvm_vcpu *vcpu,
> return 0;
> }
>
> +/*
> + * Enforce limits on guest LPCR values based on hardware availability,
> + * guest configuration, and possibly hypervisor support and security
> + * concerns.
> + */
> +unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
> +{
> + /* On POWER8 and above, userspace can modify AIL */
> + if (!cpu_has_feature(CPU_FTR_ARCH_207S))
> + lpcr &= ~LPCR_AIL;
> +
> + /*
> + * On POWER9, allow userspace to enable large decrementer for the
> + * guest, whether or not the host has it enabled.
> + */
> + if (!cpu_has_feature(CPU_FTR_ARCH_300))
> + lpcr &= ~LPCR_LD;
> +
> + return lpcr;
> +}
> +
> static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
> bool preserve_top32)
> {
> @@ -1643,6 +1664,23 @@ static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
> u64 mask;
>
> spin_lock(&vc->lock);
> +
> + /*
> + * Userspace can only modify
> + * DPFD (default prefetch depth), ILE (interrupt little-endian),
> + * TC (translation control), AIL (alternate interrupt location),
> + * LD (large decrementer).
> + * These are subject to restrictions from kvmppc_filter_lcpr_hv().
> + */
> + mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD;
> +
> + /* Broken 32-bit version of LPCR must not clear top bits */
> + if (preserve_top32)
> + mask &= 0xFFFFFFFF;
> +
> + new_lpcr = kvmppc_filter_lpcr_hv(vc,
> + (vc->lpcr & ~mask) | (new_lpcr & mask));
> +
> /*
> * If ILE (interrupt little-endian) has changed, update the
> * MSR_LE bit in the intr_msr for each vcpu in this vcore.
> @@ -1661,25 +1699,8 @@ static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
> }
> }
>
> - /*
> - * Userspace can only modify DPFD (default prefetch depth),
> - * ILE (interrupt little-endian) and TC (translation control).
> - * On POWER8 and POWER9 userspace can also modify AIL (alt. interrupt loc.).
> - */
> - mask = LPCR_DPFD | LPCR_ILE | LPCR_TC;
> - if (cpu_has_feature(CPU_FTR_ARCH_207S))
> - mask |= LPCR_AIL;
> - /*
> - * On POWER9, allow userspace to enable large decrementer for the
> - * guest, whether or not the host has it enabled.
> - */
> - if (cpu_has_feature(CPU_FTR_ARCH_300))
> - mask |= LPCR_LD;
> + vc->lpcr = new_lpcr;
>
> - /* Broken 32-bit version of LPCR must not clear top bits */
> - if (preserve_top32)
> - mask &= 0xFFFFFFFF;
> - vc->lpcr = (vc->lpcr & ~mask) | (new_lpcr & mask);
> spin_unlock(&vc->lock);
> }
>
> @@ -4641,8 +4662,9 @@ void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr, unsigned long mask)
> struct kvmppc_vcore *vc = kvm->arch.vcores[i];
> if (!vc)
> continue;
> +
> spin_lock(&vc->lock);
> - vc->lpcr = (vc->lpcr & ~mask) | lpcr;
> + vc->lpcr = kvmppc_filter_lpcr_hv(vc, (vc->lpcr & ~mask) | lpcr);
> spin_unlock(&vc->lock);
> if (++cores_done >= kvm->arch.online_vcores)
> break;
> diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
> index 2fe1fea4c934..f7b441b3eb17 100644
> --- a/arch/powerpc/kvm/book3s_hv_nested.c
> +++ b/arch/powerpc/kvm/book3s_hv_nested.c
> @@ -142,7 +142,8 @@ static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
> */
> mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
> LPCR_LPES | LPCR_MER;
> - hr->lpcr = (vc->lpcr & ~mask) | (hr->lpcr & mask);
> + hr->lpcr = kvmppc_filter_lpcr_hv(vc,
> + (vc->lpcr & ~mask) | (hr->lpcr & mask));
>
> /*
> * Don't let L1 enable features for L2 which we've disabled for L1,
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 02/46] KVM: PPC: Book3S HV: Add a function to filter guest LPCR bits
@ 2021-03-23 18:17 ` Fabiano Rosas
0 siblings, 0 replies; 188+ messages in thread
From: Fabiano Rosas @ 2021-03-23 18:17 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Nicholas Piggin <npiggin@gmail.com> writes:
> Guest LPCR depends on hardware type, and future changes will add
> restrictions based on errata and guest MMU mode. Move this logic
> to a common function and use it for the cases where the guest
> wants to update its LPCR (or the LPCR of a nested guest).
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
> ---
> arch/powerpc/include/asm/kvm_book3s.h | 2 +
> arch/powerpc/kvm/book3s_hv.c | 60 ++++++++++++++++++---------
> arch/powerpc/kvm/book3s_hv_nested.c | 3 +-
> 3 files changed, 45 insertions(+), 20 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/kvm_book3s.h b/arch/powerpc/include/asm/kvm_book3s.h
> index 2f5f919f6cd3..3eec3ef6f083 100644
> --- a/arch/powerpc/include/asm/kvm_book3s.h
> +++ b/arch/powerpc/include/asm/kvm_book3s.h
> @@ -258,6 +258,8 @@ extern long kvmppc_hv_get_dirty_log_hpt(struct kvm *kvm,
> extern void kvmppc_harvest_vpa_dirty(struct kvmppc_vpa *vpa,
> struct kvm_memory_slot *memslot,
> unsigned long *map);
> +extern unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc,
> + unsigned long lpcr);
> extern void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr,
> unsigned long mask);
> extern void kvmppc_set_fscr(struct kvm_vcpu *vcpu, u64 fscr);
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index 13bad6bf4c95..c4539c38c639 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1635,6 +1635,27 @@ static int kvm_arch_vcpu_ioctl_set_sregs_hv(struct kvm_vcpu *vcpu,
> return 0;
> }
>
> +/*
> + * Enforce limits on guest LPCR values based on hardware availability,
> + * guest configuration, and possibly hypervisor support and security
> + * concerns.
> + */
> +unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
> +{
> + /* On POWER8 and above, userspace can modify AIL */
> + if (!cpu_has_feature(CPU_FTR_ARCH_207S))
> + lpcr &= ~LPCR_AIL;
> +
> + /*
> + * On POWER9, allow userspace to enable large decrementer for the
> + * guest, whether or not the host has it enabled.
> + */
> + if (!cpu_has_feature(CPU_FTR_ARCH_300))
> + lpcr &= ~LPCR_LD;
> +
> + return lpcr;
> +}
> +
> static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
> bool preserve_top32)
> {
> @@ -1643,6 +1664,23 @@ static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
> u64 mask;
>
> spin_lock(&vc->lock);
> +
> + /*
> + * Userspace can only modify
> + * DPFD (default prefetch depth), ILE (interrupt little-endian),
> + * TC (translation control), AIL (alternate interrupt location),
> + * LD (large decrementer).
> + * These are subject to restrictions from kvmppc_filter_lcpr_hv().
> + */
> + mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD;
> +
> + /* Broken 32-bit version of LPCR must not clear top bits */
> + if (preserve_top32)
> + mask &= 0xFFFFFFFF;
> +
> + new_lpcr = kvmppc_filter_lpcr_hv(vc,
> + (vc->lpcr & ~mask) | (new_lpcr & mask));
> +
> /*
> * If ILE (interrupt little-endian) has changed, update the
> * MSR_LE bit in the intr_msr for each vcpu in this vcore.
> @@ -1661,25 +1699,8 @@ static void kvmppc_set_lpcr(struct kvm_vcpu *vcpu, u64 new_lpcr,
> }
> }
>
> - /*
> - * Userspace can only modify DPFD (default prefetch depth),
> - * ILE (interrupt little-endian) and TC (translation control).
> - * On POWER8 and POWER9 userspace can also modify AIL (alt. interrupt loc.).
> - */
> - mask = LPCR_DPFD | LPCR_ILE | LPCR_TC;
> - if (cpu_has_feature(CPU_FTR_ARCH_207S))
> - mask |= LPCR_AIL;
> - /*
> - * On POWER9, allow userspace to enable large decrementer for the
> - * guest, whether or not the host has it enabled.
> - */
> - if (cpu_has_feature(CPU_FTR_ARCH_300))
> - mask |= LPCR_LD;
> + vc->lpcr = new_lpcr;
>
> - /* Broken 32-bit version of LPCR must not clear top bits */
> - if (preserve_top32)
> - mask &= 0xFFFFFFFF;
> - vc->lpcr = (vc->lpcr & ~mask) | (new_lpcr & mask);
> spin_unlock(&vc->lock);
> }
>
> @@ -4641,8 +4662,9 @@ void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr, unsigned long mask)
> struct kvmppc_vcore *vc = kvm->arch.vcores[i];
> if (!vc)
> continue;
> +
> spin_lock(&vc->lock);
> - vc->lpcr = (vc->lpcr & ~mask) | lpcr;
> + vc->lpcr = kvmppc_filter_lpcr_hv(vc, (vc->lpcr & ~mask) | lpcr);
> spin_unlock(&vc->lock);
> if (++cores_done >= kvm->arch.online_vcores)
> break;
> diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
> index 2fe1fea4c934..f7b441b3eb17 100644
> --- a/arch/powerpc/kvm/book3s_hv_nested.c
> +++ b/arch/powerpc/kvm/book3s_hv_nested.c
> @@ -142,7 +142,8 @@ static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
> */
> mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
> LPCR_LPES | LPCR_MER;
> - hr->lpcr = (vc->lpcr & ~mask) | (hr->lpcr & mask);
> + hr->lpcr = kvmppc_filter_lpcr_hv(vc,
> + (vc->lpcr & ~mask) | (hr->lpcr & mask));
>
> /*
> * Don't let L1 enable features for L2 which we've disabled for L1,
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 02/46] KVM: PPC: Book3S HV: Add a function to filter guest LPCR bits
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-31 4:08 ` Paul Mackerras
-1 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:08 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc
On Tue, Mar 23, 2021 at 11:02:21AM +1000, Nicholas Piggin wrote:
> Guest LPCR depends on hardware type, and future changes will add
> restrictions based on errata and guest MMU mode. Move this logic
> to a common function and use it for the cases where the guest
> wants to update its LPCR (or the LPCR of a nested guest).
[snip]
> @@ -4641,8 +4662,9 @@ void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr, unsigned long mask)
> struct kvmppc_vcore *vc = kvm->arch.vcores[i];
> if (!vc)
> continue;
> +
> spin_lock(&vc->lock);
> - vc->lpcr = (vc->lpcr & ~mask) | lpcr;
> + vc->lpcr = kvmppc_filter_lpcr_hv(vc, (vc->lpcr & ~mask) | lpcr);
This change seems unnecessary, since kvmppc_update_lpcr is called only
to update MMU configuration bits, not as a result of any action by
userspace or a nested hypervisor. It's also beyond the scope of what
was mentioned in the commit message.
Paul.
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 02/46] KVM: PPC: Book3S HV: Add a function to filter guest LPCR bits
@ 2021-03-31 4:08 ` Paul Mackerras
0 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:08 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc
On Tue, Mar 23, 2021 at 11:02:21AM +1000, Nicholas Piggin wrote:
> Guest LPCR depends on hardware type, and future changes will add
> restrictions based on errata and guest MMU mode. Move this logic
> to a common function and use it for the cases where the guest
> wants to update its LPCR (or the LPCR of a nested guest).
[snip]
> @@ -4641,8 +4662,9 @@ void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr, unsigned long mask)
> struct kvmppc_vcore *vc = kvm->arch.vcores[i];
> if (!vc)
> continue;
> +
> spin_lock(&vc->lock);
> - vc->lpcr = (vc->lpcr & ~mask) | lpcr;
> + vc->lpcr = kvmppc_filter_lpcr_hv(vc, (vc->lpcr & ~mask) | lpcr);
This change seems unnecessary, since kvmppc_update_lpcr is called only
to update MMU configuration bits, not as a result of any action by
userspace or a nested hypervisor. It's also beyond the scope of what
was mentioned in the commit message.
Paul.
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 02/46] KVM: PPC: Book3S HV: Add a function to filter guest LPCR bits
2021-03-31 4:08 ` Paul Mackerras
@ 2021-04-01 9:32 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 9:32 UTC (permalink / raw)
To: Paul Mackerras; +Cc: linuxppc-dev, kvm-ppc
Excerpts from Paul Mackerras's message of March 31, 2021 2:08 pm:
> On Tue, Mar 23, 2021 at 11:02:21AM +1000, Nicholas Piggin wrote:
>> Guest LPCR depends on hardware type, and future changes will add
>> restrictions based on errata and guest MMU mode. Move this logic
>> to a common function and use it for the cases where the guest
>> wants to update its LPCR (or the LPCR of a nested guest).
>
> [snip]
>
>> @@ -4641,8 +4662,9 @@ void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr, unsigned long mask)
>> struct kvmppc_vcore *vc = kvm->arch.vcores[i];
>> if (!vc)
>> continue;
>> +
>> spin_lock(&vc->lock);
>> - vc->lpcr = (vc->lpcr & ~mask) | lpcr;
>> + vc->lpcr = kvmppc_filter_lpcr_hv(vc, (vc->lpcr & ~mask) | lpcr);
>
> This change seems unnecessary, since kvmppc_update_lpcr is called only
> to update MMU configuration bits, not as a result of any action by
> userspace or a nested hypervisor. It's also beyond the scope of what
> was mentioned in the commit message.
I didn't think it was outside the spirit of the patch, but yes
only the guest update LPCR case was enumerated. Would it be more
consistent to add it to the changelog and leave it in here or would
you prefer it left out until there is a real use?
The intention is a single location to add some of these things
(handwaving: say tlbie doesn't work on some chip and we want to
emulate it for old guests we could clear GTSE).
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 02/46] KVM: PPC: Book3S HV: Add a function to filter guest LPCR bits
@ 2021-04-01 9:32 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 9:32 UTC (permalink / raw)
To: Paul Mackerras; +Cc: linuxppc-dev, kvm-ppc
Excerpts from Paul Mackerras's message of March 31, 2021 2:08 pm:
> On Tue, Mar 23, 2021 at 11:02:21AM +1000, Nicholas Piggin wrote:
>> Guest LPCR depends on hardware type, and future changes will add
>> restrictions based on errata and guest MMU mode. Move this logic
>> to a common function and use it for the cases where the guest
>> wants to update its LPCR (or the LPCR of a nested guest).
>
> [snip]
>
>> @@ -4641,8 +4662,9 @@ void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr, unsigned long mask)
>> struct kvmppc_vcore *vc = kvm->arch.vcores[i];
>> if (!vc)
>> continue;
>> +
>> spin_lock(&vc->lock);
>> - vc->lpcr = (vc->lpcr & ~mask) | lpcr;
>> + vc->lpcr = kvmppc_filter_lpcr_hv(vc, (vc->lpcr & ~mask) | lpcr);
>
> This change seems unnecessary, since kvmppc_update_lpcr is called only
> to update MMU configuration bits, not as a result of any action by
> userspace or a nested hypervisor. It's also beyond the scope of what
> was mentioned in the commit message.
I didn't think it was outside the spirit of the patch, but yes
only the guest update LPCR case was enumerated. Would it be more
consistent to add it to the changelog and leave it in here or would
you prefer it left out until there is a real use?
The intention is a single location to add some of these things
(handwaving: say tlbie doesn't work on some chip and we want to
emulate it for old guests we could clear GTSE).
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 02/46] KVM: PPC: Book3S HV: Add a function to filter guest LPCR bits
2021-04-01 9:32 ` Nicholas Piggin
@ 2021-04-01 13:28 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 13:28 UTC (permalink / raw)
To: Paul Mackerras; +Cc: linuxppc-dev, kvm-ppc
Excerpts from Nicholas Piggin's message of April 1, 2021 7:32 pm:
> Excerpts from Paul Mackerras's message of March 31, 2021 2:08 pm:
>> On Tue, Mar 23, 2021 at 11:02:21AM +1000, Nicholas Piggin wrote:
>>> Guest LPCR depends on hardware type, and future changes will add
>>> restrictions based on errata and guest MMU mode. Move this logic
>>> to a common function and use it for the cases where the guest
>>> wants to update its LPCR (or the LPCR of a nested guest).
>>
>> [snip]
>>
>>> @@ -4641,8 +4662,9 @@ void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr, unsigned long mask)
>>> struct kvmppc_vcore *vc = kvm->arch.vcores[i];
>>> if (!vc)
>>> continue;
>>> +
>>> spin_lock(&vc->lock);
>>> - vc->lpcr = (vc->lpcr & ~mask) | lpcr;
>>> + vc->lpcr = kvmppc_filter_lpcr_hv(vc, (vc->lpcr & ~mask) | lpcr);
>>
>> This change seems unnecessary, since kvmppc_update_lpcr is called only
>> to update MMU configuration bits, not as a result of any action by
>> userspace or a nested hypervisor. It's also beyond the scope of what
>> was mentioned in the commit message.
>
> I didn't think it was outside the spirit of the patch, but yes
> only the guest update LPCR case was enumerated. Would it be more
> consistent to add it to the changelog and leave it in here or would
> you prefer it left out until there is a real use?
On second thoughts, I already left at least one other place without
such a check, so I now tend to agree with you. But I instead added
a test that just ensures the host is not out of synch with itself in
terms of what it can set the LPCR to.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 02/46] KVM: PPC: Book3S HV: Add a function to filter guest LPCR bits
@ 2021-04-01 13:28 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 13:28 UTC (permalink / raw)
To: Paul Mackerras; +Cc: linuxppc-dev, kvm-ppc
Excerpts from Nicholas Piggin's message of April 1, 2021 7:32 pm:
> Excerpts from Paul Mackerras's message of March 31, 2021 2:08 pm:
>> On Tue, Mar 23, 2021 at 11:02:21AM +1000, Nicholas Piggin wrote:
>>> Guest LPCR depends on hardware type, and future changes will add
>>> restrictions based on errata and guest MMU mode. Move this logic
>>> to a common function and use it for the cases where the guest
>>> wants to update its LPCR (or the LPCR of a nested guest).
>>
>> [snip]
>>
>>> @@ -4641,8 +4662,9 @@ void kvmppc_update_lpcr(struct kvm *kvm, unsigned long lpcr, unsigned long mask)
>>> struct kvmppc_vcore *vc = kvm->arch.vcores[i];
>>> if (!vc)
>>> continue;
>>> +
>>> spin_lock(&vc->lock);
>>> - vc->lpcr = (vc->lpcr & ~mask) | lpcr;
>>> + vc->lpcr = kvmppc_filter_lpcr_hv(vc, (vc->lpcr & ~mask) | lpcr);
>>
>> This change seems unnecessary, since kvmppc_update_lpcr is called only
>> to update MMU configuration bits, not as a result of any action by
>> userspace or a nested hypervisor. It's also beyond the scope of what
>> was mentioned in the commit message.
>
> I didn't think it was outside the spirit of the patch, but yes
> only the guest update LPCR case was enumerated. Would it be more
> consistent to add it to the changelog and leave it in here or would
> you prefer it left out until there is a real use?
On second thoughts, I already left at least one other place without
such a check, so I now tend to agree with you. But I instead added
a test that just ensures the host is not out of synch with itself in
terms of what it can set the LPCR to.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 03/46] KVM: PPC: Book3S HV: Disallow LPCR[AIL] to be set to 1 or 2
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin, Fabiano Rosas
These are already disallowed by H_SET_MODE from the guest, also disallow
these by updating LPCR directly.
AIL modes can affect the host interrupt behaviour while the guest LPCR
value is set, so filter it here too.
Suggested-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index c4539c38c639..c5de7e3f22b6 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -803,7 +803,10 @@ static int kvmppc_h_set_mode(struct kvm_vcpu *vcpu, unsigned long mflags,
vcpu->arch.dawrx1 = value2;
return H_SUCCESS;
case H_SET_MODE_RESOURCE_ADDR_TRANS_MODE:
- /* KVM does not support mflags=2 (AIL=2) */
+ /*
+ * KVM does not support mflags=2 (AIL=2) and AIL=1 is reserved.
+ * Keep this in synch with kvmppc_filter_guest_lpcr_hv.
+ */
if (mflags != 0 && mflags != 3)
return H_UNSUPPORTED_FLAG_START;
return H_TOO_HARD;
@@ -1645,6 +1648,8 @@ unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
/* On POWER8 and above, userspace can modify AIL */
if (!cpu_has_feature(CPU_FTR_ARCH_207S))
lpcr &= ~LPCR_AIL;
+ if ((lpcr & LPCR_AIL) != LPCR_AIL_3)
+ lpcr &= ~LPCR_AIL; /* LPCR[AIL]=1/2 is disallowed */
/*
* On POWER9, allow userspace to enable large decrementer for the
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 03/46] KVM: PPC: Book3S HV: Disallow LPCR[AIL] to be set to 1 or 2
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin, Fabiano Rosas
These are already disallowed by H_SET_MODE from the guest, also disallow
these by updating LPCR directly.
AIL modes can affect the host interrupt behaviour while the guest LPCR
value is set, so filter it here too.
Suggested-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index c4539c38c639..c5de7e3f22b6 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -803,7 +803,10 @@ static int kvmppc_h_set_mode(struct kvm_vcpu *vcpu, unsigned long mflags,
vcpu->arch.dawrx1 = value2;
return H_SUCCESS;
case H_SET_MODE_RESOURCE_ADDR_TRANS_MODE:
- /* KVM does not support mflags=2 (AIL=2) */
+ /*
+ * KVM does not support mflags=2 (AIL=2) and AIL=1 is reserved.
+ * Keep this in synch with kvmppc_filter_guest_lpcr_hv.
+ */
if (mflags != 0 && mflags != 3)
return H_UNSUPPORTED_FLAG_START;
return H_TOO_HARD;
@@ -1645,6 +1648,8 @@ unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
/* On POWER8 and above, userspace can modify AIL */
if (!cpu_has_feature(CPU_FTR_ARCH_207S))
lpcr &= ~LPCR_AIL;
+ if ((lpcr & LPCR_AIL) != LPCR_AIL_3)
+ lpcr &= ~LPCR_AIL; /* LPCR[AIL]=1/2 is disallowed */
/*
* On POWER9, allow userspace to enable large decrementer for the
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* Re: [PATCH v4 03/46] KVM: PPC: Book3S HV: Disallow LPCR[AIL] to be set to 1 or 2
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-31 4:28 ` Paul Mackerras
-1 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:28 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc, Fabiano Rosas
On Tue, Mar 23, 2021 at 11:02:22AM +1000, Nicholas Piggin wrote:
> These are already disallowed by H_SET_MODE from the guest, also disallow
> these by updating LPCR directly.
>
> AIL modes can affect the host interrupt behaviour while the guest LPCR
> value is set, so filter it here too.
>
> Suggested-by: Fabiano Rosas <farosas@linux.ibm.com>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 03/46] KVM: PPC: Book3S HV: Disallow LPCR[AIL] to be set to 1 or 2
@ 2021-03-31 4:28 ` Paul Mackerras
0 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:28 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc, Fabiano Rosas
On Tue, Mar 23, 2021 at 11:02:22AM +1000, Nicholas Piggin wrote:
> These are already disallowed by H_SET_MODE from the guest, also disallow
> these by updating LPCR directly.
>
> AIL modes can affect the host interrupt behaviour while the guest LPCR
> value is set, so filter it here too.
>
> Suggested-by: Fabiano Rosas <farosas@linux.ibm.com>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 04/46] KVM: PPC: Book3S HV: Prevent radix guests from setting LPCR[TC]
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
This bit only applies to hash partitions.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 6 ++++++
arch/powerpc/kvm/book3s_hv_nested.c | 3 +--
2 files changed, 7 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index c5de7e3f22b6..1ffb0902e779 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -1645,6 +1645,12 @@ static int kvm_arch_vcpu_ioctl_set_sregs_hv(struct kvm_vcpu *vcpu,
*/
unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
{
+ struct kvm *kvm = vc->kvm;
+
+ /* LPCR_TC only applies to HPT guests */
+ if (kvm_is_radix(kvm))
+ lpcr &= ~LPCR_TC;
+
/* On POWER8 and above, userspace can modify AIL */
if (!cpu_has_feature(CPU_FTR_ARCH_207S))
lpcr &= ~LPCR_AIL;
diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
index f7b441b3eb17..851e3f527eb2 100644
--- a/arch/powerpc/kvm/book3s_hv_nested.c
+++ b/arch/powerpc/kvm/book3s_hv_nested.c
@@ -140,8 +140,7 @@ static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
/*
* Don't let L1 change LPCR bits for the L2 except these:
*/
- mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
- LPCR_LPES | LPCR_MER;
+ mask = LPCR_DPFD | LPCR_ILE | LPCR_AIL | LPCR_LD | LPCR_LPES | LPCR_MER;
hr->lpcr = kvmppc_filter_lpcr_hv(vc,
(vc->lpcr & ~mask) | (hr->lpcr & mask));
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 04/46] KVM: PPC: Book3S HV: Prevent radix guests from setting LPCR[TC]
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
This bit only applies to hash partitions.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 6 ++++++
arch/powerpc/kvm/book3s_hv_nested.c | 3 +--
2 files changed, 7 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index c5de7e3f22b6..1ffb0902e779 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -1645,6 +1645,12 @@ static int kvm_arch_vcpu_ioctl_set_sregs_hv(struct kvm_vcpu *vcpu,
*/
unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
{
+ struct kvm *kvm = vc->kvm;
+
+ /* LPCR_TC only applies to HPT guests */
+ if (kvm_is_radix(kvm))
+ lpcr &= ~LPCR_TC;
+
/* On POWER8 and above, userspace can modify AIL */
if (!cpu_has_feature(CPU_FTR_ARCH_207S))
lpcr &= ~LPCR_AIL;
diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
index f7b441b3eb17..851e3f527eb2 100644
--- a/arch/powerpc/kvm/book3s_hv_nested.c
+++ b/arch/powerpc/kvm/book3s_hv_nested.c
@@ -140,8 +140,7 @@ static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
/*
* Don't let L1 change LPCR bits for the L2 except these:
*/
- mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
- LPCR_LPES | LPCR_MER;
+ mask = LPCR_DPFD | LPCR_ILE | LPCR_AIL | LPCR_LD | LPCR_LPES | LPCR_MER;
hr->lpcr = kvmppc_filter_lpcr_hv(vc,
(vc->lpcr & ~mask) | (hr->lpcr & mask));
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* Re: [PATCH v4 04/46] KVM: PPC: Book3S HV: Prevent radix guests from setting LPCR[TC]
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 8:36 ` Alexey Kardashevskiy
-1 siblings, 0 replies; 188+ messages in thread
From: Alexey Kardashevskiy @ 2021-03-23 8:36 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 23/03/2021 12:02, Nicholas Piggin wrote:
> This bit only applies to hash partitions.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
> ---
> arch/powerpc/kvm/book3s_hv.c | 6 ++++++
> arch/powerpc/kvm/book3s_hv_nested.c | 3 +--
> 2 files changed, 7 insertions(+), 2 deletions(-)
>
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index c5de7e3f22b6..1ffb0902e779 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1645,6 +1645,12 @@ static int kvm_arch_vcpu_ioctl_set_sregs_hv(struct kvm_vcpu *vcpu,
> */
> unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
> {
> + struct kvm *kvm = vc->kvm;
> +
> + /* LPCR_TC only applies to HPT guests */
> + if (kvm_is_radix(kvm))
> + lpcr &= ~LPCR_TC;
> +
> /* On POWER8 and above, userspace can modify AIL */
> if (!cpu_has_feature(CPU_FTR_ARCH_207S))
> lpcr &= ~LPCR_AIL;
> diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
> index f7b441b3eb17..851e3f527eb2 100644
> --- a/arch/powerpc/kvm/book3s_hv_nested.c
> +++ b/arch/powerpc/kvm/book3s_hv_nested.c
> @@ -140,8 +140,7 @@ static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
> /*
> * Don't let L1 change LPCR bits for the L2 except these:
> */
> - mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
> - LPCR_LPES | LPCR_MER;
> + mask = LPCR_DPFD | LPCR_ILE | LPCR_AIL | LPCR_LD | LPCR_LPES | LPCR_MER;
> hr->lpcr = kvmppc_filter_lpcr_hv(vc,
> (vc->lpcr & ~mask) | (hr->lpcr & mask));
>
>
--
Alexey
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 04/46] KVM: PPC: Book3S HV: Prevent radix guests from setting LPCR[TC]
@ 2021-03-23 8:36 ` Alexey Kardashevskiy
0 siblings, 0 replies; 188+ messages in thread
From: Alexey Kardashevskiy @ 2021-03-23 8:36 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 23/03/2021 12:02, Nicholas Piggin wrote:
> This bit only applies to hash partitions.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
> ---
> arch/powerpc/kvm/book3s_hv.c | 6 ++++++
> arch/powerpc/kvm/book3s_hv_nested.c | 3 +--
> 2 files changed, 7 insertions(+), 2 deletions(-)
>
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index c5de7e3f22b6..1ffb0902e779 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1645,6 +1645,12 @@ static int kvm_arch_vcpu_ioctl_set_sregs_hv(struct kvm_vcpu *vcpu,
> */
> unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
> {
> + struct kvm *kvm = vc->kvm;
> +
> + /* LPCR_TC only applies to HPT guests */
> + if (kvm_is_radix(kvm))
> + lpcr &= ~LPCR_TC;
> +
> /* On POWER8 and above, userspace can modify AIL */
> if (!cpu_has_feature(CPU_FTR_ARCH_207S))
> lpcr &= ~LPCR_AIL;
> diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
> index f7b441b3eb17..851e3f527eb2 100644
> --- a/arch/powerpc/kvm/book3s_hv_nested.c
> +++ b/arch/powerpc/kvm/book3s_hv_nested.c
> @@ -140,8 +140,7 @@ static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
> /*
> * Don't let L1 change LPCR bits for the L2 except these:
> */
> - mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
> - LPCR_LPES | LPCR_MER;
> + mask = LPCR_DPFD | LPCR_ILE | LPCR_AIL | LPCR_LD | LPCR_LPES | LPCR_MER;
> hr->lpcr = kvmppc_filter_lpcr_hv(vc,
> (vc->lpcr & ~mask) | (hr->lpcr & mask));
>
>
--
Alexey
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 04/46] KVM: PPC: Book3S HV: Prevent radix guests from setting LPCR[TC]
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-31 4:34 ` Paul Mackerras
-1 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:34 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc
On Tue, Mar 23, 2021 at 11:02:23AM +1000, Nicholas Piggin wrote:
> This bit only applies to hash partitions.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
> arch/powerpc/kvm/book3s_hv.c | 6 ++++++
> arch/powerpc/kvm/book3s_hv_nested.c | 3 +--
> 2 files changed, 7 insertions(+), 2 deletions(-)
>
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index c5de7e3f22b6..1ffb0902e779 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1645,6 +1645,12 @@ static int kvm_arch_vcpu_ioctl_set_sregs_hv(struct kvm_vcpu *vcpu,
> */
> unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
> {
> + struct kvm *kvm = vc->kvm;
> +
> + /* LPCR_TC only applies to HPT guests */
> + if (kvm_is_radix(kvm))
> + lpcr &= ~LPCR_TC;
I'm not sure I see any benefit from this, and it is a little extra
complexity.
> /* On POWER8 and above, userspace can modify AIL */
> if (!cpu_has_feature(CPU_FTR_ARCH_207S))
> lpcr &= ~LPCR_AIL;
> diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
> index f7b441b3eb17..851e3f527eb2 100644
> --- a/arch/powerpc/kvm/book3s_hv_nested.c
> +++ b/arch/powerpc/kvm/book3s_hv_nested.c
> @@ -140,8 +140,7 @@ static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
> /*
> * Don't let L1 change LPCR bits for the L2 except these:
> */
> - mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
> - LPCR_LPES | LPCR_MER;
> + mask = LPCR_DPFD | LPCR_ILE | LPCR_AIL | LPCR_LD | LPCR_LPES | LPCR_MER;
Doesn't this make it completely impossible to set TC for any guest?
Paul.
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 04/46] KVM: PPC: Book3S HV: Prevent radix guests from setting LPCR[TC]
@ 2021-03-31 4:34 ` Paul Mackerras
0 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:34 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc
On Tue, Mar 23, 2021 at 11:02:23AM +1000, Nicholas Piggin wrote:
> This bit only applies to hash partitions.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
> arch/powerpc/kvm/book3s_hv.c | 6 ++++++
> arch/powerpc/kvm/book3s_hv_nested.c | 3 +--
> 2 files changed, 7 insertions(+), 2 deletions(-)
>
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index c5de7e3f22b6..1ffb0902e779 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1645,6 +1645,12 @@ static int kvm_arch_vcpu_ioctl_set_sregs_hv(struct kvm_vcpu *vcpu,
> */
> unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
> {
> + struct kvm *kvm = vc->kvm;
> +
> + /* LPCR_TC only applies to HPT guests */
> + if (kvm_is_radix(kvm))
> + lpcr &= ~LPCR_TC;
I'm not sure I see any benefit from this, and it is a little extra
complexity.
> /* On POWER8 and above, userspace can modify AIL */
> if (!cpu_has_feature(CPU_FTR_ARCH_207S))
> lpcr &= ~LPCR_AIL;
> diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
> index f7b441b3eb17..851e3f527eb2 100644
> --- a/arch/powerpc/kvm/book3s_hv_nested.c
> +++ b/arch/powerpc/kvm/book3s_hv_nested.c
> @@ -140,8 +140,7 @@ static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
> /*
> * Don't let L1 change LPCR bits for the L2 except these:
> */
> - mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
> - LPCR_LPES | LPCR_MER;
> + mask = LPCR_DPFD | LPCR_ILE | LPCR_AIL | LPCR_LD | LPCR_LPES | LPCR_MER;
Doesn't this make it completely impossible to set TC for any guest?
Paul.
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 04/46] KVM: PPC: Book3S HV: Prevent radix guests from setting LPCR[TC]
2021-03-31 4:34 ` Paul Mackerras
@ 2021-04-01 9:45 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 9:45 UTC (permalink / raw)
To: Paul Mackerras; +Cc: linuxppc-dev, kvm-ppc
Excerpts from Paul Mackerras's message of March 31, 2021 2:34 pm:
> On Tue, Mar 23, 2021 at 11:02:23AM +1000, Nicholas Piggin wrote:
>> This bit only applies to hash partitions.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>> ---
>> arch/powerpc/kvm/book3s_hv.c | 6 ++++++
>> arch/powerpc/kvm/book3s_hv_nested.c | 3 +--
>> 2 files changed, 7 insertions(+), 2 deletions(-)
>>
>> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
>> index c5de7e3f22b6..1ffb0902e779 100644
>> --- a/arch/powerpc/kvm/book3s_hv.c
>> +++ b/arch/powerpc/kvm/book3s_hv.c
>> @@ -1645,6 +1645,12 @@ static int kvm_arch_vcpu_ioctl_set_sregs_hv(struct kvm_vcpu *vcpu,
>> */
>> unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
>> {
>> + struct kvm *kvm = vc->kvm;
>> +
>> + /* LPCR_TC only applies to HPT guests */
>> + if (kvm_is_radix(kvm))
>> + lpcr &= ~LPCR_TC;
>
> I'm not sure I see any benefit from this, and it is a little extra
> complexity.
Principle of allowing a guest to mess with as little HV state as
possible (but not littler), which I think is a good one to follow.
>
>> /* On POWER8 and above, userspace can modify AIL */
>> if (!cpu_has_feature(CPU_FTR_ARCH_207S))
>> lpcr &= ~LPCR_AIL;
>> diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
>> index f7b441b3eb17..851e3f527eb2 100644
>> --- a/arch/powerpc/kvm/book3s_hv_nested.c
>> +++ b/arch/powerpc/kvm/book3s_hv_nested.c
>> @@ -140,8 +140,7 @@ static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
>> /*
>> * Don't let L1 change LPCR bits for the L2 except these:
>> */
>> - mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
>> - LPCR_LPES | LPCR_MER;
>> + mask = LPCR_DPFD | LPCR_ILE | LPCR_AIL | LPCR_LD | LPCR_LPES | LPCR_MER;
>
> Doesn't this make it completely impossible to set TC for any guest?
Argh, yes Fabiano pointed this out in an earlier rev and I didn't
fix this hunk after adding the filter bit. Thanks.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 04/46] KVM: PPC: Book3S HV: Prevent radix guests from setting LPCR[TC]
@ 2021-04-01 9:45 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 9:45 UTC (permalink / raw)
To: Paul Mackerras; +Cc: linuxppc-dev, kvm-ppc
Excerpts from Paul Mackerras's message of March 31, 2021 2:34 pm:
> On Tue, Mar 23, 2021 at 11:02:23AM +1000, Nicholas Piggin wrote:
>> This bit only applies to hash partitions.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>> ---
>> arch/powerpc/kvm/book3s_hv.c | 6 ++++++
>> arch/powerpc/kvm/book3s_hv_nested.c | 3 +--
>> 2 files changed, 7 insertions(+), 2 deletions(-)
>>
>> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
>> index c5de7e3f22b6..1ffb0902e779 100644
>> --- a/arch/powerpc/kvm/book3s_hv.c
>> +++ b/arch/powerpc/kvm/book3s_hv.c
>> @@ -1645,6 +1645,12 @@ static int kvm_arch_vcpu_ioctl_set_sregs_hv(struct kvm_vcpu *vcpu,
>> */
>> unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
>> {
>> + struct kvm *kvm = vc->kvm;
>> +
>> + /* LPCR_TC only applies to HPT guests */
>> + if (kvm_is_radix(kvm))
>> + lpcr &= ~LPCR_TC;
>
> I'm not sure I see any benefit from this, and it is a little extra
> complexity.
Principle of allowing a guest to mess with as little HV state as
possible (but not littler), which I think is a good one to follow.
>
>> /* On POWER8 and above, userspace can modify AIL */
>> if (!cpu_has_feature(CPU_FTR_ARCH_207S))
>> lpcr &= ~LPCR_AIL;
>> diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
>> index f7b441b3eb17..851e3f527eb2 100644
>> --- a/arch/powerpc/kvm/book3s_hv_nested.c
>> +++ b/arch/powerpc/kvm/book3s_hv_nested.c
>> @@ -140,8 +140,7 @@ static void sanitise_hv_regs(struct kvm_vcpu *vcpu, struct hv_guest_state *hr)
>> /*
>> * Don't let L1 change LPCR bits for the L2 except these:
>> */
>> - mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
>> - LPCR_LPES | LPCR_MER;
>> + mask = LPCR_DPFD | LPCR_ILE | LPCR_AIL | LPCR_LD | LPCR_LPES | LPCR_MER;
>
> Doesn't this make it completely impossible to set TC for any guest?
Argh, yes Fabiano pointed this out in an earlier rev and I didn't
fix this hunk after adding the filter bit. Thanks.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 05/46] KVM: PPC: Book3S HV: Remove redundant mtspr PSPB
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Daniel Axtens, linuxppc-dev, Nicholas Piggin, Fabiano Rosas
This SPR is set to 0 twice when exiting the guest.
Suggested-by: Fabiano Rosas <farosas@linux.ibm.com>
Reviewed-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 1ffb0902e779..7cfaabab2c20 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3781,7 +3781,6 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
mtspr(SPRN_DSCR, host_dscr);
mtspr(SPRN_TIDR, host_tidr);
mtspr(SPRN_IAMR, host_iamr);
- mtspr(SPRN_PSPB, 0);
if (host_amr != vcpu->arch.amr)
mtspr(SPRN_AMR, host_amr);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 05/46] KVM: PPC: Book3S HV: Remove redundant mtspr PSPB
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Daniel Axtens, linuxppc-dev, Nicholas Piggin, Fabiano Rosas
This SPR is set to 0 twice when exiting the guest.
Suggested-by: Fabiano Rosas <farosas@linux.ibm.com>
Reviewed-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 1ffb0902e779..7cfaabab2c20 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3781,7 +3781,6 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
mtspr(SPRN_DSCR, host_dscr);
mtspr(SPRN_TIDR, host_tidr);
mtspr(SPRN_IAMR, host_iamr);
- mtspr(SPRN_PSPB, 0);
if (host_amr != vcpu->arch.amr)
mtspr(SPRN_AMR, host_amr);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* Re: [PATCH v4 05/46] KVM: PPC: Book3S HV: Remove redundant mtspr PSPB
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-31 4:35 ` Paul Mackerras
-1 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:35 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: Daniel Axtens, linuxppc-dev, kvm-ppc, Fabiano Rosas
On Tue, Mar 23, 2021 at 11:02:24AM +1000, Nicholas Piggin wrote:
> This SPR is set to 0 twice when exiting the guest.
>
> Suggested-by: Fabiano Rosas <farosas@linux.ibm.com>
> Reviewed-by: Daniel Axtens <dja@axtens.net>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 05/46] KVM: PPC: Book3S HV: Remove redundant mtspr PSPB
@ 2021-03-31 4:35 ` Paul Mackerras
0 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:35 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: Daniel Axtens, linuxppc-dev, kvm-ppc, Fabiano Rosas
On Tue, Mar 23, 2021 at 11:02:24AM +1000, Nicholas Piggin wrote:
> This SPR is set to 0 twice when exiting the guest.
>
> Suggested-by: Fabiano Rosas <farosas@linux.ibm.com>
> Reviewed-by: Daniel Axtens <dja@axtens.net>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 06/46] KVM: PPC: Book3S HV: remove unused kvmppc_h_protect argument
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin, Daniel Axtens
The va argument is not used in the function or set by its asm caller,
so remove it to be safe.
Reviewed-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_ppc.h | 3 +--
arch/powerpc/kvm/book3s_hv_rm_mmu.c | 3 +--
2 files changed, 2 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
index 8aacd76bb702..9531b1c1b190 100644
--- a/arch/powerpc/include/asm/kvm_ppc.h
+++ b/arch/powerpc/include/asm/kvm_ppc.h
@@ -767,8 +767,7 @@ long kvmppc_h_remove(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long pte_index, unsigned long avpn);
long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu);
long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags,
- unsigned long pte_index, unsigned long avpn,
- unsigned long va);
+ unsigned long pte_index, unsigned long avpn);
long kvmppc_h_read(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long pte_index);
long kvmppc_h_clear_ref(struct kvm_vcpu *vcpu, unsigned long flags,
diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
index 88da2764c1bb..7af7c70f1468 100644
--- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c
+++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
@@ -673,8 +673,7 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
}
long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags,
- unsigned long pte_index, unsigned long avpn,
- unsigned long va)
+ unsigned long pte_index, unsigned long avpn)
{
struct kvm *kvm = vcpu->kvm;
__be64 *hpte;
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 06/46] KVM: PPC: Book3S HV: remove unused kvmppc_h_protect argument
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin, Daniel Axtens
The va argument is not used in the function or set by its asm caller,
so remove it to be safe.
Reviewed-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_ppc.h | 3 +--
arch/powerpc/kvm/book3s_hv_rm_mmu.c | 3 +--
2 files changed, 2 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
index 8aacd76bb702..9531b1c1b190 100644
--- a/arch/powerpc/include/asm/kvm_ppc.h
+++ b/arch/powerpc/include/asm/kvm_ppc.h
@@ -767,8 +767,7 @@ long kvmppc_h_remove(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long pte_index, unsigned long avpn);
long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu);
long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags,
- unsigned long pte_index, unsigned long avpn,
- unsigned long va);
+ unsigned long pte_index, unsigned long avpn);
long kvmppc_h_read(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long pte_index);
long kvmppc_h_clear_ref(struct kvm_vcpu *vcpu, unsigned long flags,
diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
index 88da2764c1bb..7af7c70f1468 100644
--- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c
+++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
@@ -673,8 +673,7 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
}
long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags,
- unsigned long pte_index, unsigned long avpn,
- unsigned long va)
+ unsigned long pte_index, unsigned long avpn)
{
struct kvm *kvm = vcpu->kvm;
__be64 *hpte;
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* Re: [PATCH v4 06/46] KVM: PPC: Book3S HV: remove unused kvmppc_h_protect argument
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-31 4:38 ` Paul Mackerras
-1 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:38 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc, Daniel Axtens
On Tue, Mar 23, 2021 at 11:02:25AM +1000, Nicholas Piggin wrote:
> The va argument is not used in the function or set by its asm caller,
> so remove it to be safe.
>
> Reviewed-by: Daniel Axtens <dja@axtens.net>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 06/46] KVM: PPC: Book3S HV: remove unused kvmppc_h_protect argument
@ 2021-03-31 4:38 ` Paul Mackerras
0 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:38 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc, Daniel Axtens
On Tue, Mar 23, 2021 at 11:02:25AM +1000, Nicholas Piggin wrote:
> The va argument is not used in the function or set by its asm caller,
> so remove it to be safe.
>
> Reviewed-by: Daniel Axtens <dja@axtens.net>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 07/46] KVM: PPC: Book3S HV: Fix CONFIG_SPAPR_TCE_IOMMU=n default hcalls
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin, Daniel Axtens
This config option causes the warning in init_default_hcalls to fire
because the TCE handlers are in the default hcall list but not
implemented.
Reviewed-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 7cfaabab2c20..7bb4222729b0 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -5401,8 +5401,10 @@ static unsigned int default_hcall_list[] = {
H_READ,
H_PROTECT,
H_BULK_REMOVE,
+#ifdef CONFIG_SPAPR_TCE_IOMMU
H_GET_TCE,
H_PUT_TCE,
+#endif
H_SET_DABR,
H_SET_XDABR,
H_CEDE,
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 07/46] KVM: PPC: Book3S HV: Fix CONFIG_SPAPR_TCE_IOMMU=n default hcalls
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin, Daniel Axtens
This config option causes the warning in init_default_hcalls to fire
because the TCE handlers are in the default hcall list but not
implemented.
Reviewed-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 7cfaabab2c20..7bb4222729b0 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -5401,8 +5401,10 @@ static unsigned int default_hcall_list[] = {
H_READ,
H_PROTECT,
H_BULK_REMOVE,
+#ifdef CONFIG_SPAPR_TCE_IOMMU
H_GET_TCE,
H_PUT_TCE,
+#endif
H_SET_DABR,
H_SET_XDABR,
H_CEDE,
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* Re: [PATCH v4 07/46] KVM: PPC: Book3S HV: Fix CONFIG_SPAPR_TCE_IOMMU=n default hcalls
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-31 4:39 ` Paul Mackerras
-1 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:39 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc, Daniel Axtens
On Tue, Mar 23, 2021 at 11:02:26AM +1000, Nicholas Piggin wrote:
> This config option causes the warning in init_default_hcalls to fire
> because the TCE handlers are in the default hcall list but not
> implemented.
>
> Reviewed-by: Daniel Axtens <dja@axtens.net>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 07/46] KVM: PPC: Book3S HV: Fix CONFIG_SPAPR_TCE_IOMMU=n default hcalls
@ 2021-03-31 4:39 ` Paul Mackerras
0 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:39 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc, Daniel Axtens
On Tue, Mar 23, 2021 at 11:02:26AM +1000, Nicholas Piggin wrote:
> This config option causes the warning in init_default_hcalls to fire
> because the TCE handlers are in the default hcall list but not
> implemented.
>
> Reviewed-by: Daniel Axtens <dja@axtens.net>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 08/46] powerpc/64s: Remove KVM handler support from CBE_RAS interrupts
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin, Fabiano Rosas
Cell does not support KVM.
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kernel/exceptions-64s.S | 6 ------
1 file changed, 6 deletions(-)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 8082b690e874..a0515cb829c2 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -2530,8 +2530,6 @@ EXC_VIRT_NONE(0x5100, 0x100)
INT_DEFINE_BEGIN(cbe_system_error)
IVEC=0x1200
IHSRR=1
- IKVM_SKIP=1
- IKVM_REAL=1
INT_DEFINE_END(cbe_system_error)
EXC_REAL_BEGIN(cbe_system_error, 0x1200, 0x100)
@@ -2701,8 +2699,6 @@ EXC_COMMON_BEGIN(denorm_exception_common)
INT_DEFINE_BEGIN(cbe_maintenance)
IVEC=0x1600
IHSRR=1
- IKVM_SKIP=1
- IKVM_REAL=1
INT_DEFINE_END(cbe_maintenance)
EXC_REAL_BEGIN(cbe_maintenance, 0x1600, 0x100)
@@ -2754,8 +2750,6 @@ EXC_COMMON_BEGIN(altivec_assist_common)
INT_DEFINE_BEGIN(cbe_thermal)
IVEC=0x1800
IHSRR=1
- IKVM_SKIP=1
- IKVM_REAL=1
INT_DEFINE_END(cbe_thermal)
EXC_REAL_BEGIN(cbe_thermal, 0x1800, 0x100)
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 08/46] powerpc/64s: Remove KVM handler support from CBE_RAS interrupts
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin, Fabiano Rosas
Cell does not support KVM.
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kernel/exceptions-64s.S | 6 ------
1 file changed, 6 deletions(-)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 8082b690e874..a0515cb829c2 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -2530,8 +2530,6 @@ EXC_VIRT_NONE(0x5100, 0x100)
INT_DEFINE_BEGIN(cbe_system_error)
IVEC=0x1200
IHSRR=1
- IKVM_SKIP=1
- IKVM_REAL=1
INT_DEFINE_END(cbe_system_error)
EXC_REAL_BEGIN(cbe_system_error, 0x1200, 0x100)
@@ -2701,8 +2699,6 @@ EXC_COMMON_BEGIN(denorm_exception_common)
INT_DEFINE_BEGIN(cbe_maintenance)
IVEC=0x1600
IHSRR=1
- IKVM_SKIP=1
- IKVM_REAL=1
INT_DEFINE_END(cbe_maintenance)
EXC_REAL_BEGIN(cbe_maintenance, 0x1600, 0x100)
@@ -2754,8 +2750,6 @@ EXC_COMMON_BEGIN(altivec_assist_common)
INT_DEFINE_BEGIN(cbe_thermal)
IVEC=0x1800
IHSRR=1
- IKVM_SKIP=1
- IKVM_REAL=1
INT_DEFINE_END(cbe_thermal)
EXC_REAL_BEGIN(cbe_thermal, 0x1800, 0x100)
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* Re: [PATCH v4 08/46] powerpc/64s: Remove KVM handler support from CBE_RAS interrupts
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-31 4:41 ` Paul Mackerras
-1 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:41 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc, Fabiano Rosas
On Tue, Mar 23, 2021 at 11:02:27AM +1000, Nicholas Piggin wrote:
> Cell does not support KVM.
>
> Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 08/46] powerpc/64s: Remove KVM handler support from CBE_RAS interrupts
@ 2021-03-31 4:41 ` Paul Mackerras
0 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:41 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc, Fabiano Rosas
On Tue, Mar 23, 2021 at 11:02:27AM +1000, Nicholas Piggin wrote:
> Cell does not support KVM.
>
> Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 09/46] powerpc/64s: remove KVM SKIP test from instruction breakpoint handler
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Fabiano Rosas, linuxppc-dev, Nicholas Piggin, Daniel Axtens
The code being executed in KVM_GUEST_MODE_SKIP is hypervisor code with
MSR[IR]=0, so the faults of concern are the d-side ones caused by access
to guest context by the hypervisor.
Instruction breakpoint interrupts are not a concern here. It's unlikely
any good would come of causing breaks in this code, but skipping the
instruction that caused it won't help matters (e.g., skip the mtmsr that
sets MSR[DR]=0 or clears KVM_GUEST_MODE_SKIP).
Reviewed-by: Daniel Axtens <dja@axtens.net>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kernel/exceptions-64s.S | 1 -
1 file changed, 1 deletion(-)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index a0515cb829c2..c9c446ccff54 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -2553,7 +2553,6 @@ EXC_VIRT_NONE(0x5200, 0x100)
INT_DEFINE_BEGIN(instruction_breakpoint)
IVEC=0x1300
#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
- IKVM_SKIP=1
IKVM_REAL=1
#endif
INT_DEFINE_END(instruction_breakpoint)
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 09/46] powerpc/64s: remove KVM SKIP test from instruction breakpoint handler
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Fabiano Rosas, linuxppc-dev, Nicholas Piggin, Daniel Axtens
The code being executed in KVM_GUEST_MODE_SKIP is hypervisor code with
MSR[IR]=0, so the faults of concern are the d-side ones caused by access
to guest context by the hypervisor.
Instruction breakpoint interrupts are not a concern here. It's unlikely
any good would come of causing breaks in this code, but skipping the
instruction that caused it won't help matters (e.g., skip the mtmsr that
sets MSR[DR]=0 or clears KVM_GUEST_MODE_SKIP).
Reviewed-by: Daniel Axtens <dja@axtens.net>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kernel/exceptions-64s.S | 1 -
1 file changed, 1 deletion(-)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index a0515cb829c2..c9c446ccff54 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -2553,7 +2553,6 @@ EXC_VIRT_NONE(0x5200, 0x100)
INT_DEFINE_BEGIN(instruction_breakpoint)
IVEC=0x1300
#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
- IKVM_SKIP=1
IKVM_REAL=1
#endif
INT_DEFINE_END(instruction_breakpoint)
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* Re: [PATCH v4 09/46] powerpc/64s: remove KVM SKIP test from instruction breakpoint handler
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-31 4:54 ` Paul Mackerras
-1 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:54 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: Fabiano Rosas, linuxppc-dev, kvm-ppc, Daniel Axtens
On Tue, Mar 23, 2021 at 11:02:28AM +1000, Nicholas Piggin wrote:
> The code being executed in KVM_GUEST_MODE_SKIP is hypervisor code with
> MSR[IR]=0, so the faults of concern are the d-side ones caused by access
> to guest context by the hypervisor.
>
> Instruction breakpoint interrupts are not a concern here. It's unlikely
> any good would come of causing breaks in this code, but skipping the
> instruction that caused it won't help matters (e.g., skip the mtmsr that
> sets MSR[DR]=0 or clears KVM_GUEST_MODE_SKIP).
>
> Reviewed-by: Daniel Axtens <dja@axtens.net>
> Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
It might be worth noting in the commit message that the 0x1300
interrupt was dropped from the architecture a long time ago and is not
generated by P7, P8, P9 or P10.
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 09/46] powerpc/64s: remove KVM SKIP test from instruction breakpoint handler
@ 2021-03-31 4:54 ` Paul Mackerras
0 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:54 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: Fabiano Rosas, linuxppc-dev, kvm-ppc, Daniel Axtens
On Tue, Mar 23, 2021 at 11:02:28AM +1000, Nicholas Piggin wrote:
> The code being executed in KVM_GUEST_MODE_SKIP is hypervisor code with
> MSR[IR]=0, so the faults of concern are the d-side ones caused by access
> to guest context by the hypervisor.
>
> Instruction breakpoint interrupts are not a concern here. It's unlikely
> any good would come of causing breaks in this code, but skipping the
> instruction that caused it won't help matters (e.g., skip the mtmsr that
> sets MSR[DR]=0 or clears KVM_GUEST_MODE_SKIP).
>
> Reviewed-by: Daniel Axtens <dja@axtens.net>
> Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
It might be worth noting in the commit message that the 0x1300
interrupt was dropped from the architecture a long time ago and is not
generated by P7, P8, P9 or P10.
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 09/46] powerpc/64s: remove KVM SKIP test from instruction breakpoint handler
2021-03-31 4:54 ` Paul Mackerras
@ 2021-04-01 9:47 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 9:47 UTC (permalink / raw)
To: Paul Mackerras; +Cc: Fabiano Rosas, linuxppc-dev, kvm-ppc, Daniel Axtens
Excerpts from Paul Mackerras's message of March 31, 2021 2:54 pm:
> On Tue, Mar 23, 2021 at 11:02:28AM +1000, Nicholas Piggin wrote:
>> The code being executed in KVM_GUEST_MODE_SKIP is hypervisor code with
>> MSR[IR]=0, so the faults of concern are the d-side ones caused by access
>> to guest context by the hypervisor.
>>
>> Instruction breakpoint interrupts are not a concern here. It's unlikely
>> any good would come of causing breaks in this code, but skipping the
>> instruction that caused it won't help matters (e.g., skip the mtmsr that
>> sets MSR[DR]=0 or clears KVM_GUEST_MODE_SKIP).
>>
>> Reviewed-by: Daniel Axtens <dja@axtens.net>
>> Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>
> It might be worth noting in the commit message that the 0x1300
> interrupt was dropped from the architecture a long time ago and is not
> generated by P7, P8, P9 or P10.
Good background, I'll add that.
Thanks,
Nick
>
> Acked-by: Paul Mackerras <paulus@ozlabs.org>
>
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 09/46] powerpc/64s: remove KVM SKIP test from instruction breakpoint handler
@ 2021-04-01 9:47 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 9:47 UTC (permalink / raw)
To: Paul Mackerras; +Cc: Fabiano Rosas, linuxppc-dev, kvm-ppc, Daniel Axtens
Excerpts from Paul Mackerras's message of March 31, 2021 2:54 pm:
> On Tue, Mar 23, 2021 at 11:02:28AM +1000, Nicholas Piggin wrote:
>> The code being executed in KVM_GUEST_MODE_SKIP is hypervisor code with
>> MSR[IR]=0, so the faults of concern are the d-side ones caused by access
>> to guest context by the hypervisor.
>>
>> Instruction breakpoint interrupts are not a concern here. It's unlikely
>> any good would come of causing breaks in this code, but skipping the
>> instruction that caused it won't help matters (e.g., skip the mtmsr that
>> sets MSR[DR]=0 or clears KVM_GUEST_MODE_SKIP).
>>
>> Reviewed-by: Daniel Axtens <dja@axtens.net>
>> Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>
> It might be worth noting in the commit message that the 0x1300
> interrupt was dropped from the architecture a long time ago and is not
> generated by P7, P8, P9 or P10.
Good background, I'll add that.
Thanks,
Nick
>
> Acked-by: Paul Mackerras <paulus@ozlabs.org>
>
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 10/46] KVM: PPC: Book3S HV: Ensure MSR[ME] is always set in guest MSR
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Fabiano Rosas, linuxppc-dev, Nicholas Piggin, Daniel Axtens
Rather than add the ME bit to the MSR at guest entry, make it clear
that the hypervisor does not allow the guest to clear the bit.
The ME set is kept in guest entry for now, but a future patch will
warn if it's not present.
Reviewed-by: Daniel Axtens <dja@axtens.net>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_builtin.c | 3 +++
arch/powerpc/kvm/book3s_hv_nested.c | 4 +++-
2 files changed, 6 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/kvm/book3s_hv_builtin.c b/arch/powerpc/kvm/book3s_hv_builtin.c
index 158d309b42a3..41cb03d0bde4 100644
--- a/arch/powerpc/kvm/book3s_hv_builtin.c
+++ b/arch/powerpc/kvm/book3s_hv_builtin.c
@@ -662,6 +662,9 @@ static void kvmppc_end_cede(struct kvm_vcpu *vcpu)
void kvmppc_set_msr_hv(struct kvm_vcpu *vcpu, u64 msr)
{
+ /* Guest must always run with ME enabled. */
+ msr = msr | MSR_ME;
+
/*
* Check for illegal transactional state bit combination
* and if we find it, force the TS field to a safe state.
diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
index 851e3f527eb2..886c7fa86add 100644
--- a/arch/powerpc/kvm/book3s_hv_nested.c
+++ b/arch/powerpc/kvm/book3s_hv_nested.c
@@ -328,7 +328,9 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
vcpu->arch.nested = l2;
vcpu->arch.nested_vcpu_id = l2_hv.vcpu_token;
vcpu->arch.regs = l2_regs;
- vcpu->arch.shregs.msr = vcpu->arch.regs.msr;
+
+ /* Guest must always run with ME enabled. */
+ vcpu->arch.shregs.msr = vcpu->arch.regs.msr | MSR_ME;
sanitise_hv_regs(vcpu, &l2_hv);
restore_hv_regs(vcpu, &l2_hv);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 10/46] KVM: PPC: Book3S HV: Ensure MSR[ME] is always set in guest MSR
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Fabiano Rosas, linuxppc-dev, Nicholas Piggin, Daniel Axtens
Rather than add the ME bit to the MSR at guest entry, make it clear
that the hypervisor does not allow the guest to clear the bit.
The ME set is kept in guest entry for now, but a future patch will
warn if it's not present.
Reviewed-by: Daniel Axtens <dja@axtens.net>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_builtin.c | 3 +++
arch/powerpc/kvm/book3s_hv_nested.c | 4 +++-
2 files changed, 6 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/kvm/book3s_hv_builtin.c b/arch/powerpc/kvm/book3s_hv_builtin.c
index 158d309b42a3..41cb03d0bde4 100644
--- a/arch/powerpc/kvm/book3s_hv_builtin.c
+++ b/arch/powerpc/kvm/book3s_hv_builtin.c
@@ -662,6 +662,9 @@ static void kvmppc_end_cede(struct kvm_vcpu *vcpu)
void kvmppc_set_msr_hv(struct kvm_vcpu *vcpu, u64 msr)
{
+ /* Guest must always run with ME enabled. */
+ msr = msr | MSR_ME;
+
/*
* Check for illegal transactional state bit combination
* and if we find it, force the TS field to a safe state.
diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
index 851e3f527eb2..886c7fa86add 100644
--- a/arch/powerpc/kvm/book3s_hv_nested.c
+++ b/arch/powerpc/kvm/book3s_hv_nested.c
@@ -328,7 +328,9 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
vcpu->arch.nested = l2;
vcpu->arch.nested_vcpu_id = l2_hv.vcpu_token;
vcpu->arch.regs = l2_regs;
- vcpu->arch.shregs.msr = vcpu->arch.regs.msr;
+
+ /* Guest must always run with ME enabled. */
+ vcpu->arch.shregs.msr = vcpu->arch.regs.msr | MSR_ME;
sanitise_hv_regs(vcpu, &l2_hv);
restore_hv_regs(vcpu, &l2_hv);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* Re: [PATCH v4 10/46] KVM: PPC: Book3S HV: Ensure MSR[ME] is always set in guest MSR
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-31 4:56 ` Paul Mackerras
-1 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:56 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: Fabiano Rosas, linuxppc-dev, kvm-ppc, Daniel Axtens
On Tue, Mar 23, 2021 at 11:02:29AM +1000, Nicholas Piggin wrote:
> Rather than add the ME bit to the MSR at guest entry, make it clear
> that the hypervisor does not allow the guest to clear the bit.
>
> The ME set is kept in guest entry for now, but a future patch will
> warn if it's not present.
>
> Reviewed-by: Daniel Axtens <dja@axtens.net>
> Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 10/46] KVM: PPC: Book3S HV: Ensure MSR[ME] is always set in guest MSR
@ 2021-03-31 4:56 ` Paul Mackerras
0 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:56 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: Fabiano Rosas, linuxppc-dev, kvm-ppc, Daniel Axtens
On Tue, Mar 23, 2021 at 11:02:29AM +1000, Nicholas Piggin wrote:
> Rather than add the ME bit to the MSR at guest entry, make it clear
> that the hypervisor does not allow the guest to clear the bit.
>
> The ME set is kept in guest entry for now, but a future patch will
> warn if it's not present.
>
> Reviewed-by: Daniel Axtens <dja@axtens.net>
> Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 11/46] KVM: PPC: Book3S HV: Ensure MSR[HV] is always clear in guest MSR
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Rather than clear the HV bit from the MSR at guest entry, make it clear
that the hypervisor does not allow the guest to set the bit.
The HV clear is kept in guest entry for now, but a future patch will
warn if it's not present.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_builtin.c | 4 ++--
arch/powerpc/kvm/book3s_hv_nested.c | 4 ++--
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_builtin.c b/arch/powerpc/kvm/book3s_hv_builtin.c
index 41cb03d0bde4..7a0e33a9c980 100644
--- a/arch/powerpc/kvm/book3s_hv_builtin.c
+++ b/arch/powerpc/kvm/book3s_hv_builtin.c
@@ -662,8 +662,8 @@ static void kvmppc_end_cede(struct kvm_vcpu *vcpu)
void kvmppc_set_msr_hv(struct kvm_vcpu *vcpu, u64 msr)
{
- /* Guest must always run with ME enabled. */
- msr = msr | MSR_ME;
+ /* Guest must always run with ME enabled, HV disabled. */
+ msr = (msr | MSR_ME) & ~MSR_HV;
/*
* Check for illegal transactional state bit combination
diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
index 886c7fa86add..d192e799c0af 100644
--- a/arch/powerpc/kvm/book3s_hv_nested.c
+++ b/arch/powerpc/kvm/book3s_hv_nested.c
@@ -329,8 +329,8 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
vcpu->arch.nested_vcpu_id = l2_hv.vcpu_token;
vcpu->arch.regs = l2_regs;
- /* Guest must always run with ME enabled. */
- vcpu->arch.shregs.msr = vcpu->arch.regs.msr | MSR_ME;
+ /* Guest must always run with ME enabled, HV disabled. */
+ vcpu->arch.shregs.msr = (vcpu->arch.regs.msr | MSR_ME) & ~MSR_HV;
sanitise_hv_regs(vcpu, &l2_hv);
restore_hv_regs(vcpu, &l2_hv);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 11/46] KVM: PPC: Book3S HV: Ensure MSR[HV] is always clear in guest MSR
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Rather than clear the HV bit from the MSR at guest entry, make it clear
that the hypervisor does not allow the guest to set the bit.
The HV clear is kept in guest entry for now, but a future patch will
warn if it's not present.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_builtin.c | 4 ++--
arch/powerpc/kvm/book3s_hv_nested.c | 4 ++--
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_builtin.c b/arch/powerpc/kvm/book3s_hv_builtin.c
index 41cb03d0bde4..7a0e33a9c980 100644
--- a/arch/powerpc/kvm/book3s_hv_builtin.c
+++ b/arch/powerpc/kvm/book3s_hv_builtin.c
@@ -662,8 +662,8 @@ static void kvmppc_end_cede(struct kvm_vcpu *vcpu)
void kvmppc_set_msr_hv(struct kvm_vcpu *vcpu, u64 msr)
{
- /* Guest must always run with ME enabled. */
- msr = msr | MSR_ME;
+ /* Guest must always run with ME enabled, HV disabled. */
+ msr = (msr | MSR_ME) & ~MSR_HV;
/*
* Check for illegal transactional state bit combination
diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
index 886c7fa86add..d192e799c0af 100644
--- a/arch/powerpc/kvm/book3s_hv_nested.c
+++ b/arch/powerpc/kvm/book3s_hv_nested.c
@@ -329,8 +329,8 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
vcpu->arch.nested_vcpu_id = l2_hv.vcpu_token;
vcpu->arch.regs = l2_regs;
- /* Guest must always run with ME enabled. */
- vcpu->arch.shregs.msr = vcpu->arch.regs.msr | MSR_ME;
+ /* Guest must always run with ME enabled, HV disabled. */
+ vcpu->arch.shregs.msr = (vcpu->arch.regs.msr | MSR_ME) & ~MSR_HV;
sanitise_hv_regs(vcpu, &l2_hv);
restore_hv_regs(vcpu, &l2_hv);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* Re: [PATCH v4 11/46] KVM: PPC: Book3S HV: Ensure MSR[HV] is always clear in guest MSR
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-31 4:58 ` Paul Mackerras
-1 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:58 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc
On Tue, Mar 23, 2021 at 11:02:30AM +1000, Nicholas Piggin wrote:
> Rather than clear the HV bit from the MSR at guest entry, make it clear
> that the hypervisor does not allow the guest to set the bit.
>
> The HV clear is kept in guest entry for now, but a future patch will
> warn if it's not present.
Will warn if it *is* present, surely?
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 11/46] KVM: PPC: Book3S HV: Ensure MSR[HV] is always clear in guest MSR
@ 2021-03-31 4:58 ` Paul Mackerras
0 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-03-31 4:58 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc
On Tue, Mar 23, 2021 at 11:02:30AM +1000, Nicholas Piggin wrote:
> Rather than clear the HV bit from the MSR at guest entry, make it clear
> that the hypervisor does not allow the guest to set the bit.
>
> The HV clear is kept in guest entry for now, but a future patch will
> warn if it's not present.
Will warn if it *is* present, surely?
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 11/46] KVM: PPC: Book3S HV: Ensure MSR[HV] is always clear in guest MSR
2021-03-31 4:58 ` Paul Mackerras
@ 2021-04-01 9:49 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 9:49 UTC (permalink / raw)
To: Paul Mackerras; +Cc: linuxppc-dev, kvm-ppc
Excerpts from Paul Mackerras's message of March 31, 2021 2:58 pm:
> On Tue, Mar 23, 2021 at 11:02:30AM +1000, Nicholas Piggin wrote:
>> Rather than clear the HV bit from the MSR at guest entry, make it clear
>> that the hypervisor does not allow the guest to set the bit.
>>
>> The HV clear is kept in guest entry for now, but a future patch will
>> warn if it's not present.
>
> Will warn if it *is* present, surely?
Just making sure you were awake, definitely wasn't a copy-paste bug...
Thanks,
Nick
>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>
> Acked-by: Paul Mackerras <paulus@ozlabs.org>
>
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 11/46] KVM: PPC: Book3S HV: Ensure MSR[HV] is always clear in guest MSR
@ 2021-04-01 9:49 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 9:49 UTC (permalink / raw)
To: Paul Mackerras; +Cc: linuxppc-dev, kvm-ppc
Excerpts from Paul Mackerras's message of March 31, 2021 2:58 pm:
> On Tue, Mar 23, 2021 at 11:02:30AM +1000, Nicholas Piggin wrote:
>> Rather than clear the HV bit from the MSR at guest entry, make it clear
>> that the hypervisor does not allow the guest to set the bit.
>>
>> The HV clear is kept in guest entry for now, but a future patch will
>> warn if it's not present.
>
> Will warn if it *is* present, surely?
Just making sure you were awake, definitely wasn't a copy-paste bug...
Thanks,
Nick
>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>
> Acked-by: Paul Mackerras <paulus@ozlabs.org>
>
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 12/46] KVM: PPC: Book3S 64: move KVM interrupt entry to a common entry point
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Fabiano Rosas, linuxppc-dev, Nicholas Piggin, Daniel Axtens
Rather than bifurcate the call depending on whether or not HV is
possible, and have the HV entry test for PR, just make a single
common point which does the demultiplexing. This makes it simpler
to add another type of exit handler.
Reviewed-by: Daniel Axtens <dja@axtens.net>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kernel/exceptions-64s.S | 8 +-----
arch/powerpc/kvm/Makefile | 3 +++
arch/powerpc/kvm/book3s_64_entry.S | 36 +++++++++++++++++++++++++
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 11 ++------
4 files changed, 42 insertions(+), 16 deletions(-)
create mode 100644 arch/powerpc/kvm/book3s_64_entry.S
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index c9c446ccff54..162595af1ac7 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -208,7 +208,6 @@ do_define_int n
.endm
#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
-#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
/*
* All interrupts which set HSRR registers, as well as SRESET and MCE and
* syscall when invoked with "sc 1" switch to MSR[HV]=1 (HVMODE) to be taken,
@@ -238,13 +237,8 @@ do_define_int n
/*
* If an interrupt is taken while a guest is running, it is immediately routed
- * to KVM to handle. If both HV and PR KVM arepossible, KVM interrupts go first
- * to kvmppc_interrupt_hv, which handles the PR guest case.
+ * to KVM to handle.
*/
-#define kvmppc_interrupt kvmppc_interrupt_hv
-#else
-#define kvmppc_interrupt kvmppc_interrupt_pr
-#endif
.macro KVMTEST name
lbz r10,HSTATE_IN_GUEST(r13)
diff --git a/arch/powerpc/kvm/Makefile b/arch/powerpc/kvm/Makefile
index 2bfeaa13befb..cdd119028f64 100644
--- a/arch/powerpc/kvm/Makefile
+++ b/arch/powerpc/kvm/Makefile
@@ -59,6 +59,9 @@ kvm-pr-y := \
kvm-book3s_64-builtin-objs-$(CONFIG_KVM_BOOK3S_64_HANDLER) += \
tm.o
+kvm-book3s_64-builtin-objs-y += \
+ book3s_64_entry.o
+
ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
kvm-book3s_64-builtin-objs-$(CONFIG_KVM_BOOK3S_64_HANDLER) += \
book3s_rmhandlers.o
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
new file mode 100644
index 000000000000..7a039ea78f15
--- /dev/null
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#include <asm/asm-offsets.h>
+#include <asm/cache.h>
+#include <asm/kvm_asm.h>
+#include <asm/kvm_book3s_asm.h>
+#include <asm/ppc_asm.h>
+#include <asm/reg.h>
+
+/*
+ * This is branched to from interrupt handlers in exception-64s.S which set
+ * IKVM_REAL or IKVM_VIRT, if HSTATE_IN_GUEST was found to be non-zero.
+ */
+.global kvmppc_interrupt
+.balign IFETCH_ALIGN_BYTES
+kvmppc_interrupt:
+ /*
+ * Register contents:
+ * R12 = (guest CR << 32) | interrupt vector
+ * R13 = PACA
+ * guest R12 saved in shadow VCPU SCRATCH0
+ * guest R13 saved in SPRN_SCRATCH0
+ */
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+ std r9,HSTATE_SCRATCH2(r13)
+ lbz r9,HSTATE_IN_GUEST(r13)
+ cmpwi r9,KVM_GUEST_MODE_HOST_HV
+ beq kvmppc_bad_host_intr
+#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
+ cmpwi r9,KVM_GUEST_MODE_GUEST
+ ld r9,HSTATE_SCRATCH2(r13)
+ beq kvmppc_interrupt_pr
+#endif
+ b kvmppc_interrupt_hv
+#else
+ b kvmppc_interrupt_pr
+#endif
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 5e634db4809b..f976efb7e4a9 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -1269,16 +1269,8 @@ kvmppc_interrupt_hv:
* R13 = PACA
* guest R12 saved in shadow VCPU SCRATCH0
* guest R13 saved in SPRN_SCRATCH0
+ * guest R9 saved in HSTATE_SCRATCH2
*/
- std r9, HSTATE_SCRATCH2(r13)
- lbz r9, HSTATE_IN_GUEST(r13)
- cmpwi r9, KVM_GUEST_MODE_HOST_HV
- beq kvmppc_bad_host_intr
-#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
- cmpwi r9, KVM_GUEST_MODE_GUEST
- ld r9, HSTATE_SCRATCH2(r13)
- beq kvmppc_interrupt_pr
-#endif
/* We're now back in the host but in guest MMU context */
li r9, KVM_GUEST_MODE_HOST_HV
stb r9, HSTATE_IN_GUEST(r13)
@@ -3280,6 +3272,7 @@ END_FTR_SECTION_IFCLR(CPU_FTR_P9_TM_HV_ASSIST)
* cfar is saved in HSTATE_CFAR(r13)
* ppr is saved in HSTATE_PPR(r13)
*/
+.global kvmppc_bad_host_intr
kvmppc_bad_host_intr:
/*
* Switch to the emergency stack, but start half-way down in
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 12/46] KVM: PPC: Book3S 64: move KVM interrupt entry to a common entry point
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Fabiano Rosas, linuxppc-dev, Nicholas Piggin, Daniel Axtens
Rather than bifurcate the call depending on whether or not HV is
possible, and have the HV entry test for PR, just make a single
common point which does the demultiplexing. This makes it simpler
to add another type of exit handler.
Reviewed-by: Daniel Axtens <dja@axtens.net>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kernel/exceptions-64s.S | 8 +-----
arch/powerpc/kvm/Makefile | 3 +++
arch/powerpc/kvm/book3s_64_entry.S | 36 +++++++++++++++++++++++++
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 11 ++------
4 files changed, 42 insertions(+), 16 deletions(-)
create mode 100644 arch/powerpc/kvm/book3s_64_entry.S
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index c9c446ccff54..162595af1ac7 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -208,7 +208,6 @@ do_define_int n
.endm
#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
-#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
/*
* All interrupts which set HSRR registers, as well as SRESET and MCE and
* syscall when invoked with "sc 1" switch to MSR[HV]=1 (HVMODE) to be taken,
@@ -238,13 +237,8 @@ do_define_int n
/*
* If an interrupt is taken while a guest is running, it is immediately routed
- * to KVM to handle. If both HV and PR KVM arepossible, KVM interrupts go first
- * to kvmppc_interrupt_hv, which handles the PR guest case.
+ * to KVM to handle.
*/
-#define kvmppc_interrupt kvmppc_interrupt_hv
-#else
-#define kvmppc_interrupt kvmppc_interrupt_pr
-#endif
.macro KVMTEST name
lbz r10,HSTATE_IN_GUEST(r13)
diff --git a/arch/powerpc/kvm/Makefile b/arch/powerpc/kvm/Makefile
index 2bfeaa13befb..cdd119028f64 100644
--- a/arch/powerpc/kvm/Makefile
+++ b/arch/powerpc/kvm/Makefile
@@ -59,6 +59,9 @@ kvm-pr-y := \
kvm-book3s_64-builtin-objs-$(CONFIG_KVM_BOOK3S_64_HANDLER) += \
tm.o
+kvm-book3s_64-builtin-objs-y += \
+ book3s_64_entry.o
+
ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
kvm-book3s_64-builtin-objs-$(CONFIG_KVM_BOOK3S_64_HANDLER) += \
book3s_rmhandlers.o
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
new file mode 100644
index 000000000000..7a039ea78f15
--- /dev/null
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#include <asm/asm-offsets.h>
+#include <asm/cache.h>
+#include <asm/kvm_asm.h>
+#include <asm/kvm_book3s_asm.h>
+#include <asm/ppc_asm.h>
+#include <asm/reg.h>
+
+/*
+ * This is branched to from interrupt handlers in exception-64s.S which set
+ * IKVM_REAL or IKVM_VIRT, if HSTATE_IN_GUEST was found to be non-zero.
+ */
+.global kvmppc_interrupt
+.balign IFETCH_ALIGN_BYTES
+kvmppc_interrupt:
+ /*
+ * Register contents:
+ * R12 = (guest CR << 32) | interrupt vector
+ * R13 = PACA
+ * guest R12 saved in shadow VCPU SCRATCH0
+ * guest R13 saved in SPRN_SCRATCH0
+ */
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+ std r9,HSTATE_SCRATCH2(r13)
+ lbz r9,HSTATE_IN_GUEST(r13)
+ cmpwi r9,KVM_GUEST_MODE_HOST_HV
+ beq kvmppc_bad_host_intr
+#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
+ cmpwi r9,KVM_GUEST_MODE_GUEST
+ ld r9,HSTATE_SCRATCH2(r13)
+ beq kvmppc_interrupt_pr
+#endif
+ b kvmppc_interrupt_hv
+#else
+ b kvmppc_interrupt_pr
+#endif
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 5e634db4809b..f976efb7e4a9 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -1269,16 +1269,8 @@ kvmppc_interrupt_hv:
* R13 = PACA
* guest R12 saved in shadow VCPU SCRATCH0
* guest R13 saved in SPRN_SCRATCH0
+ * guest R9 saved in HSTATE_SCRATCH2
*/
- std r9, HSTATE_SCRATCH2(r13)
- lbz r9, HSTATE_IN_GUEST(r13)
- cmpwi r9, KVM_GUEST_MODE_HOST_HV
- beq kvmppc_bad_host_intr
-#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
- cmpwi r9, KVM_GUEST_MODE_GUEST
- ld r9, HSTATE_SCRATCH2(r13)
- beq kvmppc_interrupt_pr
-#endif
/* We're now back in the host but in guest MMU context */
li r9, KVM_GUEST_MODE_HOST_HV
stb r9, HSTATE_IN_GUEST(r13)
@@ -3280,6 +3272,7 @@ END_FTR_SECTION_IFCLR(CPU_FTR_P9_TM_HV_ASSIST)
* cfar is saved in HSTATE_CFAR(r13)
* ppr is saved in HSTATE_PPR(r13)
*/
+.global kvmppc_bad_host_intr
kvmppc_bad_host_intr:
/*
* Switch to the emergency stack, but start half-way down in
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* Re: [PATCH v4 12/46] KVM: PPC: Book3S 64: move KVM interrupt entry to a common entry point
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-04-01 4:29 ` Paul Mackerras
-1 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-04-01 4:29 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: Fabiano Rosas, linuxppc-dev, kvm-ppc, Daniel Axtens
On Tue, Mar 23, 2021 at 11:02:31AM +1000, Nicholas Piggin wrote:
> Rather than bifurcate the call depending on whether or not HV is
> possible, and have the HV entry test for PR, just make a single
> common point which does the demultiplexing. This makes it simpler
> to add another type of exit handler.
>
> Reviewed-by: Daniel Axtens <dja@axtens.net>
> Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 12/46] KVM: PPC: Book3S 64: move KVM interrupt entry to a common entry point
@ 2021-04-01 4:29 ` Paul Mackerras
0 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-04-01 4:29 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: Fabiano Rosas, linuxppc-dev, kvm-ppc, Daniel Axtens
On Tue, Mar 23, 2021 at 11:02:31AM +1000, Nicholas Piggin wrote:
> Rather than bifurcate the call depending on whether or not HV is
> possible, and have the HV entry test for PR, just make a single
> common point which does the demultiplexing. This makes it simpler
> to add another type of exit handler.
>
> Reviewed-by: Daniel Axtens <dja@axtens.net>
> Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Paul Mackerras <paulus@ozlabs.org>
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 13/46] KVM: PPC: Book3S 64: Move GUEST_MODE_SKIP test into KVM
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Fabiano Rosas, linuxppc-dev, Nicholas Piggin, Daniel Axtens
Move the GUEST_MODE_SKIP logic into KVM code. This is quite a KVM
internal detail that has no real need to be in common handlers.
Also add a comment explaining why this thing exists.
Reviewed-by: Daniel Axtens <dja@axtens.net>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kernel/exceptions-64s.S | 60 ----------------------------
arch/powerpc/kvm/book3s_64_entry.S | 51 ++++++++++++++++++++++-
2 files changed, 50 insertions(+), 61 deletions(-)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 162595af1ac7..16fbfde960e7 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -133,7 +133,6 @@ name:
#define IBRANCH_TO_COMMON .L_IBRANCH_TO_COMMON_\name\() /* ENTRY branch to common */
#define IREALMODE_COMMON .L_IREALMODE_COMMON_\name\() /* Common runs in realmode */
#define IMASK .L_IMASK_\name\() /* IRQ soft-mask bit */
-#define IKVM_SKIP .L_IKVM_SKIP_\name\() /* Generate KVM skip handler */
#define IKVM_REAL .L_IKVM_REAL_\name\() /* Real entry tests KVM */
#define __IKVM_REAL(name) .L_IKVM_REAL_ ## name
#define IKVM_VIRT .L_IKVM_VIRT_\name\() /* Virt entry tests KVM */
@@ -190,9 +189,6 @@ do_define_int n
.ifndef IMASK
IMASK=0
.endif
- .ifndef IKVM_SKIP
- IKVM_SKIP=0
- .endif
.ifndef IKVM_REAL
IKVM_REAL=0
.endif
@@ -250,15 +246,10 @@ do_define_int n
.balign IFETCH_ALIGN_BYTES
\name\()_kvm:
- .if IKVM_SKIP
- cmpwi r10,KVM_GUEST_MODE_SKIP
- beq 89f
- .else
BEGIN_FTR_SECTION
ld r10,IAREA+EX_CFAR(r13)
std r10,HSTATE_CFAR(r13)
END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
- .endif
ld r10,IAREA+EX_CTR(r13)
mtctr r10
@@ -285,27 +276,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
ori r12,r12,(IVEC)
.endif
b kvmppc_interrupt
-
- .if IKVM_SKIP
-89: mtocrf 0x80,r9
- ld r10,IAREA+EX_CTR(r13)
- mtctr r10
- ld r9,IAREA+EX_R9(r13)
- ld r10,IAREA+EX_R10(r13)
- ld r11,IAREA+EX_R11(r13)
- ld r12,IAREA+EX_R12(r13)
- .if IHSRR_IF_HVMODE
- BEGIN_FTR_SECTION
- b kvmppc_skip_Hinterrupt
- FTR_SECTION_ELSE
- b kvmppc_skip_interrupt
- ALT_FTR_SECTION_END_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206)
- .elseif IHSRR
- b kvmppc_skip_Hinterrupt
- .else
- b kvmppc_skip_interrupt
- .endif
- .endif
.endm
#else
@@ -1083,7 +1053,6 @@ INT_DEFINE_BEGIN(machine_check)
ISET_RI=0
IDAR=1
IDSISR=1
- IKVM_SKIP=1
IKVM_REAL=1
INT_DEFINE_END(machine_check)
@@ -1356,7 +1325,6 @@ INT_DEFINE_BEGIN(data_access)
IVEC=0x300
IDAR=1
IDSISR=1
- IKVM_SKIP=1
IKVM_REAL=1
INT_DEFINE_END(data_access)
@@ -1410,7 +1378,6 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX)
INT_DEFINE_BEGIN(data_access_slb)
IVEC=0x380
IDAR=1
- IKVM_SKIP=1
IKVM_REAL=1
INT_DEFINE_END(data_access_slb)
@@ -2080,7 +2047,6 @@ INT_DEFINE_BEGIN(h_data_storage)
IHSRR=1
IDAR=1
IDSISR=1
- IKVM_SKIP=1
IKVM_REAL=1
IKVM_VIRT=1
INT_DEFINE_END(h_data_storage)
@@ -3024,32 +2990,6 @@ EXPORT_SYMBOL(do_uaccess_flush)
MASKED_INTERRUPT
MASKED_INTERRUPT hsrr=1
-#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
-kvmppc_skip_interrupt:
- /*
- * Here all GPRs are unchanged from when the interrupt happened
- * except for r13, which is saved in SPRG_SCRATCH0.
- */
- mfspr r13, SPRN_SRR0
- addi r13, r13, 4
- mtspr SPRN_SRR0, r13
- GET_SCRATCH0(r13)
- RFI_TO_KERNEL
- b .
-
-kvmppc_skip_Hinterrupt:
- /*
- * Here all GPRs are unchanged from when the interrupt happened
- * except for r13, which is saved in SPRG_SCRATCH0.
- */
- mfspr r13, SPRN_HSRR0
- addi r13, r13, 4
- mtspr SPRN_HSRR0, r13
- GET_SCRATCH0(r13)
- HRFI_TO_KERNEL
- b .
-#endif
-
/*
* Relocation-on interrupts: A subset of the interrupts can be delivered
* with IR=1/DR=1, if AIL==2 and MSR.HV won't be changed by delivering
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index 7a039ea78f15..a5412e24cc05 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -1,6 +1,7 @@
/* SPDX-License-Identifier: GPL-2.0-only */
#include <asm/asm-offsets.h>
#include <asm/cache.h>
+#include <asm/exception-64s.h>
#include <asm/kvm_asm.h>
#include <asm/kvm_book3s_asm.h>
#include <asm/ppc_asm.h>
@@ -20,9 +21,12 @@ kvmppc_interrupt:
* guest R12 saved in shadow VCPU SCRATCH0
* guest R13 saved in SPRN_SCRATCH0
*/
-#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
std r9,HSTATE_SCRATCH2(r13)
lbz r9,HSTATE_IN_GUEST(r13)
+ cmpwi r9,KVM_GUEST_MODE_SKIP
+ beq- .Lmaybe_skip
+.Lno_skip:
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
cmpwi r9,KVM_GUEST_MODE_HOST_HV
beq kvmppc_bad_host_intr
#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
@@ -34,3 +38,48 @@ kvmppc_interrupt:
#else
b kvmppc_interrupt_pr
#endif
+
+/*
+ * KVM uses a trick where it is running in MSR[HV]=1 mode in real-mode with the
+ * guest MMU context loaded, and it sets KVM_GUEST_MODE_SKIP and enables
+ * MSR[DR]=1 while leaving MSR[IR]=0, so it continues to fetch HV instructions
+ * but loads and stores will access the guest context. This is used to load
+ * the faulting instruction without walking page tables.
+ *
+ * However the guest context may not be able to translate, or it may cause a
+ * machine check or other issue, which will result in a fault in the host
+ * (even with KVM-HV).
+ *
+ * These faults are caught here and if the fault was (or was likely) due to
+ * that load, then we just return with the PC advanced +4 and skip the load,
+ * which then goes to the slow path and walks the page tables.
+ */
+.Lmaybe_skip:
+ cmpwi r12,BOOK3S_INTERRUPT_MACHINE_CHECK
+ beq 1f
+ cmpwi r12,BOOK3S_INTERRUPT_DATA_STORAGE
+ beq 1f
+ cmpwi r12,BOOK3S_INTERRUPT_DATA_SEGMENT
+ beq 1f
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+ /* HSRR interrupts have 2 added to trap vector */
+ cmpwi r12,BOOK3S_INTERRUPT_H_DATA_STORAGE | 0x2
+ beq 2f
+#endif
+ b .Lno_skip
+1: mfspr r9,SPRN_SRR0
+ addi r9,r9,4
+ mtspr SPRN_SRR0,r9
+ ld r12,HSTATE_SCRATCH0(r13)
+ ld r9,HSTATE_SCRATCH2(r13)
+ GET_SCRATCH0(r13)
+ RFI_TO_KERNEL
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+2: mfspr r9,SPRN_HSRR0
+ addi r9,r9,4
+ mtspr SPRN_HSRR0,r9
+ ld r12,HSTATE_SCRATCH0(r13)
+ ld r9,HSTATE_SCRATCH2(r13)
+ GET_SCRATCH0(r13)
+ HRFI_TO_KERNEL
+#endif
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 13/46] KVM: PPC: Book3S 64: Move GUEST_MODE_SKIP test into KVM
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Fabiano Rosas, linuxppc-dev, Nicholas Piggin, Daniel Axtens
Move the GUEST_MODE_SKIP logic into KVM code. This is quite a KVM
internal detail that has no real need to be in common handlers.
Also add a comment explaining why this thing exists.
Reviewed-by: Daniel Axtens <dja@axtens.net>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kernel/exceptions-64s.S | 60 ----------------------------
arch/powerpc/kvm/book3s_64_entry.S | 51 ++++++++++++++++++++++-
2 files changed, 50 insertions(+), 61 deletions(-)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 162595af1ac7..16fbfde960e7 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -133,7 +133,6 @@ name:
#define IBRANCH_TO_COMMON .L_IBRANCH_TO_COMMON_\name\() /* ENTRY branch to common */
#define IREALMODE_COMMON .L_IREALMODE_COMMON_\name\() /* Common runs in realmode */
#define IMASK .L_IMASK_\name\() /* IRQ soft-mask bit */
-#define IKVM_SKIP .L_IKVM_SKIP_\name\() /* Generate KVM skip handler */
#define IKVM_REAL .L_IKVM_REAL_\name\() /* Real entry tests KVM */
#define __IKVM_REAL(name) .L_IKVM_REAL_ ## name
#define IKVM_VIRT .L_IKVM_VIRT_\name\() /* Virt entry tests KVM */
@@ -190,9 +189,6 @@ do_define_int n
.ifndef IMASK
IMASK=0
.endif
- .ifndef IKVM_SKIP
- IKVM_SKIP=0
- .endif
.ifndef IKVM_REAL
IKVM_REAL=0
.endif
@@ -250,15 +246,10 @@ do_define_int n
.balign IFETCH_ALIGN_BYTES
\name\()_kvm:
- .if IKVM_SKIP
- cmpwi r10,KVM_GUEST_MODE_SKIP
- beq 89f
- .else
BEGIN_FTR_SECTION
ld r10,IAREA+EX_CFAR(r13)
std r10,HSTATE_CFAR(r13)
END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
- .endif
ld r10,IAREA+EX_CTR(r13)
mtctr r10
@@ -285,27 +276,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
ori r12,r12,(IVEC)
.endif
b kvmppc_interrupt
-
- .if IKVM_SKIP
-89: mtocrf 0x80,r9
- ld r10,IAREA+EX_CTR(r13)
- mtctr r10
- ld r9,IAREA+EX_R9(r13)
- ld r10,IAREA+EX_R10(r13)
- ld r11,IAREA+EX_R11(r13)
- ld r12,IAREA+EX_R12(r13)
- .if IHSRR_IF_HVMODE
- BEGIN_FTR_SECTION
- b kvmppc_skip_Hinterrupt
- FTR_SECTION_ELSE
- b kvmppc_skip_interrupt
- ALT_FTR_SECTION_END_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206)
- .elseif IHSRR
- b kvmppc_skip_Hinterrupt
- .else
- b kvmppc_skip_interrupt
- .endif
- .endif
.endm
#else
@@ -1083,7 +1053,6 @@ INT_DEFINE_BEGIN(machine_check)
ISET_RI=0
IDAR=1
IDSISR=1
- IKVM_SKIP=1
IKVM_REAL=1
INT_DEFINE_END(machine_check)
@@ -1356,7 +1325,6 @@ INT_DEFINE_BEGIN(data_access)
IVEC=0x300
IDAR=1
IDSISR=1
- IKVM_SKIP=1
IKVM_REAL=1
INT_DEFINE_END(data_access)
@@ -1410,7 +1378,6 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX)
INT_DEFINE_BEGIN(data_access_slb)
IVEC=0x380
IDAR=1
- IKVM_SKIP=1
IKVM_REAL=1
INT_DEFINE_END(data_access_slb)
@@ -2080,7 +2047,6 @@ INT_DEFINE_BEGIN(h_data_storage)
IHSRR=1
IDAR=1
IDSISR=1
- IKVM_SKIP=1
IKVM_REAL=1
IKVM_VIRT=1
INT_DEFINE_END(h_data_storage)
@@ -3024,32 +2990,6 @@ EXPORT_SYMBOL(do_uaccess_flush)
MASKED_INTERRUPT
MASKED_INTERRUPT hsrr=1
-#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
-kvmppc_skip_interrupt:
- /*
- * Here all GPRs are unchanged from when the interrupt happened
- * except for r13, which is saved in SPRG_SCRATCH0.
- */
- mfspr r13, SPRN_SRR0
- addi r13, r13, 4
- mtspr SPRN_SRR0, r13
- GET_SCRATCH0(r13)
- RFI_TO_KERNEL
- b .
-
-kvmppc_skip_Hinterrupt:
- /*
- * Here all GPRs are unchanged from when the interrupt happened
- * except for r13, which is saved in SPRG_SCRATCH0.
- */
- mfspr r13, SPRN_HSRR0
- addi r13, r13, 4
- mtspr SPRN_HSRR0, r13
- GET_SCRATCH0(r13)
- HRFI_TO_KERNEL
- b .
-#endif
-
/*
* Relocation-on interrupts: A subset of the interrupts can be delivered
* with IR=1/DR=1, if AIL=2 and MSR.HV won't be changed by delivering
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index 7a039ea78f15..a5412e24cc05 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -1,6 +1,7 @@
/* SPDX-License-Identifier: GPL-2.0-only */
#include <asm/asm-offsets.h>
#include <asm/cache.h>
+#include <asm/exception-64s.h>
#include <asm/kvm_asm.h>
#include <asm/kvm_book3s_asm.h>
#include <asm/ppc_asm.h>
@@ -20,9 +21,12 @@ kvmppc_interrupt:
* guest R12 saved in shadow VCPU SCRATCH0
* guest R13 saved in SPRN_SCRATCH0
*/
-#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
std r9,HSTATE_SCRATCH2(r13)
lbz r9,HSTATE_IN_GUEST(r13)
+ cmpwi r9,KVM_GUEST_MODE_SKIP
+ beq- .Lmaybe_skip
+.Lno_skip:
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
cmpwi r9,KVM_GUEST_MODE_HOST_HV
beq kvmppc_bad_host_intr
#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
@@ -34,3 +38,48 @@ kvmppc_interrupt:
#else
b kvmppc_interrupt_pr
#endif
+
+/*
+ * KVM uses a trick where it is running in MSR[HV]=1 mode in real-mode with the
+ * guest MMU context loaded, and it sets KVM_GUEST_MODE_SKIP and enables
+ * MSR[DR]=1 while leaving MSR[IR]=0, so it continues to fetch HV instructions
+ * but loads and stores will access the guest context. This is used to load
+ * the faulting instruction without walking page tables.
+ *
+ * However the guest context may not be able to translate, or it may cause a
+ * machine check or other issue, which will result in a fault in the host
+ * (even with KVM-HV).
+ *
+ * These faults are caught here and if the fault was (or was likely) due to
+ * that load, then we just return with the PC advanced +4 and skip the load,
+ * which then goes to the slow path and walks the page tables.
+ */
+.Lmaybe_skip:
+ cmpwi r12,BOOK3S_INTERRUPT_MACHINE_CHECK
+ beq 1f
+ cmpwi r12,BOOK3S_INTERRUPT_DATA_STORAGE
+ beq 1f
+ cmpwi r12,BOOK3S_INTERRUPT_DATA_SEGMENT
+ beq 1f
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+ /* HSRR interrupts have 2 added to trap vector */
+ cmpwi r12,BOOK3S_INTERRUPT_H_DATA_STORAGE | 0x2
+ beq 2f
+#endif
+ b .Lno_skip
+1: mfspr r9,SPRN_SRR0
+ addi r9,r9,4
+ mtspr SPRN_SRR0,r9
+ ld r12,HSTATE_SCRATCH0(r13)
+ ld r9,HSTATE_SCRATCH2(r13)
+ GET_SCRATCH0(r13)
+ RFI_TO_KERNEL
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+2: mfspr r9,SPRN_HSRR0
+ addi r9,r9,4
+ mtspr SPRN_HSRR0,r9
+ ld r12,HSTATE_SCRATCH0(r13)
+ ld r9,HSTATE_SCRATCH2(r13)
+ GET_SCRATCH0(r13)
+ HRFI_TO_KERNEL
+#endif
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* Re: [PATCH v4 13/46] KVM: PPC: Book3S 64: Move GUEST_MODE_SKIP test into KVM
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-04-01 5:32 ` Paul Mackerras
-1 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-04-01 5:32 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: Fabiano Rosas, linuxppc-dev, kvm-ppc, Daniel Axtens
On Tue, Mar 23, 2021 at 11:02:32AM +1000, Nicholas Piggin wrote:
> Move the GUEST_MODE_SKIP logic into KVM code. This is quite a KVM
> internal detail that has no real need to be in common handlers.
>
> Also add a comment explaining why this thing exists.
[snip]
> diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
> index 7a039ea78f15..a5412e24cc05 100644
> --- a/arch/powerpc/kvm/book3s_64_entry.S
> +++ b/arch/powerpc/kvm/book3s_64_entry.S
> @@ -1,6 +1,7 @@
> /* SPDX-License-Identifier: GPL-2.0-only */
> #include <asm/asm-offsets.h>
> #include <asm/cache.h>
> +#include <asm/exception-64s.h>
> #include <asm/kvm_asm.h>
> #include <asm/kvm_book3s_asm.h>
> #include <asm/ppc_asm.h>
> @@ -20,9 +21,12 @@ kvmppc_interrupt:
> * guest R12 saved in shadow VCPU SCRATCH0
> * guest R13 saved in SPRN_SCRATCH0
> */
> -#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
> std r9,HSTATE_SCRATCH2(r13)
> lbz r9,HSTATE_IN_GUEST(r13)
> + cmpwi r9,KVM_GUEST_MODE_SKIP
> + beq- .Lmaybe_skip
> +.Lno_skip:
> +#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
> cmpwi r9,KVM_GUEST_MODE_HOST_HV
> beq kvmppc_bad_host_intr
> #ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
> @@ -34,3 +38,48 @@ kvmppc_interrupt:
> #else
> b kvmppc_interrupt_pr
> #endif
It's a bit hard to see without more context, but I think that in the
PR-only case (CONFIG_KVM_BOOK3S_HV_POSSIBLE undefined), this will
corrupt R9. You need to restore R9 before the unconditional branch to
kvmppc_interrupt_pr. (I realize this code gets modified further, but
I'd rather not break bisection.)
Paul.
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 13/46] KVM: PPC: Book3S 64: Move GUEST_MODE_SKIP test into KVM
@ 2021-04-01 5:32 ` Paul Mackerras
0 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-04-01 5:32 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: Fabiano Rosas, linuxppc-dev, kvm-ppc, Daniel Axtens
On Tue, Mar 23, 2021 at 11:02:32AM +1000, Nicholas Piggin wrote:
> Move the GUEST_MODE_SKIP logic into KVM code. This is quite a KVM
> internal detail that has no real need to be in common handlers.
>
> Also add a comment explaining why this thing exists.
[snip]
> diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
> index 7a039ea78f15..a5412e24cc05 100644
> --- a/arch/powerpc/kvm/book3s_64_entry.S
> +++ b/arch/powerpc/kvm/book3s_64_entry.S
> @@ -1,6 +1,7 @@
> /* SPDX-License-Identifier: GPL-2.0-only */
> #include <asm/asm-offsets.h>
> #include <asm/cache.h>
> +#include <asm/exception-64s.h>
> #include <asm/kvm_asm.h>
> #include <asm/kvm_book3s_asm.h>
> #include <asm/ppc_asm.h>
> @@ -20,9 +21,12 @@ kvmppc_interrupt:
> * guest R12 saved in shadow VCPU SCRATCH0
> * guest R13 saved in SPRN_SCRATCH0
> */
> -#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
> std r9,HSTATE_SCRATCH2(r13)
> lbz r9,HSTATE_IN_GUEST(r13)
> + cmpwi r9,KVM_GUEST_MODE_SKIP
> + beq- .Lmaybe_skip
> +.Lno_skip:
> +#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
> cmpwi r9,KVM_GUEST_MODE_HOST_HV
> beq kvmppc_bad_host_intr
> #ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
> @@ -34,3 +38,48 @@ kvmppc_interrupt:
> #else
> b kvmppc_interrupt_pr
> #endif
It's a bit hard to see without more context, but I think that in the
PR-only case (CONFIG_KVM_BOOK3S_HV_POSSIBLE undefined), this will
corrupt R9. You need to restore R9 before the unconditional branch to
kvmppc_interrupt_pr. (I realize this code gets modified further, but
I'd rather not break bisection.)
Paul.
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 13/46] KVM: PPC: Book3S 64: Move GUEST_MODE_SKIP test into KVM
2021-04-01 5:32 ` Paul Mackerras
@ 2021-04-01 9:53 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 9:53 UTC (permalink / raw)
To: Paul Mackerras; +Cc: Fabiano Rosas, linuxppc-dev, kvm-ppc, Daniel Axtens
Excerpts from Paul Mackerras's message of April 1, 2021 3:32 pm:
> On Tue, Mar 23, 2021 at 11:02:32AM +1000, Nicholas Piggin wrote:
>> Move the GUEST_MODE_SKIP logic into KVM code. This is quite a KVM
>> internal detail that has no real need to be in common handlers.
>>
>> Also add a comment explaining why this thing exists.
>
> [snip]
>
>> diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
>> index 7a039ea78f15..a5412e24cc05 100644
>> --- a/arch/powerpc/kvm/book3s_64_entry.S
>> +++ b/arch/powerpc/kvm/book3s_64_entry.S
>> @@ -1,6 +1,7 @@
>> /* SPDX-License-Identifier: GPL-2.0-only */
>> #include <asm/asm-offsets.h>
>> #include <asm/cache.h>
>> +#include <asm/exception-64s.h>
>> #include <asm/kvm_asm.h>
>> #include <asm/kvm_book3s_asm.h>
>> #include <asm/ppc_asm.h>
>> @@ -20,9 +21,12 @@ kvmppc_interrupt:
>> * guest R12 saved in shadow VCPU SCRATCH0
>> * guest R13 saved in SPRN_SCRATCH0
>> */
>> -#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
>> std r9,HSTATE_SCRATCH2(r13)
>> lbz r9,HSTATE_IN_GUEST(r13)
>> + cmpwi r9,KVM_GUEST_MODE_SKIP
>> + beq- .Lmaybe_skip
>> +.Lno_skip:
>> +#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
>> cmpwi r9,KVM_GUEST_MODE_HOST_HV
>> beq kvmppc_bad_host_intr
>> #ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
>> @@ -34,3 +38,48 @@ kvmppc_interrupt:
>> #else
>> b kvmppc_interrupt_pr
>> #endif
>
> It's a bit hard to see without more context, but I think that in the
> PR-only case (CONFIG_KVM_BOOK3S_HV_POSSIBLE undefined), this will
> corrupt R9. You need to restore R9 before the unconditional branch to
> kvmppc_interrupt_pr. (I realize this code gets modified further, but
> I'd rather not break bisection.)
Very good catch, thanks.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 13/46] KVM: PPC: Book3S 64: Move GUEST_MODE_SKIP test into KVM
@ 2021-04-01 9:53 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 9:53 UTC (permalink / raw)
To: Paul Mackerras; +Cc: Fabiano Rosas, linuxppc-dev, kvm-ppc, Daniel Axtens
Excerpts from Paul Mackerras's message of April 1, 2021 3:32 pm:
> On Tue, Mar 23, 2021 at 11:02:32AM +1000, Nicholas Piggin wrote:
>> Move the GUEST_MODE_SKIP logic into KVM code. This is quite a KVM
>> internal detail that has no real need to be in common handlers.
>>
>> Also add a comment explaining why this thing exists.
>
> [snip]
>
>> diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
>> index 7a039ea78f15..a5412e24cc05 100644
>> --- a/arch/powerpc/kvm/book3s_64_entry.S
>> +++ b/arch/powerpc/kvm/book3s_64_entry.S
>> @@ -1,6 +1,7 @@
>> /* SPDX-License-Identifier: GPL-2.0-only */
>> #include <asm/asm-offsets.h>
>> #include <asm/cache.h>
>> +#include <asm/exception-64s.h>
>> #include <asm/kvm_asm.h>
>> #include <asm/kvm_book3s_asm.h>
>> #include <asm/ppc_asm.h>
>> @@ -20,9 +21,12 @@ kvmppc_interrupt:
>> * guest R12 saved in shadow VCPU SCRATCH0
>> * guest R13 saved in SPRN_SCRATCH0
>> */
>> -#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
>> std r9,HSTATE_SCRATCH2(r13)
>> lbz r9,HSTATE_IN_GUEST(r13)
>> + cmpwi r9,KVM_GUEST_MODE_SKIP
>> + beq- .Lmaybe_skip
>> +.Lno_skip:
>> +#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
>> cmpwi r9,KVM_GUEST_MODE_HOST_HV
>> beq kvmppc_bad_host_intr
>> #ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
>> @@ -34,3 +38,48 @@ kvmppc_interrupt:
>> #else
>> b kvmppc_interrupt_pr
>> #endif
>
> It's a bit hard to see without more context, but I think that in the
> PR-only case (CONFIG_KVM_BOOK3S_HV_POSSIBLE undefined), this will
> corrupt R9. You need to restore R9 before the unconditional branch to
> kvmppc_interrupt_pr. (I realize this code gets modified further, but
> I'd rather not break bisection.)
Very good catch, thanks.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 14/46] KVM: PPC: Book3S 64: add hcall interrupt handler
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Fabiano Rosas, linuxppc-dev, Nicholas Piggin, Daniel Axtens
Add a separate hcall entry point. This can be used to deal with the
different calling convention.
Reviewed-by: Daniel Axtens <dja@axtens.net>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kernel/exceptions-64s.S | 6 +++---
arch/powerpc/kvm/book3s_64_entry.S | 6 +++++-
2 files changed, 8 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 16fbfde960e7..98bf73df0f57 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -1989,16 +1989,16 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
ori r12,r12,0xc00
#ifdef CONFIG_RELOCATABLE
/*
- * Requires __LOAD_FAR_HANDLER beause kvmppc_interrupt lives
+ * Requires __LOAD_FAR_HANDLER beause kvmppc_hcall lives
* outside the head section.
*/
- __LOAD_FAR_HANDLER(r10, kvmppc_interrupt)
+ __LOAD_FAR_HANDLER(r10, kvmppc_hcall)
mtctr r10
ld r10,PACA_EXGEN+EX_R10(r13)
bctr
#else
ld r10,PACA_EXGEN+EX_R10(r13)
- b kvmppc_interrupt
+ b kvmppc_hcall
#endif
#endif
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index a5412e24cc05..8cf5e24a81eb 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -8,9 +8,13 @@
#include <asm/reg.h>
/*
- * This is branched to from interrupt handlers in exception-64s.S which set
+ * These are branched to from interrupt handlers in exception-64s.S which set
* IKVM_REAL or IKVM_VIRT, if HSTATE_IN_GUEST was found to be non-zero.
*/
+.global kvmppc_hcall
+.balign IFETCH_ALIGN_BYTES
+kvmppc_hcall:
+
.global kvmppc_interrupt
.balign IFETCH_ALIGN_BYTES
kvmppc_interrupt:
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 14/46] KVM: PPC: Book3S 64: add hcall interrupt handler
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Fabiano Rosas, linuxppc-dev, Nicholas Piggin, Daniel Axtens
Add a separate hcall entry point. This can be used to deal with the
different calling convention.
Reviewed-by: Daniel Axtens <dja@axtens.net>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kernel/exceptions-64s.S | 6 +++---
arch/powerpc/kvm/book3s_64_entry.S | 6 +++++-
2 files changed, 8 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 16fbfde960e7..98bf73df0f57 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -1989,16 +1989,16 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
ori r12,r12,0xc00
#ifdef CONFIG_RELOCATABLE
/*
- * Requires __LOAD_FAR_HANDLER beause kvmppc_interrupt lives
+ * Requires __LOAD_FAR_HANDLER beause kvmppc_hcall lives
* outside the head section.
*/
- __LOAD_FAR_HANDLER(r10, kvmppc_interrupt)
+ __LOAD_FAR_HANDLER(r10, kvmppc_hcall)
mtctr r10
ld r10,PACA_EXGEN+EX_R10(r13)
bctr
#else
ld r10,PACA_EXGEN+EX_R10(r13)
- b kvmppc_interrupt
+ b kvmppc_hcall
#endif
#endif
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index a5412e24cc05..8cf5e24a81eb 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -8,9 +8,13 @@
#include <asm/reg.h>
/*
- * This is branched to from interrupt handlers in exception-64s.S which set
+ * These are branched to from interrupt handlers in exception-64s.S which set
* IKVM_REAL or IKVM_VIRT, if HSTATE_IN_GUEST was found to be non-zero.
*/
+.global kvmppc_hcall
+.balign IFETCH_ALIGN_BYTES
+kvmppc_hcall:
+
.global kvmppc_interrupt
.balign IFETCH_ALIGN_BYTES
kvmppc_interrupt:
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 15/46] KVM: PPC: Book3S 64: Move hcall early register setup to KVM
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
System calls / hcalls have a different calling convention than
other interrupts, so there is code in the KVMTEST to massage these
into the same form as other interrupt handlers.
Move this work into the KVM hcall handler. This means teaching KVM
a little more about the low level interrupt handler setup, PACA save
areas, etc., although that's not obviously worse than the current
approach of coming up with an entirely different interrupt register
/ save convention.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/exception-64s.h | 13 ++++++++
arch/powerpc/kernel/exceptions-64s.S | 42 +-----------------------
arch/powerpc/kvm/book3s_64_entry.S | 23 +++++++++++++
3 files changed, 37 insertions(+), 41 deletions(-)
diff --git a/arch/powerpc/include/asm/exception-64s.h b/arch/powerpc/include/asm/exception-64s.h
index c1a8aac01cf9..bb6f78fcf981 100644
--- a/arch/powerpc/include/asm/exception-64s.h
+++ b/arch/powerpc/include/asm/exception-64s.h
@@ -35,6 +35,19 @@
/* PACA save area size in u64 units (exgen, exmc, etc) */
#define EX_SIZE 10
+/* PACA save area offsets */
+#define EX_R9 0
+#define EX_R10 8
+#define EX_R11 16
+#define EX_R12 24
+#define EX_R13 32
+#define EX_DAR 40
+#define EX_DSISR 48
+#define EX_CCR 52
+#define EX_CFAR 56
+#define EX_PPR 64
+#define EX_CTR 72
+
/*
* maximum recursive depth of MCE exceptions
*/
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 98bf73df0f57..a23feaa445b5 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -21,22 +21,6 @@
#include <asm/feature-fixups.h>
#include <asm/kup.h>
-/* PACA save area offsets (exgen, exmc, etc) */
-#define EX_R9 0
-#define EX_R10 8
-#define EX_R11 16
-#define EX_R12 24
-#define EX_R13 32
-#define EX_DAR 40
-#define EX_DSISR 48
-#define EX_CCR 52
-#define EX_CFAR 56
-#define EX_PPR 64
-#define EX_CTR 72
-.if EX_SIZE != 10
- .error "EX_SIZE is wrong"
-.endif
-
/*
* Following are fixed section helper macros.
*
@@ -1964,29 +1948,8 @@ EXC_VIRT_END(system_call, 0x4c00, 0x100)
#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
TRAMP_REAL_BEGIN(system_call_kvm)
- /*
- * This is a hcall, so register convention is as above, with these
- * differences:
- * r13 = PACA
- * ctr = orig r13
- * orig r10 saved in PACA
- */
- /*
- * Save the PPR (on systems that support it) before changing to
- * HMT_MEDIUM. That allows the KVM code to save that value into the
- * guest state (it is the guest's PPR value).
- */
-BEGIN_FTR_SECTION
- mfspr r10,SPRN_PPR
- std r10,HSTATE_PPR(r13)
-END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
- HMT_MEDIUM
mfctr r10
- SET_SCRATCH0(r10)
- mfcr r10
- std r12,HSTATE_SCRATCH0(r13)
- sldi r12,r10,32
- ori r12,r12,0xc00
+ SET_SCRATCH0(r10) /* Save r13 in SCRATCH0 */
#ifdef CONFIG_RELOCATABLE
/*
* Requires __LOAD_FAR_HANDLER beause kvmppc_hcall lives
@@ -1994,15 +1957,12 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
*/
__LOAD_FAR_HANDLER(r10, kvmppc_hcall)
mtctr r10
- ld r10,PACA_EXGEN+EX_R10(r13)
bctr
#else
- ld r10,PACA_EXGEN+EX_R10(r13)
b kvmppc_hcall
#endif
#endif
-
/**
* Interrupt 0xd00 - Trace Interrupt.
* This is a synchronous interrupt in response to instruction step or
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index 8cf5e24a81eb..dea8c5bb207b 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -14,6 +14,29 @@
.global kvmppc_hcall
.balign IFETCH_ALIGN_BYTES
kvmppc_hcall:
+ /*
+ * This is a hcall, so register convention is as
+ * Documentation/powerpc/papr_hcalls.rst, with these additions:
+ * R13 = PACA
+ * guest R13 saved in SPRN_SCRATCH0
+ * R10 = free
+ * guest r10 saved in PACA_EXGEN
+ */
+ /*
+ * Save the PPR (on systems that support it) before changing to
+ * HMT_MEDIUM. That allows the KVM code to save that value into the
+ * guest state (it is the guest's PPR value).
+ */
+BEGIN_FTR_SECTION
+ mfspr r10,SPRN_PPR
+ std r10,HSTATE_PPR(r13)
+END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
+ HMT_MEDIUM
+ mfcr r10
+ std r12,HSTATE_SCRATCH0(r13)
+ sldi r12,r10,32
+ ori r12,r12,0xc00
+ ld r10,PACA_EXGEN+EX_R10(r13)
.global kvmppc_interrupt
.balign IFETCH_ALIGN_BYTES
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 15/46] KVM: PPC: Book3S 64: Move hcall early register setup to KVM
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
System calls / hcalls have a different calling convention than
other interrupts, so there is code in the KVMTEST to massage these
into the same form as other interrupt handlers.
Move this work into the KVM hcall handler. This means teaching KVM
a little more about the low level interrupt handler setup, PACA save
areas, etc., although that's not obviously worse than the current
approach of coming up with an entirely different interrupt register
/ save convention.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/exception-64s.h | 13 ++++++++
arch/powerpc/kernel/exceptions-64s.S | 42 +-----------------------
arch/powerpc/kvm/book3s_64_entry.S | 23 +++++++++++++
3 files changed, 37 insertions(+), 41 deletions(-)
diff --git a/arch/powerpc/include/asm/exception-64s.h b/arch/powerpc/include/asm/exception-64s.h
index c1a8aac01cf9..bb6f78fcf981 100644
--- a/arch/powerpc/include/asm/exception-64s.h
+++ b/arch/powerpc/include/asm/exception-64s.h
@@ -35,6 +35,19 @@
/* PACA save area size in u64 units (exgen, exmc, etc) */
#define EX_SIZE 10
+/* PACA save area offsets */
+#define EX_R9 0
+#define EX_R10 8
+#define EX_R11 16
+#define EX_R12 24
+#define EX_R13 32
+#define EX_DAR 40
+#define EX_DSISR 48
+#define EX_CCR 52
+#define EX_CFAR 56
+#define EX_PPR 64
+#define EX_CTR 72
+
/*
* maximum recursive depth of MCE exceptions
*/
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 98bf73df0f57..a23feaa445b5 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -21,22 +21,6 @@
#include <asm/feature-fixups.h>
#include <asm/kup.h>
-/* PACA save area offsets (exgen, exmc, etc) */
-#define EX_R9 0
-#define EX_R10 8
-#define EX_R11 16
-#define EX_R12 24
-#define EX_R13 32
-#define EX_DAR 40
-#define EX_DSISR 48
-#define EX_CCR 52
-#define EX_CFAR 56
-#define EX_PPR 64
-#define EX_CTR 72
-.if EX_SIZE != 10
- .error "EX_SIZE is wrong"
-.endif
-
/*
* Following are fixed section helper macros.
*
@@ -1964,29 +1948,8 @@ EXC_VIRT_END(system_call, 0x4c00, 0x100)
#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
TRAMP_REAL_BEGIN(system_call_kvm)
- /*
- * This is a hcall, so register convention is as above, with these
- * differences:
- * r13 = PACA
- * ctr = orig r13
- * orig r10 saved in PACA
- */
- /*
- * Save the PPR (on systems that support it) before changing to
- * HMT_MEDIUM. That allows the KVM code to save that value into the
- * guest state (it is the guest's PPR value).
- */
-BEGIN_FTR_SECTION
- mfspr r10,SPRN_PPR
- std r10,HSTATE_PPR(r13)
-END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
- HMT_MEDIUM
mfctr r10
- SET_SCRATCH0(r10)
- mfcr r10
- std r12,HSTATE_SCRATCH0(r13)
- sldi r12,r10,32
- ori r12,r12,0xc00
+ SET_SCRATCH0(r10) /* Save r13 in SCRATCH0 */
#ifdef CONFIG_RELOCATABLE
/*
* Requires __LOAD_FAR_HANDLER beause kvmppc_hcall lives
@@ -1994,15 +1957,12 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
*/
__LOAD_FAR_HANDLER(r10, kvmppc_hcall)
mtctr r10
- ld r10,PACA_EXGEN+EX_R10(r13)
bctr
#else
- ld r10,PACA_EXGEN+EX_R10(r13)
b kvmppc_hcall
#endif
#endif
-
/**
* Interrupt 0xd00 - Trace Interrupt.
* This is a synchronous interrupt in response to instruction step or
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index 8cf5e24a81eb..dea8c5bb207b 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -14,6 +14,29 @@
.global kvmppc_hcall
.balign IFETCH_ALIGN_BYTES
kvmppc_hcall:
+ /*
+ * This is a hcall, so register convention is as
+ * Documentation/powerpc/papr_hcalls.rst, with these additions:
+ * R13 = PACA
+ * guest R13 saved in SPRN_SCRATCH0
+ * R10 = free
+ * guest r10 saved in PACA_EXGEN
+ */
+ /*
+ * Save the PPR (on systems that support it) before changing to
+ * HMT_MEDIUM. That allows the KVM code to save that value into the
+ * guest state (it is the guest's PPR value).
+ */
+BEGIN_FTR_SECTION
+ mfspr r10,SPRN_PPR
+ std r10,HSTATE_PPR(r13)
+END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
+ HMT_MEDIUM
+ mfcr r10
+ std r12,HSTATE_SCRATCH0(r13)
+ sldi r12,r10,32
+ ori r12,r12,0xc00
+ ld r10,PACA_EXGEN+EX_R10(r13)
.global kvmppc_interrupt
.balign IFETCH_ALIGN_BYTES
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* Re: [PATCH v4 15/46] KVM: PPC: Book3S 64: Move hcall early register setup to KVM
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-04-01 5:41 ` Paul Mackerras
-1 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-04-01 5:41 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc
On Tue, Mar 23, 2021 at 11:02:34AM +1000, Nicholas Piggin wrote:
> System calls / hcalls have a different calling convention than
> other interrupts, so there is code in the KVMTEST to massage these
> into the same form as other interrupt handlers.
>
> Move this work into the KVM hcall handler. This means teaching KVM
> a little more about the low level interrupt handler setup, PACA save
> areas, etc., although that's not obviously worse than the current
> approach of coming up with an entirely different interrupt register
> / save convention.
[snip]
> @@ -1964,29 +1948,8 @@ EXC_VIRT_END(system_call, 0x4c00, 0x100)
>
> #ifdef CONFIG_KVM_BOOK3S_64_HANDLER
> TRAMP_REAL_BEGIN(system_call_kvm)
> - /*
> - * This is a hcall, so register convention is as above, with these
> - * differences:
I haven't checked all the code changes in detail yet, but this comment
at least is slightly misleading, since under PR KVM, system calls (to
the guest kernel) and hypercalls both come through this path.
Paul.
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 15/46] KVM: PPC: Book3S 64: Move hcall early register setup to KVM
@ 2021-04-01 5:41 ` Paul Mackerras
0 siblings, 0 replies; 188+ messages in thread
From: Paul Mackerras @ 2021-04-01 5:41 UTC (permalink / raw)
To: Nicholas Piggin; +Cc: linuxppc-dev, kvm-ppc
On Tue, Mar 23, 2021 at 11:02:34AM +1000, Nicholas Piggin wrote:
> System calls / hcalls have a different calling convention than
> other interrupts, so there is code in the KVMTEST to massage these
> into the same form as other interrupt handlers.
>
> Move this work into the KVM hcall handler. This means teaching KVM
> a little more about the low level interrupt handler setup, PACA save
> areas, etc., although that's not obviously worse than the current
> approach of coming up with an entirely different interrupt register
> / save convention.
[snip]
> @@ -1964,29 +1948,8 @@ EXC_VIRT_END(system_call, 0x4c00, 0x100)
>
> #ifdef CONFIG_KVM_BOOK3S_64_HANDLER
> TRAMP_REAL_BEGIN(system_call_kvm)
> - /*
> - * This is a hcall, so register convention is as above, with these
> - * differences:
I haven't checked all the code changes in detail yet, but this comment
at least is slightly misleading, since under PR KVM, system calls (to
the guest kernel) and hypercalls both come through this path.
Paul.
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 15/46] KVM: PPC: Book3S 64: Move hcall early register setup to KVM
2021-04-01 5:41 ` Paul Mackerras
@ 2021-04-01 10:03 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 10:03 UTC (permalink / raw)
To: Paul Mackerras; +Cc: linuxppc-dev, kvm-ppc
Excerpts from Paul Mackerras's message of April 1, 2021 3:41 pm:
> On Tue, Mar 23, 2021 at 11:02:34AM +1000, Nicholas Piggin wrote:
>> System calls / hcalls have a different calling convention than
>> other interrupts, so there is code in the KVMTEST to massage these
>> into the same form as other interrupt handlers.
>>
>> Move this work into the KVM hcall handler. This means teaching KVM
>> a little more about the low level interrupt handler setup, PACA save
>> areas, etc., although that's not obviously worse than the current
>> approach of coming up with an entirely different interrupt register
>> / save convention.
>
> [snip]
>
>> @@ -1964,29 +1948,8 @@ EXC_VIRT_END(system_call, 0x4c00, 0x100)
>>
>> #ifdef CONFIG_KVM_BOOK3S_64_HANDLER
>> TRAMP_REAL_BEGIN(system_call_kvm)
>> - /*
>> - * This is a hcall, so register convention is as above, with these
>> - * differences:
>
> I haven't checked all the code changes in detail yet, but this comment
> at least is slightly misleading, since under PR KVM, system calls (to
> the guest kernel) and hypercalls both come through this path.
Yeah good point, I'll update the comment at its destination.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 15/46] KVM: PPC: Book3S 64: Move hcall early register setup to KVM
@ 2021-04-01 10:03 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 10:03 UTC (permalink / raw)
To: Paul Mackerras; +Cc: linuxppc-dev, kvm-ppc
Excerpts from Paul Mackerras's message of April 1, 2021 3:41 pm:
> On Tue, Mar 23, 2021 at 11:02:34AM +1000, Nicholas Piggin wrote:
>> System calls / hcalls have a different calling convention than
>> other interrupts, so there is code in the KVMTEST to massage these
>> into the same form as other interrupt handlers.
>>
>> Move this work into the KVM hcall handler. This means teaching KVM
>> a little more about the low level interrupt handler setup, PACA save
>> areas, etc., although that's not obviously worse than the current
>> approach of coming up with an entirely different interrupt register
>> / save convention.
>
> [snip]
>
>> @@ -1964,29 +1948,8 @@ EXC_VIRT_END(system_call, 0x4c00, 0x100)
>>
>> #ifdef CONFIG_KVM_BOOK3S_64_HANDLER
>> TRAMP_REAL_BEGIN(system_call_kvm)
>> - /*
>> - * This is a hcall, so register convention is as above, with these
>> - * differences:
>
> I haven't checked all the code changes in detail yet, but this comment
> at least is slightly misleading, since under PR KVM, system calls (to
> the guest kernel) and hypercalls both come through this path.
Yeah good point, I'll update the comment at its destination.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 16/46] KVM: PPC: Book3S 64: Move interrupt early register setup to KVM
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Like the earlier patch for hcalls, KVM interrupt entry requires a
different calling convention than the Linux interrupt handlers
set up. Move the code that converts from one to the other into KVM.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kernel/exceptions-64s.S | 131 +++++----------------------
arch/powerpc/kvm/book3s_64_entry.S | 50 +++++++++-
2 files changed, 71 insertions(+), 110 deletions(-)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index a23feaa445b5..115cf79f3e82 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -187,7 +187,6 @@ do_define_int n
.endif
.endm
-#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
/*
* All interrupts which set HSRR registers, as well as SRESET and MCE and
* syscall when invoked with "sc 1" switch to MSR[HV]=1 (HVMODE) to be taken,
@@ -220,54 +219,25 @@ do_define_int n
* to KVM to handle.
*/
-.macro KVMTEST name
+.macro KVMTEST name handler
+#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
lbz r10,HSTATE_IN_GUEST(r13)
cmpwi r10,0
- bne \name\()_kvm
-.endm
-
-.macro GEN_KVM name
- .balign IFETCH_ALIGN_BYTES
-\name\()_kvm:
-
-BEGIN_FTR_SECTION
- ld r10,IAREA+EX_CFAR(r13)
- std r10,HSTATE_CFAR(r13)
-END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
-
- ld r10,IAREA+EX_CTR(r13)
- mtctr r10
-BEGIN_FTR_SECTION
- ld r10,IAREA+EX_PPR(r13)
- std r10,HSTATE_PPR(r13)
-END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
- ld r11,IAREA+EX_R11(r13)
- ld r12,IAREA+EX_R12(r13)
- std r12,HSTATE_SCRATCH0(r13)
- sldi r12,r9,32
- ld r9,IAREA+EX_R9(r13)
- ld r10,IAREA+EX_R10(r13)
/* HSRR variants have the 0x2 bit added to their trap number */
.if IHSRR_IF_HVMODE
BEGIN_FTR_SECTION
- ori r12,r12,(IVEC + 0x2)
+ li r10,(IVEC + 0x2)
FTR_SECTION_ELSE
- ori r12,r12,(IVEC)
+ li r10,(IVEC)
ALT_FTR_SECTION_END_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206)
.elseif IHSRR
- ori r12,r12,(IVEC+ 0x2)
+ li r10,(IVEC + 0x2)
.else
- ori r12,r12,(IVEC)
+ li r10,(IVEC)
.endif
- b kvmppc_interrupt
-.endm
-
-#else
-.macro KVMTEST name
-.endm
-.macro GEN_KVM name
-.endm
+ bne \handler
#endif
+.endm
/*
* This is the BOOK3S interrupt entry code macro.
@@ -409,7 +379,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
DEFINE_FIXED_SYMBOL(\name\()_common_real)
\name\()_common_real:
.if IKVM_REAL
- KVMTEST \name
+ KVMTEST \name kvm_interrupt
.endif
ld r10,PACAKMSR(r13) /* get MSR value for kernel */
@@ -432,7 +402,7 @@ DEFINE_FIXED_SYMBOL(\name\()_common_real)
DEFINE_FIXED_SYMBOL(\name\()_common_virt)
\name\()_common_virt:
.if IKVM_VIRT
- KVMTEST \name
+ KVMTEST \name kvm_interrupt
1:
.endif
.endif /* IVIRT */
@@ -446,7 +416,7 @@ DEFINE_FIXED_SYMBOL(\name\()_common_virt)
DEFINE_FIXED_SYMBOL(\name\()_common_real)
\name\()_common_real:
.if IKVM_REAL
- KVMTEST \name
+ KVMTEST \name kvm_interrupt
.endif
.endm
@@ -967,8 +937,6 @@ EXC_COMMON_BEGIN(system_reset_common)
EXCEPTION_RESTORE_REGS
RFI_TO_USER_OR_KERNEL
- GEN_KVM system_reset
-
/**
* Interrupt 0x200 - Machine Check Interrupt (MCE).
@@ -1132,7 +1100,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206)
/*
* Check if we are coming from guest. If yes, then run the normal
* exception handler which will take the
- * machine_check_kvm->kvmppc_interrupt branch to deliver the MC event
+ * machine_check_kvm->kvm_interrupt branch to deliver the MC event
* to guest.
*/
lbz r11,HSTATE_IN_GUEST(r13)
@@ -1203,8 +1171,6 @@ EXC_COMMON_BEGIN(machine_check_common)
bl machine_check_exception
b interrupt_return
- GEN_KVM machine_check
-
#ifdef CONFIG_PPC_P7_NAP
/*
@@ -1339,8 +1305,6 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX)
REST_NVGPRS(r1)
b interrupt_return
- GEN_KVM data_access
-
/**
* Interrupt 0x380 - Data Segment Interrupt (DSLB).
@@ -1390,8 +1354,6 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX)
bl do_bad_slb_fault
b interrupt_return
- GEN_KVM data_access_slb
-
/**
* Interrupt 0x400 - Instruction Storage Interrupt (ISI).
@@ -1428,8 +1390,6 @@ MMU_FTR_SECTION_ELSE
ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX)
b interrupt_return
- GEN_KVM instruction_access
-
/**
* Interrupt 0x480 - Instruction Segment Interrupt (ISLB).
@@ -1474,8 +1434,6 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX)
bl do_bad_slb_fault
b interrupt_return
- GEN_KVM instruction_access_slb
-
/**
* Interrupt 0x500 - External Interrupt.
@@ -1521,8 +1479,6 @@ EXC_COMMON_BEGIN(hardware_interrupt_common)
bl do_IRQ
b interrupt_return
- GEN_KVM hardware_interrupt
-
/**
* Interrupt 0x600 - Alignment Interrupt
@@ -1550,8 +1506,6 @@ EXC_COMMON_BEGIN(alignment_common)
REST_NVGPRS(r1) /* instruction emulation may change GPRs */
b interrupt_return
- GEN_KVM alignment
-
/**
* Interrupt 0x700 - Program Interrupt (program check).
@@ -1659,8 +1613,6 @@ EXC_COMMON_BEGIN(program_check_common)
REST_NVGPRS(r1) /* instruction emulation may change GPRs */
b interrupt_return
- GEN_KVM program_check
-
/*
* Interrupt 0x800 - Floating-Point Unavailable Interrupt.
@@ -1710,8 +1662,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_TM)
b interrupt_return
#endif
- GEN_KVM fp_unavailable
-
/**
* Interrupt 0x900 - Decrementer Interrupt.
@@ -1751,8 +1701,6 @@ EXC_COMMON_BEGIN(decrementer_common)
bl timer_interrupt
b interrupt_return
- GEN_KVM decrementer
-
/**
* Interrupt 0x980 - Hypervisor Decrementer Interrupt.
@@ -1798,8 +1746,6 @@ EXC_COMMON_BEGIN(hdecrementer_common)
ld r13,PACA_EXGEN+EX_R13(r13)
HRFI_TO_KERNEL
- GEN_KVM hdecrementer
-
/**
* Interrupt 0xa00 - Directed Privileged Doorbell Interrupt.
@@ -1840,8 +1786,6 @@ EXC_COMMON_BEGIN(doorbell_super_common)
#endif
b interrupt_return
- GEN_KVM doorbell_super
-
EXC_REAL_NONE(0xb00, 0x100)
EXC_VIRT_NONE(0x4b00, 0x100)
@@ -1891,7 +1835,7 @@ INT_DEFINE_END(system_call)
GET_PACA(r13)
std r10,PACA_EXGEN+EX_R10(r13)
INTERRUPT_TO_KERNEL
- KVMTEST system_call /* uses r10, branch to system_call_kvm */
+ KVMTEST system_call kvm_hcall /* uses r10, branch to kvm_hcall */
mfctr r9
#else
mr r9,r13
@@ -1947,7 +1891,7 @@ EXC_VIRT_BEGIN(system_call, 0x4c00, 0x100)
EXC_VIRT_END(system_call, 0x4c00, 0x100)
#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
-TRAMP_REAL_BEGIN(system_call_kvm)
+TRAMP_REAL_BEGIN(kvm_hcall)
mfctr r10
SET_SCRATCH0(r10) /* Save r13 in SCRATCH0 */
#ifdef CONFIG_RELOCATABLE
@@ -1987,8 +1931,6 @@ EXC_COMMON_BEGIN(single_step_common)
bl single_step_exception
b interrupt_return
- GEN_KVM single_step
-
/**
* Interrupt 0xe00 - Hypervisor Data Storage Interrupt (HDSI).
@@ -2027,8 +1969,6 @@ MMU_FTR_SECTION_ELSE
ALT_MMU_FTR_SECTION_END_IFSET(MMU_FTR_TYPE_RADIX)
b interrupt_return
- GEN_KVM h_data_storage
-
/**
* Interrupt 0xe20 - Hypervisor Instruction Storage Interrupt (HISI).
@@ -2054,8 +1994,6 @@ EXC_COMMON_BEGIN(h_instr_storage_common)
bl unknown_exception
b interrupt_return
- GEN_KVM h_instr_storage
-
/**
* Interrupt 0xe40 - Hypervisor Emulation Assistance Interrupt.
@@ -2080,8 +2018,6 @@ EXC_COMMON_BEGIN(emulation_assist_common)
REST_NVGPRS(r1) /* instruction emulation may change GPRs */
b interrupt_return
- GEN_KVM emulation_assist
-
/**
* Interrupt 0xe60 - Hypervisor Maintenance Interrupt (HMI).
@@ -2153,8 +2089,6 @@ EXC_COMMON_BEGIN(hmi_exception_early_common)
EXCEPTION_RESTORE_REGS hsrr=1
GEN_INT_ENTRY hmi_exception, virt=0
- GEN_KVM hmi_exception_early
-
EXC_COMMON_BEGIN(hmi_exception_common)
GEN_COMMON hmi_exception
FINISH_NAP
@@ -2162,8 +2096,6 @@ EXC_COMMON_BEGIN(hmi_exception_common)
bl handle_hmi_exception
b interrupt_return
- GEN_KVM hmi_exception
-
/**
* Interrupt 0xe80 - Directed Hypervisor Doorbell Interrupt.
@@ -2195,8 +2127,6 @@ EXC_COMMON_BEGIN(h_doorbell_common)
#endif
b interrupt_return
- GEN_KVM h_doorbell
-
/**
* Interrupt 0xea0 - Hypervisor Virtualization Interrupt.
@@ -2224,8 +2154,6 @@ EXC_COMMON_BEGIN(h_virt_irq_common)
bl do_IRQ
b interrupt_return
- GEN_KVM h_virt_irq
-
EXC_REAL_NONE(0xec0, 0x20)
EXC_VIRT_NONE(0x4ec0, 0x20)
@@ -2270,8 +2198,6 @@ EXC_COMMON_BEGIN(performance_monitor_common)
bl performance_monitor_exception
b interrupt_return
- GEN_KVM performance_monitor
-
/**
* Interrupt 0xf20 - Vector Unavailable Interrupt.
@@ -2321,8 +2247,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
bl altivec_unavailable_exception
b interrupt_return
- GEN_KVM altivec_unavailable
-
/**
* Interrupt 0xf40 - VSX Unavailable Interrupt.
@@ -2371,8 +2295,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
bl vsx_unavailable_exception
b interrupt_return
- GEN_KVM vsx_unavailable
-
/**
* Interrupt 0xf60 - Facility Unavailable Interrupt.
@@ -2401,8 +2323,6 @@ EXC_COMMON_BEGIN(facility_unavailable_common)
REST_NVGPRS(r1) /* instruction emulation may change GPRs */
b interrupt_return
- GEN_KVM facility_unavailable
-
/**
* Interrupt 0xf60 - Hypervisor Facility Unavailable Interrupt.
@@ -2431,8 +2351,6 @@ EXC_COMMON_BEGIN(h_facility_unavailable_common)
REST_NVGPRS(r1) /* XXX Shouldn't be necessary in practice */
b interrupt_return
- GEN_KVM h_facility_unavailable
-
EXC_REAL_NONE(0xfa0, 0x20)
EXC_VIRT_NONE(0x4fa0, 0x20)
@@ -2462,8 +2380,6 @@ EXC_COMMON_BEGIN(cbe_system_error_common)
bl cbe_system_error_exception
b interrupt_return
- GEN_KVM cbe_system_error
-
#else /* CONFIG_CBE_RAS */
EXC_REAL_NONE(0x1200, 0x100)
EXC_VIRT_NONE(0x5200, 0x100)
@@ -2489,8 +2405,6 @@ EXC_COMMON_BEGIN(instruction_breakpoint_common)
bl instruction_breakpoint_exception
b interrupt_return
- GEN_KVM instruction_breakpoint
-
EXC_REAL_NONE(0x1400, 0x100)
EXC_VIRT_NONE(0x5400, 0x100)
@@ -2611,8 +2525,6 @@ EXC_COMMON_BEGIN(denorm_exception_common)
bl unknown_exception
b interrupt_return
- GEN_KVM denorm_exception
-
#ifdef CONFIG_CBE_RAS
INT_DEFINE_BEGIN(cbe_maintenance)
@@ -2630,8 +2542,6 @@ EXC_COMMON_BEGIN(cbe_maintenance_common)
bl cbe_maintenance_exception
b interrupt_return
- GEN_KVM cbe_maintenance
-
#else /* CONFIG_CBE_RAS */
EXC_REAL_NONE(0x1600, 0x100)
EXC_VIRT_NONE(0x5600, 0x100)
@@ -2662,8 +2572,6 @@ EXC_COMMON_BEGIN(altivec_assist_common)
#endif
b interrupt_return
- GEN_KVM altivec_assist
-
#ifdef CONFIG_CBE_RAS
INT_DEFINE_BEGIN(cbe_thermal)
@@ -2681,8 +2589,6 @@ EXC_COMMON_BEGIN(cbe_thermal_common)
bl cbe_thermal_exception
b interrupt_return
- GEN_KVM cbe_thermal
-
#else /* CONFIG_CBE_RAS */
EXC_REAL_NONE(0x1800, 0x100)
EXC_VIRT_NONE(0x5800, 0x100)
@@ -2935,6 +2841,15 @@ TRAMP_REAL_BEGIN(rfscv_flush_fallback)
USE_TEXT_SECTION()
+#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
+kvm_interrupt:
+ /*
+ * The conditional branch in KVMTEST can't reach all the way,
+ * make a stub.
+ */
+ b kvmppc_interrupt
+#endif
+
_GLOBAL(do_uaccess_flush)
UACCESS_FLUSH_FIXUP_SECTION
nop
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index dea8c5bb207b..642d5fdc641e 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -37,15 +37,61 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
sldi r12,r10,32
ori r12,r12,0xc00
ld r10,PACA_EXGEN+EX_R10(r13)
+ b do_kvm_interrupt
+/*
+ * KVM interrupt entry occurs after GEN_INT_ENTRY runs, and follows that
+ * call convention:
+ *
+ * guest R9-R13, CTR, CFAR, PPR saved in PACA EX_xxx save area
+ * guest (H)DAR, (H)DSISR are also in the save area for relevant interrupts
+ * guest R13 also saved in SCRATCH0
+ * R13 = PACA
+ * R11 = (H)SRR0
+ * R12 = (H)SRR1
+ * R9 = guest CR
+ * PPR is set to medium
+ *
+ * With the addition for KVM:
+ * R10 = trap vector
+ */
.global kvmppc_interrupt
.balign IFETCH_ALIGN_BYTES
kvmppc_interrupt:
+ li r11,PACA_EXGEN
+ cmpdi r10,0x200
+ bgt+ 1f
+ li r11,PACA_EXMC
+ beq 1f
+ li r11,PACA_EXNMI
+1: add r11,r11,r13
+
+BEGIN_FTR_SECTION
+ ld r12,EX_CFAR(r11)
+ std r12,HSTATE_CFAR(r13)
+END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
+ ld r12,EX_CTR(r11)
+ mtctr r12
+BEGIN_FTR_SECTION
+ ld r12,EX_PPR(r11)
+ std r12,HSTATE_PPR(r13)
+END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
+ ld r12,EX_R12(r11)
+ std r12,HSTATE_SCRATCH0(r13)
+ sldi r12,r9,32
+ or r12,r12,r10
+ ld r9,EX_R9(r11)
+ ld r10,EX_R10(r11)
+ ld r11,EX_R11(r11)
+
+do_kvm_interrupt:
/*
- * Register contents:
+ * Hcalls and other interrupts come here after normalising register
+ * contents and save locations:
+ *
* R12 = (guest CR << 32) | interrupt vector
* R13 = PACA
- * guest R12 saved in shadow VCPU SCRATCH0
+ * guest R12 saved in shadow HSTATE_SCRATCH0
* guest R13 saved in SPRN_SCRATCH0
*/
std r9,HSTATE_SCRATCH2(r13)
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 16/46] KVM: PPC: Book3S 64: Move interrupt early register setup to KVM
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Like the earlier patch for hcalls, KVM interrupt entry requires a
different calling convention than the Linux interrupt handlers
set up. Move the code that converts from one to the other into KVM.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kernel/exceptions-64s.S | 131 +++++----------------------
arch/powerpc/kvm/book3s_64_entry.S | 50 +++++++++-
2 files changed, 71 insertions(+), 110 deletions(-)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index a23feaa445b5..115cf79f3e82 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -187,7 +187,6 @@ do_define_int n
.endif
.endm
-#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
/*
* All interrupts which set HSRR registers, as well as SRESET and MCE and
* syscall when invoked with "sc 1" switch to MSR[HV]=1 (HVMODE) to be taken,
@@ -220,54 +219,25 @@ do_define_int n
* to KVM to handle.
*/
-.macro KVMTEST name
+.macro KVMTEST name handler
+#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
lbz r10,HSTATE_IN_GUEST(r13)
cmpwi r10,0
- bne \name\()_kvm
-.endm
-
-.macro GEN_KVM name
- .balign IFETCH_ALIGN_BYTES
-\name\()_kvm:
-
-BEGIN_FTR_SECTION
- ld r10,IAREA+EX_CFAR(r13)
- std r10,HSTATE_CFAR(r13)
-END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
-
- ld r10,IAREA+EX_CTR(r13)
- mtctr r10
-BEGIN_FTR_SECTION
- ld r10,IAREA+EX_PPR(r13)
- std r10,HSTATE_PPR(r13)
-END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
- ld r11,IAREA+EX_R11(r13)
- ld r12,IAREA+EX_R12(r13)
- std r12,HSTATE_SCRATCH0(r13)
- sldi r12,r9,32
- ld r9,IAREA+EX_R9(r13)
- ld r10,IAREA+EX_R10(r13)
/* HSRR variants have the 0x2 bit added to their trap number */
.if IHSRR_IF_HVMODE
BEGIN_FTR_SECTION
- ori r12,r12,(IVEC + 0x2)
+ li r10,(IVEC + 0x2)
FTR_SECTION_ELSE
- ori r12,r12,(IVEC)
+ li r10,(IVEC)
ALT_FTR_SECTION_END_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206)
.elseif IHSRR
- ori r12,r12,(IVEC+ 0x2)
+ li r10,(IVEC + 0x2)
.else
- ori r12,r12,(IVEC)
+ li r10,(IVEC)
.endif
- b kvmppc_interrupt
-.endm
-
-#else
-.macro KVMTEST name
-.endm
-.macro GEN_KVM name
-.endm
+ bne \handler
#endif
+.endm
/*
* This is the BOOK3S interrupt entry code macro.
@@ -409,7 +379,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
DEFINE_FIXED_SYMBOL(\name\()_common_real)
\name\()_common_real:
.if IKVM_REAL
- KVMTEST \name
+ KVMTEST \name kvm_interrupt
.endif
ld r10,PACAKMSR(r13) /* get MSR value for kernel */
@@ -432,7 +402,7 @@ DEFINE_FIXED_SYMBOL(\name\()_common_real)
DEFINE_FIXED_SYMBOL(\name\()_common_virt)
\name\()_common_virt:
.if IKVM_VIRT
- KVMTEST \name
+ KVMTEST \name kvm_interrupt
1:
.endif
.endif /* IVIRT */
@@ -446,7 +416,7 @@ DEFINE_FIXED_SYMBOL(\name\()_common_virt)
DEFINE_FIXED_SYMBOL(\name\()_common_real)
\name\()_common_real:
.if IKVM_REAL
- KVMTEST \name
+ KVMTEST \name kvm_interrupt
.endif
.endm
@@ -967,8 +937,6 @@ EXC_COMMON_BEGIN(system_reset_common)
EXCEPTION_RESTORE_REGS
RFI_TO_USER_OR_KERNEL
- GEN_KVM system_reset
-
/**
* Interrupt 0x200 - Machine Check Interrupt (MCE).
@@ -1132,7 +1100,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206)
/*
* Check if we are coming from guest. If yes, then run the normal
* exception handler which will take the
- * machine_check_kvm->kvmppc_interrupt branch to deliver the MC event
+ * machine_check_kvm->kvm_interrupt branch to deliver the MC event
* to guest.
*/
lbz r11,HSTATE_IN_GUEST(r13)
@@ -1203,8 +1171,6 @@ EXC_COMMON_BEGIN(machine_check_common)
bl machine_check_exception
b interrupt_return
- GEN_KVM machine_check
-
#ifdef CONFIG_PPC_P7_NAP
/*
@@ -1339,8 +1305,6 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX)
REST_NVGPRS(r1)
b interrupt_return
- GEN_KVM data_access
-
/**
* Interrupt 0x380 - Data Segment Interrupt (DSLB).
@@ -1390,8 +1354,6 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX)
bl do_bad_slb_fault
b interrupt_return
- GEN_KVM data_access_slb
-
/**
* Interrupt 0x400 - Instruction Storage Interrupt (ISI).
@@ -1428,8 +1390,6 @@ MMU_FTR_SECTION_ELSE
ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX)
b interrupt_return
- GEN_KVM instruction_access
-
/**
* Interrupt 0x480 - Instruction Segment Interrupt (ISLB).
@@ -1474,8 +1434,6 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX)
bl do_bad_slb_fault
b interrupt_return
- GEN_KVM instruction_access_slb
-
/**
* Interrupt 0x500 - External Interrupt.
@@ -1521,8 +1479,6 @@ EXC_COMMON_BEGIN(hardware_interrupt_common)
bl do_IRQ
b interrupt_return
- GEN_KVM hardware_interrupt
-
/**
* Interrupt 0x600 - Alignment Interrupt
@@ -1550,8 +1506,6 @@ EXC_COMMON_BEGIN(alignment_common)
REST_NVGPRS(r1) /* instruction emulation may change GPRs */
b interrupt_return
- GEN_KVM alignment
-
/**
* Interrupt 0x700 - Program Interrupt (program check).
@@ -1659,8 +1613,6 @@ EXC_COMMON_BEGIN(program_check_common)
REST_NVGPRS(r1) /* instruction emulation may change GPRs */
b interrupt_return
- GEN_KVM program_check
-
/*
* Interrupt 0x800 - Floating-Point Unavailable Interrupt.
@@ -1710,8 +1662,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_TM)
b interrupt_return
#endif
- GEN_KVM fp_unavailable
-
/**
* Interrupt 0x900 - Decrementer Interrupt.
@@ -1751,8 +1701,6 @@ EXC_COMMON_BEGIN(decrementer_common)
bl timer_interrupt
b interrupt_return
- GEN_KVM decrementer
-
/**
* Interrupt 0x980 - Hypervisor Decrementer Interrupt.
@@ -1798,8 +1746,6 @@ EXC_COMMON_BEGIN(hdecrementer_common)
ld r13,PACA_EXGEN+EX_R13(r13)
HRFI_TO_KERNEL
- GEN_KVM hdecrementer
-
/**
* Interrupt 0xa00 - Directed Privileged Doorbell Interrupt.
@@ -1840,8 +1786,6 @@ EXC_COMMON_BEGIN(doorbell_super_common)
#endif
b interrupt_return
- GEN_KVM doorbell_super
-
EXC_REAL_NONE(0xb00, 0x100)
EXC_VIRT_NONE(0x4b00, 0x100)
@@ -1891,7 +1835,7 @@ INT_DEFINE_END(system_call)
GET_PACA(r13)
std r10,PACA_EXGEN+EX_R10(r13)
INTERRUPT_TO_KERNEL
- KVMTEST system_call /* uses r10, branch to system_call_kvm */
+ KVMTEST system_call kvm_hcall /* uses r10, branch to kvm_hcall */
mfctr r9
#else
mr r9,r13
@@ -1947,7 +1891,7 @@ EXC_VIRT_BEGIN(system_call, 0x4c00, 0x100)
EXC_VIRT_END(system_call, 0x4c00, 0x100)
#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
-TRAMP_REAL_BEGIN(system_call_kvm)
+TRAMP_REAL_BEGIN(kvm_hcall)
mfctr r10
SET_SCRATCH0(r10) /* Save r13 in SCRATCH0 */
#ifdef CONFIG_RELOCATABLE
@@ -1987,8 +1931,6 @@ EXC_COMMON_BEGIN(single_step_common)
bl single_step_exception
b interrupt_return
- GEN_KVM single_step
-
/**
* Interrupt 0xe00 - Hypervisor Data Storage Interrupt (HDSI).
@@ -2027,8 +1969,6 @@ MMU_FTR_SECTION_ELSE
ALT_MMU_FTR_SECTION_END_IFSET(MMU_FTR_TYPE_RADIX)
b interrupt_return
- GEN_KVM h_data_storage
-
/**
* Interrupt 0xe20 - Hypervisor Instruction Storage Interrupt (HISI).
@@ -2054,8 +1994,6 @@ EXC_COMMON_BEGIN(h_instr_storage_common)
bl unknown_exception
b interrupt_return
- GEN_KVM h_instr_storage
-
/**
* Interrupt 0xe40 - Hypervisor Emulation Assistance Interrupt.
@@ -2080,8 +2018,6 @@ EXC_COMMON_BEGIN(emulation_assist_common)
REST_NVGPRS(r1) /* instruction emulation may change GPRs */
b interrupt_return
- GEN_KVM emulation_assist
-
/**
* Interrupt 0xe60 - Hypervisor Maintenance Interrupt (HMI).
@@ -2153,8 +2089,6 @@ EXC_COMMON_BEGIN(hmi_exception_early_common)
EXCEPTION_RESTORE_REGS hsrr=1
GEN_INT_ENTRY hmi_exception, virt=0
- GEN_KVM hmi_exception_early
-
EXC_COMMON_BEGIN(hmi_exception_common)
GEN_COMMON hmi_exception
FINISH_NAP
@@ -2162,8 +2096,6 @@ EXC_COMMON_BEGIN(hmi_exception_common)
bl handle_hmi_exception
b interrupt_return
- GEN_KVM hmi_exception
-
/**
* Interrupt 0xe80 - Directed Hypervisor Doorbell Interrupt.
@@ -2195,8 +2127,6 @@ EXC_COMMON_BEGIN(h_doorbell_common)
#endif
b interrupt_return
- GEN_KVM h_doorbell
-
/**
* Interrupt 0xea0 - Hypervisor Virtualization Interrupt.
@@ -2224,8 +2154,6 @@ EXC_COMMON_BEGIN(h_virt_irq_common)
bl do_IRQ
b interrupt_return
- GEN_KVM h_virt_irq
-
EXC_REAL_NONE(0xec0, 0x20)
EXC_VIRT_NONE(0x4ec0, 0x20)
@@ -2270,8 +2198,6 @@ EXC_COMMON_BEGIN(performance_monitor_common)
bl performance_monitor_exception
b interrupt_return
- GEN_KVM performance_monitor
-
/**
* Interrupt 0xf20 - Vector Unavailable Interrupt.
@@ -2321,8 +2247,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
bl altivec_unavailable_exception
b interrupt_return
- GEN_KVM altivec_unavailable
-
/**
* Interrupt 0xf40 - VSX Unavailable Interrupt.
@@ -2371,8 +2295,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
bl vsx_unavailable_exception
b interrupt_return
- GEN_KVM vsx_unavailable
-
/**
* Interrupt 0xf60 - Facility Unavailable Interrupt.
@@ -2401,8 +2323,6 @@ EXC_COMMON_BEGIN(facility_unavailable_common)
REST_NVGPRS(r1) /* instruction emulation may change GPRs */
b interrupt_return
- GEN_KVM facility_unavailable
-
/**
* Interrupt 0xf60 - Hypervisor Facility Unavailable Interrupt.
@@ -2431,8 +2351,6 @@ EXC_COMMON_BEGIN(h_facility_unavailable_common)
REST_NVGPRS(r1) /* XXX Shouldn't be necessary in practice */
b interrupt_return
- GEN_KVM h_facility_unavailable
-
EXC_REAL_NONE(0xfa0, 0x20)
EXC_VIRT_NONE(0x4fa0, 0x20)
@@ -2462,8 +2380,6 @@ EXC_COMMON_BEGIN(cbe_system_error_common)
bl cbe_system_error_exception
b interrupt_return
- GEN_KVM cbe_system_error
-
#else /* CONFIG_CBE_RAS */
EXC_REAL_NONE(0x1200, 0x100)
EXC_VIRT_NONE(0x5200, 0x100)
@@ -2489,8 +2405,6 @@ EXC_COMMON_BEGIN(instruction_breakpoint_common)
bl instruction_breakpoint_exception
b interrupt_return
- GEN_KVM instruction_breakpoint
-
EXC_REAL_NONE(0x1400, 0x100)
EXC_VIRT_NONE(0x5400, 0x100)
@@ -2611,8 +2525,6 @@ EXC_COMMON_BEGIN(denorm_exception_common)
bl unknown_exception
b interrupt_return
- GEN_KVM denorm_exception
-
#ifdef CONFIG_CBE_RAS
INT_DEFINE_BEGIN(cbe_maintenance)
@@ -2630,8 +2542,6 @@ EXC_COMMON_BEGIN(cbe_maintenance_common)
bl cbe_maintenance_exception
b interrupt_return
- GEN_KVM cbe_maintenance
-
#else /* CONFIG_CBE_RAS */
EXC_REAL_NONE(0x1600, 0x100)
EXC_VIRT_NONE(0x5600, 0x100)
@@ -2662,8 +2572,6 @@ EXC_COMMON_BEGIN(altivec_assist_common)
#endif
b interrupt_return
- GEN_KVM altivec_assist
-
#ifdef CONFIG_CBE_RAS
INT_DEFINE_BEGIN(cbe_thermal)
@@ -2681,8 +2589,6 @@ EXC_COMMON_BEGIN(cbe_thermal_common)
bl cbe_thermal_exception
b interrupt_return
- GEN_KVM cbe_thermal
-
#else /* CONFIG_CBE_RAS */
EXC_REAL_NONE(0x1800, 0x100)
EXC_VIRT_NONE(0x5800, 0x100)
@@ -2935,6 +2841,15 @@ TRAMP_REAL_BEGIN(rfscv_flush_fallback)
USE_TEXT_SECTION()
+#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
+kvm_interrupt:
+ /*
+ * The conditional branch in KVMTEST can't reach all the way,
+ * make a stub.
+ */
+ b kvmppc_interrupt
+#endif
+
_GLOBAL(do_uaccess_flush)
UACCESS_FLUSH_FIXUP_SECTION
nop
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index dea8c5bb207b..642d5fdc641e 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -37,15 +37,61 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
sldi r12,r10,32
ori r12,r12,0xc00
ld r10,PACA_EXGEN+EX_R10(r13)
+ b do_kvm_interrupt
+/*
+ * KVM interrupt entry occurs after GEN_INT_ENTRY runs, and follows that
+ * call convention:
+ *
+ * guest R9-R13, CTR, CFAR, PPR saved in PACA EX_xxx save area
+ * guest (H)DAR, (H)DSISR are also in the save area for relevant interrupts
+ * guest R13 also saved in SCRATCH0
+ * R13 = PACA
+ * R11 = (H)SRR0
+ * R12 = (H)SRR1
+ * R9 = guest CR
+ * PPR is set to medium
+ *
+ * With the addition for KVM:
+ * R10 = trap vector
+ */
.global kvmppc_interrupt
.balign IFETCH_ALIGN_BYTES
kvmppc_interrupt:
+ li r11,PACA_EXGEN
+ cmpdi r10,0x200
+ bgt+ 1f
+ li r11,PACA_EXMC
+ beq 1f
+ li r11,PACA_EXNMI
+1: add r11,r11,r13
+
+BEGIN_FTR_SECTION
+ ld r12,EX_CFAR(r11)
+ std r12,HSTATE_CFAR(r13)
+END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
+ ld r12,EX_CTR(r11)
+ mtctr r12
+BEGIN_FTR_SECTION
+ ld r12,EX_PPR(r11)
+ std r12,HSTATE_PPR(r13)
+END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
+ ld r12,EX_R12(r11)
+ std r12,HSTATE_SCRATCH0(r13)
+ sldi r12,r9,32
+ or r12,r12,r10
+ ld r9,EX_R9(r11)
+ ld r10,EX_R10(r11)
+ ld r11,EX_R11(r11)
+
+do_kvm_interrupt:
/*
- * Register contents:
+ * Hcalls and other interrupts come here after normalising register
+ * contents and save locations:
+ *
* R12 = (guest CR << 32) | interrupt vector
* R13 = PACA
- * guest R12 saved in shadow VCPU SCRATCH0
+ * guest R12 saved in shadow HSTATE_SCRATCH0
* guest R13 saved in SPRN_SCRATCH0
*/
std r9,HSTATE_SCRATCH2(r13)
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 17/46] KVM: PPC: Book3S 64: move bad_host_intr check to HV handler
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Alexey Kardashevskiy, linuxppc-dev, Nicholas Piggin
This is not used by PR KVM.
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_64_entry.S | 3 ---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 4 +++-
arch/powerpc/kvm/book3s_segment.S | 7 +++++++
3 files changed, 10 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index 642d5fdc641e..b6149df21de3 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -100,11 +100,8 @@ do_kvm_interrupt:
beq- .Lmaybe_skip
.Lno_skip:
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
- cmpwi r9,KVM_GUEST_MODE_HOST_HV
- beq kvmppc_bad_host_intr
#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
cmpwi r9,KVM_GUEST_MODE_GUEST
- ld r9,HSTATE_SCRATCH2(r13)
beq kvmppc_interrupt_pr
#endif
b kvmppc_interrupt_hv
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index f976efb7e4a9..75405ef53238 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -1265,6 +1265,7 @@ hdec_soon:
kvmppc_interrupt_hv:
/*
* Register contents:
+ * R9 = HSTATE_IN_GUEST
* R12 = (guest CR << 32) | interrupt vector
* R13 = PACA
* guest R12 saved in shadow VCPU SCRATCH0
@@ -1272,6 +1273,8 @@ kvmppc_interrupt_hv:
* guest R9 saved in HSTATE_SCRATCH2
*/
/* We're now back in the host but in guest MMU context */
+ cmpwi r9,KVM_GUEST_MODE_HOST_HV
+ beq kvmppc_bad_host_intr
li r9, KVM_GUEST_MODE_HOST_HV
stb r9, HSTATE_IN_GUEST(r13)
@@ -3272,7 +3275,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_P9_TM_HV_ASSIST)
* cfar is saved in HSTATE_CFAR(r13)
* ppr is saved in HSTATE_PPR(r13)
*/
-.global kvmppc_bad_host_intr
kvmppc_bad_host_intr:
/*
* Switch to the emergency stack, but start half-way down in
diff --git a/arch/powerpc/kvm/book3s_segment.S b/arch/powerpc/kvm/book3s_segment.S
index 1f492aa4c8d6..ef1d88b869bf 100644
--- a/arch/powerpc/kvm/book3s_segment.S
+++ b/arch/powerpc/kvm/book3s_segment.S
@@ -167,8 +167,15 @@ kvmppc_interrupt_pr:
* R12 = (guest CR << 32) | exit handler id
* R13 = PACA
* HSTATE.SCRATCH0 = guest R12
+ *
+ * If HV is possible, additionally:
+ * R9 = HSTATE_IN_GUEST
+ * HSTATE.SCRATCH2 = guest R9
*/
#ifdef CONFIG_PPC64
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+ ld r9,HSTATE_SCRATCH2(r13)
+#endif
/* Match 32-bit entry */
rotldi r12, r12, 32 /* Flip R12 halves for stw */
stw r12, HSTATE_SCRATCH1(r13) /* CR is now in the low half */
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 17/46] KVM: PPC: Book3S 64: move bad_host_intr check to HV handler
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Alexey Kardashevskiy, linuxppc-dev, Nicholas Piggin
This is not used by PR KVM.
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_64_entry.S | 3 ---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 4 +++-
arch/powerpc/kvm/book3s_segment.S | 7 +++++++
3 files changed, 10 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index 642d5fdc641e..b6149df21de3 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -100,11 +100,8 @@ do_kvm_interrupt:
beq- .Lmaybe_skip
.Lno_skip:
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
- cmpwi r9,KVM_GUEST_MODE_HOST_HV
- beq kvmppc_bad_host_intr
#ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE
cmpwi r9,KVM_GUEST_MODE_GUEST
- ld r9,HSTATE_SCRATCH2(r13)
beq kvmppc_interrupt_pr
#endif
b kvmppc_interrupt_hv
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index f976efb7e4a9..75405ef53238 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -1265,6 +1265,7 @@ hdec_soon:
kvmppc_interrupt_hv:
/*
* Register contents:
+ * R9 = HSTATE_IN_GUEST
* R12 = (guest CR << 32) | interrupt vector
* R13 = PACA
* guest R12 saved in shadow VCPU SCRATCH0
@@ -1272,6 +1273,8 @@ kvmppc_interrupt_hv:
* guest R9 saved in HSTATE_SCRATCH2
*/
/* We're now back in the host but in guest MMU context */
+ cmpwi r9,KVM_GUEST_MODE_HOST_HV
+ beq kvmppc_bad_host_intr
li r9, KVM_GUEST_MODE_HOST_HV
stb r9, HSTATE_IN_GUEST(r13)
@@ -3272,7 +3275,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_P9_TM_HV_ASSIST)
* cfar is saved in HSTATE_CFAR(r13)
* ppr is saved in HSTATE_PPR(r13)
*/
-.global kvmppc_bad_host_intr
kvmppc_bad_host_intr:
/*
* Switch to the emergency stack, but start half-way down in
diff --git a/arch/powerpc/kvm/book3s_segment.S b/arch/powerpc/kvm/book3s_segment.S
index 1f492aa4c8d6..ef1d88b869bf 100644
--- a/arch/powerpc/kvm/book3s_segment.S
+++ b/arch/powerpc/kvm/book3s_segment.S
@@ -167,8 +167,15 @@ kvmppc_interrupt_pr:
* R12 = (guest CR << 32) | exit handler id
* R13 = PACA
* HSTATE.SCRATCH0 = guest R12
+ *
+ * If HV is possible, additionally:
+ * R9 = HSTATE_IN_GUEST
+ * HSTATE.SCRATCH2 = guest R9
*/
#ifdef CONFIG_PPC64
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+ ld r9,HSTATE_SCRATCH2(r13)
+#endif
/* Match 32-bit entry */
rotldi r12, r12, 32 /* Flip R12 halves for stw */
stw r12, HSTATE_SCRATCH1(r13) /* CR is now in the low half */
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 18/46] KVM: PPC: Book3S 64: Minimise hcall handler calling convention differences
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Alexey Kardashevskiy, linuxppc-dev, Nicholas Piggin
This sets up the same calling convention from interrupt entry to
KVM interrupt handler for system calls as exists for other interrupt
types.
This is a better API, it uses a save area rather than SPR, and it has
more registers free to use. Using a single common API helps maintain
it, and it becomes easier to use in C in a later patch.
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kernel/exceptions-64s.S | 21 +++++++++++++++-
arch/powerpc/kvm/book3s_64_entry.S | 37 +++++++++-------------------
2 files changed, 32 insertions(+), 26 deletions(-)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 115cf79f3e82..4615057681c3 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -1892,8 +1892,27 @@ EXC_VIRT_END(system_call, 0x4c00, 0x100)
#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
TRAMP_REAL_BEGIN(kvm_hcall)
+ std r9,PACA_EXGEN+EX_R9(r13)
+ std r11,PACA_EXGEN+EX_R11(r13)
+ std r12,PACA_EXGEN+EX_R12(r13)
+ mfcr r9
mfctr r10
- SET_SCRATCH0(r10) /* Save r13 in SCRATCH0 */
+ std r10,PACA_EXGEN+EX_R13(r13)
+ li r10,0
+ std r10,PACA_EXGEN+EX_CFAR(r13)
+ std r10,PACA_EXGEN+EX_CTR(r13)
+ /*
+ * Save the PPR (on systems that support it) before changing to
+ * HMT_MEDIUM. That allows the KVM code to save that value into the
+ * guest state (it is the guest's PPR value).
+ */
+BEGIN_FTR_SECTION
+ mfspr r10,SPRN_PPR
+ std r10,PACA_EXGEN+EX_PPR(r13)
+END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
+
+ HMT_MEDIUM
+
#ifdef CONFIG_RELOCATABLE
/*
* Requires __LOAD_FAR_HANDLER beause kvmppc_hcall lives
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index b6149df21de3..de81ab69555b 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -11,33 +11,21 @@
* These are branched to from interrupt handlers in exception-64s.S which set
* IKVM_REAL or IKVM_VIRT, if HSTATE_IN_GUEST was found to be non-zero.
*/
+
+/*
+ * This is a hcall, so guest register call convention is as
+ * Documentation/powerpc/papr_hcalls.rst.
+ *
+ * The state of registers is as below, except CFAR is not saved, R13 is not
+ * in SCRATCH0, and R10 does not contain the trap.
+ */
.global kvmppc_hcall
.balign IFETCH_ALIGN_BYTES
kvmppc_hcall:
- /*
- * This is a hcall, so register convention is as
- * Documentation/powerpc/papr_hcalls.rst, with these additions:
- * R13 = PACA
- * guest R13 saved in SPRN_SCRATCH0
- * R10 = free
- * guest r10 saved in PACA_EXGEN
- */
- /*
- * Save the PPR (on systems that support it) before changing to
- * HMT_MEDIUM. That allows the KVM code to save that value into the
- * guest state (it is the guest's PPR value).
- */
-BEGIN_FTR_SECTION
- mfspr r10,SPRN_PPR
- std r10,HSTATE_PPR(r13)
-END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
- HMT_MEDIUM
- mfcr r10
- std r12,HSTATE_SCRATCH0(r13)
- sldi r12,r10,32
- ori r12,r12,0xc00
- ld r10,PACA_EXGEN+EX_R10(r13)
- b do_kvm_interrupt
+ ld r10,PACA_EXGEN+EX_R13(r13)
+ SET_SCRATCH0(r10)
+ li r10,0xc00
+ /* Now we look like kvmppc_interrupt */
/*
* KVM interrupt entry occurs after GEN_INT_ENTRY runs, and follows that
@@ -84,7 +72,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
ld r10,EX_R10(r11)
ld r11,EX_R11(r11)
-do_kvm_interrupt:
/*
* Hcalls and other interrupts come here after normalising register
* contents and save locations:
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 18/46] KVM: PPC: Book3S 64: Minimise hcall handler calling convention differences
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Alexey Kardashevskiy, linuxppc-dev, Nicholas Piggin
This sets up the same calling convention from interrupt entry to
KVM interrupt handler for system calls as exists for other interrupt
types.
This is a better API, it uses a save area rather than SPR, and it has
more registers free to use. Using a single common API helps maintain
it, and it becomes easier to use in C in a later patch.
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kernel/exceptions-64s.S | 21 +++++++++++++++-
arch/powerpc/kvm/book3s_64_entry.S | 37 +++++++++-------------------
2 files changed, 32 insertions(+), 26 deletions(-)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 115cf79f3e82..4615057681c3 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -1892,8 +1892,27 @@ EXC_VIRT_END(system_call, 0x4c00, 0x100)
#ifdef CONFIG_KVM_BOOK3S_64_HANDLER
TRAMP_REAL_BEGIN(kvm_hcall)
+ std r9,PACA_EXGEN+EX_R9(r13)
+ std r11,PACA_EXGEN+EX_R11(r13)
+ std r12,PACA_EXGEN+EX_R12(r13)
+ mfcr r9
mfctr r10
- SET_SCRATCH0(r10) /* Save r13 in SCRATCH0 */
+ std r10,PACA_EXGEN+EX_R13(r13)
+ li r10,0
+ std r10,PACA_EXGEN+EX_CFAR(r13)
+ std r10,PACA_EXGEN+EX_CTR(r13)
+ /*
+ * Save the PPR (on systems that support it) before changing to
+ * HMT_MEDIUM. That allows the KVM code to save that value into the
+ * guest state (it is the guest's PPR value).
+ */
+BEGIN_FTR_SECTION
+ mfspr r10,SPRN_PPR
+ std r10,PACA_EXGEN+EX_PPR(r13)
+END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
+
+ HMT_MEDIUM
+
#ifdef CONFIG_RELOCATABLE
/*
* Requires __LOAD_FAR_HANDLER beause kvmppc_hcall lives
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index b6149df21de3..de81ab69555b 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -11,33 +11,21 @@
* These are branched to from interrupt handlers in exception-64s.S which set
* IKVM_REAL or IKVM_VIRT, if HSTATE_IN_GUEST was found to be non-zero.
*/
+
+/*
+ * This is a hcall, so guest register call convention is as
+ * Documentation/powerpc/papr_hcalls.rst.
+ *
+ * The state of registers is as below, except CFAR is not saved, R13 is not
+ * in SCRATCH0, and R10 does not contain the trap.
+ */
.global kvmppc_hcall
.balign IFETCH_ALIGN_BYTES
kvmppc_hcall:
- /*
- * This is a hcall, so register convention is as
- * Documentation/powerpc/papr_hcalls.rst, with these additions:
- * R13 = PACA
- * guest R13 saved in SPRN_SCRATCH0
- * R10 = free
- * guest r10 saved in PACA_EXGEN
- */
- /*
- * Save the PPR (on systems that support it) before changing to
- * HMT_MEDIUM. That allows the KVM code to save that value into the
- * guest state (it is the guest's PPR value).
- */
-BEGIN_FTR_SECTION
- mfspr r10,SPRN_PPR
- std r10,HSTATE_PPR(r13)
-END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
- HMT_MEDIUM
- mfcr r10
- std r12,HSTATE_SCRATCH0(r13)
- sldi r12,r10,32
- ori r12,r12,0xc00
- ld r10,PACA_EXGEN+EX_R10(r13)
- b do_kvm_interrupt
+ ld r10,PACA_EXGEN+EX_R13(r13)
+ SET_SCRATCH0(r10)
+ li r10,0xc00
+ /* Now we look like kvmppc_interrupt */
/*
* KVM interrupt entry occurs after GEN_INT_ENTRY runs, and follows that
@@ -84,7 +72,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
ld r10,EX_R10(r11)
ld r11,EX_R11(r11)
-do_kvm_interrupt:
/*
* Hcalls and other interrupts come here after normalising register
* contents and save locations:
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 19/46] KVM: PPC: Book3S HV P9: Move radix MMU switching instructions together
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Alexey Kardashevskiy, linuxppc-dev, Nicholas Piggin
Switching the MMU from radix<->radix mode is tricky particularly as the
MMU can remain enabled and requires a certain sequence of SPR updates.
Move these together into their own functions.
This also includes the radix TLB check / flush because it's tied in to
MMU switching due to tlbiel getting LPID from LPIDR.
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 66 +++++++++++++++++++++++-------------
1 file changed, 43 insertions(+), 23 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 7bb4222729b0..b822c573dc08 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3472,12 +3472,49 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
trace_kvmppc_run_core(vc, 1);
}
+static void switch_mmu_to_guest_radix(struct kvm *kvm, struct kvm_vcpu *vcpu, u64 lpcr)
+{
+ struct kvmppc_vcore *vc = vcpu->arch.vcore;
+ struct kvm_nested_guest *nested = vcpu->arch.nested;
+ u32 lpid;
+
+ lpid = nested ? nested->shadow_lpid : kvm->arch.lpid;
+
+ /*
+ * All the isync()s are overkill but trivially follow the ISA
+ * requirements. Some can likely be replaced with justification
+ * comment for why they are not needed.
+ */
+ isync();
+ mtspr(SPRN_LPID, lpid);
+ isync();
+ mtspr(SPRN_LPCR, lpcr);
+ isync();
+ mtspr(SPRN_PID, vcpu->arch.pid);
+ isync();
+
+ /* TLBIEL must have LPIDR set, so set guest LPID before flushing. */
+ kvmppc_check_need_tlb_flush(kvm, vc->pcpu, nested);
+}
+
+static void switch_mmu_to_host_radix(struct kvm *kvm, u32 pid)
+{
+ isync();
+ mtspr(SPRN_PID, pid);
+ isync();
+ mtspr(SPRN_LPID, kvm->arch.host_lpid);
+ isync();
+ mtspr(SPRN_LPCR, kvm->arch.host_lpcr);
+ isync();
+}
+
/*
* Load up hypervisor-mode registers on P9.
*/
static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
unsigned long lpcr)
{
+ struct kvm *kvm = vcpu->kvm;
struct kvmppc_vcore *vc = vcpu->arch.vcore;
s64 hdec;
u64 tb, purr, spurr;
@@ -3500,12 +3537,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
* P8 and P9 suppress the HDEC exception when LPCR[HDICE] = 0,
* so set HDICE before writing HDEC.
*/
- mtspr(SPRN_LPCR, vcpu->kvm->arch.host_lpcr | LPCR_HDICE);
+ mtspr(SPRN_LPCR, kvm->arch.host_lpcr | LPCR_HDICE);
isync();
hdec = time_limit - mftb();
if (hdec < 0) {
- mtspr(SPRN_LPCR, vcpu->kvm->arch.host_lpcr);
+ mtspr(SPRN_LPCR, kvm->arch.host_lpcr);
isync();
return BOOK3S_INTERRUPT_HV_DECREMENTER;
}
@@ -3540,7 +3577,6 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
}
mtspr(SPRN_CIABR, vcpu->arch.ciabr);
mtspr(SPRN_IC, vcpu->arch.ic);
- mtspr(SPRN_PID, vcpu->arch.pid);
mtspr(SPRN_PSSCR, vcpu->arch.psscr | PSSCR_EC |
(local_paca->kvm_hstate.fake_suspend << PSSCR_FAKE_SUSPEND_LG));
@@ -3554,8 +3590,7 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
mtspr(SPRN_AMOR, ~0UL);
- mtspr(SPRN_LPCR, lpcr);
- isync();
+ switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
kvmppc_xive_push_vcpu(vcpu);
@@ -3594,7 +3629,6 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
mtspr(SPRN_DAWR1, host_dawr1);
mtspr(SPRN_DAWRX1, host_dawrx1);
}
- mtspr(SPRN_PID, host_pidr);
/*
* Since this is radix, do a eieio; tlbsync; ptesync sequence in
@@ -3609,9 +3643,6 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
if (cpu_has_feature(CPU_FTR_ARCH_31))
asm volatile(PPC_CP_ABORT);
- mtspr(SPRN_LPID, vcpu->kvm->arch.host_lpid); /* restore host LPID */
- isync();
-
vc->dpdes = mfspr(SPRN_DPDES);
vc->vtb = mfspr(SPRN_VTB);
mtspr(SPRN_DPDES, 0);
@@ -3628,7 +3659,8 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
}
mtspr(SPRN_HDEC, 0x7fffffff);
- mtspr(SPRN_LPCR, vcpu->kvm->arch.host_lpcr);
+
+ switch_mmu_to_host_radix(kvm, host_pidr);
return trap;
}
@@ -4161,7 +4193,7 @@ int kvmhv_run_single_vcpu(struct kvm_vcpu *vcpu, u64 time_limit,
{
struct kvm_run *run = vcpu->run;
int trap, r, pcpu;
- int srcu_idx, lpid;
+ int srcu_idx;
struct kvmppc_vcore *vc;
struct kvm *kvm = vcpu->kvm;
struct kvm_nested_guest *nested = vcpu->arch.nested;
@@ -4235,13 +4267,6 @@ int kvmhv_run_single_vcpu(struct kvm_vcpu *vcpu, u64 time_limit,
vc->vcore_state = VCORE_RUNNING;
trace_kvmppc_run_core(vc, 0);
- if (cpu_has_feature(CPU_FTR_HVMODE)) {
- lpid = nested ? nested->shadow_lpid : kvm->arch.lpid;
- mtspr(SPRN_LPID, lpid);
- isync();
- kvmppc_check_need_tlb_flush(kvm, pcpu, nested);
- }
-
guest_enter_irqoff();
srcu_idx = srcu_read_lock(&kvm->srcu);
@@ -4260,11 +4285,6 @@ int kvmhv_run_single_vcpu(struct kvm_vcpu *vcpu, u64 time_limit,
srcu_read_unlock(&kvm->srcu, srcu_idx);
- if (cpu_has_feature(CPU_FTR_HVMODE)) {
- mtspr(SPRN_LPID, kvm->arch.host_lpid);
- isync();
- }
-
set_irq_happened(trap);
kvmppc_set_host_core(pcpu);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 19/46] KVM: PPC: Book3S HV P9: Move radix MMU switching instructions together
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Alexey Kardashevskiy, linuxppc-dev, Nicholas Piggin
Switching the MMU from radix<->radix mode is tricky particularly as the
MMU can remain enabled and requires a certain sequence of SPR updates.
Move these together into their own functions.
This also includes the radix TLB check / flush because it's tied in to
MMU switching due to tlbiel getting LPID from LPIDR.
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 66 +++++++++++++++++++++++-------------
1 file changed, 43 insertions(+), 23 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 7bb4222729b0..b822c573dc08 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3472,12 +3472,49 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
trace_kvmppc_run_core(vc, 1);
}
+static void switch_mmu_to_guest_radix(struct kvm *kvm, struct kvm_vcpu *vcpu, u64 lpcr)
+{
+ struct kvmppc_vcore *vc = vcpu->arch.vcore;
+ struct kvm_nested_guest *nested = vcpu->arch.nested;
+ u32 lpid;
+
+ lpid = nested ? nested->shadow_lpid : kvm->arch.lpid;
+
+ /*
+ * All the isync()s are overkill but trivially follow the ISA
+ * requirements. Some can likely be replaced with justification
+ * comment for why they are not needed.
+ */
+ isync();
+ mtspr(SPRN_LPID, lpid);
+ isync();
+ mtspr(SPRN_LPCR, lpcr);
+ isync();
+ mtspr(SPRN_PID, vcpu->arch.pid);
+ isync();
+
+ /* TLBIEL must have LPIDR set, so set guest LPID before flushing. */
+ kvmppc_check_need_tlb_flush(kvm, vc->pcpu, nested);
+}
+
+static void switch_mmu_to_host_radix(struct kvm *kvm, u32 pid)
+{
+ isync();
+ mtspr(SPRN_PID, pid);
+ isync();
+ mtspr(SPRN_LPID, kvm->arch.host_lpid);
+ isync();
+ mtspr(SPRN_LPCR, kvm->arch.host_lpcr);
+ isync();
+}
+
/*
* Load up hypervisor-mode registers on P9.
*/
static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
unsigned long lpcr)
{
+ struct kvm *kvm = vcpu->kvm;
struct kvmppc_vcore *vc = vcpu->arch.vcore;
s64 hdec;
u64 tb, purr, spurr;
@@ -3500,12 +3537,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
* P8 and P9 suppress the HDEC exception when LPCR[HDICE] = 0,
* so set HDICE before writing HDEC.
*/
- mtspr(SPRN_LPCR, vcpu->kvm->arch.host_lpcr | LPCR_HDICE);
+ mtspr(SPRN_LPCR, kvm->arch.host_lpcr | LPCR_HDICE);
isync();
hdec = time_limit - mftb();
if (hdec < 0) {
- mtspr(SPRN_LPCR, vcpu->kvm->arch.host_lpcr);
+ mtspr(SPRN_LPCR, kvm->arch.host_lpcr);
isync();
return BOOK3S_INTERRUPT_HV_DECREMENTER;
}
@@ -3540,7 +3577,6 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
}
mtspr(SPRN_CIABR, vcpu->arch.ciabr);
mtspr(SPRN_IC, vcpu->arch.ic);
- mtspr(SPRN_PID, vcpu->arch.pid);
mtspr(SPRN_PSSCR, vcpu->arch.psscr | PSSCR_EC |
(local_paca->kvm_hstate.fake_suspend << PSSCR_FAKE_SUSPEND_LG));
@@ -3554,8 +3590,7 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
mtspr(SPRN_AMOR, ~0UL);
- mtspr(SPRN_LPCR, lpcr);
- isync();
+ switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
kvmppc_xive_push_vcpu(vcpu);
@@ -3594,7 +3629,6 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
mtspr(SPRN_DAWR1, host_dawr1);
mtspr(SPRN_DAWRX1, host_dawrx1);
}
- mtspr(SPRN_PID, host_pidr);
/*
* Since this is radix, do a eieio; tlbsync; ptesync sequence in
@@ -3609,9 +3643,6 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
if (cpu_has_feature(CPU_FTR_ARCH_31))
asm volatile(PPC_CP_ABORT);
- mtspr(SPRN_LPID, vcpu->kvm->arch.host_lpid); /* restore host LPID */
- isync();
-
vc->dpdes = mfspr(SPRN_DPDES);
vc->vtb = mfspr(SPRN_VTB);
mtspr(SPRN_DPDES, 0);
@@ -3628,7 +3659,8 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
}
mtspr(SPRN_HDEC, 0x7fffffff);
- mtspr(SPRN_LPCR, vcpu->kvm->arch.host_lpcr);
+
+ switch_mmu_to_host_radix(kvm, host_pidr);
return trap;
}
@@ -4161,7 +4193,7 @@ int kvmhv_run_single_vcpu(struct kvm_vcpu *vcpu, u64 time_limit,
{
struct kvm_run *run = vcpu->run;
int trap, r, pcpu;
- int srcu_idx, lpid;
+ int srcu_idx;
struct kvmppc_vcore *vc;
struct kvm *kvm = vcpu->kvm;
struct kvm_nested_guest *nested = vcpu->arch.nested;
@@ -4235,13 +4267,6 @@ int kvmhv_run_single_vcpu(struct kvm_vcpu *vcpu, u64 time_limit,
vc->vcore_state = VCORE_RUNNING;
trace_kvmppc_run_core(vc, 0);
- if (cpu_has_feature(CPU_FTR_HVMODE)) {
- lpid = nested ? nested->shadow_lpid : kvm->arch.lpid;
- mtspr(SPRN_LPID, lpid);
- isync();
- kvmppc_check_need_tlb_flush(kvm, pcpu, nested);
- }
-
guest_enter_irqoff();
srcu_idx = srcu_read_lock(&kvm->srcu);
@@ -4260,11 +4285,6 @@ int kvmhv_run_single_vcpu(struct kvm_vcpu *vcpu, u64 time_limit,
srcu_read_unlock(&kvm->srcu, srcu_idx);
- if (cpu_has_feature(CPU_FTR_HVMODE)) {
- mtspr(SPRN_LPID, kvm->arch.host_lpid);
- isync();
- }
-
set_irq_happened(trap);
kvmppc_set_host_core(pcpu);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 20/46] KVM: PPC: Book3S HV P9: implement kvmppc_xive_pull_vcpu in C
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc
Cc: Alexey Kardashevskiy, linuxppc-dev, Cédric Le Goater,
Nicholas Piggin
This is more symmetric with kvmppc_xive_push_vcpu. The extra test in
the asm will go away in a later change.
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_ppc.h | 2 ++
arch/powerpc/kvm/book3s_hv.c | 2 ++
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++++
arch/powerpc/kvm/book3s_xive.c | 31 +++++++++++++++++++++++++
4 files changed, 40 insertions(+)
diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
index 9531b1c1b190..73b1ca5a6471 100644
--- a/arch/powerpc/include/asm/kvm_ppc.h
+++ b/arch/powerpc/include/asm/kvm_ppc.h
@@ -672,6 +672,7 @@ extern int kvmppc_xive_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
int level, bool line_status);
extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
+extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
{
@@ -712,6 +713,7 @@ static inline int kvmppc_xive_set_icp(struct kvm_vcpu *vcpu, u64 icpval) { retur
static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
int level, bool line_status) { return -ENODEV; }
static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
+static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
{ return 0; }
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index b822c573dc08..d28fb76b08e1 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3599,6 +3599,8 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
trap = __kvmhv_vcpu_entry_p9(vcpu);
+ kvmppc_xive_pull_vcpu(vcpu);
+
/* Advance host PURR/SPURR by the amount used by guest */
purr = mfspr(SPRN_PURR);
spurr = mfspr(SPRN_SPURR);
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 75405ef53238..c11597f815e4 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -1442,6 +1442,11 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
bl kvmhv_accumulate_time
#endif
#ifdef CONFIG_KVM_XICS
+ /* If we came in through the P9 short path, xive pull is done in C */
+ lwz r0, STACK_SLOT_SHORT_PATH(r1)
+ cmpwi r0, 0
+ bne 1f
+
/* We are exiting, pull the VP from the XIVE */
lbz r0, VCPU_XIVE_PUSHED(r9)
cmpwi cr0, r0, 0
diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
index e7219b6f5f9a..741bf1f4387a 100644
--- a/arch/powerpc/kvm/book3s_xive.c
+++ b/arch/powerpc/kvm/book3s_xive.c
@@ -127,6 +127,37 @@ void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu)
}
EXPORT_SYMBOL_GPL(kvmppc_xive_push_vcpu);
+/*
+ * Pull a vcpu's context from the XIVE on guest exit.
+ * This assumes we are in virtual mode (MMU on)
+ */
+void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu)
+{
+ void __iomem *tima = local_paca->kvm_hstate.xive_tima_virt;
+
+ if (!vcpu->arch.xive_pushed)
+ return;
+
+ /*
+ * Should not have been pushed if there is no tima
+ */
+ if (WARN_ON(!tima))
+ return;
+
+ eieio();
+ /* First load to pull the context, we ignore the value */
+ __raw_readl(tima + TM_SPC_PULL_OS_CTX);
+ /* Second load to recover the context state (Words 0 and 1) */
+ vcpu->arch.xive_saved_state.w01 = __raw_readq(tima + TM_QW1_OS);
+
+ /* Fixup some of the state for the next load */
+ vcpu->arch.xive_saved_state.lsmfb = 0;
+ vcpu->arch.xive_saved_state.ack = 0xff;
+ vcpu->arch.xive_pushed = 0;
+ eieio();
+}
+EXPORT_SYMBOL_GPL(kvmppc_xive_pull_vcpu);
+
/*
* This is a simple trigger for a generic XIVE IRQ. This must
* only be called for interrupts that support a trigger page
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 20/46] KVM: PPC: Book3S HV P9: implement kvmppc_xive_pull_vcpu in C
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc
Cc: Alexey Kardashevskiy, linuxppc-dev, Cédric Le Goater,
Nicholas Piggin
This is more symmetric with kvmppc_xive_push_vcpu. The extra test in
the asm will go away in a later change.
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_ppc.h | 2 ++
arch/powerpc/kvm/book3s_hv.c | 2 ++
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++++
arch/powerpc/kvm/book3s_xive.c | 31 +++++++++++++++++++++++++
4 files changed, 40 insertions(+)
diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
index 9531b1c1b190..73b1ca5a6471 100644
--- a/arch/powerpc/include/asm/kvm_ppc.h
+++ b/arch/powerpc/include/asm/kvm_ppc.h
@@ -672,6 +672,7 @@ extern int kvmppc_xive_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
int level, bool line_status);
extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
+extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
{
@@ -712,6 +713,7 @@ static inline int kvmppc_xive_set_icp(struct kvm_vcpu *vcpu, u64 icpval) { retur
static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
int level, bool line_status) { return -ENODEV; }
static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
+static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
{ return 0; }
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index b822c573dc08..d28fb76b08e1 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3599,6 +3599,8 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
trap = __kvmhv_vcpu_entry_p9(vcpu);
+ kvmppc_xive_pull_vcpu(vcpu);
+
/* Advance host PURR/SPURR by the amount used by guest */
purr = mfspr(SPRN_PURR);
spurr = mfspr(SPRN_SPURR);
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 75405ef53238..c11597f815e4 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -1442,6 +1442,11 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
bl kvmhv_accumulate_time
#endif
#ifdef CONFIG_KVM_XICS
+ /* If we came in through the P9 short path, xive pull is done in C */
+ lwz r0, STACK_SLOT_SHORT_PATH(r1)
+ cmpwi r0, 0
+ bne 1f
+
/* We are exiting, pull the VP from the XIVE */
lbz r0, VCPU_XIVE_PUSHED(r9)
cmpwi cr0, r0, 0
diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
index e7219b6f5f9a..741bf1f4387a 100644
--- a/arch/powerpc/kvm/book3s_xive.c
+++ b/arch/powerpc/kvm/book3s_xive.c
@@ -127,6 +127,37 @@ void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu)
}
EXPORT_SYMBOL_GPL(kvmppc_xive_push_vcpu);
+/*
+ * Pull a vcpu's context from the XIVE on guest exit.
+ * This assumes we are in virtual mode (MMU on)
+ */
+void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu)
+{
+ void __iomem *tima = local_paca->kvm_hstate.xive_tima_virt;
+
+ if (!vcpu->arch.xive_pushed)
+ return;
+
+ /*
+ * Should not have been pushed if there is no tima
+ */
+ if (WARN_ON(!tima))
+ return;
+
+ eieio();
+ /* First load to pull the context, we ignore the value */
+ __raw_readl(tima + TM_SPC_PULL_OS_CTX);
+ /* Second load to recover the context state (Words 0 and 1) */
+ vcpu->arch.xive_saved_state.w01 = __raw_readq(tima + TM_QW1_OS);
+
+ /* Fixup some of the state for the next load */
+ vcpu->arch.xive_saved_state.lsmfb = 0;
+ vcpu->arch.xive_saved_state.ack = 0xff;
+ vcpu->arch.xive_pushed = 0;
+ eieio();
+}
+EXPORT_SYMBOL_GPL(kvmppc_xive_pull_vcpu);
+
/*
* This is a simple trigger for a generic XIVE IRQ. This must
* only be called for interrupts that support a trigger page
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 21/46] KVM: PPC: Book3S HV P9: Move xive vcpu context management into kvmhv_p9_guest_entry
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Alexey Kardashevskiy, linuxppc-dev, Nicholas Piggin
Move the xive management up so the low level register switching can be
pushed further down in a later patch. XIVE MMIO CI operations can run in
higher level code with machine checks, tracing, etc., available.
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 7 +++----
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index d28fb76b08e1..fa7614c37e08 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3592,15 +3592,11 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
- kvmppc_xive_push_vcpu(vcpu);
-
mtspr(SPRN_SRR0, vcpu->arch.shregs.srr0);
mtspr(SPRN_SRR1, vcpu->arch.shregs.srr1);
trap = __kvmhv_vcpu_entry_p9(vcpu);
- kvmppc_xive_pull_vcpu(vcpu);
-
/* Advance host PURR/SPURR by the amount used by guest */
purr = mfspr(SPRN_PURR);
spurr = mfspr(SPRN_SPURR);
@@ -3783,7 +3779,10 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
trap = 0;
}
} else {
+ kvmppc_xive_push_vcpu(vcpu);
trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
+ kvmppc_xive_pull_vcpu(vcpu);
+
}
vcpu->arch.slb_max = 0;
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 21/46] KVM: PPC: Book3S HV P9: Move xive vcpu context management into kvmhv_p9_guest_entry
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Alexey Kardashevskiy, linuxppc-dev, Nicholas Piggin
Move the xive management up so the low level register switching can be
pushed further down in a later patch. XIVE MMIO CI operations can run in
higher level code with machine checks, tracing, etc., available.
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 7 +++----
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index d28fb76b08e1..fa7614c37e08 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3592,15 +3592,11 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
- kvmppc_xive_push_vcpu(vcpu);
-
mtspr(SPRN_SRR0, vcpu->arch.shregs.srr0);
mtspr(SPRN_SRR1, vcpu->arch.shregs.srr1);
trap = __kvmhv_vcpu_entry_p9(vcpu);
- kvmppc_xive_pull_vcpu(vcpu);
-
/* Advance host PURR/SPURR by the amount used by guest */
purr = mfspr(SPRN_PURR);
spurr = mfspr(SPRN_SPURR);
@@ -3783,7 +3779,10 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
trap = 0;
}
} else {
+ kvmppc_xive_push_vcpu(vcpu);
trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
+ kvmppc_xive_pull_vcpu(vcpu);
+
}
vcpu->arch.slb_max = 0;
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
In the interest of minimising the amount of code that is run in
"real-mode", don't handle hcalls in real mode in the P9 path.
POWER8 and earlier are much more expensive to exit from HV real mode
and switch to host mode, because on those processors HV interrupts get
to the hypervisor with the MMU off, and the other threads in the core
need to be pulled out of the guest, and SLBs all need to be saved,
ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
in host mode. Hash guests also require a lot of hcalls to run. The
XICS interrupt controller requires hcalls to run.
By contrast, POWER9 has independent thread switching, and in radix mode
the hypervisor is already in a host virtual memory mode when the HV
interrupt is taken. Radix + xive guests don't need hcalls to handle
interrupts or manage translations.
So it's much less important to handle hcalls in real mode in P9.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_ppc.h | 5 ++
arch/powerpc/kvm/book3s_hv.c | 57 ++++++++++++++++----
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++
arch/powerpc/kvm/book3s_xive.c | 70 +++++++++++++++++++++++++
4 files changed, 127 insertions(+), 10 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
index 73b1ca5a6471..db6646c2ade2 100644
--- a/arch/powerpc/include/asm/kvm_ppc.h
+++ b/arch/powerpc/include/asm/kvm_ppc.h
@@ -607,6 +607,7 @@ extern void kvmppc_free_pimap(struct kvm *kvm);
extern int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall);
extern void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu);
extern int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd);
+extern int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req);
extern u64 kvmppc_xics_get_icp(struct kvm_vcpu *vcpu);
extern int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
extern int kvmppc_xics_connect_vcpu(struct kvm_device *dev,
@@ -639,6 +640,8 @@ static inline int kvmppc_xics_enabled(struct kvm_vcpu *vcpu)
static inline void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) { }
static inline int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd)
{ return 0; }
+static inline int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
+ { return 0; }
#endif
#ifdef CONFIG_KVM_XIVE
@@ -673,6 +676,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
int level, bool line_status);
extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
+extern void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu);
static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
{
@@ -714,6 +718,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
int level, bool line_status) { return -ENODEV; }
static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
+static inline void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu) { }
static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
{ return 0; }
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index fa7614c37e08..17739aaee3d8 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
}
/*
- * Handle H_CEDE in the nested virtualization case where we haven't
- * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
+ * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
+ * handlers in book3s_hv_rmhandlers.S.
+ *
* This has to be done early, not in kvmppc_pseries_do_hcall(), so
* that the cede logic in kvmppc_run_single_vcpu() works properly.
*/
-static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
+static void kvmppc_cede(struct kvm_vcpu *vcpu)
{
vcpu->arch.shregs.msr |= MSR_EE;
vcpu->arch.ceded = 1;
@@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
/* hcall - punt to userspace */
int i;
- /* hypercall with MSR_PR has already been handled in rmode,
- * and never reaches here.
- */
+ if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
+ /*
+ * Guest userspace executed sc 1, reflect it back as a
+ * privileged program check interrupt.
+ */
+ kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
+ r = RESUME_GUEST;
+ break;
+ }
run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
for (i = 0; i < 9; ++i)
@@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
return trap;
}
+static inline bool hcall_is_xics(unsigned long req)
+{
+ return (req == H_EOI || req == H_CPPR || req == H_IPI ||
+ req == H_IPOLL || req == H_XIRR || req == H_XIRR_X);
+}
+
/*
* Virtual-mode guest entry for POWER9 and later when the host and
* guest are both using the radix MMU. The LPIDR has already been set.
@@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
/* H_CEDE has to be handled now, not later */
if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
kvmppc_get_gpr(vcpu, 3) == H_CEDE) {
- kvmppc_nested_cede(vcpu);
+ kvmppc_cede(vcpu);
kvmppc_set_gpr(vcpu, 3, 0);
trap = 0;
}
} else {
kvmppc_xive_push_vcpu(vcpu);
trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
+ if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
+ !(vcpu->arch.shregs.msr & MSR_PR)) {
+ unsigned long req = kvmppc_get_gpr(vcpu, 3);
+
+ /* H_CEDE has to be handled now, not later */
+ if (req == H_CEDE) {
+ kvmppc_cede(vcpu);
+ kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */
+ kvmppc_set_gpr(vcpu, 3, 0);
+ trap = 0;
+
+ /* XICS hcalls must be handled before xive is pulled */
+ } else if (hcall_is_xics(req)) {
+ int ret;
+
+ ret = kvmppc_xive_xics_hcall(vcpu, req);
+ if (ret != H_TOO_HARD) {
+ kvmppc_set_gpr(vcpu, 3, ret);
+ trap = 0;
+ }
+ }
+ }
kvmppc_xive_pull_vcpu(vcpu);
-
}
vcpu->arch.slb_max = 0;
@@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
else
r = kvmppc_run_vcpu(vcpu);
- if (run->exit_reason == KVM_EXIT_PAPR_HCALL &&
- !(vcpu->arch.shregs.msr & MSR_PR)) {
+ if (run->exit_reason == KVM_EXIT_PAPR_HCALL) {
+ if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
+ r = RESUME_GUEST;
+ continue;
+ }
trace_kvm_hcall_enter(vcpu);
r = kvmppc_pseries_do_hcall(vcpu);
trace_kvm_hcall_exit(vcpu, r);
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index c11597f815e4..2d0d14ed1d92 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
mr r4,r9
bge fast_guest_return
2:
+ /* If we came in through the P9 short path, no real mode hcalls */
+ lwz r0, STACK_SLOT_SHORT_PATH(r1)
+ cmpwi r0, 0
+ bne no_try_real
/* See if this is an hcall we can handle in real mode */
cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
beq hcall_try_real_mode
+no_try_real:
/* Hypervisor doorbell - exit only if host IPI flag set */
cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
index 741bf1f4387a..dcc07ceaf5ca 100644
--- a/arch/powerpc/kvm/book3s_xive.c
+++ b/arch/powerpc/kvm/book3s_xive.c
@@ -158,6 +158,40 @@ void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu)
}
EXPORT_SYMBOL_GPL(kvmppc_xive_pull_vcpu);
+void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu)
+{
+ void __iomem *esc_vaddr = (void __iomem *)vcpu->arch.xive_esc_vaddr;
+
+ if (!esc_vaddr)
+ return;
+
+ /* we are using XIVE with single escalation */
+
+ if (vcpu->arch.xive_esc_on) {
+ /*
+ * If we still have a pending escalation, abort the cede,
+ * and we must set PQ to 10 rather than 00 so that we don't
+ * potentially end up with two entries for the escalation
+ * interrupt in the XIVE interrupt queue. In that case
+ * we also don't want to set xive_esc_on to 1 here in
+ * case we race with xive_esc_irq().
+ */
+ vcpu->arch.ceded = 0;
+ /*
+ * The escalation interrupts are special as we don't EOI them.
+ * There is no need to use the load-after-store ordering offset
+ * to set PQ to 10 as we won't use StoreEOI.
+ */
+ __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_10);
+ } else {
+ vcpu->arch.xive_esc_on = true;
+ mb();
+ __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_00);
+ }
+ mb();
+}
+EXPORT_SYMBOL_GPL(kvmppc_xive_cede_vcpu);
+
/*
* This is a simple trigger for a generic XIVE IRQ. This must
* only be called for interrupts that support a trigger page
@@ -2106,6 +2140,42 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type)
return 0;
}
+int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
+{
+ struct kvmppc_vcore *vc = vcpu->arch.vcore;
+
+ /*
+ * This test covers the case in which a vCPU does XICS hcalls without
+ * QEMU having connected the vCPU to a XICS ICP. The ICP is the KVM
+ * XICS device on P8 or XICS-on-XIVE on P9. It catches QEMU errors when
+ * the interrupt mode is negotiated, we don't want the OS to do XICS
+ * hcalls after having negotiated the XIVE interrupt mode.
+ */
+ if (!kvmppc_xics_enabled(vcpu))
+ return H_TOO_HARD;
+
+ switch (req) {
+ case H_XIRR:
+ return xive_vm_h_xirr(vcpu);
+ case H_CPPR:
+ return xive_vm_h_cppr(vcpu, kvmppc_get_gpr(vcpu, 4));
+ case H_EOI:
+ return xive_vm_h_eoi(vcpu, kvmppc_get_gpr(vcpu, 4));
+ case H_IPI:
+ return xive_vm_h_ipi(vcpu, kvmppc_get_gpr(vcpu, 4),
+ kvmppc_get_gpr(vcpu, 5));
+ case H_IPOLL:
+ return xive_vm_h_ipoll(vcpu, kvmppc_get_gpr(vcpu, 4));
+ case H_XIRR_X:
+ xive_vm_h_xirr(vcpu);
+ kvmppc_set_gpr(vcpu, 5, get_tb() + vc->tb_offset);
+ return H_SUCCESS;
+ }
+
+ return H_UNSUPPORTED;
+}
+EXPORT_SYMBOL_GPL(kvmppc_xive_xics_hcall);
+
int kvmppc_xive_debug_show_queues(struct seq_file *m, struct kvm_vcpu *vcpu)
{
struct kvmppc_xive_vcpu *xc = vcpu->arch.xive_vcpu;
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
In the interest of minimising the amount of code that is run in
"real-mode", don't handle hcalls in real mode in the P9 path.
POWER8 and earlier are much more expensive to exit from HV real mode
and switch to host mode, because on those processors HV interrupts get
to the hypervisor with the MMU off, and the other threads in the core
need to be pulled out of the guest, and SLBs all need to be saved,
ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
in host mode. Hash guests also require a lot of hcalls to run. The
XICS interrupt controller requires hcalls to run.
By contrast, POWER9 has independent thread switching, and in radix mode
the hypervisor is already in a host virtual memory mode when the HV
interrupt is taken. Radix + xive guests don't need hcalls to handle
interrupts or manage translations.
So it's much less important to handle hcalls in real mode in P9.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_ppc.h | 5 ++
arch/powerpc/kvm/book3s_hv.c | 57 ++++++++++++++++----
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++
arch/powerpc/kvm/book3s_xive.c | 70 +++++++++++++++++++++++++
4 files changed, 127 insertions(+), 10 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
index 73b1ca5a6471..db6646c2ade2 100644
--- a/arch/powerpc/include/asm/kvm_ppc.h
+++ b/arch/powerpc/include/asm/kvm_ppc.h
@@ -607,6 +607,7 @@ extern void kvmppc_free_pimap(struct kvm *kvm);
extern int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall);
extern void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu);
extern int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd);
+extern int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req);
extern u64 kvmppc_xics_get_icp(struct kvm_vcpu *vcpu);
extern int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
extern int kvmppc_xics_connect_vcpu(struct kvm_device *dev,
@@ -639,6 +640,8 @@ static inline int kvmppc_xics_enabled(struct kvm_vcpu *vcpu)
static inline void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) { }
static inline int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd)
{ return 0; }
+static inline int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
+ { return 0; }
#endif
#ifdef CONFIG_KVM_XIVE
@@ -673,6 +676,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
int level, bool line_status);
extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
+extern void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu);
static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
{
@@ -714,6 +718,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
int level, bool line_status) { return -ENODEV; }
static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
+static inline void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu) { }
static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
{ return 0; }
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index fa7614c37e08..17739aaee3d8 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
}
/*
- * Handle H_CEDE in the nested virtualization case where we haven't
- * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
+ * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
+ * handlers in book3s_hv_rmhandlers.S.
+ *
* This has to be done early, not in kvmppc_pseries_do_hcall(), so
* that the cede logic in kvmppc_run_single_vcpu() works properly.
*/
-static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
+static void kvmppc_cede(struct kvm_vcpu *vcpu)
{
vcpu->arch.shregs.msr |= MSR_EE;
vcpu->arch.ceded = 1;
@@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
/* hcall - punt to userspace */
int i;
- /* hypercall with MSR_PR has already been handled in rmode,
- * and never reaches here.
- */
+ if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
+ /*
+ * Guest userspace executed sc 1, reflect it back as a
+ * privileged program check interrupt.
+ */
+ kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
+ r = RESUME_GUEST;
+ break;
+ }
run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
for (i = 0; i < 9; ++i)
@@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
return trap;
}
+static inline bool hcall_is_xics(unsigned long req)
+{
+ return (req = H_EOI || req = H_CPPR || req = H_IPI ||
+ req = H_IPOLL || req = H_XIRR || req = H_XIRR_X);
+}
+
/*
* Virtual-mode guest entry for POWER9 and later when the host and
* guest are both using the radix MMU. The LPIDR has already been set.
@@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
/* H_CEDE has to be handled now, not later */
if (trap = BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
kvmppc_get_gpr(vcpu, 3) = H_CEDE) {
- kvmppc_nested_cede(vcpu);
+ kvmppc_cede(vcpu);
kvmppc_set_gpr(vcpu, 3, 0);
trap = 0;
}
} else {
kvmppc_xive_push_vcpu(vcpu);
trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
+ if (trap = BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
+ !(vcpu->arch.shregs.msr & MSR_PR)) {
+ unsigned long req = kvmppc_get_gpr(vcpu, 3);
+
+ /* H_CEDE has to be handled now, not later */
+ if (req = H_CEDE) {
+ kvmppc_cede(vcpu);
+ kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */
+ kvmppc_set_gpr(vcpu, 3, 0);
+ trap = 0;
+
+ /* XICS hcalls must be handled before xive is pulled */
+ } else if (hcall_is_xics(req)) {
+ int ret;
+
+ ret = kvmppc_xive_xics_hcall(vcpu, req);
+ if (ret != H_TOO_HARD) {
+ kvmppc_set_gpr(vcpu, 3, ret);
+ trap = 0;
+ }
+ }
+ }
kvmppc_xive_pull_vcpu(vcpu);
-
}
vcpu->arch.slb_max = 0;
@@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
else
r = kvmppc_run_vcpu(vcpu);
- if (run->exit_reason = KVM_EXIT_PAPR_HCALL &&
- !(vcpu->arch.shregs.msr & MSR_PR)) {
+ if (run->exit_reason = KVM_EXIT_PAPR_HCALL) {
+ if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
+ r = RESUME_GUEST;
+ continue;
+ }
trace_kvm_hcall_enter(vcpu);
r = kvmppc_pseries_do_hcall(vcpu);
trace_kvm_hcall_exit(vcpu, r);
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index c11597f815e4..2d0d14ed1d92 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
mr r4,r9
bge fast_guest_return
2:
+ /* If we came in through the P9 short path, no real mode hcalls */
+ lwz r0, STACK_SLOT_SHORT_PATH(r1)
+ cmpwi r0, 0
+ bne no_try_real
/* See if this is an hcall we can handle in real mode */
cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
beq hcall_try_real_mode
+no_try_real:
/* Hypervisor doorbell - exit only if host IPI flag set */
cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
index 741bf1f4387a..dcc07ceaf5ca 100644
--- a/arch/powerpc/kvm/book3s_xive.c
+++ b/arch/powerpc/kvm/book3s_xive.c
@@ -158,6 +158,40 @@ void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu)
}
EXPORT_SYMBOL_GPL(kvmppc_xive_pull_vcpu);
+void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu)
+{
+ void __iomem *esc_vaddr = (void __iomem *)vcpu->arch.xive_esc_vaddr;
+
+ if (!esc_vaddr)
+ return;
+
+ /* we are using XIVE with single escalation */
+
+ if (vcpu->arch.xive_esc_on) {
+ /*
+ * If we still have a pending escalation, abort the cede,
+ * and we must set PQ to 10 rather than 00 so that we don't
+ * potentially end up with two entries for the escalation
+ * interrupt in the XIVE interrupt queue. In that case
+ * we also don't want to set xive_esc_on to 1 here in
+ * case we race with xive_esc_irq().
+ */
+ vcpu->arch.ceded = 0;
+ /*
+ * The escalation interrupts are special as we don't EOI them.
+ * There is no need to use the load-after-store ordering offset
+ * to set PQ to 10 as we won't use StoreEOI.
+ */
+ __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_10);
+ } else {
+ vcpu->arch.xive_esc_on = true;
+ mb();
+ __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_00);
+ }
+ mb();
+}
+EXPORT_SYMBOL_GPL(kvmppc_xive_cede_vcpu);
+
/*
* This is a simple trigger for a generic XIVE IRQ. This must
* only be called for interrupts that support a trigger page
@@ -2106,6 +2140,42 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type)
return 0;
}
+int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
+{
+ struct kvmppc_vcore *vc = vcpu->arch.vcore;
+
+ /*
+ * This test covers the case in which a vCPU does XICS hcalls without
+ * QEMU having connected the vCPU to a XICS ICP. The ICP is the KVM
+ * XICS device on P8 or XICS-on-XIVE on P9. It catches QEMU errors when
+ * the interrupt mode is negotiated, we don't want the OS to do XICS
+ * hcalls after having negotiated the XIVE interrupt mode.
+ */
+ if (!kvmppc_xics_enabled(vcpu))
+ return H_TOO_HARD;
+
+ switch (req) {
+ case H_XIRR:
+ return xive_vm_h_xirr(vcpu);
+ case H_CPPR:
+ return xive_vm_h_cppr(vcpu, kvmppc_get_gpr(vcpu, 4));
+ case H_EOI:
+ return xive_vm_h_eoi(vcpu, kvmppc_get_gpr(vcpu, 4));
+ case H_IPI:
+ return xive_vm_h_ipi(vcpu, kvmppc_get_gpr(vcpu, 4),
+ kvmppc_get_gpr(vcpu, 5));
+ case H_IPOLL:
+ return xive_vm_h_ipoll(vcpu, kvmppc_get_gpr(vcpu, 4));
+ case H_XIRR_X:
+ xive_vm_h_xirr(vcpu);
+ kvmppc_set_gpr(vcpu, 5, get_tb() + vc->tb_offset);
+ return H_SUCCESS;
+ }
+
+ return H_UNSUPPORTED;
+}
+EXPORT_SYMBOL_GPL(kvmppc_xive_xics_hcall);
+
int kvmppc_xive_debug_show_queues(struct seq_file *m, struct kvm_vcpu *vcpu)
{
struct kvmppc_xive_vcpu *xc = vcpu->arch.xive_vcpu;
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 9:02 ` Alexey Kardashevskiy
-1 siblings, 0 replies; 188+ messages in thread
From: Alexey Kardashevskiy @ 2021-03-23 9:02 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 23/03/2021 12:02, Nicholas Piggin wrote:
> In the interest of minimising the amount of code that is run in
> "real-mode", don't handle hcalls in real mode in the P9 path.
>
> POWER8 and earlier are much more expensive to exit from HV real mode
> and switch to host mode, because on those processors HV interrupts get
> to the hypervisor with the MMU off, and the other threads in the core
> need to be pulled out of the guest, and SLBs all need to be saved,
> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
> in host mode. Hash guests also require a lot of hcalls to run. The
> XICS interrupt controller requires hcalls to run.
>
> By contrast, POWER9 has independent thread switching, and in radix mode
> the hypervisor is already in a host virtual memory mode when the HV
> interrupt is taken. Radix + xive guests don't need hcalls to handle
> interrupts or manage translations.
>
> So it's much less important to handle hcalls in real mode in P9.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
> arch/powerpc/include/asm/kvm_ppc.h | 5 ++
> arch/powerpc/kvm/book3s_hv.c | 57 ++++++++++++++++----
> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++
> arch/powerpc/kvm/book3s_xive.c | 70 +++++++++++++++++++++++++
> 4 files changed, 127 insertions(+), 10 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
> index 73b1ca5a6471..db6646c2ade2 100644
> --- a/arch/powerpc/include/asm/kvm_ppc.h
> +++ b/arch/powerpc/include/asm/kvm_ppc.h
> @@ -607,6 +607,7 @@ extern void kvmppc_free_pimap(struct kvm *kvm);
> extern int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall);
> extern void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu);
> extern int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd);
> +extern int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req);
> extern u64 kvmppc_xics_get_icp(struct kvm_vcpu *vcpu);
> extern int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
> extern int kvmppc_xics_connect_vcpu(struct kvm_device *dev,
> @@ -639,6 +640,8 @@ static inline int kvmppc_xics_enabled(struct kvm_vcpu *vcpu)
> static inline void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) { }
> static inline int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd)
> { return 0; }
> +static inline int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
> + { return 0; }
> #endif
>
> #ifdef CONFIG_KVM_XIVE
> @@ -673,6 +676,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
> int level, bool line_status);
> extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
> extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
> +extern void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu);
>
> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> {
> @@ -714,6 +718,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
> int level, bool line_status) { return -ENODEV; }
> static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
> static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
> +static inline void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu) { }
>
> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> { return 0; }
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index fa7614c37e08..17739aaee3d8 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
> }
>
> /*
> - * Handle H_CEDE in the nested virtualization case where we haven't
> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
> + * handlers in book3s_hv_rmhandlers.S.
> + *
> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
> * that the cede logic in kvmppc_run_single_vcpu() works properly.
> */
> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
> {
> vcpu->arch.shregs.msr |= MSR_EE;
> vcpu->arch.ceded = 1;
> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
> /* hcall - punt to userspace */
> int i;
>
> - /* hypercall with MSR_PR has already been handled in rmode,
> - * and never reaches here.
> - */
> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
> + /*
> + * Guest userspace executed sc 1, reflect it back as a
> + * privileged program check interrupt.
> + */
> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
> + r = RESUME_GUEST;
> + break;
> + }
>
> run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
> for (i = 0; i < 9; ++i)
> @@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
> return trap;
> }
>
> +static inline bool hcall_is_xics(unsigned long req)
> +{
> + return (req == H_EOI || req == H_CPPR || req == H_IPI ||
> + req == H_IPOLL || req == H_XIRR || req == H_XIRR_X);
Do not need braces :)
> +}
> +
> /*
> * Virtual-mode guest entry for POWER9 and later when the host and
> * guest are both using the radix MMU. The LPIDR has already been set.
> @@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> /* H_CEDE has to be handled now, not later */
> if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> kvmppc_get_gpr(vcpu, 3) == H_CEDE) {
> - kvmppc_nested_cede(vcpu);
> + kvmppc_cede(vcpu);
> kvmppc_set_gpr(vcpu, 3, 0);
> trap = 0;
> }
> } else {
> kvmppc_xive_push_vcpu(vcpu);
> trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
> + if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> + !(vcpu->arch.shregs.msr & MSR_PR)) {
> + unsigned long req = kvmppc_get_gpr(vcpu, 3);
> +
> + /* H_CEDE has to be handled now, not later */
> + if (req == H_CEDE) {
> + kvmppc_cede(vcpu);
> + kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */
> + kvmppc_set_gpr(vcpu, 3, 0);
> + trap = 0;
> +
> + /* XICS hcalls must be handled before xive is pulled */
> + } else if (hcall_is_xics(req)) {
> + int ret;
> +
> + ret = kvmppc_xive_xics_hcall(vcpu, req);
> + if (ret != H_TOO_HARD) {
> + kvmppc_set_gpr(vcpu, 3, ret);
> + trap = 0;
> + }
> + }
> + }
> kvmppc_xive_pull_vcpu(vcpu);
> -
> }
>
> vcpu->arch.slb_max = 0;
> @@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
> else
> r = kvmppc_run_vcpu(vcpu);
>
> - if (run->exit_reason == KVM_EXIT_PAPR_HCALL &&
> - !(vcpu->arch.shregs.msr & MSR_PR)) {
> + if (run->exit_reason == KVM_EXIT_PAPR_HCALL) {
> + if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
> + r = RESUME_GUEST;
> + continue;
> + }
> trace_kvm_hcall_enter(vcpu);
> r = kvmppc_pseries_do_hcall(vcpu);
> trace_kvm_hcall_exit(vcpu, r);
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index c11597f815e4..2d0d14ed1d92 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
> mr r4,r9
> bge fast_guest_return
> 2:
> + /* If we came in through the P9 short path, no real mode hcalls */
> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
> + cmpwi r0, 0
> + bne no_try_real
btw is mmu on at this point? or it gets enabled by rfid at the end of
guest_exit_short_path?
anyway,
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
> /* See if this is an hcall we can handle in real mode */
> cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
> beq hcall_try_real_mode
> +no_try_real:
>
> /* Hypervisor doorbell - exit only if host IPI flag set */
> cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
> diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
> index 741bf1f4387a..dcc07ceaf5ca 100644
> --- a/arch/powerpc/kvm/book3s_xive.c
> +++ b/arch/powerpc/kvm/book3s_xive.c
> @@ -158,6 +158,40 @@ void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu)
> }
> EXPORT_SYMBOL_GPL(kvmppc_xive_pull_vcpu);
>
> +void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu)
> +{
> + void __iomem *esc_vaddr = (void __iomem *)vcpu->arch.xive_esc_vaddr;
> +
> + if (!esc_vaddr)
> + return;
> +
> + /* we are using XIVE with single escalation */
> +
> + if (vcpu->arch.xive_esc_on) {
> + /*
> + * If we still have a pending escalation, abort the cede,
> + * and we must set PQ to 10 rather than 00 so that we don't
> + * potentially end up with two entries for the escalation
> + * interrupt in the XIVE interrupt queue. In that case
> + * we also don't want to set xive_esc_on to 1 here in
> + * case we race with xive_esc_irq().
> + */
> + vcpu->arch.ceded = 0;
> + /*
> + * The escalation interrupts are special as we don't EOI them.
> + * There is no need to use the load-after-store ordering offset
> + * to set PQ to 10 as we won't use StoreEOI.
> + */
> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_10);
> + } else {
> + vcpu->arch.xive_esc_on = true;
> + mb();
> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_00);
> + }
> + mb();
> +}
> +EXPORT_SYMBOL_GPL(kvmppc_xive_cede_vcpu);
> +
> /*
> * This is a simple trigger for a generic XIVE IRQ. This must
> * only be called for interrupts that support a trigger page
> @@ -2106,6 +2140,42 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type)
> return 0;
> }
>
> +int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
> +{
> + struct kvmppc_vcore *vc = vcpu->arch.vcore;
> +
> + /*
> + * This test covers the case in which a vCPU does XICS hcalls without
> + * QEMU having connected the vCPU to a XICS ICP. The ICP is the KVM
> + * XICS device on P8 or XICS-on-XIVE on P9. It catches QEMU errors when
> + * the interrupt mode is negotiated, we don't want the OS to do XICS
> + * hcalls after having negotiated the XIVE interrupt mode.
> + */
> + if (!kvmppc_xics_enabled(vcpu))
> + return H_TOO_HARD;
> +
> + switch (req) {
> + case H_XIRR:
> + return xive_vm_h_xirr(vcpu);
> + case H_CPPR:
> + return xive_vm_h_cppr(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_EOI:
> + return xive_vm_h_eoi(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_IPI:
> + return xive_vm_h_ipi(vcpu, kvmppc_get_gpr(vcpu, 4),
> + kvmppc_get_gpr(vcpu, 5));
> + case H_IPOLL:
> + return xive_vm_h_ipoll(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_XIRR_X:
> + xive_vm_h_xirr(vcpu);
> + kvmppc_set_gpr(vcpu, 5, get_tb() + vc->tb_offset);
> + return H_SUCCESS;
> + }
> +
> + return H_UNSUPPORTED;
> +}
> +EXPORT_SYMBOL_GPL(kvmppc_xive_xics_hcall);
> +
> int kvmppc_xive_debug_show_queues(struct seq_file *m, struct kvm_vcpu *vcpu)
> {
> struct kvmppc_xive_vcpu *xc = vcpu->arch.xive_vcpu;
>
--
Alexey
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
@ 2021-03-23 9:02 ` Alexey Kardashevskiy
0 siblings, 0 replies; 188+ messages in thread
From: Alexey Kardashevskiy @ 2021-03-23 9:02 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 23/03/2021 12:02, Nicholas Piggin wrote:
> In the interest of minimising the amount of code that is run in
> "real-mode", don't handle hcalls in real mode in the P9 path.
>
> POWER8 and earlier are much more expensive to exit from HV real mode
> and switch to host mode, because on those processors HV interrupts get
> to the hypervisor with the MMU off, and the other threads in the core
> need to be pulled out of the guest, and SLBs all need to be saved,
> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
> in host mode. Hash guests also require a lot of hcalls to run. The
> XICS interrupt controller requires hcalls to run.
>
> By contrast, POWER9 has independent thread switching, and in radix mode
> the hypervisor is already in a host virtual memory mode when the HV
> interrupt is taken. Radix + xive guests don't need hcalls to handle
> interrupts or manage translations.
>
> So it's much less important to handle hcalls in real mode in P9.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
> arch/powerpc/include/asm/kvm_ppc.h | 5 ++
> arch/powerpc/kvm/book3s_hv.c | 57 ++++++++++++++++----
> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++
> arch/powerpc/kvm/book3s_xive.c | 70 +++++++++++++++++++++++++
> 4 files changed, 127 insertions(+), 10 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
> index 73b1ca5a6471..db6646c2ade2 100644
> --- a/arch/powerpc/include/asm/kvm_ppc.h
> +++ b/arch/powerpc/include/asm/kvm_ppc.h
> @@ -607,6 +607,7 @@ extern void kvmppc_free_pimap(struct kvm *kvm);
> extern int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall);
> extern void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu);
> extern int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd);
> +extern int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req);
> extern u64 kvmppc_xics_get_icp(struct kvm_vcpu *vcpu);
> extern int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
> extern int kvmppc_xics_connect_vcpu(struct kvm_device *dev,
> @@ -639,6 +640,8 @@ static inline int kvmppc_xics_enabled(struct kvm_vcpu *vcpu)
> static inline void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) { }
> static inline int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd)
> { return 0; }
> +static inline int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
> + { return 0; }
> #endif
>
> #ifdef CONFIG_KVM_XIVE
> @@ -673,6 +676,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
> int level, bool line_status);
> extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
> extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
> +extern void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu);
>
> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> {
> @@ -714,6 +718,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
> int level, bool line_status) { return -ENODEV; }
> static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
> static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
> +static inline void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu) { }
>
> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> { return 0; }
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index fa7614c37e08..17739aaee3d8 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
> }
>
> /*
> - * Handle H_CEDE in the nested virtualization case where we haven't
> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
> + * handlers in book3s_hv_rmhandlers.S.
> + *
> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
> * that the cede logic in kvmppc_run_single_vcpu() works properly.
> */
> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
> {
> vcpu->arch.shregs.msr |= MSR_EE;
> vcpu->arch.ceded = 1;
> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
> /* hcall - punt to userspace */
> int i;
>
> - /* hypercall with MSR_PR has already been handled in rmode,
> - * and never reaches here.
> - */
> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
> + /*
> + * Guest userspace executed sc 1, reflect it back as a
> + * privileged program check interrupt.
> + */
> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
> + r = RESUME_GUEST;
> + break;
> + }
>
> run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
> for (i = 0; i < 9; ++i)
> @@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
> return trap;
> }
>
> +static inline bool hcall_is_xics(unsigned long req)
> +{
> + return (req = H_EOI || req = H_CPPR || req = H_IPI ||
> + req = H_IPOLL || req = H_XIRR || req = H_XIRR_X);
Do not need braces :)
> +}
> +
> /*
> * Virtual-mode guest entry for POWER9 and later when the host and
> * guest are both using the radix MMU. The LPIDR has already been set.
> @@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> /* H_CEDE has to be handled now, not later */
> if (trap = BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> kvmppc_get_gpr(vcpu, 3) = H_CEDE) {
> - kvmppc_nested_cede(vcpu);
> + kvmppc_cede(vcpu);
> kvmppc_set_gpr(vcpu, 3, 0);
> trap = 0;
> }
> } else {
> kvmppc_xive_push_vcpu(vcpu);
> trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
> + if (trap = BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> + !(vcpu->arch.shregs.msr & MSR_PR)) {
> + unsigned long req = kvmppc_get_gpr(vcpu, 3);
> +
> + /* H_CEDE has to be handled now, not later */
> + if (req = H_CEDE) {
> + kvmppc_cede(vcpu);
> + kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */
> + kvmppc_set_gpr(vcpu, 3, 0);
> + trap = 0;
> +
> + /* XICS hcalls must be handled before xive is pulled */
> + } else if (hcall_is_xics(req)) {
> + int ret;
> +
> + ret = kvmppc_xive_xics_hcall(vcpu, req);
> + if (ret != H_TOO_HARD) {
> + kvmppc_set_gpr(vcpu, 3, ret);
> + trap = 0;
> + }
> + }
> + }
> kvmppc_xive_pull_vcpu(vcpu);
> -
> }
>
> vcpu->arch.slb_max = 0;
> @@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
> else
> r = kvmppc_run_vcpu(vcpu);
>
> - if (run->exit_reason = KVM_EXIT_PAPR_HCALL &&
> - !(vcpu->arch.shregs.msr & MSR_PR)) {
> + if (run->exit_reason = KVM_EXIT_PAPR_HCALL) {
> + if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
> + r = RESUME_GUEST;
> + continue;
> + }
> trace_kvm_hcall_enter(vcpu);
> r = kvmppc_pseries_do_hcall(vcpu);
> trace_kvm_hcall_exit(vcpu, r);
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index c11597f815e4..2d0d14ed1d92 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
> mr r4,r9
> bge fast_guest_return
> 2:
> + /* If we came in through the P9 short path, no real mode hcalls */
> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
> + cmpwi r0, 0
> + bne no_try_real
btw is mmu on at this point? or it gets enabled by rfid at the end of
guest_exit_short_path?
anyway,
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
> /* See if this is an hcall we can handle in real mode */
> cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
> beq hcall_try_real_mode
> +no_try_real:
>
> /* Hypervisor doorbell - exit only if host IPI flag set */
> cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
> diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
> index 741bf1f4387a..dcc07ceaf5ca 100644
> --- a/arch/powerpc/kvm/book3s_xive.c
> +++ b/arch/powerpc/kvm/book3s_xive.c
> @@ -158,6 +158,40 @@ void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu)
> }
> EXPORT_SYMBOL_GPL(kvmppc_xive_pull_vcpu);
>
> +void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu)
> +{
> + void __iomem *esc_vaddr = (void __iomem *)vcpu->arch.xive_esc_vaddr;
> +
> + if (!esc_vaddr)
> + return;
> +
> + /* we are using XIVE with single escalation */
> +
> + if (vcpu->arch.xive_esc_on) {
> + /*
> + * If we still have a pending escalation, abort the cede,
> + * and we must set PQ to 10 rather than 00 so that we don't
> + * potentially end up with two entries for the escalation
> + * interrupt in the XIVE interrupt queue. In that case
> + * we also don't want to set xive_esc_on to 1 here in
> + * case we race with xive_esc_irq().
> + */
> + vcpu->arch.ceded = 0;
> + /*
> + * The escalation interrupts are special as we don't EOI them.
> + * There is no need to use the load-after-store ordering offset
> + * to set PQ to 10 as we won't use StoreEOI.
> + */
> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_10);
> + } else {
> + vcpu->arch.xive_esc_on = true;
> + mb();
> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_00);
> + }
> + mb();
> +}
> +EXPORT_SYMBOL_GPL(kvmppc_xive_cede_vcpu);
> +
> /*
> * This is a simple trigger for a generic XIVE IRQ. This must
> * only be called for interrupts that support a trigger page
> @@ -2106,6 +2140,42 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type)
> return 0;
> }
>
> +int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
> +{
> + struct kvmppc_vcore *vc = vcpu->arch.vcore;
> +
> + /*
> + * This test covers the case in which a vCPU does XICS hcalls without
> + * QEMU having connected the vCPU to a XICS ICP. The ICP is the KVM
> + * XICS device on P8 or XICS-on-XIVE on P9. It catches QEMU errors when
> + * the interrupt mode is negotiated, we don't want the OS to do XICS
> + * hcalls after having negotiated the XIVE interrupt mode.
> + */
> + if (!kvmppc_xics_enabled(vcpu))
> + return H_TOO_HARD;
> +
> + switch (req) {
> + case H_XIRR:
> + return xive_vm_h_xirr(vcpu);
> + case H_CPPR:
> + return xive_vm_h_cppr(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_EOI:
> + return xive_vm_h_eoi(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_IPI:
> + return xive_vm_h_ipi(vcpu, kvmppc_get_gpr(vcpu, 4),
> + kvmppc_get_gpr(vcpu, 5));
> + case H_IPOLL:
> + return xive_vm_h_ipoll(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_XIRR_X:
> + xive_vm_h_xirr(vcpu);
> + kvmppc_set_gpr(vcpu, 5, get_tb() + vc->tb_offset);
> + return H_SUCCESS;
> + }
> +
> + return H_UNSUPPORTED;
> +}
> +EXPORT_SYMBOL_GPL(kvmppc_xive_xics_hcall);
> +
> int kvmppc_xive_debug_show_queues(struct seq_file *m, struct kvm_vcpu *vcpu)
> {
> struct kvmppc_xive_vcpu *xc = vcpu->arch.xive_vcpu;
>
--
Alexey
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
2021-03-23 9:02 ` Alexey Kardashevskiy
@ 2021-03-23 9:16 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 9:16 UTC (permalink / raw)
To: Alexey Kardashevskiy, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Alexey Kardashevskiy's message of March 23, 2021 7:02 pm:
>
>
> On 23/03/2021 12:02, Nicholas Piggin wrote:
>> In the interest of minimising the amount of code that is run in
>> "real-mode", don't handle hcalls in real mode in the P9 path.
>>
>> POWER8 and earlier are much more expensive to exit from HV real mode
>> and switch to host mode, because on those processors HV interrupts get
>> to the hypervisor with the MMU off, and the other threads in the core
>> need to be pulled out of the guest, and SLBs all need to be saved,
>> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
>> in host mode. Hash guests also require a lot of hcalls to run. The
>> XICS interrupt controller requires hcalls to run.
>>
>> By contrast, POWER9 has independent thread switching, and in radix mode
>> the hypervisor is already in a host virtual memory mode when the HV
>> interrupt is taken. Radix + xive guests don't need hcalls to handle
>> interrupts or manage translations.
>>
>> So it's much less important to handle hcalls in real mode in P9.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>> ---
>> arch/powerpc/include/asm/kvm_ppc.h | 5 ++
>> arch/powerpc/kvm/book3s_hv.c | 57 ++++++++++++++++----
>> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++
>> arch/powerpc/kvm/book3s_xive.c | 70 +++++++++++++++++++++++++
>> 4 files changed, 127 insertions(+), 10 deletions(-)
>>
>> diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
>> index 73b1ca5a6471..db6646c2ade2 100644
>> --- a/arch/powerpc/include/asm/kvm_ppc.h
>> +++ b/arch/powerpc/include/asm/kvm_ppc.h
>> @@ -607,6 +607,7 @@ extern void kvmppc_free_pimap(struct kvm *kvm);
>> extern int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall);
>> extern void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu);
>> extern int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd);
>> +extern int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req);
>> extern u64 kvmppc_xics_get_icp(struct kvm_vcpu *vcpu);
>> extern int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
>> extern int kvmppc_xics_connect_vcpu(struct kvm_device *dev,
>> @@ -639,6 +640,8 @@ static inline int kvmppc_xics_enabled(struct kvm_vcpu *vcpu)
>> static inline void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) { }
>> static inline int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd)
>> { return 0; }
>> +static inline int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
>> + { return 0; }
>> #endif
>>
>> #ifdef CONFIG_KVM_XIVE
>> @@ -673,6 +676,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
>> int level, bool line_status);
>> extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
>> extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
>> +extern void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu);
>>
>> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
>> {
>> @@ -714,6 +718,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
>> int level, bool line_status) { return -ENODEV; }
>> static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
>> static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
>> +static inline void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu) { }
>>
>> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
>> { return 0; }
>> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
>> index fa7614c37e08..17739aaee3d8 100644
>> --- a/arch/powerpc/kvm/book3s_hv.c
>> +++ b/arch/powerpc/kvm/book3s_hv.c
>> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
>> }
>>
>> /*
>> - * Handle H_CEDE in the nested virtualization case where we haven't
>> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
>> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
>> + * handlers in book3s_hv_rmhandlers.S.
>> + *
>> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
>> * that the cede logic in kvmppc_run_single_vcpu() works properly.
>> */
>> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
>> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
>> {
>> vcpu->arch.shregs.msr |= MSR_EE;
>> vcpu->arch.ceded = 1;
>> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
>> /* hcall - punt to userspace */
>> int i;
>>
>> - /* hypercall with MSR_PR has already been handled in rmode,
>> - * and never reaches here.
>> - */
>> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
>> + /*
>> + * Guest userspace executed sc 1, reflect it back as a
>> + * privileged program check interrupt.
>> + */
>> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
>> + r = RESUME_GUEST;
>> + break;
>> + }
>>
>> run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
>> for (i = 0; i < 9; ++i)
>> @@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
>> return trap;
>> }
>>
>> +static inline bool hcall_is_xics(unsigned long req)
>> +{
>> + return (req == H_EOI || req == H_CPPR || req == H_IPI ||
>> + req == H_IPOLL || req == H_XIRR || req == H_XIRR_X);
>
> Do not need braces :)
>
>
>> +}
>> +
>> /*
>> * Virtual-mode guest entry for POWER9 and later when the host and
>> * guest are both using the radix MMU. The LPIDR has already been set.
>> @@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
>> /* H_CEDE has to be handled now, not later */
>> if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
>> kvmppc_get_gpr(vcpu, 3) == H_CEDE) {
>> - kvmppc_nested_cede(vcpu);
>> + kvmppc_cede(vcpu);
>> kvmppc_set_gpr(vcpu, 3, 0);
>> trap = 0;
>> }
>> } else {
>> kvmppc_xive_push_vcpu(vcpu);
>> trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
>> + if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
>> + !(vcpu->arch.shregs.msr & MSR_PR)) {
>> + unsigned long req = kvmppc_get_gpr(vcpu, 3);
>> +
>> + /* H_CEDE has to be handled now, not later */
>> + if (req == H_CEDE) {
>> + kvmppc_cede(vcpu);
>> + kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */
>> + kvmppc_set_gpr(vcpu, 3, 0);
>> + trap = 0;
>> +
>> + /* XICS hcalls must be handled before xive is pulled */
>> + } else if (hcall_is_xics(req)) {
>> + int ret;
>> +
>> + ret = kvmppc_xive_xics_hcall(vcpu, req);
>> + if (ret != H_TOO_HARD) {
>> + kvmppc_set_gpr(vcpu, 3, ret);
>> + trap = 0;
>> + }
>> + }
>> + }
>> kvmppc_xive_pull_vcpu(vcpu);
>> -
>> }
>>
>> vcpu->arch.slb_max = 0;
>> @@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
>> else
>> r = kvmppc_run_vcpu(vcpu);
>>
>> - if (run->exit_reason == KVM_EXIT_PAPR_HCALL &&
>> - !(vcpu->arch.shregs.msr & MSR_PR)) {
>> + if (run->exit_reason == KVM_EXIT_PAPR_HCALL) {
>> + if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
>> + r = RESUME_GUEST;
>> + continue;
>> + }
>> trace_kvm_hcall_enter(vcpu);
>> r = kvmppc_pseries_do_hcall(vcpu);
>> trace_kvm_hcall_exit(vcpu, r);
>> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>> index c11597f815e4..2d0d14ed1d92 100644
>> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
>> mr r4,r9
>> bge fast_guest_return
>> 2:
>> + /* If we came in through the P9 short path, no real mode hcalls */
>> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
>> + cmpwi r0, 0
>> + bne no_try_real
>
>
> btw is mmu on at this point? or it gets enabled by rfid at the end of
> guest_exit_short_path?
Hash guest it's off. Radix guest it can be on or off depending on the
interrupt type and MSR and LPCR[AIL] values.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
@ 2021-03-23 9:16 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 9:16 UTC (permalink / raw)
To: Alexey Kardashevskiy, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Alexey Kardashevskiy's message of March 23, 2021 7:02 pm:
>
>
> On 23/03/2021 12:02, Nicholas Piggin wrote:
>> In the interest of minimising the amount of code that is run in
>> "real-mode", don't handle hcalls in real mode in the P9 path.
>>
>> POWER8 and earlier are much more expensive to exit from HV real mode
>> and switch to host mode, because on those processors HV interrupts get
>> to the hypervisor with the MMU off, and the other threads in the core
>> need to be pulled out of the guest, and SLBs all need to be saved,
>> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
>> in host mode. Hash guests also require a lot of hcalls to run. The
>> XICS interrupt controller requires hcalls to run.
>>
>> By contrast, POWER9 has independent thread switching, and in radix mode
>> the hypervisor is already in a host virtual memory mode when the HV
>> interrupt is taken. Radix + xive guests don't need hcalls to handle
>> interrupts or manage translations.
>>
>> So it's much less important to handle hcalls in real mode in P9.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>> ---
>> arch/powerpc/include/asm/kvm_ppc.h | 5 ++
>> arch/powerpc/kvm/book3s_hv.c | 57 ++++++++++++++++----
>> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++
>> arch/powerpc/kvm/book3s_xive.c | 70 +++++++++++++++++++++++++
>> 4 files changed, 127 insertions(+), 10 deletions(-)
>>
>> diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
>> index 73b1ca5a6471..db6646c2ade2 100644
>> --- a/arch/powerpc/include/asm/kvm_ppc.h
>> +++ b/arch/powerpc/include/asm/kvm_ppc.h
>> @@ -607,6 +607,7 @@ extern void kvmppc_free_pimap(struct kvm *kvm);
>> extern int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall);
>> extern void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu);
>> extern int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd);
>> +extern int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req);
>> extern u64 kvmppc_xics_get_icp(struct kvm_vcpu *vcpu);
>> extern int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
>> extern int kvmppc_xics_connect_vcpu(struct kvm_device *dev,
>> @@ -639,6 +640,8 @@ static inline int kvmppc_xics_enabled(struct kvm_vcpu *vcpu)
>> static inline void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) { }
>> static inline int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd)
>> { return 0; }
>> +static inline int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
>> + { return 0; }
>> #endif
>>
>> #ifdef CONFIG_KVM_XIVE
>> @@ -673,6 +676,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
>> int level, bool line_status);
>> extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
>> extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
>> +extern void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu);
>>
>> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
>> {
>> @@ -714,6 +718,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
>> int level, bool line_status) { return -ENODEV; }
>> static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
>> static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
>> +static inline void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu) { }
>>
>> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
>> { return 0; }
>> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
>> index fa7614c37e08..17739aaee3d8 100644
>> --- a/arch/powerpc/kvm/book3s_hv.c
>> +++ b/arch/powerpc/kvm/book3s_hv.c
>> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
>> }
>>
>> /*
>> - * Handle H_CEDE in the nested virtualization case where we haven't
>> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
>> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
>> + * handlers in book3s_hv_rmhandlers.S.
>> + *
>> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
>> * that the cede logic in kvmppc_run_single_vcpu() works properly.
>> */
>> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
>> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
>> {
>> vcpu->arch.shregs.msr |= MSR_EE;
>> vcpu->arch.ceded = 1;
>> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
>> /* hcall - punt to userspace */
>> int i;
>>
>> - /* hypercall with MSR_PR has already been handled in rmode,
>> - * and never reaches here.
>> - */
>> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
>> + /*
>> + * Guest userspace executed sc 1, reflect it back as a
>> + * privileged program check interrupt.
>> + */
>> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
>> + r = RESUME_GUEST;
>> + break;
>> + }
>>
>> run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
>> for (i = 0; i < 9; ++i)
>> @@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
>> return trap;
>> }
>>
>> +static inline bool hcall_is_xics(unsigned long req)
>> +{
>> + return (req == H_EOI || req == H_CPPR || req == H_IPI ||
>> + req == H_IPOLL || req == H_XIRR || req == H_XIRR_X);
>
> Do not need braces :)
>
>
>> +}
>> +
>> /*
>> * Virtual-mode guest entry for POWER9 and later when the host and
>> * guest are both using the radix MMU. The LPIDR has already been set.
>> @@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
>> /* H_CEDE has to be handled now, not later */
>> if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
>> kvmppc_get_gpr(vcpu, 3) == H_CEDE) {
>> - kvmppc_nested_cede(vcpu);
>> + kvmppc_cede(vcpu);
>> kvmppc_set_gpr(vcpu, 3, 0);
>> trap = 0;
>> }
>> } else {
>> kvmppc_xive_push_vcpu(vcpu);
>> trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
>> + if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
>> + !(vcpu->arch.shregs.msr & MSR_PR)) {
>> + unsigned long req = kvmppc_get_gpr(vcpu, 3);
>> +
>> + /* H_CEDE has to be handled now, not later */
>> + if (req == H_CEDE) {
>> + kvmppc_cede(vcpu);
>> + kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */
>> + kvmppc_set_gpr(vcpu, 3, 0);
>> + trap = 0;
>> +
>> + /* XICS hcalls must be handled before xive is pulled */
>> + } else if (hcall_is_xics(req)) {
>> + int ret;
>> +
>> + ret = kvmppc_xive_xics_hcall(vcpu, req);
>> + if (ret != H_TOO_HARD) {
>> + kvmppc_set_gpr(vcpu, 3, ret);
>> + trap = 0;
>> + }
>> + }
>> + }
>> kvmppc_xive_pull_vcpu(vcpu);
>> -
>> }
>>
>> vcpu->arch.slb_max = 0;
>> @@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
>> else
>> r = kvmppc_run_vcpu(vcpu);
>>
>> - if (run->exit_reason == KVM_EXIT_PAPR_HCALL &&
>> - !(vcpu->arch.shregs.msr & MSR_PR)) {
>> + if (run->exit_reason == KVM_EXIT_PAPR_HCALL) {
>> + if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
>> + r = RESUME_GUEST;
>> + continue;
>> + }
>> trace_kvm_hcall_enter(vcpu);
>> r = kvmppc_pseries_do_hcall(vcpu);
>> trace_kvm_hcall_exit(vcpu, r);
>> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>> index c11597f815e4..2d0d14ed1d92 100644
>> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
>> mr r4,r9
>> bge fast_guest_return
>> 2:
>> + /* If we came in through the P9 short path, no real mode hcalls */
>> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
>> + cmpwi r0, 0
>> + bne no_try_real
>
>
> btw is mmu on at this point? or it gets enabled by rfid at the end of
> guest_exit_short_path?
Hash guest it's off. Radix guest it can be on or off depending on the
interrupt type and MSR and LPCR[AIL] values.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
2021-03-23 9:16 ` Nicholas Piggin
@ 2021-03-23 9:24 ` Alexey Kardashevskiy
-1 siblings, 0 replies; 188+ messages in thread
From: Alexey Kardashevskiy @ 2021-03-23 9:24 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 23/03/2021 20:16, Nicholas Piggin wrote:
> Excerpts from Alexey Kardashevskiy's message of March 23, 2021 7:02 pm:
>>
>>
>> On 23/03/2021 12:02, Nicholas Piggin wrote:
>>> In the interest of minimising the amount of code that is run in
>>> "real-mode", don't handle hcalls in real mode in the P9 path.
>>>
>>> POWER8 and earlier are much more expensive to exit from HV real mode
>>> and switch to host mode, because on those processors HV interrupts get
>>> to the hypervisor with the MMU off, and the other threads in the core
>>> need to be pulled out of the guest, and SLBs all need to be saved,
>>> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
>>> in host mode. Hash guests also require a lot of hcalls to run. The
>>> XICS interrupt controller requires hcalls to run.
>>>
>>> By contrast, POWER9 has independent thread switching, and in radix mode
>>> the hypervisor is already in a host virtual memory mode when the HV
>>> interrupt is taken. Radix + xive guests don't need hcalls to handle
>>> interrupts or manage translations.
>>>
>>> So it's much less important to handle hcalls in real mode in P9.
>>>
>>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>>> ---
>>> arch/powerpc/include/asm/kvm_ppc.h | 5 ++
>>> arch/powerpc/kvm/book3s_hv.c | 57 ++++++++++++++++----
>>> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++
>>> arch/powerpc/kvm/book3s_xive.c | 70 +++++++++++++++++++++++++
>>> 4 files changed, 127 insertions(+), 10 deletions(-)
>>>
>>> diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
>>> index 73b1ca5a6471..db6646c2ade2 100644
>>> --- a/arch/powerpc/include/asm/kvm_ppc.h
>>> +++ b/arch/powerpc/include/asm/kvm_ppc.h
>>> @@ -607,6 +607,7 @@ extern void kvmppc_free_pimap(struct kvm *kvm);
>>> extern int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall);
>>> extern void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu);
>>> extern int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd);
>>> +extern int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req);
>>> extern u64 kvmppc_xics_get_icp(struct kvm_vcpu *vcpu);
>>> extern int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
>>> extern int kvmppc_xics_connect_vcpu(struct kvm_device *dev,
>>> @@ -639,6 +640,8 @@ static inline int kvmppc_xics_enabled(struct kvm_vcpu *vcpu)
>>> static inline void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) { }
>>> static inline int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd)
>>> { return 0; }
>>> +static inline int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
>>> + { return 0; }
>>> #endif
>>>
>>> #ifdef CONFIG_KVM_XIVE
>>> @@ -673,6 +676,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
>>> int level, bool line_status);
>>> extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
>>> extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
>>> +extern void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu);
>>>
>>> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
>>> {
>>> @@ -714,6 +718,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
>>> int level, bool line_status) { return -ENODEV; }
>>> static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
>>> static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
>>> +static inline void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu) { }
>>>
>>> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
>>> { return 0; }
>>> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
>>> index fa7614c37e08..17739aaee3d8 100644
>>> --- a/arch/powerpc/kvm/book3s_hv.c
>>> +++ b/arch/powerpc/kvm/book3s_hv.c
>>> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
>>> }
>>>
>>> /*
>>> - * Handle H_CEDE in the nested virtualization case where we haven't
>>> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
>>> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
>>> + * handlers in book3s_hv_rmhandlers.S.
>>> + *
>>> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
>>> * that the cede logic in kvmppc_run_single_vcpu() works properly.
>>> */
>>> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
>>> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
>>> {
>>> vcpu->arch.shregs.msr |= MSR_EE;
>>> vcpu->arch.ceded = 1;
>>> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
>>> /* hcall - punt to userspace */
>>> int i;
>>>
>>> - /* hypercall with MSR_PR has already been handled in rmode,
>>> - * and never reaches here.
>>> - */
>>> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
>>> + /*
>>> + * Guest userspace executed sc 1, reflect it back as a
>>> + * privileged program check interrupt.
>>> + */
>>> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
>>> + r = RESUME_GUEST;
>>> + break;
>>> + }
>>>
>>> run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
>>> for (i = 0; i < 9; ++i)
>>> @@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
>>> return trap;
>>> }
>>>
>>> +static inline bool hcall_is_xics(unsigned long req)
>>> +{
>>> + return (req == H_EOI || req == H_CPPR || req == H_IPI ||
>>> + req == H_IPOLL || req == H_XIRR || req == H_XIRR_X);
>>
>> Do not need braces :)
>>
>>
>>> +}
>>> +
>>> /*
>>> * Virtual-mode guest entry for POWER9 and later when the host and
>>> * guest are both using the radix MMU. The LPIDR has already been set.
>>> @@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
>>> /* H_CEDE has to be handled now, not later */
>>> if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
>>> kvmppc_get_gpr(vcpu, 3) == H_CEDE) {
>>> - kvmppc_nested_cede(vcpu);
>>> + kvmppc_cede(vcpu);
>>> kvmppc_set_gpr(vcpu, 3, 0);
>>> trap = 0;
>>> }
>>> } else {
>>> kvmppc_xive_push_vcpu(vcpu);
>>> trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
>>> + if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
>>> + !(vcpu->arch.shregs.msr & MSR_PR)) {
>>> + unsigned long req = kvmppc_get_gpr(vcpu, 3);
>>> +
>>> + /* H_CEDE has to be handled now, not later */
>>> + if (req == H_CEDE) {
>>> + kvmppc_cede(vcpu);
>>> + kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */
>>> + kvmppc_set_gpr(vcpu, 3, 0);
>>> + trap = 0;
>>> +
>>> + /* XICS hcalls must be handled before xive is pulled */
>>> + } else if (hcall_is_xics(req)) {
>>> + int ret;
>>> +
>>> + ret = kvmppc_xive_xics_hcall(vcpu, req);
>>> + if (ret != H_TOO_HARD) {
>>> + kvmppc_set_gpr(vcpu, 3, ret);
>>> + trap = 0;
>>> + }
>>> + }
>>> + }
>>> kvmppc_xive_pull_vcpu(vcpu);
>>> -
>>> }
>>>
>>> vcpu->arch.slb_max = 0;
>>> @@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
>>> else
>>> r = kvmppc_run_vcpu(vcpu);
>>>
>>> - if (run->exit_reason == KVM_EXIT_PAPR_HCALL &&
>>> - !(vcpu->arch.shregs.msr & MSR_PR)) {
>>> + if (run->exit_reason == KVM_EXIT_PAPR_HCALL) {
>>> + if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
>>> + r = RESUME_GUEST;
>>> + continue;
>>> + }
>>> trace_kvm_hcall_enter(vcpu);
>>> r = kvmppc_pseries_do_hcall(vcpu);
>>> trace_kvm_hcall_exit(vcpu, r);
>>> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>>> index c11597f815e4..2d0d14ed1d92 100644
>>> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>>> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>>> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
>>> mr r4,r9
>>> bge fast_guest_return
>>> 2:
>>> + /* If we came in through the P9 short path, no real mode hcalls */
>>> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
>>> + cmpwi r0, 0
>>> + bne no_try_real
>>
>>
>> btw is mmu on at this point? or it gets enabled by rfid at the end of
>> guest_exit_short_path?
>
> Hash guest it's off. Radix guest it can be on or off depending on the
> interrupt type and MSR and LPCR[AIL] values.
What I meant was - what do we expect here on p9? mmu on? ^w^w^w^w^w^w^w^w^w
I just realized - it is radix so there is no problem with vmalloc
addresses in real mode as these do not use top 2 bits as on hash and the
exact mmu state is less important here. Cheers.
>
> Thanks,
> Nick
>
--
Alexey
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
@ 2021-03-23 9:24 ` Alexey Kardashevskiy
0 siblings, 0 replies; 188+ messages in thread
From: Alexey Kardashevskiy @ 2021-03-23 9:24 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 23/03/2021 20:16, Nicholas Piggin wrote:
> Excerpts from Alexey Kardashevskiy's message of March 23, 2021 7:02 pm:
>>
>>
>> On 23/03/2021 12:02, Nicholas Piggin wrote:
>>> In the interest of minimising the amount of code that is run in
>>> "real-mode", don't handle hcalls in real mode in the P9 path.
>>>
>>> POWER8 and earlier are much more expensive to exit from HV real mode
>>> and switch to host mode, because on those processors HV interrupts get
>>> to the hypervisor with the MMU off, and the other threads in the core
>>> need to be pulled out of the guest, and SLBs all need to be saved,
>>> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
>>> in host mode. Hash guests also require a lot of hcalls to run. The
>>> XICS interrupt controller requires hcalls to run.
>>>
>>> By contrast, POWER9 has independent thread switching, and in radix mode
>>> the hypervisor is already in a host virtual memory mode when the HV
>>> interrupt is taken. Radix + xive guests don't need hcalls to handle
>>> interrupts or manage translations.
>>>
>>> So it's much less important to handle hcalls in real mode in P9.
>>>
>>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>>> ---
>>> arch/powerpc/include/asm/kvm_ppc.h | 5 ++
>>> arch/powerpc/kvm/book3s_hv.c | 57 ++++++++++++++++----
>>> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++
>>> arch/powerpc/kvm/book3s_xive.c | 70 +++++++++++++++++++++++++
>>> 4 files changed, 127 insertions(+), 10 deletions(-)
>>>
>>> diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
>>> index 73b1ca5a6471..db6646c2ade2 100644
>>> --- a/arch/powerpc/include/asm/kvm_ppc.h
>>> +++ b/arch/powerpc/include/asm/kvm_ppc.h
>>> @@ -607,6 +607,7 @@ extern void kvmppc_free_pimap(struct kvm *kvm);
>>> extern int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall);
>>> extern void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu);
>>> extern int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd);
>>> +extern int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req);
>>> extern u64 kvmppc_xics_get_icp(struct kvm_vcpu *vcpu);
>>> extern int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
>>> extern int kvmppc_xics_connect_vcpu(struct kvm_device *dev,
>>> @@ -639,6 +640,8 @@ static inline int kvmppc_xics_enabled(struct kvm_vcpu *vcpu)
>>> static inline void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) { }
>>> static inline int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd)
>>> { return 0; }
>>> +static inline int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
>>> + { return 0; }
>>> #endif
>>>
>>> #ifdef CONFIG_KVM_XIVE
>>> @@ -673,6 +676,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
>>> int level, bool line_status);
>>> extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
>>> extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
>>> +extern void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu);
>>>
>>> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
>>> {
>>> @@ -714,6 +718,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
>>> int level, bool line_status) { return -ENODEV; }
>>> static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
>>> static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
>>> +static inline void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu) { }
>>>
>>> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
>>> { return 0; }
>>> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
>>> index fa7614c37e08..17739aaee3d8 100644
>>> --- a/arch/powerpc/kvm/book3s_hv.c
>>> +++ b/arch/powerpc/kvm/book3s_hv.c
>>> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
>>> }
>>>
>>> /*
>>> - * Handle H_CEDE in the nested virtualization case where we haven't
>>> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
>>> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
>>> + * handlers in book3s_hv_rmhandlers.S.
>>> + *
>>> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
>>> * that the cede logic in kvmppc_run_single_vcpu() works properly.
>>> */
>>> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
>>> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
>>> {
>>> vcpu->arch.shregs.msr |= MSR_EE;
>>> vcpu->arch.ceded = 1;
>>> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
>>> /* hcall - punt to userspace */
>>> int i;
>>>
>>> - /* hypercall with MSR_PR has already been handled in rmode,
>>> - * and never reaches here.
>>> - */
>>> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
>>> + /*
>>> + * Guest userspace executed sc 1, reflect it back as a
>>> + * privileged program check interrupt.
>>> + */
>>> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
>>> + r = RESUME_GUEST;
>>> + break;
>>> + }
>>>
>>> run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
>>> for (i = 0; i < 9; ++i)
>>> @@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
>>> return trap;
>>> }
>>>
>>> +static inline bool hcall_is_xics(unsigned long req)
>>> +{
>>> + return (req = H_EOI || req = H_CPPR || req = H_IPI ||
>>> + req = H_IPOLL || req = H_XIRR || req = H_XIRR_X);
>>
>> Do not need braces :)
>>
>>
>>> +}
>>> +
>>> /*
>>> * Virtual-mode guest entry for POWER9 and later when the host and
>>> * guest are both using the radix MMU. The LPIDR has already been set.
>>> @@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
>>> /* H_CEDE has to be handled now, not later */
>>> if (trap = BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
>>> kvmppc_get_gpr(vcpu, 3) = H_CEDE) {
>>> - kvmppc_nested_cede(vcpu);
>>> + kvmppc_cede(vcpu);
>>> kvmppc_set_gpr(vcpu, 3, 0);
>>> trap = 0;
>>> }
>>> } else {
>>> kvmppc_xive_push_vcpu(vcpu);
>>> trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
>>> + if (trap = BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
>>> + !(vcpu->arch.shregs.msr & MSR_PR)) {
>>> + unsigned long req = kvmppc_get_gpr(vcpu, 3);
>>> +
>>> + /* H_CEDE has to be handled now, not later */
>>> + if (req = H_CEDE) {
>>> + kvmppc_cede(vcpu);
>>> + kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */
>>> + kvmppc_set_gpr(vcpu, 3, 0);
>>> + trap = 0;
>>> +
>>> + /* XICS hcalls must be handled before xive is pulled */
>>> + } else if (hcall_is_xics(req)) {
>>> + int ret;
>>> +
>>> + ret = kvmppc_xive_xics_hcall(vcpu, req);
>>> + if (ret != H_TOO_HARD) {
>>> + kvmppc_set_gpr(vcpu, 3, ret);
>>> + trap = 0;
>>> + }
>>> + }
>>> + }
>>> kvmppc_xive_pull_vcpu(vcpu);
>>> -
>>> }
>>>
>>> vcpu->arch.slb_max = 0;
>>> @@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
>>> else
>>> r = kvmppc_run_vcpu(vcpu);
>>>
>>> - if (run->exit_reason = KVM_EXIT_PAPR_HCALL &&
>>> - !(vcpu->arch.shregs.msr & MSR_PR)) {
>>> + if (run->exit_reason = KVM_EXIT_PAPR_HCALL) {
>>> + if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
>>> + r = RESUME_GUEST;
>>> + continue;
>>> + }
>>> trace_kvm_hcall_enter(vcpu);
>>> r = kvmppc_pseries_do_hcall(vcpu);
>>> trace_kvm_hcall_exit(vcpu, r);
>>> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>>> index c11597f815e4..2d0d14ed1d92 100644
>>> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>>> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>>> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
>>> mr r4,r9
>>> bge fast_guest_return
>>> 2:
>>> + /* If we came in through the P9 short path, no real mode hcalls */
>>> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
>>> + cmpwi r0, 0
>>> + bne no_try_real
>>
>>
>> btw is mmu on at this point? or it gets enabled by rfid at the end of
>> guest_exit_short_path?
>
> Hash guest it's off. Radix guest it can be on or off depending on the
> interrupt type and MSR and LPCR[AIL] values.
What I meant was - what do we expect here on p9? mmu on? ^w^w^w^w^w^w^w^w^w
I just realized - it is radix so there is no problem with vmalloc
addresses in real mode as these do not use top 2 bits as on hash and the
exact mmu state is less important here. Cheers.
>
> Thanks,
> Nick
>
--
Alexey
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
2021-03-23 9:24 ` Alexey Kardashevskiy
@ 2021-03-23 9:48 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 9:48 UTC (permalink / raw)
To: Alexey Kardashevskiy, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Alexey Kardashevskiy's message of March 23, 2021 7:24 pm:
>
>
> On 23/03/2021 20:16, Nicholas Piggin wrote:
>> Excerpts from Alexey Kardashevskiy's message of March 23, 2021 7:02 pm:
>>>
>>>
>>> On 23/03/2021 12:02, Nicholas Piggin wrote:
>>>> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>>>> index c11597f815e4..2d0d14ed1d92 100644
>>>> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>>>> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>>>> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
>>>> mr r4,r9
>>>> bge fast_guest_return
>>>> 2:
>>>> + /* If we came in through the P9 short path, no real mode hcalls */
>>>> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
>>>> + cmpwi r0, 0
>>>> + bne no_try_real
>>>
>>>
>>> btw is mmu on at this point? or it gets enabled by rfid at the end of
>>> guest_exit_short_path?
>>
>> Hash guest it's off. Radix guest it can be on or off depending on the
>> interrupt type and MSR and LPCR[AIL] values.
>
> What I meant was - what do we expect here on p9? mmu on? ^w^w^w^w^w^w^w^w^w
P9 radix can be on or off. If the guest had MSR[IR] or MSR[DR] clear, or
if the guest is running AIL=0 mode, or if this is a machine check,
system reset, or HMI interrupt then the MMU will be off here.
> I just realized - it is radix so there is no problem with vmalloc
> addresses in real mode as these do not use top 2 bits as on hash and the
> exact mmu state is less important here. Cheers.
We still can't use vmalloc addresses in real mode on radix because they
don't translate with the page tables.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
@ 2021-03-23 9:48 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 9:48 UTC (permalink / raw)
To: Alexey Kardashevskiy, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Alexey Kardashevskiy's message of March 23, 2021 7:24 pm:
>
>
> On 23/03/2021 20:16, Nicholas Piggin wrote:
>> Excerpts from Alexey Kardashevskiy's message of March 23, 2021 7:02 pm:
>>>
>>>
>>> On 23/03/2021 12:02, Nicholas Piggin wrote:
>>>> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>>>> index c11597f815e4..2d0d14ed1d92 100644
>>>> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>>>> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>>>> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
>>>> mr r4,r9
>>>> bge fast_guest_return
>>>> 2:
>>>> + /* If we came in through the P9 short path, no real mode hcalls */
>>>> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
>>>> + cmpwi r0, 0
>>>> + bne no_try_real
>>>
>>>
>>> btw is mmu on at this point? or it gets enabled by rfid at the end of
>>> guest_exit_short_path?
>>
>> Hash guest it's off. Radix guest it can be on or off depending on the
>> interrupt type and MSR and LPCR[AIL] values.
>
> What I meant was - what do we expect here on p9? mmu on? ^w^w^w^w^w^w^w^w^w
P9 radix can be on or off. If the guest had MSR[IR] or MSR[DR] clear, or
if the guest is running AIL=0 mode, or if this is a machine check,
system reset, or HMI interrupt then the MMU will be off here.
> I just realized - it is radix so there is no problem with vmalloc
> addresses in real mode as these do not use top 2 bits as on hash and the
> exact mmu state is less important here. Cheers.
We still can't use vmalloc addresses in real mode on radix because they
don't translate with the page tables.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 13:23 ` Cédric Le Goater
-1 siblings, 0 replies; 188+ messages in thread
From: Cédric Le Goater @ 2021-03-23 13:23 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 3/23/21 2:02 AM, Nicholas Piggin wrote:
> In the interest of minimising the amount of code that is run in
> "real-mode", don't handle hcalls in real mode in the P9 path.
>
> POWER8 and earlier are much more expensive to exit from HV real mode
> and switch to host mode, because on those processors HV interrupts get
> to the hypervisor with the MMU off, and the other threads in the core
> need to be pulled out of the guest, and SLBs all need to be saved,
> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
> in host mode. Hash guests also require a lot of hcalls to run. The
> XICS interrupt controller requires hcalls to run.
>
> By contrast, POWER9 has independent thread switching, and in radix mode
> the hypervisor is already in a host virtual memory mode when the HV
> interrupt is taken. Radix + xive guests don't need hcalls to handle
> interrupts or manage translations.
>
> So it's much less important to handle hcalls in real mode in P9.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
> arch/powerpc/include/asm/kvm_ppc.h | 5 ++
> arch/powerpc/kvm/book3s_hv.c | 57 ++++++++++++++++----
> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++
> arch/powerpc/kvm/book3s_xive.c | 70 +++++++++++++++++++++++++
> 4 files changed, 127 insertions(+), 10 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
> index 73b1ca5a6471..db6646c2ade2 100644
> --- a/arch/powerpc/include/asm/kvm_ppc.h
> +++ b/arch/powerpc/include/asm/kvm_ppc.h
> @@ -607,6 +607,7 @@ extern void kvmppc_free_pimap(struct kvm *kvm);
> extern int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall);
> extern void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu);
> extern int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd);
> +extern int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req);
> extern u64 kvmppc_xics_get_icp(struct kvm_vcpu *vcpu);
> extern int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
> extern int kvmppc_xics_connect_vcpu(struct kvm_device *dev,
> @@ -639,6 +640,8 @@ static inline int kvmppc_xics_enabled(struct kvm_vcpu *vcpu)
> static inline void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) { }
> static inline int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd)
> { return 0; }
> +static inline int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
> + { return 0; }
> #endif
>
> #ifdef CONFIG_KVM_XIVE
> @@ -673,6 +676,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
> int level, bool line_status);
> extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
> extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
> +extern void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu);
>
> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> {
> @@ -714,6 +718,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
> int level, bool line_status) { return -ENODEV; }
> static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
> static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
> +static inline void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu) { }
>
> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> { return 0; }
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index fa7614c37e08..17739aaee3d8 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
> }
>
> /*
> - * Handle H_CEDE in the nested virtualization case where we haven't
> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
> + * handlers in book3s_hv_rmhandlers.S.
> + *
> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
> * that the cede logic in kvmppc_run_single_vcpu() works properly.
> */
> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
> {
> vcpu->arch.shregs.msr |= MSR_EE;
> vcpu->arch.ceded = 1;
> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
> /* hcall - punt to userspace */
> int i;
>
> - /* hypercall with MSR_PR has already been handled in rmode,
> - * and never reaches here.
> - */
> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
> + /*
> + * Guest userspace executed sc 1, reflect it back as a
> + * privileged program check interrupt.
> + */
> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
> + r = RESUME_GUEST;
> + break;
> + }
>
> run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
> for (i = 0; i < 9; ++i)
> @@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
> return trap;
> }
>
> +static inline bool hcall_is_xics(unsigned long req)
> +{
> + return (req == H_EOI || req == H_CPPR || req == H_IPI ||
> + req == H_IPOLL || req == H_XIRR || req == H_XIRR_X);
> +}
> +
> /*
> * Virtual-mode guest entry for POWER9 and later when the host and
> * guest are both using the radix MMU. The LPIDR has already been set.
> @@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> /* H_CEDE has to be handled now, not later */
> if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> kvmppc_get_gpr(vcpu, 3) == H_CEDE) {
> - kvmppc_nested_cede(vcpu);
> + kvmppc_cede(vcpu);
> kvmppc_set_gpr(vcpu, 3, 0);
> trap = 0;
> }
> } else {
> kvmppc_xive_push_vcpu(vcpu);
> trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
> + if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> + !(vcpu->arch.shregs.msr & MSR_PR)) {
> + unsigned long req = kvmppc_get_gpr(vcpu, 3);
> +
> + /* H_CEDE has to be handled now, not later */
> + if (req == H_CEDE) {
> + kvmppc_cede(vcpu);
> + kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */> + kvmppc_set_gpr(vcpu, 3, 0);
> + trap = 0;
> +
> + /* XICS hcalls must be handled before xive is pulled */
> + } else if (hcall_is_xics(req)) {
> + int ret;
> +
> + ret = kvmppc_xive_xics_hcall(vcpu, req);
> + if (ret != H_TOO_HARD) {
> + kvmppc_set_gpr(vcpu, 3, ret);
> + trap = 0;
> + }
> + }
> + }
> kvmppc_xive_pull_vcpu(vcpu);
> -
> }
>
> vcpu->arch.slb_max = 0;
> @@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
> else
> r = kvmppc_run_vcpu(vcpu);
>
> - if (run->exit_reason == KVM_EXIT_PAPR_HCALL &&
> - !(vcpu->arch.shregs.msr & MSR_PR)) {
> + if (run->exit_reason == KVM_EXIT_PAPR_HCALL) {
> + if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
> + r = RESUME_GUEST;
> + continue;
> + }
> trace_kvm_hcall_enter(vcpu);
> r = kvmppc_pseries_do_hcall(vcpu);
> trace_kvm_hcall_exit(vcpu, r);
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index c11597f815e4..2d0d14ed1d92 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
> mr r4,r9
> bge fast_guest_return
> 2:
> + /* If we came in through the P9 short path, no real mode hcalls */
> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
> + cmpwi r0, 0
> + bne no_try_real
> /* See if this is an hcall we can handle in real mode */
> cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
> beq hcall_try_real_mode
> +no_try_real:
>
> /* Hypervisor doorbell - exit only if host IPI flag set */
> cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
> diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
> index 741bf1f4387a..dcc07ceaf5ca 100644
> --- a/arch/powerpc/kvm/book3s_xive.c
> +++ b/arch/powerpc/kvm/book3s_xive.c
> @@ -158,6 +158,40 @@ void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu)
> }
> EXPORT_SYMBOL_GPL(kvmppc_xive_pull_vcpu);
>
> +void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu)
kvmppc_xive_rearm_escalation() may be ? It has more meaning to me.
> +{
> + void __iomem *esc_vaddr = (void __iomem *)vcpu->arch.xive_esc_vaddr;
> +
> + if (!esc_vaddr)
> + return;
> +
> + /* we are using XIVE with single escalation */
> +
> + if (vcpu->arch.xive_esc_on) {
> + /*
> + * If we still have a pending escalation, abort the cede,
> + * and we must set PQ to 10 rather than 00 so that we don't
> + * potentially end up with two entries for the escalation
> + * interrupt in the XIVE interrupt queue. In that case
> + * we also don't want to set xive_esc_on to 1 here in
> + * case we race with xive_esc_irq().
> + */
> + vcpu->arch.ceded = 0;
> + /*
> + * The escalation interrupts are special as we don't EOI them.
> + * There is no need to use the load-after-store ordering offset
> + * to set PQ to 10 as we won't use StoreEOI.
> + */
> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_10);
> + } else {
> + vcpu->arch.xive_esc_on = true;> + mb();
> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_00);
> + }
> + mb();
> +}
> +EXPORT_SYMBOL_GPL(kvmppc_xive_cede_vcpu);
> +
> /*
> * This is a simple trigger for a generic XIVE IRQ. This must
> * only be called for interrupts that support a trigger page
> @@ -2106,6 +2140,42 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type)
> return 0;
> }
>
> +int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
> +{
> + struct kvmppc_vcore *vc = vcpu->arch.vcore;
> +
> + /*
> + * This test covers the case in which a vCPU does XICS hcalls without
> + * QEMU having connected the vCPU to a XICS ICP. The ICP is the KVM
> + * XICS device on P8 or XICS-on-XIVE on P9. It catches QEMU errors when
> + * the interrupt mode is negotiated, we don't want the OS to do XICS
> + * hcalls after having negotiated the XIVE interrupt mode.
> + */
I think a comment like the following should be enough.
The VM should have configured XICS mode before doing XICS hcalls.
No need to resend for that though.
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Thanks,
C.
> + if (!kvmppc_xics_enabled(vcpu))
> + return H_TOO_HARD;
> +
> + switch (req) {
> + case H_XIRR:
> + return xive_vm_h_xirr(vcpu);
> + case H_CPPR:
> + return xive_vm_h_cppr(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_EOI:
> + return xive_vm_h_eoi(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_IPI:
> + return xive_vm_h_ipi(vcpu, kvmppc_get_gpr(vcpu, 4),
> + kvmppc_get_gpr(vcpu, 5));
> + case H_IPOLL:
> + return xive_vm_h_ipoll(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_XIRR_X:
> + xive_vm_h_xirr(vcpu);
> + kvmppc_set_gpr(vcpu, 5, get_tb() + vc->tb_offset);
> + return H_SUCCESS;
> + }
> +
> + return H_UNSUPPORTED;
> +}
> +EXPORT_SYMBOL_GPL(kvmppc_xive_xics_hcall);
> +
> int kvmppc_xive_debug_show_queues(struct seq_file *m, struct kvm_vcpu *vcpu)
> {
> struct kvmppc_xive_vcpu *xc = vcpu->arch.xive_vcpu;
>
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
@ 2021-03-23 13:23 ` Cédric Le Goater
0 siblings, 0 replies; 188+ messages in thread
From: Cédric Le Goater @ 2021-03-23 13:23 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 3/23/21 2:02 AM, Nicholas Piggin wrote:
> In the interest of minimising the amount of code that is run in
> "real-mode", don't handle hcalls in real mode in the P9 path.
>
> POWER8 and earlier are much more expensive to exit from HV real mode
> and switch to host mode, because on those processors HV interrupts get
> to the hypervisor with the MMU off, and the other threads in the core
> need to be pulled out of the guest, and SLBs all need to be saved,
> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
> in host mode. Hash guests also require a lot of hcalls to run. The
> XICS interrupt controller requires hcalls to run.
>
> By contrast, POWER9 has independent thread switching, and in radix mode
> the hypervisor is already in a host virtual memory mode when the HV
> interrupt is taken. Radix + xive guests don't need hcalls to handle
> interrupts or manage translations.
>
> So it's much less important to handle hcalls in real mode in P9.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
> arch/powerpc/include/asm/kvm_ppc.h | 5 ++
> arch/powerpc/kvm/book3s_hv.c | 57 ++++++++++++++++----
> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++
> arch/powerpc/kvm/book3s_xive.c | 70 +++++++++++++++++++++++++
> 4 files changed, 127 insertions(+), 10 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
> index 73b1ca5a6471..db6646c2ade2 100644
> --- a/arch/powerpc/include/asm/kvm_ppc.h
> +++ b/arch/powerpc/include/asm/kvm_ppc.h
> @@ -607,6 +607,7 @@ extern void kvmppc_free_pimap(struct kvm *kvm);
> extern int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall);
> extern void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu);
> extern int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd);
> +extern int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req);
> extern u64 kvmppc_xics_get_icp(struct kvm_vcpu *vcpu);
> extern int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
> extern int kvmppc_xics_connect_vcpu(struct kvm_device *dev,
> @@ -639,6 +640,8 @@ static inline int kvmppc_xics_enabled(struct kvm_vcpu *vcpu)
> static inline void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) { }
> static inline int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd)
> { return 0; }
> +static inline int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
> + { return 0; }
> #endif
>
> #ifdef CONFIG_KVM_XIVE
> @@ -673,6 +676,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
> int level, bool line_status);
> extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
> extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
> +extern void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu);
>
> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> {
> @@ -714,6 +718,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
> int level, bool line_status) { return -ENODEV; }
> static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
> static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
> +static inline void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu) { }
>
> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> { return 0; }
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index fa7614c37e08..17739aaee3d8 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
> }
>
> /*
> - * Handle H_CEDE in the nested virtualization case where we haven't
> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
> + * handlers in book3s_hv_rmhandlers.S.
> + *
> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
> * that the cede logic in kvmppc_run_single_vcpu() works properly.
> */
> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
> {
> vcpu->arch.shregs.msr |= MSR_EE;
> vcpu->arch.ceded = 1;
> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
> /* hcall - punt to userspace */
> int i;
>
> - /* hypercall with MSR_PR has already been handled in rmode,
> - * and never reaches here.
> - */
> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
> + /*
> + * Guest userspace executed sc 1, reflect it back as a
> + * privileged program check interrupt.
> + */
> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
> + r = RESUME_GUEST;
> + break;
> + }
>
> run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
> for (i = 0; i < 9; ++i)
> @@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
> return trap;
> }
>
> +static inline bool hcall_is_xics(unsigned long req)
> +{
> + return (req = H_EOI || req = H_CPPR || req = H_IPI ||
> + req = H_IPOLL || req = H_XIRR || req = H_XIRR_X);
> +}
> +
> /*
> * Virtual-mode guest entry for POWER9 and later when the host and
> * guest are both using the radix MMU. The LPIDR has already been set.
> @@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> /* H_CEDE has to be handled now, not later */
> if (trap = BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> kvmppc_get_gpr(vcpu, 3) = H_CEDE) {
> - kvmppc_nested_cede(vcpu);
> + kvmppc_cede(vcpu);
> kvmppc_set_gpr(vcpu, 3, 0);
> trap = 0;
> }
> } else {
> kvmppc_xive_push_vcpu(vcpu);
> trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
> + if (trap = BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> + !(vcpu->arch.shregs.msr & MSR_PR)) {
> + unsigned long req = kvmppc_get_gpr(vcpu, 3);
> +
> + /* H_CEDE has to be handled now, not later */
> + if (req = H_CEDE) {
> + kvmppc_cede(vcpu);
> + kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */> + kvmppc_set_gpr(vcpu, 3, 0);
> + trap = 0;
> +
> + /* XICS hcalls must be handled before xive is pulled */
> + } else if (hcall_is_xics(req)) {
> + int ret;
> +
> + ret = kvmppc_xive_xics_hcall(vcpu, req);
> + if (ret != H_TOO_HARD) {
> + kvmppc_set_gpr(vcpu, 3, ret);
> + trap = 0;
> + }
> + }
> + }
> kvmppc_xive_pull_vcpu(vcpu);
> -
> }
>
> vcpu->arch.slb_max = 0;
> @@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
> else
> r = kvmppc_run_vcpu(vcpu);
>
> - if (run->exit_reason = KVM_EXIT_PAPR_HCALL &&
> - !(vcpu->arch.shregs.msr & MSR_PR)) {
> + if (run->exit_reason = KVM_EXIT_PAPR_HCALL) {
> + if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
> + r = RESUME_GUEST;
> + continue;
> + }
> trace_kvm_hcall_enter(vcpu);
> r = kvmppc_pseries_do_hcall(vcpu);
> trace_kvm_hcall_exit(vcpu, r);
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index c11597f815e4..2d0d14ed1d92 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
> mr r4,r9
> bge fast_guest_return
> 2:
> + /* If we came in through the P9 short path, no real mode hcalls */
> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
> + cmpwi r0, 0
> + bne no_try_real
> /* See if this is an hcall we can handle in real mode */
> cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
> beq hcall_try_real_mode
> +no_try_real:
>
> /* Hypervisor doorbell - exit only if host IPI flag set */
> cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
> diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
> index 741bf1f4387a..dcc07ceaf5ca 100644
> --- a/arch/powerpc/kvm/book3s_xive.c
> +++ b/arch/powerpc/kvm/book3s_xive.c
> @@ -158,6 +158,40 @@ void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu)
> }
> EXPORT_SYMBOL_GPL(kvmppc_xive_pull_vcpu);
>
> +void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu)
kvmppc_xive_rearm_escalation() may be ? It has more meaning to me.
> +{
> + void __iomem *esc_vaddr = (void __iomem *)vcpu->arch.xive_esc_vaddr;
> +
> + if (!esc_vaddr)
> + return;
> +
> + /* we are using XIVE with single escalation */
> +
> + if (vcpu->arch.xive_esc_on) {
> + /*
> + * If we still have a pending escalation, abort the cede,
> + * and we must set PQ to 10 rather than 00 so that we don't
> + * potentially end up with two entries for the escalation
> + * interrupt in the XIVE interrupt queue. In that case
> + * we also don't want to set xive_esc_on to 1 here in
> + * case we race with xive_esc_irq().
> + */
> + vcpu->arch.ceded = 0;
> + /*
> + * The escalation interrupts are special as we don't EOI them.
> + * There is no need to use the load-after-store ordering offset
> + * to set PQ to 10 as we won't use StoreEOI.
> + */
> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_10);
> + } else {
> + vcpu->arch.xive_esc_on = true;> + mb();
> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_00);
> + }
> + mb();
> +}
> +EXPORT_SYMBOL_GPL(kvmppc_xive_cede_vcpu);
> +
> /*
> * This is a simple trigger for a generic XIVE IRQ. This must
> * only be called for interrupts that support a trigger page
> @@ -2106,6 +2140,42 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type)
> return 0;
> }
>
> +int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
> +{
> + struct kvmppc_vcore *vc = vcpu->arch.vcore;
> +
> + /*
> + * This test covers the case in which a vCPU does XICS hcalls without
> + * QEMU having connected the vCPU to a XICS ICP. The ICP is the KVM
> + * XICS device on P8 or XICS-on-XIVE on P9. It catches QEMU errors when
> + * the interrupt mode is negotiated, we don't want the OS to do XICS
> + * hcalls after having negotiated the XIVE interrupt mode.
> + */
I think a comment like the following should be enough.
The VM should have configured XICS mode before doing XICS hcalls.
No need to resend for that though.
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Thanks,
C.
> + if (!kvmppc_xics_enabled(vcpu))
> + return H_TOO_HARD;
> +
> + switch (req) {
> + case H_XIRR:
> + return xive_vm_h_xirr(vcpu);
> + case H_CPPR:
> + return xive_vm_h_cppr(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_EOI:
> + return xive_vm_h_eoi(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_IPI:
> + return xive_vm_h_ipi(vcpu, kvmppc_get_gpr(vcpu, 4),
> + kvmppc_get_gpr(vcpu, 5));
> + case H_IPOLL:
> + return xive_vm_h_ipoll(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_XIRR_X:
> + xive_vm_h_xirr(vcpu);
> + kvmppc_set_gpr(vcpu, 5, get_tb() + vc->tb_offset);
> + return H_SUCCESS;
> + }
> +
> + return H_UNSUPPORTED;
> +}
> +EXPORT_SYMBOL_GPL(kvmppc_xive_xics_hcall);
> +
> int kvmppc_xive_debug_show_queues(struct seq_file *m, struct kvm_vcpu *vcpu)
> {
> struct kvmppc_xive_vcpu *xc = vcpu->arch.xive_vcpu;
>
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
2021-03-23 13:23 ` Cédric Le Goater
@ 2021-03-24 1:21 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-24 1:21 UTC (permalink / raw)
To: Cédric Le Goater, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Cédric Le Goater's message of March 23, 2021 11:23 pm:
> On 3/23/21 2:02 AM, Nicholas Piggin wrote:
>> In the interest of minimising the amount of code that is run in
>> "real-mode", don't handle hcalls in real mode in the P9 path.
>>
>> POWER8 and earlier are much more expensive to exit from HV real mode
>> and switch to host mode, because on those processors HV interrupts get
>> to the hypervisor with the MMU off, and the other threads in the core
>> need to be pulled out of the guest, and SLBs all need to be saved,
>> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
>> in host mode. Hash guests also require a lot of hcalls to run. The
>> XICS interrupt controller requires hcalls to run.
>>
>> By contrast, POWER9 has independent thread switching, and in radix mode
>> the hypervisor is already in a host virtual memory mode when the HV
>> interrupt is taken. Radix + xive guests don't need hcalls to handle
>> interrupts or manage translations.
>>
>> So it's much less important to handle hcalls in real mode in P9.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>> ---
>> arch/powerpc/include/asm/kvm_ppc.h | 5 ++
>> arch/powerpc/kvm/book3s_hv.c | 57 ++++++++++++++++----
>> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++
>> arch/powerpc/kvm/book3s_xive.c | 70 +++++++++++++++++++++++++
>> 4 files changed, 127 insertions(+), 10 deletions(-)
>>
>> diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
>> index 73b1ca5a6471..db6646c2ade2 100644
>> --- a/arch/powerpc/include/asm/kvm_ppc.h
>> +++ b/arch/powerpc/include/asm/kvm_ppc.h
>> @@ -607,6 +607,7 @@ extern void kvmppc_free_pimap(struct kvm *kvm);
>> extern int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall);
>> extern void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu);
>> extern int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd);
>> +extern int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req);
>> extern u64 kvmppc_xics_get_icp(struct kvm_vcpu *vcpu);
>> extern int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
>> extern int kvmppc_xics_connect_vcpu(struct kvm_device *dev,
>> @@ -639,6 +640,8 @@ static inline int kvmppc_xics_enabled(struct kvm_vcpu *vcpu)
>> static inline void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) { }
>> static inline int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd)
>> { return 0; }
>> +static inline int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
>> + { return 0; }
>> #endif
>>
>> #ifdef CONFIG_KVM_XIVE
>> @@ -673,6 +676,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
>> int level, bool line_status);
>> extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
>> extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
>> +extern void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu);
>>
>> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
>> {
>> @@ -714,6 +718,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
>> int level, bool line_status) { return -ENODEV; }
>> static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
>> static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
>> +static inline void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu) { }
>>
>> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
>> { return 0; }
>> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
>> index fa7614c37e08..17739aaee3d8 100644
>> --- a/arch/powerpc/kvm/book3s_hv.c
>> +++ b/arch/powerpc/kvm/book3s_hv.c
>> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
>> }
>>
>> /*
>> - * Handle H_CEDE in the nested virtualization case where we haven't
>> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
>> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
>> + * handlers in book3s_hv_rmhandlers.S.
>> + *
>> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
>> * that the cede logic in kvmppc_run_single_vcpu() works properly.
>> */
>> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
>> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
>> {
>> vcpu->arch.shregs.msr |= MSR_EE;
>> vcpu->arch.ceded = 1;
>> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
>> /* hcall - punt to userspace */
>> int i;
>>
>> - /* hypercall with MSR_PR has already been handled in rmode,
>> - * and never reaches here.
>> - */
>> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
>> + /*
>> + * Guest userspace executed sc 1, reflect it back as a
>> + * privileged program check interrupt.
>> + */
>> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
>> + r = RESUME_GUEST;
>> + break;
>> + }
>>
>> run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
>> for (i = 0; i < 9; ++i)
>> @@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
>> return trap;
>> }
>>
>> +static inline bool hcall_is_xics(unsigned long req)
>> +{
>> + return (req == H_EOI || req == H_CPPR || req == H_IPI ||
>> + req == H_IPOLL || req == H_XIRR || req == H_XIRR_X);
>> +}
>> +
>> /*
>> * Virtual-mode guest entry for POWER9 and later when the host and
>> * guest are both using the radix MMU. The LPIDR has already been set.
>> @@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
>> /* H_CEDE has to be handled now, not later */
>> if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
>> kvmppc_get_gpr(vcpu, 3) == H_CEDE) {
>> - kvmppc_nested_cede(vcpu);
>> + kvmppc_cede(vcpu);
>> kvmppc_set_gpr(vcpu, 3, 0);
>> trap = 0;
>> }
>> } else {
>> kvmppc_xive_push_vcpu(vcpu);
>> trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
>> + if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
>> + !(vcpu->arch.shregs.msr & MSR_PR)) {
>> + unsigned long req = kvmppc_get_gpr(vcpu, 3);
>> +
>> + /* H_CEDE has to be handled now, not later */
>> + if (req == H_CEDE) {
>> + kvmppc_cede(vcpu);
>> + kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */> + kvmppc_set_gpr(vcpu, 3, 0);
>> + trap = 0;
>> +
>> + /* XICS hcalls must be handled before xive is pulled */
>> + } else if (hcall_is_xics(req)) {
>> + int ret;
>> +
>> + ret = kvmppc_xive_xics_hcall(vcpu, req);
>> + if (ret != H_TOO_HARD) {
>> + kvmppc_set_gpr(vcpu, 3, ret);
>> + trap = 0;
>> + }
>> + }
>> + }
>> kvmppc_xive_pull_vcpu(vcpu);
>> -
>> }
>>
>> vcpu->arch.slb_max = 0;
>> @@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
>> else
>> r = kvmppc_run_vcpu(vcpu);
>>
>> - if (run->exit_reason == KVM_EXIT_PAPR_HCALL &&
>> - !(vcpu->arch.shregs.msr & MSR_PR)) {
>> + if (run->exit_reason == KVM_EXIT_PAPR_HCALL) {
>> + if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
>> + r = RESUME_GUEST;
>> + continue;
>> + }
>> trace_kvm_hcall_enter(vcpu);
>> r = kvmppc_pseries_do_hcall(vcpu);
>> trace_kvm_hcall_exit(vcpu, r);
>> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>> index c11597f815e4..2d0d14ed1d92 100644
>> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
>> mr r4,r9
>> bge fast_guest_return
>> 2:
>> + /* If we came in through the P9 short path, no real mode hcalls */
>> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
>> + cmpwi r0, 0
>> + bne no_try_real
>> /* See if this is an hcall we can handle in real mode */
>> cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
>> beq hcall_try_real_mode
>> +no_try_real:
>>
>> /* Hypervisor doorbell - exit only if host IPI flag set */
>> cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
>> diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
>> index 741bf1f4387a..dcc07ceaf5ca 100644
>> --- a/arch/powerpc/kvm/book3s_xive.c
>> +++ b/arch/powerpc/kvm/book3s_xive.c
>> @@ -158,6 +158,40 @@ void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu)
>> }
>> EXPORT_SYMBOL_GPL(kvmppc_xive_pull_vcpu);
>>
>> +void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu)
>
> kvmppc_xive_rearm_escalation() may be ? It has more meaning to me.
Sure whatever you prefer.
>
>> +{
>> + void __iomem *esc_vaddr = (void __iomem *)vcpu->arch.xive_esc_vaddr;
>> +
>> + if (!esc_vaddr)
>> + return;
>> +
>> + /* we are using XIVE with single escalation */
>> +
>> + if (vcpu->arch.xive_esc_on) {
>> + /*
>> + * If we still have a pending escalation, abort the cede,
>> + * and we must set PQ to 10 rather than 00 so that we don't
>> + * potentially end up with two entries for the escalation
>> + * interrupt in the XIVE interrupt queue. In that case
>> + * we also don't want to set xive_esc_on to 1 here in
>> + * case we race with xive_esc_irq().
>> + */
>> + vcpu->arch.ceded = 0;
>> + /*
>> + * The escalation interrupts are special as we don't EOI them.
>> + * There is no need to use the load-after-store ordering offset
>> + * to set PQ to 10 as we won't use StoreEOI.
>> + */
>> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_10);
>> + } else {
>> + vcpu->arch.xive_esc_on = true;> + mb();
>> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_00);
>> + }
>> + mb();
>> +}
>> +EXPORT_SYMBOL_GPL(kvmppc_xive_cede_vcpu);
>> +
>> /*
>> * This is a simple trigger for a generic XIVE IRQ. This must
>> * only be called for interrupts that support a trigger page
>> @@ -2106,6 +2140,42 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type)
>> return 0;
>> }
>>
>> +int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
>> +{
>> + struct kvmppc_vcore *vc = vcpu->arch.vcore;
>> +
>> + /*
>> + * This test covers the case in which a vCPU does XICS hcalls without
>> + * QEMU having connected the vCPU to a XICS ICP. The ICP is the KVM
>> + * XICS device on P8 or XICS-on-XIVE on P9. It catches QEMU errors when
>> + * the interrupt mode is negotiated, we don't want the OS to do XICS
>> + * hcalls after having negotiated the XIVE interrupt mode.
>> + */
>
> I think a comment like the following should be enough.
>
> The VM should have configured XICS mode before doing XICS hcalls.
I'll change it.
> No need to resend for that though.
>
> Reviewed-by: Cédric Le Goater <clg@kaod.org>
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
@ 2021-03-24 1:21 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-24 1:21 UTC (permalink / raw)
To: Cédric Le Goater, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Cédric Le Goater's message of March 23, 2021 11:23 pm:
> On 3/23/21 2:02 AM, Nicholas Piggin wrote:
>> In the interest of minimising the amount of code that is run in
>> "real-mode", don't handle hcalls in real mode in the P9 path.
>>
>> POWER8 and earlier are much more expensive to exit from HV real mode
>> and switch to host mode, because on those processors HV interrupts get
>> to the hypervisor with the MMU off, and the other threads in the core
>> need to be pulled out of the guest, and SLBs all need to be saved,
>> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
>> in host mode. Hash guests also require a lot of hcalls to run. The
>> XICS interrupt controller requires hcalls to run.
>>
>> By contrast, POWER9 has independent thread switching, and in radix mode
>> the hypervisor is already in a host virtual memory mode when the HV
>> interrupt is taken. Radix + xive guests don't need hcalls to handle
>> interrupts or manage translations.
>>
>> So it's much less important to handle hcalls in real mode in P9.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>> ---
>> arch/powerpc/include/asm/kvm_ppc.h | 5 ++
>> arch/powerpc/kvm/book3s_hv.c | 57 ++++++++++++++++----
>> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++
>> arch/powerpc/kvm/book3s_xive.c | 70 +++++++++++++++++++++++++
>> 4 files changed, 127 insertions(+), 10 deletions(-)
>>
>> diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
>> index 73b1ca5a6471..db6646c2ade2 100644
>> --- a/arch/powerpc/include/asm/kvm_ppc.h
>> +++ b/arch/powerpc/include/asm/kvm_ppc.h
>> @@ -607,6 +607,7 @@ extern void kvmppc_free_pimap(struct kvm *kvm);
>> extern int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall);
>> extern void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu);
>> extern int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd);
>> +extern int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req);
>> extern u64 kvmppc_xics_get_icp(struct kvm_vcpu *vcpu);
>> extern int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
>> extern int kvmppc_xics_connect_vcpu(struct kvm_device *dev,
>> @@ -639,6 +640,8 @@ static inline int kvmppc_xics_enabled(struct kvm_vcpu *vcpu)
>> static inline void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) { }
>> static inline int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd)
>> { return 0; }
>> +static inline int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
>> + { return 0; }
>> #endif
>>
>> #ifdef CONFIG_KVM_XIVE
>> @@ -673,6 +676,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
>> int level, bool line_status);
>> extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
>> extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
>> +extern void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu);
>>
>> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
>> {
>> @@ -714,6 +718,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
>> int level, bool line_status) { return -ENODEV; }
>> static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
>> static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
>> +static inline void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu) { }
>>
>> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
>> { return 0; }
>> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
>> index fa7614c37e08..17739aaee3d8 100644
>> --- a/arch/powerpc/kvm/book3s_hv.c
>> +++ b/arch/powerpc/kvm/book3s_hv.c
>> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
>> }
>>
>> /*
>> - * Handle H_CEDE in the nested virtualization case where we haven't
>> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
>> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
>> + * handlers in book3s_hv_rmhandlers.S.
>> + *
>> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
>> * that the cede logic in kvmppc_run_single_vcpu() works properly.
>> */
>> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
>> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
>> {
>> vcpu->arch.shregs.msr |= MSR_EE;
>> vcpu->arch.ceded = 1;
>> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
>> /* hcall - punt to userspace */
>> int i;
>>
>> - /* hypercall with MSR_PR has already been handled in rmode,
>> - * and never reaches here.
>> - */
>> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
>> + /*
>> + * Guest userspace executed sc 1, reflect it back as a
>> + * privileged program check interrupt.
>> + */
>> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
>> + r = RESUME_GUEST;
>> + break;
>> + }
>>
>> run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
>> for (i = 0; i < 9; ++i)
>> @@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
>> return trap;
>> }
>>
>> +static inline bool hcall_is_xics(unsigned long req)
>> +{
>> + return (req == H_EOI || req == H_CPPR || req == H_IPI ||
>> + req == H_IPOLL || req == H_XIRR || req == H_XIRR_X);
>> +}
>> +
>> /*
>> * Virtual-mode guest entry for POWER9 and later when the host and
>> * guest are both using the radix MMU. The LPIDR has already been set.
>> @@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
>> /* H_CEDE has to be handled now, not later */
>> if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
>> kvmppc_get_gpr(vcpu, 3) == H_CEDE) {
>> - kvmppc_nested_cede(vcpu);
>> + kvmppc_cede(vcpu);
>> kvmppc_set_gpr(vcpu, 3, 0);
>> trap = 0;
>> }
>> } else {
>> kvmppc_xive_push_vcpu(vcpu);
>> trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
>> + if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
>> + !(vcpu->arch.shregs.msr & MSR_PR)) {
>> + unsigned long req = kvmppc_get_gpr(vcpu, 3);
>> +
>> + /* H_CEDE has to be handled now, not later */
>> + if (req == H_CEDE) {
>> + kvmppc_cede(vcpu);
>> + kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */> + kvmppc_set_gpr(vcpu, 3, 0);
>> + trap = 0;
>> +
>> + /* XICS hcalls must be handled before xive is pulled */
>> + } else if (hcall_is_xics(req)) {
>> + int ret;
>> +
>> + ret = kvmppc_xive_xics_hcall(vcpu, req);
>> + if (ret != H_TOO_HARD) {
>> + kvmppc_set_gpr(vcpu, 3, ret);
>> + trap = 0;
>> + }
>> + }
>> + }
>> kvmppc_xive_pull_vcpu(vcpu);
>> -
>> }
>>
>> vcpu->arch.slb_max = 0;
>> @@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
>> else
>> r = kvmppc_run_vcpu(vcpu);
>>
>> - if (run->exit_reason == KVM_EXIT_PAPR_HCALL &&
>> - !(vcpu->arch.shregs.msr & MSR_PR)) {
>> + if (run->exit_reason == KVM_EXIT_PAPR_HCALL) {
>> + if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
>> + r = RESUME_GUEST;
>> + continue;
>> + }
>> trace_kvm_hcall_enter(vcpu);
>> r = kvmppc_pseries_do_hcall(vcpu);
>> trace_kvm_hcall_exit(vcpu, r);
>> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>> index c11597f815e4..2d0d14ed1d92 100644
>> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
>> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
>> mr r4,r9
>> bge fast_guest_return
>> 2:
>> + /* If we came in through the P9 short path, no real mode hcalls */
>> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
>> + cmpwi r0, 0
>> + bne no_try_real
>> /* See if this is an hcall we can handle in real mode */
>> cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
>> beq hcall_try_real_mode
>> +no_try_real:
>>
>> /* Hypervisor doorbell - exit only if host IPI flag set */
>> cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
>> diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
>> index 741bf1f4387a..dcc07ceaf5ca 100644
>> --- a/arch/powerpc/kvm/book3s_xive.c
>> +++ b/arch/powerpc/kvm/book3s_xive.c
>> @@ -158,6 +158,40 @@ void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu)
>> }
>> EXPORT_SYMBOL_GPL(kvmppc_xive_pull_vcpu);
>>
>> +void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu)
>
> kvmppc_xive_rearm_escalation() may be ? It has more meaning to me.
Sure whatever you prefer.
>
>> +{
>> + void __iomem *esc_vaddr = (void __iomem *)vcpu->arch.xive_esc_vaddr;
>> +
>> + if (!esc_vaddr)
>> + return;
>> +
>> + /* we are using XIVE with single escalation */
>> +
>> + if (vcpu->arch.xive_esc_on) {
>> + /*
>> + * If we still have a pending escalation, abort the cede,
>> + * and we must set PQ to 10 rather than 00 so that we don't
>> + * potentially end up with two entries for the escalation
>> + * interrupt in the XIVE interrupt queue. In that case
>> + * we also don't want to set xive_esc_on to 1 here in
>> + * case we race with xive_esc_irq().
>> + */
>> + vcpu->arch.ceded = 0;
>> + /*
>> + * The escalation interrupts are special as we don't EOI them.
>> + * There is no need to use the load-after-store ordering offset
>> + * to set PQ to 10 as we won't use StoreEOI.
>> + */
>> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_10);
>> + } else {
>> + vcpu->arch.xive_esc_on = true;> + mb();
>> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_00);
>> + }
>> + mb();
>> +}
>> +EXPORT_SYMBOL_GPL(kvmppc_xive_cede_vcpu);
>> +
>> /*
>> * This is a simple trigger for a generic XIVE IRQ. This must
>> * only be called for interrupts that support a trigger page
>> @@ -2106,6 +2140,42 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type)
>> return 0;
>> }
>>
>> +int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
>> +{
>> + struct kvmppc_vcore *vc = vcpu->arch.vcore;
>> +
>> + /*
>> + * This test covers the case in which a vCPU does XICS hcalls without
>> + * QEMU having connected the vCPU to a XICS ICP. The ICP is the KVM
>> + * XICS device on P8 or XICS-on-XIVE on P9. It catches QEMU errors when
>> + * the interrupt mode is negotiated, we don't want the OS to do XICS
>> + * hcalls after having negotiated the XIVE interrupt mode.
>> + */
>
> I think a comment like the following should be enough.
>
> The VM should have configured XICS mode before doing XICS hcalls.
I'll change it.
> No need to resend for that though.
>
> Reviewed-by: Cédric Le Goater <clg@kaod.org>
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 18:03 ` Fabiano Rosas
-1 siblings, 0 replies; 188+ messages in thread
From: Fabiano Rosas @ 2021-03-23 18:03 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Nicholas Piggin <npiggin@gmail.com> writes:
> In the interest of minimising the amount of code that is run in
> "real-mode", don't handle hcalls in real mode in the P9 path.
>
> POWER8 and earlier are much more expensive to exit from HV real mode
> and switch to host mode, because on those processors HV interrupts get
> to the hypervisor with the MMU off, and the other threads in the core
> need to be pulled out of the guest, and SLBs all need to be saved,
> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
> in host mode. Hash guests also require a lot of hcalls to run. The
> XICS interrupt controller requires hcalls to run.
>
> By contrast, POWER9 has independent thread switching, and in radix mode
> the hypervisor is already in a host virtual memory mode when the HV
> interrupt is taken. Radix + xive guests don't need hcalls to handle
> interrupts or manage translations.
>
> So it's much less important to handle hcalls in real mode in P9.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
I tried this again in the L2 with xive=off and it works as expected now.
Tested-by: Fabiano Rosas <farosas@linux.ibm.com>
> ---
> arch/powerpc/include/asm/kvm_ppc.h | 5 ++
> arch/powerpc/kvm/book3s_hv.c | 57 ++++++++++++++++----
> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++
> arch/powerpc/kvm/book3s_xive.c | 70 +++++++++++++++++++++++++
> 4 files changed, 127 insertions(+), 10 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
> index 73b1ca5a6471..db6646c2ade2 100644
> --- a/arch/powerpc/include/asm/kvm_ppc.h
> +++ b/arch/powerpc/include/asm/kvm_ppc.h
> @@ -607,6 +607,7 @@ extern void kvmppc_free_pimap(struct kvm *kvm);
> extern int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall);
> extern void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu);
> extern int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd);
> +extern int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req);
> extern u64 kvmppc_xics_get_icp(struct kvm_vcpu *vcpu);
> extern int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
> extern int kvmppc_xics_connect_vcpu(struct kvm_device *dev,
> @@ -639,6 +640,8 @@ static inline int kvmppc_xics_enabled(struct kvm_vcpu *vcpu)
> static inline void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) { }
> static inline int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd)
> { return 0; }
> +static inline int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
> + { return 0; }
> #endif
>
> #ifdef CONFIG_KVM_XIVE
> @@ -673,6 +676,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
> int level, bool line_status);
> extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
> extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
> +extern void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu);
>
> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> {
> @@ -714,6 +718,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
> int level, bool line_status) { return -ENODEV; }
> static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
> static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
> +static inline void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu) { }
>
> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> { return 0; }
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index fa7614c37e08..17739aaee3d8 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
> }
>
> /*
> - * Handle H_CEDE in the nested virtualization case where we haven't
> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
> + * handlers in book3s_hv_rmhandlers.S.
> + *
> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
> * that the cede logic in kvmppc_run_single_vcpu() works properly.
> */
> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
> {
> vcpu->arch.shregs.msr |= MSR_EE;
> vcpu->arch.ceded = 1;
> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
> /* hcall - punt to userspace */
> int i;
>
> - /* hypercall with MSR_PR has already been handled in rmode,
> - * and never reaches here.
> - */
> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
> + /*
> + * Guest userspace executed sc 1, reflect it back as a
> + * privileged program check interrupt.
> + */
> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
> + r = RESUME_GUEST;
> + break;
> + }
>
> run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
> for (i = 0; i < 9; ++i)
> @@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
> return trap;
> }
>
> +static inline bool hcall_is_xics(unsigned long req)
> +{
> + return (req == H_EOI || req == H_CPPR || req == H_IPI ||
> + req == H_IPOLL || req == H_XIRR || req == H_XIRR_X);
> +}
> +
> /*
> * Virtual-mode guest entry for POWER9 and later when the host and
> * guest are both using the radix MMU. The LPIDR has already been set.
> @@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> /* H_CEDE has to be handled now, not later */
> if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> kvmppc_get_gpr(vcpu, 3) == H_CEDE) {
> - kvmppc_nested_cede(vcpu);
> + kvmppc_cede(vcpu);
> kvmppc_set_gpr(vcpu, 3, 0);
> trap = 0;
> }
> } else {
> kvmppc_xive_push_vcpu(vcpu);
> trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
> + if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> + !(vcpu->arch.shregs.msr & MSR_PR)) {
> + unsigned long req = kvmppc_get_gpr(vcpu, 3);
> +
> + /* H_CEDE has to be handled now, not later */
> + if (req == H_CEDE) {
> + kvmppc_cede(vcpu);
> + kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */
> + kvmppc_set_gpr(vcpu, 3, 0);
> + trap = 0;
> +
> + /* XICS hcalls must be handled before xive is pulled */
> + } else if (hcall_is_xics(req)) {
> + int ret;
> +
> + ret = kvmppc_xive_xics_hcall(vcpu, req);
> + if (ret != H_TOO_HARD) {
> + kvmppc_set_gpr(vcpu, 3, ret);
> + trap = 0;
> + }
> + }
> + }
> kvmppc_xive_pull_vcpu(vcpu);
> -
> }
>
> vcpu->arch.slb_max = 0;
> @@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
> else
> r = kvmppc_run_vcpu(vcpu);
>
> - if (run->exit_reason == KVM_EXIT_PAPR_HCALL &&
> - !(vcpu->arch.shregs.msr & MSR_PR)) {
> + if (run->exit_reason == KVM_EXIT_PAPR_HCALL) {
> + if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
> + r = RESUME_GUEST;
> + continue;
> + }
> trace_kvm_hcall_enter(vcpu);
> r = kvmppc_pseries_do_hcall(vcpu);
> trace_kvm_hcall_exit(vcpu, r);
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index c11597f815e4..2d0d14ed1d92 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
> mr r4,r9
> bge fast_guest_return
> 2:
> + /* If we came in through the P9 short path, no real mode hcalls */
> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
> + cmpwi r0, 0
> + bne no_try_real
> /* See if this is an hcall we can handle in real mode */
> cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
> beq hcall_try_real_mode
> +no_try_real:
>
> /* Hypervisor doorbell - exit only if host IPI flag set */
> cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
> diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
> index 741bf1f4387a..dcc07ceaf5ca 100644
> --- a/arch/powerpc/kvm/book3s_xive.c
> +++ b/arch/powerpc/kvm/book3s_xive.c
> @@ -158,6 +158,40 @@ void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu)
> }
> EXPORT_SYMBOL_GPL(kvmppc_xive_pull_vcpu);
>
> +void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu)
> +{
> + void __iomem *esc_vaddr = (void __iomem *)vcpu->arch.xive_esc_vaddr;
> +
> + if (!esc_vaddr)
> + return;
> +
> + /* we are using XIVE with single escalation */
> +
> + if (vcpu->arch.xive_esc_on) {
> + /*
> + * If we still have a pending escalation, abort the cede,
> + * and we must set PQ to 10 rather than 00 so that we don't
> + * potentially end up with two entries for the escalation
> + * interrupt in the XIVE interrupt queue. In that case
> + * we also don't want to set xive_esc_on to 1 here in
> + * case we race with xive_esc_irq().
> + */
> + vcpu->arch.ceded = 0;
> + /*
> + * The escalation interrupts are special as we don't EOI them.
> + * There is no need to use the load-after-store ordering offset
> + * to set PQ to 10 as we won't use StoreEOI.
> + */
> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_10);
> + } else {
> + vcpu->arch.xive_esc_on = true;
> + mb();
> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_00);
> + }
> + mb();
> +}
> +EXPORT_SYMBOL_GPL(kvmppc_xive_cede_vcpu);
> +
> /*
> * This is a simple trigger for a generic XIVE IRQ. This must
> * only be called for interrupts that support a trigger page
> @@ -2106,6 +2140,42 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type)
> return 0;
> }
>
> +int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
> +{
> + struct kvmppc_vcore *vc = vcpu->arch.vcore;
> +
> + /*
> + * This test covers the case in which a vCPU does XICS hcalls without
> + * QEMU having connected the vCPU to a XICS ICP. The ICP is the KVM
> + * XICS device on P8 or XICS-on-XIVE on P9. It catches QEMU errors when
> + * the interrupt mode is negotiated, we don't want the OS to do XICS
> + * hcalls after having negotiated the XIVE interrupt mode.
> + */
> + if (!kvmppc_xics_enabled(vcpu))
> + return H_TOO_HARD;
> +
> + switch (req) {
> + case H_XIRR:
> + return xive_vm_h_xirr(vcpu);
> + case H_CPPR:
> + return xive_vm_h_cppr(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_EOI:
> + return xive_vm_h_eoi(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_IPI:
> + return xive_vm_h_ipi(vcpu, kvmppc_get_gpr(vcpu, 4),
> + kvmppc_get_gpr(vcpu, 5));
> + case H_IPOLL:
> + return xive_vm_h_ipoll(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_XIRR_X:
> + xive_vm_h_xirr(vcpu);
> + kvmppc_set_gpr(vcpu, 5, get_tb() + vc->tb_offset);
> + return H_SUCCESS;
> + }
> +
> + return H_UNSUPPORTED;
> +}
> +EXPORT_SYMBOL_GPL(kvmppc_xive_xics_hcall);
> +
> int kvmppc_xive_debug_show_queues(struct seq_file *m, struct kvm_vcpu *vcpu)
> {
> struct kvmppc_xive_vcpu *xc = vcpu->arch.xive_vcpu;
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
@ 2021-03-23 18:03 ` Fabiano Rosas
0 siblings, 0 replies; 188+ messages in thread
From: Fabiano Rosas @ 2021-03-23 18:03 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Nicholas Piggin <npiggin@gmail.com> writes:
> In the interest of minimising the amount of code that is run in
> "real-mode", don't handle hcalls in real mode in the P9 path.
>
> POWER8 and earlier are much more expensive to exit from HV real mode
> and switch to host mode, because on those processors HV interrupts get
> to the hypervisor with the MMU off, and the other threads in the core
> need to be pulled out of the guest, and SLBs all need to be saved,
> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
> in host mode. Hash guests also require a lot of hcalls to run. The
> XICS interrupt controller requires hcalls to run.
>
> By contrast, POWER9 has independent thread switching, and in radix mode
> the hypervisor is already in a host virtual memory mode when the HV
> interrupt is taken. Radix + xive guests don't need hcalls to handle
> interrupts or manage translations.
>
> So it's much less important to handle hcalls in real mode in P9.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
I tried this again in the L2 with xive=off and it works as expected now.
Tested-by: Fabiano Rosas <farosas@linux.ibm.com>
> ---
> arch/powerpc/include/asm/kvm_ppc.h | 5 ++
> arch/powerpc/kvm/book3s_hv.c | 57 ++++++++++++++++----
> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 5 ++
> arch/powerpc/kvm/book3s_xive.c | 70 +++++++++++++++++++++++++
> 4 files changed, 127 insertions(+), 10 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
> index 73b1ca5a6471..db6646c2ade2 100644
> --- a/arch/powerpc/include/asm/kvm_ppc.h
> +++ b/arch/powerpc/include/asm/kvm_ppc.h
> @@ -607,6 +607,7 @@ extern void kvmppc_free_pimap(struct kvm *kvm);
> extern int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall);
> extern void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu);
> extern int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd);
> +extern int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req);
> extern u64 kvmppc_xics_get_icp(struct kvm_vcpu *vcpu);
> extern int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval);
> extern int kvmppc_xics_connect_vcpu(struct kvm_device *dev,
> @@ -639,6 +640,8 @@ static inline int kvmppc_xics_enabled(struct kvm_vcpu *vcpu)
> static inline void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) { }
> static inline int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd)
> { return 0; }
> +static inline int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
> + { return 0; }
> #endif
>
> #ifdef CONFIG_KVM_XIVE
> @@ -673,6 +676,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
> int level, bool line_status);
> extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
> extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
> +extern void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu);
>
> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> {
> @@ -714,6 +718,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
> int level, bool line_status) { return -ENODEV; }
> static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
> static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
> +static inline void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu) { }
>
> static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> { return 0; }
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index fa7614c37e08..17739aaee3d8 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
> }
>
> /*
> - * Handle H_CEDE in the nested virtualization case where we haven't
> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
> + * handlers in book3s_hv_rmhandlers.S.
> + *
> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
> * that the cede logic in kvmppc_run_single_vcpu() works properly.
> */
> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
> {
> vcpu->arch.shregs.msr |= MSR_EE;
> vcpu->arch.ceded = 1;
> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
> /* hcall - punt to userspace */
> int i;
>
> - /* hypercall with MSR_PR has already been handled in rmode,
> - * and never reaches here.
> - */
> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
> + /*
> + * Guest userspace executed sc 1, reflect it back as a
> + * privileged program check interrupt.
> + */
> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
> + r = RESUME_GUEST;
> + break;
> + }
>
> run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
> for (i = 0; i < 9; ++i)
> @@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
> return trap;
> }
>
> +static inline bool hcall_is_xics(unsigned long req)
> +{
> + return (req = H_EOI || req = H_CPPR || req = H_IPI ||
> + req = H_IPOLL || req = H_XIRR || req = H_XIRR_X);
> +}
> +
> /*
> * Virtual-mode guest entry for POWER9 and later when the host and
> * guest are both using the radix MMU. The LPIDR has already been set.
> @@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> /* H_CEDE has to be handled now, not later */
> if (trap = BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> kvmppc_get_gpr(vcpu, 3) = H_CEDE) {
> - kvmppc_nested_cede(vcpu);
> + kvmppc_cede(vcpu);
> kvmppc_set_gpr(vcpu, 3, 0);
> trap = 0;
> }
> } else {
> kvmppc_xive_push_vcpu(vcpu);
> trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
> + if (trap = BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> + !(vcpu->arch.shregs.msr & MSR_PR)) {
> + unsigned long req = kvmppc_get_gpr(vcpu, 3);
> +
> + /* H_CEDE has to be handled now, not later */
> + if (req = H_CEDE) {
> + kvmppc_cede(vcpu);
> + kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */
> + kvmppc_set_gpr(vcpu, 3, 0);
> + trap = 0;
> +
> + /* XICS hcalls must be handled before xive is pulled */
> + } else if (hcall_is_xics(req)) {
> + int ret;
> +
> + ret = kvmppc_xive_xics_hcall(vcpu, req);
> + if (ret != H_TOO_HARD) {
> + kvmppc_set_gpr(vcpu, 3, ret);
> + trap = 0;
> + }
> + }
> + }
> kvmppc_xive_pull_vcpu(vcpu);
> -
> }
>
> vcpu->arch.slb_max = 0;
> @@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
> else
> r = kvmppc_run_vcpu(vcpu);
>
> - if (run->exit_reason = KVM_EXIT_PAPR_HCALL &&
> - !(vcpu->arch.shregs.msr & MSR_PR)) {
> + if (run->exit_reason = KVM_EXIT_PAPR_HCALL) {
> + if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
> + r = RESUME_GUEST;
> + continue;
> + }
> trace_kvm_hcall_enter(vcpu);
> r = kvmppc_pseries_do_hcall(vcpu);
> trace_kvm_hcall_exit(vcpu, r);
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index c11597f815e4..2d0d14ed1d92 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
> mr r4,r9
> bge fast_guest_return
> 2:
> + /* If we came in through the P9 short path, no real mode hcalls */
> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
> + cmpwi r0, 0
> + bne no_try_real
> /* See if this is an hcall we can handle in real mode */
> cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
> beq hcall_try_real_mode
> +no_try_real:
>
> /* Hypervisor doorbell - exit only if host IPI flag set */
> cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
> diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
> index 741bf1f4387a..dcc07ceaf5ca 100644
> --- a/arch/powerpc/kvm/book3s_xive.c
> +++ b/arch/powerpc/kvm/book3s_xive.c
> @@ -158,6 +158,40 @@ void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu)
> }
> EXPORT_SYMBOL_GPL(kvmppc_xive_pull_vcpu);
>
> +void kvmppc_xive_cede_vcpu(struct kvm_vcpu *vcpu)
> +{
> + void __iomem *esc_vaddr = (void __iomem *)vcpu->arch.xive_esc_vaddr;
> +
> + if (!esc_vaddr)
> + return;
> +
> + /* we are using XIVE with single escalation */
> +
> + if (vcpu->arch.xive_esc_on) {
> + /*
> + * If we still have a pending escalation, abort the cede,
> + * and we must set PQ to 10 rather than 00 so that we don't
> + * potentially end up with two entries for the escalation
> + * interrupt in the XIVE interrupt queue. In that case
> + * we also don't want to set xive_esc_on to 1 here in
> + * case we race with xive_esc_irq().
> + */
> + vcpu->arch.ceded = 0;
> + /*
> + * The escalation interrupts are special as we don't EOI them.
> + * There is no need to use the load-after-store ordering offset
> + * to set PQ to 10 as we won't use StoreEOI.
> + */
> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_10);
> + } else {
> + vcpu->arch.xive_esc_on = true;
> + mb();
> + __raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_00);
> + }
> + mb();
> +}
> +EXPORT_SYMBOL_GPL(kvmppc_xive_cede_vcpu);
> +
> /*
> * This is a simple trigger for a generic XIVE IRQ. This must
> * only be called for interrupts that support a trigger page
> @@ -2106,6 +2140,42 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type)
> return 0;
> }
>
> +int kvmppc_xive_xics_hcall(struct kvm_vcpu *vcpu, u32 req)
> +{
> + struct kvmppc_vcore *vc = vcpu->arch.vcore;
> +
> + /*
> + * This test covers the case in which a vCPU does XICS hcalls without
> + * QEMU having connected the vCPU to a XICS ICP. The ICP is the KVM
> + * XICS device on P8 or XICS-on-XIVE on P9. It catches QEMU errors when
> + * the interrupt mode is negotiated, we don't want the OS to do XICS
> + * hcalls after having negotiated the XIVE interrupt mode.
> + */
> + if (!kvmppc_xics_enabled(vcpu))
> + return H_TOO_HARD;
> +
> + switch (req) {
> + case H_XIRR:
> + return xive_vm_h_xirr(vcpu);
> + case H_CPPR:
> + return xive_vm_h_cppr(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_EOI:
> + return xive_vm_h_eoi(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_IPI:
> + return xive_vm_h_ipi(vcpu, kvmppc_get_gpr(vcpu, 4),
> + kvmppc_get_gpr(vcpu, 5));
> + case H_IPOLL:
> + return xive_vm_h_ipoll(vcpu, kvmppc_get_gpr(vcpu, 4));
> + case H_XIRR_X:
> + xive_vm_h_xirr(vcpu);
> + kvmppc_set_gpr(vcpu, 5, get_tb() + vc->tb_offset);
> + return H_SUCCESS;
> + }
> +
> + return H_UNSUPPORTED;
> +}
> +EXPORT_SYMBOL_GPL(kvmppc_xive_xics_hcall);
> +
> int kvmppc_xive_debug_show_queues(struct seq_file *m, struct kvm_vcpu *vcpu)
> {
> struct kvmppc_xive_vcpu *xc = vcpu->arch.xive_vcpu;
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
2021-03-23 18:03 ` Fabiano Rosas
@ 2021-03-24 1:22 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-24 1:22 UTC (permalink / raw)
To: Fabiano Rosas, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Fabiano Rosas's message of March 24, 2021 4:03 am:
> Nicholas Piggin <npiggin@gmail.com> writes:
>
>> In the interest of minimising the amount of code that is run in
>> "real-mode", don't handle hcalls in real mode in the P9 path.
>>
>> POWER8 and earlier are much more expensive to exit from HV real mode
>> and switch to host mode, because on those processors HV interrupts get
>> to the hypervisor with the MMU off, and the other threads in the core
>> need to be pulled out of the guest, and SLBs all need to be saved,
>> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
>> in host mode. Hash guests also require a lot of hcalls to run. The
>> XICS interrupt controller requires hcalls to run.
>>
>> By contrast, POWER9 has independent thread switching, and in radix mode
>> the hypervisor is already in a host virtual memory mode when the HV
>> interrupt is taken. Radix + xive guests don't need hcalls to handle
>> interrupts or manage translations.
>>
>> So it's much less important to handle hcalls in real mode in P9.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>
> I tried this again in the L2 with xive=off and it works as expected now.
>
> Tested-by: Fabiano Rosas <farosas@linux.ibm.com>
Oh good, thanks for spotting the problem and re testing.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
@ 2021-03-24 1:22 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-24 1:22 UTC (permalink / raw)
To: Fabiano Rosas, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Fabiano Rosas's message of March 24, 2021 4:03 am:
> Nicholas Piggin <npiggin@gmail.com> writes:
>
>> In the interest of minimising the amount of code that is run in
>> "real-mode", don't handle hcalls in real mode in the P9 path.
>>
>> POWER8 and earlier are much more expensive to exit from HV real mode
>> and switch to host mode, because on those processors HV interrupts get
>> to the hypervisor with the MMU off, and the other threads in the core
>> need to be pulled out of the guest, and SLBs all need to be saved,
>> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
>> in host mode. Hash guests also require a lot of hcalls to run. The
>> XICS interrupt controller requires hcalls to run.
>>
>> By contrast, POWER9 has independent thread switching, and in radix mode
>> the hypervisor is already in a host virtual memory mode when the HV
>> interrupt is taken. Radix + xive guests don't need hcalls to handle
>> interrupts or manage translations.
>>
>> So it's much less important to handle hcalls in real mode in P9.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>
> I tried this again in the L2 with xive=off and it works as expected now.
>
> Tested-by: Fabiano Rosas <farosas@linux.ibm.com>
Oh good, thanks for spotting the problem and re testing.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 22:57 ` Fabiano Rosas
-1 siblings, 0 replies; 188+ messages in thread
From: Fabiano Rosas @ 2021-03-23 22:57 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Nicholas Piggin <npiggin@gmail.com> writes:
> In the interest of minimising the amount of code that is run in
> "real-mode", don't handle hcalls in real mode in the P9 path.
>
> POWER8 and earlier are much more expensive to exit from HV real mode
> and switch to host mode, because on those processors HV interrupts get
> to the hypervisor with the MMU off, and the other threads in the core
> need to be pulled out of the guest, and SLBs all need to be saved,
> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
> in host mode. Hash guests also require a lot of hcalls to run. The
> XICS interrupt controller requires hcalls to run.
>
> By contrast, POWER9 has independent thread switching, and in radix mode
> the hypervisor is already in a host virtual memory mode when the HV
> interrupt is taken. Radix + xive guests don't need hcalls to handle
> interrupts or manage translations.
>
> So it's much less important to handle hcalls in real mode in P9.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
<snip>
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index fa7614c37e08..17739aaee3d8 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
> }
>
> /*
> - * Handle H_CEDE in the nested virtualization case where we haven't
> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
> + * handlers in book3s_hv_rmhandlers.S.
> + *
> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
> * that the cede logic in kvmppc_run_single_vcpu() works properly.
> */
> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
> {
> vcpu->arch.shregs.msr |= MSR_EE;
> vcpu->arch.ceded = 1;
> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
> /* hcall - punt to userspace */
> int i;
>
> - /* hypercall with MSR_PR has already been handled in rmode,
> - * and never reaches here.
> - */
> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
> + /*
> + * Guest userspace executed sc 1, reflect it back as a
> + * privileged program check interrupt.
> + */
> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
> + r = RESUME_GUEST;
> + break;
> + }
This patch bypasses sc_1_fast_return so it breaks KVM-PR. L1 loops with
the following output:
[ 9.503929][ T3443] Couldn't emulate instruction 0x4e800020 (op 19 xop 16)
[ 9.503990][ T3443] kvmppc_exit_pr_progint: emulation at 48f4 failed (4e800020)
[ 9.504080][ T3443] Couldn't emulate instruction 0x4e800020 (op 19 xop 16)
[ 9.504170][ T3443] kvmppc_exit_pr_progint: emulation at 48f4 failed (4e800020)
0x4e800020 is a blr after a sc 1 in SLOF.
For KVM-PR we need to inject a 0xc00 at some point, either here or
before branching to no_try_real in book3s_hv_rmhandlers.S.
>
> run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
> for (i = 0; i < 9; ++i)
> @@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
> return trap;
> }
>
> +static inline bool hcall_is_xics(unsigned long req)
> +{
> + return (req == H_EOI || req == H_CPPR || req == H_IPI ||
> + req == H_IPOLL || req == H_XIRR || req == H_XIRR_X);
> +}
> +
> /*
> * Virtual-mode guest entry for POWER9 and later when the host and
> * guest are both using the radix MMU. The LPIDR has already been set.
> @@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> /* H_CEDE has to be handled now, not later */
> if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> kvmppc_get_gpr(vcpu, 3) == H_CEDE) {
> - kvmppc_nested_cede(vcpu);
> + kvmppc_cede(vcpu);
> kvmppc_set_gpr(vcpu, 3, 0);
> trap = 0;
> }
> } else {
> kvmppc_xive_push_vcpu(vcpu);
> trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
> + if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> + !(vcpu->arch.shregs.msr & MSR_PR)) {
> + unsigned long req = kvmppc_get_gpr(vcpu, 3);
> +
> + /* H_CEDE has to be handled now, not later */
> + if (req == H_CEDE) {
> + kvmppc_cede(vcpu);
> + kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */
> + kvmppc_set_gpr(vcpu, 3, 0);
> + trap = 0;
> +
> + /* XICS hcalls must be handled before xive is pulled */
> + } else if (hcall_is_xics(req)) {
> + int ret;
> +
> + ret = kvmppc_xive_xics_hcall(vcpu, req);
> + if (ret != H_TOO_HARD) {
> + kvmppc_set_gpr(vcpu, 3, ret);
> + trap = 0;
> + }
> + }
> + }
> kvmppc_xive_pull_vcpu(vcpu);
> -
> }
>
> vcpu->arch.slb_max = 0;
> @@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
> else
> r = kvmppc_run_vcpu(vcpu);
>
> - if (run->exit_reason == KVM_EXIT_PAPR_HCALL &&
> - !(vcpu->arch.shregs.msr & MSR_PR)) {
> + if (run->exit_reason == KVM_EXIT_PAPR_HCALL) {
> + if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
> + r = RESUME_GUEST;
> + continue;
> + }
Note that this hunk might need to be dropped.
> trace_kvm_hcall_enter(vcpu);
> r = kvmppc_pseries_do_hcall(vcpu);
> trace_kvm_hcall_exit(vcpu, r);
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index c11597f815e4..2d0d14ed1d92 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
> mr r4,r9
> bge fast_guest_return
> 2:
> + /* If we came in through the P9 short path, no real mode hcalls */
> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
> + cmpwi r0, 0
> + bne no_try_real
> /* See if this is an hcall we can handle in real mode */
> cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
> beq hcall_try_real_mode
> +no_try_real:
>
> /* Hypervisor doorbell - exit only if host IPI flag set */
> cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
@ 2021-03-23 22:57 ` Fabiano Rosas
0 siblings, 0 replies; 188+ messages in thread
From: Fabiano Rosas @ 2021-03-23 22:57 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Nicholas Piggin <npiggin@gmail.com> writes:
> In the interest of minimising the amount of code that is run in
> "real-mode", don't handle hcalls in real mode in the P9 path.
>
> POWER8 and earlier are much more expensive to exit from HV real mode
> and switch to host mode, because on those processors HV interrupts get
> to the hypervisor with the MMU off, and the other threads in the core
> need to be pulled out of the guest, and SLBs all need to be saved,
> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
> in host mode. Hash guests also require a lot of hcalls to run. The
> XICS interrupt controller requires hcalls to run.
>
> By contrast, POWER9 has independent thread switching, and in radix mode
> the hypervisor is already in a host virtual memory mode when the HV
> interrupt is taken. Radix + xive guests don't need hcalls to handle
> interrupts or manage translations.
>
> So it's much less important to handle hcalls in real mode in P9.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
<snip>
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index fa7614c37e08..17739aaee3d8 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
> }
>
> /*
> - * Handle H_CEDE in the nested virtualization case where we haven't
> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
> + * handlers in book3s_hv_rmhandlers.S.
> + *
> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
> * that the cede logic in kvmppc_run_single_vcpu() works properly.
> */
> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
> {
> vcpu->arch.shregs.msr |= MSR_EE;
> vcpu->arch.ceded = 1;
> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
> /* hcall - punt to userspace */
> int i;
>
> - /* hypercall with MSR_PR has already been handled in rmode,
> - * and never reaches here.
> - */
> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
> + /*
> + * Guest userspace executed sc 1, reflect it back as a
> + * privileged program check interrupt.
> + */
> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
> + r = RESUME_GUEST;
> + break;
> + }
This patch bypasses sc_1_fast_return so it breaks KVM-PR. L1 loops with
the following output:
[ 9.503929][ T3443] Couldn't emulate instruction 0x4e800020 (op 19 xop 16)
[ 9.503990][ T3443] kvmppc_exit_pr_progint: emulation at 48f4 failed (4e800020)
[ 9.504080][ T3443] Couldn't emulate instruction 0x4e800020 (op 19 xop 16)
[ 9.504170][ T3443] kvmppc_exit_pr_progint: emulation at 48f4 failed (4e800020)
0x4e800020 is a blr after a sc 1 in SLOF.
For KVM-PR we need to inject a 0xc00 at some point, either here or
before branching to no_try_real in book3s_hv_rmhandlers.S.
>
> run->papr_hcall.nr = kvmppc_get_gpr(vcpu, 3);
> for (i = 0; i < 9; ++i)
> @@ -3663,6 +3670,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
> return trap;
> }
>
> +static inline bool hcall_is_xics(unsigned long req)
> +{
> + return (req = H_EOI || req = H_CPPR || req = H_IPI ||
> + req = H_IPOLL || req = H_XIRR || req = H_XIRR_X);
> +}
> +
> /*
> * Virtual-mode guest entry for POWER9 and later when the host and
> * guest are both using the radix MMU. The LPIDR has already been set.
> @@ -3774,15 +3787,36 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> /* H_CEDE has to be handled now, not later */
> if (trap = BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> kvmppc_get_gpr(vcpu, 3) = H_CEDE) {
> - kvmppc_nested_cede(vcpu);
> + kvmppc_cede(vcpu);
> kvmppc_set_gpr(vcpu, 3, 0);
> trap = 0;
> }
> } else {
> kvmppc_xive_push_vcpu(vcpu);
> trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
> + if (trap = BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
> + !(vcpu->arch.shregs.msr & MSR_PR)) {
> + unsigned long req = kvmppc_get_gpr(vcpu, 3);
> +
> + /* H_CEDE has to be handled now, not later */
> + if (req = H_CEDE) {
> + kvmppc_cede(vcpu);
> + kvmppc_xive_cede_vcpu(vcpu); /* may un-cede */
> + kvmppc_set_gpr(vcpu, 3, 0);
> + trap = 0;
> +
> + /* XICS hcalls must be handled before xive is pulled */
> + } else if (hcall_is_xics(req)) {
> + int ret;
> +
> + ret = kvmppc_xive_xics_hcall(vcpu, req);
> + if (ret != H_TOO_HARD) {
> + kvmppc_set_gpr(vcpu, 3, ret);
> + trap = 0;
> + }
> + }
> + }
> kvmppc_xive_pull_vcpu(vcpu);
> -
> }
>
> vcpu->arch.slb_max = 0;
> @@ -4442,8 +4476,11 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
> else
> r = kvmppc_run_vcpu(vcpu);
>
> - if (run->exit_reason = KVM_EXIT_PAPR_HCALL &&
> - !(vcpu->arch.shregs.msr & MSR_PR)) {
> + if (run->exit_reason = KVM_EXIT_PAPR_HCALL) {
> + if (WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_PR)) {
> + r = RESUME_GUEST;
> + continue;
> + }
Note that this hunk might need to be dropped.
> trace_kvm_hcall_enter(vcpu);
> r = kvmppc_pseries_do_hcall(vcpu);
> trace_kvm_hcall_exit(vcpu, r);
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index c11597f815e4..2d0d14ed1d92 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -1397,9 +1397,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
> mr r4,r9
> bge fast_guest_return
> 2:
> + /* If we came in through the P9 short path, no real mode hcalls */
> + lwz r0, STACK_SLOT_SHORT_PATH(r1)
> + cmpwi r0, 0
> + bne no_try_real
> /* See if this is an hcall we can handle in real mode */
> cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
> beq hcall_try_real_mode
> +no_try_real:
>
> /* Hypervisor doorbell - exit only if host IPI flag set */
> cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
2021-03-23 22:57 ` Fabiano Rosas
@ 2021-03-24 1:27 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-24 1:27 UTC (permalink / raw)
To: Fabiano Rosas, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Fabiano Rosas's message of March 24, 2021 8:57 am:
> Nicholas Piggin <npiggin@gmail.com> writes:
>
>> In the interest of minimising the amount of code that is run in
>> "real-mode", don't handle hcalls in real mode in the P9 path.
>>
>> POWER8 and earlier are much more expensive to exit from HV real mode
>> and switch to host mode, because on those processors HV interrupts get
>> to the hypervisor with the MMU off, and the other threads in the core
>> need to be pulled out of the guest, and SLBs all need to be saved,
>> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
>> in host mode. Hash guests also require a lot of hcalls to run. The
>> XICS interrupt controller requires hcalls to run.
>>
>> By contrast, POWER9 has independent thread switching, and in radix mode
>> the hypervisor is already in a host virtual memory mode when the HV
>> interrupt is taken. Radix + xive guests don't need hcalls to handle
>> interrupts or manage translations.
>>
>> So it's much less important to handle hcalls in real mode in P9.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>> ---
>
> <snip>
>
>> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
>> index fa7614c37e08..17739aaee3d8 100644
>> --- a/arch/powerpc/kvm/book3s_hv.c
>> +++ b/arch/powerpc/kvm/book3s_hv.c
>> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
>> }
>>
>> /*
>> - * Handle H_CEDE in the nested virtualization case where we haven't
>> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
>> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
>> + * handlers in book3s_hv_rmhandlers.S.
>> + *
>> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
>> * that the cede logic in kvmppc_run_single_vcpu() works properly.
>> */
>> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
>> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
>> {
>> vcpu->arch.shregs.msr |= MSR_EE;
>> vcpu->arch.ceded = 1;
>> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
>> /* hcall - punt to userspace */
>> int i;
>>
>> - /* hypercall with MSR_PR has already been handled in rmode,
>> - * and never reaches here.
>> - */
>> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
>> + /*
>> + * Guest userspace executed sc 1, reflect it back as a
>> + * privileged program check interrupt.
>> + */
>> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
>> + r = RESUME_GUEST;
>> + break;
>> + }
>
> This patch bypasses sc_1_fast_return so it breaks KVM-PR. L1 loops with
> the following output:
>
> [ 9.503929][ T3443] Couldn't emulate instruction 0x4e800020 (op 19 xop 16)
> [ 9.503990][ T3443] kvmppc_exit_pr_progint: emulation at 48f4 failed (4e800020)
> [ 9.504080][ T3443] Couldn't emulate instruction 0x4e800020 (op 19 xop 16)
> [ 9.504170][ T3443] kvmppc_exit_pr_progint: emulation at 48f4 failed (4e800020)
>
> 0x4e800020 is a blr after a sc 1 in SLOF.
>
> For KVM-PR we need to inject a 0xc00 at some point, either here or
> before branching to no_try_real in book3s_hv_rmhandlers.S.
Ah, I didn't know about that PR KVM (I suppose I should test it but I
haven't been able to get it running in the past).
Should be able to deal with that. This patch probably shouldn't change
the syscall behaviour like this anyway.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 22/46] KVM: PPC: Book3S HV P9: Stop handling hcalls in real-mode in the P9 path
@ 2021-03-24 1:27 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-24 1:27 UTC (permalink / raw)
To: Fabiano Rosas, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Fabiano Rosas's message of March 24, 2021 8:57 am:
> Nicholas Piggin <npiggin@gmail.com> writes:
>
>> In the interest of minimising the amount of code that is run in
>> "real-mode", don't handle hcalls in real mode in the P9 path.
>>
>> POWER8 and earlier are much more expensive to exit from HV real mode
>> and switch to host mode, because on those processors HV interrupts get
>> to the hypervisor with the MMU off, and the other threads in the core
>> need to be pulled out of the guest, and SLBs all need to be saved,
>> ERATs invalidated, and host SLB reloaded before the MMU is re-enabled
>> in host mode. Hash guests also require a lot of hcalls to run. The
>> XICS interrupt controller requires hcalls to run.
>>
>> By contrast, POWER9 has independent thread switching, and in radix mode
>> the hypervisor is already in a host virtual memory mode when the HV
>> interrupt is taken. Radix + xive guests don't need hcalls to handle
>> interrupts or manage translations.
>>
>> So it's much less important to handle hcalls in real mode in P9.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>> ---
>
> <snip>
>
>> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
>> index fa7614c37e08..17739aaee3d8 100644
>> --- a/arch/powerpc/kvm/book3s_hv.c
>> +++ b/arch/powerpc/kvm/book3s_hv.c
>> @@ -1142,12 +1142,13 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
>> }
>>
>> /*
>> - * Handle H_CEDE in the nested virtualization case where we haven't
>> - * called the real-mode hcall handlers in book3s_hv_rmhandlers.S.
>> + * Handle H_CEDE in the P9 path where we don't call the real-mode hcall
>> + * handlers in book3s_hv_rmhandlers.S.
>> + *
>> * This has to be done early, not in kvmppc_pseries_do_hcall(), so
>> * that the cede logic in kvmppc_run_single_vcpu() works properly.
>> */
>> -static void kvmppc_nested_cede(struct kvm_vcpu *vcpu)
>> +static void kvmppc_cede(struct kvm_vcpu *vcpu)
>> {
>> vcpu->arch.shregs.msr |= MSR_EE;
>> vcpu->arch.ceded = 1;
>> @@ -1403,9 +1404,15 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
>> /* hcall - punt to userspace */
>> int i;
>>
>> - /* hypercall with MSR_PR has already been handled in rmode,
>> - * and never reaches here.
>> - */
>> + if (unlikely(vcpu->arch.shregs.msr & MSR_PR)) {
>> + /*
>> + * Guest userspace executed sc 1, reflect it back as a
>> + * privileged program check interrupt.
>> + */
>> + kvmppc_core_queue_program(vcpu, SRR1_PROGPRIV);
>> + r = RESUME_GUEST;
>> + break;
>> + }
>
> This patch bypasses sc_1_fast_return so it breaks KVM-PR. L1 loops with
> the following output:
>
> [ 9.503929][ T3443] Couldn't emulate instruction 0x4e800020 (op 19 xop 16)
> [ 9.503990][ T3443] kvmppc_exit_pr_progint: emulation at 48f4 failed (4e800020)
> [ 9.504080][ T3443] Couldn't emulate instruction 0x4e800020 (op 19 xop 16)
> [ 9.504170][ T3443] kvmppc_exit_pr_progint: emulation at 48f4 failed (4e800020)
>
> 0x4e800020 is a blr after a sc 1 in SLOF.
>
> For KVM-PR we need to inject a 0xc00 at some point, either here or
> before branching to no_try_real in book3s_hv_rmhandlers.S.
Ah, I didn't know about that PR KVM (I suppose I should test it but I
haven't been able to get it running in the past).
Should be able to deal with that. This patch probably shouldn't change
the syscall behaviour like this anyway.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 23/46] KVM: PPC: Book3S HV P9: Move setting HDEC after switching to guest LPCR
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc
Cc: Alexey Kardashevskiy, linuxppc-dev, Nicholas Piggin, Fabiano Rosas
LPCR[HDICE]=0 suppresses hypervisor decrementer exceptions on some
processors, so it must be enabled before HDEC is set.
Rather than set it in the host LPCR then setting HDEC, move the HDEC
update to after the guest MMU context (including LPCR) is loaded.
There shouldn't be much concern with delaying HDEC by some 10s or 100s
of nanoseconds by setting it a bit later.
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 19 +++++++------------
1 file changed, 7 insertions(+), 12 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 17739aaee3d8..8215430e6d5e 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3540,20 +3540,9 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
host_dawrx1 = mfspr(SPRN_DAWRX1);
}
- /*
- * P8 and P9 suppress the HDEC exception when LPCR[HDICE] = 0,
- * so set HDICE before writing HDEC.
- */
- mtspr(SPRN_LPCR, kvm->arch.host_lpcr | LPCR_HDICE);
- isync();
-
hdec = time_limit - mftb();
- if (hdec < 0) {
- mtspr(SPRN_LPCR, kvm->arch.host_lpcr);
- isync();
+ if (hdec < 0)
return BOOK3S_INTERRUPT_HV_DECREMENTER;
- }
- mtspr(SPRN_HDEC, hdec);
if (vc->tb_offset) {
u64 new_tb = mftb() + vc->tb_offset;
@@ -3599,6 +3588,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
+ /*
+ * P9 suppresses the HDEC exception when LPCR[HDICE] = 0,
+ * so set guest LPCR (with HDICE) before writing HDEC.
+ */
+ mtspr(SPRN_HDEC, hdec);
+
mtspr(SPRN_SRR0, vcpu->arch.shregs.srr0);
mtspr(SPRN_SRR1, vcpu->arch.shregs.srr1);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 23/46] KVM: PPC: Book3S HV P9: Move setting HDEC after switching to guest LPCR
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc
Cc: Alexey Kardashevskiy, linuxppc-dev, Nicholas Piggin, Fabiano Rosas
LPCR[HDICE]=0 suppresses hypervisor decrementer exceptions on some
processors, so it must be enabled before HDEC is set.
Rather than set it in the host LPCR then setting HDEC, move the HDEC
update to after the guest MMU context (including LPCR) is loaded.
There shouldn't be much concern with delaying HDEC by some 10s or 100s
of nanoseconds by setting it a bit later.
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 19 +++++++------------
1 file changed, 7 insertions(+), 12 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 17739aaee3d8..8215430e6d5e 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3540,20 +3540,9 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
host_dawrx1 = mfspr(SPRN_DAWRX1);
}
- /*
- * P8 and P9 suppress the HDEC exception when LPCR[HDICE] = 0,
- * so set HDICE before writing HDEC.
- */
- mtspr(SPRN_LPCR, kvm->arch.host_lpcr | LPCR_HDICE);
- isync();
-
hdec = time_limit - mftb();
- if (hdec < 0) {
- mtspr(SPRN_LPCR, kvm->arch.host_lpcr);
- isync();
+ if (hdec < 0)
return BOOK3S_INTERRUPT_HV_DECREMENTER;
- }
- mtspr(SPRN_HDEC, hdec);
if (vc->tb_offset) {
u64 new_tb = mftb() + vc->tb_offset;
@@ -3599,6 +3588,12 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
+ /*
+ * P9 suppresses the HDEC exception when LPCR[HDICE] = 0,
+ * so set guest LPCR (with HDICE) before writing HDEC.
+ */
+ mtspr(SPRN_HDEC, hdec);
+
mtspr(SPRN_SRR0, vcpu->arch.shregs.srr0);
mtspr(SPRN_SRR1, vcpu->arch.shregs.srr1);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 24/46] KVM: PPC: Book3S HV P9: Use large decrementer for HDEC
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Alexey Kardashevskiy, linuxppc-dev, Nicholas Piggin
On processors that don't suppress the HDEC exceptions when LPCR[HDICE]=0,
this could help reduce needless guest exits due to leftover exceptions on
entering the guest.
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/time.h | 2 ++
arch/powerpc/kvm/book3s_hv.c | 3 ++-
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/include/asm/time.h b/arch/powerpc/include/asm/time.h
index 8dd3cdb25338..68d94711811e 100644
--- a/arch/powerpc/include/asm/time.h
+++ b/arch/powerpc/include/asm/time.h
@@ -18,6 +18,8 @@
#include <asm/vdso/timebase.h>
/* time.c */
+extern u64 decrementer_max;
+
extern unsigned long tb_ticks_per_jiffy;
extern unsigned long tb_ticks_per_usec;
extern unsigned long tb_ticks_per_sec;
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 8215430e6d5e..bb30c5ab53d1 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3658,7 +3658,8 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
vc->tb_offset_applied = 0;
}
- mtspr(SPRN_HDEC, 0x7fffffff);
+ /* HDEC must be at least as large as DEC, so decrementer_max fits */
+ mtspr(SPRN_HDEC, decrementer_max);
switch_mmu_to_host_radix(kvm, host_pidr);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 24/46] KVM: PPC: Book3S HV P9: Use large decrementer for HDEC
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Alexey Kardashevskiy, linuxppc-dev, Nicholas Piggin
On processors that don't suppress the HDEC exceptions when LPCR[HDICE]=0,
this could help reduce needless guest exits due to leftover exceptions on
entering the guest.
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/time.h | 2 ++
arch/powerpc/kvm/book3s_hv.c | 3 ++-
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/include/asm/time.h b/arch/powerpc/include/asm/time.h
index 8dd3cdb25338..68d94711811e 100644
--- a/arch/powerpc/include/asm/time.h
+++ b/arch/powerpc/include/asm/time.h
@@ -18,6 +18,8 @@
#include <asm/vdso/timebase.h>
/* time.c */
+extern u64 decrementer_max;
+
extern unsigned long tb_ticks_per_jiffy;
extern unsigned long tb_ticks_per_usec;
extern unsigned long tb_ticks_per_sec;
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 8215430e6d5e..bb30c5ab53d1 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3658,7 +3658,8 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
vc->tb_offset_applied = 0;
}
- mtspr(SPRN_HDEC, 0x7fffffff);
+ /* HDEC must be at least as large as DEC, so decrementer_max fits */
+ mtspr(SPRN_HDEC, decrementer_max);
switch_mmu_to_host_radix(kvm, host_pidr);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* Re: [PATCH v4 24/46] KVM: PPC: Book3S HV P9: Use large decrementer for HDEC
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-26 2:05 ` Alexey Kardashevskiy
-1 siblings, 0 replies; 188+ messages in thread
From: Alexey Kardashevskiy @ 2021-03-26 2:05 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 23/03/2021 12:02, Nicholas Piggin wrote:
> On processors that don't suppress the HDEC exceptions when LPCR[HDICE]=0,
> this could help reduce needless guest exits due to leftover exceptions on
> entering the guest.
>
> Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
ERROR: modpost: "decrementer_max" [arch/powerpc/kvm/kvm-hv.ko] undefined!
need this:
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -89,6 +89,7 @@ static struct clocksource clocksource_timebase = {
#define DECREMENTER_DEFAULT_MAX 0x7FFFFFFF
u64 decrementer_max = DECREMENTER_DEFAULT_MAX;
+EXPORT_SYMBOL_GPL(decrementer_max);
> ---
> arch/powerpc/include/asm/time.h | 2 ++
> arch/powerpc/kvm/book3s_hv.c | 3 ++-
> 2 files changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/arch/powerpc/include/asm/time.h b/arch/powerpc/include/asm/time.h
> index 8dd3cdb25338..68d94711811e 100644
> --- a/arch/powerpc/include/asm/time.h
> +++ b/arch/powerpc/include/asm/time.h
> @@ -18,6 +18,8 @@
> #include <asm/vdso/timebase.h>
>
> /* time.c */
> +extern u64 decrementer_max;
> +
> extern unsigned long tb_ticks_per_jiffy;
> extern unsigned long tb_ticks_per_usec;
> extern unsigned long tb_ticks_per_sec;
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index 8215430e6d5e..bb30c5ab53d1 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -3658,7 +3658,8 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
> vc->tb_offset_applied = 0;
> }
>
> - mtspr(SPRN_HDEC, 0x7fffffff);
> + /* HDEC must be at least as large as DEC, so decrementer_max fits */
> + mtspr(SPRN_HDEC, decrementer_max);
>
> switch_mmu_to_host_radix(kvm, host_pidr);
>
>
--
Alexey
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 24/46] KVM: PPC: Book3S HV P9: Use large decrementer for HDEC
@ 2021-03-26 2:05 ` Alexey Kardashevskiy
0 siblings, 0 replies; 188+ messages in thread
From: Alexey Kardashevskiy @ 2021-03-26 2:05 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 23/03/2021 12:02, Nicholas Piggin wrote:
> On processors that don't suppress the HDEC exceptions when LPCR[HDICE]=0,
> this could help reduce needless guest exits due to leftover exceptions on
> entering the guest.
>
> Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
ERROR: modpost: "decrementer_max" [arch/powerpc/kvm/kvm-hv.ko] undefined!
need this:
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -89,6 +89,7 @@ static struct clocksource clocksource_timebase = {
#define DECREMENTER_DEFAULT_MAX 0x7FFFFFFF
u64 decrementer_max = DECREMENTER_DEFAULT_MAX;
+EXPORT_SYMBOL_GPL(decrementer_max);
> ---
> arch/powerpc/include/asm/time.h | 2 ++
> arch/powerpc/kvm/book3s_hv.c | 3 ++-
> 2 files changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/arch/powerpc/include/asm/time.h b/arch/powerpc/include/asm/time.h
> index 8dd3cdb25338..68d94711811e 100644
> --- a/arch/powerpc/include/asm/time.h
> +++ b/arch/powerpc/include/asm/time.h
> @@ -18,6 +18,8 @@
> #include <asm/vdso/timebase.h>
>
> /* time.c */
> +extern u64 decrementer_max;
> +
> extern unsigned long tb_ticks_per_jiffy;
> extern unsigned long tb_ticks_per_usec;
> extern unsigned long tb_ticks_per_sec;
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index 8215430e6d5e..bb30c5ab53d1 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -3658,7 +3658,8 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
> vc->tb_offset_applied = 0;
> }
>
> - mtspr(SPRN_HDEC, 0x7fffffff);
> + /* HDEC must be at least as large as DEC, so decrementer_max fits */
> + mtspr(SPRN_HDEC, decrementer_max);
>
> switch_mmu_to_host_radix(kvm, host_pidr);
>
>
--
Alexey
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 24/46] KVM: PPC: Book3S HV P9: Use large decrementer for HDEC
2021-03-26 2:05 ` Alexey Kardashevskiy
@ 2021-04-01 10:14 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 10:14 UTC (permalink / raw)
To: Alexey Kardashevskiy, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Alexey Kardashevskiy's message of March 26, 2021 12:05 pm:
>
>
> On 23/03/2021 12:02, Nicholas Piggin wrote:
>> On processors that don't suppress the HDEC exceptions when LPCR[HDICE]=0,
>> this could help reduce needless guest exits due to leftover exceptions on
>> entering the guest.
>>
>> Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>
>
> ERROR: modpost: "decrementer_max" [arch/powerpc/kvm/kvm-hv.ko] undefined!
>
>
> need this:
>
> --- a/arch/powerpc/kernel/time.c
> +++ b/arch/powerpc/kernel/time.c
> @@ -89,6 +89,7 @@ static struct clocksource clocksource_timebase = {
>
> #define DECREMENTER_DEFAULT_MAX 0x7FFFFFFF
> u64 decrementer_max = DECREMENTER_DEFAULT_MAX;
> +EXPORT_SYMBOL_GPL(decrementer_max);
Added.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 24/46] KVM: PPC: Book3S HV P9: Use large decrementer for HDEC
@ 2021-04-01 10:14 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 10:14 UTC (permalink / raw)
To: Alexey Kardashevskiy, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Alexey Kardashevskiy's message of March 26, 2021 12:05 pm:
>
>
> On 23/03/2021 12:02, Nicholas Piggin wrote:
>> On processors that don't suppress the HDEC exceptions when LPCR[HDICE]=0,
>> this could help reduce needless guest exits due to leftover exceptions on
>> entering the guest.
>>
>> Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>
>
> ERROR: modpost: "decrementer_max" [arch/powerpc/kvm/kvm-hv.ko] undefined!
>
>
> need this:
>
> --- a/arch/powerpc/kernel/time.c
> +++ b/arch/powerpc/kernel/time.c
> @@ -89,6 +89,7 @@ static struct clocksource clocksource_timebase = {
>
> #define DECREMENTER_DEFAULT_MAX 0x7FFFFFFF
> u64 decrementer_max = DECREMENTER_DEFAULT_MAX;
> +EXPORT_SYMBOL_GPL(decrementer_max);
Added.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 25/46] KVM: PPC: Book3S HV P9: Use host timer accounting to avoid decrementer read
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
There is no need to save away the host DEC value, as it is derived
from the host timer subsystem, which maintains the next timer time.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/time.h | 5 +++++
arch/powerpc/kernel/time.c | 1 +
arch/powerpc/kvm/book3s_hv.c | 12 ++++++------
3 files changed, 12 insertions(+), 6 deletions(-)
diff --git a/arch/powerpc/include/asm/time.h b/arch/powerpc/include/asm/time.h
index 68d94711811e..0128cd9769bc 100644
--- a/arch/powerpc/include/asm/time.h
+++ b/arch/powerpc/include/asm/time.h
@@ -101,6 +101,11 @@ extern void __init time_init(void);
DECLARE_PER_CPU(u64, decrementers_next_tb);
+static inline u64 timer_get_next_tb(void)
+{
+ return __this_cpu_read(decrementers_next_tb);
+}
+
/* Convert timebase ticks to nanoseconds */
unsigned long long tb_to_ns(unsigned long long tb_ticks);
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index b67d93a609a2..c5d524622c17 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -108,6 +108,7 @@ struct clock_event_device decrementer_clockevent = {
EXPORT_SYMBOL(decrementer_clockevent);
DEFINE_PER_CPU(u64, decrementers_next_tb);
+EXPORT_SYMBOL_GPL(decrementers_next_tb);
static DEFINE_PER_CPU(struct clock_event_device, decrementers);
#define XSEC_PER_SEC (1024*1024)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index bb30c5ab53d1..db807eebb3bd 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3686,16 +3686,16 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
unsigned long host_amr = mfspr(SPRN_AMR);
unsigned long host_fscr = mfspr(SPRN_FSCR);
s64 dec;
- u64 tb;
+ u64 tb, next_timer;
int trap, save_pmu;
- dec = mfspr(SPRN_DEC);
tb = mftb();
- if (dec < 0)
+ next_timer = timer_get_next_tb();
+ if (tb >= next_timer)
return BOOK3S_INTERRUPT_HV_DECREMENTER;
- local_paca->kvm_hstate.dec_expires = dec + tb;
- if (local_paca->kvm_hstate.dec_expires < time_limit)
- time_limit = local_paca->kvm_hstate.dec_expires;
+ local_paca->kvm_hstate.dec_expires = next_timer;
+ if (next_timer < time_limit)
+ time_limit = next_timer;
vcpu->arch.ceded = 0;
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 25/46] KVM: PPC: Book3S HV P9: Use host timer accounting to avoid decrementer read
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
There is no need to save away the host DEC value, as it is derived
from the host timer subsystem, which maintains the next timer time.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/time.h | 5 +++++
arch/powerpc/kernel/time.c | 1 +
arch/powerpc/kvm/book3s_hv.c | 12 ++++++------
3 files changed, 12 insertions(+), 6 deletions(-)
diff --git a/arch/powerpc/include/asm/time.h b/arch/powerpc/include/asm/time.h
index 68d94711811e..0128cd9769bc 100644
--- a/arch/powerpc/include/asm/time.h
+++ b/arch/powerpc/include/asm/time.h
@@ -101,6 +101,11 @@ extern void __init time_init(void);
DECLARE_PER_CPU(u64, decrementers_next_tb);
+static inline u64 timer_get_next_tb(void)
+{
+ return __this_cpu_read(decrementers_next_tb);
+}
+
/* Convert timebase ticks to nanoseconds */
unsigned long long tb_to_ns(unsigned long long tb_ticks);
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index b67d93a609a2..c5d524622c17 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -108,6 +108,7 @@ struct clock_event_device decrementer_clockevent = {
EXPORT_SYMBOL(decrementer_clockevent);
DEFINE_PER_CPU(u64, decrementers_next_tb);
+EXPORT_SYMBOL_GPL(decrementers_next_tb);
static DEFINE_PER_CPU(struct clock_event_device, decrementers);
#define XSEC_PER_SEC (1024*1024)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index bb30c5ab53d1..db807eebb3bd 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3686,16 +3686,16 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
unsigned long host_amr = mfspr(SPRN_AMR);
unsigned long host_fscr = mfspr(SPRN_FSCR);
s64 dec;
- u64 tb;
+ u64 tb, next_timer;
int trap, save_pmu;
- dec = mfspr(SPRN_DEC);
tb = mftb();
- if (dec < 0)
+ next_timer = timer_get_next_tb();
+ if (tb >= next_timer)
return BOOK3S_INTERRUPT_HV_DECREMENTER;
- local_paca->kvm_hstate.dec_expires = dec + tb;
- if (local_paca->kvm_hstate.dec_expires < time_limit)
- time_limit = local_paca->kvm_hstate.dec_expires;
+ local_paca->kvm_hstate.dec_expires = next_timer;
+ if (next_timer < time_limit)
+ time_limit = next_timer;
vcpu->arch.ceded = 0;
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 26/46] KVM: PPC: Book3S HV P9: Reduce mftb per guest entry/exit
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin, Fabiano Rosas
mftb is serialising (dispatch next-to-complete) so it is heavy weight
for a mfspr. Avoid reading it multiple times in the entry or exit paths.
A small number of cycles delay to timers is tolerable.
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index db807eebb3bd..1f38a0abc611 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3540,12 +3540,13 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
host_dawrx1 = mfspr(SPRN_DAWRX1);
}
- hdec = time_limit - mftb();
+ tb = mftb();
+ hdec = time_limit - tb;
if (hdec < 0)
return BOOK3S_INTERRUPT_HV_DECREMENTER;
if (vc->tb_offset) {
- u64 new_tb = mftb() + vc->tb_offset;
+ u64 new_tb = tb + vc->tb_offset;
mtspr(SPRN_TBU40, new_tb);
tb = mftb();
if ((tb & 0xffffff) < (new_tb & 0xffffff))
@@ -3744,7 +3745,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
if (!(vcpu->arch.ctrl & 1))
mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
- mtspr(SPRN_DEC, vcpu->arch.dec_expires - mftb());
+ mtspr(SPRN_DEC, vcpu->arch.dec_expires - tb);
if (kvmhv_on_pseries()) {
/*
@@ -3878,7 +3879,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
vc->entry_exit_map = 0x101;
vc->in_guest = 0;
- mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - mftb());
+ mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - tb);
mtspr(SPRN_SPRG_VDSO_WRITE, local_paca->sprg_vdso);
kvmhv_load_host_pmu();
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 26/46] KVM: PPC: Book3S HV P9: Reduce mftb per guest entry/exit
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin, Fabiano Rosas
mftb is serialising (dispatch next-to-complete) so it is heavy weight
for a mfspr. Avoid reading it multiple times in the entry or exit paths.
A small number of cycles delay to timers is tolerable.
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index db807eebb3bd..1f38a0abc611 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3540,12 +3540,13 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
host_dawrx1 = mfspr(SPRN_DAWRX1);
}
- hdec = time_limit - mftb();
+ tb = mftb();
+ hdec = time_limit - tb;
if (hdec < 0)
return BOOK3S_INTERRUPT_HV_DECREMENTER;
if (vc->tb_offset) {
- u64 new_tb = mftb() + vc->tb_offset;
+ u64 new_tb = tb + vc->tb_offset;
mtspr(SPRN_TBU40, new_tb);
tb = mftb();
if ((tb & 0xffffff) < (new_tb & 0xffffff))
@@ -3744,7 +3745,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
if (!(vcpu->arch.ctrl & 1))
mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
- mtspr(SPRN_DEC, vcpu->arch.dec_expires - mftb());
+ mtspr(SPRN_DEC, vcpu->arch.dec_expires - tb);
if (kvmhv_on_pseries()) {
/*
@@ -3878,7 +3879,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
vc->entry_exit_map = 0x101;
vc->in_guest = 0;
- mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - mftb());
+ mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - tb);
mtspr(SPRN_SPRG_VDSO_WRITE, local_paca->sprg_vdso);
kvmhv_load_host_pmu();
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 27/46] powerpc: add set_dec_or_work API for safely updating decrementer
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Alexey Kardashevskiy, linuxppc-dev, Nicholas Piggin
Decrementer updates must always check for new irq work to avoid an
irq work decrementer interrupt being lost.
Add an API for this in the timer code so callers don't have to care
about details.
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/time.h | 9 +++++++++
arch/powerpc/kernel/time.c | 20 +++++++++++---------
2 files changed, 20 insertions(+), 9 deletions(-)
diff --git a/arch/powerpc/include/asm/time.h b/arch/powerpc/include/asm/time.h
index 0128cd9769bc..d62bde57bf02 100644
--- a/arch/powerpc/include/asm/time.h
+++ b/arch/powerpc/include/asm/time.h
@@ -78,6 +78,15 @@ static inline void set_dec(u64 val)
mtspr(SPRN_DEC, val - 1);
}
+#ifdef CONFIG_IRQ_WORK
+void set_dec_or_work(u64 val);
+#else
+static inline void set_dec_or_work(u64 val)
+{
+ set_dec(val);
+}
+#endif
+
static inline unsigned long tb_ticks_since(unsigned long tstamp)
{
return mftb() - tstamp;
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index c5d524622c17..341cc8442e5e 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -562,6 +562,15 @@ void arch_irq_work_raise(void)
preempt_enable();
}
+void set_dec_or_work(u64 val)
+{
+ set_dec(val);
+ /* We may have raced with new irq work */
+ if (unlikely(test_irq_work_pending()))
+ set_dec(1);
+}
+EXPORT_SYMBOL_GPL(set_dec_or_work);
+
#else /* CONFIG_IRQ_WORK */
#define test_irq_work_pending() 0
@@ -629,10 +638,7 @@ DEFINE_INTERRUPT_HANDLER_ASYNC(timer_interrupt)
} else {
now = *next_tb - now;
if (now <= decrementer_max)
- set_dec(now);
- /* We may have raced with new irq work */
- if (test_irq_work_pending())
- set_dec(1);
+ set_dec_or_work(now);
__this_cpu_inc(irq_stat.timer_irqs_others);
}
@@ -874,11 +880,7 @@ static int decrementer_set_next_event(unsigned long evt,
struct clock_event_device *dev)
{
__this_cpu_write(decrementers_next_tb, get_tb() + evt);
- set_dec(evt);
-
- /* We may have raced with new irq work */
- if (test_irq_work_pending())
- set_dec(1);
+ set_dec_or_work(evt);
return 0;
}
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 27/46] powerpc: add set_dec_or_work API for safely updating decrementer
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: Alexey Kardashevskiy, linuxppc-dev, Nicholas Piggin
Decrementer updates must always check for new irq work to avoid an
irq work decrementer interrupt being lost.
Add an API for this in the timer code so callers don't have to care
about details.
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/time.h | 9 +++++++++
arch/powerpc/kernel/time.c | 20 +++++++++++---------
2 files changed, 20 insertions(+), 9 deletions(-)
diff --git a/arch/powerpc/include/asm/time.h b/arch/powerpc/include/asm/time.h
index 0128cd9769bc..d62bde57bf02 100644
--- a/arch/powerpc/include/asm/time.h
+++ b/arch/powerpc/include/asm/time.h
@@ -78,6 +78,15 @@ static inline void set_dec(u64 val)
mtspr(SPRN_DEC, val - 1);
}
+#ifdef CONFIG_IRQ_WORK
+void set_dec_or_work(u64 val);
+#else
+static inline void set_dec_or_work(u64 val)
+{
+ set_dec(val);
+}
+#endif
+
static inline unsigned long tb_ticks_since(unsigned long tstamp)
{
return mftb() - tstamp;
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index c5d524622c17..341cc8442e5e 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -562,6 +562,15 @@ void arch_irq_work_raise(void)
preempt_enable();
}
+void set_dec_or_work(u64 val)
+{
+ set_dec(val);
+ /* We may have raced with new irq work */
+ if (unlikely(test_irq_work_pending()))
+ set_dec(1);
+}
+EXPORT_SYMBOL_GPL(set_dec_or_work);
+
#else /* CONFIG_IRQ_WORK */
#define test_irq_work_pending() 0
@@ -629,10 +638,7 @@ DEFINE_INTERRUPT_HANDLER_ASYNC(timer_interrupt)
} else {
now = *next_tb - now;
if (now <= decrementer_max)
- set_dec(now);
- /* We may have raced with new irq work */
- if (test_irq_work_pending())
- set_dec(1);
+ set_dec_or_work(now);
__this_cpu_inc(irq_stat.timer_irqs_others);
}
@@ -874,11 +880,7 @@ static int decrementer_set_next_event(unsigned long evt,
struct clock_event_device *dev)
{
__this_cpu_write(decrementers_next_tb, get_tb() + evt);
- set_dec(evt);
-
- /* We may have raced with new irq work */
- if (test_irq_work_pending())
- set_dec(1);
+ set_dec_or_work(evt);
return 0;
}
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 28/46] KVM: PPC: Book3S HV P9: Reduce irq_work vs guest decrementer races
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
irq_work's use of the DEC SPR is racy with guest<->host switch and guest
entry which flips the DEC interrupt to guest, which could lose a host
work interrupt.
This patch closes one race, and attempts to comment another class of
races.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 15 ++++++++++++++-
1 file changed, 14 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 1f38a0abc611..989a1ff5ad11 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3745,6 +3745,18 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
if (!(vcpu->arch.ctrl & 1))
mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
+ /*
+ * When setting DEC, we must always deal with irq_work_raise via NMI vs
+ * setting DEC. The problem occurs right as we switch into guest mode
+ * if a NMI hits and sets pending work and sets DEC, then that will
+ * apply to the guest and not bring us back to the host.
+ *
+ * irq_work_raise could check a flag (or possibly LPCR[HDICE] for
+ * example) and set HDEC to 1? That wouldn't solve the nested hv
+ * case which needs to abort the hcall or zero the time limit.
+ *
+ * XXX: Another day's problem.
+ */
mtspr(SPRN_DEC, vcpu->arch.dec_expires - tb);
if (kvmhv_on_pseries()) {
@@ -3879,7 +3891,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
vc->entry_exit_map = 0x101;
vc->in_guest = 0;
- mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - tb);
+ set_dec_or_work(local_paca->kvm_hstate.dec_expires - tb);
+
mtspr(SPRN_SPRG_VDSO_WRITE, local_paca->sprg_vdso);
kvmhv_load_host_pmu();
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 28/46] KVM: PPC: Book3S HV P9: Reduce irq_work vs guest decrementer races
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
irq_work's use of the DEC SPR is racy with guest<->host switch and guest
entry which flips the DEC interrupt to guest, which could lose a host
work interrupt.
This patch closes one race, and attempts to comment another class of
races.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 15 ++++++++++++++-
1 file changed, 14 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 1f38a0abc611..989a1ff5ad11 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3745,6 +3745,18 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
if (!(vcpu->arch.ctrl & 1))
mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
+ /*
+ * When setting DEC, we must always deal with irq_work_raise via NMI vs
+ * setting DEC. The problem occurs right as we switch into guest mode
+ * if a NMI hits and sets pending work and sets DEC, then that will
+ * apply to the guest and not bring us back to the host.
+ *
+ * irq_work_raise could check a flag (or possibly LPCR[HDICE] for
+ * example) and set HDEC to 1? That wouldn't solve the nested hv
+ * case which needs to abort the hcall or zero the time limit.
+ *
+ * XXX: Another day's problem.
+ */
mtspr(SPRN_DEC, vcpu->arch.dec_expires - tb);
if (kvmhv_on_pseries()) {
@@ -3879,7 +3891,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
vc->entry_exit_map = 0x101;
vc->in_guest = 0;
- mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - tb);
+ set_dec_or_work(local_paca->kvm_hstate.dec_expires - tb);
+
mtspr(SPRN_SPRG_VDSO_WRITE, local_paca->sprg_vdso);
kvmhv_load_host_pmu();
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* Re: [PATCH v4 28/46] KVM: PPC: Book3S HV P9: Reduce irq_work vs guest decrementer races
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 10:13 ` Alexey Kardashevskiy
-1 siblings, 0 replies; 188+ messages in thread
From: Alexey Kardashevskiy @ 2021-03-23 10:13 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 23/03/2021 12:02, Nicholas Piggin wrote:
> irq_work's use of the DEC SPR is racy with guest<->host switch and guest
> entry which flips the DEC interrupt to guest, which could lose a host
> work interrupt.
>
> This patch closes one race, and attempts to comment another class of
> races.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
> arch/powerpc/kvm/book3s_hv.c | 15 ++++++++++++++-
> 1 file changed, 14 insertions(+), 1 deletion(-)
>
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index 1f38a0abc611..989a1ff5ad11 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -3745,6 +3745,18 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> if (!(vcpu->arch.ctrl & 1))
> mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
>
> + /*
> + * When setting DEC, we must always deal with irq_work_raise via NMI vs
> + * setting DEC. The problem occurs right as we switch into guest mode
> + * if a NMI hits and sets pending work and sets DEC, then that will
> + * apply to the guest and not bring us back to the host.
> + *
> + * irq_work_raise could check a flag (or possibly LPCR[HDICE] for
> + * example) and set HDEC to 1? That wouldn't solve the nested hv
> + * case which needs to abort the hcall or zero the time limit.
> + *
> + * XXX: Another day's problem.
> + */
> mtspr(SPRN_DEC, vcpu->arch.dec_expires - tb);
>
> if (kvmhv_on_pseries()) {
> @@ -3879,7 +3891,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> vc->entry_exit_map = 0x101;
> vc->in_guest = 0;
>
> - mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - tb);
> + set_dec_or_work(local_paca->kvm_hstate.dec_expires - tb);
set_dec_or_work() will write local_paca->kvm_hstate.dec_expires - tb - 1
to SPRN_DEC which is not exactly the same, is this still alright?
I asked in v3 but it is probably lost :)
> +
> mtspr(SPRN_SPRG_VDSO_WRITE, local_paca->sprg_vdso);
>
> kvmhv_load_host_pmu();
>
--
Alexey
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 28/46] KVM: PPC: Book3S HV P9: Reduce irq_work vs guest decrementer races
@ 2021-03-23 10:13 ` Alexey Kardashevskiy
0 siblings, 0 replies; 188+ messages in thread
From: Alexey Kardashevskiy @ 2021-03-23 10:13 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 23/03/2021 12:02, Nicholas Piggin wrote:
> irq_work's use of the DEC SPR is racy with guest<->host switch and guest
> entry which flips the DEC interrupt to guest, which could lose a host
> work interrupt.
>
> This patch closes one race, and attempts to comment another class of
> races.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
> arch/powerpc/kvm/book3s_hv.c | 15 ++++++++++++++-
> 1 file changed, 14 insertions(+), 1 deletion(-)
>
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index 1f38a0abc611..989a1ff5ad11 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -3745,6 +3745,18 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> if (!(vcpu->arch.ctrl & 1))
> mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
>
> + /*
> + * When setting DEC, we must always deal with irq_work_raise via NMI vs
> + * setting DEC. The problem occurs right as we switch into guest mode
> + * if a NMI hits and sets pending work and sets DEC, then that will
> + * apply to the guest and not bring us back to the host.
> + *
> + * irq_work_raise could check a flag (or possibly LPCR[HDICE] for
> + * example) and set HDEC to 1? That wouldn't solve the nested hv
> + * case which needs to abort the hcall or zero the time limit.
> + *
> + * XXX: Another day's problem.
> + */
> mtspr(SPRN_DEC, vcpu->arch.dec_expires - tb);
>
> if (kvmhv_on_pseries()) {
> @@ -3879,7 +3891,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> vc->entry_exit_map = 0x101;
> vc->in_guest = 0;
>
> - mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - tb);
> + set_dec_or_work(local_paca->kvm_hstate.dec_expires - tb);
set_dec_or_work() will write local_paca->kvm_hstate.dec_expires - tb - 1
to SPRN_DEC which is not exactly the same, is this still alright?
I asked in v3 but it is probably lost :)
> +
> mtspr(SPRN_SPRG_VDSO_WRITE, local_paca->sprg_vdso);
>
> kvmhv_load_host_pmu();
>
--
Alexey
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 28/46] KVM: PPC: Book3S HV P9: Reduce irq_work vs guest decrementer races
2021-03-23 10:13 ` Alexey Kardashevskiy
@ 2021-03-23 10:36 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 10:36 UTC (permalink / raw)
To: Alexey Kardashevskiy, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Alexey Kardashevskiy's message of March 23, 2021 8:13 pm:
>
>
> On 23/03/2021 12:02, Nicholas Piggin wrote:
>> irq_work's use of the DEC SPR is racy with guest<->host switch and guest
>> entry which flips the DEC interrupt to guest, which could lose a host
>> work interrupt.
>>
>> This patch closes one race, and attempts to comment another class of
>> races.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>> ---
>> arch/powerpc/kvm/book3s_hv.c | 15 ++++++++++++++-
>> 1 file changed, 14 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
>> index 1f38a0abc611..989a1ff5ad11 100644
>> --- a/arch/powerpc/kvm/book3s_hv.c
>> +++ b/arch/powerpc/kvm/book3s_hv.c
>> @@ -3745,6 +3745,18 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
>> if (!(vcpu->arch.ctrl & 1))
>> mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
>>
>> + /*
>> + * When setting DEC, we must always deal with irq_work_raise via NMI vs
>> + * setting DEC. The problem occurs right as we switch into guest mode
>> + * if a NMI hits and sets pending work and sets DEC, then that will
>> + * apply to the guest and not bring us back to the host.
>> + *
>> + * irq_work_raise could check a flag (or possibly LPCR[HDICE] for
>> + * example) and set HDEC to 1? That wouldn't solve the nested hv
>> + * case which needs to abort the hcall or zero the time limit.
>> + *
>> + * XXX: Another day's problem.
>> + */
>> mtspr(SPRN_DEC, vcpu->arch.dec_expires - tb);
>>
>> if (kvmhv_on_pseries()) {
>> @@ -3879,7 +3891,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
>> vc->entry_exit_map = 0x101;
>> vc->in_guest = 0;
>>
>> - mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - tb);
>> + set_dec_or_work(local_paca->kvm_hstate.dec_expires - tb);
>
>
> set_dec_or_work() will write local_paca->kvm_hstate.dec_expires - tb - 1
> to SPRN_DEC which is not exactly the same, is this still alright?
>
> I asked in v3 but it is probably lost :)
Oh I did see that then forgot.
It will write dec_expires - tb, then it will write 1 if it found irq_work
was pending.
The change is intentional to fixes one of the lost irq_work races.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 28/46] KVM: PPC: Book3S HV P9: Reduce irq_work vs guest decrementer races
@ 2021-03-23 10:36 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 10:36 UTC (permalink / raw)
To: Alexey Kardashevskiy, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Alexey Kardashevskiy's message of March 23, 2021 8:13 pm:
>
>
> On 23/03/2021 12:02, Nicholas Piggin wrote:
>> irq_work's use of the DEC SPR is racy with guest<->host switch and guest
>> entry which flips the DEC interrupt to guest, which could lose a host
>> work interrupt.
>>
>> This patch closes one race, and attempts to comment another class of
>> races.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>> ---
>> arch/powerpc/kvm/book3s_hv.c | 15 ++++++++++++++-
>> 1 file changed, 14 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
>> index 1f38a0abc611..989a1ff5ad11 100644
>> --- a/arch/powerpc/kvm/book3s_hv.c
>> +++ b/arch/powerpc/kvm/book3s_hv.c
>> @@ -3745,6 +3745,18 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
>> if (!(vcpu->arch.ctrl & 1))
>> mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
>>
>> + /*
>> + * When setting DEC, we must always deal with irq_work_raise via NMI vs
>> + * setting DEC. The problem occurs right as we switch into guest mode
>> + * if a NMI hits and sets pending work and sets DEC, then that will
>> + * apply to the guest and not bring us back to the host.
>> + *
>> + * irq_work_raise could check a flag (or possibly LPCR[HDICE] for
>> + * example) and set HDEC to 1? That wouldn't solve the nested hv
>> + * case which needs to abort the hcall or zero the time limit.
>> + *
>> + * XXX: Another day's problem.
>> + */
>> mtspr(SPRN_DEC, vcpu->arch.dec_expires - tb);
>>
>> if (kvmhv_on_pseries()) {
>> @@ -3879,7 +3891,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
>> vc->entry_exit_map = 0x101;
>> vc->in_guest = 0;
>>
>> - mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - tb);
>> + set_dec_or_work(local_paca->kvm_hstate.dec_expires - tb);
>
>
> set_dec_or_work() will write local_paca->kvm_hstate.dec_expires - tb - 1
> to SPRN_DEC which is not exactly the same, is this still alright?
>
> I asked in v3 but it is probably lost :)
Oh I did see that then forgot.
It will write dec_expires - tb, then it will write 1 if it found irq_work
was pending.
The change is intentional to fixes one of the lost irq_work races.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 28/46] KVM: PPC: Book3S HV P9: Reduce irq_work vs guest decrementer races
2021-03-23 10:36 ` Nicholas Piggin
@ 2021-03-23 11:15 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 11:15 UTC (permalink / raw)
To: Alexey Kardashevskiy, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Nicholas Piggin's message of March 23, 2021 8:36 pm:
> Excerpts from Alexey Kardashevskiy's message of March 23, 2021 8:13 pm:
>>
>>
>> On 23/03/2021 12:02, Nicholas Piggin wrote:
>>> irq_work's use of the DEC SPR is racy with guest<->host switch and guest
>>> entry which flips the DEC interrupt to guest, which could lose a host
>>> work interrupt.
>>>
>>> This patch closes one race, and attempts to comment another class of
>>> races.
>>>
>>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>>> ---
>>> arch/powerpc/kvm/book3s_hv.c | 15 ++++++++++++++-
>>> 1 file changed, 14 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
>>> index 1f38a0abc611..989a1ff5ad11 100644
>>> --- a/arch/powerpc/kvm/book3s_hv.c
>>> +++ b/arch/powerpc/kvm/book3s_hv.c
>>> @@ -3745,6 +3745,18 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
>>> if (!(vcpu->arch.ctrl & 1))
>>> mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
>>>
>>> + /*
>>> + * When setting DEC, we must always deal with irq_work_raise via NMI vs
>>> + * setting DEC. The problem occurs right as we switch into guest mode
>>> + * if a NMI hits and sets pending work and sets DEC, then that will
>>> + * apply to the guest and not bring us back to the host.
>>> + *
>>> + * irq_work_raise could check a flag (or possibly LPCR[HDICE] for
>>> + * example) and set HDEC to 1? That wouldn't solve the nested hv
>>> + * case which needs to abort the hcall or zero the time limit.
>>> + *
>>> + * XXX: Another day's problem.
>>> + */
>>> mtspr(SPRN_DEC, vcpu->arch.dec_expires - tb);
>>>
>>> if (kvmhv_on_pseries()) {
>>> @@ -3879,7 +3891,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
>>> vc->entry_exit_map = 0x101;
>>> vc->in_guest = 0;
>>>
>>> - mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - tb);
>>> + set_dec_or_work(local_paca->kvm_hstate.dec_expires - tb);
>>
>>
>> set_dec_or_work() will write local_paca->kvm_hstate.dec_expires - tb - 1
>> to SPRN_DEC which is not exactly the same, is this still alright?
>>
>> I asked in v3 but it is probably lost :)
>
> Oh I did see that then forgot.
>
> It will write dec_expires - tb, then it will write 1 if it found irq_work
> was pending.
Ah you were actually asking about set_dec writing val - 1. I totally
missed that.
Yes that was an unintentional change. This is the way timer.c code works
with respect to the decrementers_next_tb value, so it's probably better
to make them so it seems like it should be okay (and better to bring the
KVM code up to match timer code rather than be different or the other
way around). The difference should be noted in the changelog though.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 28/46] KVM: PPC: Book3S HV P9: Reduce irq_work vs guest decrementer races
@ 2021-03-23 11:15 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 11:15 UTC (permalink / raw)
To: Alexey Kardashevskiy, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Nicholas Piggin's message of March 23, 2021 8:36 pm:
> Excerpts from Alexey Kardashevskiy's message of March 23, 2021 8:13 pm:
>>
>>
>> On 23/03/2021 12:02, Nicholas Piggin wrote:
>>> irq_work's use of the DEC SPR is racy with guest<->host switch and guest
>>> entry which flips the DEC interrupt to guest, which could lose a host
>>> work interrupt.
>>>
>>> This patch closes one race, and attempts to comment another class of
>>> races.
>>>
>>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>>> ---
>>> arch/powerpc/kvm/book3s_hv.c | 15 ++++++++++++++-
>>> 1 file changed, 14 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
>>> index 1f38a0abc611..989a1ff5ad11 100644
>>> --- a/arch/powerpc/kvm/book3s_hv.c
>>> +++ b/arch/powerpc/kvm/book3s_hv.c
>>> @@ -3745,6 +3745,18 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
>>> if (!(vcpu->arch.ctrl & 1))
>>> mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
>>>
>>> + /*
>>> + * When setting DEC, we must always deal with irq_work_raise via NMI vs
>>> + * setting DEC. The problem occurs right as we switch into guest mode
>>> + * if a NMI hits and sets pending work and sets DEC, then that will
>>> + * apply to the guest and not bring us back to the host.
>>> + *
>>> + * irq_work_raise could check a flag (or possibly LPCR[HDICE] for
>>> + * example) and set HDEC to 1? That wouldn't solve the nested hv
>>> + * case which needs to abort the hcall or zero the time limit.
>>> + *
>>> + * XXX: Another day's problem.
>>> + */
>>> mtspr(SPRN_DEC, vcpu->arch.dec_expires - tb);
>>>
>>> if (kvmhv_on_pseries()) {
>>> @@ -3879,7 +3891,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
>>> vc->entry_exit_map = 0x101;
>>> vc->in_guest = 0;
>>>
>>> - mtspr(SPRN_DEC, local_paca->kvm_hstate.dec_expires - tb);
>>> + set_dec_or_work(local_paca->kvm_hstate.dec_expires - tb);
>>
>>
>> set_dec_or_work() will write local_paca->kvm_hstate.dec_expires - tb - 1
>> to SPRN_DEC which is not exactly the same, is this still alright?
>>
>> I asked in v3 but it is probably lost :)
>
> Oh I did see that then forgot.
>
> It will write dec_expires - tb, then it will write 1 if it found irq_work
> was pending.
Ah you were actually asking about set_dec writing val - 1. I totally
missed that.
Yes that was an unintentional change. This is the way timer.c code works
with respect to the decrementers_next_tb value, so it's probably better
to make them so it seems like it should be okay (and better to bring the
KVM code up to match timer code rather than be different or the other
way around). The difference should be noted in the changelog though.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 29/46] KVM: PPC: Book3S HV P9: Implement the rest of the P9 path in C
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Almost all logic is moved to C, by introducing a new in_guest mode that
selects and branches very early in the interrupt handler to the P9 exit
code.
The remaining assembly is only about 160 lines of low level stack setup,
with VCPU vs host register save and restore, plus a small shim to the
legacy paths in the interrupt handler.
There are two motivations for this, the first is just make the code more
maintainable being in C. The second is to reduce the amount of code
running in a special KVM mode, "realmode". I put that in quotes because
with radix it is no longer necessarily real-mode in the MMU, but it
still has to be treated specially because it may be in real-mode, and
has various important registers like PID, DEC, TB, etc set to guest.
This is hostile to the rest of Linux and can't use arbitrary kernel
functionality or be instrumented well.
This initial patch is a reasonably faithful conversion of the asm code.
It does lack any loop to return quickly back into the guest without
switching out of realmode in the case of unimportant or easily handled
interrupts, as explained in the previous change, handling HV interrupts
in real mode is not so important for P9.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/asm-prototypes.h | 3 +-
arch/powerpc/include/asm/kvm_asm.h | 3 +-
arch/powerpc/include/asm/kvm_book3s_64.h | 8 +
arch/powerpc/kernel/security.c | 5 +-
arch/powerpc/kvm/Makefile | 3 +
arch/powerpc/kvm/book3s_64_entry.S | 246 ++++++++++++++++++++++
arch/powerpc/kvm/book3s_hv.c | 9 +-
arch/powerpc/kvm/book3s_hv_interrupt.c | 223 ++++++++++++++++++++
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 123 +----------
9 files changed, 500 insertions(+), 123 deletions(-)
create mode 100644 arch/powerpc/kvm/book3s_hv_interrupt.c
diff --git a/arch/powerpc/include/asm/asm-prototypes.h b/arch/powerpc/include/asm/asm-prototypes.h
index 939f3c94c8f3..7c74c80ed994 100644
--- a/arch/powerpc/include/asm/asm-prototypes.h
+++ b/arch/powerpc/include/asm/asm-prototypes.h
@@ -122,6 +122,7 @@ extern s32 patch__call_flush_branch_caches3;
extern s32 patch__flush_count_cache_return;
extern s32 patch__flush_link_stack_return;
extern s32 patch__call_kvm_flush_link_stack;
+extern s32 patch__call_kvm_flush_link_stack_p9;
extern s32 patch__memset_nocache, patch__memcpy_nocache;
extern long flush_branch_caches;
@@ -142,7 +143,7 @@ void kvmhv_load_host_pmu(void);
void kvmhv_save_guest_pmu(struct kvm_vcpu *vcpu, bool pmu_in_use);
void kvmhv_load_guest_pmu(struct kvm_vcpu *vcpu);
-int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu);
+void kvmppc_p9_enter_guest(struct kvm_vcpu *vcpu);
long kvmppc_h_set_dabr(struct kvm_vcpu *vcpu, unsigned long dabr);
long kvmppc_h_set_xdabr(struct kvm_vcpu *vcpu, unsigned long dabr,
diff --git a/arch/powerpc/include/asm/kvm_asm.h b/arch/powerpc/include/asm/kvm_asm.h
index a3633560493b..b4f9996bd331 100644
--- a/arch/powerpc/include/asm/kvm_asm.h
+++ b/arch/powerpc/include/asm/kvm_asm.h
@@ -146,7 +146,8 @@
#define KVM_GUEST_MODE_GUEST 1
#define KVM_GUEST_MODE_SKIP 2
#define KVM_GUEST_MODE_GUEST_HV 3
-#define KVM_GUEST_MODE_HOST_HV 4
+#define KVM_GUEST_MODE_GUEST_HV_FAST 4 /* ISA v3.0 with host radix mode */
+#define KVM_GUEST_MODE_HOST_HV 5
#define KVM_INST_FETCH_FAILED -1
diff --git a/arch/powerpc/include/asm/kvm_book3s_64.h b/arch/powerpc/include/asm/kvm_book3s_64.h
index 9bb9bb370b53..c214bcffb441 100644
--- a/arch/powerpc/include/asm/kvm_book3s_64.h
+++ b/arch/powerpc/include/asm/kvm_book3s_64.h
@@ -153,9 +153,17 @@ static inline bool kvmhv_vcpu_is_radix(struct kvm_vcpu *vcpu)
return radix;
}
+int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu);
+
#define KVM_DEFAULT_HPT_ORDER 24 /* 16MB HPT by default */
#endif
+/*
+ * Invalid HDSISR value which is used to indicate when HW has not set the reg.
+ * Used to work around an errata.
+ */
+#define HDSISR_CANARY 0x7fff
+
/*
* We use a lock bit in HPTE dword 0 to synchronize updates and
* accesses to each HPTE, and another bit to indicate non-present
diff --git a/arch/powerpc/kernel/security.c b/arch/powerpc/kernel/security.c
index e4e1a94ccf6a..3a607c11f20f 100644
--- a/arch/powerpc/kernel/security.c
+++ b/arch/powerpc/kernel/security.c
@@ -430,16 +430,19 @@ device_initcall(stf_barrier_debugfs_init);
static void update_branch_cache_flush(void)
{
- u32 *site;
+ u32 *site, __maybe_unused *site2;
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
site = &patch__call_kvm_flush_link_stack;
+ site2 = &patch__call_kvm_flush_link_stack_p9;
// This controls the branch from guest_exit_cont to kvm_flush_link_stack
if (link_stack_flush_type == BRANCH_CACHE_FLUSH_NONE) {
patch_instruction_site(site, ppc_inst(PPC_INST_NOP));
+ patch_instruction_site(site2, ppc_inst(PPC_INST_NOP));
} else {
// Could use HW flush, but that could also flush count cache
patch_branch_site(site, (u64)&kvm_flush_link_stack, BRANCH_SET_LINK);
+ patch_branch_site(site2, (u64)&kvm_flush_link_stack, BRANCH_SET_LINK);
}
#endif
diff --git a/arch/powerpc/kvm/Makefile b/arch/powerpc/kvm/Makefile
index cdd119028f64..40a55a5ba4ff 100644
--- a/arch/powerpc/kvm/Makefile
+++ b/arch/powerpc/kvm/Makefile
@@ -43,6 +43,9 @@ kvm-objs-$(CONFIG_KVM_E500MC) := $(kvm-e500mc-objs)
kvm-book3s_64-builtin-objs-$(CONFIG_SPAPR_TCE_IOMMU) := \
book3s_64_vio_hv.o
+kvm-book3s_64-builtin-objs-y += \
+ book3s_hv_interrupt.o
+
kvm-pr-y := \
fpu.o \
emulate.o \
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index de81ab69555b..845df5fefdbd 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -1,11 +1,16 @@
/* SPDX-License-Identifier: GPL-2.0-only */
#include <asm/asm-offsets.h>
#include <asm/cache.h>
+#include <asm/code-patching-asm.h>
#include <asm/exception-64s.h>
+#include <asm/export.h>
#include <asm/kvm_asm.h>
#include <asm/kvm_book3s_asm.h>
+#include <asm/mmu.h>
#include <asm/ppc_asm.h>
+#include <asm/ptrace.h>
#include <asm/reg.h>
+#include <asm/ultravisor-api.h>
/*
* These are branched to from interrupt handlers in exception-64s.S which set
@@ -22,10 +27,15 @@
.global kvmppc_hcall
.balign IFETCH_ALIGN_BYTES
kvmppc_hcall:
+ lbz r10,HSTATE_IN_GUEST(r13)
+ cmpwi r10,KVM_GUEST_MODE_GUEST_HV_FAST
+ beq kvmppc_p9_exit_hcall
ld r10,PACA_EXGEN+EX_R13(r13)
SET_SCRATCH0(r10)
li r10,0xc00
/* Now we look like kvmppc_interrupt */
+ li r11,PACA_EXGEN
+ b 1f
/*
* KVM interrupt entry occurs after GEN_INT_ENTRY runs, and follows that
@@ -46,6 +56,12 @@ kvmppc_hcall:
.global kvmppc_interrupt
.balign IFETCH_ALIGN_BYTES
kvmppc_interrupt:
+ std r10,HSTATE_SCRATCH0(r13)
+ lbz r10,HSTATE_IN_GUEST(r13)
+ cmpwi r10,KVM_GUEST_MODE_GUEST_HV_FAST
+ beq kvmppc_p9_exit_interrupt
+ ld r10,HSTATE_SCRATCH0(r13)
+ lbz r11,HSTATE_IN_GUEST(r13)
li r11,PACA_EXGEN
cmpdi r10,0x200
bgt+ 1f
@@ -140,3 +156,233 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
GET_SCRATCH0(r13)
HRFI_TO_KERNEL
#endif
+
+/* Stack frame offsets for kvmppc_hv_entry */
+#define SFS (144 + STACK_FRAME_MIN_SIZE)
+#define STACK_SLOT_NVGPRS (SFS - 144) /* 18 gprs */
+
+/*
+ * void kvmppc_p9_enter_guest(struct vcpu *vcpu);
+ *
+ * Enter the guest on a ISAv3.0 or later system where we have exactly
+ * one vcpu per vcore, and both the host and guest are radix, and threads
+ * are set to "indepdent mode".
+ */
+.balign IFETCH_ALIGN_BYTES
+_GLOBAL(kvmppc_p9_enter_guest)
+EXPORT_SYMBOL_GPL(kvmppc_p9_enter_guest)
+ mflr r0
+ std r0,PPC_LR_STKOFF(r1)
+ stdu r1,-SFS(r1)
+
+ std r1,HSTATE_HOST_R1(r13)
+
+ mfcr r4
+ stw r4,SFS+8(r1)
+
+ reg = 14
+ .rept 18
+ std reg,STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
+ reg = reg + 1
+ .endr
+
+ ld r4,VCPU_LR(r3)
+ mtlr r4
+ ld r4,VCPU_CTR(r3)
+ mtctr r4
+ ld r4,VCPU_XER(r3)
+ mtspr SPRN_XER,r4
+
+ ld r1,VCPU_CR(r3)
+
+BEGIN_FTR_SECTION
+ ld r4,VCPU_CFAR(r3)
+ mtspr SPRN_CFAR,r4
+END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
+BEGIN_FTR_SECTION
+ ld r4,VCPU_PPR(r3)
+ mtspr SPRN_PPR,r4
+END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
+
+ reg = 4
+ .rept 28
+ ld reg,__VCPU_GPR(reg)(r3)
+ reg = reg + 1
+ .endr
+
+ ld r4,VCPU_KVM(r3)
+ lbz r4,KVM_SECURE_GUEST(r4)
+ cmpdi r4,0
+ ld r4,VCPU_GPR(R4)(r3)
+ bne .Lret_to_ultra
+
+ mtcr r1
+
+ ld r0,VCPU_GPR(R0)(r3)
+ ld r1,VCPU_GPR(R1)(r3)
+ ld r2,VCPU_GPR(R2)(r3)
+ ld r3,VCPU_GPR(R3)(r3)
+
+ HRFI_TO_GUEST
+ b .
+
+ /*
+ * Use UV_RETURN ultracall to return control back to the Ultravisor
+ * after processing an hypercall or interrupt that was forwarded
+ * (a.k.a. reflected) to the Hypervisor.
+ *
+ * All registers have already been reloaded except the ucall requires:
+ * R0 = hcall result
+ * R2 = SRR1, so UV can detect a synthesized interrupt (if any)
+ * R3 = UV_RETURN
+ */
+.Lret_to_ultra:
+ mtcr r1
+ ld r1,VCPU_GPR(R1)(r3)
+
+ ld r0,VCPU_GPR(R3)(r3)
+ mfspr r2,SPRN_SRR1
+ LOAD_REG_IMMEDIATE(r3, UV_RETURN)
+ sc 2
+
+/*
+ * kvmppc_p9_exit_hcall and kvmppc_p9_exit_interrupt are branched to from
+ * above if the interrupt was taken for a guest that was entered via
+ * kvmppc_p9_enter_guest().
+ *
+ * This code recovers the host stack and vcpu pointer, saves all GPRs and
+ * CR, LR, CTR, XER as well as guest MSR and NIA into the VCPU, then re-
+ * establishes the host stack and registers to return from the
+ * kvmppc_p9_enter_guest() function.
+ */
+.balign IFETCH_ALIGN_BYTES
+kvmppc_p9_exit_hcall:
+ mfspr r11,SPRN_SRR0
+ mfspr r12,SPRN_SRR1
+ li r10,0xc00
+ std r10,HSTATE_SCRATCH0(r13)
+
+.balign IFETCH_ALIGN_BYTES
+kvmppc_p9_exit_interrupt:
+ /*
+ * If set to KVM_GUEST_MODE_GUEST_HV_FAST but we're still in the
+ * hypervisor, that means we can't return from the entry stack.
+ */
+ rldicl. r10,r12,64-MSR_HV_LG,63
+ bne- kvmppc_p9_bad_interrupt
+
+ std r1,HSTATE_SCRATCH1(r13)
+ std r3,HSTATE_SCRATCH2(r13)
+ ld r1,HSTATE_HOST_R1(r13)
+ ld r3,HSTATE_KVM_VCPU(r13)
+
+ std r9,VCPU_CR(r3)
+
+1:
+ std r11,VCPU_PC(r3)
+ std r12,VCPU_MSR(r3)
+
+ reg = 14
+ .rept 18
+ std reg,__VCPU_GPR(reg)(r3)
+ reg = reg + 1
+ .endr
+
+ /* r1, r3, r9-r13 are saved to vcpu by C code */
+ std r0,VCPU_GPR(R0)(r3)
+ std r2,VCPU_GPR(R2)(r3)
+ reg = 4
+ .rept 5
+ std reg,__VCPU_GPR(reg)(r3)
+ reg = reg + 1
+ .endr
+
+ ld r2,PACATOC(r13)
+
+ mflr r4
+ std r4,VCPU_LR(r3)
+ mfspr r4,SPRN_XER
+ std r4,VCPU_XER(r3)
+
+ reg = 14
+ .rept 18
+ ld reg,STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
+ reg = reg + 1
+ .endr
+
+ lwz r4,SFS+8(r1)
+ mtcr r4
+
+ /*
+ * Flush the link stack here, before executing the first blr on the
+ * way out of the guest.
+ *
+ * The link stack won't match coming out of the guest anyway so the
+ * only cost is the flush itself. The call clobbers r0.
+ */
+1: nop
+ patch_site 1b patch__call_kvm_flush_link_stack_p9
+
+ addi r1,r1,SFS
+ ld r0,PPC_LR_STKOFF(r1)
+ mtlr r0
+ blr
+
+/*
+ * Took an interrupt somewhere right before HRFID to guest, so registers are
+ * in a bad way. Return things hopefully enough to run host virtual code and
+ * run the Linux interrupt handler (SRESET or MCE) to print something useful.
+ *
+ * We could be really clever and save all host registers in known locations
+ * before setting HSTATE_IN_GUEST, then restoring them all here, and setting
+ * return address to a fixup that sets them up again. But that's a lot of
+ * effort for a small bit of code. Lots of other things to do first.
+ */
+kvmppc_p9_bad_interrupt:
+ /*
+ * Set GUEST_MODE_NONE so the handler won't branch to KVM, and clear
+ * MSR_RI in r12 ([H]SRR1) so the handler won't try to return.
+ */
+ li r10,KVM_GUEST_MODE_NONE
+ stb r10,HSTATE_IN_GUEST(r13)
+ li r10,MSR_RI
+ andc r12,r12,r10
+
+ /*
+ * Clean up guest registers to give host a chance to run.
+ */
+ li r10,0
+ mtspr SPRN_AMR,r10
+ mtspr SPRN_IAMR,r10
+ mtspr SPRN_CIABR,r10
+ mtspr SPRN_DAWRX0,r10
+BEGIN_FTR_SECTION
+ mtspr SPRN_DAWRX1,r10
+END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
+ mtspr SPRN_PID,r10
+
+ /*
+ * Switch to host MMU mode
+ */
+ ld r10, HSTATE_KVM_VCPU(r13)
+ ld r10, VCPU_KVM(r10)
+ lwz r10, KVM_HOST_LPID(r10)
+ mtspr SPRN_LPID,r10
+
+ ld r10, HSTATE_KVM_VCPU(r13)
+ ld r10, VCPU_KVM(r10)
+ ld r10, KVM_HOST_LPCR(r10)
+ mtspr SPRN_LPCR,r10
+
+ /*
+ * Go back to interrupt handler
+ */
+ ld r10,HSTATE_SCRATCH0(r13)
+ cmpwi r10,BOOK3S_INTERRUPT_MACHINE_CHECK
+ beq machine_check_common
+
+ ld r10,HSTATE_SCRATCH0(r13)
+ cmpwi r10,BOOK3S_INTERRUPT_SYSTEM_RESET
+ beq system_reset_common
+
+ b .
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 989a1ff5ad11..6dca639ed997 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -1431,6 +1431,8 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
*/
case BOOK3S_INTERRUPT_H_DATA_STORAGE:
r = RESUME_PAGE_FAULT;
+ if (vcpu->arch.fault_dsisr == HDSISR_CANARY)
+ r = RESUME_GUEST; /* Just retry if it's the canary */
break;
case BOOK3S_INTERRUPT_H_INST_STORAGE:
vcpu->arch.fault_dar = kvmppc_get_pc(vcpu);
@@ -3690,6 +3692,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
u64 tb, next_timer;
int trap, save_pmu;
+ WARN_ON_ONCE(vcpu->arch.ceded);
+
tb = mftb();
next_timer = timer_get_next_tb();
if (tb >= next_timer)
@@ -3698,8 +3702,6 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
if (next_timer < time_limit)
time_limit = next_timer;
- vcpu->arch.ceded = 0;
-
kvmhv_save_host_pmu(); /* saves it to PACA kvm_hstate */
kvmppc_subcore_enter_guest();
@@ -3826,9 +3828,10 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
}
}
kvmppc_xive_pull_vcpu(vcpu);
+
+ vcpu->arch.slb_max = 0;
}
- vcpu->arch.slb_max = 0;
dec = mfspr(SPRN_DEC);
if (!(lpcr & LPCR_LD)) /* Sign extend if not using large decrementer */
dec = (s32) dec;
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
new file mode 100644
index 000000000000..3151b3d62c01
--- /dev/null
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -0,0 +1,223 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include <linux/kernel.h>
+#include <linux/kvm_host.h>
+#include <asm/asm-prototypes.h>
+#include <asm/dbell.h>
+#include <asm/kvm_ppc.h>
+
+#ifdef CONFIG_KVM_BOOK3S_HV_EXIT_TIMING
+static void __start_timing(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator *next)
+{
+ struct kvmppc_vcore *vc = vcpu->arch.vcore;
+ u64 tb = mftb() - vc->tb_offset_applied;
+
+ vcpu->arch.cur_activity = next;
+ vcpu->arch.cur_tb_start = tb;
+}
+
+static void __accumulate_time(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator *next)
+{
+ struct kvmppc_vcore *vc = vcpu->arch.vcore;
+ struct kvmhv_tb_accumulator *curr;
+ u64 tb = mftb() - vc->tb_offset_applied;
+ u64 prev_tb;
+ u64 delta;
+ u64 seq;
+
+ curr = vcpu->arch.cur_activity;
+ vcpu->arch.cur_activity = next;
+ prev_tb = vcpu->arch.cur_tb_start;
+ vcpu->arch.cur_tb_start = tb;
+
+ if (!curr)
+ return;
+
+ delta = tb - prev_tb;
+
+ seq = curr->seqcount;
+ curr->seqcount = seq + 1;
+ smp_wmb();
+ curr->tb_total += delta;
+ if (seq == 0 || delta < curr->tb_min)
+ curr->tb_min = delta;
+ if (delta > curr->tb_max)
+ curr->tb_max = delta;
+ smp_wmb();
+ curr->seqcount = seq + 2;
+}
+
+#define start_timing(vcpu, next) __start_timing(vcpu, next)
+#define end_timing(vcpu) __start_timing(vcpu, NULL)
+#define accumulate_time(vcpu, next) __accumulate_time(vcpu, next)
+#else
+#define start_timing(vcpu, next) do {} while (0)
+#define end_timing(vcpu) do {} while (0)
+#define accumulate_time(vcpu, next) do {} while (0)
+#endif
+
+static inline void mfslb(unsigned int idx, u64 *slbee, u64 *slbev)
+{
+ asm volatile("slbmfev %0,%1" : "=r" (*slbev) : "r" (idx));
+ asm volatile("slbmfee %0,%1" : "=r" (*slbee) : "r" (idx));
+}
+
+static inline void mtslb(unsigned int idx, u64 slbee, u64 slbev)
+{
+ BUG_ON((slbee & 0xfff) != idx);
+
+ asm volatile("slbmte %0,%1" :: "r" (slbev), "r" (slbee));
+}
+
+static inline void slb_invalidate(unsigned int ih)
+{
+ asm volatile("slbia %0" :: "i"(ih));
+}
+
+/*
+ * Malicious or buggy radix guests may have inserted SLB entries
+ * (only 0..3 because radix always runs with UPRT=1), so these must
+ * be cleared here to avoid side-channels. slbmte is used rather
+ * than slbia, as it won't clear cached translations.
+ */
+static void radix_clear_slb(void)
+{
+ u64 slbee, slbev;
+ int i;
+
+ for (i = 0; i < 4; i++) {
+ mfslb(i, &slbee, &slbev);
+ if (unlikely(slbee || slbev)) {
+ slbee = i;
+ slbev = 0;
+ mtslb(i, slbee, slbev);
+ }
+ }
+}
+
+int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu)
+{
+ u64 *exsave;
+ unsigned long msr = mfmsr();
+ int trap;
+
+ start_timing(vcpu, &vcpu->arch.rm_entry);
+
+ vcpu->arch.ceded = 0;
+
+ WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
+ WARN_ON_ONCE(!(vcpu->arch.shregs.msr & MSR_ME));
+
+ mtspr(SPRN_HSRR0, vcpu->arch.regs.nip);
+ mtspr(SPRN_HSRR1, (vcpu->arch.shregs.msr & ~MSR_HV) | MSR_ME);
+
+ /*
+ * On POWER9 DD2.1 and below, sometimes on a Hypervisor Data Storage
+ * Interrupt (HDSI) the HDSISR is not be updated at all.
+ *
+ * To work around this we put a canary value into the HDSISR before
+ * returning to a guest and then check for this canary when we take a
+ * HDSI. If we find the canary on a HDSI, we know the hardware didn't
+ * update the HDSISR. In this case we return to the guest to retake the
+ * HDSI which should correctly update the HDSISR the second time HDSI
+ * entry.
+ *
+ * Just do this on all p9 processors for now.
+ */
+ mtspr(SPRN_HDSISR, HDSISR_CANARY);
+
+ accumulate_time(vcpu, &vcpu->arch.guest_time);
+
+ local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
+ kvmppc_p9_enter_guest(vcpu);
+ // Radix host and guest means host never runs with guest MMU state
+ local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
+
+ accumulate_time(vcpu, &vcpu->arch.rm_intr);
+
+ /* Get these from r11/12 and paca exsave */
+ vcpu->arch.shregs.srr0 = mfspr(SPRN_SRR0);
+ vcpu->arch.shregs.srr1 = mfspr(SPRN_SRR1);
+ vcpu->arch.shregs.dar = mfspr(SPRN_DAR);
+ vcpu->arch.shregs.dsisr = mfspr(SPRN_DSISR);
+
+ /* 0x2 bit for HSRR is only used by PR and P7/8 HV paths, clear it */
+ trap = local_paca->kvm_hstate.scratch0 & ~0x2;
+ if (likely(trap > BOOK3S_INTERRUPT_MACHINE_CHECK)) {
+ exsave = local_paca->exgen;
+ } else if (trap == BOOK3S_INTERRUPT_SYSTEM_RESET) {
+ exsave = local_paca->exnmi;
+ } else { /* trap == 0x200 */
+ exsave = local_paca->exmc;
+ }
+
+ vcpu->arch.regs.gpr[1] = local_paca->kvm_hstate.scratch1;
+ vcpu->arch.regs.gpr[3] = local_paca->kvm_hstate.scratch2;
+ vcpu->arch.regs.gpr[9] = exsave[EX_R9/sizeof(u64)];
+ vcpu->arch.regs.gpr[10] = exsave[EX_R10/sizeof(u64)];
+ vcpu->arch.regs.gpr[11] = exsave[EX_R11/sizeof(u64)];
+ vcpu->arch.regs.gpr[12] = exsave[EX_R12/sizeof(u64)];
+ vcpu->arch.regs.gpr[13] = exsave[EX_R13/sizeof(u64)];
+ vcpu->arch.ppr = exsave[EX_PPR/sizeof(u64)];
+ vcpu->arch.cfar = exsave[EX_CFAR/sizeof(u64)];
+ vcpu->arch.regs.ctr = exsave[EX_CTR/sizeof(u64)];
+
+ vcpu->arch.last_inst = KVM_INST_FETCH_FAILED;
+
+ if (unlikely(trap == BOOK3S_INTERRUPT_MACHINE_CHECK)) {
+ vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
+ vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
+ kvmppc_realmode_machine_check(vcpu);
+
+ } else if (unlikely(trap == BOOK3S_INTERRUPT_HMI)) {
+ kvmppc_realmode_hmi_handler();
+
+ } else if (trap == BOOK3S_INTERRUPT_H_EMUL_ASSIST) {
+ vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
+
+ } else if (trap == BOOK3S_INTERRUPT_H_DATA_STORAGE) {
+ vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
+ vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
+ vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
+
+ } else if (trap == BOOK3S_INTERRUPT_H_INST_STORAGE) {
+ vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
+
+ } else if (trap == BOOK3S_INTERRUPT_H_FAC_UNAVAIL) {
+ vcpu->arch.hfscr = mfspr(SPRN_HFSCR);
+
+#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
+ /*
+ * Softpatch interrupt for transactional memory emulation cases
+ * on POWER9 DD2.2. This is early in the guest exit path - we
+ * haven't saved registers or done a treclaim yet.
+ */
+ } else if (trap == BOOK3S_INTERRUPT_HV_SOFTPATCH) {
+ vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
+
+ /*
+ * The cases we want to handle here are those where the guest
+ * is in real suspend mode and is trying to transition to
+ * transactional mode.
+ */
+ if (local_paca->kvm_hstate.fake_suspend &&
+ (vcpu->arch.shregs.msr & MSR_TS_S)) {
+ if (kvmhv_p9_tm_emulation_early(vcpu)) {
+ /* Prevent it being handled again. */
+ trap = 0;
+ }
+ }
+#endif
+ }
+
+ radix_clear_slb();
+
+ __mtmsrd(msr, 0);
+ mtspr(SPRN_CTRLT, 1);
+
+ accumulate_time(vcpu, &vcpu->arch.rm_exit);
+
+ end_timing(vcpu);
+
+ return trap;
+}
+EXPORT_SYMBOL_GPL(__kvmhv_vcpu_entry_p9);
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 2d0d14ed1d92..6118e8a97ddd 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -44,9 +44,8 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
#define NAPPING_UNSPLIT 3
/* Stack frame offsets for kvmppc_hv_entry */
-#define SFS 208
+#define SFS 160
#define STACK_SLOT_TRAP (SFS-4)
-#define STACK_SLOT_SHORT_PATH (SFS-8)
#define STACK_SLOT_TID (SFS-16)
#define STACK_SLOT_PSSCR (SFS-24)
#define STACK_SLOT_PID (SFS-32)
@@ -59,8 +58,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
#define STACK_SLOT_UAMOR (SFS-88)
#define STACK_SLOT_DAWR1 (SFS-96)
#define STACK_SLOT_DAWRX1 (SFS-104)
-/* the following is used by the P9 short path */
-#define STACK_SLOT_NVGPRS (SFS-152) /* 18 gprs */
/*
* Call kvmppc_hv_entry in real mode.
@@ -1008,9 +1005,6 @@ END_MMU_FTR_SECTION_IFCLR(MMU_FTR_TYPE_RADIX)
no_xive:
#endif /* CONFIG_KVM_XICS */
- li r0, 0
- stw r0, STACK_SLOT_SHORT_PATH(r1)
-
deliver_guest_interrupt: /* r4 = vcpu, r13 = paca */
/* Check if we can deliver an external or decrementer interrupt now */
ld r0, VCPU_PENDING_EXC(r4)
@@ -1030,7 +1024,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
mtspr SPRN_SRR0, r6
mtspr SPRN_SRR1, r7
-fast_guest_entry_c:
ld r10, VCPU_PC(r4)
ld r11, VCPU_MSR(r4)
/* r11 = vcpu->arch.msr & ~MSR_HV */
@@ -1135,97 +1128,6 @@ ret_to_ultra:
ld r4, VCPU_GPR(R4)(r4)
sc 2
-/*
- * Enter the guest on a P9 or later system where we have exactly
- * one vcpu per vcore and we don't need to go to real mode
- * (which implies that host and guest are both using radix MMU mode).
- * r3 = vcpu pointer
- * Most SPRs and all the VSRs have been loaded already.
- */
-_GLOBAL(__kvmhv_vcpu_entry_p9)
-EXPORT_SYMBOL_GPL(__kvmhv_vcpu_entry_p9)
- mflr r0
- std r0, PPC_LR_STKOFF(r1)
- stdu r1, -SFS(r1)
-
- li r0, 1
- stw r0, STACK_SLOT_SHORT_PATH(r1)
-
- std r3, HSTATE_KVM_VCPU(r13)
- mfcr r4
- stw r4, SFS+8(r1)
-
- std r1, HSTATE_HOST_R1(r13)
-
- reg = 14
- .rept 18
- std reg, STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
- reg = reg + 1
- .endr
-
- reg = 14
- .rept 18
- ld reg, __VCPU_GPR(reg)(r3)
- reg = reg + 1
- .endr
-
- mfmsr r10
- std r10, HSTATE_HOST_MSR(r13)
-
- mr r4, r3
- b fast_guest_entry_c
-guest_exit_short_path:
- /*
- * Malicious or buggy radix guests may have inserted SLB entries
- * (only 0..3 because radix always runs with UPRT=1), so these must
- * be cleared here to avoid side-channels. slbmte is used rather
- * than slbia, as it won't clear cached translations.
- */
- li r0,0
- slbmte r0,r0
- li r4,1
- slbmte r0,r4
- li r4,2
- slbmte r0,r4
- li r4,3
- slbmte r0,r4
-
- li r0, KVM_GUEST_MODE_NONE
- stb r0, HSTATE_IN_GUEST(r13)
-
- reg = 14
- .rept 18
- std reg, __VCPU_GPR(reg)(r9)
- reg = reg + 1
- .endr
-
- reg = 14
- .rept 18
- ld reg, STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
- reg = reg + 1
- .endr
-
- lwz r4, SFS+8(r1)
- mtcr r4
-
- mr r3, r12 /* trap number */
-
- addi r1, r1, SFS
- ld r0, PPC_LR_STKOFF(r1)
- mtlr r0
-
- /* If we are in real mode, do a rfid to get back to the caller */
- mfmsr r4
- andi. r5, r4, MSR_IR
- bnelr
- rldicl r5, r4, 64 - MSR_TS_S_LG, 62 /* extract TS field */
- mtspr SPRN_SRR0, r0
- ld r10, HSTATE_HOST_MSR(r13)
- rldimi r10, r5, MSR_TS_S_LG, 63 - MSR_TS_T_LG
- mtspr SPRN_SRR1, r10
- RFI_TO_KERNEL
- b .
-
secondary_too_late:
li r12, 0
stw r12, STACK_SLOT_TRAP(r1)
@@ -1397,14 +1299,9 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
mr r4,r9
bge fast_guest_return
2:
- /* If we came in through the P9 short path, no real mode hcalls */
- lwz r0, STACK_SLOT_SHORT_PATH(r1)
- cmpwi r0, 0
- bne no_try_real
/* See if this is an hcall we can handle in real mode */
cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
beq hcall_try_real_mode
-no_try_real:
/* Hypervisor doorbell - exit only if host IPI flag set */
cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
@@ -1447,11 +1344,6 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
bl kvmhv_accumulate_time
#endif
#ifdef CONFIG_KVM_XICS
- /* If we came in through the P9 short path, xive pull is done in C */
- lwz r0, STACK_SLOT_SHORT_PATH(r1)
- cmpwi r0, 0
- bne 1f
-
/* We are exiting, pull the VP from the XIVE */
lbz r0, VCPU_XIVE_PUSHED(r9)
cmpwi cr0, r0, 0
@@ -1496,11 +1388,6 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
1: nop
patch_site 1b patch__call_kvm_flush_link_stack
- /* If we came in through the P9 short path, go back out to C now */
- lwz r0, STACK_SLOT_SHORT_PATH(r1)
- cmpwi r0, 0
- bne guest_exit_short_path
-
/* For hash guest, read the guest SLB and save it away */
ld r5, VCPU_KVM(r9)
lbz r0, KVM_RADIX(r5)
@@ -1548,8 +1435,10 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
b guest_bypass
0: /*
- * Sanitise radix guest SLB, see guest_exit_short_path comment.
- * We clear vcpu->arch.slb_max to match earlier behaviour.
+ * Malicious or buggy radix guests may have inserted SLB entries
+ * (only 0..3 because radix always runs with UPRT=1), so these must
+ * be cleared here to avoid side-channels. slbmte is used rather
+ * than slbia, as it won't clear cached translations.
*/
li r0,0
stw r0,VCPU_SLB_MAX(r9)
@@ -3362,7 +3251,7 @@ BEGIN_FTR_SECTION
mtspr SPRN_DAWRX1, r0
END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
- /* Clear hash and radix guest SLB, see guest_exit_short_path comment. */
+ /* Clear hash and radix guest SLB. */
slbmte r0, r0
PPC_SLBIA(6)
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 29/46] KVM: PPC: Book3S HV P9: Implement the rest of the P9 path in C
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Almost all logic is moved to C, by introducing a new in_guest mode that
selects and branches very early in the interrupt handler to the P9 exit
code.
The remaining assembly is only about 160 lines of low level stack setup,
with VCPU vs host register save and restore, plus a small shim to the
legacy paths in the interrupt handler.
There are two motivations for this, the first is just make the code more
maintainable being in C. The second is to reduce the amount of code
running in a special KVM mode, "realmode". I put that in quotes because
with radix it is no longer necessarily real-mode in the MMU, but it
still has to be treated specially because it may be in real-mode, and
has various important registers like PID, DEC, TB, etc set to guest.
This is hostile to the rest of Linux and can't use arbitrary kernel
functionality or be instrumented well.
This initial patch is a reasonably faithful conversion of the asm code.
It does lack any loop to return quickly back into the guest without
switching out of realmode in the case of unimportant or easily handled
interrupts, as explained in the previous change, handling HV interrupts
in real mode is not so important for P9.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/asm-prototypes.h | 3 +-
arch/powerpc/include/asm/kvm_asm.h | 3 +-
arch/powerpc/include/asm/kvm_book3s_64.h | 8 +
arch/powerpc/kernel/security.c | 5 +-
arch/powerpc/kvm/Makefile | 3 +
arch/powerpc/kvm/book3s_64_entry.S | 246 ++++++++++++++++++++++
arch/powerpc/kvm/book3s_hv.c | 9 +-
arch/powerpc/kvm/book3s_hv_interrupt.c | 223 ++++++++++++++++++++
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 123 +----------
9 files changed, 500 insertions(+), 123 deletions(-)
create mode 100644 arch/powerpc/kvm/book3s_hv_interrupt.c
diff --git a/arch/powerpc/include/asm/asm-prototypes.h b/arch/powerpc/include/asm/asm-prototypes.h
index 939f3c94c8f3..7c74c80ed994 100644
--- a/arch/powerpc/include/asm/asm-prototypes.h
+++ b/arch/powerpc/include/asm/asm-prototypes.h
@@ -122,6 +122,7 @@ extern s32 patch__call_flush_branch_caches3;
extern s32 patch__flush_count_cache_return;
extern s32 patch__flush_link_stack_return;
extern s32 patch__call_kvm_flush_link_stack;
+extern s32 patch__call_kvm_flush_link_stack_p9;
extern s32 patch__memset_nocache, patch__memcpy_nocache;
extern long flush_branch_caches;
@@ -142,7 +143,7 @@ void kvmhv_load_host_pmu(void);
void kvmhv_save_guest_pmu(struct kvm_vcpu *vcpu, bool pmu_in_use);
void kvmhv_load_guest_pmu(struct kvm_vcpu *vcpu);
-int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu);
+void kvmppc_p9_enter_guest(struct kvm_vcpu *vcpu);
long kvmppc_h_set_dabr(struct kvm_vcpu *vcpu, unsigned long dabr);
long kvmppc_h_set_xdabr(struct kvm_vcpu *vcpu, unsigned long dabr,
diff --git a/arch/powerpc/include/asm/kvm_asm.h b/arch/powerpc/include/asm/kvm_asm.h
index a3633560493b..b4f9996bd331 100644
--- a/arch/powerpc/include/asm/kvm_asm.h
+++ b/arch/powerpc/include/asm/kvm_asm.h
@@ -146,7 +146,8 @@
#define KVM_GUEST_MODE_GUEST 1
#define KVM_GUEST_MODE_SKIP 2
#define KVM_GUEST_MODE_GUEST_HV 3
-#define KVM_GUEST_MODE_HOST_HV 4
+#define KVM_GUEST_MODE_GUEST_HV_FAST 4 /* ISA v3.0 with host radix mode */
+#define KVM_GUEST_MODE_HOST_HV 5
#define KVM_INST_FETCH_FAILED -1
diff --git a/arch/powerpc/include/asm/kvm_book3s_64.h b/arch/powerpc/include/asm/kvm_book3s_64.h
index 9bb9bb370b53..c214bcffb441 100644
--- a/arch/powerpc/include/asm/kvm_book3s_64.h
+++ b/arch/powerpc/include/asm/kvm_book3s_64.h
@@ -153,9 +153,17 @@ static inline bool kvmhv_vcpu_is_radix(struct kvm_vcpu *vcpu)
return radix;
}
+int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu);
+
#define KVM_DEFAULT_HPT_ORDER 24 /* 16MB HPT by default */
#endif
+/*
+ * Invalid HDSISR value which is used to indicate when HW has not set the reg.
+ * Used to work around an errata.
+ */
+#define HDSISR_CANARY 0x7fff
+
/*
* We use a lock bit in HPTE dword 0 to synchronize updates and
* accesses to each HPTE, and another bit to indicate non-present
diff --git a/arch/powerpc/kernel/security.c b/arch/powerpc/kernel/security.c
index e4e1a94ccf6a..3a607c11f20f 100644
--- a/arch/powerpc/kernel/security.c
+++ b/arch/powerpc/kernel/security.c
@@ -430,16 +430,19 @@ device_initcall(stf_barrier_debugfs_init);
static void update_branch_cache_flush(void)
{
- u32 *site;
+ u32 *site, __maybe_unused *site2;
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
site = &patch__call_kvm_flush_link_stack;
+ site2 = &patch__call_kvm_flush_link_stack_p9;
// This controls the branch from guest_exit_cont to kvm_flush_link_stack
if (link_stack_flush_type = BRANCH_CACHE_FLUSH_NONE) {
patch_instruction_site(site, ppc_inst(PPC_INST_NOP));
+ patch_instruction_site(site2, ppc_inst(PPC_INST_NOP));
} else {
// Could use HW flush, but that could also flush count cache
patch_branch_site(site, (u64)&kvm_flush_link_stack, BRANCH_SET_LINK);
+ patch_branch_site(site2, (u64)&kvm_flush_link_stack, BRANCH_SET_LINK);
}
#endif
diff --git a/arch/powerpc/kvm/Makefile b/arch/powerpc/kvm/Makefile
index cdd119028f64..40a55a5ba4ff 100644
--- a/arch/powerpc/kvm/Makefile
+++ b/arch/powerpc/kvm/Makefile
@@ -43,6 +43,9 @@ kvm-objs-$(CONFIG_KVM_E500MC) := $(kvm-e500mc-objs)
kvm-book3s_64-builtin-objs-$(CONFIG_SPAPR_TCE_IOMMU) := \
book3s_64_vio_hv.o
+kvm-book3s_64-builtin-objs-y += \
+ book3s_hv_interrupt.o
+
kvm-pr-y := \
fpu.o \
emulate.o \
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index de81ab69555b..845df5fefdbd 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -1,11 +1,16 @@
/* SPDX-License-Identifier: GPL-2.0-only */
#include <asm/asm-offsets.h>
#include <asm/cache.h>
+#include <asm/code-patching-asm.h>
#include <asm/exception-64s.h>
+#include <asm/export.h>
#include <asm/kvm_asm.h>
#include <asm/kvm_book3s_asm.h>
+#include <asm/mmu.h>
#include <asm/ppc_asm.h>
+#include <asm/ptrace.h>
#include <asm/reg.h>
+#include <asm/ultravisor-api.h>
/*
* These are branched to from interrupt handlers in exception-64s.S which set
@@ -22,10 +27,15 @@
.global kvmppc_hcall
.balign IFETCH_ALIGN_BYTES
kvmppc_hcall:
+ lbz r10,HSTATE_IN_GUEST(r13)
+ cmpwi r10,KVM_GUEST_MODE_GUEST_HV_FAST
+ beq kvmppc_p9_exit_hcall
ld r10,PACA_EXGEN+EX_R13(r13)
SET_SCRATCH0(r10)
li r10,0xc00
/* Now we look like kvmppc_interrupt */
+ li r11,PACA_EXGEN
+ b 1f
/*
* KVM interrupt entry occurs after GEN_INT_ENTRY runs, and follows that
@@ -46,6 +56,12 @@ kvmppc_hcall:
.global kvmppc_interrupt
.balign IFETCH_ALIGN_BYTES
kvmppc_interrupt:
+ std r10,HSTATE_SCRATCH0(r13)
+ lbz r10,HSTATE_IN_GUEST(r13)
+ cmpwi r10,KVM_GUEST_MODE_GUEST_HV_FAST
+ beq kvmppc_p9_exit_interrupt
+ ld r10,HSTATE_SCRATCH0(r13)
+ lbz r11,HSTATE_IN_GUEST(r13)
li r11,PACA_EXGEN
cmpdi r10,0x200
bgt+ 1f
@@ -140,3 +156,233 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
GET_SCRATCH0(r13)
HRFI_TO_KERNEL
#endif
+
+/* Stack frame offsets for kvmppc_hv_entry */
+#define SFS (144 + STACK_FRAME_MIN_SIZE)
+#define STACK_SLOT_NVGPRS (SFS - 144) /* 18 gprs */
+
+/*
+ * void kvmppc_p9_enter_guest(struct vcpu *vcpu);
+ *
+ * Enter the guest on a ISAv3.0 or later system where we have exactly
+ * one vcpu per vcore, and both the host and guest are radix, and threads
+ * are set to "indepdent mode".
+ */
+.balign IFETCH_ALIGN_BYTES
+_GLOBAL(kvmppc_p9_enter_guest)
+EXPORT_SYMBOL_GPL(kvmppc_p9_enter_guest)
+ mflr r0
+ std r0,PPC_LR_STKOFF(r1)
+ stdu r1,-SFS(r1)
+
+ std r1,HSTATE_HOST_R1(r13)
+
+ mfcr r4
+ stw r4,SFS+8(r1)
+
+ reg = 14
+ .rept 18
+ std reg,STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
+ reg = reg + 1
+ .endr
+
+ ld r4,VCPU_LR(r3)
+ mtlr r4
+ ld r4,VCPU_CTR(r3)
+ mtctr r4
+ ld r4,VCPU_XER(r3)
+ mtspr SPRN_XER,r4
+
+ ld r1,VCPU_CR(r3)
+
+BEGIN_FTR_SECTION
+ ld r4,VCPU_CFAR(r3)
+ mtspr SPRN_CFAR,r4
+END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
+BEGIN_FTR_SECTION
+ ld r4,VCPU_PPR(r3)
+ mtspr SPRN_PPR,r4
+END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
+
+ reg = 4
+ .rept 28
+ ld reg,__VCPU_GPR(reg)(r3)
+ reg = reg + 1
+ .endr
+
+ ld r4,VCPU_KVM(r3)
+ lbz r4,KVM_SECURE_GUEST(r4)
+ cmpdi r4,0
+ ld r4,VCPU_GPR(R4)(r3)
+ bne .Lret_to_ultra
+
+ mtcr r1
+
+ ld r0,VCPU_GPR(R0)(r3)
+ ld r1,VCPU_GPR(R1)(r3)
+ ld r2,VCPU_GPR(R2)(r3)
+ ld r3,VCPU_GPR(R3)(r3)
+
+ HRFI_TO_GUEST
+ b .
+
+ /*
+ * Use UV_RETURN ultracall to return control back to the Ultravisor
+ * after processing an hypercall or interrupt that was forwarded
+ * (a.k.a. reflected) to the Hypervisor.
+ *
+ * All registers have already been reloaded except the ucall requires:
+ * R0 = hcall result
+ * R2 = SRR1, so UV can detect a synthesized interrupt (if any)
+ * R3 = UV_RETURN
+ */
+.Lret_to_ultra:
+ mtcr r1
+ ld r1,VCPU_GPR(R1)(r3)
+
+ ld r0,VCPU_GPR(R3)(r3)
+ mfspr r2,SPRN_SRR1
+ LOAD_REG_IMMEDIATE(r3, UV_RETURN)
+ sc 2
+
+/*
+ * kvmppc_p9_exit_hcall and kvmppc_p9_exit_interrupt are branched to from
+ * above if the interrupt was taken for a guest that was entered via
+ * kvmppc_p9_enter_guest().
+ *
+ * This code recovers the host stack and vcpu pointer, saves all GPRs and
+ * CR, LR, CTR, XER as well as guest MSR and NIA into the VCPU, then re-
+ * establishes the host stack and registers to return from the
+ * kvmppc_p9_enter_guest() function.
+ */
+.balign IFETCH_ALIGN_BYTES
+kvmppc_p9_exit_hcall:
+ mfspr r11,SPRN_SRR0
+ mfspr r12,SPRN_SRR1
+ li r10,0xc00
+ std r10,HSTATE_SCRATCH0(r13)
+
+.balign IFETCH_ALIGN_BYTES
+kvmppc_p9_exit_interrupt:
+ /*
+ * If set to KVM_GUEST_MODE_GUEST_HV_FAST but we're still in the
+ * hypervisor, that means we can't return from the entry stack.
+ */
+ rldicl. r10,r12,64-MSR_HV_LG,63
+ bne- kvmppc_p9_bad_interrupt
+
+ std r1,HSTATE_SCRATCH1(r13)
+ std r3,HSTATE_SCRATCH2(r13)
+ ld r1,HSTATE_HOST_R1(r13)
+ ld r3,HSTATE_KVM_VCPU(r13)
+
+ std r9,VCPU_CR(r3)
+
+1:
+ std r11,VCPU_PC(r3)
+ std r12,VCPU_MSR(r3)
+
+ reg = 14
+ .rept 18
+ std reg,__VCPU_GPR(reg)(r3)
+ reg = reg + 1
+ .endr
+
+ /* r1, r3, r9-r13 are saved to vcpu by C code */
+ std r0,VCPU_GPR(R0)(r3)
+ std r2,VCPU_GPR(R2)(r3)
+ reg = 4
+ .rept 5
+ std reg,__VCPU_GPR(reg)(r3)
+ reg = reg + 1
+ .endr
+
+ ld r2,PACATOC(r13)
+
+ mflr r4
+ std r4,VCPU_LR(r3)
+ mfspr r4,SPRN_XER
+ std r4,VCPU_XER(r3)
+
+ reg = 14
+ .rept 18
+ ld reg,STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
+ reg = reg + 1
+ .endr
+
+ lwz r4,SFS+8(r1)
+ mtcr r4
+
+ /*
+ * Flush the link stack here, before executing the first blr on the
+ * way out of the guest.
+ *
+ * The link stack won't match coming out of the guest anyway so the
+ * only cost is the flush itself. The call clobbers r0.
+ */
+1: nop
+ patch_site 1b patch__call_kvm_flush_link_stack_p9
+
+ addi r1,r1,SFS
+ ld r0,PPC_LR_STKOFF(r1)
+ mtlr r0
+ blr
+
+/*
+ * Took an interrupt somewhere right before HRFID to guest, so registers are
+ * in a bad way. Return things hopefully enough to run host virtual code and
+ * run the Linux interrupt handler (SRESET or MCE) to print something useful.
+ *
+ * We could be really clever and save all host registers in known locations
+ * before setting HSTATE_IN_GUEST, then restoring them all here, and setting
+ * return address to a fixup that sets them up again. But that's a lot of
+ * effort for a small bit of code. Lots of other things to do first.
+ */
+kvmppc_p9_bad_interrupt:
+ /*
+ * Set GUEST_MODE_NONE so the handler won't branch to KVM, and clear
+ * MSR_RI in r12 ([H]SRR1) so the handler won't try to return.
+ */
+ li r10,KVM_GUEST_MODE_NONE
+ stb r10,HSTATE_IN_GUEST(r13)
+ li r10,MSR_RI
+ andc r12,r12,r10
+
+ /*
+ * Clean up guest registers to give host a chance to run.
+ */
+ li r10,0
+ mtspr SPRN_AMR,r10
+ mtspr SPRN_IAMR,r10
+ mtspr SPRN_CIABR,r10
+ mtspr SPRN_DAWRX0,r10
+BEGIN_FTR_SECTION
+ mtspr SPRN_DAWRX1,r10
+END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
+ mtspr SPRN_PID,r10
+
+ /*
+ * Switch to host MMU mode
+ */
+ ld r10, HSTATE_KVM_VCPU(r13)
+ ld r10, VCPU_KVM(r10)
+ lwz r10, KVM_HOST_LPID(r10)
+ mtspr SPRN_LPID,r10
+
+ ld r10, HSTATE_KVM_VCPU(r13)
+ ld r10, VCPU_KVM(r10)
+ ld r10, KVM_HOST_LPCR(r10)
+ mtspr SPRN_LPCR,r10
+
+ /*
+ * Go back to interrupt handler
+ */
+ ld r10,HSTATE_SCRATCH0(r13)
+ cmpwi r10,BOOK3S_INTERRUPT_MACHINE_CHECK
+ beq machine_check_common
+
+ ld r10,HSTATE_SCRATCH0(r13)
+ cmpwi r10,BOOK3S_INTERRUPT_SYSTEM_RESET
+ beq system_reset_common
+
+ b .
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 989a1ff5ad11..6dca639ed997 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -1431,6 +1431,8 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
*/
case BOOK3S_INTERRUPT_H_DATA_STORAGE:
r = RESUME_PAGE_FAULT;
+ if (vcpu->arch.fault_dsisr = HDSISR_CANARY)
+ r = RESUME_GUEST; /* Just retry if it's the canary */
break;
case BOOK3S_INTERRUPT_H_INST_STORAGE:
vcpu->arch.fault_dar = kvmppc_get_pc(vcpu);
@@ -3690,6 +3692,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
u64 tb, next_timer;
int trap, save_pmu;
+ WARN_ON_ONCE(vcpu->arch.ceded);
+
tb = mftb();
next_timer = timer_get_next_tb();
if (tb >= next_timer)
@@ -3698,8 +3702,6 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
if (next_timer < time_limit)
time_limit = next_timer;
- vcpu->arch.ceded = 0;
-
kvmhv_save_host_pmu(); /* saves it to PACA kvm_hstate */
kvmppc_subcore_enter_guest();
@@ -3826,9 +3828,10 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
}
}
kvmppc_xive_pull_vcpu(vcpu);
+
+ vcpu->arch.slb_max = 0;
}
- vcpu->arch.slb_max = 0;
dec = mfspr(SPRN_DEC);
if (!(lpcr & LPCR_LD)) /* Sign extend if not using large decrementer */
dec = (s32) dec;
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
new file mode 100644
index 000000000000..3151b3d62c01
--- /dev/null
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -0,0 +1,223 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include <linux/kernel.h>
+#include <linux/kvm_host.h>
+#include <asm/asm-prototypes.h>
+#include <asm/dbell.h>
+#include <asm/kvm_ppc.h>
+
+#ifdef CONFIG_KVM_BOOK3S_HV_EXIT_TIMING
+static void __start_timing(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator *next)
+{
+ struct kvmppc_vcore *vc = vcpu->arch.vcore;
+ u64 tb = mftb() - vc->tb_offset_applied;
+
+ vcpu->arch.cur_activity = next;
+ vcpu->arch.cur_tb_start = tb;
+}
+
+static void __accumulate_time(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator *next)
+{
+ struct kvmppc_vcore *vc = vcpu->arch.vcore;
+ struct kvmhv_tb_accumulator *curr;
+ u64 tb = mftb() - vc->tb_offset_applied;
+ u64 prev_tb;
+ u64 delta;
+ u64 seq;
+
+ curr = vcpu->arch.cur_activity;
+ vcpu->arch.cur_activity = next;
+ prev_tb = vcpu->arch.cur_tb_start;
+ vcpu->arch.cur_tb_start = tb;
+
+ if (!curr)
+ return;
+
+ delta = tb - prev_tb;
+
+ seq = curr->seqcount;
+ curr->seqcount = seq + 1;
+ smp_wmb();
+ curr->tb_total += delta;
+ if (seq = 0 || delta < curr->tb_min)
+ curr->tb_min = delta;
+ if (delta > curr->tb_max)
+ curr->tb_max = delta;
+ smp_wmb();
+ curr->seqcount = seq + 2;
+}
+
+#define start_timing(vcpu, next) __start_timing(vcpu, next)
+#define end_timing(vcpu) __start_timing(vcpu, NULL)
+#define accumulate_time(vcpu, next) __accumulate_time(vcpu, next)
+#else
+#define start_timing(vcpu, next) do {} while (0)
+#define end_timing(vcpu) do {} while (0)
+#define accumulate_time(vcpu, next) do {} while (0)
+#endif
+
+static inline void mfslb(unsigned int idx, u64 *slbee, u64 *slbev)
+{
+ asm volatile("slbmfev %0,%1" : "=r" (*slbev) : "r" (idx));
+ asm volatile("slbmfee %0,%1" : "=r" (*slbee) : "r" (idx));
+}
+
+static inline void mtslb(unsigned int idx, u64 slbee, u64 slbev)
+{
+ BUG_ON((slbee & 0xfff) != idx);
+
+ asm volatile("slbmte %0,%1" :: "r" (slbev), "r" (slbee));
+}
+
+static inline void slb_invalidate(unsigned int ih)
+{
+ asm volatile("slbia %0" :: "i"(ih));
+}
+
+/*
+ * Malicious or buggy radix guests may have inserted SLB entries
+ * (only 0..3 because radix always runs with UPRT=1), so these must
+ * be cleared here to avoid side-channels. slbmte is used rather
+ * than slbia, as it won't clear cached translations.
+ */
+static void radix_clear_slb(void)
+{
+ u64 slbee, slbev;
+ int i;
+
+ for (i = 0; i < 4; i++) {
+ mfslb(i, &slbee, &slbev);
+ if (unlikely(slbee || slbev)) {
+ slbee = i;
+ slbev = 0;
+ mtslb(i, slbee, slbev);
+ }
+ }
+}
+
+int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu)
+{
+ u64 *exsave;
+ unsigned long msr = mfmsr();
+ int trap;
+
+ start_timing(vcpu, &vcpu->arch.rm_entry);
+
+ vcpu->arch.ceded = 0;
+
+ WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
+ WARN_ON_ONCE(!(vcpu->arch.shregs.msr & MSR_ME));
+
+ mtspr(SPRN_HSRR0, vcpu->arch.regs.nip);
+ mtspr(SPRN_HSRR1, (vcpu->arch.shregs.msr & ~MSR_HV) | MSR_ME);
+
+ /*
+ * On POWER9 DD2.1 and below, sometimes on a Hypervisor Data Storage
+ * Interrupt (HDSI) the HDSISR is not be updated at all.
+ *
+ * To work around this we put a canary value into the HDSISR before
+ * returning to a guest and then check for this canary when we take a
+ * HDSI. If we find the canary on a HDSI, we know the hardware didn't
+ * update the HDSISR. In this case we return to the guest to retake the
+ * HDSI which should correctly update the HDSISR the second time HDSI
+ * entry.
+ *
+ * Just do this on all p9 processors for now.
+ */
+ mtspr(SPRN_HDSISR, HDSISR_CANARY);
+
+ accumulate_time(vcpu, &vcpu->arch.guest_time);
+
+ local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
+ kvmppc_p9_enter_guest(vcpu);
+ // Radix host and guest means host never runs with guest MMU state
+ local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
+
+ accumulate_time(vcpu, &vcpu->arch.rm_intr);
+
+ /* Get these from r11/12 and paca exsave */
+ vcpu->arch.shregs.srr0 = mfspr(SPRN_SRR0);
+ vcpu->arch.shregs.srr1 = mfspr(SPRN_SRR1);
+ vcpu->arch.shregs.dar = mfspr(SPRN_DAR);
+ vcpu->arch.shregs.dsisr = mfspr(SPRN_DSISR);
+
+ /* 0x2 bit for HSRR is only used by PR and P7/8 HV paths, clear it */
+ trap = local_paca->kvm_hstate.scratch0 & ~0x2;
+ if (likely(trap > BOOK3S_INTERRUPT_MACHINE_CHECK)) {
+ exsave = local_paca->exgen;
+ } else if (trap = BOOK3S_INTERRUPT_SYSTEM_RESET) {
+ exsave = local_paca->exnmi;
+ } else { /* trap = 0x200 */
+ exsave = local_paca->exmc;
+ }
+
+ vcpu->arch.regs.gpr[1] = local_paca->kvm_hstate.scratch1;
+ vcpu->arch.regs.gpr[3] = local_paca->kvm_hstate.scratch2;
+ vcpu->arch.regs.gpr[9] = exsave[EX_R9/sizeof(u64)];
+ vcpu->arch.regs.gpr[10] = exsave[EX_R10/sizeof(u64)];
+ vcpu->arch.regs.gpr[11] = exsave[EX_R11/sizeof(u64)];
+ vcpu->arch.regs.gpr[12] = exsave[EX_R12/sizeof(u64)];
+ vcpu->arch.regs.gpr[13] = exsave[EX_R13/sizeof(u64)];
+ vcpu->arch.ppr = exsave[EX_PPR/sizeof(u64)];
+ vcpu->arch.cfar = exsave[EX_CFAR/sizeof(u64)];
+ vcpu->arch.regs.ctr = exsave[EX_CTR/sizeof(u64)];
+
+ vcpu->arch.last_inst = KVM_INST_FETCH_FAILED;
+
+ if (unlikely(trap = BOOK3S_INTERRUPT_MACHINE_CHECK)) {
+ vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
+ vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
+ kvmppc_realmode_machine_check(vcpu);
+
+ } else if (unlikely(trap = BOOK3S_INTERRUPT_HMI)) {
+ kvmppc_realmode_hmi_handler();
+
+ } else if (trap = BOOK3S_INTERRUPT_H_EMUL_ASSIST) {
+ vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
+
+ } else if (trap = BOOK3S_INTERRUPT_H_DATA_STORAGE) {
+ vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
+ vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
+ vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
+
+ } else if (trap = BOOK3S_INTERRUPT_H_INST_STORAGE) {
+ vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
+
+ } else if (trap = BOOK3S_INTERRUPT_H_FAC_UNAVAIL) {
+ vcpu->arch.hfscr = mfspr(SPRN_HFSCR);
+
+#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
+ /*
+ * Softpatch interrupt for transactional memory emulation cases
+ * on POWER9 DD2.2. This is early in the guest exit path - we
+ * haven't saved registers or done a treclaim yet.
+ */
+ } else if (trap = BOOK3S_INTERRUPT_HV_SOFTPATCH) {
+ vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
+
+ /*
+ * The cases we want to handle here are those where the guest
+ * is in real suspend mode and is trying to transition to
+ * transactional mode.
+ */
+ if (local_paca->kvm_hstate.fake_suspend &&
+ (vcpu->arch.shregs.msr & MSR_TS_S)) {
+ if (kvmhv_p9_tm_emulation_early(vcpu)) {
+ /* Prevent it being handled again. */
+ trap = 0;
+ }
+ }
+#endif
+ }
+
+ radix_clear_slb();
+
+ __mtmsrd(msr, 0);
+ mtspr(SPRN_CTRLT, 1);
+
+ accumulate_time(vcpu, &vcpu->arch.rm_exit);
+
+ end_timing(vcpu);
+
+ return trap;
+}
+EXPORT_SYMBOL_GPL(__kvmhv_vcpu_entry_p9);
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 2d0d14ed1d92..6118e8a97ddd 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -44,9 +44,8 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
#define NAPPING_UNSPLIT 3
/* Stack frame offsets for kvmppc_hv_entry */
-#define SFS 208
+#define SFS 160
#define STACK_SLOT_TRAP (SFS-4)
-#define STACK_SLOT_SHORT_PATH (SFS-8)
#define STACK_SLOT_TID (SFS-16)
#define STACK_SLOT_PSSCR (SFS-24)
#define STACK_SLOT_PID (SFS-32)
@@ -59,8 +58,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
#define STACK_SLOT_UAMOR (SFS-88)
#define STACK_SLOT_DAWR1 (SFS-96)
#define STACK_SLOT_DAWRX1 (SFS-104)
-/* the following is used by the P9 short path */
-#define STACK_SLOT_NVGPRS (SFS-152) /* 18 gprs */
/*
* Call kvmppc_hv_entry in real mode.
@@ -1008,9 +1005,6 @@ END_MMU_FTR_SECTION_IFCLR(MMU_FTR_TYPE_RADIX)
no_xive:
#endif /* CONFIG_KVM_XICS */
- li r0, 0
- stw r0, STACK_SLOT_SHORT_PATH(r1)
-
deliver_guest_interrupt: /* r4 = vcpu, r13 = paca */
/* Check if we can deliver an external or decrementer interrupt now */
ld r0, VCPU_PENDING_EXC(r4)
@@ -1030,7 +1024,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
mtspr SPRN_SRR0, r6
mtspr SPRN_SRR1, r7
-fast_guest_entry_c:
ld r10, VCPU_PC(r4)
ld r11, VCPU_MSR(r4)
/* r11 = vcpu->arch.msr & ~MSR_HV */
@@ -1135,97 +1128,6 @@ ret_to_ultra:
ld r4, VCPU_GPR(R4)(r4)
sc 2
-/*
- * Enter the guest on a P9 or later system where we have exactly
- * one vcpu per vcore and we don't need to go to real mode
- * (which implies that host and guest are both using radix MMU mode).
- * r3 = vcpu pointer
- * Most SPRs and all the VSRs have been loaded already.
- */
-_GLOBAL(__kvmhv_vcpu_entry_p9)
-EXPORT_SYMBOL_GPL(__kvmhv_vcpu_entry_p9)
- mflr r0
- std r0, PPC_LR_STKOFF(r1)
- stdu r1, -SFS(r1)
-
- li r0, 1
- stw r0, STACK_SLOT_SHORT_PATH(r1)
-
- std r3, HSTATE_KVM_VCPU(r13)
- mfcr r4
- stw r4, SFS+8(r1)
-
- std r1, HSTATE_HOST_R1(r13)
-
- reg = 14
- .rept 18
- std reg, STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
- reg = reg + 1
- .endr
-
- reg = 14
- .rept 18
- ld reg, __VCPU_GPR(reg)(r3)
- reg = reg + 1
- .endr
-
- mfmsr r10
- std r10, HSTATE_HOST_MSR(r13)
-
- mr r4, r3
- b fast_guest_entry_c
-guest_exit_short_path:
- /*
- * Malicious or buggy radix guests may have inserted SLB entries
- * (only 0..3 because radix always runs with UPRT=1), so these must
- * be cleared here to avoid side-channels. slbmte is used rather
- * than slbia, as it won't clear cached translations.
- */
- li r0,0
- slbmte r0,r0
- li r4,1
- slbmte r0,r4
- li r4,2
- slbmte r0,r4
- li r4,3
- slbmte r0,r4
-
- li r0, KVM_GUEST_MODE_NONE
- stb r0, HSTATE_IN_GUEST(r13)
-
- reg = 14
- .rept 18
- std reg, __VCPU_GPR(reg)(r9)
- reg = reg + 1
- .endr
-
- reg = 14
- .rept 18
- ld reg, STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
- reg = reg + 1
- .endr
-
- lwz r4, SFS+8(r1)
- mtcr r4
-
- mr r3, r12 /* trap number */
-
- addi r1, r1, SFS
- ld r0, PPC_LR_STKOFF(r1)
- mtlr r0
-
- /* If we are in real mode, do a rfid to get back to the caller */
- mfmsr r4
- andi. r5, r4, MSR_IR
- bnelr
- rldicl r5, r4, 64 - MSR_TS_S_LG, 62 /* extract TS field */
- mtspr SPRN_SRR0, r0
- ld r10, HSTATE_HOST_MSR(r13)
- rldimi r10, r5, MSR_TS_S_LG, 63 - MSR_TS_T_LG
- mtspr SPRN_SRR1, r10
- RFI_TO_KERNEL
- b .
-
secondary_too_late:
li r12, 0
stw r12, STACK_SLOT_TRAP(r1)
@@ -1397,14 +1299,9 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
mr r4,r9
bge fast_guest_return
2:
- /* If we came in through the P9 short path, no real mode hcalls */
- lwz r0, STACK_SLOT_SHORT_PATH(r1)
- cmpwi r0, 0
- bne no_try_real
/* See if this is an hcall we can handle in real mode */
cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
beq hcall_try_real_mode
-no_try_real:
/* Hypervisor doorbell - exit only if host IPI flag set */
cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
@@ -1447,11 +1344,6 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
bl kvmhv_accumulate_time
#endif
#ifdef CONFIG_KVM_XICS
- /* If we came in through the P9 short path, xive pull is done in C */
- lwz r0, STACK_SLOT_SHORT_PATH(r1)
- cmpwi r0, 0
- bne 1f
-
/* We are exiting, pull the VP from the XIVE */
lbz r0, VCPU_XIVE_PUSHED(r9)
cmpwi cr0, r0, 0
@@ -1496,11 +1388,6 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
1: nop
patch_site 1b patch__call_kvm_flush_link_stack
- /* If we came in through the P9 short path, go back out to C now */
- lwz r0, STACK_SLOT_SHORT_PATH(r1)
- cmpwi r0, 0
- bne guest_exit_short_path
-
/* For hash guest, read the guest SLB and save it away */
ld r5, VCPU_KVM(r9)
lbz r0, KVM_RADIX(r5)
@@ -1548,8 +1435,10 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
b guest_bypass
0: /*
- * Sanitise radix guest SLB, see guest_exit_short_path comment.
- * We clear vcpu->arch.slb_max to match earlier behaviour.
+ * Malicious or buggy radix guests may have inserted SLB entries
+ * (only 0..3 because radix always runs with UPRT=1), so these must
+ * be cleared here to avoid side-channels. slbmte is used rather
+ * than slbia, as it won't clear cached translations.
*/
li r0,0
stw r0,VCPU_SLB_MAX(r9)
@@ -3362,7 +3251,7 @@ BEGIN_FTR_SECTION
mtspr SPRN_DAWRX1, r0
END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
- /* Clear hash and radix guest SLB, see guest_exit_short_path comment. */
+ /* Clear hash and radix guest SLB. */
slbmte r0, r0
PPC_SLBIA(6)
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* Re: [PATCH v4 29/46] KVM: PPC: Book3S HV P9: Implement the rest of the P9 path in C
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-04-01 5:30 ` Alexey Kardashevskiy
-1 siblings, 0 replies; 188+ messages in thread
From: Alexey Kardashevskiy @ 2021-04-01 5:30 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 3/23/21 12:02 PM, Nicholas Piggin wrote:
> Almost all logic is moved to C, by introducing a new in_guest mode that
> selects and branches very early in the interrupt handler to the P9 exit
> code.
>
> The remaining assembly is only about 160 lines of low level stack setup,
> with VCPU vs host register save and restore, plus a small shim to the
> legacy paths in the interrupt handler.
>
> There are two motivations for this, the first is just make the code more
> maintainable being in C. The second is to reduce the amount of code
> running in a special KVM mode, "realmode". I put that in quotes because
> with radix it is no longer necessarily real-mode in the MMU, but it
> still has to be treated specially because it may be in real-mode, and
> has various important registers like PID, DEC, TB, etc set to guest.
> This is hostile to the rest of Linux and can't use arbitrary kernel
> functionality or be instrumented well.
>
> This initial patch is a reasonably faithful conversion of the asm code.
> It does lack any loop to return quickly back into the guest without
> switching out of realmode in the case of unimportant or easily handled
> interrupts, as explained in the previous change, handling HV interrupts
> in real mode is not so important for P9.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
> arch/powerpc/include/asm/asm-prototypes.h | 3 +-
> arch/powerpc/include/asm/kvm_asm.h | 3 +-
> arch/powerpc/include/asm/kvm_book3s_64.h | 8 +
> arch/powerpc/kernel/security.c | 5 +-
> arch/powerpc/kvm/Makefile | 3 +
> arch/powerpc/kvm/book3s_64_entry.S | 246 ++++++++++++++++++++++
> arch/powerpc/kvm/book3s_hv.c | 9 +-
> arch/powerpc/kvm/book3s_hv_interrupt.c | 223 ++++++++++++++++++++
> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 123 +----------
> 9 files changed, 500 insertions(+), 123 deletions(-)
> create mode 100644 arch/powerpc/kvm/book3s_hv_interrupt.c
>
> diff --git a/arch/powerpc/include/asm/asm-prototypes.h b/arch/powerpc/include/asm/asm-prototypes.h
> index 939f3c94c8f3..7c74c80ed994 100644
> --- a/arch/powerpc/include/asm/asm-prototypes.h
> +++ b/arch/powerpc/include/asm/asm-prototypes.h
> @@ -122,6 +122,7 @@ extern s32 patch__call_flush_branch_caches3;
> extern s32 patch__flush_count_cache_return;
> extern s32 patch__flush_link_stack_return;
> extern s32 patch__call_kvm_flush_link_stack;
> +extern s32 patch__call_kvm_flush_link_stack_p9;
> extern s32 patch__memset_nocache, patch__memcpy_nocache;
>
> extern long flush_branch_caches;
> @@ -142,7 +143,7 @@ void kvmhv_load_host_pmu(void);
> void kvmhv_save_guest_pmu(struct kvm_vcpu *vcpu, bool pmu_in_use);
> void kvmhv_load_guest_pmu(struct kvm_vcpu *vcpu);
>
> -int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu);
> +void kvmppc_p9_enter_guest(struct kvm_vcpu *vcpu);
>
> long kvmppc_h_set_dabr(struct kvm_vcpu *vcpu, unsigned long dabr);
> long kvmppc_h_set_xdabr(struct kvm_vcpu *vcpu, unsigned long dabr,
> diff --git a/arch/powerpc/include/asm/kvm_asm.h b/arch/powerpc/include/asm/kvm_asm.h
> index a3633560493b..b4f9996bd331 100644
> --- a/arch/powerpc/include/asm/kvm_asm.h
> +++ b/arch/powerpc/include/asm/kvm_asm.h
> @@ -146,7 +146,8 @@
> #define KVM_GUEST_MODE_GUEST 1
> #define KVM_GUEST_MODE_SKIP 2
> #define KVM_GUEST_MODE_GUEST_HV 3
> -#define KVM_GUEST_MODE_HOST_HV 4
> +#define KVM_GUEST_MODE_GUEST_HV_FAST 4 /* ISA v3.0 with host radix mode */
> +#define KVM_GUEST_MODE_HOST_HV 5
>
> #define KVM_INST_FETCH_FAILED -1
>
> diff --git a/arch/powerpc/include/asm/kvm_book3s_64.h b/arch/powerpc/include/asm/kvm_book3s_64.h
> index 9bb9bb370b53..c214bcffb441 100644
> --- a/arch/powerpc/include/asm/kvm_book3s_64.h
> +++ b/arch/powerpc/include/asm/kvm_book3s_64.h
> @@ -153,9 +153,17 @@ static inline bool kvmhv_vcpu_is_radix(struct kvm_vcpu *vcpu)
> return radix;
> }
>
> +int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu);
> +
> #define KVM_DEFAULT_HPT_ORDER 24 /* 16MB HPT by default */
> #endif
>
> +/*
> + * Invalid HDSISR value which is used to indicate when HW has not set the reg.
> + * Used to work around an errata.
> + */
> +#define HDSISR_CANARY 0x7fff
> +
> /*
> * We use a lock bit in HPTE dword 0 to synchronize updates and
> * accesses to each HPTE, and another bit to indicate non-present
> diff --git a/arch/powerpc/kernel/security.c b/arch/powerpc/kernel/security.c
> index e4e1a94ccf6a..3a607c11f20f 100644
> --- a/arch/powerpc/kernel/security.c
> +++ b/arch/powerpc/kernel/security.c
> @@ -430,16 +430,19 @@ device_initcall(stf_barrier_debugfs_init);
>
> static void update_branch_cache_flush(void)
> {
> - u32 *site;
> + u32 *site, __maybe_unused *site2;
>
> #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
> site = &patch__call_kvm_flush_link_stack;
> + site2 = &patch__call_kvm_flush_link_stack_p9;
> // This controls the branch from guest_exit_cont to kvm_flush_link_stack
> if (link_stack_flush_type == BRANCH_CACHE_FLUSH_NONE) {
> patch_instruction_site(site, ppc_inst(PPC_INST_NOP));
> + patch_instruction_site(site2, ppc_inst(PPC_INST_NOP));
> } else {
> // Could use HW flush, but that could also flush count cache
> patch_branch_site(site, (u64)&kvm_flush_link_stack, BRANCH_SET_LINK);
> + patch_branch_site(site2, (u64)&kvm_flush_link_stack, BRANCH_SET_LINK);
> }
> #endif
>
> diff --git a/arch/powerpc/kvm/Makefile b/arch/powerpc/kvm/Makefile
> index cdd119028f64..40a55a5ba4ff 100644
> --- a/arch/powerpc/kvm/Makefile
> +++ b/arch/powerpc/kvm/Makefile
> @@ -43,6 +43,9 @@ kvm-objs-$(CONFIG_KVM_E500MC) := $(kvm-e500mc-objs)
> kvm-book3s_64-builtin-objs-$(CONFIG_SPAPR_TCE_IOMMU) := \
> book3s_64_vio_hv.o
>
> +kvm-book3s_64-builtin-objs-y += \
> + book3s_hv_interrupt.o
> +
> kvm-pr-y := \
> fpu.o \
> emulate.o \
> diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
> index de81ab69555b..845df5fefdbd 100644
> --- a/arch/powerpc/kvm/book3s_64_entry.S
> +++ b/arch/powerpc/kvm/book3s_64_entry.S
> @@ -1,11 +1,16 @@
> /* SPDX-License-Identifier: GPL-2.0-only */
> #include <asm/asm-offsets.h>
> #include <asm/cache.h>
> +#include <asm/code-patching-asm.h>
> #include <asm/exception-64s.h>
> +#include <asm/export.h>
> #include <asm/kvm_asm.h>
> #include <asm/kvm_book3s_asm.h>
> +#include <asm/mmu.h>
> #include <asm/ppc_asm.h>
> +#include <asm/ptrace.h>
> #include <asm/reg.h>
> +#include <asm/ultravisor-api.h>
>
> /*
> * These are branched to from interrupt handlers in exception-64s.S which set
> @@ -22,10 +27,15 @@
> .global kvmppc_hcall
> .balign IFETCH_ALIGN_BYTES
> kvmppc_hcall:
> + lbz r10,HSTATE_IN_GUEST(r13)
> + cmpwi r10,KVM_GUEST_MODE_GUEST_HV_FAST
> + beq kvmppc_p9_exit_hcall
> ld r10,PACA_EXGEN+EX_R13(r13)
> SET_SCRATCH0(r10)
> li r10,0xc00
> /* Now we look like kvmppc_interrupt */
> + li r11,PACA_EXGEN
> + b 1f
>
> /*
> * KVM interrupt entry occurs after GEN_INT_ENTRY runs, and follows that
> @@ -46,6 +56,12 @@ kvmppc_hcall:
> .global kvmppc_interrupt
> .balign IFETCH_ALIGN_BYTES
> kvmppc_interrupt:
> + std r10,HSTATE_SCRATCH0(r13)
> + lbz r10,HSTATE_IN_GUEST(r13)
> + cmpwi r10,KVM_GUEST_MODE_GUEST_HV_FAST
> + beq kvmppc_p9_exit_interrupt
> + ld r10,HSTATE_SCRATCH0(r13)
> + lbz r11,HSTATE_IN_GUEST(r13)
> li r11,PACA_EXGEN
> cmpdi r10,0x200
> bgt+ 1f
> @@ -140,3 +156,233 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
> GET_SCRATCH0(r13)
> HRFI_TO_KERNEL
> #endif
> +
> +/* Stack frame offsets for kvmppc_hv_entry */
> +#define SFS (144 + STACK_FRAME_MIN_SIZE)
> +#define STACK_SLOT_NVGPRS (SFS - 144) /* 18 gprs */
> +
> +/*
> + * void kvmppc_p9_enter_guest(struct vcpu *vcpu);
> + *
> + * Enter the guest on a ISAv3.0 or later system where we have exactly
> + * one vcpu per vcore, and both the host and guest are radix, and threads
> + * are set to "indepdent mode".
> + */
> +.balign IFETCH_ALIGN_BYTES
> +_GLOBAL(kvmppc_p9_enter_guest)
> +EXPORT_SYMBOL_GPL(kvmppc_p9_enter_guest)
> + mflr r0
> + std r0,PPC_LR_STKOFF(r1)
> + stdu r1,-SFS(r1)
> +
> + std r1,HSTATE_HOST_R1(r13)
> +
> + mfcr r4
> + stw r4,SFS+8(r1)
> +
> + reg = 14
> + .rept 18
> + std reg,STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
> + reg = reg + 1
> + .endr
> +
> + ld r4,VCPU_LR(r3)
> + mtlr r4
> + ld r4,VCPU_CTR(r3)
> + mtctr r4
> + ld r4,VCPU_XER(r3)
> + mtspr SPRN_XER,r4
> +
> + ld r1,VCPU_CR(r3)
> +
> +BEGIN_FTR_SECTION
> + ld r4,VCPU_CFAR(r3)
> + mtspr SPRN_CFAR,r4
> +END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
> +BEGIN_FTR_SECTION
> + ld r4,VCPU_PPR(r3)
> + mtspr SPRN_PPR,r4
> +END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
> +
> + reg = 4
> + .rept 28
> + ld reg,__VCPU_GPR(reg)(r3)
> + reg = reg + 1
> + .endr
> +
> + ld r4,VCPU_KVM(r3)
> + lbz r4,KVM_SECURE_GUEST(r4)
> + cmpdi r4,0
> + ld r4,VCPU_GPR(R4)(r3)
> + bne .Lret_to_ultra
> +
> + mtcr r1
> +
> + ld r0,VCPU_GPR(R0)(r3)
> + ld r1,VCPU_GPR(R1)(r3)
> + ld r2,VCPU_GPR(R2)(r3)
> + ld r3,VCPU_GPR(R3)(r3)
> +
> + HRFI_TO_GUEST
> + b .
> +
> + /*
> + * Use UV_RETURN ultracall to return control back to the Ultravisor
> + * after processing an hypercall or interrupt that was forwarded
> + * (a.k.a. reflected) to the Hypervisor.
> + *
> + * All registers have already been reloaded except the ucall requires:
> + * R0 = hcall result
> + * R2 = SRR1, so UV can detect a synthesized interrupt (if any)
> + * R3 = UV_RETURN
> + */
> +.Lret_to_ultra:
> + mtcr r1
> + ld r1,VCPU_GPR(R1)(r3)
> +
> + ld r0,VCPU_GPR(R3)(r3)
> + mfspr r2,SPRN_SRR1
> + LOAD_REG_IMMEDIATE(r3, UV_RETURN)
> + sc 2
> +
> +/*
> + * kvmppc_p9_exit_hcall and kvmppc_p9_exit_interrupt are branched to from
> + * above if the interrupt was taken for a guest that was entered via
> + * kvmppc_p9_enter_guest().
> + *
> + * This code recovers the host stack and vcpu pointer, saves all GPRs and
> + * CR, LR, CTR, XER as well as guest MSR and NIA into the VCPU, then re-
> + * establishes the host stack and registers to return from the
> + * kvmppc_p9_enter_guest() function.
What does "this code" refer to? If it is the asm below, then it does not
save CTR, it is in the c code. Otherwise it is confusing (to me) :)
> + */
> +.balign IFETCH_ALIGN_BYTES
> +kvmppc_p9_exit_hcall:
> + mfspr r11,SPRN_SRR0
> + mfspr r12,SPRN_SRR1
> + li r10,0xc00
> + std r10,HSTATE_SCRATCH0(r13)
> +
> +.balign IFETCH_ALIGN_BYTES
> +kvmppc_p9_exit_interrupt:
> + /*
> + * If set to KVM_GUEST_MODE_GUEST_HV_FAST but we're still in the
> + * hypervisor, that means we can't return from the entry stack.
> + */
> + rldicl. r10,r12,64-MSR_HV_LG,63
> + bne- kvmppc_p9_bad_interrupt
> +
> + std r1,HSTATE_SCRATCH1(r13)
> + std r3,HSTATE_SCRATCH2(r13)
> + ld r1,HSTATE_HOST_R1(r13)
> + ld r3,HSTATE_KVM_VCPU(r13)
> +
> + std r9,VCPU_CR(r3)
> +
> +1:
> + std r11,VCPU_PC(r3)
> + std r12,VCPU_MSR(r3)
> +
> + reg = 14
> + .rept 18
> + std reg,__VCPU_GPR(reg)(r3)
> + reg = reg + 1
> + .endr
> +
> + /* r1, r3, r9-r13 are saved to vcpu by C code */
> + std r0,VCPU_GPR(R0)(r3)
> + std r2,VCPU_GPR(R2)(r3)
> + reg = 4
> + .rept 5
> + std reg,__VCPU_GPR(reg)(r3)
> + reg = reg + 1
> + .endr
> +
> + ld r2,PACATOC(r13)
> +
> + mflr r4
> + std r4,VCPU_LR(r3)
> + mfspr r4,SPRN_XER
> + std r4,VCPU_XER(r3)
> +
> + reg = 14
> + .rept 18
> + ld reg,STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
> + reg = reg + 1
> + .endr
> +
> + lwz r4,SFS+8(r1)
> + mtcr r4
> +
> + /*
> + * Flush the link stack here, before executing the first blr on the
> + * way out of the guest.
> + *
> + * The link stack won't match coming out of the guest anyway so the
> + * only cost is the flush itself. The call clobbers r0.
> + */
> +1: nop
> + patch_site 1b patch__call_kvm_flush_link_stack_p9
> +
> + addi r1,r1,SFS
> + ld r0,PPC_LR_STKOFF(r1)
> + mtlr r0
> + blr
> +
> +/*
> + * Took an interrupt somewhere right before HRFID to guest, so registers are
> + * in a bad way. Return things hopefully enough to run host virtual code and
> + * run the Linux interrupt handler (SRESET or MCE) to print something useful.
> + *
> + * We could be really clever and save all host registers in known locations
> + * before setting HSTATE_IN_GUEST, then restoring them all here, and setting
> + * return address to a fixup that sets them up again. But that's a lot of
> + * effort for a small bit of code. Lots of other things to do first.
> + */
> +kvmppc_p9_bad_interrupt:
> + /*
> + * Set GUEST_MODE_NONE so the handler won't branch to KVM, and clear
> + * MSR_RI in r12 ([H]SRR1) so the handler won't try to return.
> + */
> + li r10,KVM_GUEST_MODE_NONE
> + stb r10,HSTATE_IN_GUEST(r13)
> + li r10,MSR_RI
> + andc r12,r12,r10
> +
> + /*
> + * Clean up guest registers to give host a chance to run.
> + */
> + li r10,0
> + mtspr SPRN_AMR,r10
> + mtspr SPRN_IAMR,r10
> + mtspr SPRN_CIABR,r10
> + mtspr SPRN_DAWRX0,r10
> +BEGIN_FTR_SECTION
> + mtspr SPRN_DAWRX1,r10
> +END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
> + mtspr SPRN_PID,r10
> +
> + /*
> + * Switch to host MMU mode
> + */
> + ld r10, HSTATE_KVM_VCPU(r13)
> + ld r10, VCPU_KVM(r10)
> + lwz r10, KVM_HOST_LPID(r10)
> + mtspr SPRN_LPID,r10
> +
> + ld r10, HSTATE_KVM_VCPU(r13)
> + ld r10, VCPU_KVM(r10)
> + ld r10, KVM_HOST_LPCR(r10)
> + mtspr SPRN_LPCR,r10
> +
> + /*
> + * Go back to interrupt handler
> + */
> + ld r10,HSTATE_SCRATCH0(r13)
> + cmpwi r10,BOOK3S_INTERRUPT_MACHINE_CHECK
> + beq machine_check_common
> +
> + ld r10,HSTATE_SCRATCH0(r13)
> + cmpwi r10,BOOK3S_INTERRUPT_SYSTEM_RESET
> + beq system_reset_common
> +
> + b .
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index 989a1ff5ad11..6dca639ed997 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1431,6 +1431,8 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
> */
> case BOOK3S_INTERRUPT_H_DATA_STORAGE:
> r = RESUME_PAGE_FAULT;
> + if (vcpu->arch.fault_dsisr == HDSISR_CANARY)
> + r = RESUME_GUEST; /* Just retry if it's the canary */
> break;
> case BOOK3S_INTERRUPT_H_INST_STORAGE:
> vcpu->arch.fault_dar = kvmppc_get_pc(vcpu);
> @@ -3690,6 +3692,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> u64 tb, next_timer;
> int trap, save_pmu;
>
> + WARN_ON_ONCE(vcpu->arch.ceded);
> +
> tb = mftb();
> next_timer = timer_get_next_tb();
> if (tb >= next_timer)
> @@ -3698,8 +3702,6 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> if (next_timer < time_limit)
> time_limit = next_timer;
>
> - vcpu->arch.ceded = 0;
> -
> kvmhv_save_host_pmu(); /* saves it to PACA kvm_hstate */
>
> kvmppc_subcore_enter_guest();
> @@ -3826,9 +3828,10 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> }
> }
> kvmppc_xive_pull_vcpu(vcpu);
> +
> + vcpu->arch.slb_max = 0;
> }
>
> - vcpu->arch.slb_max = 0;
> dec = mfspr(SPRN_DEC);
> if (!(lpcr & LPCR_LD)) /* Sign extend if not using large decrementer */
> dec = (s32) dec;
> diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
> new file mode 100644
> index 000000000000..3151b3d62c01
> --- /dev/null
> +++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
> @@ -0,0 +1,223 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +#include <linux/kernel.h>
> +#include <linux/kvm_host.h>
> +#include <asm/asm-prototypes.h>
> +#include <asm/dbell.h>
> +#include <asm/kvm_ppc.h>
> +
> +#ifdef CONFIG_KVM_BOOK3S_HV_EXIT_TIMING
> +static void __start_timing(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator *next)
> +{
> + struct kvmppc_vcore *vc = vcpu->arch.vcore;
> + u64 tb = mftb() - vc->tb_offset_applied;
> +
> + vcpu->arch.cur_activity = next;
> + vcpu->arch.cur_tb_start = tb;
> +}
> +
> +static void __accumulate_time(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator *next)
> +{
> + struct kvmppc_vcore *vc = vcpu->arch.vcore;
> + struct kvmhv_tb_accumulator *curr;
> + u64 tb = mftb() - vc->tb_offset_applied;
> + u64 prev_tb;
> + u64 delta;
> + u64 seq;
> +
> + curr = vcpu->arch.cur_activity;
> + vcpu->arch.cur_activity = next;
> + prev_tb = vcpu->arch.cur_tb_start;
> + vcpu->arch.cur_tb_start = tb;
> +
> + if (!curr)
> + return;
> +
> + delta = tb - prev_tb;
> +
> + seq = curr->seqcount;
> + curr->seqcount = seq + 1;
> + smp_wmb();
> + curr->tb_total += delta;
> + if (seq == 0 || delta < curr->tb_min)
> + curr->tb_min = delta;
> + if (delta > curr->tb_max)
> + curr->tb_max = delta;
> + smp_wmb();
> + curr->seqcount = seq + 2;
> +}
> +
> +#define start_timing(vcpu, next) __start_timing(vcpu, next)
> +#define end_timing(vcpu) __start_timing(vcpu, NULL)
> +#define accumulate_time(vcpu, next) __accumulate_time(vcpu, next)
> +#else
> +#define start_timing(vcpu, next) do {} while (0)
> +#define end_timing(vcpu) do {} while (0)
> +#define accumulate_time(vcpu, next) do {} while (0)
> +#endif
> +
> +static inline void mfslb(unsigned int idx, u64 *slbee, u64 *slbev)
> +{
> + asm volatile("slbmfev %0,%1" : "=r" (*slbev) : "r" (idx));
> + asm volatile("slbmfee %0,%1" : "=r" (*slbee) : "r" (idx));
> +}
> +
> +static inline void mtslb(unsigned int idx, u64 slbee, u64 slbev)
> +{
> + BUG_ON((slbee & 0xfff) != idx);
> +
> + asm volatile("slbmte %0,%1" :: "r" (slbev), "r" (slbee));
> +}
> +
> +static inline void slb_invalidate(unsigned int ih)
> +{
> + asm volatile("slbia %0" :: "i"(ih));
> +}
This one is not used.
> +
> +/*
> + * Malicious or buggy radix guests may have inserted SLB entries
> + * (only 0..3 because radix always runs with UPRT=1), so these must
> + * be cleared here to avoid side-channels. slbmte is used rather
> + * than slbia, as it won't clear cached translations.
> + */
> +static void radix_clear_slb(void)
> +{
> + u64 slbee, slbev;
> + int i;
> +
> + for (i = 0; i < 4; i++) {
> + mfslb(i, &slbee, &slbev);
> + if (unlikely(slbee || slbev)) {
> + slbee = i;
> + slbev = 0;
> + mtslb(i, slbee, slbev);
> + }
> + }
> +}
> +
> +int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu)
> +{
> + u64 *exsave;
> + unsigned long msr = mfmsr();
> + int trap;
> +
> + start_timing(vcpu, &vcpu->arch.rm_entry);
> +
> + vcpu->arch.ceded = 0;
> +
> + WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
> + WARN_ON_ONCE(!(vcpu->arch.shregs.msr & MSR_ME));
> +
> + mtspr(SPRN_HSRR0, vcpu->arch.regs.nip);
> + mtspr(SPRN_HSRR1, (vcpu->arch.shregs.msr & ~MSR_HV) | MSR_ME);
> +
> + /*
> + * On POWER9 DD2.1 and below, sometimes on a Hypervisor Data Storage
> + * Interrupt (HDSI) the HDSISR is not be updated at all.
> + *
> + * To work around this we put a canary value into the HDSISR before
> + * returning to a guest and then check for this canary when we take a
> + * HDSI. If we find the canary on a HDSI, we know the hardware didn't
> + * update the HDSISR. In this case we return to the guest to retake the
> + * HDSI which should correctly update the HDSISR the second time HDSI
> + * entry.
> + *
> + * Just do this on all p9 processors for now.
> + */
> + mtspr(SPRN_HDSISR, HDSISR_CANARY);
> +
> + accumulate_time(vcpu, &vcpu->arch.guest_time);
> +
> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
> + kvmppc_p9_enter_guest(vcpu);
> + // Radix host and guest means host never runs with guest MMU state
> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
> +
> + accumulate_time(vcpu, &vcpu->arch.rm_intr);
> +
> + /* Get these from r11/12 and paca exsave */
> + vcpu->arch.shregs.srr0 = mfspr(SPRN_SRR0);
> + vcpu->arch.shregs.srr1 = mfspr(SPRN_SRR1);
> + vcpu->arch.shregs.dar = mfspr(SPRN_DAR);
> + vcpu->arch.shregs.dsisr = mfspr(SPRN_DSISR);
> +
> + /* 0x2 bit for HSRR is only used by PR and P7/8 HV paths, clear it */
> + trap = local_paca->kvm_hstate.scratch0 & ~0x2;
> + if (likely(trap > BOOK3S_INTERRUPT_MACHINE_CHECK)) {
> + exsave = local_paca->exgen;
> + } else if (trap == BOOK3S_INTERRUPT_SYSTEM_RESET) {
> + exsave = local_paca->exnmi;
> + } else { /* trap == 0x200 */
> + exsave = local_paca->exmc;
> + }
> +
> + vcpu->arch.regs.gpr[1] = local_paca->kvm_hstate.scratch1;
> + vcpu->arch.regs.gpr[3] = local_paca->kvm_hstate.scratch2;
> + vcpu->arch.regs.gpr[9] = exsave[EX_R9/sizeof(u64)];
> + vcpu->arch.regs.gpr[10] = exsave[EX_R10/sizeof(u64)];
> + vcpu->arch.regs.gpr[11] = exsave[EX_R11/sizeof(u64)];
> + vcpu->arch.regs.gpr[12] = exsave[EX_R12/sizeof(u64)];
> + vcpu->arch.regs.gpr[13] = exsave[EX_R13/sizeof(u64)];
> + vcpu->arch.ppr = exsave[EX_PPR/sizeof(u64)];
> + vcpu->arch.cfar = exsave[EX_CFAR/sizeof(u64)];
> + vcpu->arch.regs.ctr = exsave[EX_CTR/sizeof(u64)];
> +
> + vcpu->arch.last_inst = KVM_INST_FETCH_FAILED;
> +
> + if (unlikely(trap == BOOK3S_INTERRUPT_MACHINE_CHECK)) {
> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
> + kvmppc_realmode_machine_check(vcpu);
> +
> + } else if (unlikely(trap == BOOK3S_INTERRUPT_HMI)) {
> + kvmppc_realmode_hmi_handler();
> +
> + } else if (trap == BOOK3S_INTERRUPT_H_EMUL_ASSIST) {
> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
> +
> + } else if (trap == BOOK3S_INTERRUPT_H_DATA_STORAGE) {
> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
> +
> + } else if (trap == BOOK3S_INTERRUPT_H_INST_STORAGE) {
> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
> +
> + } else if (trap == BOOK3S_INTERRUPT_H_FAC_UNAVAIL) {
> + vcpu->arch.hfscr = mfspr(SPRN_HFSCR);
> +
> +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
> + /*
> + * Softpatch interrupt for transactional memory emulation cases
> + * on POWER9 DD2.2. This is early in the guest exit path - we
> + * haven't saved registers or done a treclaim yet.
> + */
> + } else if (trap == BOOK3S_INTERRUPT_HV_SOFTPATCH) {
> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
> +
> + /*
> + * The cases we want to handle here are those where the guest
> + * is in real suspend mode and is trying to transition to
> + * transactional mode.
> + */
> + if (local_paca->kvm_hstate.fake_suspend &&
> + (vcpu->arch.shregs.msr & MSR_TS_S)) {
> + if (kvmhv_p9_tm_emulation_early(vcpu)) {
> + /* Prevent it being handled again. */
> + trap = 0;
> + }
> + }
> +#endif
> + }
> +
> + radix_clear_slb();
> +
> + __mtmsrd(msr, 0);
The asm code only sets RI but this potentially sets more bits including
MSR_EE, is it expected to be 0 when __kvmhv_vcpu_entry_p9() is called?
> + mtspr(SPRN_CTRLT, 1);
What is this for? ISA does not shed much light:
===
63 RUN This bit controls an external I/O pin.
===
> +
> + accumulate_time(vcpu, &vcpu->arch.rm_exit);
This should not compile without CONFIG_KVM_BOOK3S_HV_EXIT_TIMING.
> +
> + end_timing(vcpu);
> +
> + return trap;
The asm does "For hash guest, read the guest SLB and save it away", this
code does not. Is this new fast-path-in-c only for radix-on-radix or
hash VMs are supported too?
> +}
> +EXPORT_SYMBOL_GPL(__kvmhv_vcpu_entry_p9);
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index 2d0d14ed1d92..6118e8a97ddd 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -44,9 +44,8 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
> #define NAPPING_UNSPLIT 3
>
> /* Stack frame offsets for kvmppc_hv_entry */
> -#define SFS 208
> +#define SFS 160
> #define STACK_SLOT_TRAP (SFS-4)
> -#define STACK_SLOT_SHORT_PATH (SFS-8)
> #define STACK_SLOT_TID (SFS-16)
> #define STACK_SLOT_PSSCR (SFS-24)
> #define STACK_SLOT_PID (SFS-32)
> @@ -59,8 +58,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
> #define STACK_SLOT_UAMOR (SFS-88)
> #define STACK_SLOT_DAWR1 (SFS-96)
> #define STACK_SLOT_DAWRX1 (SFS-104)
> -/* the following is used by the P9 short path */
> -#define STACK_SLOT_NVGPRS (SFS-152) /* 18 gprs */
>
> /*
> * Call kvmppc_hv_entry in real mode.
> @@ -1008,9 +1005,6 @@ END_MMU_FTR_SECTION_IFCLR(MMU_FTR_TYPE_RADIX)
> no_xive:
> #endif /* CONFIG_KVM_XICS */
>
> - li r0, 0
> - stw r0, STACK_SLOT_SHORT_PATH(r1)
> -
> deliver_guest_interrupt: /* r4 = vcpu, r13 = paca */
> /* Check if we can deliver an external or decrementer interrupt now */
> ld r0, VCPU_PENDING_EXC(r4)
> @@ -1030,7 +1024,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
> mtspr SPRN_SRR0, r6
> mtspr SPRN_SRR1, r7
>
> -fast_guest_entry_c:
> ld r10, VCPU_PC(r4)
> ld r11, VCPU_MSR(r4)
> /* r11 = vcpu->arch.msr & ~MSR_HV */
> @@ -1135,97 +1128,6 @@ ret_to_ultra:
> ld r4, VCPU_GPR(R4)(r4)
> sc 2
>
> -/*
> - * Enter the guest on a P9 or later system where we have exactly
> - * one vcpu per vcore and we don't need to go to real mode
> - * (which implies that host and guest are both using radix MMU mode).
> - * r3 = vcpu pointer
> - * Most SPRs and all the VSRs have been loaded already.
> - */
> -_GLOBAL(__kvmhv_vcpu_entry_p9)
> -EXPORT_SYMBOL_GPL(__kvmhv_vcpu_entry_p9)
> - mflr r0
> - std r0, PPC_LR_STKOFF(r1)
> - stdu r1, -SFS(r1)
> -
> - li r0, 1
> - stw r0, STACK_SLOT_SHORT_PATH(r1)
> -
> - std r3, HSTATE_KVM_VCPU(r13)
> - mfcr r4
> - stw r4, SFS+8(r1)
> -
> - std r1, HSTATE_HOST_R1(r13)
> -
> - reg = 14
> - .rept 18
> - std reg, STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
> - reg = reg + 1
> - .endr
> -
> - reg = 14
> - .rept 18
> - ld reg, __VCPU_GPR(reg)(r3)
> - reg = reg + 1
> - .endr
> -
> - mfmsr r10
> - std r10, HSTATE_HOST_MSR(r13)
> -
> - mr r4, r3
> - b fast_guest_entry_c
> -guest_exit_short_path:
> - /*
> - * Malicious or buggy radix guests may have inserted SLB entries
> - * (only 0..3 because radix always runs with UPRT=1), so these must
> - * be cleared here to avoid side-channels. slbmte is used rather
> - * than slbia, as it won't clear cached translations.
> - */
> - li r0,0
> - slbmte r0,r0
> - li r4,1
> - slbmte r0,r4
> - li r4,2
> - slbmte r0,r4
> - li r4,3
> - slbmte r0,r4
> -
> - li r0, KVM_GUEST_MODE_NONE
> - stb r0, HSTATE_IN_GUEST(r13)
> -
> - reg = 14
> - .rept 18
> - std reg, __VCPU_GPR(reg)(r9)
> - reg = reg + 1
> - .endr
> -
> - reg = 14
> - .rept 18
> - ld reg, STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
> - reg = reg + 1
> - .endr
> -
> - lwz r4, SFS+8(r1)
> - mtcr r4
> -
> - mr r3, r12 /* trap number */
> -
> - addi r1, r1, SFS
> - ld r0, PPC_LR_STKOFF(r1)
> - mtlr r0
> -
> - /* If we are in real mode, do a rfid to get back to the caller */
> - mfmsr r4
> - andi. r5, r4, MSR_IR
> - bnelr
> - rldicl r5, r4, 64 - MSR_TS_S_LG, 62 /* extract TS field */
> - mtspr SPRN_SRR0, r0
> - ld r10, HSTATE_HOST_MSR(r13)
> - rldimi r10, r5, MSR_TS_S_LG, 63 - MSR_TS_T_LG
> - mtspr SPRN_SRR1, r10
> - RFI_TO_KERNEL
> - b .
> -
> secondary_too_late:
> li r12, 0
> stw r12, STACK_SLOT_TRAP(r1)
> @@ -1397,14 +1299,9 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
> mr r4,r9
> bge fast_guest_return
> 2:
> - /* If we came in through the P9 short path, no real mode hcalls */
> - lwz r0, STACK_SLOT_SHORT_PATH(r1)
> - cmpwi r0, 0
> - bne no_try_real
> /* See if this is an hcall we can handle in real mode */
> cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
> beq hcall_try_real_mode
> -no_try_real:
>
> /* Hypervisor doorbell - exit only if host IPI flag set */
> cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
> @@ -1447,11 +1344,6 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
> bl kvmhv_accumulate_time
> #endif
> #ifdef CONFIG_KVM_XICS
> - /* If we came in through the P9 short path, xive pull is done in C */
> - lwz r0, STACK_SLOT_SHORT_PATH(r1)
> - cmpwi r0, 0
> - bne 1f
> -
> /* We are exiting, pull the VP from the XIVE */
> lbz r0, VCPU_XIVE_PUSHED(r9)
> cmpwi cr0, r0, 0
> @@ -1496,11 +1388,6 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
> 1: nop
> patch_site 1b patch__call_kvm_flush_link_stack
>
> - /* If we came in through the P9 short path, go back out to C now */
> - lwz r0, STACK_SLOT_SHORT_PATH(r1)
> - cmpwi r0, 0
> - bne guest_exit_short_path
> -
> /* For hash guest, read the guest SLB and save it away */
> ld r5, VCPU_KVM(r9)
> lbz r0, KVM_RADIX(r5)
> @@ -1548,8 +1435,10 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
> b guest_bypass
>
> 0: /*
> - * Sanitise radix guest SLB, see guest_exit_short_path comment.
> - * We clear vcpu->arch.slb_max to match earlier behaviour.
> + * Malicious or buggy radix guests may have inserted SLB entries
> + * (only 0..3 because radix always runs with UPRT=1), so these must
> + * be cleared here to avoid side-channels. slbmte is used rather
> + * than slbia, as it won't clear cached translations.
> */
> li r0,0
> stw r0,VCPU_SLB_MAX(r9)
> @@ -3362,7 +3251,7 @@ BEGIN_FTR_SECTION
> mtspr SPRN_DAWRX1, r0
> END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
>
> - /* Clear hash and radix guest SLB, see guest_exit_short_path comment. */
> + /* Clear hash and radix guest SLB. */
> slbmte r0, r0
> PPC_SLBIA(6)
>
>
--
Alexey
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 29/46] KVM: PPC: Book3S HV P9: Implement the rest of the P9 path in C
@ 2021-04-01 5:30 ` Alexey Kardashevskiy
0 siblings, 0 replies; 188+ messages in thread
From: Alexey Kardashevskiy @ 2021-04-01 5:30 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 3/23/21 12:02 PM, Nicholas Piggin wrote:
> Almost all logic is moved to C, by introducing a new in_guest mode that
> selects and branches very early in the interrupt handler to the P9 exit
> code.
>
> The remaining assembly is only about 160 lines of low level stack setup,
> with VCPU vs host register save and restore, plus a small shim to the
> legacy paths in the interrupt handler.
>
> There are two motivations for this, the first is just make the code more
> maintainable being in C. The second is to reduce the amount of code
> running in a special KVM mode, "realmode". I put that in quotes because
> with radix it is no longer necessarily real-mode in the MMU, but it
> still has to be treated specially because it may be in real-mode, and
> has various important registers like PID, DEC, TB, etc set to guest.
> This is hostile to the rest of Linux and can't use arbitrary kernel
> functionality or be instrumented well.
>
> This initial patch is a reasonably faithful conversion of the asm code.
> It does lack any loop to return quickly back into the guest without
> switching out of realmode in the case of unimportant or easily handled
> interrupts, as explained in the previous change, handling HV interrupts
> in real mode is not so important for P9.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
> arch/powerpc/include/asm/asm-prototypes.h | 3 +-
> arch/powerpc/include/asm/kvm_asm.h | 3 +-
> arch/powerpc/include/asm/kvm_book3s_64.h | 8 +
> arch/powerpc/kernel/security.c | 5 +-
> arch/powerpc/kvm/Makefile | 3 +
> arch/powerpc/kvm/book3s_64_entry.S | 246 ++++++++++++++++++++++
> arch/powerpc/kvm/book3s_hv.c | 9 +-
> arch/powerpc/kvm/book3s_hv_interrupt.c | 223 ++++++++++++++++++++
> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 123 +----------
> 9 files changed, 500 insertions(+), 123 deletions(-)
> create mode 100644 arch/powerpc/kvm/book3s_hv_interrupt.c
>
> diff --git a/arch/powerpc/include/asm/asm-prototypes.h b/arch/powerpc/include/asm/asm-prototypes.h
> index 939f3c94c8f3..7c74c80ed994 100644
> --- a/arch/powerpc/include/asm/asm-prototypes.h
> +++ b/arch/powerpc/include/asm/asm-prototypes.h
> @@ -122,6 +122,7 @@ extern s32 patch__call_flush_branch_caches3;
> extern s32 patch__flush_count_cache_return;
> extern s32 patch__flush_link_stack_return;
> extern s32 patch__call_kvm_flush_link_stack;
> +extern s32 patch__call_kvm_flush_link_stack_p9;
> extern s32 patch__memset_nocache, patch__memcpy_nocache;
>
> extern long flush_branch_caches;
> @@ -142,7 +143,7 @@ void kvmhv_load_host_pmu(void);
> void kvmhv_save_guest_pmu(struct kvm_vcpu *vcpu, bool pmu_in_use);
> void kvmhv_load_guest_pmu(struct kvm_vcpu *vcpu);
>
> -int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu);
> +void kvmppc_p9_enter_guest(struct kvm_vcpu *vcpu);
>
> long kvmppc_h_set_dabr(struct kvm_vcpu *vcpu, unsigned long dabr);
> long kvmppc_h_set_xdabr(struct kvm_vcpu *vcpu, unsigned long dabr,
> diff --git a/arch/powerpc/include/asm/kvm_asm.h b/arch/powerpc/include/asm/kvm_asm.h
> index a3633560493b..b4f9996bd331 100644
> --- a/arch/powerpc/include/asm/kvm_asm.h
> +++ b/arch/powerpc/include/asm/kvm_asm.h
> @@ -146,7 +146,8 @@
> #define KVM_GUEST_MODE_GUEST 1
> #define KVM_GUEST_MODE_SKIP 2
> #define KVM_GUEST_MODE_GUEST_HV 3
> -#define KVM_GUEST_MODE_HOST_HV 4
> +#define KVM_GUEST_MODE_GUEST_HV_FAST 4 /* ISA v3.0 with host radix mode */
> +#define KVM_GUEST_MODE_HOST_HV 5
>
> #define KVM_INST_FETCH_FAILED -1
>
> diff --git a/arch/powerpc/include/asm/kvm_book3s_64.h b/arch/powerpc/include/asm/kvm_book3s_64.h
> index 9bb9bb370b53..c214bcffb441 100644
> --- a/arch/powerpc/include/asm/kvm_book3s_64.h
> +++ b/arch/powerpc/include/asm/kvm_book3s_64.h
> @@ -153,9 +153,17 @@ static inline bool kvmhv_vcpu_is_radix(struct kvm_vcpu *vcpu)
> return radix;
> }
>
> +int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu);
> +
> #define KVM_DEFAULT_HPT_ORDER 24 /* 16MB HPT by default */
> #endif
>
> +/*
> + * Invalid HDSISR value which is used to indicate when HW has not set the reg.
> + * Used to work around an errata.
> + */
> +#define HDSISR_CANARY 0x7fff
> +
> /*
> * We use a lock bit in HPTE dword 0 to synchronize updates and
> * accesses to each HPTE, and another bit to indicate non-present
> diff --git a/arch/powerpc/kernel/security.c b/arch/powerpc/kernel/security.c
> index e4e1a94ccf6a..3a607c11f20f 100644
> --- a/arch/powerpc/kernel/security.c
> +++ b/arch/powerpc/kernel/security.c
> @@ -430,16 +430,19 @@ device_initcall(stf_barrier_debugfs_init);
>
> static void update_branch_cache_flush(void)
> {
> - u32 *site;
> + u32 *site, __maybe_unused *site2;
>
> #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
> site = &patch__call_kvm_flush_link_stack;
> + site2 = &patch__call_kvm_flush_link_stack_p9;
> // This controls the branch from guest_exit_cont to kvm_flush_link_stack
> if (link_stack_flush_type = BRANCH_CACHE_FLUSH_NONE) {
> patch_instruction_site(site, ppc_inst(PPC_INST_NOP));
> + patch_instruction_site(site2, ppc_inst(PPC_INST_NOP));
> } else {
> // Could use HW flush, but that could also flush count cache
> patch_branch_site(site, (u64)&kvm_flush_link_stack, BRANCH_SET_LINK);
> + patch_branch_site(site2, (u64)&kvm_flush_link_stack, BRANCH_SET_LINK);
> }
> #endif
>
> diff --git a/arch/powerpc/kvm/Makefile b/arch/powerpc/kvm/Makefile
> index cdd119028f64..40a55a5ba4ff 100644
> --- a/arch/powerpc/kvm/Makefile
> +++ b/arch/powerpc/kvm/Makefile
> @@ -43,6 +43,9 @@ kvm-objs-$(CONFIG_KVM_E500MC) := $(kvm-e500mc-objs)
> kvm-book3s_64-builtin-objs-$(CONFIG_SPAPR_TCE_IOMMU) := \
> book3s_64_vio_hv.o
>
> +kvm-book3s_64-builtin-objs-y += \
> + book3s_hv_interrupt.o
> +
> kvm-pr-y := \
> fpu.o \
> emulate.o \
> diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
> index de81ab69555b..845df5fefdbd 100644
> --- a/arch/powerpc/kvm/book3s_64_entry.S
> +++ b/arch/powerpc/kvm/book3s_64_entry.S
> @@ -1,11 +1,16 @@
> /* SPDX-License-Identifier: GPL-2.0-only */
> #include <asm/asm-offsets.h>
> #include <asm/cache.h>
> +#include <asm/code-patching-asm.h>
> #include <asm/exception-64s.h>
> +#include <asm/export.h>
> #include <asm/kvm_asm.h>
> #include <asm/kvm_book3s_asm.h>
> +#include <asm/mmu.h>
> #include <asm/ppc_asm.h>
> +#include <asm/ptrace.h>
> #include <asm/reg.h>
> +#include <asm/ultravisor-api.h>
>
> /*
> * These are branched to from interrupt handlers in exception-64s.S which set
> @@ -22,10 +27,15 @@
> .global kvmppc_hcall
> .balign IFETCH_ALIGN_BYTES
> kvmppc_hcall:
> + lbz r10,HSTATE_IN_GUEST(r13)
> + cmpwi r10,KVM_GUEST_MODE_GUEST_HV_FAST
> + beq kvmppc_p9_exit_hcall
> ld r10,PACA_EXGEN+EX_R13(r13)
> SET_SCRATCH0(r10)
> li r10,0xc00
> /* Now we look like kvmppc_interrupt */
> + li r11,PACA_EXGEN
> + b 1f
>
> /*
> * KVM interrupt entry occurs after GEN_INT_ENTRY runs, and follows that
> @@ -46,6 +56,12 @@ kvmppc_hcall:
> .global kvmppc_interrupt
> .balign IFETCH_ALIGN_BYTES
> kvmppc_interrupt:
> + std r10,HSTATE_SCRATCH0(r13)
> + lbz r10,HSTATE_IN_GUEST(r13)
> + cmpwi r10,KVM_GUEST_MODE_GUEST_HV_FAST
> + beq kvmppc_p9_exit_interrupt
> + ld r10,HSTATE_SCRATCH0(r13)
> + lbz r11,HSTATE_IN_GUEST(r13)
> li r11,PACA_EXGEN
> cmpdi r10,0x200
> bgt+ 1f
> @@ -140,3 +156,233 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
> GET_SCRATCH0(r13)
> HRFI_TO_KERNEL
> #endif
> +
> +/* Stack frame offsets for kvmppc_hv_entry */
> +#define SFS (144 + STACK_FRAME_MIN_SIZE)
> +#define STACK_SLOT_NVGPRS (SFS - 144) /* 18 gprs */
> +
> +/*
> + * void kvmppc_p9_enter_guest(struct vcpu *vcpu);
> + *
> + * Enter the guest on a ISAv3.0 or later system where we have exactly
> + * one vcpu per vcore, and both the host and guest are radix, and threads
> + * are set to "indepdent mode".
> + */
> +.balign IFETCH_ALIGN_BYTES
> +_GLOBAL(kvmppc_p9_enter_guest)
> +EXPORT_SYMBOL_GPL(kvmppc_p9_enter_guest)
> + mflr r0
> + std r0,PPC_LR_STKOFF(r1)
> + stdu r1,-SFS(r1)
> +
> + std r1,HSTATE_HOST_R1(r13)
> +
> + mfcr r4
> + stw r4,SFS+8(r1)
> +
> + reg = 14
> + .rept 18
> + std reg,STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
> + reg = reg + 1
> + .endr
> +
> + ld r4,VCPU_LR(r3)
> + mtlr r4
> + ld r4,VCPU_CTR(r3)
> + mtctr r4
> + ld r4,VCPU_XER(r3)
> + mtspr SPRN_XER,r4
> +
> + ld r1,VCPU_CR(r3)
> +
> +BEGIN_FTR_SECTION
> + ld r4,VCPU_CFAR(r3)
> + mtspr SPRN_CFAR,r4
> +END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
> +BEGIN_FTR_SECTION
> + ld r4,VCPU_PPR(r3)
> + mtspr SPRN_PPR,r4
> +END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
> +
> + reg = 4
> + .rept 28
> + ld reg,__VCPU_GPR(reg)(r3)
> + reg = reg + 1
> + .endr
> +
> + ld r4,VCPU_KVM(r3)
> + lbz r4,KVM_SECURE_GUEST(r4)
> + cmpdi r4,0
> + ld r4,VCPU_GPR(R4)(r3)
> + bne .Lret_to_ultra
> +
> + mtcr r1
> +
> + ld r0,VCPU_GPR(R0)(r3)
> + ld r1,VCPU_GPR(R1)(r3)
> + ld r2,VCPU_GPR(R2)(r3)
> + ld r3,VCPU_GPR(R3)(r3)
> +
> + HRFI_TO_GUEST
> + b .
> +
> + /*
> + * Use UV_RETURN ultracall to return control back to the Ultravisor
> + * after processing an hypercall or interrupt that was forwarded
> + * (a.k.a. reflected) to the Hypervisor.
> + *
> + * All registers have already been reloaded except the ucall requires:
> + * R0 = hcall result
> + * R2 = SRR1, so UV can detect a synthesized interrupt (if any)
> + * R3 = UV_RETURN
> + */
> +.Lret_to_ultra:
> + mtcr r1
> + ld r1,VCPU_GPR(R1)(r3)
> +
> + ld r0,VCPU_GPR(R3)(r3)
> + mfspr r2,SPRN_SRR1
> + LOAD_REG_IMMEDIATE(r3, UV_RETURN)
> + sc 2
> +
> +/*
> + * kvmppc_p9_exit_hcall and kvmppc_p9_exit_interrupt are branched to from
> + * above if the interrupt was taken for a guest that was entered via
> + * kvmppc_p9_enter_guest().
> + *
> + * This code recovers the host stack and vcpu pointer, saves all GPRs and
> + * CR, LR, CTR, XER as well as guest MSR and NIA into the VCPU, then re-
> + * establishes the host stack and registers to return from the
> + * kvmppc_p9_enter_guest() function.
What does "this code" refer to? If it is the asm below, then it does not
save CTR, it is in the c code. Otherwise it is confusing (to me) :)
> + */
> +.balign IFETCH_ALIGN_BYTES
> +kvmppc_p9_exit_hcall:
> + mfspr r11,SPRN_SRR0
> + mfspr r12,SPRN_SRR1
> + li r10,0xc00
> + std r10,HSTATE_SCRATCH0(r13)
> +
> +.balign IFETCH_ALIGN_BYTES
> +kvmppc_p9_exit_interrupt:
> + /*
> + * If set to KVM_GUEST_MODE_GUEST_HV_FAST but we're still in the
> + * hypervisor, that means we can't return from the entry stack.
> + */
> + rldicl. r10,r12,64-MSR_HV_LG,63
> + bne- kvmppc_p9_bad_interrupt
> +
> + std r1,HSTATE_SCRATCH1(r13)
> + std r3,HSTATE_SCRATCH2(r13)
> + ld r1,HSTATE_HOST_R1(r13)
> + ld r3,HSTATE_KVM_VCPU(r13)
> +
> + std r9,VCPU_CR(r3)
> +
> +1:
> + std r11,VCPU_PC(r3)
> + std r12,VCPU_MSR(r3)
> +
> + reg = 14
> + .rept 18
> + std reg,__VCPU_GPR(reg)(r3)
> + reg = reg + 1
> + .endr
> +
> + /* r1, r3, r9-r13 are saved to vcpu by C code */
> + std r0,VCPU_GPR(R0)(r3)
> + std r2,VCPU_GPR(R2)(r3)
> + reg = 4
> + .rept 5
> + std reg,__VCPU_GPR(reg)(r3)
> + reg = reg + 1
> + .endr
> +
> + ld r2,PACATOC(r13)
> +
> + mflr r4
> + std r4,VCPU_LR(r3)
> + mfspr r4,SPRN_XER
> + std r4,VCPU_XER(r3)
> +
> + reg = 14
> + .rept 18
> + ld reg,STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
> + reg = reg + 1
> + .endr
> +
> + lwz r4,SFS+8(r1)
> + mtcr r4
> +
> + /*
> + * Flush the link stack here, before executing the first blr on the
> + * way out of the guest.
> + *
> + * The link stack won't match coming out of the guest anyway so the
> + * only cost is the flush itself. The call clobbers r0.
> + */
> +1: nop
> + patch_site 1b patch__call_kvm_flush_link_stack_p9
> +
> + addi r1,r1,SFS
> + ld r0,PPC_LR_STKOFF(r1)
> + mtlr r0
> + blr
> +
> +/*
> + * Took an interrupt somewhere right before HRFID to guest, so registers are
> + * in a bad way. Return things hopefully enough to run host virtual code and
> + * run the Linux interrupt handler (SRESET or MCE) to print something useful.
> + *
> + * We could be really clever and save all host registers in known locations
> + * before setting HSTATE_IN_GUEST, then restoring them all here, and setting
> + * return address to a fixup that sets them up again. But that's a lot of
> + * effort for a small bit of code. Lots of other things to do first.
> + */
> +kvmppc_p9_bad_interrupt:
> + /*
> + * Set GUEST_MODE_NONE so the handler won't branch to KVM, and clear
> + * MSR_RI in r12 ([H]SRR1) so the handler won't try to return.
> + */
> + li r10,KVM_GUEST_MODE_NONE
> + stb r10,HSTATE_IN_GUEST(r13)
> + li r10,MSR_RI
> + andc r12,r12,r10
> +
> + /*
> + * Clean up guest registers to give host a chance to run.
> + */
> + li r10,0
> + mtspr SPRN_AMR,r10
> + mtspr SPRN_IAMR,r10
> + mtspr SPRN_CIABR,r10
> + mtspr SPRN_DAWRX0,r10
> +BEGIN_FTR_SECTION
> + mtspr SPRN_DAWRX1,r10
> +END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
> + mtspr SPRN_PID,r10
> +
> + /*
> + * Switch to host MMU mode
> + */
> + ld r10, HSTATE_KVM_VCPU(r13)
> + ld r10, VCPU_KVM(r10)
> + lwz r10, KVM_HOST_LPID(r10)
> + mtspr SPRN_LPID,r10
> +
> + ld r10, HSTATE_KVM_VCPU(r13)
> + ld r10, VCPU_KVM(r10)
> + ld r10, KVM_HOST_LPCR(r10)
> + mtspr SPRN_LPCR,r10
> +
> + /*
> + * Go back to interrupt handler
> + */
> + ld r10,HSTATE_SCRATCH0(r13)
> + cmpwi r10,BOOK3S_INTERRUPT_MACHINE_CHECK
> + beq machine_check_common
> +
> + ld r10,HSTATE_SCRATCH0(r13)
> + cmpwi r10,BOOK3S_INTERRUPT_SYSTEM_RESET
> + beq system_reset_common
> +
> + b .
> diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
> index 989a1ff5ad11..6dca639ed997 100644
> --- a/arch/powerpc/kvm/book3s_hv.c
> +++ b/arch/powerpc/kvm/book3s_hv.c
> @@ -1431,6 +1431,8 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
> */
> case BOOK3S_INTERRUPT_H_DATA_STORAGE:
> r = RESUME_PAGE_FAULT;
> + if (vcpu->arch.fault_dsisr = HDSISR_CANARY)
> + r = RESUME_GUEST; /* Just retry if it's the canary */
> break;
> case BOOK3S_INTERRUPT_H_INST_STORAGE:
> vcpu->arch.fault_dar = kvmppc_get_pc(vcpu);
> @@ -3690,6 +3692,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> u64 tb, next_timer;
> int trap, save_pmu;
>
> + WARN_ON_ONCE(vcpu->arch.ceded);
> +
> tb = mftb();
> next_timer = timer_get_next_tb();
> if (tb >= next_timer)
> @@ -3698,8 +3702,6 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> if (next_timer < time_limit)
> time_limit = next_timer;
>
> - vcpu->arch.ceded = 0;
> -
> kvmhv_save_host_pmu(); /* saves it to PACA kvm_hstate */
>
> kvmppc_subcore_enter_guest();
> @@ -3826,9 +3828,10 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
> }
> }
> kvmppc_xive_pull_vcpu(vcpu);
> +
> + vcpu->arch.slb_max = 0;
> }
>
> - vcpu->arch.slb_max = 0;
> dec = mfspr(SPRN_DEC);
> if (!(lpcr & LPCR_LD)) /* Sign extend if not using large decrementer */
> dec = (s32) dec;
> diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
> new file mode 100644
> index 000000000000..3151b3d62c01
> --- /dev/null
> +++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
> @@ -0,0 +1,223 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +#include <linux/kernel.h>
> +#include <linux/kvm_host.h>
> +#include <asm/asm-prototypes.h>
> +#include <asm/dbell.h>
> +#include <asm/kvm_ppc.h>
> +
> +#ifdef CONFIG_KVM_BOOK3S_HV_EXIT_TIMING
> +static void __start_timing(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator *next)
> +{
> + struct kvmppc_vcore *vc = vcpu->arch.vcore;
> + u64 tb = mftb() - vc->tb_offset_applied;
> +
> + vcpu->arch.cur_activity = next;
> + vcpu->arch.cur_tb_start = tb;
> +}
> +
> +static void __accumulate_time(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator *next)
> +{
> + struct kvmppc_vcore *vc = vcpu->arch.vcore;
> + struct kvmhv_tb_accumulator *curr;
> + u64 tb = mftb() - vc->tb_offset_applied;
> + u64 prev_tb;
> + u64 delta;
> + u64 seq;
> +
> + curr = vcpu->arch.cur_activity;
> + vcpu->arch.cur_activity = next;
> + prev_tb = vcpu->arch.cur_tb_start;
> + vcpu->arch.cur_tb_start = tb;
> +
> + if (!curr)
> + return;
> +
> + delta = tb - prev_tb;
> +
> + seq = curr->seqcount;
> + curr->seqcount = seq + 1;
> + smp_wmb();
> + curr->tb_total += delta;
> + if (seq = 0 || delta < curr->tb_min)
> + curr->tb_min = delta;
> + if (delta > curr->tb_max)
> + curr->tb_max = delta;
> + smp_wmb();
> + curr->seqcount = seq + 2;
> +}
> +
> +#define start_timing(vcpu, next) __start_timing(vcpu, next)
> +#define end_timing(vcpu) __start_timing(vcpu, NULL)
> +#define accumulate_time(vcpu, next) __accumulate_time(vcpu, next)
> +#else
> +#define start_timing(vcpu, next) do {} while (0)
> +#define end_timing(vcpu) do {} while (0)
> +#define accumulate_time(vcpu, next) do {} while (0)
> +#endif
> +
> +static inline void mfslb(unsigned int idx, u64 *slbee, u64 *slbev)
> +{
> + asm volatile("slbmfev %0,%1" : "=r" (*slbev) : "r" (idx));
> + asm volatile("slbmfee %0,%1" : "=r" (*slbee) : "r" (idx));
> +}
> +
> +static inline void mtslb(unsigned int idx, u64 slbee, u64 slbev)
> +{
> + BUG_ON((slbee & 0xfff) != idx);
> +
> + asm volatile("slbmte %0,%1" :: "r" (slbev), "r" (slbee));
> +}
> +
> +static inline void slb_invalidate(unsigned int ih)
> +{
> + asm volatile("slbia %0" :: "i"(ih));
> +}
This one is not used.
> +
> +/*
> + * Malicious or buggy radix guests may have inserted SLB entries
> + * (only 0..3 because radix always runs with UPRT=1), so these must
> + * be cleared here to avoid side-channels. slbmte is used rather
> + * than slbia, as it won't clear cached translations.
> + */
> +static void radix_clear_slb(void)
> +{
> + u64 slbee, slbev;
> + int i;
> +
> + for (i = 0; i < 4; i++) {
> + mfslb(i, &slbee, &slbev);
> + if (unlikely(slbee || slbev)) {
> + slbee = i;
> + slbev = 0;
> + mtslb(i, slbee, slbev);
> + }
> + }
> +}
> +
> +int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu)
> +{
> + u64 *exsave;
> + unsigned long msr = mfmsr();
> + int trap;
> +
> + start_timing(vcpu, &vcpu->arch.rm_entry);
> +
> + vcpu->arch.ceded = 0;
> +
> + WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
> + WARN_ON_ONCE(!(vcpu->arch.shregs.msr & MSR_ME));
> +
> + mtspr(SPRN_HSRR0, vcpu->arch.regs.nip);
> + mtspr(SPRN_HSRR1, (vcpu->arch.shregs.msr & ~MSR_HV) | MSR_ME);
> +
> + /*
> + * On POWER9 DD2.1 and below, sometimes on a Hypervisor Data Storage
> + * Interrupt (HDSI) the HDSISR is not be updated at all.
> + *
> + * To work around this we put a canary value into the HDSISR before
> + * returning to a guest and then check for this canary when we take a
> + * HDSI. If we find the canary on a HDSI, we know the hardware didn't
> + * update the HDSISR. In this case we return to the guest to retake the
> + * HDSI which should correctly update the HDSISR the second time HDSI
> + * entry.
> + *
> + * Just do this on all p9 processors for now.
> + */
> + mtspr(SPRN_HDSISR, HDSISR_CANARY);
> +
> + accumulate_time(vcpu, &vcpu->arch.guest_time);
> +
> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
> + kvmppc_p9_enter_guest(vcpu);
> + // Radix host and guest means host never runs with guest MMU state
> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
> +
> + accumulate_time(vcpu, &vcpu->arch.rm_intr);
> +
> + /* Get these from r11/12 and paca exsave */
> + vcpu->arch.shregs.srr0 = mfspr(SPRN_SRR0);
> + vcpu->arch.shregs.srr1 = mfspr(SPRN_SRR1);
> + vcpu->arch.shregs.dar = mfspr(SPRN_DAR);
> + vcpu->arch.shregs.dsisr = mfspr(SPRN_DSISR);
> +
> + /* 0x2 bit for HSRR is only used by PR and P7/8 HV paths, clear it */
> + trap = local_paca->kvm_hstate.scratch0 & ~0x2;
> + if (likely(trap > BOOK3S_INTERRUPT_MACHINE_CHECK)) {
> + exsave = local_paca->exgen;
> + } else if (trap = BOOK3S_INTERRUPT_SYSTEM_RESET) {
> + exsave = local_paca->exnmi;
> + } else { /* trap = 0x200 */
> + exsave = local_paca->exmc;
> + }
> +
> + vcpu->arch.regs.gpr[1] = local_paca->kvm_hstate.scratch1;
> + vcpu->arch.regs.gpr[3] = local_paca->kvm_hstate.scratch2;
> + vcpu->arch.regs.gpr[9] = exsave[EX_R9/sizeof(u64)];
> + vcpu->arch.regs.gpr[10] = exsave[EX_R10/sizeof(u64)];
> + vcpu->arch.regs.gpr[11] = exsave[EX_R11/sizeof(u64)];
> + vcpu->arch.regs.gpr[12] = exsave[EX_R12/sizeof(u64)];
> + vcpu->arch.regs.gpr[13] = exsave[EX_R13/sizeof(u64)];
> + vcpu->arch.ppr = exsave[EX_PPR/sizeof(u64)];
> + vcpu->arch.cfar = exsave[EX_CFAR/sizeof(u64)];
> + vcpu->arch.regs.ctr = exsave[EX_CTR/sizeof(u64)];
> +
> + vcpu->arch.last_inst = KVM_INST_FETCH_FAILED;
> +
> + if (unlikely(trap = BOOK3S_INTERRUPT_MACHINE_CHECK)) {
> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
> + kvmppc_realmode_machine_check(vcpu);
> +
> + } else if (unlikely(trap = BOOK3S_INTERRUPT_HMI)) {
> + kvmppc_realmode_hmi_handler();
> +
> + } else if (trap = BOOK3S_INTERRUPT_H_EMUL_ASSIST) {
> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
> +
> + } else if (trap = BOOK3S_INTERRUPT_H_DATA_STORAGE) {
> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
> +
> + } else if (trap = BOOK3S_INTERRUPT_H_INST_STORAGE) {
> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
> +
> + } else if (trap = BOOK3S_INTERRUPT_H_FAC_UNAVAIL) {
> + vcpu->arch.hfscr = mfspr(SPRN_HFSCR);
> +
> +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
> + /*
> + * Softpatch interrupt for transactional memory emulation cases
> + * on POWER9 DD2.2. This is early in the guest exit path - we
> + * haven't saved registers or done a treclaim yet.
> + */
> + } else if (trap = BOOK3S_INTERRUPT_HV_SOFTPATCH) {
> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
> +
> + /*
> + * The cases we want to handle here are those where the guest
> + * is in real suspend mode and is trying to transition to
> + * transactional mode.
> + */
> + if (local_paca->kvm_hstate.fake_suspend &&
> + (vcpu->arch.shregs.msr & MSR_TS_S)) {
> + if (kvmhv_p9_tm_emulation_early(vcpu)) {
> + /* Prevent it being handled again. */
> + trap = 0;
> + }
> + }
> +#endif
> + }
> +
> + radix_clear_slb();
> +
> + __mtmsrd(msr, 0);
The asm code only sets RI but this potentially sets more bits including
MSR_EE, is it expected to be 0 when __kvmhv_vcpu_entry_p9() is called?
> + mtspr(SPRN_CTRLT, 1);
What is this for? ISA does not shed much light:
=63 RUN This bit controls an external I/O pin.
=
> +
> + accumulate_time(vcpu, &vcpu->arch.rm_exit);
This should not compile without CONFIG_KVM_BOOK3S_HV_EXIT_TIMING.
> +
> + end_timing(vcpu);
> +
> + return trap;
The asm does "For hash guest, read the guest SLB and save it away", this
code does not. Is this new fast-path-in-c only for radix-on-radix or
hash VMs are supported too?
> +}
> +EXPORT_SYMBOL_GPL(__kvmhv_vcpu_entry_p9);
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index 2d0d14ed1d92..6118e8a97ddd 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -44,9 +44,8 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
> #define NAPPING_UNSPLIT 3
>
> /* Stack frame offsets for kvmppc_hv_entry */
> -#define SFS 208
> +#define SFS 160
> #define STACK_SLOT_TRAP (SFS-4)
> -#define STACK_SLOT_SHORT_PATH (SFS-8)
> #define STACK_SLOT_TID (SFS-16)
> #define STACK_SLOT_PSSCR (SFS-24)
> #define STACK_SLOT_PID (SFS-32)
> @@ -59,8 +58,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
> #define STACK_SLOT_UAMOR (SFS-88)
> #define STACK_SLOT_DAWR1 (SFS-96)
> #define STACK_SLOT_DAWRX1 (SFS-104)
> -/* the following is used by the P9 short path */
> -#define STACK_SLOT_NVGPRS (SFS-152) /* 18 gprs */
>
> /*
> * Call kvmppc_hv_entry in real mode.
> @@ -1008,9 +1005,6 @@ END_MMU_FTR_SECTION_IFCLR(MMU_FTR_TYPE_RADIX)
> no_xive:
> #endif /* CONFIG_KVM_XICS */
>
> - li r0, 0
> - stw r0, STACK_SLOT_SHORT_PATH(r1)
> -
> deliver_guest_interrupt: /* r4 = vcpu, r13 = paca */
> /* Check if we can deliver an external or decrementer interrupt now */
> ld r0, VCPU_PENDING_EXC(r4)
> @@ -1030,7 +1024,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
> mtspr SPRN_SRR0, r6
> mtspr SPRN_SRR1, r7
>
> -fast_guest_entry_c:
> ld r10, VCPU_PC(r4)
> ld r11, VCPU_MSR(r4)
> /* r11 = vcpu->arch.msr & ~MSR_HV */
> @@ -1135,97 +1128,6 @@ ret_to_ultra:
> ld r4, VCPU_GPR(R4)(r4)
> sc 2
>
> -/*
> - * Enter the guest on a P9 or later system where we have exactly
> - * one vcpu per vcore and we don't need to go to real mode
> - * (which implies that host and guest are both using radix MMU mode).
> - * r3 = vcpu pointer
> - * Most SPRs and all the VSRs have been loaded already.
> - */
> -_GLOBAL(__kvmhv_vcpu_entry_p9)
> -EXPORT_SYMBOL_GPL(__kvmhv_vcpu_entry_p9)
> - mflr r0
> - std r0, PPC_LR_STKOFF(r1)
> - stdu r1, -SFS(r1)
> -
> - li r0, 1
> - stw r0, STACK_SLOT_SHORT_PATH(r1)
> -
> - std r3, HSTATE_KVM_VCPU(r13)
> - mfcr r4
> - stw r4, SFS+8(r1)
> -
> - std r1, HSTATE_HOST_R1(r13)
> -
> - reg = 14
> - .rept 18
> - std reg, STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
> - reg = reg + 1
> - .endr
> -
> - reg = 14
> - .rept 18
> - ld reg, __VCPU_GPR(reg)(r3)
> - reg = reg + 1
> - .endr
> -
> - mfmsr r10
> - std r10, HSTATE_HOST_MSR(r13)
> -
> - mr r4, r3
> - b fast_guest_entry_c
> -guest_exit_short_path:
> - /*
> - * Malicious or buggy radix guests may have inserted SLB entries
> - * (only 0..3 because radix always runs with UPRT=1), so these must
> - * be cleared here to avoid side-channels. slbmte is used rather
> - * than slbia, as it won't clear cached translations.
> - */
> - li r0,0
> - slbmte r0,r0
> - li r4,1
> - slbmte r0,r4
> - li r4,2
> - slbmte r0,r4
> - li r4,3
> - slbmte r0,r4
> -
> - li r0, KVM_GUEST_MODE_NONE
> - stb r0, HSTATE_IN_GUEST(r13)
> -
> - reg = 14
> - .rept 18
> - std reg, __VCPU_GPR(reg)(r9)
> - reg = reg + 1
> - .endr
> -
> - reg = 14
> - .rept 18
> - ld reg, STACK_SLOT_NVGPRS + ((reg - 14) * 8)(r1)
> - reg = reg + 1
> - .endr
> -
> - lwz r4, SFS+8(r1)
> - mtcr r4
> -
> - mr r3, r12 /* trap number */
> -
> - addi r1, r1, SFS
> - ld r0, PPC_LR_STKOFF(r1)
> - mtlr r0
> -
> - /* If we are in real mode, do a rfid to get back to the caller */
> - mfmsr r4
> - andi. r5, r4, MSR_IR
> - bnelr
> - rldicl r5, r4, 64 - MSR_TS_S_LG, 62 /* extract TS field */
> - mtspr SPRN_SRR0, r0
> - ld r10, HSTATE_HOST_MSR(r13)
> - rldimi r10, r5, MSR_TS_S_LG, 63 - MSR_TS_T_LG
> - mtspr SPRN_SRR1, r10
> - RFI_TO_KERNEL
> - b .
> -
> secondary_too_late:
> li r12, 0
> stw r12, STACK_SLOT_TRAP(r1)
> @@ -1397,14 +1299,9 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
> mr r4,r9
> bge fast_guest_return
> 2:
> - /* If we came in through the P9 short path, no real mode hcalls */
> - lwz r0, STACK_SLOT_SHORT_PATH(r1)
> - cmpwi r0, 0
> - bne no_try_real
> /* See if this is an hcall we can handle in real mode */
> cmpwi r12,BOOK3S_INTERRUPT_SYSCALL
> beq hcall_try_real_mode
> -no_try_real:
>
> /* Hypervisor doorbell - exit only if host IPI flag set */
> cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
> @@ -1447,11 +1344,6 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
> bl kvmhv_accumulate_time
> #endif
> #ifdef CONFIG_KVM_XICS
> - /* If we came in through the P9 short path, xive pull is done in C */
> - lwz r0, STACK_SLOT_SHORT_PATH(r1)
> - cmpwi r0, 0
> - bne 1f
> -
> /* We are exiting, pull the VP from the XIVE */
> lbz r0, VCPU_XIVE_PUSHED(r9)
> cmpwi cr0, r0, 0
> @@ -1496,11 +1388,6 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
> 1: nop
> patch_site 1b patch__call_kvm_flush_link_stack
>
> - /* If we came in through the P9 short path, go back out to C now */
> - lwz r0, STACK_SLOT_SHORT_PATH(r1)
> - cmpwi r0, 0
> - bne guest_exit_short_path
> -
> /* For hash guest, read the guest SLB and save it away */
> ld r5, VCPU_KVM(r9)
> lbz r0, KVM_RADIX(r5)
> @@ -1548,8 +1435,10 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
> b guest_bypass
>
> 0: /*
> - * Sanitise radix guest SLB, see guest_exit_short_path comment.
> - * We clear vcpu->arch.slb_max to match earlier behaviour.
> + * Malicious or buggy radix guests may have inserted SLB entries
> + * (only 0..3 because radix always runs with UPRT=1), so these must
> + * be cleared here to avoid side-channels. slbmte is used rather
> + * than slbia, as it won't clear cached translations.
> */
> li r0,0
> stw r0,VCPU_SLB_MAX(r9)
> @@ -3362,7 +3251,7 @@ BEGIN_FTR_SECTION
> mtspr SPRN_DAWRX1, r0
> END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
>
> - /* Clear hash and radix guest SLB, see guest_exit_short_path comment. */
> + /* Clear hash and radix guest SLB. */
> slbmte r0, r0
> PPC_SLBIA(6)
>
>
--
Alexey
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 29/46] KVM: PPC: Book3S HV P9: Implement the rest of the P9 path in C
2021-04-01 5:30 ` Alexey Kardashevskiy
@ 2021-04-01 10:35 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 10:35 UTC (permalink / raw)
To: Alexey Kardashevskiy, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Alexey Kardashevskiy's message of April 1, 2021 3:30 pm:
>
>
> On 3/23/21 12:02 PM, Nicholas Piggin wrote:
>> Almost all logic is moved to C, by introducing a new in_guest mode that
>> selects and branches very early in the interrupt handler to the P9 exit
>> code.
[...]
>> +/*
>> + * kvmppc_p9_exit_hcall and kvmppc_p9_exit_interrupt are branched to from
>> + * above if the interrupt was taken for a guest that was entered via
>> + * kvmppc_p9_enter_guest().
>> + *
>> + * This code recovers the host stack and vcpu pointer, saves all GPRs and
>> + * CR, LR, CTR, XER as well as guest MSR and NIA into the VCPU, then re-
>> + * establishes the host stack and registers to return from the
>> + * kvmppc_p9_enter_guest() function.
>
> What does "this code" refer to? If it is the asm below, then it does not
> save CTR, it is in the c code. Otherwise it is confusing (to me) :)
Yes you're right, CTR is saved in C.
>> + */
>> +.balign IFETCH_ALIGN_BYTES
>> +kvmppc_p9_exit_hcall:
>> + mfspr r11,SPRN_SRR0
>> + mfspr r12,SPRN_SRR1
>> + li r10,0xc00
>> + std r10,HSTATE_SCRATCH0(r13)
>> +
>> +.balign IFETCH_ALIGN_BYTES
>> +kvmppc_p9_exit_interrupt:
[...]
>> +static inline void slb_invalidate(unsigned int ih)
>> +{
>> + asm volatile("slbia %0" :: "i"(ih));
>> +}
>
> This one is not used.
It gets used in a later patch, I guess I should move it there.
[...]
>> +int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu)
>> +{
>> + u64 *exsave;
>> + unsigned long msr = mfmsr();
>> + int trap;
>> +
>> + start_timing(vcpu, &vcpu->arch.rm_entry);
>> +
>> + vcpu->arch.ceded = 0;
>> +
>> + WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
>> + WARN_ON_ONCE(!(vcpu->arch.shregs.msr & MSR_ME));
>> +
>> + mtspr(SPRN_HSRR0, vcpu->arch.regs.nip);
>> + mtspr(SPRN_HSRR1, (vcpu->arch.shregs.msr & ~MSR_HV) | MSR_ME);
>> +
>> + /*
>> + * On POWER9 DD2.1 and below, sometimes on a Hypervisor Data Storage
>> + * Interrupt (HDSI) the HDSISR is not be updated at all.
>> + *
>> + * To work around this we put a canary value into the HDSISR before
>> + * returning to a guest and then check for this canary when we take a
>> + * HDSI. If we find the canary on a HDSI, we know the hardware didn't
>> + * update the HDSISR. In this case we return to the guest to retake the
>> + * HDSI which should correctly update the HDSISR the second time HDSI
>> + * entry.
>> + *
>> + * Just do this on all p9 processors for now.
>> + */
>> + mtspr(SPRN_HDSISR, HDSISR_CANARY);
>> +
>> + accumulate_time(vcpu, &vcpu->arch.guest_time);
>> +
>> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
>> + kvmppc_p9_enter_guest(vcpu);
>> + // Radix host and guest means host never runs with guest MMU state
>> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
>> +
>> + accumulate_time(vcpu, &vcpu->arch.rm_intr);
>> +
>> + /* Get these from r11/12 and paca exsave */
>> + vcpu->arch.shregs.srr0 = mfspr(SPRN_SRR0);
>> + vcpu->arch.shregs.srr1 = mfspr(SPRN_SRR1);
>> + vcpu->arch.shregs.dar = mfspr(SPRN_DAR);
>> + vcpu->arch.shregs.dsisr = mfspr(SPRN_DSISR);
>> +
>> + /* 0x2 bit for HSRR is only used by PR and P7/8 HV paths, clear it */
>> + trap = local_paca->kvm_hstate.scratch0 & ~0x2;
>> + if (likely(trap > BOOK3S_INTERRUPT_MACHINE_CHECK)) {
>> + exsave = local_paca->exgen;
>> + } else if (trap == BOOK3S_INTERRUPT_SYSTEM_RESET) {
>> + exsave = local_paca->exnmi;
>> + } else { /* trap == 0x200 */
>> + exsave = local_paca->exmc;
>> + }
>> +
>> + vcpu->arch.regs.gpr[1] = local_paca->kvm_hstate.scratch1;
>> + vcpu->arch.regs.gpr[3] = local_paca->kvm_hstate.scratch2;
>> + vcpu->arch.regs.gpr[9] = exsave[EX_R9/sizeof(u64)];
>> + vcpu->arch.regs.gpr[10] = exsave[EX_R10/sizeof(u64)];
>> + vcpu->arch.regs.gpr[11] = exsave[EX_R11/sizeof(u64)];
>> + vcpu->arch.regs.gpr[12] = exsave[EX_R12/sizeof(u64)];
>> + vcpu->arch.regs.gpr[13] = exsave[EX_R13/sizeof(u64)];
>> + vcpu->arch.ppr = exsave[EX_PPR/sizeof(u64)];
>> + vcpu->arch.cfar = exsave[EX_CFAR/sizeof(u64)];
>> + vcpu->arch.regs.ctr = exsave[EX_CTR/sizeof(u64)];
>> +
>> + vcpu->arch.last_inst = KVM_INST_FETCH_FAILED;
>> +
>> + if (unlikely(trap == BOOK3S_INTERRUPT_MACHINE_CHECK)) {
>> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
>> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
>> + kvmppc_realmode_machine_check(vcpu);
>> +
>> + } else if (unlikely(trap == BOOK3S_INTERRUPT_HMI)) {
>> + kvmppc_realmode_hmi_handler();
>> +
>> + } else if (trap == BOOK3S_INTERRUPT_H_EMUL_ASSIST) {
>> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
>> +
>> + } else if (trap == BOOK3S_INTERRUPT_H_DATA_STORAGE) {
>> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
>> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
>> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
>> +
>> + } else if (trap == BOOK3S_INTERRUPT_H_INST_STORAGE) {
>> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
>> +
>> + } else if (trap == BOOK3S_INTERRUPT_H_FAC_UNAVAIL) {
>> + vcpu->arch.hfscr = mfspr(SPRN_HFSCR);
>> +
>> +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
>> + /*
>> + * Softpatch interrupt for transactional memory emulation cases
>> + * on POWER9 DD2.2. This is early in the guest exit path - we
>> + * haven't saved registers or done a treclaim yet.
>> + */
>> + } else if (trap == BOOK3S_INTERRUPT_HV_SOFTPATCH) {
>> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
>> +
>> + /*
>> + * The cases we want to handle here are those where the guest
>> + * is in real suspend mode and is trying to transition to
>> + * transactional mode.
>> + */
>> + if (local_paca->kvm_hstate.fake_suspend &&
>> + (vcpu->arch.shregs.msr & MSR_TS_S)) {
>> + if (kvmhv_p9_tm_emulation_early(vcpu)) {
>> + /* Prevent it being handled again. */
>> + trap = 0;
>> + }
>> + }
>> +#endif
>> + }
>> +
>> + radix_clear_slb();
>> +
>> + __mtmsrd(msr, 0);
>
>
> The asm code only sets RI but this potentially sets more bits including
> MSR_EE, is it expected to be 0 when __kvmhv_vcpu_entry_p9() is called?
Yes.
>> + mtspr(SPRN_CTRLT, 1);
>
> What is this for? ISA does not shed much light:
> ===
> 63 RUN This bit controls an external I/O pin.
> ===
I don't think it even does that these days. It interacts with the PMU.
I was looking whether it's feasible to move it into PMU code entirely,
but apparently some tool or something might sample it. I'm a bit
suspicious about that because an untrusted guest could be running and
claim not to so I don't know what said tool really achieves, but I'll
go through that fight another day.
But KVM has to set it to 1 at exit because Linux host has it set to 1
except in CPU idle.
>
>
>> +
>> + accumulate_time(vcpu, &vcpu->arch.rm_exit);
>
> This should not compile without CONFIG_KVM_BOOK3S_HV_EXIT_TIMING.
It has an ifdef wrapper so it should work (it does on my local tree
which is slightly newer than what you have but I don't think I fixed
anything around this recently).
>> +
>> + end_timing(vcpu);
>> +
>> + return trap;
>
>
> The asm does "For hash guest, read the guest SLB and save it away", this
> code does not. Is this new fast-path-in-c only for radix-on-radix or
> hash VMs are supported too?
That asm code does not run for "guest_exit_short_path" case (aka the
p9 path aka the fast path).
Upstream code only supports radix host and radix guest in this path.
The old path supports hash and radix. That's unchanged with this patch.
After the series, the new path supports all P9 modes (hash/hash,
radix/radix, and radix/hash), and the old path supports P7 and P8 only.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 29/46] KVM: PPC: Book3S HV P9: Implement the rest of the P9 path in C
@ 2021-04-01 10:35 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-01 10:35 UTC (permalink / raw)
To: Alexey Kardashevskiy, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Alexey Kardashevskiy's message of April 1, 2021 3:30 pm:
>
>
> On 3/23/21 12:02 PM, Nicholas Piggin wrote:
>> Almost all logic is moved to C, by introducing a new in_guest mode that
>> selects and branches very early in the interrupt handler to the P9 exit
>> code.
[...]
>> +/*
>> + * kvmppc_p9_exit_hcall and kvmppc_p9_exit_interrupt are branched to from
>> + * above if the interrupt was taken for a guest that was entered via
>> + * kvmppc_p9_enter_guest().
>> + *
>> + * This code recovers the host stack and vcpu pointer, saves all GPRs and
>> + * CR, LR, CTR, XER as well as guest MSR and NIA into the VCPU, then re-
>> + * establishes the host stack and registers to return from the
>> + * kvmppc_p9_enter_guest() function.
>
> What does "this code" refer to? If it is the asm below, then it does not
> save CTR, it is in the c code. Otherwise it is confusing (to me) :)
Yes you're right, CTR is saved in C.
>> + */
>> +.balign IFETCH_ALIGN_BYTES
>> +kvmppc_p9_exit_hcall:
>> + mfspr r11,SPRN_SRR0
>> + mfspr r12,SPRN_SRR1
>> + li r10,0xc00
>> + std r10,HSTATE_SCRATCH0(r13)
>> +
>> +.balign IFETCH_ALIGN_BYTES
>> +kvmppc_p9_exit_interrupt:
[...]
>> +static inline void slb_invalidate(unsigned int ih)
>> +{
>> + asm volatile("slbia %0" :: "i"(ih));
>> +}
>
> This one is not used.
It gets used in a later patch, I guess I should move it there.
[...]
>> +int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu)
>> +{
>> + u64 *exsave;
>> + unsigned long msr = mfmsr();
>> + int trap;
>> +
>> + start_timing(vcpu, &vcpu->arch.rm_entry);
>> +
>> + vcpu->arch.ceded = 0;
>> +
>> + WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
>> + WARN_ON_ONCE(!(vcpu->arch.shregs.msr & MSR_ME));
>> +
>> + mtspr(SPRN_HSRR0, vcpu->arch.regs.nip);
>> + mtspr(SPRN_HSRR1, (vcpu->arch.shregs.msr & ~MSR_HV) | MSR_ME);
>> +
>> + /*
>> + * On POWER9 DD2.1 and below, sometimes on a Hypervisor Data Storage
>> + * Interrupt (HDSI) the HDSISR is not be updated at all.
>> + *
>> + * To work around this we put a canary value into the HDSISR before
>> + * returning to a guest and then check for this canary when we take a
>> + * HDSI. If we find the canary on a HDSI, we know the hardware didn't
>> + * update the HDSISR. In this case we return to the guest to retake the
>> + * HDSI which should correctly update the HDSISR the second time HDSI
>> + * entry.
>> + *
>> + * Just do this on all p9 processors for now.
>> + */
>> + mtspr(SPRN_HDSISR, HDSISR_CANARY);
>> +
>> + accumulate_time(vcpu, &vcpu->arch.guest_time);
>> +
>> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
>> + kvmppc_p9_enter_guest(vcpu);
>> + // Radix host and guest means host never runs with guest MMU state
>> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
>> +
>> + accumulate_time(vcpu, &vcpu->arch.rm_intr);
>> +
>> + /* Get these from r11/12 and paca exsave */
>> + vcpu->arch.shregs.srr0 = mfspr(SPRN_SRR0);
>> + vcpu->arch.shregs.srr1 = mfspr(SPRN_SRR1);
>> + vcpu->arch.shregs.dar = mfspr(SPRN_DAR);
>> + vcpu->arch.shregs.dsisr = mfspr(SPRN_DSISR);
>> +
>> + /* 0x2 bit for HSRR is only used by PR and P7/8 HV paths, clear it */
>> + trap = local_paca->kvm_hstate.scratch0 & ~0x2;
>> + if (likely(trap > BOOK3S_INTERRUPT_MACHINE_CHECK)) {
>> + exsave = local_paca->exgen;
>> + } else if (trap == BOOK3S_INTERRUPT_SYSTEM_RESET) {
>> + exsave = local_paca->exnmi;
>> + } else { /* trap == 0x200 */
>> + exsave = local_paca->exmc;
>> + }
>> +
>> + vcpu->arch.regs.gpr[1] = local_paca->kvm_hstate.scratch1;
>> + vcpu->arch.regs.gpr[3] = local_paca->kvm_hstate.scratch2;
>> + vcpu->arch.regs.gpr[9] = exsave[EX_R9/sizeof(u64)];
>> + vcpu->arch.regs.gpr[10] = exsave[EX_R10/sizeof(u64)];
>> + vcpu->arch.regs.gpr[11] = exsave[EX_R11/sizeof(u64)];
>> + vcpu->arch.regs.gpr[12] = exsave[EX_R12/sizeof(u64)];
>> + vcpu->arch.regs.gpr[13] = exsave[EX_R13/sizeof(u64)];
>> + vcpu->arch.ppr = exsave[EX_PPR/sizeof(u64)];
>> + vcpu->arch.cfar = exsave[EX_CFAR/sizeof(u64)];
>> + vcpu->arch.regs.ctr = exsave[EX_CTR/sizeof(u64)];
>> +
>> + vcpu->arch.last_inst = KVM_INST_FETCH_FAILED;
>> +
>> + if (unlikely(trap == BOOK3S_INTERRUPT_MACHINE_CHECK)) {
>> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
>> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
>> + kvmppc_realmode_machine_check(vcpu);
>> +
>> + } else if (unlikely(trap == BOOK3S_INTERRUPT_HMI)) {
>> + kvmppc_realmode_hmi_handler();
>> +
>> + } else if (trap == BOOK3S_INTERRUPT_H_EMUL_ASSIST) {
>> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
>> +
>> + } else if (trap == BOOK3S_INTERRUPT_H_DATA_STORAGE) {
>> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
>> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
>> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
>> +
>> + } else if (trap == BOOK3S_INTERRUPT_H_INST_STORAGE) {
>> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
>> +
>> + } else if (trap == BOOK3S_INTERRUPT_H_FAC_UNAVAIL) {
>> + vcpu->arch.hfscr = mfspr(SPRN_HFSCR);
>> +
>> +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
>> + /*
>> + * Softpatch interrupt for transactional memory emulation cases
>> + * on POWER9 DD2.2. This is early in the guest exit path - we
>> + * haven't saved registers or done a treclaim yet.
>> + */
>> + } else if (trap == BOOK3S_INTERRUPT_HV_SOFTPATCH) {
>> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
>> +
>> + /*
>> + * The cases we want to handle here are those where the guest
>> + * is in real suspend mode and is trying to transition to
>> + * transactional mode.
>> + */
>> + if (local_paca->kvm_hstate.fake_suspend &&
>> + (vcpu->arch.shregs.msr & MSR_TS_S)) {
>> + if (kvmhv_p9_tm_emulation_early(vcpu)) {
>> + /* Prevent it being handled again. */
>> + trap = 0;
>> + }
>> + }
>> +#endif
>> + }
>> +
>> + radix_clear_slb();
>> +
>> + __mtmsrd(msr, 0);
>
>
> The asm code only sets RI but this potentially sets more bits including
> MSR_EE, is it expected to be 0 when __kvmhv_vcpu_entry_p9() is called?
Yes.
>> + mtspr(SPRN_CTRLT, 1);
>
> What is this for? ISA does not shed much light:
> ===
> 63 RUN This bit controls an external I/O pin.
> ===
I don't think it even does that these days. It interacts with the PMU.
I was looking whether it's feasible to move it into PMU code entirely,
but apparently some tool or something might sample it. I'm a bit
suspicious about that because an untrusted guest could be running and
claim not to so I don't know what said tool really achieves, but I'll
go through that fight another day.
But KVM has to set it to 1 at exit because Linux host has it set to 1
except in CPU idle.
>
>
>> +
>> + accumulate_time(vcpu, &vcpu->arch.rm_exit);
>
> This should not compile without CONFIG_KVM_BOOK3S_HV_EXIT_TIMING.
It has an ifdef wrapper so it should work (it does on my local tree
which is slightly newer than what you have but I don't think I fixed
anything around this recently).
>> +
>> + end_timing(vcpu);
>> +
>> + return trap;
>
>
> The asm does "For hash guest, read the guest SLB and save it away", this
> code does not. Is this new fast-path-in-c only for radix-on-radix or
> hash VMs are supported too?
That asm code does not run for "guest_exit_short_path" case (aka the
p9 path aka the fast path).
Upstream code only supports radix host and radix guest in this path.
The old path supports hash and radix. That's unchanged with this patch.
After the series, the new path supports all P9 modes (hash/hash,
radix/radix, and radix/hash), and the old path supports P7 and P8 only.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 29/46] KVM: PPC: Book3S HV P9: Implement the rest of the P9 path in C
2021-04-01 10:35 ` Nicholas Piggin
@ 2021-04-02 4:36 ` Alexey Kardashevskiy
-1 siblings, 0 replies; 188+ messages in thread
From: Alexey Kardashevskiy @ 2021-04-02 4:36 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 01/04/2021 21:35, Nicholas Piggin wrote:
> Excerpts from Alexey Kardashevskiy's message of April 1, 2021 3:30 pm:
>>
>>
>> On 3/23/21 12:02 PM, Nicholas Piggin wrote:
>>> Almost all logic is moved to C, by introducing a new in_guest mode that
>>> selects and branches very early in the interrupt handler to the P9 exit
>>> code.
>
> [...]
>
>>> +/*
>>> + * kvmppc_p9_exit_hcall and kvmppc_p9_exit_interrupt are branched to from
>>> + * above if the interrupt was taken for a guest that was entered via
>>> + * kvmppc_p9_enter_guest().
>>> + *
>>> + * This code recovers the host stack and vcpu pointer, saves all GPRs and
>>> + * CR, LR, CTR, XER as well as guest MSR and NIA into the VCPU, then re-
>>> + * establishes the host stack and registers to return from the
>>> + * kvmppc_p9_enter_guest() function.
>>
>> What does "this code" refer to? If it is the asm below, then it does not
>> save CTR, it is in the c code. Otherwise it is confusing (to me) :)
>
> Yes you're right, CTR is saved in C.
>
>>> + */
>>> +.balign IFETCH_ALIGN_BYTES
>>> +kvmppc_p9_exit_hcall:
>>> + mfspr r11,SPRN_SRR0
>>> + mfspr r12,SPRN_SRR1
>>> + li r10,0xc00
>>> + std r10,HSTATE_SCRATCH0(r13)
>>> +
>>> +.balign IFETCH_ALIGN_BYTES
>>> +kvmppc_p9_exit_interrupt:
>
> [...]
>
>>> +static inline void slb_invalidate(unsigned int ih)
>>> +{
>>> + asm volatile("slbia %0" :: "i"(ih));
>>> +}
>>
>> This one is not used.
>
> It gets used in a later patch, I guess I should move it there.
>
> [...]
>
>>> +int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu)
>>> +{
>>> + u64 *exsave;
>>> + unsigned long msr = mfmsr();
>>> + int trap;
>>> +
>>> + start_timing(vcpu, &vcpu->arch.rm_entry);
>>> +
>>> + vcpu->arch.ceded = 0;
>>> +
>>> + WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
>>> + WARN_ON_ONCE(!(vcpu->arch.shregs.msr & MSR_ME));
>>> +
>>> + mtspr(SPRN_HSRR0, vcpu->arch.regs.nip);
>>> + mtspr(SPRN_HSRR1, (vcpu->arch.shregs.msr & ~MSR_HV) | MSR_ME);
>>> +
>>> + /*
>>> + * On POWER9 DD2.1 and below, sometimes on a Hypervisor Data Storage
>>> + * Interrupt (HDSI) the HDSISR is not be updated at all.
>>> + *
>>> + * To work around this we put a canary value into the HDSISR before
>>> + * returning to a guest and then check for this canary when we take a
>>> + * HDSI. If we find the canary on a HDSI, we know the hardware didn't
>>> + * update the HDSISR. In this case we return to the guest to retake the
>>> + * HDSI which should correctly update the HDSISR the second time HDSI
>>> + * entry.
>>> + *
>>> + * Just do this on all p9 processors for now.
>>> + */
>>> + mtspr(SPRN_HDSISR, HDSISR_CANARY);
>>> +
>>> + accumulate_time(vcpu, &vcpu->arch.guest_time);
>>> +
>>> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
>>> + kvmppc_p9_enter_guest(vcpu);
>>> + // Radix host and guest means host never runs with guest MMU state
>>> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
>>> +
>>> + accumulate_time(vcpu, &vcpu->arch.rm_intr);
>>> +
>>> + /* Get these from r11/12 and paca exsave */
>>> + vcpu->arch.shregs.srr0 = mfspr(SPRN_SRR0);
>>> + vcpu->arch.shregs.srr1 = mfspr(SPRN_SRR1);
>>> + vcpu->arch.shregs.dar = mfspr(SPRN_DAR);
>>> + vcpu->arch.shregs.dsisr = mfspr(SPRN_DSISR);
>>> +
>>> + /* 0x2 bit for HSRR is only used by PR and P7/8 HV paths, clear it */
>>> + trap = local_paca->kvm_hstate.scratch0 & ~0x2;
>>> + if (likely(trap > BOOK3S_INTERRUPT_MACHINE_CHECK)) {
>>> + exsave = local_paca->exgen;
>>> + } else if (trap == BOOK3S_INTERRUPT_SYSTEM_RESET) {
>>> + exsave = local_paca->exnmi;
>>> + } else { /* trap == 0x200 */
>>> + exsave = local_paca->exmc;
>>> + }
>>> +
>>> + vcpu->arch.regs.gpr[1] = local_paca->kvm_hstate.scratch1;
>>> + vcpu->arch.regs.gpr[3] = local_paca->kvm_hstate.scratch2;
>>> + vcpu->arch.regs.gpr[9] = exsave[EX_R9/sizeof(u64)];
>>> + vcpu->arch.regs.gpr[10] = exsave[EX_R10/sizeof(u64)];
>>> + vcpu->arch.regs.gpr[11] = exsave[EX_R11/sizeof(u64)];
>>> + vcpu->arch.regs.gpr[12] = exsave[EX_R12/sizeof(u64)];
>>> + vcpu->arch.regs.gpr[13] = exsave[EX_R13/sizeof(u64)];
>>> + vcpu->arch.ppr = exsave[EX_PPR/sizeof(u64)];
>>> + vcpu->arch.cfar = exsave[EX_CFAR/sizeof(u64)];
>>> + vcpu->arch.regs.ctr = exsave[EX_CTR/sizeof(u64)];
>>> +
>>> + vcpu->arch.last_inst = KVM_INST_FETCH_FAILED;
>>> +
>>> + if (unlikely(trap == BOOK3S_INTERRUPT_MACHINE_CHECK)) {
>>> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
>>> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
>>> + kvmppc_realmode_machine_check(vcpu);
>>> +
>>> + } else if (unlikely(trap == BOOK3S_INTERRUPT_HMI)) {
>>> + kvmppc_realmode_hmi_handler();
>>> +
>>> + } else if (trap == BOOK3S_INTERRUPT_H_EMUL_ASSIST) {
>>> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
>>> +
>>> + } else if (trap == BOOK3S_INTERRUPT_H_DATA_STORAGE) {
>>> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
>>> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
>>> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
>>> +
>>> + } else if (trap == BOOK3S_INTERRUPT_H_INST_STORAGE) {
>>> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
>>> +
>>> + } else if (trap == BOOK3S_INTERRUPT_H_FAC_UNAVAIL) {
>>> + vcpu->arch.hfscr = mfspr(SPRN_HFSCR);
>>> +
>>> +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
>>> + /*
>>> + * Softpatch interrupt for transactional memory emulation cases
>>> + * on POWER9 DD2.2. This is early in the guest exit path - we
>>> + * haven't saved registers or done a treclaim yet.
>>> + */
>>> + } else if (trap == BOOK3S_INTERRUPT_HV_SOFTPATCH) {
>>> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
>>> +
>>> + /*
>>> + * The cases we want to handle here are those where the guest
>>> + * is in real suspend mode and is trying to transition to
>>> + * transactional mode.
>>> + */
>>> + if (local_paca->kvm_hstate.fake_suspend &&
>>> + (vcpu->arch.shregs.msr & MSR_TS_S)) {
>>> + if (kvmhv_p9_tm_emulation_early(vcpu)) {
>>> + /* Prevent it being handled again. */
>>> + trap = 0;
>>> + }
>>> + }
>>> +#endif
>>> + }
>>> +
>>> + radix_clear_slb();
>>> +
>>> + __mtmsrd(msr, 0);
>>
>>
>> The asm code only sets RI but this potentially sets more bits including
>> MSR_EE, is it expected to be 0 when __kvmhv_vcpu_entry_p9() is called?
>
> Yes.
>
>>> + mtspr(SPRN_CTRLT, 1);
>>
>> What is this for? ISA does not shed much light:
>> ===
>> 63 RUN This bit controls an external I/O pin.
>> ===
>
> I don't think it even does that these days. It interacts with the PMU.
> I was looking whether it's feasible to move it into PMU code entirely,
> but apparently some tool or something might sample it. I'm a bit
> suspicious about that because an untrusted guest could be running and
> claim not to so I don't know what said tool really achieves, but I'll
> go through that fight another day.
>
> But KVM has to set it to 1 at exit because Linux host has it set to 1
> except in CPU idle.
It this CTRLT setting a new thing or the asm does it too? I could not
spot it.
>>
>>
>>> +
>>> + accumulate_time(vcpu, &vcpu->arch.rm_exit);
>>
>> This should not compile without CONFIG_KVM_BOOK3S_HV_EXIT_TIMING.
>
> It has an ifdef wrapper so it should work (it does on my local tree
> which is slightly newer than what you have but I don't think I fixed
> anything around this recently).
You are absolutely right, my bad.
>
>>> +
>>> + end_timing(vcpu);
>>> +
>>> + return trap;
>>
>>
>> The asm does "For hash guest, read the guest SLB and save it away", this
>> code does not. Is this new fast-path-in-c only for radix-on-radix or
>> hash VMs are supported too?
>
> That asm code does not run for "guest_exit_short_path" case (aka the
> p9 path aka the fast path).
>
> Upstream code only supports radix host and radix guest in this path.
> The old path supports hash and radix. That's unchanged with this patch.
>
> After the series, the new path supports all P9 modes (hash/hash,
> radix/radix, and radix/hash), and the old path supports P7 and P8 only.
Thanks for clarification. Besides that CTRLT, I checked if the new c
code matches the old asm code (which made diving into ISA incredible fun
:) ) so fwiw
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
I'd really like to see longer commit logs clarifying all intended
changes but it is probably just me.
>
> Thanks,
> Nick
>
--
Alexey
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 29/46] KVM: PPC: Book3S HV P9: Implement the rest of the P9 path in C
@ 2021-04-02 4:36 ` Alexey Kardashevskiy
0 siblings, 0 replies; 188+ messages in thread
From: Alexey Kardashevskiy @ 2021-04-02 4:36 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 01/04/2021 21:35, Nicholas Piggin wrote:
> Excerpts from Alexey Kardashevskiy's message of April 1, 2021 3:30 pm:
>>
>>
>> On 3/23/21 12:02 PM, Nicholas Piggin wrote:
>>> Almost all logic is moved to C, by introducing a new in_guest mode that
>>> selects and branches very early in the interrupt handler to the P9 exit
>>> code.
>
> [...]
>
>>> +/*
>>> + * kvmppc_p9_exit_hcall and kvmppc_p9_exit_interrupt are branched to from
>>> + * above if the interrupt was taken for a guest that was entered via
>>> + * kvmppc_p9_enter_guest().
>>> + *
>>> + * This code recovers the host stack and vcpu pointer, saves all GPRs and
>>> + * CR, LR, CTR, XER as well as guest MSR and NIA into the VCPU, then re-
>>> + * establishes the host stack and registers to return from the
>>> + * kvmppc_p9_enter_guest() function.
>>
>> What does "this code" refer to? If it is the asm below, then it does not
>> save CTR, it is in the c code. Otherwise it is confusing (to me) :)
>
> Yes you're right, CTR is saved in C.
>
>>> + */
>>> +.balign IFETCH_ALIGN_BYTES
>>> +kvmppc_p9_exit_hcall:
>>> + mfspr r11,SPRN_SRR0
>>> + mfspr r12,SPRN_SRR1
>>> + li r10,0xc00
>>> + std r10,HSTATE_SCRATCH0(r13)
>>> +
>>> +.balign IFETCH_ALIGN_BYTES
>>> +kvmppc_p9_exit_interrupt:
>
> [...]
>
>>> +static inline void slb_invalidate(unsigned int ih)
>>> +{
>>> + asm volatile("slbia %0" :: "i"(ih));
>>> +}
>>
>> This one is not used.
>
> It gets used in a later patch, I guess I should move it there.
>
> [...]
>
>>> +int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu)
>>> +{
>>> + u64 *exsave;
>>> + unsigned long msr = mfmsr();
>>> + int trap;
>>> +
>>> + start_timing(vcpu, &vcpu->arch.rm_entry);
>>> +
>>> + vcpu->arch.ceded = 0;
>>> +
>>> + WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
>>> + WARN_ON_ONCE(!(vcpu->arch.shregs.msr & MSR_ME));
>>> +
>>> + mtspr(SPRN_HSRR0, vcpu->arch.regs.nip);
>>> + mtspr(SPRN_HSRR1, (vcpu->arch.shregs.msr & ~MSR_HV) | MSR_ME);
>>> +
>>> + /*
>>> + * On POWER9 DD2.1 and below, sometimes on a Hypervisor Data Storage
>>> + * Interrupt (HDSI) the HDSISR is not be updated at all.
>>> + *
>>> + * To work around this we put a canary value into the HDSISR before
>>> + * returning to a guest and then check for this canary when we take a
>>> + * HDSI. If we find the canary on a HDSI, we know the hardware didn't
>>> + * update the HDSISR. In this case we return to the guest to retake the
>>> + * HDSI which should correctly update the HDSISR the second time HDSI
>>> + * entry.
>>> + *
>>> + * Just do this on all p9 processors for now.
>>> + */
>>> + mtspr(SPRN_HDSISR, HDSISR_CANARY);
>>> +
>>> + accumulate_time(vcpu, &vcpu->arch.guest_time);
>>> +
>>> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
>>> + kvmppc_p9_enter_guest(vcpu);
>>> + // Radix host and guest means host never runs with guest MMU state
>>> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
>>> +
>>> + accumulate_time(vcpu, &vcpu->arch.rm_intr);
>>> +
>>> + /* Get these from r11/12 and paca exsave */
>>> + vcpu->arch.shregs.srr0 = mfspr(SPRN_SRR0);
>>> + vcpu->arch.shregs.srr1 = mfspr(SPRN_SRR1);
>>> + vcpu->arch.shregs.dar = mfspr(SPRN_DAR);
>>> + vcpu->arch.shregs.dsisr = mfspr(SPRN_DSISR);
>>> +
>>> + /* 0x2 bit for HSRR is only used by PR and P7/8 HV paths, clear it */
>>> + trap = local_paca->kvm_hstate.scratch0 & ~0x2;
>>> + if (likely(trap > BOOK3S_INTERRUPT_MACHINE_CHECK)) {
>>> + exsave = local_paca->exgen;
>>> + } else if (trap = BOOK3S_INTERRUPT_SYSTEM_RESET) {
>>> + exsave = local_paca->exnmi;
>>> + } else { /* trap = 0x200 */
>>> + exsave = local_paca->exmc;
>>> + }
>>> +
>>> + vcpu->arch.regs.gpr[1] = local_paca->kvm_hstate.scratch1;
>>> + vcpu->arch.regs.gpr[3] = local_paca->kvm_hstate.scratch2;
>>> + vcpu->arch.regs.gpr[9] = exsave[EX_R9/sizeof(u64)];
>>> + vcpu->arch.regs.gpr[10] = exsave[EX_R10/sizeof(u64)];
>>> + vcpu->arch.regs.gpr[11] = exsave[EX_R11/sizeof(u64)];
>>> + vcpu->arch.regs.gpr[12] = exsave[EX_R12/sizeof(u64)];
>>> + vcpu->arch.regs.gpr[13] = exsave[EX_R13/sizeof(u64)];
>>> + vcpu->arch.ppr = exsave[EX_PPR/sizeof(u64)];
>>> + vcpu->arch.cfar = exsave[EX_CFAR/sizeof(u64)];
>>> + vcpu->arch.regs.ctr = exsave[EX_CTR/sizeof(u64)];
>>> +
>>> + vcpu->arch.last_inst = KVM_INST_FETCH_FAILED;
>>> +
>>> + if (unlikely(trap = BOOK3S_INTERRUPT_MACHINE_CHECK)) {
>>> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
>>> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
>>> + kvmppc_realmode_machine_check(vcpu);
>>> +
>>> + } else if (unlikely(trap = BOOK3S_INTERRUPT_HMI)) {
>>> + kvmppc_realmode_hmi_handler();
>>> +
>>> + } else if (trap = BOOK3S_INTERRUPT_H_EMUL_ASSIST) {
>>> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
>>> +
>>> + } else if (trap = BOOK3S_INTERRUPT_H_DATA_STORAGE) {
>>> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
>>> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
>>> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
>>> +
>>> + } else if (trap = BOOK3S_INTERRUPT_H_INST_STORAGE) {
>>> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
>>> +
>>> + } else if (trap = BOOK3S_INTERRUPT_H_FAC_UNAVAIL) {
>>> + vcpu->arch.hfscr = mfspr(SPRN_HFSCR);
>>> +
>>> +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
>>> + /*
>>> + * Softpatch interrupt for transactional memory emulation cases
>>> + * on POWER9 DD2.2. This is early in the guest exit path - we
>>> + * haven't saved registers or done a treclaim yet.
>>> + */
>>> + } else if (trap = BOOK3S_INTERRUPT_HV_SOFTPATCH) {
>>> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
>>> +
>>> + /*
>>> + * The cases we want to handle here are those where the guest
>>> + * is in real suspend mode and is trying to transition to
>>> + * transactional mode.
>>> + */
>>> + if (local_paca->kvm_hstate.fake_suspend &&
>>> + (vcpu->arch.shregs.msr & MSR_TS_S)) {
>>> + if (kvmhv_p9_tm_emulation_early(vcpu)) {
>>> + /* Prevent it being handled again. */
>>> + trap = 0;
>>> + }
>>> + }
>>> +#endif
>>> + }
>>> +
>>> + radix_clear_slb();
>>> +
>>> + __mtmsrd(msr, 0);
>>
>>
>> The asm code only sets RI but this potentially sets more bits including
>> MSR_EE, is it expected to be 0 when __kvmhv_vcpu_entry_p9() is called?
>
> Yes.
>
>>> + mtspr(SPRN_CTRLT, 1);
>>
>> What is this for? ISA does not shed much light:
>> =>> 63 RUN This bit controls an external I/O pin.
>> =>
> I don't think it even does that these days. It interacts with the PMU.
> I was looking whether it's feasible to move it into PMU code entirely,
> but apparently some tool or something might sample it. I'm a bit
> suspicious about that because an untrusted guest could be running and
> claim not to so I don't know what said tool really achieves, but I'll
> go through that fight another day.
>
> But KVM has to set it to 1 at exit because Linux host has it set to 1
> except in CPU idle.
It this CTRLT setting a new thing or the asm does it too? I could not
spot it.
>>
>>
>>> +
>>> + accumulate_time(vcpu, &vcpu->arch.rm_exit);
>>
>> This should not compile without CONFIG_KVM_BOOK3S_HV_EXIT_TIMING.
>
> It has an ifdef wrapper so it should work (it does on my local tree
> which is slightly newer than what you have but I don't think I fixed
> anything around this recently).
You are absolutely right, my bad.
>
>>> +
>>> + end_timing(vcpu);
>>> +
>>> + return trap;
>>
>>
>> The asm does "For hash guest, read the guest SLB and save it away", this
>> code does not. Is this new fast-path-in-c only for radix-on-radix or
>> hash VMs are supported too?
>
> That asm code does not run for "guest_exit_short_path" case (aka the
> p9 path aka the fast path).
>
> Upstream code only supports radix host and radix guest in this path.
> The old path supports hash and radix. That's unchanged with this patch.
>
> After the series, the new path supports all P9 modes (hash/hash,
> radix/radix, and radix/hash), and the old path supports P7 and P8 only.
Thanks for clarification. Besides that CTRLT, I checked if the new c
code matches the old asm code (which made diving into ISA incredible fun
:) ) so fwiw
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
I'd really like to see longer commit logs clarifying all intended
changes but it is probably just me.
>
> Thanks,
> Nick
>
--
Alexey
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 29/46] KVM: PPC: Book3S HV P9: Implement the rest of the P9 path in C
2021-04-02 4:36 ` Alexey Kardashevskiy
@ 2021-04-02 7:58 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-02 7:58 UTC (permalink / raw)
To: Alexey Kardashevskiy, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Alexey Kardashevskiy's message of April 2, 2021 2:36 pm:
>
>
> On 01/04/2021 21:35, Nicholas Piggin wrote:
>> Excerpts from Alexey Kardashevskiy's message of April 1, 2021 3:30 pm:
>>>
>>>
>>> On 3/23/21 12:02 PM, Nicholas Piggin wrote:
>>>> Almost all logic is moved to C, by introducing a new in_guest mode that
>>>> selects and branches very early in the interrupt handler to the P9 exit
>>>> code.
>>
>> [...]
>>
>>>> +/*
>>>> + * kvmppc_p9_exit_hcall and kvmppc_p9_exit_interrupt are branched to from
>>>> + * above if the interrupt was taken for a guest that was entered via
>>>> + * kvmppc_p9_enter_guest().
>>>> + *
>>>> + * This code recovers the host stack and vcpu pointer, saves all GPRs and
>>>> + * CR, LR, CTR, XER as well as guest MSR and NIA into the VCPU, then re-
>>>> + * establishes the host stack and registers to return from the
>>>> + * kvmppc_p9_enter_guest() function.
>>>
>>> What does "this code" refer to? If it is the asm below, then it does not
>>> save CTR, it is in the c code. Otherwise it is confusing (to me) :)
>>
>> Yes you're right, CTR is saved in C.
>>
>>>> + */
>>>> +.balign IFETCH_ALIGN_BYTES
>>>> +kvmppc_p9_exit_hcall:
>>>> + mfspr r11,SPRN_SRR0
>>>> + mfspr r12,SPRN_SRR1
>>>> + li r10,0xc00
>>>> + std r10,HSTATE_SCRATCH0(r13)
>>>> +
>>>> +.balign IFETCH_ALIGN_BYTES
>>>> +kvmppc_p9_exit_interrupt:
>>
>> [...]
>>
>>>> +static inline void slb_invalidate(unsigned int ih)
>>>> +{
>>>> + asm volatile("slbia %0" :: "i"(ih));
>>>> +}
>>>
>>> This one is not used.
>>
>> It gets used in a later patch, I guess I should move it there.
>>
>> [...]
>>
>>>> +int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu)
>>>> +{
>>>> + u64 *exsave;
>>>> + unsigned long msr = mfmsr();
>>>> + int trap;
>>>> +
>>>> + start_timing(vcpu, &vcpu->arch.rm_entry);
>>>> +
>>>> + vcpu->arch.ceded = 0;
>>>> +
>>>> + WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
>>>> + WARN_ON_ONCE(!(vcpu->arch.shregs.msr & MSR_ME));
>>>> +
>>>> + mtspr(SPRN_HSRR0, vcpu->arch.regs.nip);
>>>> + mtspr(SPRN_HSRR1, (vcpu->arch.shregs.msr & ~MSR_HV) | MSR_ME);
>>>> +
>>>> + /*
>>>> + * On POWER9 DD2.1 and below, sometimes on a Hypervisor Data Storage
>>>> + * Interrupt (HDSI) the HDSISR is not be updated at all.
>>>> + *
>>>> + * To work around this we put a canary value into the HDSISR before
>>>> + * returning to a guest and then check for this canary when we take a
>>>> + * HDSI. If we find the canary on a HDSI, we know the hardware didn't
>>>> + * update the HDSISR. In this case we return to the guest to retake the
>>>> + * HDSI which should correctly update the HDSISR the second time HDSI
>>>> + * entry.
>>>> + *
>>>> + * Just do this on all p9 processors for now.
>>>> + */
>>>> + mtspr(SPRN_HDSISR, HDSISR_CANARY);
>>>> +
>>>> + accumulate_time(vcpu, &vcpu->arch.guest_time);
>>>> +
>>>> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
>>>> + kvmppc_p9_enter_guest(vcpu);
>>>> + // Radix host and guest means host never runs with guest MMU state
>>>> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
>>>> +
>>>> + accumulate_time(vcpu, &vcpu->arch.rm_intr);
>>>> +
>>>> + /* Get these from r11/12 and paca exsave */
>>>> + vcpu->arch.shregs.srr0 = mfspr(SPRN_SRR0);
>>>> + vcpu->arch.shregs.srr1 = mfspr(SPRN_SRR1);
>>>> + vcpu->arch.shregs.dar = mfspr(SPRN_DAR);
>>>> + vcpu->arch.shregs.dsisr = mfspr(SPRN_DSISR);
>>>> +
>>>> + /* 0x2 bit for HSRR is only used by PR and P7/8 HV paths, clear it */
>>>> + trap = local_paca->kvm_hstate.scratch0 & ~0x2;
>>>> + if (likely(trap > BOOK3S_INTERRUPT_MACHINE_CHECK)) {
>>>> + exsave = local_paca->exgen;
>>>> + } else if (trap == BOOK3S_INTERRUPT_SYSTEM_RESET) {
>>>> + exsave = local_paca->exnmi;
>>>> + } else { /* trap == 0x200 */
>>>> + exsave = local_paca->exmc;
>>>> + }
>>>> +
>>>> + vcpu->arch.regs.gpr[1] = local_paca->kvm_hstate.scratch1;
>>>> + vcpu->arch.regs.gpr[3] = local_paca->kvm_hstate.scratch2;
>>>> + vcpu->arch.regs.gpr[9] = exsave[EX_R9/sizeof(u64)];
>>>> + vcpu->arch.regs.gpr[10] = exsave[EX_R10/sizeof(u64)];
>>>> + vcpu->arch.regs.gpr[11] = exsave[EX_R11/sizeof(u64)];
>>>> + vcpu->arch.regs.gpr[12] = exsave[EX_R12/sizeof(u64)];
>>>> + vcpu->arch.regs.gpr[13] = exsave[EX_R13/sizeof(u64)];
>>>> + vcpu->arch.ppr = exsave[EX_PPR/sizeof(u64)];
>>>> + vcpu->arch.cfar = exsave[EX_CFAR/sizeof(u64)];
>>>> + vcpu->arch.regs.ctr = exsave[EX_CTR/sizeof(u64)];
>>>> +
>>>> + vcpu->arch.last_inst = KVM_INST_FETCH_FAILED;
>>>> +
>>>> + if (unlikely(trap == BOOK3S_INTERRUPT_MACHINE_CHECK)) {
>>>> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
>>>> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
>>>> + kvmppc_realmode_machine_check(vcpu);
>>>> +
>>>> + } else if (unlikely(trap == BOOK3S_INTERRUPT_HMI)) {
>>>> + kvmppc_realmode_hmi_handler();
>>>> +
>>>> + } else if (trap == BOOK3S_INTERRUPT_H_EMUL_ASSIST) {
>>>> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
>>>> +
>>>> + } else if (trap == BOOK3S_INTERRUPT_H_DATA_STORAGE) {
>>>> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
>>>> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
>>>> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
>>>> +
>>>> + } else if (trap == BOOK3S_INTERRUPT_H_INST_STORAGE) {
>>>> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
>>>> +
>>>> + } else if (trap == BOOK3S_INTERRUPT_H_FAC_UNAVAIL) {
>>>> + vcpu->arch.hfscr = mfspr(SPRN_HFSCR);
>>>> +
>>>> +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
>>>> + /*
>>>> + * Softpatch interrupt for transactional memory emulation cases
>>>> + * on POWER9 DD2.2. This is early in the guest exit path - we
>>>> + * haven't saved registers or done a treclaim yet.
>>>> + */
>>>> + } else if (trap == BOOK3S_INTERRUPT_HV_SOFTPATCH) {
>>>> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
>>>> +
>>>> + /*
>>>> + * The cases we want to handle here are those where the guest
>>>> + * is in real suspend mode and is trying to transition to
>>>> + * transactional mode.
>>>> + */
>>>> + if (local_paca->kvm_hstate.fake_suspend &&
>>>> + (vcpu->arch.shregs.msr & MSR_TS_S)) {
>>>> + if (kvmhv_p9_tm_emulation_early(vcpu)) {
>>>> + /* Prevent it being handled again. */
>>>> + trap = 0;
>>>> + }
>>>> + }
>>>> +#endif
>>>> + }
>>>> +
>>>> + radix_clear_slb();
>>>> +
>>>> + __mtmsrd(msr, 0);
>>>
>>>
>>> The asm code only sets RI but this potentially sets more bits including
>>> MSR_EE, is it expected to be 0 when __kvmhv_vcpu_entry_p9() is called?
>>
>> Yes.
>>
>>>> + mtspr(SPRN_CTRLT, 1);
>>>
>>> What is this for? ISA does not shed much light:
>>> ===
>>> 63 RUN This bit controls an external I/O pin.
>>> ===
>>
>> I don't think it even does that these days. It interacts with the PMU.
>> I was looking whether it's feasible to move it into PMU code entirely,
>> but apparently some tool or something might sample it. I'm a bit
>> suspicious about that because an untrusted guest could be running and
>> claim not to so I don't know what said tool really achieves, but I'll
>> go through that fight another day.
>>
>> But KVM has to set it to 1 at exit because Linux host has it set to 1
>> except in CPU idle.
>
>
> It this CTRLT setting a new thing or the asm does it too? I could not
> spot it.
It's quite old actually. Earlier processors (maybe POWER6) you had to
even read-modify-write but new ones you can just store 1:
Guest exit:
/* Save guest CTRL register, set runlatch to 1 */
mfspr r6,SPRN_CTRLF
stw r6,VCPU_CTRL(r9)
andi. r0,r6,1
bne 4f
ori r6,r6,1
mtspr SPRN_CTRLT,r6
4:
entry:
/* Restore state of CTRL run bit; assume 1 on entry */
lwz r5,VCPU_CTRL(r4)
andi. r5,r5,1
bne 4f
mfspr r6,SPRN_CTRLF
clrrdi r6,r6,1
mtspr SPRN_CTRLT,r6
It used to light an indicator on the front of the system once upon a
time and I think on some processors (Cell maybe?) it actually controlled
SMT threads in some way. But certainly in P9 it does almost nothing and
we'll probably try to phase it out.
>>> The asm does "For hash guest, read the guest SLB and save it away", this
>>> code does not. Is this new fast-path-in-c only for radix-on-radix or
>>> hash VMs are supported too?
>>
>> That asm code does not run for "guest_exit_short_path" case (aka the
>> p9 path aka the fast path).
>>
>> Upstream code only supports radix host and radix guest in this path.
>> The old path supports hash and radix. That's unchanged with this patch.
>>
>> After the series, the new path supports all P9 modes (hash/hash,
>> radix/radix, and radix/hash), and the old path supports P7 and P8 only.
>
>
> Thanks for clarification. Besides that CTRLT, I checked if the new c
> code matches the old asm code (which made diving into ISA incredible fun
> :) ) so fwiw
>
> Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Thanks for reviewing.
> I'd really like to see longer commit logs clarifying all intended
> changes but it is probably just me.
I'm not sure what the best balance is, at some point code is a more
precise description. For this particular patch I probably do need to go
over the changelog again and try to make sure it makes sense and covers
things. If you have specifics that are missing or changes you'd like
I would definitely consider them.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 29/46] KVM: PPC: Book3S HV P9: Implement the rest of the P9 path in C
@ 2021-04-02 7:58 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-04-02 7:58 UTC (permalink / raw)
To: Alexey Kardashevskiy, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Alexey Kardashevskiy's message of April 2, 2021 2:36 pm:
>
>
> On 01/04/2021 21:35, Nicholas Piggin wrote:
>> Excerpts from Alexey Kardashevskiy's message of April 1, 2021 3:30 pm:
>>>
>>>
>>> On 3/23/21 12:02 PM, Nicholas Piggin wrote:
>>>> Almost all logic is moved to C, by introducing a new in_guest mode that
>>>> selects and branches very early in the interrupt handler to the P9 exit
>>>> code.
>>
>> [...]
>>
>>>> +/*
>>>> + * kvmppc_p9_exit_hcall and kvmppc_p9_exit_interrupt are branched to from
>>>> + * above if the interrupt was taken for a guest that was entered via
>>>> + * kvmppc_p9_enter_guest().
>>>> + *
>>>> + * This code recovers the host stack and vcpu pointer, saves all GPRs and
>>>> + * CR, LR, CTR, XER as well as guest MSR and NIA into the VCPU, then re-
>>>> + * establishes the host stack and registers to return from the
>>>> + * kvmppc_p9_enter_guest() function.
>>>
>>> What does "this code" refer to? If it is the asm below, then it does not
>>> save CTR, it is in the c code. Otherwise it is confusing (to me) :)
>>
>> Yes you're right, CTR is saved in C.
>>
>>>> + */
>>>> +.balign IFETCH_ALIGN_BYTES
>>>> +kvmppc_p9_exit_hcall:
>>>> + mfspr r11,SPRN_SRR0
>>>> + mfspr r12,SPRN_SRR1
>>>> + li r10,0xc00
>>>> + std r10,HSTATE_SCRATCH0(r13)
>>>> +
>>>> +.balign IFETCH_ALIGN_BYTES
>>>> +kvmppc_p9_exit_interrupt:
>>
>> [...]
>>
>>>> +static inline void slb_invalidate(unsigned int ih)
>>>> +{
>>>> + asm volatile("slbia %0" :: "i"(ih));
>>>> +}
>>>
>>> This one is not used.
>>
>> It gets used in a later patch, I guess I should move it there.
>>
>> [...]
>>
>>>> +int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu)
>>>> +{
>>>> + u64 *exsave;
>>>> + unsigned long msr = mfmsr();
>>>> + int trap;
>>>> +
>>>> + start_timing(vcpu, &vcpu->arch.rm_entry);
>>>> +
>>>> + vcpu->arch.ceded = 0;
>>>> +
>>>> + WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
>>>> + WARN_ON_ONCE(!(vcpu->arch.shregs.msr & MSR_ME));
>>>> +
>>>> + mtspr(SPRN_HSRR0, vcpu->arch.regs.nip);
>>>> + mtspr(SPRN_HSRR1, (vcpu->arch.shregs.msr & ~MSR_HV) | MSR_ME);
>>>> +
>>>> + /*
>>>> + * On POWER9 DD2.1 and below, sometimes on a Hypervisor Data Storage
>>>> + * Interrupt (HDSI) the HDSISR is not be updated at all.
>>>> + *
>>>> + * To work around this we put a canary value into the HDSISR before
>>>> + * returning to a guest and then check for this canary when we take a
>>>> + * HDSI. If we find the canary on a HDSI, we know the hardware didn't
>>>> + * update the HDSISR. In this case we return to the guest to retake the
>>>> + * HDSI which should correctly update the HDSISR the second time HDSI
>>>> + * entry.
>>>> + *
>>>> + * Just do this on all p9 processors for now.
>>>> + */
>>>> + mtspr(SPRN_HDSISR, HDSISR_CANARY);
>>>> +
>>>> + accumulate_time(vcpu, &vcpu->arch.guest_time);
>>>> +
>>>> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
>>>> + kvmppc_p9_enter_guest(vcpu);
>>>> + // Radix host and guest means host never runs with guest MMU state
>>>> + local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
>>>> +
>>>> + accumulate_time(vcpu, &vcpu->arch.rm_intr);
>>>> +
>>>> + /* Get these from r11/12 and paca exsave */
>>>> + vcpu->arch.shregs.srr0 = mfspr(SPRN_SRR0);
>>>> + vcpu->arch.shregs.srr1 = mfspr(SPRN_SRR1);
>>>> + vcpu->arch.shregs.dar = mfspr(SPRN_DAR);
>>>> + vcpu->arch.shregs.dsisr = mfspr(SPRN_DSISR);
>>>> +
>>>> + /* 0x2 bit for HSRR is only used by PR and P7/8 HV paths, clear it */
>>>> + trap = local_paca->kvm_hstate.scratch0 & ~0x2;
>>>> + if (likely(trap > BOOK3S_INTERRUPT_MACHINE_CHECK)) {
>>>> + exsave = local_paca->exgen;
>>>> + } else if (trap == BOOK3S_INTERRUPT_SYSTEM_RESET) {
>>>> + exsave = local_paca->exnmi;
>>>> + } else { /* trap == 0x200 */
>>>> + exsave = local_paca->exmc;
>>>> + }
>>>> +
>>>> + vcpu->arch.regs.gpr[1] = local_paca->kvm_hstate.scratch1;
>>>> + vcpu->arch.regs.gpr[3] = local_paca->kvm_hstate.scratch2;
>>>> + vcpu->arch.regs.gpr[9] = exsave[EX_R9/sizeof(u64)];
>>>> + vcpu->arch.regs.gpr[10] = exsave[EX_R10/sizeof(u64)];
>>>> + vcpu->arch.regs.gpr[11] = exsave[EX_R11/sizeof(u64)];
>>>> + vcpu->arch.regs.gpr[12] = exsave[EX_R12/sizeof(u64)];
>>>> + vcpu->arch.regs.gpr[13] = exsave[EX_R13/sizeof(u64)];
>>>> + vcpu->arch.ppr = exsave[EX_PPR/sizeof(u64)];
>>>> + vcpu->arch.cfar = exsave[EX_CFAR/sizeof(u64)];
>>>> + vcpu->arch.regs.ctr = exsave[EX_CTR/sizeof(u64)];
>>>> +
>>>> + vcpu->arch.last_inst = KVM_INST_FETCH_FAILED;
>>>> +
>>>> + if (unlikely(trap == BOOK3S_INTERRUPT_MACHINE_CHECK)) {
>>>> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
>>>> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
>>>> + kvmppc_realmode_machine_check(vcpu);
>>>> +
>>>> + } else if (unlikely(trap == BOOK3S_INTERRUPT_HMI)) {
>>>> + kvmppc_realmode_hmi_handler();
>>>> +
>>>> + } else if (trap == BOOK3S_INTERRUPT_H_EMUL_ASSIST) {
>>>> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
>>>> +
>>>> + } else if (trap == BOOK3S_INTERRUPT_H_DATA_STORAGE) {
>>>> + vcpu->arch.fault_dar = exsave[EX_DAR/sizeof(u64)];
>>>> + vcpu->arch.fault_dsisr = exsave[EX_DSISR/sizeof(u64)];
>>>> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
>>>> +
>>>> + } else if (trap == BOOK3S_INTERRUPT_H_INST_STORAGE) {
>>>> + vcpu->arch.fault_gpa = mfspr(SPRN_ASDR);
>>>> +
>>>> + } else if (trap == BOOK3S_INTERRUPT_H_FAC_UNAVAIL) {
>>>> + vcpu->arch.hfscr = mfspr(SPRN_HFSCR);
>>>> +
>>>> +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
>>>> + /*
>>>> + * Softpatch interrupt for transactional memory emulation cases
>>>> + * on POWER9 DD2.2. This is early in the guest exit path - we
>>>> + * haven't saved registers or done a treclaim yet.
>>>> + */
>>>> + } else if (trap == BOOK3S_INTERRUPT_HV_SOFTPATCH) {
>>>> + vcpu->arch.emul_inst = mfspr(SPRN_HEIR);
>>>> +
>>>> + /*
>>>> + * The cases we want to handle here are those where the guest
>>>> + * is in real suspend mode and is trying to transition to
>>>> + * transactional mode.
>>>> + */
>>>> + if (local_paca->kvm_hstate.fake_suspend &&
>>>> + (vcpu->arch.shregs.msr & MSR_TS_S)) {
>>>> + if (kvmhv_p9_tm_emulation_early(vcpu)) {
>>>> + /* Prevent it being handled again. */
>>>> + trap = 0;
>>>> + }
>>>> + }
>>>> +#endif
>>>> + }
>>>> +
>>>> + radix_clear_slb();
>>>> +
>>>> + __mtmsrd(msr, 0);
>>>
>>>
>>> The asm code only sets RI but this potentially sets more bits including
>>> MSR_EE, is it expected to be 0 when __kvmhv_vcpu_entry_p9() is called?
>>
>> Yes.
>>
>>>> + mtspr(SPRN_CTRLT, 1);
>>>
>>> What is this for? ISA does not shed much light:
>>> ===
>>> 63 RUN This bit controls an external I/O pin.
>>> ===
>>
>> I don't think it even does that these days. It interacts with the PMU.
>> I was looking whether it's feasible to move it into PMU code entirely,
>> but apparently some tool or something might sample it. I'm a bit
>> suspicious about that because an untrusted guest could be running and
>> claim not to so I don't know what said tool really achieves, but I'll
>> go through that fight another day.
>>
>> But KVM has to set it to 1 at exit because Linux host has it set to 1
>> except in CPU idle.
>
>
> It this CTRLT setting a new thing or the asm does it too? I could not
> spot it.
It's quite old actually. Earlier processors (maybe POWER6) you had to
even read-modify-write but new ones you can just store 1:
Guest exit:
/* Save guest CTRL register, set runlatch to 1 */
mfspr r6,SPRN_CTRLF
stw r6,VCPU_CTRL(r9)
andi. r0,r6,1
bne 4f
ori r6,r6,1
mtspr SPRN_CTRLT,r6
4:
entry:
/* Restore state of CTRL run bit; assume 1 on entry */
lwz r5,VCPU_CTRL(r4)
andi. r5,r5,1
bne 4f
mfspr r6,SPRN_CTRLF
clrrdi r6,r6,1
mtspr SPRN_CTRLT,r6
It used to light an indicator on the front of the system once upon a
time and I think on some processors (Cell maybe?) it actually controlled
SMT threads in some way. But certainly in P9 it does almost nothing and
we'll probably try to phase it out.
>>> The asm does "For hash guest, read the guest SLB and save it away", this
>>> code does not. Is this new fast-path-in-c only for radix-on-radix or
>>> hash VMs are supported too?
>>
>> That asm code does not run for "guest_exit_short_path" case (aka the
>> p9 path aka the fast path).
>>
>> Upstream code only supports radix host and radix guest in this path.
>> The old path supports hash and radix. That's unchanged with this patch.
>>
>> After the series, the new path supports all P9 modes (hash/hash,
>> radix/radix, and radix/hash), and the old path supports P7 and P8 only.
>
>
> Thanks for clarification. Besides that CTRLT, I checked if the new c
> code matches the old asm code (which made diving into ISA incredible fun
> :) ) so fwiw
>
> Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Thanks for reviewing.
> I'd really like to see longer commit logs clarifying all intended
> changes but it is probably just me.
I'm not sure what the best balance is, at some point code is a more
precise description. For this particular patch I probably do need to go
over the changelog again and try to make sure it makes sense and covers
things. If you have specifics that are missing or changes you'd like
I would definitely consider them.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 30/46] KVM: PPC: Book3S HV P9: inline kvmhv_load_hv_regs_and_go into __kvmhv_vcpu_entry_p9
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin, Fabiano Rosas
Now the initial C implementation is done, inline more HV code to make
rearranging things easier.
And rename __kvmhv_vcpu_entry_p9 to drop the leading underscores as it's
now C, and is now a more complete vcpu entry.
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_book3s_64.h | 2 +-
arch/powerpc/kvm/book3s_hv.c | 192 +----------------------
arch/powerpc/kvm/book3s_hv_interrupt.c | 179 ++++++++++++++++++++-
3 files changed, 180 insertions(+), 193 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_book3s_64.h b/arch/powerpc/include/asm/kvm_book3s_64.h
index c214bcffb441..eaf3a562bf1e 100644
--- a/arch/powerpc/include/asm/kvm_book3s_64.h
+++ b/arch/powerpc/include/asm/kvm_book3s_64.h
@@ -153,7 +153,7 @@ static inline bool kvmhv_vcpu_is_radix(struct kvm_vcpu *vcpu)
return radix;
}
-int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu);
+int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpcr);
#define KVM_DEFAULT_HPT_ORDER 24 /* 16MB HPT by default */
#endif
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 6dca639ed997..52f844a06899 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3481,194 +3481,6 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
trace_kvmppc_run_core(vc, 1);
}
-static void switch_mmu_to_guest_radix(struct kvm *kvm, struct kvm_vcpu *vcpu, u64 lpcr)
-{
- struct kvmppc_vcore *vc = vcpu->arch.vcore;
- struct kvm_nested_guest *nested = vcpu->arch.nested;
- u32 lpid;
-
- lpid = nested ? nested->shadow_lpid : kvm->arch.lpid;
-
- /*
- * All the isync()s are overkill but trivially follow the ISA
- * requirements. Some can likely be replaced with justification
- * comment for why they are not needed.
- */
- isync();
- mtspr(SPRN_LPID, lpid);
- isync();
- mtspr(SPRN_LPCR, lpcr);
- isync();
- mtspr(SPRN_PID, vcpu->arch.pid);
- isync();
-
- /* TLBIEL must have LPIDR set, so set guest LPID before flushing. */
- kvmppc_check_need_tlb_flush(kvm, vc->pcpu, nested);
-}
-
-static void switch_mmu_to_host_radix(struct kvm *kvm, u32 pid)
-{
- isync();
- mtspr(SPRN_PID, pid);
- isync();
- mtspr(SPRN_LPID, kvm->arch.host_lpid);
- isync();
- mtspr(SPRN_LPCR, kvm->arch.host_lpcr);
- isync();
-}
-
-/*
- * Load up hypervisor-mode registers on P9.
- */
-static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
- unsigned long lpcr)
-{
- struct kvm *kvm = vcpu->kvm;
- struct kvmppc_vcore *vc = vcpu->arch.vcore;
- s64 hdec;
- u64 tb, purr, spurr;
- int trap;
- unsigned long host_hfscr = mfspr(SPRN_HFSCR);
- unsigned long host_ciabr = mfspr(SPRN_CIABR);
- unsigned long host_dawr0 = mfspr(SPRN_DAWR0);
- unsigned long host_dawrx0 = mfspr(SPRN_DAWRX0);
- unsigned long host_psscr = mfspr(SPRN_PSSCR);
- unsigned long host_pidr = mfspr(SPRN_PID);
- unsigned long host_dawr1 = 0;
- unsigned long host_dawrx1 = 0;
-
- if (cpu_has_feature(CPU_FTR_DAWR1)) {
- host_dawr1 = mfspr(SPRN_DAWR1);
- host_dawrx1 = mfspr(SPRN_DAWRX1);
- }
-
- tb = mftb();
- hdec = time_limit - tb;
- if (hdec < 0)
- return BOOK3S_INTERRUPT_HV_DECREMENTER;
-
- if (vc->tb_offset) {
- u64 new_tb = tb + vc->tb_offset;
- mtspr(SPRN_TBU40, new_tb);
- tb = mftb();
- if ((tb & 0xffffff) < (new_tb & 0xffffff))
- mtspr(SPRN_TBU40, new_tb + 0x1000000);
- vc->tb_offset_applied = vc->tb_offset;
- }
-
- if (vc->pcr)
- mtspr(SPRN_PCR, vc->pcr | PCR_MASK);
- mtspr(SPRN_DPDES, vc->dpdes);
- mtspr(SPRN_VTB, vc->vtb);
-
- local_paca->kvm_hstate.host_purr = mfspr(SPRN_PURR);
- local_paca->kvm_hstate.host_spurr = mfspr(SPRN_SPURR);
- mtspr(SPRN_PURR, vcpu->arch.purr);
- mtspr(SPRN_SPURR, vcpu->arch.spurr);
-
- if (dawr_enabled()) {
- mtspr(SPRN_DAWR0, vcpu->arch.dawr0);
- mtspr(SPRN_DAWRX0, vcpu->arch.dawrx0);
- if (cpu_has_feature(CPU_FTR_DAWR1)) {
- mtspr(SPRN_DAWR1, vcpu->arch.dawr1);
- mtspr(SPRN_DAWRX1, vcpu->arch.dawrx1);
- }
- }
- mtspr(SPRN_CIABR, vcpu->arch.ciabr);
- mtspr(SPRN_IC, vcpu->arch.ic);
-
- mtspr(SPRN_PSSCR, vcpu->arch.psscr | PSSCR_EC |
- (local_paca->kvm_hstate.fake_suspend << PSSCR_FAKE_SUSPEND_LG));
-
- mtspr(SPRN_HFSCR, vcpu->arch.hfscr);
-
- mtspr(SPRN_SPRG0, vcpu->arch.shregs.sprg0);
- mtspr(SPRN_SPRG1, vcpu->arch.shregs.sprg1);
- mtspr(SPRN_SPRG2, vcpu->arch.shregs.sprg2);
- mtspr(SPRN_SPRG3, vcpu->arch.shregs.sprg3);
-
- mtspr(SPRN_AMOR, ~0UL);
-
- switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
-
- /*
- * P9 suppresses the HDEC exception when LPCR[HDICE] = 0,
- * so set guest LPCR (with HDICE) before writing HDEC.
- */
- mtspr(SPRN_HDEC, hdec);
-
- mtspr(SPRN_SRR0, vcpu->arch.shregs.srr0);
- mtspr(SPRN_SRR1, vcpu->arch.shregs.srr1);
-
- trap = __kvmhv_vcpu_entry_p9(vcpu);
-
- /* Advance host PURR/SPURR by the amount used by guest */
- purr = mfspr(SPRN_PURR);
- spurr = mfspr(SPRN_SPURR);
- mtspr(SPRN_PURR, local_paca->kvm_hstate.host_purr +
- purr - vcpu->arch.purr);
- mtspr(SPRN_SPURR, local_paca->kvm_hstate.host_spurr +
- spurr - vcpu->arch.spurr);
- vcpu->arch.purr = purr;
- vcpu->arch.spurr = spurr;
-
- vcpu->arch.ic = mfspr(SPRN_IC);
- vcpu->arch.pid = mfspr(SPRN_PID);
- vcpu->arch.psscr = mfspr(SPRN_PSSCR) & PSSCR_GUEST_VIS;
-
- vcpu->arch.shregs.sprg0 = mfspr(SPRN_SPRG0);
- vcpu->arch.shregs.sprg1 = mfspr(SPRN_SPRG1);
- vcpu->arch.shregs.sprg2 = mfspr(SPRN_SPRG2);
- vcpu->arch.shregs.sprg3 = mfspr(SPRN_SPRG3);
-
- /* Preserve PSSCR[FAKE_SUSPEND] until we've called kvmppc_save_tm_hv */
- mtspr(SPRN_PSSCR, host_psscr |
- (local_paca->kvm_hstate.fake_suspend << PSSCR_FAKE_SUSPEND_LG));
- mtspr(SPRN_HFSCR, host_hfscr);
- mtspr(SPRN_CIABR, host_ciabr);
- mtspr(SPRN_DAWR0, host_dawr0);
- mtspr(SPRN_DAWRX0, host_dawrx0);
- if (cpu_has_feature(CPU_FTR_DAWR1)) {
- mtspr(SPRN_DAWR1, host_dawr1);
- mtspr(SPRN_DAWRX1, host_dawrx1);
- }
-
- /*
- * Since this is radix, do a eieio; tlbsync; ptesync sequence in
- * case we interrupted the guest between a tlbie and a ptesync.
- */
- asm volatile("eieio; tlbsync; ptesync");
-
- /*
- * cp_abort is required if the processor supports local copy-paste
- * to clear the copy buffer that was under control of the guest.
- */
- if (cpu_has_feature(CPU_FTR_ARCH_31))
- asm volatile(PPC_CP_ABORT);
-
- vc->dpdes = mfspr(SPRN_DPDES);
- vc->vtb = mfspr(SPRN_VTB);
- mtspr(SPRN_DPDES, 0);
- if (vc->pcr)
- mtspr(SPRN_PCR, PCR_MASK);
-
- if (vc->tb_offset_applied) {
- u64 new_tb = mftb() - vc->tb_offset_applied;
- mtspr(SPRN_TBU40, new_tb);
- tb = mftb();
- if ((tb & 0xffffff) < (new_tb & 0xffffff))
- mtspr(SPRN_TBU40, new_tb + 0x1000000);
- vc->tb_offset_applied = 0;
- }
-
- /* HDEC must be at least as large as DEC, so decrementer_max fits */
- mtspr(SPRN_HDEC, decrementer_max);
-
- switch_mmu_to_host_radix(kvm, host_pidr);
-
- return trap;
-}
-
static inline bool hcall_is_xics(unsigned long req)
{
return (req == H_EOI || req == H_CPPR || req == H_IPI ||
@@ -3766,7 +3578,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
* We need to save and restore the guest visible part of the
* psscr (i.e. using SPRN_PSSCR_PR) since the hypervisor
* doesn't do this for us. Note only required if pseries since
- * this is done in kvmhv_load_hv_regs_and_go() below otherwise.
+ * this is done in kvmhv_vcpu_entry_p9() below otherwise.
*/
unsigned long host_psscr;
/* call our hypervisor to load up HV regs and go */
@@ -3804,7 +3616,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
}
} else {
kvmppc_xive_push_vcpu(vcpu);
- trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
+ trap = kvmhv_vcpu_entry_p9(vcpu, time_limit, lpcr);
if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
!(vcpu->arch.shregs.msr & MSR_PR)) {
unsigned long req = kvmppc_get_gpr(vcpu, 3);
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index 3151b3d62c01..c1105a4cc9b7 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -55,6 +55,42 @@ static void __accumulate_time(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator
#define accumulate_time(vcpu, next) do {} while (0)
#endif
+static void switch_mmu_to_guest_radix(struct kvm *kvm, struct kvm_vcpu *vcpu, u64 lpcr)
+{
+ struct kvmppc_vcore *vc = vcpu->arch.vcore;
+ struct kvm_nested_guest *nested = vcpu->arch.nested;
+ u32 lpid;
+
+ lpid = nested ? nested->shadow_lpid : kvm->arch.lpid;
+
+ /*
+ * All the isync()s are overkill but trivially follow the ISA
+ * requirements. Some can likely be replaced with justification
+ * comment for why they are not needed.
+ */
+ isync();
+ mtspr(SPRN_LPID, lpid);
+ isync();
+ mtspr(SPRN_LPCR, lpcr);
+ isync();
+ mtspr(SPRN_PID, vcpu->arch.pid);
+ isync();
+
+ /* TLBIEL must have LPIDR set, so set guest LPID before flushing. */
+ kvmppc_check_need_tlb_flush(kvm, vc->pcpu, nested);
+}
+
+static void switch_mmu_to_host_radix(struct kvm *kvm, u32 pid)
+{
+ isync();
+ mtspr(SPRN_PID, pid);
+ isync();
+ mtspr(SPRN_LPID, kvm->arch.host_lpid);
+ isync();
+ mtspr(SPRN_LPCR, kvm->arch.host_lpcr);
+ isync();
+}
+
static inline void mfslb(unsigned int idx, u64 *slbee, u64 *slbev)
{
asm volatile("slbmfev %0,%1" : "=r" (*slbev) : "r" (idx));
@@ -94,11 +130,86 @@ static void radix_clear_slb(void)
}
}
-int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu)
+int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpcr)
{
+ struct kvm *kvm = vcpu->kvm;
+ struct kvmppc_vcore *vc = vcpu->arch.vcore;
+ s64 hdec;
+ u64 tb, purr, spurr;
u64 *exsave;
unsigned long msr = mfmsr();
int trap;
+ unsigned long host_hfscr = mfspr(SPRN_HFSCR);
+ unsigned long host_ciabr = mfspr(SPRN_CIABR);
+ unsigned long host_dawr0 = mfspr(SPRN_DAWR0);
+ unsigned long host_dawrx0 = mfspr(SPRN_DAWRX0);
+ unsigned long host_psscr = mfspr(SPRN_PSSCR);
+ unsigned long host_pidr = mfspr(SPRN_PID);
+ unsigned long host_dawr1 = 0;
+ unsigned long host_dawrx1 = 0;
+
+ if (cpu_has_feature(CPU_FTR_DAWR1)) {
+ host_dawr1 = mfspr(SPRN_DAWR1);
+ host_dawrx1 = mfspr(SPRN_DAWRX1);
+ }
+
+ tb = mftb();
+ hdec = time_limit - tb;
+ if (hdec < 0)
+ return BOOK3S_INTERRUPT_HV_DECREMENTER;
+
+ if (vc->tb_offset) {
+ u64 new_tb = tb + vc->tb_offset;
+ mtspr(SPRN_TBU40, new_tb);
+ tb = mftb();
+ if ((tb & 0xffffff) < (new_tb & 0xffffff))
+ mtspr(SPRN_TBU40, new_tb + 0x1000000);
+ vc->tb_offset_applied = vc->tb_offset;
+ }
+
+ if (vc->pcr)
+ mtspr(SPRN_PCR, vc->pcr | PCR_MASK);
+ mtspr(SPRN_DPDES, vc->dpdes);
+ mtspr(SPRN_VTB, vc->vtb);
+
+ local_paca->kvm_hstate.host_purr = mfspr(SPRN_PURR);
+ local_paca->kvm_hstate.host_spurr = mfspr(SPRN_SPURR);
+ mtspr(SPRN_PURR, vcpu->arch.purr);
+ mtspr(SPRN_SPURR, vcpu->arch.spurr);
+
+ if (dawr_enabled()) {
+ mtspr(SPRN_DAWR0, vcpu->arch.dawr0);
+ mtspr(SPRN_DAWRX0, vcpu->arch.dawrx0);
+ if (cpu_has_feature(CPU_FTR_DAWR1)) {
+ mtspr(SPRN_DAWR1, vcpu->arch.dawr1);
+ mtspr(SPRN_DAWRX1, vcpu->arch.dawrx1);
+ }
+ }
+ mtspr(SPRN_CIABR, vcpu->arch.ciabr);
+ mtspr(SPRN_IC, vcpu->arch.ic);
+
+ mtspr(SPRN_PSSCR, vcpu->arch.psscr | PSSCR_EC |
+ (local_paca->kvm_hstate.fake_suspend << PSSCR_FAKE_SUSPEND_LG));
+
+ mtspr(SPRN_HFSCR, vcpu->arch.hfscr);
+
+ mtspr(SPRN_SPRG0, vcpu->arch.shregs.sprg0);
+ mtspr(SPRN_SPRG1, vcpu->arch.shregs.sprg1);
+ mtspr(SPRN_SPRG2, vcpu->arch.shregs.sprg2);
+ mtspr(SPRN_SPRG3, vcpu->arch.shregs.sprg3);
+
+ mtspr(SPRN_AMOR, ~0UL);
+
+ switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
+
+ /*
+ * P9 suppresses the HDEC exception when LPCR[HDICE] = 0,
+ * so set guest LPCR (with HDICE) before writing HDEC.
+ */
+ mtspr(SPRN_HDEC, hdec);
+
+ mtspr(SPRN_SRR0, vcpu->arch.shregs.srr0);
+ mtspr(SPRN_SRR1, vcpu->arch.shregs.srr1);
start_timing(vcpu, &vcpu->arch.rm_entry);
@@ -218,6 +329,70 @@ int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu)
end_timing(vcpu);
+ /* Advance host PURR/SPURR by the amount used by guest */
+ purr = mfspr(SPRN_PURR);
+ spurr = mfspr(SPRN_SPURR);
+ mtspr(SPRN_PURR, local_paca->kvm_hstate.host_purr +
+ purr - vcpu->arch.purr);
+ mtspr(SPRN_SPURR, local_paca->kvm_hstate.host_spurr +
+ spurr - vcpu->arch.spurr);
+ vcpu->arch.purr = purr;
+ vcpu->arch.spurr = spurr;
+
+ vcpu->arch.ic = mfspr(SPRN_IC);
+ vcpu->arch.pid = mfspr(SPRN_PID);
+ vcpu->arch.psscr = mfspr(SPRN_PSSCR) & PSSCR_GUEST_VIS;
+
+ vcpu->arch.shregs.sprg0 = mfspr(SPRN_SPRG0);
+ vcpu->arch.shregs.sprg1 = mfspr(SPRN_SPRG1);
+ vcpu->arch.shregs.sprg2 = mfspr(SPRN_SPRG2);
+ vcpu->arch.shregs.sprg3 = mfspr(SPRN_SPRG3);
+
+ /* Preserve PSSCR[FAKE_SUSPEND] until we've called kvmppc_save_tm_hv */
+ mtspr(SPRN_PSSCR, host_psscr |
+ (local_paca->kvm_hstate.fake_suspend << PSSCR_FAKE_SUSPEND_LG));
+ mtspr(SPRN_HFSCR, host_hfscr);
+ mtspr(SPRN_CIABR, host_ciabr);
+ mtspr(SPRN_DAWR0, host_dawr0);
+ mtspr(SPRN_DAWRX0, host_dawrx0);
+ if (cpu_has_feature(CPU_FTR_DAWR1)) {
+ mtspr(SPRN_DAWR1, host_dawr1);
+ mtspr(SPRN_DAWRX1, host_dawrx1);
+ }
+
+ /*
+ * Since this is radix, do a eieio; tlbsync; ptesync sequence in
+ * case we interrupted the guest between a tlbie and a ptesync.
+ */
+ asm volatile("eieio; tlbsync; ptesync");
+
+ /*
+ * cp_abort is required if the processor supports local copy-paste
+ * to clear the copy buffer that was under control of the guest.
+ */
+ if (cpu_has_feature(CPU_FTR_ARCH_31))
+ asm volatile(PPC_CP_ABORT);
+
+ vc->dpdes = mfspr(SPRN_DPDES);
+ vc->vtb = mfspr(SPRN_VTB);
+ mtspr(SPRN_DPDES, 0);
+ if (vc->pcr)
+ mtspr(SPRN_PCR, PCR_MASK);
+
+ if (vc->tb_offset_applied) {
+ u64 new_tb = mftb() - vc->tb_offset_applied;
+ mtspr(SPRN_TBU40, new_tb);
+ tb = mftb();
+ if ((tb & 0xffffff) < (new_tb & 0xffffff))
+ mtspr(SPRN_TBU40, new_tb + 0x1000000);
+ vc->tb_offset_applied = 0;
+ }
+
+ /* HDEC must be at least as large as DEC, so decrementer_max fits */
+ mtspr(SPRN_HDEC, decrementer_max);
+
+ switch_mmu_to_host_radix(kvm, host_pidr);
+
return trap;
}
-EXPORT_SYMBOL_GPL(__kvmhv_vcpu_entry_p9);
+EXPORT_SYMBOL_GPL(kvmhv_vcpu_entry_p9);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 30/46] KVM: PPC: Book3S HV P9: inline kvmhv_load_hv_regs_and_go into __kvmhv_vcpu_entry_p9
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin, Fabiano Rosas
Now the initial C implementation is done, inline more HV code to make
rearranging things easier.
And rename __kvmhv_vcpu_entry_p9 to drop the leading underscores as it's
now C, and is now a more complete vcpu entry.
Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_book3s_64.h | 2 +-
arch/powerpc/kvm/book3s_hv.c | 192 +----------------------
arch/powerpc/kvm/book3s_hv_interrupt.c | 179 ++++++++++++++++++++-
3 files changed, 180 insertions(+), 193 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_book3s_64.h b/arch/powerpc/include/asm/kvm_book3s_64.h
index c214bcffb441..eaf3a562bf1e 100644
--- a/arch/powerpc/include/asm/kvm_book3s_64.h
+++ b/arch/powerpc/include/asm/kvm_book3s_64.h
@@ -153,7 +153,7 @@ static inline bool kvmhv_vcpu_is_radix(struct kvm_vcpu *vcpu)
return radix;
}
-int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu);
+int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpcr);
#define KVM_DEFAULT_HPT_ORDER 24 /* 16MB HPT by default */
#endif
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 6dca639ed997..52f844a06899 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3481,194 +3481,6 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
trace_kvmppc_run_core(vc, 1);
}
-static void switch_mmu_to_guest_radix(struct kvm *kvm, struct kvm_vcpu *vcpu, u64 lpcr)
-{
- struct kvmppc_vcore *vc = vcpu->arch.vcore;
- struct kvm_nested_guest *nested = vcpu->arch.nested;
- u32 lpid;
-
- lpid = nested ? nested->shadow_lpid : kvm->arch.lpid;
-
- /*
- * All the isync()s are overkill but trivially follow the ISA
- * requirements. Some can likely be replaced with justification
- * comment for why they are not needed.
- */
- isync();
- mtspr(SPRN_LPID, lpid);
- isync();
- mtspr(SPRN_LPCR, lpcr);
- isync();
- mtspr(SPRN_PID, vcpu->arch.pid);
- isync();
-
- /* TLBIEL must have LPIDR set, so set guest LPID before flushing. */
- kvmppc_check_need_tlb_flush(kvm, vc->pcpu, nested);
-}
-
-static void switch_mmu_to_host_radix(struct kvm *kvm, u32 pid)
-{
- isync();
- mtspr(SPRN_PID, pid);
- isync();
- mtspr(SPRN_LPID, kvm->arch.host_lpid);
- isync();
- mtspr(SPRN_LPCR, kvm->arch.host_lpcr);
- isync();
-}
-
-/*
- * Load up hypervisor-mode registers on P9.
- */
-static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,
- unsigned long lpcr)
-{
- struct kvm *kvm = vcpu->kvm;
- struct kvmppc_vcore *vc = vcpu->arch.vcore;
- s64 hdec;
- u64 tb, purr, spurr;
- int trap;
- unsigned long host_hfscr = mfspr(SPRN_HFSCR);
- unsigned long host_ciabr = mfspr(SPRN_CIABR);
- unsigned long host_dawr0 = mfspr(SPRN_DAWR0);
- unsigned long host_dawrx0 = mfspr(SPRN_DAWRX0);
- unsigned long host_psscr = mfspr(SPRN_PSSCR);
- unsigned long host_pidr = mfspr(SPRN_PID);
- unsigned long host_dawr1 = 0;
- unsigned long host_dawrx1 = 0;
-
- if (cpu_has_feature(CPU_FTR_DAWR1)) {
- host_dawr1 = mfspr(SPRN_DAWR1);
- host_dawrx1 = mfspr(SPRN_DAWRX1);
- }
-
- tb = mftb();
- hdec = time_limit - tb;
- if (hdec < 0)
- return BOOK3S_INTERRUPT_HV_DECREMENTER;
-
- if (vc->tb_offset) {
- u64 new_tb = tb + vc->tb_offset;
- mtspr(SPRN_TBU40, new_tb);
- tb = mftb();
- if ((tb & 0xffffff) < (new_tb & 0xffffff))
- mtspr(SPRN_TBU40, new_tb + 0x1000000);
- vc->tb_offset_applied = vc->tb_offset;
- }
-
- if (vc->pcr)
- mtspr(SPRN_PCR, vc->pcr | PCR_MASK);
- mtspr(SPRN_DPDES, vc->dpdes);
- mtspr(SPRN_VTB, vc->vtb);
-
- local_paca->kvm_hstate.host_purr = mfspr(SPRN_PURR);
- local_paca->kvm_hstate.host_spurr = mfspr(SPRN_SPURR);
- mtspr(SPRN_PURR, vcpu->arch.purr);
- mtspr(SPRN_SPURR, vcpu->arch.spurr);
-
- if (dawr_enabled()) {
- mtspr(SPRN_DAWR0, vcpu->arch.dawr0);
- mtspr(SPRN_DAWRX0, vcpu->arch.dawrx0);
- if (cpu_has_feature(CPU_FTR_DAWR1)) {
- mtspr(SPRN_DAWR1, vcpu->arch.dawr1);
- mtspr(SPRN_DAWRX1, vcpu->arch.dawrx1);
- }
- }
- mtspr(SPRN_CIABR, vcpu->arch.ciabr);
- mtspr(SPRN_IC, vcpu->arch.ic);
-
- mtspr(SPRN_PSSCR, vcpu->arch.psscr | PSSCR_EC |
- (local_paca->kvm_hstate.fake_suspend << PSSCR_FAKE_SUSPEND_LG));
-
- mtspr(SPRN_HFSCR, vcpu->arch.hfscr);
-
- mtspr(SPRN_SPRG0, vcpu->arch.shregs.sprg0);
- mtspr(SPRN_SPRG1, vcpu->arch.shregs.sprg1);
- mtspr(SPRN_SPRG2, vcpu->arch.shregs.sprg2);
- mtspr(SPRN_SPRG3, vcpu->arch.shregs.sprg3);
-
- mtspr(SPRN_AMOR, ~0UL);
-
- switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
-
- /*
- * P9 suppresses the HDEC exception when LPCR[HDICE] = 0,
- * so set guest LPCR (with HDICE) before writing HDEC.
- */
- mtspr(SPRN_HDEC, hdec);
-
- mtspr(SPRN_SRR0, vcpu->arch.shregs.srr0);
- mtspr(SPRN_SRR1, vcpu->arch.shregs.srr1);
-
- trap = __kvmhv_vcpu_entry_p9(vcpu);
-
- /* Advance host PURR/SPURR by the amount used by guest */
- purr = mfspr(SPRN_PURR);
- spurr = mfspr(SPRN_SPURR);
- mtspr(SPRN_PURR, local_paca->kvm_hstate.host_purr +
- purr - vcpu->arch.purr);
- mtspr(SPRN_SPURR, local_paca->kvm_hstate.host_spurr +
- spurr - vcpu->arch.spurr);
- vcpu->arch.purr = purr;
- vcpu->arch.spurr = spurr;
-
- vcpu->arch.ic = mfspr(SPRN_IC);
- vcpu->arch.pid = mfspr(SPRN_PID);
- vcpu->arch.psscr = mfspr(SPRN_PSSCR) & PSSCR_GUEST_VIS;
-
- vcpu->arch.shregs.sprg0 = mfspr(SPRN_SPRG0);
- vcpu->arch.shregs.sprg1 = mfspr(SPRN_SPRG1);
- vcpu->arch.shregs.sprg2 = mfspr(SPRN_SPRG2);
- vcpu->arch.shregs.sprg3 = mfspr(SPRN_SPRG3);
-
- /* Preserve PSSCR[FAKE_SUSPEND] until we've called kvmppc_save_tm_hv */
- mtspr(SPRN_PSSCR, host_psscr |
- (local_paca->kvm_hstate.fake_suspend << PSSCR_FAKE_SUSPEND_LG));
- mtspr(SPRN_HFSCR, host_hfscr);
- mtspr(SPRN_CIABR, host_ciabr);
- mtspr(SPRN_DAWR0, host_dawr0);
- mtspr(SPRN_DAWRX0, host_dawrx0);
- if (cpu_has_feature(CPU_FTR_DAWR1)) {
- mtspr(SPRN_DAWR1, host_dawr1);
- mtspr(SPRN_DAWRX1, host_dawrx1);
- }
-
- /*
- * Since this is radix, do a eieio; tlbsync; ptesync sequence in
- * case we interrupted the guest between a tlbie and a ptesync.
- */
- asm volatile("eieio; tlbsync; ptesync");
-
- /*
- * cp_abort is required if the processor supports local copy-paste
- * to clear the copy buffer that was under control of the guest.
- */
- if (cpu_has_feature(CPU_FTR_ARCH_31))
- asm volatile(PPC_CP_ABORT);
-
- vc->dpdes = mfspr(SPRN_DPDES);
- vc->vtb = mfspr(SPRN_VTB);
- mtspr(SPRN_DPDES, 0);
- if (vc->pcr)
- mtspr(SPRN_PCR, PCR_MASK);
-
- if (vc->tb_offset_applied) {
- u64 new_tb = mftb() - vc->tb_offset_applied;
- mtspr(SPRN_TBU40, new_tb);
- tb = mftb();
- if ((tb & 0xffffff) < (new_tb & 0xffffff))
- mtspr(SPRN_TBU40, new_tb + 0x1000000);
- vc->tb_offset_applied = 0;
- }
-
- /* HDEC must be at least as large as DEC, so decrementer_max fits */
- mtspr(SPRN_HDEC, decrementer_max);
-
- switch_mmu_to_host_radix(kvm, host_pidr);
-
- return trap;
-}
-
static inline bool hcall_is_xics(unsigned long req)
{
return (req = H_EOI || req = H_CPPR || req = H_IPI ||
@@ -3766,7 +3578,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
* We need to save and restore the guest visible part of the
* psscr (i.e. using SPRN_PSSCR_PR) since the hypervisor
* doesn't do this for us. Note only required if pseries since
- * this is done in kvmhv_load_hv_regs_and_go() below otherwise.
+ * this is done in kvmhv_vcpu_entry_p9() below otherwise.
*/
unsigned long host_psscr;
/* call our hypervisor to load up HV regs and go */
@@ -3804,7 +3616,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
}
} else {
kvmppc_xive_push_vcpu(vcpu);
- trap = kvmhv_load_hv_regs_and_go(vcpu, time_limit, lpcr);
+ trap = kvmhv_vcpu_entry_p9(vcpu, time_limit, lpcr);
if (trap = BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
!(vcpu->arch.shregs.msr & MSR_PR)) {
unsigned long req = kvmppc_get_gpr(vcpu, 3);
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index 3151b3d62c01..c1105a4cc9b7 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -55,6 +55,42 @@ static void __accumulate_time(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator
#define accumulate_time(vcpu, next) do {} while (0)
#endif
+static void switch_mmu_to_guest_radix(struct kvm *kvm, struct kvm_vcpu *vcpu, u64 lpcr)
+{
+ struct kvmppc_vcore *vc = vcpu->arch.vcore;
+ struct kvm_nested_guest *nested = vcpu->arch.nested;
+ u32 lpid;
+
+ lpid = nested ? nested->shadow_lpid : kvm->arch.lpid;
+
+ /*
+ * All the isync()s are overkill but trivially follow the ISA
+ * requirements. Some can likely be replaced with justification
+ * comment for why they are not needed.
+ */
+ isync();
+ mtspr(SPRN_LPID, lpid);
+ isync();
+ mtspr(SPRN_LPCR, lpcr);
+ isync();
+ mtspr(SPRN_PID, vcpu->arch.pid);
+ isync();
+
+ /* TLBIEL must have LPIDR set, so set guest LPID before flushing. */
+ kvmppc_check_need_tlb_flush(kvm, vc->pcpu, nested);
+}
+
+static void switch_mmu_to_host_radix(struct kvm *kvm, u32 pid)
+{
+ isync();
+ mtspr(SPRN_PID, pid);
+ isync();
+ mtspr(SPRN_LPID, kvm->arch.host_lpid);
+ isync();
+ mtspr(SPRN_LPCR, kvm->arch.host_lpcr);
+ isync();
+}
+
static inline void mfslb(unsigned int idx, u64 *slbee, u64 *slbev)
{
asm volatile("slbmfev %0,%1" : "=r" (*slbev) : "r" (idx));
@@ -94,11 +130,86 @@ static void radix_clear_slb(void)
}
}
-int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu)
+int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpcr)
{
+ struct kvm *kvm = vcpu->kvm;
+ struct kvmppc_vcore *vc = vcpu->arch.vcore;
+ s64 hdec;
+ u64 tb, purr, spurr;
u64 *exsave;
unsigned long msr = mfmsr();
int trap;
+ unsigned long host_hfscr = mfspr(SPRN_HFSCR);
+ unsigned long host_ciabr = mfspr(SPRN_CIABR);
+ unsigned long host_dawr0 = mfspr(SPRN_DAWR0);
+ unsigned long host_dawrx0 = mfspr(SPRN_DAWRX0);
+ unsigned long host_psscr = mfspr(SPRN_PSSCR);
+ unsigned long host_pidr = mfspr(SPRN_PID);
+ unsigned long host_dawr1 = 0;
+ unsigned long host_dawrx1 = 0;
+
+ if (cpu_has_feature(CPU_FTR_DAWR1)) {
+ host_dawr1 = mfspr(SPRN_DAWR1);
+ host_dawrx1 = mfspr(SPRN_DAWRX1);
+ }
+
+ tb = mftb();
+ hdec = time_limit - tb;
+ if (hdec < 0)
+ return BOOK3S_INTERRUPT_HV_DECREMENTER;
+
+ if (vc->tb_offset) {
+ u64 new_tb = tb + vc->tb_offset;
+ mtspr(SPRN_TBU40, new_tb);
+ tb = mftb();
+ if ((tb & 0xffffff) < (new_tb & 0xffffff))
+ mtspr(SPRN_TBU40, new_tb + 0x1000000);
+ vc->tb_offset_applied = vc->tb_offset;
+ }
+
+ if (vc->pcr)
+ mtspr(SPRN_PCR, vc->pcr | PCR_MASK);
+ mtspr(SPRN_DPDES, vc->dpdes);
+ mtspr(SPRN_VTB, vc->vtb);
+
+ local_paca->kvm_hstate.host_purr = mfspr(SPRN_PURR);
+ local_paca->kvm_hstate.host_spurr = mfspr(SPRN_SPURR);
+ mtspr(SPRN_PURR, vcpu->arch.purr);
+ mtspr(SPRN_SPURR, vcpu->arch.spurr);
+
+ if (dawr_enabled()) {
+ mtspr(SPRN_DAWR0, vcpu->arch.dawr0);
+ mtspr(SPRN_DAWRX0, vcpu->arch.dawrx0);
+ if (cpu_has_feature(CPU_FTR_DAWR1)) {
+ mtspr(SPRN_DAWR1, vcpu->arch.dawr1);
+ mtspr(SPRN_DAWRX1, vcpu->arch.dawrx1);
+ }
+ }
+ mtspr(SPRN_CIABR, vcpu->arch.ciabr);
+ mtspr(SPRN_IC, vcpu->arch.ic);
+
+ mtspr(SPRN_PSSCR, vcpu->arch.psscr | PSSCR_EC |
+ (local_paca->kvm_hstate.fake_suspend << PSSCR_FAKE_SUSPEND_LG));
+
+ mtspr(SPRN_HFSCR, vcpu->arch.hfscr);
+
+ mtspr(SPRN_SPRG0, vcpu->arch.shregs.sprg0);
+ mtspr(SPRN_SPRG1, vcpu->arch.shregs.sprg1);
+ mtspr(SPRN_SPRG2, vcpu->arch.shregs.sprg2);
+ mtspr(SPRN_SPRG3, vcpu->arch.shregs.sprg3);
+
+ mtspr(SPRN_AMOR, ~0UL);
+
+ switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
+
+ /*
+ * P9 suppresses the HDEC exception when LPCR[HDICE] = 0,
+ * so set guest LPCR (with HDICE) before writing HDEC.
+ */
+ mtspr(SPRN_HDEC, hdec);
+
+ mtspr(SPRN_SRR0, vcpu->arch.shregs.srr0);
+ mtspr(SPRN_SRR1, vcpu->arch.shregs.srr1);
start_timing(vcpu, &vcpu->arch.rm_entry);
@@ -218,6 +329,70 @@ int __kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu)
end_timing(vcpu);
+ /* Advance host PURR/SPURR by the amount used by guest */
+ purr = mfspr(SPRN_PURR);
+ spurr = mfspr(SPRN_SPURR);
+ mtspr(SPRN_PURR, local_paca->kvm_hstate.host_purr +
+ purr - vcpu->arch.purr);
+ mtspr(SPRN_SPURR, local_paca->kvm_hstate.host_spurr +
+ spurr - vcpu->arch.spurr);
+ vcpu->arch.purr = purr;
+ vcpu->arch.spurr = spurr;
+
+ vcpu->arch.ic = mfspr(SPRN_IC);
+ vcpu->arch.pid = mfspr(SPRN_PID);
+ vcpu->arch.psscr = mfspr(SPRN_PSSCR) & PSSCR_GUEST_VIS;
+
+ vcpu->arch.shregs.sprg0 = mfspr(SPRN_SPRG0);
+ vcpu->arch.shregs.sprg1 = mfspr(SPRN_SPRG1);
+ vcpu->arch.shregs.sprg2 = mfspr(SPRN_SPRG2);
+ vcpu->arch.shregs.sprg3 = mfspr(SPRN_SPRG3);
+
+ /* Preserve PSSCR[FAKE_SUSPEND] until we've called kvmppc_save_tm_hv */
+ mtspr(SPRN_PSSCR, host_psscr |
+ (local_paca->kvm_hstate.fake_suspend << PSSCR_FAKE_SUSPEND_LG));
+ mtspr(SPRN_HFSCR, host_hfscr);
+ mtspr(SPRN_CIABR, host_ciabr);
+ mtspr(SPRN_DAWR0, host_dawr0);
+ mtspr(SPRN_DAWRX0, host_dawrx0);
+ if (cpu_has_feature(CPU_FTR_DAWR1)) {
+ mtspr(SPRN_DAWR1, host_dawr1);
+ mtspr(SPRN_DAWRX1, host_dawrx1);
+ }
+
+ /*
+ * Since this is radix, do a eieio; tlbsync; ptesync sequence in
+ * case we interrupted the guest between a tlbie and a ptesync.
+ */
+ asm volatile("eieio; tlbsync; ptesync");
+
+ /*
+ * cp_abort is required if the processor supports local copy-paste
+ * to clear the copy buffer that was under control of the guest.
+ */
+ if (cpu_has_feature(CPU_FTR_ARCH_31))
+ asm volatile(PPC_CP_ABORT);
+
+ vc->dpdes = mfspr(SPRN_DPDES);
+ vc->vtb = mfspr(SPRN_VTB);
+ mtspr(SPRN_DPDES, 0);
+ if (vc->pcr)
+ mtspr(SPRN_PCR, PCR_MASK);
+
+ if (vc->tb_offset_applied) {
+ u64 new_tb = mftb() - vc->tb_offset_applied;
+ mtspr(SPRN_TBU40, new_tb);
+ tb = mftb();
+ if ((tb & 0xffffff) < (new_tb & 0xffffff))
+ mtspr(SPRN_TBU40, new_tb + 0x1000000);
+ vc->tb_offset_applied = 0;
+ }
+
+ /* HDEC must be at least as large as DEC, so decrementer_max fits */
+ mtspr(SPRN_HDEC, decrementer_max);
+
+ switch_mmu_to_host_radix(kvm, host_pidr);
+
return trap;
}
-EXPORT_SYMBOL_GPL(__kvmhv_vcpu_entry_p9);
+EXPORT_SYMBOL_GPL(kvmhv_vcpu_entry_p9);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 31/46] KVM: PPC: Book3S HV P9: Read machine check registers while MSR[RI] is 0
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
SRR0/1, DAR, DSISR must all be protected from machine check which can
clobber them. Ensure MSR[RI] is clear while they are live.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 11 +++++++--
arch/powerpc/kvm/book3s_hv_interrupt.c | 33 +++++++++++++++++++++++---
arch/powerpc/kvm/book3s_hv_ras.c | 2 ++
3 files changed, 41 insertions(+), 5 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 52f844a06899..98e5ae897fad 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3551,11 +3551,16 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
mtspr(SPRN_BESCR, vcpu->arch.bescr);
mtspr(SPRN_WORT, vcpu->arch.wort);
mtspr(SPRN_TIDR, vcpu->arch.tid);
- mtspr(SPRN_DAR, vcpu->arch.shregs.dar);
- mtspr(SPRN_DSISR, vcpu->arch.shregs.dsisr);
mtspr(SPRN_AMR, vcpu->arch.amr);
mtspr(SPRN_UAMOR, vcpu->arch.uamor);
+ /*
+ * DAR, DSISR, and for nested HV, SPRGs must be set with MSR[RI]
+ * clear (or hstate set appropriately to catch those registers
+ * being clobbered if we take a MCE or SRESET), so those are done
+ * later.
+ */
+
if (!(vcpu->arch.ctrl & 1))
mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
@@ -3598,6 +3603,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
hvregs.vcpu_token = vcpu->vcpu_id;
}
hvregs.hdec_expiry = time_limit;
+ mtspr(SPRN_DAR, vcpu->arch.shregs.dar);
+ mtspr(SPRN_DSISR, vcpu->arch.shregs.dsisr);
trap = plpar_hcall_norets(H_ENTER_NESTED, __pa(&hvregs),
__pa(&vcpu->arch.regs));
kvmhv_restore_hv_return_state(vcpu, &hvregs);
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index c1105a4cc9b7..a7e5628ac36c 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -137,6 +137,7 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
s64 hdec;
u64 tb, purr, spurr;
u64 *exsave;
+ bool ri_set;
unsigned long msr = mfmsr();
int trap;
unsigned long host_hfscr = mfspr(SPRN_HFSCR);
@@ -208,9 +209,6 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
*/
mtspr(SPRN_HDEC, hdec);
- mtspr(SPRN_SRR0, vcpu->arch.shregs.srr0);
- mtspr(SPRN_SRR1, vcpu->arch.shregs.srr1);
-
start_timing(vcpu, &vcpu->arch.rm_entry);
vcpu->arch.ceded = 0;
@@ -236,6 +234,13 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
*/
mtspr(SPRN_HDSISR, HDSISR_CANARY);
+ __mtmsrd(0, 1); /* clear RI */
+
+ mtspr(SPRN_DAR, vcpu->arch.shregs.dar);
+ mtspr(SPRN_DSISR, vcpu->arch.shregs.dsisr);
+ mtspr(SPRN_SRR0, vcpu->arch.shregs.srr0);
+ mtspr(SPRN_SRR1, vcpu->arch.shregs.srr1);
+
accumulate_time(vcpu, &vcpu->arch.guest_time);
local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
@@ -253,7 +258,13 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
/* 0x2 bit for HSRR is only used by PR and P7/8 HV paths, clear it */
trap = local_paca->kvm_hstate.scratch0 & ~0x2;
+
+ /* HSRR interrupts leave MSR[RI] unchanged, SRR interrupts clear it. */
+ ri_set = false;
if (likely(trap > BOOK3S_INTERRUPT_MACHINE_CHECK)) {
+ if (trap != BOOK3S_INTERRUPT_SYSCALL &&
+ (vcpu->arch.shregs.msr & MSR_RI))
+ ri_set = true;
exsave = local_paca->exgen;
} else if (trap == BOOK3S_INTERRUPT_SYSTEM_RESET) {
exsave = local_paca->exnmi;
@@ -263,6 +274,22 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
vcpu->arch.regs.gpr[1] = local_paca->kvm_hstate.scratch1;
vcpu->arch.regs.gpr[3] = local_paca->kvm_hstate.scratch2;
+
+ /*
+ * Only set RI after reading machine check regs (DAR, DSISR, SRR0/1)
+ * and hstate scratch (which we need to move into exsave to make
+ * re-entrant vs SRESET/MCE)
+ */
+ if (ri_set) {
+ if (unlikely(!(mfmsr() & MSR_RI))) {
+ __mtmsrd(MSR_RI, 1);
+ WARN_ON_ONCE(1);
+ }
+ } else {
+ WARN_ON_ONCE(mfmsr() & MSR_RI);
+ __mtmsrd(MSR_RI, 1);
+ }
+
vcpu->arch.regs.gpr[9] = exsave[EX_R9/sizeof(u64)];
vcpu->arch.regs.gpr[10] = exsave[EX_R10/sizeof(u64)];
vcpu->arch.regs.gpr[11] = exsave[EX_R11/sizeof(u64)];
diff --git a/arch/powerpc/kvm/book3s_hv_ras.c b/arch/powerpc/kvm/book3s_hv_ras.c
index d4bca93b79f6..8d8a4d5f0b55 100644
--- a/arch/powerpc/kvm/book3s_hv_ras.c
+++ b/arch/powerpc/kvm/book3s_hv_ras.c
@@ -199,6 +199,8 @@ static void kvmppc_tb_resync_done(void)
* know about the exact state of the TB value. Resync TB call will
* restore TB to host timebase.
*
+ * This could use the new OPAL_HANDLE_HMI2 to avoid resyncing TB every time.
+ *
* Things to consider:
* - On TB error, HMI interrupt is reported on all the threads of the core
* that has encountered TB error irrespective of split-core mode.
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 31/46] KVM: PPC: Book3S HV P9: Read machine check registers while MSR[RI] is 0
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
SRR0/1, DAR, DSISR must all be protected from machine check which can
clobber them. Ensure MSR[RI] is clear while they are live.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 11 +++++++--
arch/powerpc/kvm/book3s_hv_interrupt.c | 33 +++++++++++++++++++++++---
arch/powerpc/kvm/book3s_hv_ras.c | 2 ++
3 files changed, 41 insertions(+), 5 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 52f844a06899..98e5ae897fad 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3551,11 +3551,16 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
mtspr(SPRN_BESCR, vcpu->arch.bescr);
mtspr(SPRN_WORT, vcpu->arch.wort);
mtspr(SPRN_TIDR, vcpu->arch.tid);
- mtspr(SPRN_DAR, vcpu->arch.shregs.dar);
- mtspr(SPRN_DSISR, vcpu->arch.shregs.dsisr);
mtspr(SPRN_AMR, vcpu->arch.amr);
mtspr(SPRN_UAMOR, vcpu->arch.uamor);
+ /*
+ * DAR, DSISR, and for nested HV, SPRGs must be set with MSR[RI]
+ * clear (or hstate set appropriately to catch those registers
+ * being clobbered if we take a MCE or SRESET), so those are done
+ * later.
+ */
+
if (!(vcpu->arch.ctrl & 1))
mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
@@ -3598,6 +3603,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
hvregs.vcpu_token = vcpu->vcpu_id;
}
hvregs.hdec_expiry = time_limit;
+ mtspr(SPRN_DAR, vcpu->arch.shregs.dar);
+ mtspr(SPRN_DSISR, vcpu->arch.shregs.dsisr);
trap = plpar_hcall_norets(H_ENTER_NESTED, __pa(&hvregs),
__pa(&vcpu->arch.regs));
kvmhv_restore_hv_return_state(vcpu, &hvregs);
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index c1105a4cc9b7..a7e5628ac36c 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -137,6 +137,7 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
s64 hdec;
u64 tb, purr, spurr;
u64 *exsave;
+ bool ri_set;
unsigned long msr = mfmsr();
int trap;
unsigned long host_hfscr = mfspr(SPRN_HFSCR);
@@ -208,9 +209,6 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
*/
mtspr(SPRN_HDEC, hdec);
- mtspr(SPRN_SRR0, vcpu->arch.shregs.srr0);
- mtspr(SPRN_SRR1, vcpu->arch.shregs.srr1);
-
start_timing(vcpu, &vcpu->arch.rm_entry);
vcpu->arch.ceded = 0;
@@ -236,6 +234,13 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
*/
mtspr(SPRN_HDSISR, HDSISR_CANARY);
+ __mtmsrd(0, 1); /* clear RI */
+
+ mtspr(SPRN_DAR, vcpu->arch.shregs.dar);
+ mtspr(SPRN_DSISR, vcpu->arch.shregs.dsisr);
+ mtspr(SPRN_SRR0, vcpu->arch.shregs.srr0);
+ mtspr(SPRN_SRR1, vcpu->arch.shregs.srr1);
+
accumulate_time(vcpu, &vcpu->arch.guest_time);
local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
@@ -253,7 +258,13 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
/* 0x2 bit for HSRR is only used by PR and P7/8 HV paths, clear it */
trap = local_paca->kvm_hstate.scratch0 & ~0x2;
+
+ /* HSRR interrupts leave MSR[RI] unchanged, SRR interrupts clear it. */
+ ri_set = false;
if (likely(trap > BOOK3S_INTERRUPT_MACHINE_CHECK)) {
+ if (trap != BOOK3S_INTERRUPT_SYSCALL &&
+ (vcpu->arch.shregs.msr & MSR_RI))
+ ri_set = true;
exsave = local_paca->exgen;
} else if (trap = BOOK3S_INTERRUPT_SYSTEM_RESET) {
exsave = local_paca->exnmi;
@@ -263,6 +274,22 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
vcpu->arch.regs.gpr[1] = local_paca->kvm_hstate.scratch1;
vcpu->arch.regs.gpr[3] = local_paca->kvm_hstate.scratch2;
+
+ /*
+ * Only set RI after reading machine check regs (DAR, DSISR, SRR0/1)
+ * and hstate scratch (which we need to move into exsave to make
+ * re-entrant vs SRESET/MCE)
+ */
+ if (ri_set) {
+ if (unlikely(!(mfmsr() & MSR_RI))) {
+ __mtmsrd(MSR_RI, 1);
+ WARN_ON_ONCE(1);
+ }
+ } else {
+ WARN_ON_ONCE(mfmsr() & MSR_RI);
+ __mtmsrd(MSR_RI, 1);
+ }
+
vcpu->arch.regs.gpr[9] = exsave[EX_R9/sizeof(u64)];
vcpu->arch.regs.gpr[10] = exsave[EX_R10/sizeof(u64)];
vcpu->arch.regs.gpr[11] = exsave[EX_R11/sizeof(u64)];
diff --git a/arch/powerpc/kvm/book3s_hv_ras.c b/arch/powerpc/kvm/book3s_hv_ras.c
index d4bca93b79f6..8d8a4d5f0b55 100644
--- a/arch/powerpc/kvm/book3s_hv_ras.c
+++ b/arch/powerpc/kvm/book3s_hv_ras.c
@@ -199,6 +199,8 @@ static void kvmppc_tb_resync_done(void)
* know about the exact state of the TB value. Resync TB call will
* restore TB to host timebase.
*
+ * This could use the new OPAL_HANDLE_HMI2 to avoid resyncing TB every time.
+ *
* Things to consider:
* - On TB error, HMI interrupt is reported on all the threads of the core
* that has encountered TB error irrespective of split-core mode.
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 32/46] KVM: PPC: Book3S HV P9: Improve exit timing accounting coverage
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
The C conversion caused exit timing to become a bit cramped. Expand it
to cover more of the entry and exit code.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_interrupt.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index a7e5628ac36c..4058a325a7f0 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -159,6 +159,8 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
if (hdec < 0)
return BOOK3S_INTERRUPT_HV_DECREMENTER;
+ start_timing(vcpu, &vcpu->arch.rm_entry);
+
if (vc->tb_offset) {
u64 new_tb = tb + vc->tb_offset;
mtspr(SPRN_TBU40, new_tb);
@@ -209,8 +211,6 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
*/
mtspr(SPRN_HDEC, hdec);
- start_timing(vcpu, &vcpu->arch.rm_entry);
-
vcpu->arch.ceded = 0;
WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
@@ -354,8 +354,6 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
accumulate_time(vcpu, &vcpu->arch.rm_exit);
- end_timing(vcpu);
-
/* Advance host PURR/SPURR by the amount used by guest */
purr = mfspr(SPRN_PURR);
spurr = mfspr(SPRN_SPURR);
@@ -420,6 +418,8 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
switch_mmu_to_host_radix(kvm, host_pidr);
+ end_timing(vcpu);
+
return trap;
}
EXPORT_SYMBOL_GPL(kvmhv_vcpu_entry_p9);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 32/46] KVM: PPC: Book3S HV P9: Improve exit timing accounting coverage
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
The C conversion caused exit timing to become a bit cramped. Expand it
to cover more of the entry and exit code.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_interrupt.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index a7e5628ac36c..4058a325a7f0 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -159,6 +159,8 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
if (hdec < 0)
return BOOK3S_INTERRUPT_HV_DECREMENTER;
+ start_timing(vcpu, &vcpu->arch.rm_entry);
+
if (vc->tb_offset) {
u64 new_tb = tb + vc->tb_offset;
mtspr(SPRN_TBU40, new_tb);
@@ -209,8 +211,6 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
*/
mtspr(SPRN_HDEC, hdec);
- start_timing(vcpu, &vcpu->arch.rm_entry);
-
vcpu->arch.ceded = 0;
WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
@@ -354,8 +354,6 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
accumulate_time(vcpu, &vcpu->arch.rm_exit);
- end_timing(vcpu);
-
/* Advance host PURR/SPURR by the amount used by guest */
purr = mfspr(SPRN_PURR);
spurr = mfspr(SPRN_SPURR);
@@ -420,6 +418,8 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
switch_mmu_to_host_radix(kvm, host_pidr);
+ end_timing(vcpu);
+
return trap;
}
EXPORT_SYMBOL_GPL(kvmhv_vcpu_entry_p9);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 33/46] KVM: PPC: Book3S HV P9: Move SPR loading after expiry time check
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
This is wasted work if the time limit is exceeded.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_interrupt.c | 36 ++++++++++++++++----------
1 file changed, 22 insertions(+), 14 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index 4058a325a7f0..f57379e73b5c 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -138,21 +138,16 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
u64 tb, purr, spurr;
u64 *exsave;
bool ri_set;
- unsigned long msr = mfmsr();
int trap;
- unsigned long host_hfscr = mfspr(SPRN_HFSCR);
- unsigned long host_ciabr = mfspr(SPRN_CIABR);
- unsigned long host_dawr0 = mfspr(SPRN_DAWR0);
- unsigned long host_dawrx0 = mfspr(SPRN_DAWRX0);
- unsigned long host_psscr = mfspr(SPRN_PSSCR);
- unsigned long host_pidr = mfspr(SPRN_PID);
- unsigned long host_dawr1 = 0;
- unsigned long host_dawrx1 = 0;
-
- if (cpu_has_feature(CPU_FTR_DAWR1)) {
- host_dawr1 = mfspr(SPRN_DAWR1);
- host_dawrx1 = mfspr(SPRN_DAWRX1);
- }
+ unsigned long msr;
+ unsigned long host_hfscr;
+ unsigned long host_ciabr;
+ unsigned long host_dawr0;
+ unsigned long host_dawrx0;
+ unsigned long host_psscr;
+ unsigned long host_pidr;
+ unsigned long host_dawr1;
+ unsigned long host_dawrx1;
tb = mftb();
hdec = time_limit - tb;
@@ -170,6 +165,19 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
vc->tb_offset_applied = vc->tb_offset;
}
+ msr = mfmsr();
+
+ host_hfscr = mfspr(SPRN_HFSCR);
+ host_ciabr = mfspr(SPRN_CIABR);
+ host_dawr0 = mfspr(SPRN_DAWR0);
+ host_dawrx0 = mfspr(SPRN_DAWRX0);
+ host_psscr = mfspr(SPRN_PSSCR);
+ host_pidr = mfspr(SPRN_PID);
+ if (cpu_has_feature(CPU_FTR_DAWR1)) {
+ host_dawr1 = mfspr(SPRN_DAWR1);
+ host_dawrx1 = mfspr(SPRN_DAWRX1);
+ }
+
if (vc->pcr)
mtspr(SPRN_PCR, vc->pcr | PCR_MASK);
mtspr(SPRN_DPDES, vc->dpdes);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 33/46] KVM: PPC: Book3S HV P9: Move SPR loading after expiry time check
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
This is wasted work if the time limit is exceeded.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_interrupt.c | 36 ++++++++++++++++----------
1 file changed, 22 insertions(+), 14 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index 4058a325a7f0..f57379e73b5c 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -138,21 +138,16 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
u64 tb, purr, spurr;
u64 *exsave;
bool ri_set;
- unsigned long msr = mfmsr();
int trap;
- unsigned long host_hfscr = mfspr(SPRN_HFSCR);
- unsigned long host_ciabr = mfspr(SPRN_CIABR);
- unsigned long host_dawr0 = mfspr(SPRN_DAWR0);
- unsigned long host_dawrx0 = mfspr(SPRN_DAWRX0);
- unsigned long host_psscr = mfspr(SPRN_PSSCR);
- unsigned long host_pidr = mfspr(SPRN_PID);
- unsigned long host_dawr1 = 0;
- unsigned long host_dawrx1 = 0;
-
- if (cpu_has_feature(CPU_FTR_DAWR1)) {
- host_dawr1 = mfspr(SPRN_DAWR1);
- host_dawrx1 = mfspr(SPRN_DAWRX1);
- }
+ unsigned long msr;
+ unsigned long host_hfscr;
+ unsigned long host_ciabr;
+ unsigned long host_dawr0;
+ unsigned long host_dawrx0;
+ unsigned long host_psscr;
+ unsigned long host_pidr;
+ unsigned long host_dawr1;
+ unsigned long host_dawrx1;
tb = mftb();
hdec = time_limit - tb;
@@ -170,6 +165,19 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
vc->tb_offset_applied = vc->tb_offset;
}
+ msr = mfmsr();
+
+ host_hfscr = mfspr(SPRN_HFSCR);
+ host_ciabr = mfspr(SPRN_CIABR);
+ host_dawr0 = mfspr(SPRN_DAWR0);
+ host_dawrx0 = mfspr(SPRN_DAWRX0);
+ host_psscr = mfspr(SPRN_PSSCR);
+ host_pidr = mfspr(SPRN_PID);
+ if (cpu_has_feature(CPU_FTR_DAWR1)) {
+ host_dawr1 = mfspr(SPRN_DAWR1);
+ host_dawrx1 = mfspr(SPRN_DAWRX1);
+ }
+
if (vc->pcr)
mtspr(SPRN_PCR, vc->pcr | PCR_MASK);
mtspr(SPRN_DPDES, vc->dpdes);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 34/46] KVM: PPC: Book3S HV P9: Add helpers for OS SPR handling
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
This is a first step to wrapping supervisor and user SPR saving and
loading up into helpers, which will then be called independently in
bare metal and nested HV cases in order to optimise SPR access.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 141 ++++++++++++++++++++++-------------
1 file changed, 89 insertions(+), 52 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 98e5ae897fad..74c67f713a26 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3481,6 +3481,89 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
trace_kvmppc_run_core(vc, 1);
}
+static void load_spr_state(struct kvm_vcpu *vcpu)
+{
+ mtspr(SPRN_DSCR, vcpu->arch.dscr);
+ mtspr(SPRN_IAMR, vcpu->arch.iamr);
+ mtspr(SPRN_PSPB, vcpu->arch.pspb);
+ mtspr(SPRN_FSCR, vcpu->arch.fscr);
+ mtspr(SPRN_TAR, vcpu->arch.tar);
+ mtspr(SPRN_EBBHR, vcpu->arch.ebbhr);
+ mtspr(SPRN_EBBRR, vcpu->arch.ebbrr);
+ mtspr(SPRN_BESCR, vcpu->arch.bescr);
+ mtspr(SPRN_WORT, vcpu->arch.wort);
+ mtspr(SPRN_TIDR, vcpu->arch.tid);
+ mtspr(SPRN_AMR, vcpu->arch.amr);
+ mtspr(SPRN_UAMOR, vcpu->arch.uamor);
+
+ /*
+ * DAR, DSISR, and for nested HV, SPRGs must be set with MSR[RI]
+ * clear (or hstate set appropriately to catch those registers
+ * being clobbered if we take a MCE or SRESET), so those are done
+ * later.
+ */
+
+ if (!(vcpu->arch.ctrl & 1))
+ mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
+}
+
+static void store_spr_state(struct kvm_vcpu *vcpu)
+{
+ vcpu->arch.ctrl = mfspr(SPRN_CTRLF);
+
+ vcpu->arch.iamr = mfspr(SPRN_IAMR);
+ vcpu->arch.pspb = mfspr(SPRN_PSPB);
+ vcpu->arch.fscr = mfspr(SPRN_FSCR);
+ vcpu->arch.tar = mfspr(SPRN_TAR);
+ vcpu->arch.ebbhr = mfspr(SPRN_EBBHR);
+ vcpu->arch.ebbrr = mfspr(SPRN_EBBRR);
+ vcpu->arch.bescr = mfspr(SPRN_BESCR);
+ vcpu->arch.wort = mfspr(SPRN_WORT);
+ vcpu->arch.tid = mfspr(SPRN_TIDR);
+ vcpu->arch.amr = mfspr(SPRN_AMR);
+ vcpu->arch.uamor = mfspr(SPRN_UAMOR);
+ vcpu->arch.dscr = mfspr(SPRN_DSCR);
+}
+
+/*
+ * Privileged (non-hypervisor) host registers to save.
+ */
+struct p9_host_os_sprs {
+ unsigned long dscr;
+ unsigned long tidr;
+ unsigned long iamr;
+ unsigned long amr;
+ unsigned long fscr;
+};
+
+static void save_p9_host_os_sprs(struct p9_host_os_sprs *host_os_sprs)
+{
+ host_os_sprs->dscr = mfspr(SPRN_DSCR);
+ host_os_sprs->tidr = mfspr(SPRN_TIDR);
+ host_os_sprs->iamr = mfspr(SPRN_IAMR);
+ host_os_sprs->amr = mfspr(SPRN_AMR);
+ host_os_sprs->fscr = mfspr(SPRN_FSCR);
+}
+
+/* vcpu guest regs must already be saved */
+static void restore_p9_host_os_sprs(struct kvm_vcpu *vcpu,
+ struct p9_host_os_sprs *host_os_sprs)
+{
+ mtspr(SPRN_PSPB, 0);
+ mtspr(SPRN_WORT, 0);
+ mtspr(SPRN_UAMOR, 0);
+
+ mtspr(SPRN_DSCR, host_os_sprs->dscr);
+ mtspr(SPRN_TIDR, host_os_sprs->tidr);
+ mtspr(SPRN_IAMR, host_os_sprs->iamr);
+
+ if (host_os_sprs->amr != vcpu->arch.amr)
+ mtspr(SPRN_AMR, host_os_sprs->amr);
+
+ if (host_os_sprs->fscr != vcpu->arch.fscr)
+ mtspr(SPRN_FSCR, host_os_sprs->fscr);
+}
+
static inline bool hcall_is_xics(unsigned long req)
{
return (req == H_EOI || req == H_CPPR || req == H_IPI ||
@@ -3495,11 +3578,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
unsigned long lpcr)
{
struct kvmppc_vcore *vc = vcpu->arch.vcore;
- unsigned long host_dscr = mfspr(SPRN_DSCR);
- unsigned long host_tidr = mfspr(SPRN_TIDR);
- unsigned long host_iamr = mfspr(SPRN_IAMR);
- unsigned long host_amr = mfspr(SPRN_AMR);
- unsigned long host_fscr = mfspr(SPRN_FSCR);
+ struct p9_host_os_sprs host_os_sprs;
s64 dec;
u64 tb, next_timer;
int trap, save_pmu;
@@ -3514,6 +3593,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
if (next_timer < time_limit)
time_limit = next_timer;
+ save_p9_host_os_sprs(&host_os_sprs);
+
kvmhv_save_host_pmu(); /* saves it to PACA kvm_hstate */
kvmppc_subcore_enter_guest();
@@ -3541,28 +3622,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
#endif
mtspr(SPRN_VRSAVE, vcpu->arch.vrsave);
- mtspr(SPRN_DSCR, vcpu->arch.dscr);
- mtspr(SPRN_IAMR, vcpu->arch.iamr);
- mtspr(SPRN_PSPB, vcpu->arch.pspb);
- mtspr(SPRN_FSCR, vcpu->arch.fscr);
- mtspr(SPRN_TAR, vcpu->arch.tar);
- mtspr(SPRN_EBBHR, vcpu->arch.ebbhr);
- mtspr(SPRN_EBBRR, vcpu->arch.ebbrr);
- mtspr(SPRN_BESCR, vcpu->arch.bescr);
- mtspr(SPRN_WORT, vcpu->arch.wort);
- mtspr(SPRN_TIDR, vcpu->arch.tid);
- mtspr(SPRN_AMR, vcpu->arch.amr);
- mtspr(SPRN_UAMOR, vcpu->arch.uamor);
-
- /*
- * DAR, DSISR, and for nested HV, SPRGs must be set with MSR[RI]
- * clear (or hstate set appropriately to catch those registers
- * being clobbered if we take a MCE or SRESET), so those are done
- * later.
- */
-
- if (!(vcpu->arch.ctrl & 1))
- mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
+ load_spr_state(vcpu);
/*
* When setting DEC, we must always deal with irq_work_raise via NMI vs
@@ -3658,33 +3718,10 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
vcpu->arch.dec_expires = dec + tb;
vcpu->cpu = -1;
vcpu->arch.thread_cpu = -1;
- vcpu->arch.ctrl = mfspr(SPRN_CTRLF);
-
- vcpu->arch.iamr = mfspr(SPRN_IAMR);
- vcpu->arch.pspb = mfspr(SPRN_PSPB);
- vcpu->arch.fscr = mfspr(SPRN_FSCR);
- vcpu->arch.tar = mfspr(SPRN_TAR);
- vcpu->arch.ebbhr = mfspr(SPRN_EBBHR);
- vcpu->arch.ebbrr = mfspr(SPRN_EBBRR);
- vcpu->arch.bescr = mfspr(SPRN_BESCR);
- vcpu->arch.wort = mfspr(SPRN_WORT);
- vcpu->arch.tid = mfspr(SPRN_TIDR);
- vcpu->arch.amr = mfspr(SPRN_AMR);
- vcpu->arch.uamor = mfspr(SPRN_UAMOR);
- vcpu->arch.dscr = mfspr(SPRN_DSCR);
-
- mtspr(SPRN_PSPB, 0);
- mtspr(SPRN_WORT, 0);
- mtspr(SPRN_UAMOR, 0);
- mtspr(SPRN_DSCR, host_dscr);
- mtspr(SPRN_TIDR, host_tidr);
- mtspr(SPRN_IAMR, host_iamr);
- if (host_amr != vcpu->arch.amr)
- mtspr(SPRN_AMR, host_amr);
+ store_spr_state(vcpu);
- if (host_fscr != vcpu->arch.fscr)
- mtspr(SPRN_FSCR, host_fscr);
+ restore_p9_host_os_sprs(vcpu, &host_os_sprs);
msr_check_and_set(MSR_FP | MSR_VEC | MSR_VSX);
store_fp_state(&vcpu->arch.fp);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 34/46] KVM: PPC: Book3S HV P9: Add helpers for OS SPR handling
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
This is a first step to wrapping supervisor and user SPR saving and
loading up into helpers, which will then be called independently in
bare metal and nested HV cases in order to optimise SPR access.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 141 ++++++++++++++++++++++-------------
1 file changed, 89 insertions(+), 52 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 98e5ae897fad..74c67f713a26 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3481,6 +3481,89 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
trace_kvmppc_run_core(vc, 1);
}
+static void load_spr_state(struct kvm_vcpu *vcpu)
+{
+ mtspr(SPRN_DSCR, vcpu->arch.dscr);
+ mtspr(SPRN_IAMR, vcpu->arch.iamr);
+ mtspr(SPRN_PSPB, vcpu->arch.pspb);
+ mtspr(SPRN_FSCR, vcpu->arch.fscr);
+ mtspr(SPRN_TAR, vcpu->arch.tar);
+ mtspr(SPRN_EBBHR, vcpu->arch.ebbhr);
+ mtspr(SPRN_EBBRR, vcpu->arch.ebbrr);
+ mtspr(SPRN_BESCR, vcpu->arch.bescr);
+ mtspr(SPRN_WORT, vcpu->arch.wort);
+ mtspr(SPRN_TIDR, vcpu->arch.tid);
+ mtspr(SPRN_AMR, vcpu->arch.amr);
+ mtspr(SPRN_UAMOR, vcpu->arch.uamor);
+
+ /*
+ * DAR, DSISR, and for nested HV, SPRGs must be set with MSR[RI]
+ * clear (or hstate set appropriately to catch those registers
+ * being clobbered if we take a MCE or SRESET), so those are done
+ * later.
+ */
+
+ if (!(vcpu->arch.ctrl & 1))
+ mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
+}
+
+static void store_spr_state(struct kvm_vcpu *vcpu)
+{
+ vcpu->arch.ctrl = mfspr(SPRN_CTRLF);
+
+ vcpu->arch.iamr = mfspr(SPRN_IAMR);
+ vcpu->arch.pspb = mfspr(SPRN_PSPB);
+ vcpu->arch.fscr = mfspr(SPRN_FSCR);
+ vcpu->arch.tar = mfspr(SPRN_TAR);
+ vcpu->arch.ebbhr = mfspr(SPRN_EBBHR);
+ vcpu->arch.ebbrr = mfspr(SPRN_EBBRR);
+ vcpu->arch.bescr = mfspr(SPRN_BESCR);
+ vcpu->arch.wort = mfspr(SPRN_WORT);
+ vcpu->arch.tid = mfspr(SPRN_TIDR);
+ vcpu->arch.amr = mfspr(SPRN_AMR);
+ vcpu->arch.uamor = mfspr(SPRN_UAMOR);
+ vcpu->arch.dscr = mfspr(SPRN_DSCR);
+}
+
+/*
+ * Privileged (non-hypervisor) host registers to save.
+ */
+struct p9_host_os_sprs {
+ unsigned long dscr;
+ unsigned long tidr;
+ unsigned long iamr;
+ unsigned long amr;
+ unsigned long fscr;
+};
+
+static void save_p9_host_os_sprs(struct p9_host_os_sprs *host_os_sprs)
+{
+ host_os_sprs->dscr = mfspr(SPRN_DSCR);
+ host_os_sprs->tidr = mfspr(SPRN_TIDR);
+ host_os_sprs->iamr = mfspr(SPRN_IAMR);
+ host_os_sprs->amr = mfspr(SPRN_AMR);
+ host_os_sprs->fscr = mfspr(SPRN_FSCR);
+}
+
+/* vcpu guest regs must already be saved */
+static void restore_p9_host_os_sprs(struct kvm_vcpu *vcpu,
+ struct p9_host_os_sprs *host_os_sprs)
+{
+ mtspr(SPRN_PSPB, 0);
+ mtspr(SPRN_WORT, 0);
+ mtspr(SPRN_UAMOR, 0);
+
+ mtspr(SPRN_DSCR, host_os_sprs->dscr);
+ mtspr(SPRN_TIDR, host_os_sprs->tidr);
+ mtspr(SPRN_IAMR, host_os_sprs->iamr);
+
+ if (host_os_sprs->amr != vcpu->arch.amr)
+ mtspr(SPRN_AMR, host_os_sprs->amr);
+
+ if (host_os_sprs->fscr != vcpu->arch.fscr)
+ mtspr(SPRN_FSCR, host_os_sprs->fscr);
+}
+
static inline bool hcall_is_xics(unsigned long req)
{
return (req = H_EOI || req = H_CPPR || req = H_IPI ||
@@ -3495,11 +3578,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
unsigned long lpcr)
{
struct kvmppc_vcore *vc = vcpu->arch.vcore;
- unsigned long host_dscr = mfspr(SPRN_DSCR);
- unsigned long host_tidr = mfspr(SPRN_TIDR);
- unsigned long host_iamr = mfspr(SPRN_IAMR);
- unsigned long host_amr = mfspr(SPRN_AMR);
- unsigned long host_fscr = mfspr(SPRN_FSCR);
+ struct p9_host_os_sprs host_os_sprs;
s64 dec;
u64 tb, next_timer;
int trap, save_pmu;
@@ -3514,6 +3593,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
if (next_timer < time_limit)
time_limit = next_timer;
+ save_p9_host_os_sprs(&host_os_sprs);
+
kvmhv_save_host_pmu(); /* saves it to PACA kvm_hstate */
kvmppc_subcore_enter_guest();
@@ -3541,28 +3622,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
#endif
mtspr(SPRN_VRSAVE, vcpu->arch.vrsave);
- mtspr(SPRN_DSCR, vcpu->arch.dscr);
- mtspr(SPRN_IAMR, vcpu->arch.iamr);
- mtspr(SPRN_PSPB, vcpu->arch.pspb);
- mtspr(SPRN_FSCR, vcpu->arch.fscr);
- mtspr(SPRN_TAR, vcpu->arch.tar);
- mtspr(SPRN_EBBHR, vcpu->arch.ebbhr);
- mtspr(SPRN_EBBRR, vcpu->arch.ebbrr);
- mtspr(SPRN_BESCR, vcpu->arch.bescr);
- mtspr(SPRN_WORT, vcpu->arch.wort);
- mtspr(SPRN_TIDR, vcpu->arch.tid);
- mtspr(SPRN_AMR, vcpu->arch.amr);
- mtspr(SPRN_UAMOR, vcpu->arch.uamor);
-
- /*
- * DAR, DSISR, and for nested HV, SPRGs must be set with MSR[RI]
- * clear (or hstate set appropriately to catch those registers
- * being clobbered if we take a MCE or SRESET), so those are done
- * later.
- */
-
- if (!(vcpu->arch.ctrl & 1))
- mtspr(SPRN_CTRLT, mfspr(SPRN_CTRLF) & ~1);
+ load_spr_state(vcpu);
/*
* When setting DEC, we must always deal with irq_work_raise via NMI vs
@@ -3658,33 +3718,10 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
vcpu->arch.dec_expires = dec + tb;
vcpu->cpu = -1;
vcpu->arch.thread_cpu = -1;
- vcpu->arch.ctrl = mfspr(SPRN_CTRLF);
-
- vcpu->arch.iamr = mfspr(SPRN_IAMR);
- vcpu->arch.pspb = mfspr(SPRN_PSPB);
- vcpu->arch.fscr = mfspr(SPRN_FSCR);
- vcpu->arch.tar = mfspr(SPRN_TAR);
- vcpu->arch.ebbhr = mfspr(SPRN_EBBHR);
- vcpu->arch.ebbrr = mfspr(SPRN_EBBRR);
- vcpu->arch.bescr = mfspr(SPRN_BESCR);
- vcpu->arch.wort = mfspr(SPRN_WORT);
- vcpu->arch.tid = mfspr(SPRN_TIDR);
- vcpu->arch.amr = mfspr(SPRN_AMR);
- vcpu->arch.uamor = mfspr(SPRN_UAMOR);
- vcpu->arch.dscr = mfspr(SPRN_DSCR);
-
- mtspr(SPRN_PSPB, 0);
- mtspr(SPRN_WORT, 0);
- mtspr(SPRN_UAMOR, 0);
- mtspr(SPRN_DSCR, host_dscr);
- mtspr(SPRN_TIDR, host_tidr);
- mtspr(SPRN_IAMR, host_iamr);
- if (host_amr != vcpu->arch.amr)
- mtspr(SPRN_AMR, host_amr);
+ store_spr_state(vcpu);
- if (host_fscr != vcpu->arch.fscr)
- mtspr(SPRN_FSCR, host_fscr);
+ restore_p9_host_os_sprs(vcpu, &host_os_sprs);
msr_check_and_set(MSR_FP | MSR_VEC | MSR_VSX);
store_fp_state(&vcpu->arch.fp);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 35/46] KVM: PPC: Book3S HV P9: Switch to guest MMU context as late as possible
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Move MMU context switch as late as reasonably possible to minimise code
running with guest context switched in. This becomes more important when
this code may run in real-mode, with later changes.
Move WARN_ON as early as possible so program check interrupts are less
likely to tangle everything up.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_interrupt.c | 40 +++++++++++++-------------
1 file changed, 20 insertions(+), 20 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index f57379e73b5c..787a7d2aed1a 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -154,8 +154,13 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
if (hdec < 0)
return BOOK3S_INTERRUPT_HV_DECREMENTER;
+ WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
+ WARN_ON_ONCE(!(vcpu->arch.shregs.msr & MSR_ME));
+
start_timing(vcpu, &vcpu->arch.rm_entry);
+ vcpu->arch.ceded = 0;
+
if (vc->tb_offset) {
u64 new_tb = tb + vc->tb_offset;
mtspr(SPRN_TBU40, new_tb);
@@ -204,26 +209,6 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
mtspr(SPRN_HFSCR, vcpu->arch.hfscr);
- mtspr(SPRN_SPRG0, vcpu->arch.shregs.sprg0);
- mtspr(SPRN_SPRG1, vcpu->arch.shregs.sprg1);
- mtspr(SPRN_SPRG2, vcpu->arch.shregs.sprg2);
- mtspr(SPRN_SPRG3, vcpu->arch.shregs.sprg3);
-
- mtspr(SPRN_AMOR, ~0UL);
-
- switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
-
- /*
- * P9 suppresses the HDEC exception when LPCR[HDICE] = 0,
- * so set guest LPCR (with HDICE) before writing HDEC.
- */
- mtspr(SPRN_HDEC, hdec);
-
- vcpu->arch.ceded = 0;
-
- WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
- WARN_ON_ONCE(!(vcpu->arch.shregs.msr & MSR_ME));
-
mtspr(SPRN_HSRR0, vcpu->arch.regs.nip);
mtspr(SPRN_HSRR1, (vcpu->arch.shregs.msr & ~MSR_HV) | MSR_ME);
@@ -242,6 +227,21 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
*/
mtspr(SPRN_HDSISR, HDSISR_CANARY);
+ mtspr(SPRN_SPRG0, vcpu->arch.shregs.sprg0);
+ mtspr(SPRN_SPRG1, vcpu->arch.shregs.sprg1);
+ mtspr(SPRN_SPRG2, vcpu->arch.shregs.sprg2);
+ mtspr(SPRN_SPRG3, vcpu->arch.shregs.sprg3);
+
+ mtspr(SPRN_AMOR, ~0UL);
+
+ switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
+
+ /*
+ * P9 suppresses the HDEC exception when LPCR[HDICE] = 0,
+ * so set guest LPCR (with HDICE) before writing HDEC.
+ */
+ mtspr(SPRN_HDEC, hdec);
+
__mtmsrd(0, 1); /* clear RI */
mtspr(SPRN_DAR, vcpu->arch.shregs.dar);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 35/46] KVM: PPC: Book3S HV P9: Switch to guest MMU context as late as possible
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Move MMU context switch as late as reasonably possible to minimise code
running with guest context switched in. This becomes more important when
this code may run in real-mode, with later changes.
Move WARN_ON as early as possible so program check interrupts are less
likely to tangle everything up.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_interrupt.c | 40 +++++++++++++-------------
1 file changed, 20 insertions(+), 20 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index f57379e73b5c..787a7d2aed1a 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -154,8 +154,13 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
if (hdec < 0)
return BOOK3S_INTERRUPT_HV_DECREMENTER;
+ WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
+ WARN_ON_ONCE(!(vcpu->arch.shregs.msr & MSR_ME));
+
start_timing(vcpu, &vcpu->arch.rm_entry);
+ vcpu->arch.ceded = 0;
+
if (vc->tb_offset) {
u64 new_tb = tb + vc->tb_offset;
mtspr(SPRN_TBU40, new_tb);
@@ -204,26 +209,6 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
mtspr(SPRN_HFSCR, vcpu->arch.hfscr);
- mtspr(SPRN_SPRG0, vcpu->arch.shregs.sprg0);
- mtspr(SPRN_SPRG1, vcpu->arch.shregs.sprg1);
- mtspr(SPRN_SPRG2, vcpu->arch.shregs.sprg2);
- mtspr(SPRN_SPRG3, vcpu->arch.shregs.sprg3);
-
- mtspr(SPRN_AMOR, ~0UL);
-
- switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
-
- /*
- * P9 suppresses the HDEC exception when LPCR[HDICE] = 0,
- * so set guest LPCR (with HDICE) before writing HDEC.
- */
- mtspr(SPRN_HDEC, hdec);
-
- vcpu->arch.ceded = 0;
-
- WARN_ON_ONCE(vcpu->arch.shregs.msr & MSR_HV);
- WARN_ON_ONCE(!(vcpu->arch.shregs.msr & MSR_ME));
-
mtspr(SPRN_HSRR0, vcpu->arch.regs.nip);
mtspr(SPRN_HSRR1, (vcpu->arch.shregs.msr & ~MSR_HV) | MSR_ME);
@@ -242,6 +227,21 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
*/
mtspr(SPRN_HDSISR, HDSISR_CANARY);
+ mtspr(SPRN_SPRG0, vcpu->arch.shregs.sprg0);
+ mtspr(SPRN_SPRG1, vcpu->arch.shregs.sprg1);
+ mtspr(SPRN_SPRG2, vcpu->arch.shregs.sprg2);
+ mtspr(SPRN_SPRG3, vcpu->arch.shregs.sprg3);
+
+ mtspr(SPRN_AMOR, ~0UL);
+
+ switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
+
+ /*
+ * P9 suppresses the HDEC exception when LPCR[HDICE] = 0,
+ * so set guest LPCR (with HDICE) before writing HDEC.
+ */
+ mtspr(SPRN_HDEC, hdec);
+
__mtmsrd(0, 1); /* clear RI */
mtspr(SPRN_DAR, vcpu->arch.shregs.dar);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 36/46] KVM: PPC: Book3S HV: Implement radix prefetch workaround by disabling MMU
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Rather than partition the guest PID space + flush a rogue guest PID to
work around this problem, instead fix it by always disabling the MMU when
switching in or out of guest MMU context in HV mode.
This may be a bit less efficient, but it is a lot less complicated and
allows the P9 path to trivally implement the workaround too. Newer CPUs
are not subject to this issue.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/mmu_context.h | 6 ----
arch/powerpc/kvm/book3s_hv.c | 21 +++++++----
arch/powerpc/kvm/book3s_hv_interrupt.c | 16 ++++++---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 34 ------------------
arch/powerpc/mm/book3s64/radix_pgtable.c | 27 +++++---------
arch/powerpc/mm/book3s64/radix_tlb.c | 46 ------------------------
arch/powerpc/mm/mmu_context.c | 4 +--
7 files changed, 36 insertions(+), 118 deletions(-)
diff --git a/arch/powerpc/include/asm/mmu_context.h b/arch/powerpc/include/asm/mmu_context.h
index 652ce85f9410..bb5c7e5e142e 100644
--- a/arch/powerpc/include/asm/mmu_context.h
+++ b/arch/powerpc/include/asm/mmu_context.h
@@ -122,12 +122,6 @@ static inline bool need_extra_context(struct mm_struct *mm, unsigned long ea)
}
#endif
-#if defined(CONFIG_KVM_BOOK3S_HV_POSSIBLE) && defined(CONFIG_PPC_RADIX_MMU)
-extern void radix_kvm_prefetch_workaround(struct mm_struct *mm);
-#else
-static inline void radix_kvm_prefetch_workaround(struct mm_struct *mm) { }
-#endif
-
extern void switch_cop(struct mm_struct *next);
extern int use_cop(unsigned long acop, struct mm_struct *mm);
extern void drop_cop(unsigned long acop, struct mm_struct *mm);
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 74c67f713a26..8480cbe4b4fa 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -809,6 +809,9 @@ static int kvmppc_h_set_mode(struct kvm_vcpu *vcpu, unsigned long mflags,
*/
if (mflags != 0 && mflags != 3)
return H_UNSUPPORTED_FLAG_START;
+ if (cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG) &&
+ kvmhv_vcpu_is_radix(vcpu) && mflags == 3)
+ return H_UNSUPPORTED_FLAG_START;
return H_TOO_HARD;
default:
return H_TOO_HARD;
@@ -1665,6 +1668,14 @@ unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
lpcr &= ~LPCR_AIL;
if ((lpcr & LPCR_AIL) != LPCR_AIL_3)
lpcr &= ~LPCR_AIL; /* LPCR[AIL]=1/2 is disallowed */
+ /*
+ * On some POWER9s we force AIL off for radix guests to prevent
+ * executing in MSR[HV]=1 mode with the MMU enabled and PIDR set to
+ * guest, which can result in Q0 translations with LPID=0 PID=PIDR to
+ * be cached, which the host TLB management does not expect.
+ */
+ if (kvm_is_radix(kvm) && cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
+ lpcr &= ~LPCR_AIL;
/*
* On POWER9, allow userspace to enable large decrementer for the
@@ -4334,12 +4345,7 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
vcpu->arch.state = KVMPPC_VCPU_BUSY_IN_HOST;
do {
- /*
- * The TLB prefetch bug fixup is only in the kvmppc_run_vcpu
- * path, which also handles hash and dependent threads mode.
- */
- if (kvm->arch.threads_indep && kvm_is_radix(kvm) &&
- !cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
+ if (kvm->arch.threads_indep && kvm_is_radix(kvm))
r = kvmhv_run_single_vcpu(vcpu, ~(u64)0,
vcpu->arch.vcore->lpcr);
else
@@ -4962,6 +4968,9 @@ static int kvmppc_core_init_vm_hv(struct kvm *kvm)
if (!indep_threads_mode && !cpu_has_feature(CPU_FTR_HVMODE)) {
pr_warn("KVM: Ignoring indep_threads_mode=N in nested hypervisor\n");
kvm->arch.threads_indep = true;
+ } else if (!indep_threads_mode && cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG)) {
+ pr_warn("KVM: Ignoring indep_threads_mode=N on pre-DD2.2 POWER9\n");
+ kvm->arch.threads_indep = true;
} else {
kvm->arch.threads_indep = indep_threads_mode;
}
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index 787a7d2aed1a..cd84d2c37632 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -234,6 +234,9 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
mtspr(SPRN_AMOR, ~0UL);
+ if (cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
+ __mtmsrd(msr & ~(MSR_IR|MSR_DR|MSR_RI), 0);
+
switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
/*
@@ -242,7 +245,8 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
*/
mtspr(SPRN_HDEC, hdec);
- __mtmsrd(0, 1); /* clear RI */
+ if (!cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
+ __mtmsrd(0, 1); /* clear RI */
mtspr(SPRN_DAR, vcpu->arch.shregs.dar);
mtspr(SPRN_DSISR, vcpu->arch.shregs.dsisr);
@@ -357,9 +361,6 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
radix_clear_slb();
- __mtmsrd(msr, 0);
- mtspr(SPRN_CTRLT, 1);
-
accumulate_time(vcpu, &vcpu->arch.rm_exit);
/* Advance host PURR/SPURR by the amount used by guest */
@@ -426,6 +427,13 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
switch_mmu_to_host_radix(kvm, host_pidr);
+ /*
+ * If we are in real mode, only switch MMU on after the MMU is
+ * switched to host, to avoid the P9_RADIX_PREFETCH_BUG.
+ */
+ __mtmsrd(msr, 0);
+ mtspr(SPRN_CTRLT, 1);
+
end_timing(vcpu);
return trap;
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 6118e8a97ddd..61f71a7df238 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -1710,40 +1710,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
eieio
tlbsync
ptesync
-
-BEGIN_FTR_SECTION
- /* Radix: Handle the case where the guest used an illegal PID */
- LOAD_REG_ADDR(r4, mmu_base_pid)
- lwz r3, VCPU_GUEST_PID(r9)
- lwz r5, 0(r4)
- cmpw cr0,r3,r5
- blt 2f
-
- /*
- * Illegal PID, the HW might have prefetched and cached in the TLB
- * some translations for the LPID 0 / guest PID combination which
- * Linux doesn't know about, so we need to flush that PID out of
- * the TLB. First we need to set LPIDR to 0 so tlbiel applies to
- * the right context.
- */
- li r0,0
- mtspr SPRN_LPID,r0
- isync
-
- /* Then do a congruence class local flush */
- ld r6,VCPU_KVM(r9)
- lwz r0,KVM_TLB_SETS(r6)
- mtctr r0
- li r7,0x400 /* IS field = 0b01 */
- ptesync
- sldi r0,r3,32 /* RS has PID */
-1: PPC_TLBIEL(7,0,2,1,1) /* RIC=2, PRS=1, R=1 */
- addi r7,r7,0x1000
- bdnz 1b
- ptesync
-END_FTR_SECTION_IFSET(CPU_FTR_P9_RADIX_PREFETCH_BUG)
-
-2:
#endif /* CONFIG_PPC_RADIX_MMU */
/*
diff --git a/arch/powerpc/mm/book3s64/radix_pgtable.c b/arch/powerpc/mm/book3s64/radix_pgtable.c
index 98f0b243c1ab..1ea95891a79e 100644
--- a/arch/powerpc/mm/book3s64/radix_pgtable.c
+++ b/arch/powerpc/mm/book3s64/radix_pgtable.c
@@ -357,30 +357,19 @@ static void __init radix_init_pgtable(void)
}
/* Find out how many PID bits are supported */
- if (!cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG)) {
- if (!mmu_pid_bits)
- mmu_pid_bits = 20;
- mmu_base_pid = 1;
- } else if (cpu_has_feature(CPU_FTR_HVMODE)) {
- if (!mmu_pid_bits)
- mmu_pid_bits = 20;
-#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+ if (!cpu_has_feature(CPU_FTR_HVMODE) &&
+ cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG)) {
/*
- * When KVM is possible, we only use the top half of the
- * PID space to avoid collisions between host and guest PIDs
- * which can cause problems due to prefetch when exiting the
- * guest with AIL=3
+ * Older versions of KVM on these machines perfer if the
+ * guest only uses the low 19 PID bits.
*/
- mmu_base_pid = 1 << (mmu_pid_bits - 1);
-#else
- mmu_base_pid = 1;
-#endif
- } else {
- /* The guest uses the bottom half of the PID space */
if (!mmu_pid_bits)
mmu_pid_bits = 19;
- mmu_base_pid = 1;
+ } else {
+ if (!mmu_pid_bits)
+ mmu_pid_bits = 20;
}
+ mmu_base_pid = 1;
/*
* Allocate Partition table and process table for the
diff --git a/arch/powerpc/mm/book3s64/radix_tlb.c b/arch/powerpc/mm/book3s64/radix_tlb.c
index 409e61210789..312236a6b085 100644
--- a/arch/powerpc/mm/book3s64/radix_tlb.c
+++ b/arch/powerpc/mm/book3s64/radix_tlb.c
@@ -1336,49 +1336,3 @@ void radix__flush_tlb_all(void)
: : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(0) : "memory");
asm volatile("eieio; tlbsync; ptesync": : :"memory");
}
-
-#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
-extern void radix_kvm_prefetch_workaround(struct mm_struct *mm)
-{
- unsigned long pid = mm->context.id;
-
- if (unlikely(pid == MMU_NO_CONTEXT))
- return;
-
- if (!cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
- return;
-
- /*
- * If this context hasn't run on that CPU before and KVM is
- * around, there's a slim chance that the guest on another
- * CPU just brought in obsolete translation into the TLB of
- * this CPU due to a bad prefetch using the guest PID on
- * the way into the hypervisor.
- *
- * We work around this here. If KVM is possible, we check if
- * any sibling thread is in KVM. If it is, the window may exist
- * and thus we flush that PID from the core.
- *
- * A potential future improvement would be to mark which PIDs
- * have never been used on the system and avoid it if the PID
- * is new and the process has no other cpumask bit set.
- */
- if (cpu_has_feature(CPU_FTR_HVMODE) && radix_enabled()) {
- int cpu = smp_processor_id();
- int sib = cpu_first_thread_sibling(cpu);
- bool flush = false;
-
- for (; sib <= cpu_last_thread_sibling(cpu) && !flush; sib++) {
- if (sib == cpu)
- continue;
- if (!cpu_possible(sib))
- continue;
- if (paca_ptrs[sib]->kvm_hstate.kvm_vcpu)
- flush = true;
- }
- if (flush)
- _tlbiel_pid(pid, RIC_FLUSH_ALL);
- }
-}
-EXPORT_SYMBOL_GPL(radix_kvm_prefetch_workaround);
-#endif /* CONFIG_KVM_BOOK3S_HV_POSSIBLE */
diff --git a/arch/powerpc/mm/mmu_context.c b/arch/powerpc/mm/mmu_context.c
index 18f20da0d348..7479d39976c9 100644
--- a/arch/powerpc/mm/mmu_context.c
+++ b/arch/powerpc/mm/mmu_context.c
@@ -81,9 +81,7 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,
if (cpu_has_feature(CPU_FTR_ALTIVEC))
asm volatile ("dssall");
- if (new_on_cpu)
- radix_kvm_prefetch_workaround(next);
- else
+ if (!new_on_cpu)
membarrier_arch_switch_mm(prev, next, tsk);
/*
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 36/46] KVM: PPC: Book3S HV: Implement radix prefetch workaround by disabling MMU
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Rather than partition the guest PID space + flush a rogue guest PID to
work around this problem, instead fix it by always disabling the MMU when
switching in or out of guest MMU context in HV mode.
This may be a bit less efficient, but it is a lot less complicated and
allows the P9 path to trivally implement the workaround too. Newer CPUs
are not subject to this issue.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/mmu_context.h | 6 ----
arch/powerpc/kvm/book3s_hv.c | 21 +++++++----
arch/powerpc/kvm/book3s_hv_interrupt.c | 16 ++++++---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 34 ------------------
arch/powerpc/mm/book3s64/radix_pgtable.c | 27 +++++---------
arch/powerpc/mm/book3s64/radix_tlb.c | 46 ------------------------
arch/powerpc/mm/mmu_context.c | 4 +--
7 files changed, 36 insertions(+), 118 deletions(-)
diff --git a/arch/powerpc/include/asm/mmu_context.h b/arch/powerpc/include/asm/mmu_context.h
index 652ce85f9410..bb5c7e5e142e 100644
--- a/arch/powerpc/include/asm/mmu_context.h
+++ b/arch/powerpc/include/asm/mmu_context.h
@@ -122,12 +122,6 @@ static inline bool need_extra_context(struct mm_struct *mm, unsigned long ea)
}
#endif
-#if defined(CONFIG_KVM_BOOK3S_HV_POSSIBLE) && defined(CONFIG_PPC_RADIX_MMU)
-extern void radix_kvm_prefetch_workaround(struct mm_struct *mm);
-#else
-static inline void radix_kvm_prefetch_workaround(struct mm_struct *mm) { }
-#endif
-
extern void switch_cop(struct mm_struct *next);
extern int use_cop(unsigned long acop, struct mm_struct *mm);
extern void drop_cop(unsigned long acop, struct mm_struct *mm);
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 74c67f713a26..8480cbe4b4fa 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -809,6 +809,9 @@ static int kvmppc_h_set_mode(struct kvm_vcpu *vcpu, unsigned long mflags,
*/
if (mflags != 0 && mflags != 3)
return H_UNSUPPORTED_FLAG_START;
+ if (cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG) &&
+ kvmhv_vcpu_is_radix(vcpu) && mflags = 3)
+ return H_UNSUPPORTED_FLAG_START;
return H_TOO_HARD;
default:
return H_TOO_HARD;
@@ -1665,6 +1668,14 @@ unsigned long kvmppc_filter_lpcr_hv(struct kvmppc_vcore *vc, unsigned long lpcr)
lpcr &= ~LPCR_AIL;
if ((lpcr & LPCR_AIL) != LPCR_AIL_3)
lpcr &= ~LPCR_AIL; /* LPCR[AIL]=1/2 is disallowed */
+ /*
+ * On some POWER9s we force AIL off for radix guests to prevent
+ * executing in MSR[HV]=1 mode with the MMU enabled and PIDR set to
+ * guest, which can result in Q0 translations with LPID=0 PID=PIDR to
+ * be cached, which the host TLB management does not expect.
+ */
+ if (kvm_is_radix(kvm) && cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
+ lpcr &= ~LPCR_AIL;
/*
* On POWER9, allow userspace to enable large decrementer for the
@@ -4334,12 +4345,7 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
vcpu->arch.state = KVMPPC_VCPU_BUSY_IN_HOST;
do {
- /*
- * The TLB prefetch bug fixup is only in the kvmppc_run_vcpu
- * path, which also handles hash and dependent threads mode.
- */
- if (kvm->arch.threads_indep && kvm_is_radix(kvm) &&
- !cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
+ if (kvm->arch.threads_indep && kvm_is_radix(kvm))
r = kvmhv_run_single_vcpu(vcpu, ~(u64)0,
vcpu->arch.vcore->lpcr);
else
@@ -4962,6 +4968,9 @@ static int kvmppc_core_init_vm_hv(struct kvm *kvm)
if (!indep_threads_mode && !cpu_has_feature(CPU_FTR_HVMODE)) {
pr_warn("KVM: Ignoring indep_threads_mode=N in nested hypervisor\n");
kvm->arch.threads_indep = true;
+ } else if (!indep_threads_mode && cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG)) {
+ pr_warn("KVM: Ignoring indep_threads_mode=N on pre-DD2.2 POWER9\n");
+ kvm->arch.threads_indep = true;
} else {
kvm->arch.threads_indep = indep_threads_mode;
}
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index 787a7d2aed1a..cd84d2c37632 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -234,6 +234,9 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
mtspr(SPRN_AMOR, ~0UL);
+ if (cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
+ __mtmsrd(msr & ~(MSR_IR|MSR_DR|MSR_RI), 0);
+
switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
/*
@@ -242,7 +245,8 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
*/
mtspr(SPRN_HDEC, hdec);
- __mtmsrd(0, 1); /* clear RI */
+ if (!cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
+ __mtmsrd(0, 1); /* clear RI */
mtspr(SPRN_DAR, vcpu->arch.shregs.dar);
mtspr(SPRN_DSISR, vcpu->arch.shregs.dsisr);
@@ -357,9 +361,6 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
radix_clear_slb();
- __mtmsrd(msr, 0);
- mtspr(SPRN_CTRLT, 1);
-
accumulate_time(vcpu, &vcpu->arch.rm_exit);
/* Advance host PURR/SPURR by the amount used by guest */
@@ -426,6 +427,13 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
switch_mmu_to_host_radix(kvm, host_pidr);
+ /*
+ * If we are in real mode, only switch MMU on after the MMU is
+ * switched to host, to avoid the P9_RADIX_PREFETCH_BUG.
+ */
+ __mtmsrd(msr, 0);
+ mtspr(SPRN_CTRLT, 1);
+
end_timing(vcpu);
return trap;
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 6118e8a97ddd..61f71a7df238 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -1710,40 +1710,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
eieio
tlbsync
ptesync
-
-BEGIN_FTR_SECTION
- /* Radix: Handle the case where the guest used an illegal PID */
- LOAD_REG_ADDR(r4, mmu_base_pid)
- lwz r3, VCPU_GUEST_PID(r9)
- lwz r5, 0(r4)
- cmpw cr0,r3,r5
- blt 2f
-
- /*
- * Illegal PID, the HW might have prefetched and cached in the TLB
- * some translations for the LPID 0 / guest PID combination which
- * Linux doesn't know about, so we need to flush that PID out of
- * the TLB. First we need to set LPIDR to 0 so tlbiel applies to
- * the right context.
- */
- li r0,0
- mtspr SPRN_LPID,r0
- isync
-
- /* Then do a congruence class local flush */
- ld r6,VCPU_KVM(r9)
- lwz r0,KVM_TLB_SETS(r6)
- mtctr r0
- li r7,0x400 /* IS field = 0b01 */
- ptesync
- sldi r0,r3,32 /* RS has PID */
-1: PPC_TLBIEL(7,0,2,1,1) /* RIC=2, PRS=1, R=1 */
- addi r7,r7,0x1000
- bdnz 1b
- ptesync
-END_FTR_SECTION_IFSET(CPU_FTR_P9_RADIX_PREFETCH_BUG)
-
-2:
#endif /* CONFIG_PPC_RADIX_MMU */
/*
diff --git a/arch/powerpc/mm/book3s64/radix_pgtable.c b/arch/powerpc/mm/book3s64/radix_pgtable.c
index 98f0b243c1ab..1ea95891a79e 100644
--- a/arch/powerpc/mm/book3s64/radix_pgtable.c
+++ b/arch/powerpc/mm/book3s64/radix_pgtable.c
@@ -357,30 +357,19 @@ static void __init radix_init_pgtable(void)
}
/* Find out how many PID bits are supported */
- if (!cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG)) {
- if (!mmu_pid_bits)
- mmu_pid_bits = 20;
- mmu_base_pid = 1;
- } else if (cpu_has_feature(CPU_FTR_HVMODE)) {
- if (!mmu_pid_bits)
- mmu_pid_bits = 20;
-#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+ if (!cpu_has_feature(CPU_FTR_HVMODE) &&
+ cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG)) {
/*
- * When KVM is possible, we only use the top half of the
- * PID space to avoid collisions between host and guest PIDs
- * which can cause problems due to prefetch when exiting the
- * guest with AIL=3
+ * Older versions of KVM on these machines perfer if the
+ * guest only uses the low 19 PID bits.
*/
- mmu_base_pid = 1 << (mmu_pid_bits - 1);
-#else
- mmu_base_pid = 1;
-#endif
- } else {
- /* The guest uses the bottom half of the PID space */
if (!mmu_pid_bits)
mmu_pid_bits = 19;
- mmu_base_pid = 1;
+ } else {
+ if (!mmu_pid_bits)
+ mmu_pid_bits = 20;
}
+ mmu_base_pid = 1;
/*
* Allocate Partition table and process table for the
diff --git a/arch/powerpc/mm/book3s64/radix_tlb.c b/arch/powerpc/mm/book3s64/radix_tlb.c
index 409e61210789..312236a6b085 100644
--- a/arch/powerpc/mm/book3s64/radix_tlb.c
+++ b/arch/powerpc/mm/book3s64/radix_tlb.c
@@ -1336,49 +1336,3 @@ void radix__flush_tlb_all(void)
: : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(0) : "memory");
asm volatile("eieio; tlbsync; ptesync": : :"memory");
}
-
-#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
-extern void radix_kvm_prefetch_workaround(struct mm_struct *mm)
-{
- unsigned long pid = mm->context.id;
-
- if (unlikely(pid = MMU_NO_CONTEXT))
- return;
-
- if (!cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
- return;
-
- /*
- * If this context hasn't run on that CPU before and KVM is
- * around, there's a slim chance that the guest on another
- * CPU just brought in obsolete translation into the TLB of
- * this CPU due to a bad prefetch using the guest PID on
- * the way into the hypervisor.
- *
- * We work around this here. If KVM is possible, we check if
- * any sibling thread is in KVM. If it is, the window may exist
- * and thus we flush that PID from the core.
- *
- * A potential future improvement would be to mark which PIDs
- * have never been used on the system and avoid it if the PID
- * is new and the process has no other cpumask bit set.
- */
- if (cpu_has_feature(CPU_FTR_HVMODE) && radix_enabled()) {
- int cpu = smp_processor_id();
- int sib = cpu_first_thread_sibling(cpu);
- bool flush = false;
-
- for (; sib <= cpu_last_thread_sibling(cpu) && !flush; sib++) {
- if (sib = cpu)
- continue;
- if (!cpu_possible(sib))
- continue;
- if (paca_ptrs[sib]->kvm_hstate.kvm_vcpu)
- flush = true;
- }
- if (flush)
- _tlbiel_pid(pid, RIC_FLUSH_ALL);
- }
-}
-EXPORT_SYMBOL_GPL(radix_kvm_prefetch_workaround);
-#endif /* CONFIG_KVM_BOOK3S_HV_POSSIBLE */
diff --git a/arch/powerpc/mm/mmu_context.c b/arch/powerpc/mm/mmu_context.c
index 18f20da0d348..7479d39976c9 100644
--- a/arch/powerpc/mm/mmu_context.c
+++ b/arch/powerpc/mm/mmu_context.c
@@ -81,9 +81,7 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,
if (cpu_has_feature(CPU_FTR_ALTIVEC))
asm volatile ("dssall");
- if (new_on_cpu)
- radix_kvm_prefetch_workaround(next);
- else
+ if (!new_on_cpu)
membarrier_arch_switch_mm(prev, next, tsk);
/*
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 37/46] KVM: PPC: Book3S HV: Remove support for dependent threads mode on P9
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Radix guest support will be removed from the P7/8 path, so disallow
dependent threads mode on P9.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_host.h | 1 -
arch/powerpc/kvm/book3s_hv.c | 27 +++++----------------------
2 files changed, 5 insertions(+), 23 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h
index 05fb00d37609..dd017dfa4e65 100644
--- a/arch/powerpc/include/asm/kvm_host.h
+++ b/arch/powerpc/include/asm/kvm_host.h
@@ -304,7 +304,6 @@ struct kvm_arch {
u8 fwnmi_enabled;
u8 secure_guest;
u8 svm_enabled;
- bool threads_indep;
bool nested_enable;
bool dawr1_enabled;
pgd_t *pgtable;
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 8480cbe4b4fa..85a14ce0ea0e 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -103,13 +103,9 @@ static int target_smt_mode;
module_param(target_smt_mode, int, 0644);
MODULE_PARM_DESC(target_smt_mode, "Target threads per core (0 = max)");
-static bool indep_threads_mode = true;
-module_param(indep_threads_mode, bool, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(indep_threads_mode, "Independent-threads mode (only on POWER9)");
-
static bool one_vm_per_core;
module_param(one_vm_per_core, bool, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(one_vm_per_core, "Only run vCPUs from the same VM on a core (requires indep_threads_mode=N)");
+MODULE_PARM_DESC(one_vm_per_core, "Only run vCPUs from the same VM on a core (requires POWER8 or older)");
#ifdef CONFIG_KVM_XICS
static const struct kernel_param_ops module_param_ops = {
@@ -2247,7 +2243,7 @@ static int kvmppc_set_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
*/
static int threads_per_vcore(struct kvm *kvm)
{
- if (kvm->arch.threads_indep)
+ if (cpu_has_feature(CPU_FTR_ARCH_300))
return 1;
return threads_per_subcore;
}
@@ -4345,7 +4341,7 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
vcpu->arch.state = KVMPPC_VCPU_BUSY_IN_HOST;
do {
- if (kvm->arch.threads_indep && kvm_is_radix(kvm))
+ if (kvm_is_radix(kvm))
r = kvmhv_run_single_vcpu(vcpu, ~(u64)0,
vcpu->arch.vcore->lpcr);
else
@@ -4961,21 +4957,8 @@ static int kvmppc_core_init_vm_hv(struct kvm *kvm)
/*
* Track that we now have a HV mode VM active. This blocks secondary
* CPU threads from coming online.
- * On POWER9, we only need to do this if the "indep_threads_mode"
- * module parameter has been set to N.
*/
- if (cpu_has_feature(CPU_FTR_ARCH_300)) {
- if (!indep_threads_mode && !cpu_has_feature(CPU_FTR_HVMODE)) {
- pr_warn("KVM: Ignoring indep_threads_mode=N in nested hypervisor\n");
- kvm->arch.threads_indep = true;
- } else if (!indep_threads_mode && cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG)) {
- pr_warn("KVM: Ignoring indep_threads_mode=N on pre-DD2.2 POWER9\n");
- kvm->arch.threads_indep = true;
- } else {
- kvm->arch.threads_indep = indep_threads_mode;
- }
- }
- if (!kvm->arch.threads_indep)
+ if (!cpu_has_feature(CPU_FTR_ARCH_300))
kvm_hv_vm_activated();
/*
@@ -5016,7 +4999,7 @@ static void kvmppc_core_destroy_vm_hv(struct kvm *kvm)
{
debugfs_remove_recursive(kvm->arch.debugfs_dir);
- if (!kvm->arch.threads_indep)
+ if (!cpu_has_feature(CPU_FTR_ARCH_300))
kvm_hv_vm_deactivated();
kvmppc_free_vcores(kvm);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 37/46] KVM: PPC: Book3S HV: Remove support for dependent threads mode on P9
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Radix guest support will be removed from the P7/8 path, so disallow
dependent threads mode on P9.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_host.h | 1 -
arch/powerpc/kvm/book3s_hv.c | 27 +++++----------------------
2 files changed, 5 insertions(+), 23 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h
index 05fb00d37609..dd017dfa4e65 100644
--- a/arch/powerpc/include/asm/kvm_host.h
+++ b/arch/powerpc/include/asm/kvm_host.h
@@ -304,7 +304,6 @@ struct kvm_arch {
u8 fwnmi_enabled;
u8 secure_guest;
u8 svm_enabled;
- bool threads_indep;
bool nested_enable;
bool dawr1_enabled;
pgd_t *pgtable;
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 8480cbe4b4fa..85a14ce0ea0e 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -103,13 +103,9 @@ static int target_smt_mode;
module_param(target_smt_mode, int, 0644);
MODULE_PARM_DESC(target_smt_mode, "Target threads per core (0 = max)");
-static bool indep_threads_mode = true;
-module_param(indep_threads_mode, bool, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(indep_threads_mode, "Independent-threads mode (only on POWER9)");
-
static bool one_vm_per_core;
module_param(one_vm_per_core, bool, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(one_vm_per_core, "Only run vCPUs from the same VM on a core (requires indep_threads_mode=N)");
+MODULE_PARM_DESC(one_vm_per_core, "Only run vCPUs from the same VM on a core (requires POWER8 or older)");
#ifdef CONFIG_KVM_XICS
static const struct kernel_param_ops module_param_ops = {
@@ -2247,7 +2243,7 @@ static int kvmppc_set_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
*/
static int threads_per_vcore(struct kvm *kvm)
{
- if (kvm->arch.threads_indep)
+ if (cpu_has_feature(CPU_FTR_ARCH_300))
return 1;
return threads_per_subcore;
}
@@ -4345,7 +4341,7 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
vcpu->arch.state = KVMPPC_VCPU_BUSY_IN_HOST;
do {
- if (kvm->arch.threads_indep && kvm_is_radix(kvm))
+ if (kvm_is_radix(kvm))
r = kvmhv_run_single_vcpu(vcpu, ~(u64)0,
vcpu->arch.vcore->lpcr);
else
@@ -4961,21 +4957,8 @@ static int kvmppc_core_init_vm_hv(struct kvm *kvm)
/*
* Track that we now have a HV mode VM active. This blocks secondary
* CPU threads from coming online.
- * On POWER9, we only need to do this if the "indep_threads_mode"
- * module parameter has been set to N.
*/
- if (cpu_has_feature(CPU_FTR_ARCH_300)) {
- if (!indep_threads_mode && !cpu_has_feature(CPU_FTR_HVMODE)) {
- pr_warn("KVM: Ignoring indep_threads_mode=N in nested hypervisor\n");
- kvm->arch.threads_indep = true;
- } else if (!indep_threads_mode && cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG)) {
- pr_warn("KVM: Ignoring indep_threads_mode=N on pre-DD2.2 POWER9\n");
- kvm->arch.threads_indep = true;
- } else {
- kvm->arch.threads_indep = indep_threads_mode;
- }
- }
- if (!kvm->arch.threads_indep)
+ if (!cpu_has_feature(CPU_FTR_ARCH_300))
kvm_hv_vm_activated();
/*
@@ -5016,7 +4999,7 @@ static void kvmppc_core_destroy_vm_hv(struct kvm *kvm)
{
debugfs_remove_recursive(kvm->arch.debugfs_dir);
- if (!kvm->arch.threads_indep)
+ if (!cpu_has_feature(CPU_FTR_ARCH_300))
kvm_hv_vm_deactivated();
kvmppc_free_vcores(kvm);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 38/46] KVM: PPC: Book3S HV: Remove radix guest support from P7/8 path
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
The P9 path now runs all supported radix guest combinations, so
remove radix guest support from the P7/8 path.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 79 +------------------------
1 file changed, 3 insertions(+), 76 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 61f71a7df238..b1f3ee16fd84 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -133,15 +133,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
/* Return the trap number on this thread as the return value */
mr r3, r12
- /*
- * If we came back from the guest via a relocation-on interrupt,
- * we will be in virtual mode at this point, which makes it a
- * little easier to get back to the caller.
- */
- mfmsr r0
- andi. r0, r0, MSR_IR /* in real mode? */
- bne .Lvirt_return
-
/* RFI into the highmem handler */
mfmsr r6
li r0, MSR_RI
@@ -151,11 +142,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
mtsrr1 r7
RFI_TO_KERNEL
- /* Virtual-mode return */
-.Lvirt_return:
- mtlr r8
- blr
-
kvmppc_primary_no_guest:
/* We handle this much like a ceded vcpu */
/* put the HDEC into the DEC, since HDEC interrupts don't wake us */
@@ -899,11 +885,6 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
cmpdi r3, 512 /* 1 microsecond */
blt hdec_soon
- ld r6, VCPU_KVM(r4)
- lbz r0, KVM_RADIX(r6)
- cmpwi r0, 0
- bne 9f
-
/* For hash guest, clear out and reload the SLB */
BEGIN_MMU_FTR_SECTION
/* Radix host won't have populated the SLB, so no need to clear */
@@ -1389,11 +1370,7 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
patch_site 1b patch__call_kvm_flush_link_stack
/* For hash guest, read the guest SLB and save it away */
- ld r5, VCPU_KVM(r9)
- lbz r0, KVM_RADIX(r5)
li r5, 0
- cmpwi r0, 0
- bne 0f /* for radix, save 0 entries */
lwz r0,VCPU_SLB_NR(r9) /* number of entries in SLB */
mtctr r0
li r6,0
@@ -1432,23 +1409,6 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
slbmte r6,r5
1: addi r8,r8,16
.endr
- b guest_bypass
-
-0: /*
- * Malicious or buggy radix guests may have inserted SLB entries
- * (only 0..3 because radix always runs with UPRT=1), so these must
- * be cleared here to avoid side-channels. slbmte is used rather
- * than slbia, as it won't clear cached translations.
- */
- li r0,0
- stw r0,VCPU_SLB_MAX(r9)
- slbmte r0,r0
- li r4,1
- slbmte r0,r4
- li r4,2
- slbmte r0,r4
- li r4,3
- slbmte r0,r4
guest_bypass:
stw r12, STACK_SLOT_TRAP(r1)
@@ -1694,24 +1654,6 @@ BEGIN_FTR_SECTION
mtspr SPRN_PID, r7
END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
-#ifdef CONFIG_PPC_RADIX_MMU
- /*
- * Are we running hash or radix ?
- */
- ld r5, VCPU_KVM(r9)
- lbz r0, KVM_RADIX(r5)
- cmpwi cr2, r0, 0
- beq cr2, 2f
-
- /*
- * Radix: do eieio; tlbsync; ptesync sequence in case we
- * interrupted the guest between a tlbie and a ptesync.
- */
- eieio
- tlbsync
- ptesync
-#endif /* CONFIG_PPC_RADIX_MMU */
-
/*
* cp_abort is required if the processor supports local copy-paste
* to clear the copy buffer that was under control of the guest.
@@ -1970,8 +1912,6 @@ kvmppc_tm_emul:
* reflect the HDSI to the guest as a DSI.
*/
kvmppc_hdsi:
- ld r3, VCPU_KVM(r9)
- lbz r0, KVM_RADIX(r3)
mfspr r4, SPRN_HDAR
mfspr r6, SPRN_HDSISR
BEGIN_FTR_SECTION
@@ -1979,8 +1919,6 @@ BEGIN_FTR_SECTION
cmpdi r6, 0x7fff
beq 6f
END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
- cmpwi r0, 0
- bne .Lradix_hdsi /* on radix, just save DAR/DSISR/ASDR */
/* HPTE not found fault or protection fault? */
andis. r0, r6, (DSISR_NOHPTE | DSISR_PROTFAULT)@h
beq 1f /* if not, send it to the guest */
@@ -2057,23 +1995,11 @@ fast_interrupt_c_return:
stb r0, HSTATE_IN_GUEST(r13)
b guest_exit_cont
-.Lradix_hdsi:
- std r4, VCPU_FAULT_DAR(r9)
- stw r6, VCPU_FAULT_DSISR(r9)
-.Lradix_hisi:
- mfspr r5, SPRN_ASDR
- std r5, VCPU_FAULT_GPA(r9)
- b guest_exit_cont
-
/*
* Similarly for an HISI, reflect it to the guest as an ISI unless
* it is an HPTE not found fault for a page that we have paged out.
*/
kvmppc_hisi:
- ld r3, VCPU_KVM(r9)
- lbz r0, KVM_RADIX(r3)
- cmpwi r0, 0
- bne .Lradix_hisi /* for radix, just save ASDR */
andis. r0, r11, SRR1_ISI_NOPT@h
beq 1f
andi. r0, r11, MSR_IR /* instruction relocation enabled? */
@@ -3217,15 +3143,16 @@ BEGIN_FTR_SECTION
mtspr SPRN_DAWRX1, r0
END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
- /* Clear hash and radix guest SLB. */
+ /* Clear guest SLB. */
slbmte r0, r0
PPC_SLBIA(6)
+ ptesync
BEGIN_MMU_FTR_SECTION
b 4f
END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
- ptesync
+ /* load host SLB entries */
ld r8, PACA_SLBSHADOWPTR(r13)
.rept SLB_NUM_BOLTED
li r3, SLBSHADOW_SAVEAREA
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 38/46] KVM: PPC: Book3S HV: Remove radix guest support from P7/8 path
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
The P9 path now runs all supported radix guest combinations, so
remove radix guest support from the P7/8 path.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 79 +------------------------
1 file changed, 3 insertions(+), 76 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 61f71a7df238..b1f3ee16fd84 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -133,15 +133,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
/* Return the trap number on this thread as the return value */
mr r3, r12
- /*
- * If we came back from the guest via a relocation-on interrupt,
- * we will be in virtual mode at this point, which makes it a
- * little easier to get back to the caller.
- */
- mfmsr r0
- andi. r0, r0, MSR_IR /* in real mode? */
- bne .Lvirt_return
-
/* RFI into the highmem handler */
mfmsr r6
li r0, MSR_RI
@@ -151,11 +142,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
mtsrr1 r7
RFI_TO_KERNEL
- /* Virtual-mode return */
-.Lvirt_return:
- mtlr r8
- blr
-
kvmppc_primary_no_guest:
/* We handle this much like a ceded vcpu */
/* put the HDEC into the DEC, since HDEC interrupts don't wake us */
@@ -899,11 +885,6 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
cmpdi r3, 512 /* 1 microsecond */
blt hdec_soon
- ld r6, VCPU_KVM(r4)
- lbz r0, KVM_RADIX(r6)
- cmpwi r0, 0
- bne 9f
-
/* For hash guest, clear out and reload the SLB */
BEGIN_MMU_FTR_SECTION
/* Radix host won't have populated the SLB, so no need to clear */
@@ -1389,11 +1370,7 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
patch_site 1b patch__call_kvm_flush_link_stack
/* For hash guest, read the guest SLB and save it away */
- ld r5, VCPU_KVM(r9)
- lbz r0, KVM_RADIX(r5)
li r5, 0
- cmpwi r0, 0
- bne 0f /* for radix, save 0 entries */
lwz r0,VCPU_SLB_NR(r9) /* number of entries in SLB */
mtctr r0
li r6,0
@@ -1432,23 +1409,6 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
slbmte r6,r5
1: addi r8,r8,16
.endr
- b guest_bypass
-
-0: /*
- * Malicious or buggy radix guests may have inserted SLB entries
- * (only 0..3 because radix always runs with UPRT=1), so these must
- * be cleared here to avoid side-channels. slbmte is used rather
- * than slbia, as it won't clear cached translations.
- */
- li r0,0
- stw r0,VCPU_SLB_MAX(r9)
- slbmte r0,r0
- li r4,1
- slbmte r0,r4
- li r4,2
- slbmte r0,r4
- li r4,3
- slbmte r0,r4
guest_bypass:
stw r12, STACK_SLOT_TRAP(r1)
@@ -1694,24 +1654,6 @@ BEGIN_FTR_SECTION
mtspr SPRN_PID, r7
END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
-#ifdef CONFIG_PPC_RADIX_MMU
- /*
- * Are we running hash or radix ?
- */
- ld r5, VCPU_KVM(r9)
- lbz r0, KVM_RADIX(r5)
- cmpwi cr2, r0, 0
- beq cr2, 2f
-
- /*
- * Radix: do eieio; tlbsync; ptesync sequence in case we
- * interrupted the guest between a tlbie and a ptesync.
- */
- eieio
- tlbsync
- ptesync
-#endif /* CONFIG_PPC_RADIX_MMU */
-
/*
* cp_abort is required if the processor supports local copy-paste
* to clear the copy buffer that was under control of the guest.
@@ -1970,8 +1912,6 @@ kvmppc_tm_emul:
* reflect the HDSI to the guest as a DSI.
*/
kvmppc_hdsi:
- ld r3, VCPU_KVM(r9)
- lbz r0, KVM_RADIX(r3)
mfspr r4, SPRN_HDAR
mfspr r6, SPRN_HDSISR
BEGIN_FTR_SECTION
@@ -1979,8 +1919,6 @@ BEGIN_FTR_SECTION
cmpdi r6, 0x7fff
beq 6f
END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
- cmpwi r0, 0
- bne .Lradix_hdsi /* on radix, just save DAR/DSISR/ASDR */
/* HPTE not found fault or protection fault? */
andis. r0, r6, (DSISR_NOHPTE | DSISR_PROTFAULT)@h
beq 1f /* if not, send it to the guest */
@@ -2057,23 +1995,11 @@ fast_interrupt_c_return:
stb r0, HSTATE_IN_GUEST(r13)
b guest_exit_cont
-.Lradix_hdsi:
- std r4, VCPU_FAULT_DAR(r9)
- stw r6, VCPU_FAULT_DSISR(r9)
-.Lradix_hisi:
- mfspr r5, SPRN_ASDR
- std r5, VCPU_FAULT_GPA(r9)
- b guest_exit_cont
-
/*
* Similarly for an HISI, reflect it to the guest as an ISI unless
* it is an HPTE not found fault for a page that we have paged out.
*/
kvmppc_hisi:
- ld r3, VCPU_KVM(r9)
- lbz r0, KVM_RADIX(r3)
- cmpwi r0, 0
- bne .Lradix_hisi /* for radix, just save ASDR */
andis. r0, r11, SRR1_ISI_NOPT@h
beq 1f
andi. r0, r11, MSR_IR /* instruction relocation enabled? */
@@ -3217,15 +3143,16 @@ BEGIN_FTR_SECTION
mtspr SPRN_DAWRX1, r0
END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
- /* Clear hash and radix guest SLB. */
+ /* Clear guest SLB. */
slbmte r0, r0
PPC_SLBIA(6)
+ ptesync
BEGIN_MMU_FTR_SECTION
b 4f
END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
- ptesync
+ /* load host SLB entries */
ld r8, PACA_SLBSHADOWPTR(r13)
.rept SLB_NUM_BOLTED
li r3, SLBSHADOW_SAVEAREA
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 39/46] KVM: PPC: Book3S HV: Remove virt mode checks from real mode handlers
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Cédric Le Goater, Nicholas Piggin
Now that the P7/8 path no longer supports radix, real-mode handlers
do not need to deal with being called in virt mode.
This change effectively reverts commit acde25726bc6 ("KVM: PPC: Book3S
HV: Add radix checks in real-mode hypercall handlers").
It removes a few more real-mode tests in rm hcall handlers, which also
allows the indirect ops for the xive module to be removed from the
built-in xics rm handlers.
kvmppc_h_random is renamed to kvmppc_rm_h_random to be a bit more
descriptive of its function.
Cc: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_ppc.h | 10 +--
arch/powerpc/kvm/book3s.c | 11 +--
arch/powerpc/kvm/book3s_64_vio_hv.c | 12 ----
arch/powerpc/kvm/book3s_hv_builtin.c | 91 ++++++-------------------
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 2 +-
arch/powerpc/kvm/book3s_xive.c | 18 -----
arch/powerpc/kvm/book3s_xive.h | 7 --
arch/powerpc/kvm/book3s_xive_native.c | 10 ---
8 files changed, 23 insertions(+), 138 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
index db6646c2ade2..5dfb3f167f2c 100644
--- a/arch/powerpc/include/asm/kvm_ppc.h
+++ b/arch/powerpc/include/asm/kvm_ppc.h
@@ -659,8 +659,6 @@ extern int kvmppc_xive_get_xive(struct kvm *kvm, u32 irq, u32 *server,
u32 *priority);
extern int kvmppc_xive_int_on(struct kvm *kvm, u32 irq);
extern int kvmppc_xive_int_off(struct kvm *kvm, u32 irq);
-extern void kvmppc_xive_init_module(void);
-extern void kvmppc_xive_exit_module(void);
extern int kvmppc_xive_connect_vcpu(struct kvm_device *dev,
struct kvm_vcpu *vcpu, u32 cpu);
@@ -686,8 +684,6 @@ static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
extern int kvmppc_xive_native_connect_vcpu(struct kvm_device *dev,
struct kvm_vcpu *vcpu, u32 cpu);
extern void kvmppc_xive_native_cleanup_vcpu(struct kvm_vcpu *vcpu);
-extern void kvmppc_xive_native_init_module(void);
-extern void kvmppc_xive_native_exit_module(void);
extern int kvmppc_xive_native_get_vp(struct kvm_vcpu *vcpu,
union kvmppc_one_reg *val);
extern int kvmppc_xive_native_set_vp(struct kvm_vcpu *vcpu,
@@ -701,8 +697,6 @@ static inline int kvmppc_xive_get_xive(struct kvm *kvm, u32 irq, u32 *server,
u32 *priority) { return -1; }
static inline int kvmppc_xive_int_on(struct kvm *kvm, u32 irq) { return -1; }
static inline int kvmppc_xive_int_off(struct kvm *kvm, u32 irq) { return -1; }
-static inline void kvmppc_xive_init_module(void) { }
-static inline void kvmppc_xive_exit_module(void) { }
static inline int kvmppc_xive_connect_vcpu(struct kvm_device *dev,
struct kvm_vcpu *vcpu, u32 cpu) { return -EBUSY; }
@@ -725,8 +719,6 @@ static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
static inline int kvmppc_xive_native_connect_vcpu(struct kvm_device *dev,
struct kvm_vcpu *vcpu, u32 cpu) { return -EBUSY; }
static inline void kvmppc_xive_native_cleanup_vcpu(struct kvm_vcpu *vcpu) { }
-static inline void kvmppc_xive_native_init_module(void) { }
-static inline void kvmppc_xive_native_exit_module(void) { }
static inline int kvmppc_xive_native_get_vp(struct kvm_vcpu *vcpu,
union kvmppc_one_reg *val)
{ return 0; }
@@ -762,7 +754,7 @@ long kvmppc_rm_h_stuff_tce(struct kvm_vcpu *vcpu,
unsigned long tce_value, unsigned long npages);
long int kvmppc_rm_h_confer(struct kvm_vcpu *vcpu, int target,
unsigned int yield_count);
-long kvmppc_h_random(struct kvm_vcpu *vcpu);
+long kvmppc_rm_h_random(struct kvm_vcpu *vcpu);
void kvmhv_commence_exit(int trap);
void kvmppc_realmode_machine_check(struct kvm_vcpu *vcpu);
void kvmppc_subcore_enter_guest(void);
diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c
index 44bf567b6589..1888aedfd410 100644
--- a/arch/powerpc/kvm/book3s.c
+++ b/arch/powerpc/kvm/book3s.c
@@ -1046,13 +1046,10 @@ static int kvmppc_book3s_init(void)
#ifdef CONFIG_KVM_XICS
#ifdef CONFIG_KVM_XIVE
if (xics_on_xive()) {
- kvmppc_xive_init_module();
kvm_register_device_ops(&kvm_xive_ops, KVM_DEV_TYPE_XICS);
- if (kvmppc_xive_native_supported()) {
- kvmppc_xive_native_init_module();
+ if (kvmppc_xive_native_supported())
kvm_register_device_ops(&kvm_xive_native_ops,
KVM_DEV_TYPE_XIVE);
- }
} else
#endif
kvm_register_device_ops(&kvm_xics_ops, KVM_DEV_TYPE_XICS);
@@ -1062,12 +1059,6 @@ static int kvmppc_book3s_init(void)
static void kvmppc_book3s_exit(void)
{
-#ifdef CONFIG_KVM_XICS
- if (xics_on_xive()) {
- kvmppc_xive_exit_module();
- kvmppc_xive_native_exit_module();
- }
-#endif
#ifdef CONFIG_KVM_BOOK3S_32_HANDLER
kvmppc_book3s_exit_pr();
#endif
diff --git a/arch/powerpc/kvm/book3s_64_vio_hv.c b/arch/powerpc/kvm/book3s_64_vio_hv.c
index 083a4e037718..dc6591548f0c 100644
--- a/arch/powerpc/kvm/book3s_64_vio_hv.c
+++ b/arch/powerpc/kvm/book3s_64_vio_hv.c
@@ -391,10 +391,6 @@ long kvmppc_rm_h_put_tce(struct kvm_vcpu *vcpu, unsigned long liobn,
/* udbg_printf("H_PUT_TCE(): liobn=0x%lx ioba=0x%lx, tce=0x%lx\n", */
/* liobn, ioba, tce); */
- /* For radix, we might be in virtual mode, so punt */
- if (kvm_is_radix(vcpu->kvm))
- return H_TOO_HARD;
-
stt = kvmppc_find_table(vcpu->kvm, liobn);
if (!stt)
return H_TOO_HARD;
@@ -489,10 +485,6 @@ long kvmppc_rm_h_put_tce_indirect(struct kvm_vcpu *vcpu,
bool prereg = false;
struct kvmppc_spapr_tce_iommu_table *stit;
- /* For radix, we might be in virtual mode, so punt */
- if (kvm_is_radix(vcpu->kvm))
- return H_TOO_HARD;
-
/*
* used to check for invalidations in progress
*/
@@ -602,10 +594,6 @@ long kvmppc_rm_h_stuff_tce(struct kvm_vcpu *vcpu,
long i, ret;
struct kvmppc_spapr_tce_iommu_table *stit;
- /* For radix, we might be in virtual mode, so punt */
- if (kvm_is_radix(vcpu->kvm))
- return H_TOO_HARD;
-
stt = kvmppc_find_table(vcpu->kvm, liobn);
if (!stt)
return H_TOO_HARD;
diff --git a/arch/powerpc/kvm/book3s_hv_builtin.c b/arch/powerpc/kvm/book3s_hv_builtin.c
index 7a0e33a9c980..8d669a0e15f8 100644
--- a/arch/powerpc/kvm/book3s_hv_builtin.c
+++ b/arch/powerpc/kvm/book3s_hv_builtin.c
@@ -34,21 +34,6 @@
#include "book3s_xics.h"
#include "book3s_xive.h"
-/*
- * The XIVE module will populate these when it loads
- */
-unsigned long (*__xive_vm_h_xirr)(struct kvm_vcpu *vcpu);
-unsigned long (*__xive_vm_h_ipoll)(struct kvm_vcpu *vcpu, unsigned long server);
-int (*__xive_vm_h_ipi)(struct kvm_vcpu *vcpu, unsigned long server,
- unsigned long mfrr);
-int (*__xive_vm_h_cppr)(struct kvm_vcpu *vcpu, unsigned long cppr);
-int (*__xive_vm_h_eoi)(struct kvm_vcpu *vcpu, unsigned long xirr);
-EXPORT_SYMBOL_GPL(__xive_vm_h_xirr);
-EXPORT_SYMBOL_GPL(__xive_vm_h_ipoll);
-EXPORT_SYMBOL_GPL(__xive_vm_h_ipi);
-EXPORT_SYMBOL_GPL(__xive_vm_h_cppr);
-EXPORT_SYMBOL_GPL(__xive_vm_h_eoi);
-
/*
* Hash page table alignment on newer cpus(CPU_FTR_ARCH_206)
* should be power of 2.
@@ -196,16 +181,9 @@ int kvmppc_hwrng_present(void)
}
EXPORT_SYMBOL_GPL(kvmppc_hwrng_present);
-long kvmppc_h_random(struct kvm_vcpu *vcpu)
+long kvmppc_rm_h_random(struct kvm_vcpu *vcpu)
{
- int r;
-
- /* Only need to do the expensive mfmsr() on radix */
- if (kvm_is_radix(vcpu->kvm) && (mfmsr() & MSR_IR))
- r = powernv_get_random_long(&vcpu->arch.regs.gpr[4]);
- else
- r = powernv_get_random_real_mode(&vcpu->arch.regs.gpr[4]);
- if (r)
+ if (powernv_get_random_real_mode(&vcpu->arch.regs.gpr[4]))
return H_SUCCESS;
return H_HARDWARE;
@@ -541,22 +519,13 @@ static long kvmppc_read_one_intr(bool *again)
}
#ifdef CONFIG_KVM_XICS
-static inline bool is_rm(void)
-{
- return !(mfmsr() & MSR_DR);
-}
-
unsigned long kvmppc_rm_h_xirr(struct kvm_vcpu *vcpu)
{
if (!kvmppc_xics_enabled(vcpu))
return H_TOO_HARD;
- if (xics_on_xive()) {
- if (is_rm())
- return xive_rm_h_xirr(vcpu);
- if (unlikely(!__xive_vm_h_xirr))
- return H_NOT_AVAILABLE;
- return __xive_vm_h_xirr(vcpu);
- } else
+ if (xics_on_xive())
+ return xive_rm_h_xirr(vcpu);
+ else
return xics_rm_h_xirr(vcpu);
}
@@ -565,13 +534,9 @@ unsigned long kvmppc_rm_h_xirr_x(struct kvm_vcpu *vcpu)
if (!kvmppc_xics_enabled(vcpu))
return H_TOO_HARD;
vcpu->arch.regs.gpr[5] = get_tb();
- if (xics_on_xive()) {
- if (is_rm())
- return xive_rm_h_xirr(vcpu);
- if (unlikely(!__xive_vm_h_xirr))
- return H_NOT_AVAILABLE;
- return __xive_vm_h_xirr(vcpu);
- } else
+ if (xics_on_xive())
+ return xive_rm_h_xirr(vcpu);
+ else
return xics_rm_h_xirr(vcpu);
}
@@ -579,13 +544,9 @@ unsigned long kvmppc_rm_h_ipoll(struct kvm_vcpu *vcpu, unsigned long server)
{
if (!kvmppc_xics_enabled(vcpu))
return H_TOO_HARD;
- if (xics_on_xive()) {
- if (is_rm())
- return xive_rm_h_ipoll(vcpu, server);
- if (unlikely(!__xive_vm_h_ipoll))
- return H_NOT_AVAILABLE;
- return __xive_vm_h_ipoll(vcpu, server);
- } else
+ if (xics_on_xive())
+ return xive_rm_h_ipoll(vcpu, server);
+ else
return H_TOO_HARD;
}
@@ -594,13 +555,9 @@ int kvmppc_rm_h_ipi(struct kvm_vcpu *vcpu, unsigned long server,
{
if (!kvmppc_xics_enabled(vcpu))
return H_TOO_HARD;
- if (xics_on_xive()) {
- if (is_rm())
- return xive_rm_h_ipi(vcpu, server, mfrr);
- if (unlikely(!__xive_vm_h_ipi))
- return H_NOT_AVAILABLE;
- return __xive_vm_h_ipi(vcpu, server, mfrr);
- } else
+ if (xics_on_xive())
+ return xive_rm_h_ipi(vcpu, server, mfrr);
+ else
return xics_rm_h_ipi(vcpu, server, mfrr);
}
@@ -608,13 +565,9 @@ int kvmppc_rm_h_cppr(struct kvm_vcpu *vcpu, unsigned long cppr)
{
if (!kvmppc_xics_enabled(vcpu))
return H_TOO_HARD;
- if (xics_on_xive()) {
- if (is_rm())
- return xive_rm_h_cppr(vcpu, cppr);
- if (unlikely(!__xive_vm_h_cppr))
- return H_NOT_AVAILABLE;
- return __xive_vm_h_cppr(vcpu, cppr);
- } else
+ if (xics_on_xive())
+ return xive_rm_h_cppr(vcpu, cppr);
+ else
return xics_rm_h_cppr(vcpu, cppr);
}
@@ -622,13 +575,9 @@ int kvmppc_rm_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr)
{
if (!kvmppc_xics_enabled(vcpu))
return H_TOO_HARD;
- if (xics_on_xive()) {
- if (is_rm())
- return xive_rm_h_eoi(vcpu, xirr);
- if (unlikely(!__xive_vm_h_eoi))
- return H_NOT_AVAILABLE;
- return __xive_vm_h_eoi(vcpu, xirr);
- } else
+ if (xics_on_xive())
+ return xive_rm_h_eoi(vcpu, xirr);
+ else
return xics_rm_h_eoi(vcpu, xirr);
}
#endif /* CONFIG_KVM_XICS */
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index b1f3ee16fd84..564ca9feef35 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -2323,7 +2323,7 @@ hcall_real_table:
#else
.long 0 /* 0x2fc - H_XIRR_X*/
#endif
- .long DOTSYM(kvmppc_h_random) - hcall_real_table
+ .long DOTSYM(kvmppc_rm_h_random) - hcall_real_table
.globl hcall_real_table_end
hcall_real_table_end:
diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
index dcc07ceaf5ca..80d32b4eb898 100644
--- a/arch/powerpc/kvm/book3s_xive.c
+++ b/arch/powerpc/kvm/book3s_xive.c
@@ -2358,21 +2358,3 @@ struct kvm_device_ops kvm_xive_ops = {
.get_attr = xive_get_attr,
.has_attr = xive_has_attr,
};
-
-void kvmppc_xive_init_module(void)
-{
- __xive_vm_h_xirr = xive_vm_h_xirr;
- __xive_vm_h_ipoll = xive_vm_h_ipoll;
- __xive_vm_h_ipi = xive_vm_h_ipi;
- __xive_vm_h_cppr = xive_vm_h_cppr;
- __xive_vm_h_eoi = xive_vm_h_eoi;
-}
-
-void kvmppc_xive_exit_module(void)
-{
- __xive_vm_h_xirr = NULL;
- __xive_vm_h_ipoll = NULL;
- __xive_vm_h_ipi = NULL;
- __xive_vm_h_cppr = NULL;
- __xive_vm_h_eoi = NULL;
-}
diff --git a/arch/powerpc/kvm/book3s_xive.h b/arch/powerpc/kvm/book3s_xive.h
index 86c24a4ad809..afe9eeac6d56 100644
--- a/arch/powerpc/kvm/book3s_xive.h
+++ b/arch/powerpc/kvm/book3s_xive.h
@@ -289,13 +289,6 @@ extern int xive_rm_h_ipi(struct kvm_vcpu *vcpu, unsigned long server,
extern int xive_rm_h_cppr(struct kvm_vcpu *vcpu, unsigned long cppr);
extern int xive_rm_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr);
-extern unsigned long (*__xive_vm_h_xirr)(struct kvm_vcpu *vcpu);
-extern unsigned long (*__xive_vm_h_ipoll)(struct kvm_vcpu *vcpu, unsigned long server);
-extern int (*__xive_vm_h_ipi)(struct kvm_vcpu *vcpu, unsigned long server,
- unsigned long mfrr);
-extern int (*__xive_vm_h_cppr)(struct kvm_vcpu *vcpu, unsigned long cppr);
-extern int (*__xive_vm_h_eoi)(struct kvm_vcpu *vcpu, unsigned long xirr);
-
/*
* Common Xive routines for XICS-over-XIVE and XIVE native
*/
diff --git a/arch/powerpc/kvm/book3s_xive_native.c b/arch/powerpc/kvm/book3s_xive_native.c
index 76800c84f2a3..1253666dd4d8 100644
--- a/arch/powerpc/kvm/book3s_xive_native.c
+++ b/arch/powerpc/kvm/book3s_xive_native.c
@@ -1281,13 +1281,3 @@ struct kvm_device_ops kvm_xive_native_ops = {
.has_attr = kvmppc_xive_native_has_attr,
.mmap = kvmppc_xive_native_mmap,
};
-
-void kvmppc_xive_native_init_module(void)
-{
- ;
-}
-
-void kvmppc_xive_native_exit_module(void)
-{
- ;
-}
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 39/46] KVM: PPC: Book3S HV: Remove virt mode checks from real mode handlers
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Cédric Le Goater, Nicholas Piggin
Now that the P7/8 path no longer supports radix, real-mode handlers
do not need to deal with being called in virt mode.
This change effectively reverts commit acde25726bc6 ("KVM: PPC: Book3S
HV: Add radix checks in real-mode hypercall handlers").
It removes a few more real-mode tests in rm hcall handlers, which also
allows the indirect ops for the xive module to be removed from the
built-in xics rm handlers.
kvmppc_h_random is renamed to kvmppc_rm_h_random to be a bit more
descriptive of its function.
Cc: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/kvm_ppc.h | 10 +--
arch/powerpc/kvm/book3s.c | 11 +--
arch/powerpc/kvm/book3s_64_vio_hv.c | 12 ----
arch/powerpc/kvm/book3s_hv_builtin.c | 91 ++++++-------------------
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 2 +-
arch/powerpc/kvm/book3s_xive.c | 18 -----
arch/powerpc/kvm/book3s_xive.h | 7 --
arch/powerpc/kvm/book3s_xive_native.c | 10 ---
8 files changed, 23 insertions(+), 138 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
index db6646c2ade2..5dfb3f167f2c 100644
--- a/arch/powerpc/include/asm/kvm_ppc.h
+++ b/arch/powerpc/include/asm/kvm_ppc.h
@@ -659,8 +659,6 @@ extern int kvmppc_xive_get_xive(struct kvm *kvm, u32 irq, u32 *server,
u32 *priority);
extern int kvmppc_xive_int_on(struct kvm *kvm, u32 irq);
extern int kvmppc_xive_int_off(struct kvm *kvm, u32 irq);
-extern void kvmppc_xive_init_module(void);
-extern void kvmppc_xive_exit_module(void);
extern int kvmppc_xive_connect_vcpu(struct kvm_device *dev,
struct kvm_vcpu *vcpu, u32 cpu);
@@ -686,8 +684,6 @@ static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
extern int kvmppc_xive_native_connect_vcpu(struct kvm_device *dev,
struct kvm_vcpu *vcpu, u32 cpu);
extern void kvmppc_xive_native_cleanup_vcpu(struct kvm_vcpu *vcpu);
-extern void kvmppc_xive_native_init_module(void);
-extern void kvmppc_xive_native_exit_module(void);
extern int kvmppc_xive_native_get_vp(struct kvm_vcpu *vcpu,
union kvmppc_one_reg *val);
extern int kvmppc_xive_native_set_vp(struct kvm_vcpu *vcpu,
@@ -701,8 +697,6 @@ static inline int kvmppc_xive_get_xive(struct kvm *kvm, u32 irq, u32 *server,
u32 *priority) { return -1; }
static inline int kvmppc_xive_int_on(struct kvm *kvm, u32 irq) { return -1; }
static inline int kvmppc_xive_int_off(struct kvm *kvm, u32 irq) { return -1; }
-static inline void kvmppc_xive_init_module(void) { }
-static inline void kvmppc_xive_exit_module(void) { }
static inline int kvmppc_xive_connect_vcpu(struct kvm_device *dev,
struct kvm_vcpu *vcpu, u32 cpu) { return -EBUSY; }
@@ -725,8 +719,6 @@ static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
static inline int kvmppc_xive_native_connect_vcpu(struct kvm_device *dev,
struct kvm_vcpu *vcpu, u32 cpu) { return -EBUSY; }
static inline void kvmppc_xive_native_cleanup_vcpu(struct kvm_vcpu *vcpu) { }
-static inline void kvmppc_xive_native_init_module(void) { }
-static inline void kvmppc_xive_native_exit_module(void) { }
static inline int kvmppc_xive_native_get_vp(struct kvm_vcpu *vcpu,
union kvmppc_one_reg *val)
{ return 0; }
@@ -762,7 +754,7 @@ long kvmppc_rm_h_stuff_tce(struct kvm_vcpu *vcpu,
unsigned long tce_value, unsigned long npages);
long int kvmppc_rm_h_confer(struct kvm_vcpu *vcpu, int target,
unsigned int yield_count);
-long kvmppc_h_random(struct kvm_vcpu *vcpu);
+long kvmppc_rm_h_random(struct kvm_vcpu *vcpu);
void kvmhv_commence_exit(int trap);
void kvmppc_realmode_machine_check(struct kvm_vcpu *vcpu);
void kvmppc_subcore_enter_guest(void);
diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c
index 44bf567b6589..1888aedfd410 100644
--- a/arch/powerpc/kvm/book3s.c
+++ b/arch/powerpc/kvm/book3s.c
@@ -1046,13 +1046,10 @@ static int kvmppc_book3s_init(void)
#ifdef CONFIG_KVM_XICS
#ifdef CONFIG_KVM_XIVE
if (xics_on_xive()) {
- kvmppc_xive_init_module();
kvm_register_device_ops(&kvm_xive_ops, KVM_DEV_TYPE_XICS);
- if (kvmppc_xive_native_supported()) {
- kvmppc_xive_native_init_module();
+ if (kvmppc_xive_native_supported())
kvm_register_device_ops(&kvm_xive_native_ops,
KVM_DEV_TYPE_XIVE);
- }
} else
#endif
kvm_register_device_ops(&kvm_xics_ops, KVM_DEV_TYPE_XICS);
@@ -1062,12 +1059,6 @@ static int kvmppc_book3s_init(void)
static void kvmppc_book3s_exit(void)
{
-#ifdef CONFIG_KVM_XICS
- if (xics_on_xive()) {
- kvmppc_xive_exit_module();
- kvmppc_xive_native_exit_module();
- }
-#endif
#ifdef CONFIG_KVM_BOOK3S_32_HANDLER
kvmppc_book3s_exit_pr();
#endif
diff --git a/arch/powerpc/kvm/book3s_64_vio_hv.c b/arch/powerpc/kvm/book3s_64_vio_hv.c
index 083a4e037718..dc6591548f0c 100644
--- a/arch/powerpc/kvm/book3s_64_vio_hv.c
+++ b/arch/powerpc/kvm/book3s_64_vio_hv.c
@@ -391,10 +391,6 @@ long kvmppc_rm_h_put_tce(struct kvm_vcpu *vcpu, unsigned long liobn,
/* udbg_printf("H_PUT_TCE(): liobn=0x%lx ioba=0x%lx, tce=0x%lx\n", */
/* liobn, ioba, tce); */
- /* For radix, we might be in virtual mode, so punt */
- if (kvm_is_radix(vcpu->kvm))
- return H_TOO_HARD;
-
stt = kvmppc_find_table(vcpu->kvm, liobn);
if (!stt)
return H_TOO_HARD;
@@ -489,10 +485,6 @@ long kvmppc_rm_h_put_tce_indirect(struct kvm_vcpu *vcpu,
bool prereg = false;
struct kvmppc_spapr_tce_iommu_table *stit;
- /* For radix, we might be in virtual mode, so punt */
- if (kvm_is_radix(vcpu->kvm))
- return H_TOO_HARD;
-
/*
* used to check for invalidations in progress
*/
@@ -602,10 +594,6 @@ long kvmppc_rm_h_stuff_tce(struct kvm_vcpu *vcpu,
long i, ret;
struct kvmppc_spapr_tce_iommu_table *stit;
- /* For radix, we might be in virtual mode, so punt */
- if (kvm_is_radix(vcpu->kvm))
- return H_TOO_HARD;
-
stt = kvmppc_find_table(vcpu->kvm, liobn);
if (!stt)
return H_TOO_HARD;
diff --git a/arch/powerpc/kvm/book3s_hv_builtin.c b/arch/powerpc/kvm/book3s_hv_builtin.c
index 7a0e33a9c980..8d669a0e15f8 100644
--- a/arch/powerpc/kvm/book3s_hv_builtin.c
+++ b/arch/powerpc/kvm/book3s_hv_builtin.c
@@ -34,21 +34,6 @@
#include "book3s_xics.h"
#include "book3s_xive.h"
-/*
- * The XIVE module will populate these when it loads
- */
-unsigned long (*__xive_vm_h_xirr)(struct kvm_vcpu *vcpu);
-unsigned long (*__xive_vm_h_ipoll)(struct kvm_vcpu *vcpu, unsigned long server);
-int (*__xive_vm_h_ipi)(struct kvm_vcpu *vcpu, unsigned long server,
- unsigned long mfrr);
-int (*__xive_vm_h_cppr)(struct kvm_vcpu *vcpu, unsigned long cppr);
-int (*__xive_vm_h_eoi)(struct kvm_vcpu *vcpu, unsigned long xirr);
-EXPORT_SYMBOL_GPL(__xive_vm_h_xirr);
-EXPORT_SYMBOL_GPL(__xive_vm_h_ipoll);
-EXPORT_SYMBOL_GPL(__xive_vm_h_ipi);
-EXPORT_SYMBOL_GPL(__xive_vm_h_cppr);
-EXPORT_SYMBOL_GPL(__xive_vm_h_eoi);
-
/*
* Hash page table alignment on newer cpus(CPU_FTR_ARCH_206)
* should be power of 2.
@@ -196,16 +181,9 @@ int kvmppc_hwrng_present(void)
}
EXPORT_SYMBOL_GPL(kvmppc_hwrng_present);
-long kvmppc_h_random(struct kvm_vcpu *vcpu)
+long kvmppc_rm_h_random(struct kvm_vcpu *vcpu)
{
- int r;
-
- /* Only need to do the expensive mfmsr() on radix */
- if (kvm_is_radix(vcpu->kvm) && (mfmsr() & MSR_IR))
- r = powernv_get_random_long(&vcpu->arch.regs.gpr[4]);
- else
- r = powernv_get_random_real_mode(&vcpu->arch.regs.gpr[4]);
- if (r)
+ if (powernv_get_random_real_mode(&vcpu->arch.regs.gpr[4]))
return H_SUCCESS;
return H_HARDWARE;
@@ -541,22 +519,13 @@ static long kvmppc_read_one_intr(bool *again)
}
#ifdef CONFIG_KVM_XICS
-static inline bool is_rm(void)
-{
- return !(mfmsr() & MSR_DR);
-}
-
unsigned long kvmppc_rm_h_xirr(struct kvm_vcpu *vcpu)
{
if (!kvmppc_xics_enabled(vcpu))
return H_TOO_HARD;
- if (xics_on_xive()) {
- if (is_rm())
- return xive_rm_h_xirr(vcpu);
- if (unlikely(!__xive_vm_h_xirr))
- return H_NOT_AVAILABLE;
- return __xive_vm_h_xirr(vcpu);
- } else
+ if (xics_on_xive())
+ return xive_rm_h_xirr(vcpu);
+ else
return xics_rm_h_xirr(vcpu);
}
@@ -565,13 +534,9 @@ unsigned long kvmppc_rm_h_xirr_x(struct kvm_vcpu *vcpu)
if (!kvmppc_xics_enabled(vcpu))
return H_TOO_HARD;
vcpu->arch.regs.gpr[5] = get_tb();
- if (xics_on_xive()) {
- if (is_rm())
- return xive_rm_h_xirr(vcpu);
- if (unlikely(!__xive_vm_h_xirr))
- return H_NOT_AVAILABLE;
- return __xive_vm_h_xirr(vcpu);
- } else
+ if (xics_on_xive())
+ return xive_rm_h_xirr(vcpu);
+ else
return xics_rm_h_xirr(vcpu);
}
@@ -579,13 +544,9 @@ unsigned long kvmppc_rm_h_ipoll(struct kvm_vcpu *vcpu, unsigned long server)
{
if (!kvmppc_xics_enabled(vcpu))
return H_TOO_HARD;
- if (xics_on_xive()) {
- if (is_rm())
- return xive_rm_h_ipoll(vcpu, server);
- if (unlikely(!__xive_vm_h_ipoll))
- return H_NOT_AVAILABLE;
- return __xive_vm_h_ipoll(vcpu, server);
- } else
+ if (xics_on_xive())
+ return xive_rm_h_ipoll(vcpu, server);
+ else
return H_TOO_HARD;
}
@@ -594,13 +555,9 @@ int kvmppc_rm_h_ipi(struct kvm_vcpu *vcpu, unsigned long server,
{
if (!kvmppc_xics_enabled(vcpu))
return H_TOO_HARD;
- if (xics_on_xive()) {
- if (is_rm())
- return xive_rm_h_ipi(vcpu, server, mfrr);
- if (unlikely(!__xive_vm_h_ipi))
- return H_NOT_AVAILABLE;
- return __xive_vm_h_ipi(vcpu, server, mfrr);
- } else
+ if (xics_on_xive())
+ return xive_rm_h_ipi(vcpu, server, mfrr);
+ else
return xics_rm_h_ipi(vcpu, server, mfrr);
}
@@ -608,13 +565,9 @@ int kvmppc_rm_h_cppr(struct kvm_vcpu *vcpu, unsigned long cppr)
{
if (!kvmppc_xics_enabled(vcpu))
return H_TOO_HARD;
- if (xics_on_xive()) {
- if (is_rm())
- return xive_rm_h_cppr(vcpu, cppr);
- if (unlikely(!__xive_vm_h_cppr))
- return H_NOT_AVAILABLE;
- return __xive_vm_h_cppr(vcpu, cppr);
- } else
+ if (xics_on_xive())
+ return xive_rm_h_cppr(vcpu, cppr);
+ else
return xics_rm_h_cppr(vcpu, cppr);
}
@@ -622,13 +575,9 @@ int kvmppc_rm_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr)
{
if (!kvmppc_xics_enabled(vcpu))
return H_TOO_HARD;
- if (xics_on_xive()) {
- if (is_rm())
- return xive_rm_h_eoi(vcpu, xirr);
- if (unlikely(!__xive_vm_h_eoi))
- return H_NOT_AVAILABLE;
- return __xive_vm_h_eoi(vcpu, xirr);
- } else
+ if (xics_on_xive())
+ return xive_rm_h_eoi(vcpu, xirr);
+ else
return xics_rm_h_eoi(vcpu, xirr);
}
#endif /* CONFIG_KVM_XICS */
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index b1f3ee16fd84..564ca9feef35 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -2323,7 +2323,7 @@ hcall_real_table:
#else
.long 0 /* 0x2fc - H_XIRR_X*/
#endif
- .long DOTSYM(kvmppc_h_random) - hcall_real_table
+ .long DOTSYM(kvmppc_rm_h_random) - hcall_real_table
.globl hcall_real_table_end
hcall_real_table_end:
diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
index dcc07ceaf5ca..80d32b4eb898 100644
--- a/arch/powerpc/kvm/book3s_xive.c
+++ b/arch/powerpc/kvm/book3s_xive.c
@@ -2358,21 +2358,3 @@ struct kvm_device_ops kvm_xive_ops = {
.get_attr = xive_get_attr,
.has_attr = xive_has_attr,
};
-
-void kvmppc_xive_init_module(void)
-{
- __xive_vm_h_xirr = xive_vm_h_xirr;
- __xive_vm_h_ipoll = xive_vm_h_ipoll;
- __xive_vm_h_ipi = xive_vm_h_ipi;
- __xive_vm_h_cppr = xive_vm_h_cppr;
- __xive_vm_h_eoi = xive_vm_h_eoi;
-}
-
-void kvmppc_xive_exit_module(void)
-{
- __xive_vm_h_xirr = NULL;
- __xive_vm_h_ipoll = NULL;
- __xive_vm_h_ipi = NULL;
- __xive_vm_h_cppr = NULL;
- __xive_vm_h_eoi = NULL;
-}
diff --git a/arch/powerpc/kvm/book3s_xive.h b/arch/powerpc/kvm/book3s_xive.h
index 86c24a4ad809..afe9eeac6d56 100644
--- a/arch/powerpc/kvm/book3s_xive.h
+++ b/arch/powerpc/kvm/book3s_xive.h
@@ -289,13 +289,6 @@ extern int xive_rm_h_ipi(struct kvm_vcpu *vcpu, unsigned long server,
extern int xive_rm_h_cppr(struct kvm_vcpu *vcpu, unsigned long cppr);
extern int xive_rm_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr);
-extern unsigned long (*__xive_vm_h_xirr)(struct kvm_vcpu *vcpu);
-extern unsigned long (*__xive_vm_h_ipoll)(struct kvm_vcpu *vcpu, unsigned long server);
-extern int (*__xive_vm_h_ipi)(struct kvm_vcpu *vcpu, unsigned long server,
- unsigned long mfrr);
-extern int (*__xive_vm_h_cppr)(struct kvm_vcpu *vcpu, unsigned long cppr);
-extern int (*__xive_vm_h_eoi)(struct kvm_vcpu *vcpu, unsigned long xirr);
-
/*
* Common Xive routines for XICS-over-XIVE and XIVE native
*/
diff --git a/arch/powerpc/kvm/book3s_xive_native.c b/arch/powerpc/kvm/book3s_xive_native.c
index 76800c84f2a3..1253666dd4d8 100644
--- a/arch/powerpc/kvm/book3s_xive_native.c
+++ b/arch/powerpc/kvm/book3s_xive_native.c
@@ -1281,13 +1281,3 @@ struct kvm_device_ops kvm_xive_native_ops = {
.has_attr = kvmppc_xive_native_has_attr,
.mmap = kvmppc_xive_native_mmap,
};
-
-void kvmppc_xive_native_init_module(void)
-{
- ;
-}
-
-void kvmppc_xive_native_exit_module(void)
-{
- ;
-}
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* Re: [PATCH v4 39/46] KVM: PPC: Book3S HV: Remove virt mode checks from real mode handlers
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 13:38 ` Cédric Le Goater
-1 siblings, 0 replies; 188+ messages in thread
From: Cédric Le Goater @ 2021-03-23 13:38 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 3/23/21 2:02 AM, Nicholas Piggin wrote:
> Now that the P7/8 path no longer supports radix, real-mode handlers
> do not need to deal with being called in virt mode.
>
> This change effectively reverts commit acde25726bc6 ("KVM: PPC: Book3S
> HV: Add radix checks in real-mode hypercall handlers").
>
> It removes a few more real-mode tests in rm hcall handlers, which also
> allows the indirect ops for the xive module to be removed from the
> built-in xics rm handlers.
>
> kvmppc_h_random is renamed to kvmppc_rm_h_random to be a bit more
> descriptive of its function.
>
> Cc: Cédric Le Goater <clg@kaod.org>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
> ---
> arch/powerpc/include/asm/kvm_ppc.h | 10 +--
> arch/powerpc/kvm/book3s.c | 11 +--
> arch/powerpc/kvm/book3s_64_vio_hv.c | 12 ----
> arch/powerpc/kvm/book3s_hv_builtin.c | 91 ++++++-------------------
> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 2 +-
> arch/powerpc/kvm/book3s_xive.c | 18 -----
> arch/powerpc/kvm/book3s_xive.h | 7 --
> arch/powerpc/kvm/book3s_xive_native.c | 10 ---
> 8 files changed, 23 insertions(+), 138 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
> index db6646c2ade2..5dfb3f167f2c 100644
> --- a/arch/powerpc/include/asm/kvm_ppc.h
> +++ b/arch/powerpc/include/asm/kvm_ppc.h
> @@ -659,8 +659,6 @@ extern int kvmppc_xive_get_xive(struct kvm *kvm, u32 irq, u32 *server,
> u32 *priority);
> extern int kvmppc_xive_int_on(struct kvm *kvm, u32 irq);
> extern int kvmppc_xive_int_off(struct kvm *kvm, u32 irq);
> -extern void kvmppc_xive_init_module(void);
> -extern void kvmppc_xive_exit_module(void);
>
> extern int kvmppc_xive_connect_vcpu(struct kvm_device *dev,
> struct kvm_vcpu *vcpu, u32 cpu);
> @@ -686,8 +684,6 @@ static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> extern int kvmppc_xive_native_connect_vcpu(struct kvm_device *dev,
> struct kvm_vcpu *vcpu, u32 cpu);
> extern void kvmppc_xive_native_cleanup_vcpu(struct kvm_vcpu *vcpu);
> -extern void kvmppc_xive_native_init_module(void);
> -extern void kvmppc_xive_native_exit_module(void);
> extern int kvmppc_xive_native_get_vp(struct kvm_vcpu *vcpu,
> union kvmppc_one_reg *val);
> extern int kvmppc_xive_native_set_vp(struct kvm_vcpu *vcpu,
> @@ -701,8 +697,6 @@ static inline int kvmppc_xive_get_xive(struct kvm *kvm, u32 irq, u32 *server,
> u32 *priority) { return -1; }
> static inline int kvmppc_xive_int_on(struct kvm *kvm, u32 irq) { return -1; }
> static inline int kvmppc_xive_int_off(struct kvm *kvm, u32 irq) { return -1; }
> -static inline void kvmppc_xive_init_module(void) { }
> -static inline void kvmppc_xive_exit_module(void) { }
>
> static inline int kvmppc_xive_connect_vcpu(struct kvm_device *dev,
> struct kvm_vcpu *vcpu, u32 cpu) { return -EBUSY; }
> @@ -725,8 +719,6 @@ static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> static inline int kvmppc_xive_native_connect_vcpu(struct kvm_device *dev,
> struct kvm_vcpu *vcpu, u32 cpu) { return -EBUSY; }
> static inline void kvmppc_xive_native_cleanup_vcpu(struct kvm_vcpu *vcpu) { }
> -static inline void kvmppc_xive_native_init_module(void) { }
> -static inline void kvmppc_xive_native_exit_module(void) { }
> static inline int kvmppc_xive_native_get_vp(struct kvm_vcpu *vcpu,
> union kvmppc_one_reg *val)
> { return 0; }
> @@ -762,7 +754,7 @@ long kvmppc_rm_h_stuff_tce(struct kvm_vcpu *vcpu,
> unsigned long tce_value, unsigned long npages);
> long int kvmppc_rm_h_confer(struct kvm_vcpu *vcpu, int target,
> unsigned int yield_count);
> -long kvmppc_h_random(struct kvm_vcpu *vcpu);
> +long kvmppc_rm_h_random(struct kvm_vcpu *vcpu);
> void kvmhv_commence_exit(int trap);
> void kvmppc_realmode_machine_check(struct kvm_vcpu *vcpu);
> void kvmppc_subcore_enter_guest(void);
> diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c
> index 44bf567b6589..1888aedfd410 100644
> --- a/arch/powerpc/kvm/book3s.c
> +++ b/arch/powerpc/kvm/book3s.c
> @@ -1046,13 +1046,10 @@ static int kvmppc_book3s_init(void)
> #ifdef CONFIG_KVM_XICS
> #ifdef CONFIG_KVM_XIVE
> if (xics_on_xive()) {
> - kvmppc_xive_init_module();
> kvm_register_device_ops(&kvm_xive_ops, KVM_DEV_TYPE_XICS);
> - if (kvmppc_xive_native_supported()) {
> - kvmppc_xive_native_init_module();
> + if (kvmppc_xive_native_supported())
> kvm_register_device_ops(&kvm_xive_native_ops,
> KVM_DEV_TYPE_XIVE);
> - }
> } else
> #endif
> kvm_register_device_ops(&kvm_xics_ops, KVM_DEV_TYPE_XICS);
> @@ -1062,12 +1059,6 @@ static int kvmppc_book3s_init(void)
>
> static void kvmppc_book3s_exit(void)
> {
> -#ifdef CONFIG_KVM_XICS
> - if (xics_on_xive()) {
> - kvmppc_xive_exit_module();
> - kvmppc_xive_native_exit_module();
> - }
> -#endif
> #ifdef CONFIG_KVM_BOOK3S_32_HANDLER
> kvmppc_book3s_exit_pr();
> #endif
> diff --git a/arch/powerpc/kvm/book3s_64_vio_hv.c b/arch/powerpc/kvm/book3s_64_vio_hv.c
> index 083a4e037718..dc6591548f0c 100644
> --- a/arch/powerpc/kvm/book3s_64_vio_hv.c
> +++ b/arch/powerpc/kvm/book3s_64_vio_hv.c
> @@ -391,10 +391,6 @@ long kvmppc_rm_h_put_tce(struct kvm_vcpu *vcpu, unsigned long liobn,
> /* udbg_printf("H_PUT_TCE(): liobn=0x%lx ioba=0x%lx, tce=0x%lx\n", */
> /* liobn, ioba, tce); */
>
> - /* For radix, we might be in virtual mode, so punt */
> - if (kvm_is_radix(vcpu->kvm))
> - return H_TOO_HARD;
> -
> stt = kvmppc_find_table(vcpu->kvm, liobn);
> if (!stt)
> return H_TOO_HARD;
> @@ -489,10 +485,6 @@ long kvmppc_rm_h_put_tce_indirect(struct kvm_vcpu *vcpu,
> bool prereg = false;
> struct kvmppc_spapr_tce_iommu_table *stit;
>
> - /* For radix, we might be in virtual mode, so punt */
> - if (kvm_is_radix(vcpu->kvm))
> - return H_TOO_HARD;
> -
> /*
> * used to check for invalidations in progress
> */
> @@ -602,10 +594,6 @@ long kvmppc_rm_h_stuff_tce(struct kvm_vcpu *vcpu,
> long i, ret;
> struct kvmppc_spapr_tce_iommu_table *stit;
>
> - /* For radix, we might be in virtual mode, so punt */
> - if (kvm_is_radix(vcpu->kvm))
> - return H_TOO_HARD;
> -
> stt = kvmppc_find_table(vcpu->kvm, liobn);
> if (!stt)
> return H_TOO_HARD;
> diff --git a/arch/powerpc/kvm/book3s_hv_builtin.c b/arch/powerpc/kvm/book3s_hv_builtin.c
> index 7a0e33a9c980..8d669a0e15f8 100644
> --- a/arch/powerpc/kvm/book3s_hv_builtin.c
> +++ b/arch/powerpc/kvm/book3s_hv_builtin.c
> @@ -34,21 +34,6 @@
> #include "book3s_xics.h"
> #include "book3s_xive.h"
>
> -/*
> - * The XIVE module will populate these when it loads
> - */
> -unsigned long (*__xive_vm_h_xirr)(struct kvm_vcpu *vcpu);
> -unsigned long (*__xive_vm_h_ipoll)(struct kvm_vcpu *vcpu, unsigned long server);
> -int (*__xive_vm_h_ipi)(struct kvm_vcpu *vcpu, unsigned long server,
> - unsigned long mfrr);
> -int (*__xive_vm_h_cppr)(struct kvm_vcpu *vcpu, unsigned long cppr);
> -int (*__xive_vm_h_eoi)(struct kvm_vcpu *vcpu, unsigned long xirr);
> -EXPORT_SYMBOL_GPL(__xive_vm_h_xirr);
> -EXPORT_SYMBOL_GPL(__xive_vm_h_ipoll);
> -EXPORT_SYMBOL_GPL(__xive_vm_h_ipi);
> -EXPORT_SYMBOL_GPL(__xive_vm_h_cppr);
> -EXPORT_SYMBOL_GPL(__xive_vm_h_eoi);
> -
> /*
> * Hash page table alignment on newer cpus(CPU_FTR_ARCH_206)
> * should be power of 2.
> @@ -196,16 +181,9 @@ int kvmppc_hwrng_present(void)
> }
> EXPORT_SYMBOL_GPL(kvmppc_hwrng_present);
>
> -long kvmppc_h_random(struct kvm_vcpu *vcpu)
> +long kvmppc_rm_h_random(struct kvm_vcpu *vcpu)
> {
> - int r;
> -
> - /* Only need to do the expensive mfmsr() on radix */
> - if (kvm_is_radix(vcpu->kvm) && (mfmsr() & MSR_IR))
> - r = powernv_get_random_long(&vcpu->arch.regs.gpr[4]);
> - else
> - r = powernv_get_random_real_mode(&vcpu->arch.regs.gpr[4]);
> - if (r)
> + if (powernv_get_random_real_mode(&vcpu->arch.regs.gpr[4]))
> return H_SUCCESS;
>
> return H_HARDWARE;
> @@ -541,22 +519,13 @@ static long kvmppc_read_one_intr(bool *again)
> }
>
> #ifdef CONFIG_KVM_XICS
> -static inline bool is_rm(void)
> -{
> - return !(mfmsr() & MSR_DR);
> -}
> -
> unsigned long kvmppc_rm_h_xirr(struct kvm_vcpu *vcpu)
> {
> if (!kvmppc_xics_enabled(vcpu))
> return H_TOO_HARD;
> - if (xics_on_xive()) {
> - if (is_rm())
> - return xive_rm_h_xirr(vcpu);
> - if (unlikely(!__xive_vm_h_xirr))
> - return H_NOT_AVAILABLE;
> - return __xive_vm_h_xirr(vcpu);
> - } else
> + if (xics_on_xive())
> + return xive_rm_h_xirr(vcpu);
> + else
> return xics_rm_h_xirr(vcpu);
> }
>
> @@ -565,13 +534,9 @@ unsigned long kvmppc_rm_h_xirr_x(struct kvm_vcpu *vcpu)
> if (!kvmppc_xics_enabled(vcpu))
> return H_TOO_HARD;
> vcpu->arch.regs.gpr[5] = get_tb();
> - if (xics_on_xive()) {
> - if (is_rm())
> - return xive_rm_h_xirr(vcpu);
> - if (unlikely(!__xive_vm_h_xirr))
> - return H_NOT_AVAILABLE;
> - return __xive_vm_h_xirr(vcpu);
> - } else
> + if (xics_on_xive())
> + return xive_rm_h_xirr(vcpu);
> + else
> return xics_rm_h_xirr(vcpu);
> }
>
> @@ -579,13 +544,9 @@ unsigned long kvmppc_rm_h_ipoll(struct kvm_vcpu *vcpu, unsigned long server)
> {
> if (!kvmppc_xics_enabled(vcpu))
> return H_TOO_HARD;
> - if (xics_on_xive()) {
> - if (is_rm())
> - return xive_rm_h_ipoll(vcpu, server);
> - if (unlikely(!__xive_vm_h_ipoll))
> - return H_NOT_AVAILABLE;
> - return __xive_vm_h_ipoll(vcpu, server);
> - } else
> + if (xics_on_xive())
> + return xive_rm_h_ipoll(vcpu, server);
> + else
> return H_TOO_HARD;
> }
>
> @@ -594,13 +555,9 @@ int kvmppc_rm_h_ipi(struct kvm_vcpu *vcpu, unsigned long server,
> {
> if (!kvmppc_xics_enabled(vcpu))
> return H_TOO_HARD;
> - if (xics_on_xive()) {
> - if (is_rm())
> - return xive_rm_h_ipi(vcpu, server, mfrr);
> - if (unlikely(!__xive_vm_h_ipi))
> - return H_NOT_AVAILABLE;
> - return __xive_vm_h_ipi(vcpu, server, mfrr);
> - } else
> + if (xics_on_xive())
> + return xive_rm_h_ipi(vcpu, server, mfrr);
> + else
> return xics_rm_h_ipi(vcpu, server, mfrr);
> }
>
> @@ -608,13 +565,9 @@ int kvmppc_rm_h_cppr(struct kvm_vcpu *vcpu, unsigned long cppr)
> {
> if (!kvmppc_xics_enabled(vcpu))
> return H_TOO_HARD;
> - if (xics_on_xive()) {
> - if (is_rm())
> - return xive_rm_h_cppr(vcpu, cppr);
> - if (unlikely(!__xive_vm_h_cppr))
> - return H_NOT_AVAILABLE;
> - return __xive_vm_h_cppr(vcpu, cppr);
> - } else
> + if (xics_on_xive())
> + return xive_rm_h_cppr(vcpu, cppr);
> + else
> return xics_rm_h_cppr(vcpu, cppr);
> }
>
> @@ -622,13 +575,9 @@ int kvmppc_rm_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr)
> {
> if (!kvmppc_xics_enabled(vcpu))
> return H_TOO_HARD;
> - if (xics_on_xive()) {
> - if (is_rm())
> - return xive_rm_h_eoi(vcpu, xirr);
> - if (unlikely(!__xive_vm_h_eoi))
> - return H_NOT_AVAILABLE;
> - return __xive_vm_h_eoi(vcpu, xirr);
> - } else
> + if (xics_on_xive())
> + return xive_rm_h_eoi(vcpu, xirr);
> + else
> return xics_rm_h_eoi(vcpu, xirr);
> }
> #endif /* CONFIG_KVM_XICS */
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index b1f3ee16fd84..564ca9feef35 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -2323,7 +2323,7 @@ hcall_real_table:
> #else
> .long 0 /* 0x2fc - H_XIRR_X*/
> #endif
> - .long DOTSYM(kvmppc_h_random) - hcall_real_table
> + .long DOTSYM(kvmppc_rm_h_random) - hcall_real_table
> .globl hcall_real_table_end
> hcall_real_table_end:
>
> diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
> index dcc07ceaf5ca..80d32b4eb898 100644
> --- a/arch/powerpc/kvm/book3s_xive.c
> +++ b/arch/powerpc/kvm/book3s_xive.c
> @@ -2358,21 +2358,3 @@ struct kvm_device_ops kvm_xive_ops = {
> .get_attr = xive_get_attr,
> .has_attr = xive_has_attr,
> };
> -
> -void kvmppc_xive_init_module(void)
> -{
> - __xive_vm_h_xirr = xive_vm_h_xirr;
> - __xive_vm_h_ipoll = xive_vm_h_ipoll;
> - __xive_vm_h_ipi = xive_vm_h_ipi;
> - __xive_vm_h_cppr = xive_vm_h_cppr;
> - __xive_vm_h_eoi = xive_vm_h_eoi;
> -}
> -
> -void kvmppc_xive_exit_module(void)
> -{
> - __xive_vm_h_xirr = NULL;
> - __xive_vm_h_ipoll = NULL;
> - __xive_vm_h_ipi = NULL;
> - __xive_vm_h_cppr = NULL;
> - __xive_vm_h_eoi = NULL;
> -}
> diff --git a/arch/powerpc/kvm/book3s_xive.h b/arch/powerpc/kvm/book3s_xive.h
> index 86c24a4ad809..afe9eeac6d56 100644
> --- a/arch/powerpc/kvm/book3s_xive.h
> +++ b/arch/powerpc/kvm/book3s_xive.h
> @@ -289,13 +289,6 @@ extern int xive_rm_h_ipi(struct kvm_vcpu *vcpu, unsigned long server,
> extern int xive_rm_h_cppr(struct kvm_vcpu *vcpu, unsigned long cppr);
> extern int xive_rm_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr);
>
> -extern unsigned long (*__xive_vm_h_xirr)(struct kvm_vcpu *vcpu);
> -extern unsigned long (*__xive_vm_h_ipoll)(struct kvm_vcpu *vcpu, unsigned long server);
> -extern int (*__xive_vm_h_ipi)(struct kvm_vcpu *vcpu, unsigned long server,
> - unsigned long mfrr);
> -extern int (*__xive_vm_h_cppr)(struct kvm_vcpu *vcpu, unsigned long cppr);
> -extern int (*__xive_vm_h_eoi)(struct kvm_vcpu *vcpu, unsigned long xirr);
> -
> /*
> * Common Xive routines for XICS-over-XIVE and XIVE native
> */
> diff --git a/arch/powerpc/kvm/book3s_xive_native.c b/arch/powerpc/kvm/book3s_xive_native.c
> index 76800c84f2a3..1253666dd4d8 100644
> --- a/arch/powerpc/kvm/book3s_xive_native.c
> +++ b/arch/powerpc/kvm/book3s_xive_native.c
> @@ -1281,13 +1281,3 @@ struct kvm_device_ops kvm_xive_native_ops = {
> .has_attr = kvmppc_xive_native_has_attr,
> .mmap = kvmppc_xive_native_mmap,
> };
> -
> -void kvmppc_xive_native_init_module(void)
> -{
> - ;
> -}
> -
> -void kvmppc_xive_native_exit_module(void)
> -{
> - ;
> -}
>
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 39/46] KVM: PPC: Book3S HV: Remove virt mode checks from real mode handlers
@ 2021-03-23 13:38 ` Cédric Le Goater
0 siblings, 0 replies; 188+ messages in thread
From: Cédric Le Goater @ 2021-03-23 13:38 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev
On 3/23/21 2:02 AM, Nicholas Piggin wrote:
> Now that the P7/8 path no longer supports radix, real-mode handlers
> do not need to deal with being called in virt mode.
>
> This change effectively reverts commit acde25726bc6 ("KVM: PPC: Book3S
> HV: Add radix checks in real-mode hypercall handlers").
>
> It removes a few more real-mode tests in rm hcall handlers, which also
> allows the indirect ops for the xive module to be removed from the
> built-in xics rm handlers.
>
> kvmppc_h_random is renamed to kvmppc_rm_h_random to be a bit more
> descriptive of its function.
>
> Cc: Cédric Le Goater <clg@kaod.org>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
> ---
> arch/powerpc/include/asm/kvm_ppc.h | 10 +--
> arch/powerpc/kvm/book3s.c | 11 +--
> arch/powerpc/kvm/book3s_64_vio_hv.c | 12 ----
> arch/powerpc/kvm/book3s_hv_builtin.c | 91 ++++++-------------------
> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 2 +-
> arch/powerpc/kvm/book3s_xive.c | 18 -----
> arch/powerpc/kvm/book3s_xive.h | 7 --
> arch/powerpc/kvm/book3s_xive_native.c | 10 ---
> 8 files changed, 23 insertions(+), 138 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
> index db6646c2ade2..5dfb3f167f2c 100644
> --- a/arch/powerpc/include/asm/kvm_ppc.h
> +++ b/arch/powerpc/include/asm/kvm_ppc.h
> @@ -659,8 +659,6 @@ extern int kvmppc_xive_get_xive(struct kvm *kvm, u32 irq, u32 *server,
> u32 *priority);
> extern int kvmppc_xive_int_on(struct kvm *kvm, u32 irq);
> extern int kvmppc_xive_int_off(struct kvm *kvm, u32 irq);
> -extern void kvmppc_xive_init_module(void);
> -extern void kvmppc_xive_exit_module(void);
>
> extern int kvmppc_xive_connect_vcpu(struct kvm_device *dev,
> struct kvm_vcpu *vcpu, u32 cpu);
> @@ -686,8 +684,6 @@ static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> extern int kvmppc_xive_native_connect_vcpu(struct kvm_device *dev,
> struct kvm_vcpu *vcpu, u32 cpu);
> extern void kvmppc_xive_native_cleanup_vcpu(struct kvm_vcpu *vcpu);
> -extern void kvmppc_xive_native_init_module(void);
> -extern void kvmppc_xive_native_exit_module(void);
> extern int kvmppc_xive_native_get_vp(struct kvm_vcpu *vcpu,
> union kvmppc_one_reg *val);
> extern int kvmppc_xive_native_set_vp(struct kvm_vcpu *vcpu,
> @@ -701,8 +697,6 @@ static inline int kvmppc_xive_get_xive(struct kvm *kvm, u32 irq, u32 *server,
> u32 *priority) { return -1; }
> static inline int kvmppc_xive_int_on(struct kvm *kvm, u32 irq) { return -1; }
> static inline int kvmppc_xive_int_off(struct kvm *kvm, u32 irq) { return -1; }
> -static inline void kvmppc_xive_init_module(void) { }
> -static inline void kvmppc_xive_exit_module(void) { }
>
> static inline int kvmppc_xive_connect_vcpu(struct kvm_device *dev,
> struct kvm_vcpu *vcpu, u32 cpu) { return -EBUSY; }
> @@ -725,8 +719,6 @@ static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
> static inline int kvmppc_xive_native_connect_vcpu(struct kvm_device *dev,
> struct kvm_vcpu *vcpu, u32 cpu) { return -EBUSY; }
> static inline void kvmppc_xive_native_cleanup_vcpu(struct kvm_vcpu *vcpu) { }
> -static inline void kvmppc_xive_native_init_module(void) { }
> -static inline void kvmppc_xive_native_exit_module(void) { }
> static inline int kvmppc_xive_native_get_vp(struct kvm_vcpu *vcpu,
> union kvmppc_one_reg *val)
> { return 0; }
> @@ -762,7 +754,7 @@ long kvmppc_rm_h_stuff_tce(struct kvm_vcpu *vcpu,
> unsigned long tce_value, unsigned long npages);
> long int kvmppc_rm_h_confer(struct kvm_vcpu *vcpu, int target,
> unsigned int yield_count);
> -long kvmppc_h_random(struct kvm_vcpu *vcpu);
> +long kvmppc_rm_h_random(struct kvm_vcpu *vcpu);
> void kvmhv_commence_exit(int trap);
> void kvmppc_realmode_machine_check(struct kvm_vcpu *vcpu);
> void kvmppc_subcore_enter_guest(void);
> diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c
> index 44bf567b6589..1888aedfd410 100644
> --- a/arch/powerpc/kvm/book3s.c
> +++ b/arch/powerpc/kvm/book3s.c
> @@ -1046,13 +1046,10 @@ static int kvmppc_book3s_init(void)
> #ifdef CONFIG_KVM_XICS
> #ifdef CONFIG_KVM_XIVE
> if (xics_on_xive()) {
> - kvmppc_xive_init_module();
> kvm_register_device_ops(&kvm_xive_ops, KVM_DEV_TYPE_XICS);
> - if (kvmppc_xive_native_supported()) {
> - kvmppc_xive_native_init_module();
> + if (kvmppc_xive_native_supported())
> kvm_register_device_ops(&kvm_xive_native_ops,
> KVM_DEV_TYPE_XIVE);
> - }
> } else
> #endif
> kvm_register_device_ops(&kvm_xics_ops, KVM_DEV_TYPE_XICS);
> @@ -1062,12 +1059,6 @@ static int kvmppc_book3s_init(void)
>
> static void kvmppc_book3s_exit(void)
> {
> -#ifdef CONFIG_KVM_XICS
> - if (xics_on_xive()) {
> - kvmppc_xive_exit_module();
> - kvmppc_xive_native_exit_module();
> - }
> -#endif
> #ifdef CONFIG_KVM_BOOK3S_32_HANDLER
> kvmppc_book3s_exit_pr();
> #endif
> diff --git a/arch/powerpc/kvm/book3s_64_vio_hv.c b/arch/powerpc/kvm/book3s_64_vio_hv.c
> index 083a4e037718..dc6591548f0c 100644
> --- a/arch/powerpc/kvm/book3s_64_vio_hv.c
> +++ b/arch/powerpc/kvm/book3s_64_vio_hv.c
> @@ -391,10 +391,6 @@ long kvmppc_rm_h_put_tce(struct kvm_vcpu *vcpu, unsigned long liobn,
> /* udbg_printf("H_PUT_TCE(): liobn=0x%lx ioba=0x%lx, tce=0x%lx\n", */
> /* liobn, ioba, tce); */
>
> - /* For radix, we might be in virtual mode, so punt */
> - if (kvm_is_radix(vcpu->kvm))
> - return H_TOO_HARD;
> -
> stt = kvmppc_find_table(vcpu->kvm, liobn);
> if (!stt)
> return H_TOO_HARD;
> @@ -489,10 +485,6 @@ long kvmppc_rm_h_put_tce_indirect(struct kvm_vcpu *vcpu,
> bool prereg = false;
> struct kvmppc_spapr_tce_iommu_table *stit;
>
> - /* For radix, we might be in virtual mode, so punt */
> - if (kvm_is_radix(vcpu->kvm))
> - return H_TOO_HARD;
> -
> /*
> * used to check for invalidations in progress
> */
> @@ -602,10 +594,6 @@ long kvmppc_rm_h_stuff_tce(struct kvm_vcpu *vcpu,
> long i, ret;
> struct kvmppc_spapr_tce_iommu_table *stit;
>
> - /* For radix, we might be in virtual mode, so punt */
> - if (kvm_is_radix(vcpu->kvm))
> - return H_TOO_HARD;
> -
> stt = kvmppc_find_table(vcpu->kvm, liobn);
> if (!stt)
> return H_TOO_HARD;
> diff --git a/arch/powerpc/kvm/book3s_hv_builtin.c b/arch/powerpc/kvm/book3s_hv_builtin.c
> index 7a0e33a9c980..8d669a0e15f8 100644
> --- a/arch/powerpc/kvm/book3s_hv_builtin.c
> +++ b/arch/powerpc/kvm/book3s_hv_builtin.c
> @@ -34,21 +34,6 @@
> #include "book3s_xics.h"
> #include "book3s_xive.h"
>
> -/*
> - * The XIVE module will populate these when it loads
> - */
> -unsigned long (*__xive_vm_h_xirr)(struct kvm_vcpu *vcpu);
> -unsigned long (*__xive_vm_h_ipoll)(struct kvm_vcpu *vcpu, unsigned long server);
> -int (*__xive_vm_h_ipi)(struct kvm_vcpu *vcpu, unsigned long server,
> - unsigned long mfrr);
> -int (*__xive_vm_h_cppr)(struct kvm_vcpu *vcpu, unsigned long cppr);
> -int (*__xive_vm_h_eoi)(struct kvm_vcpu *vcpu, unsigned long xirr);
> -EXPORT_SYMBOL_GPL(__xive_vm_h_xirr);
> -EXPORT_SYMBOL_GPL(__xive_vm_h_ipoll);
> -EXPORT_SYMBOL_GPL(__xive_vm_h_ipi);
> -EXPORT_SYMBOL_GPL(__xive_vm_h_cppr);
> -EXPORT_SYMBOL_GPL(__xive_vm_h_eoi);
> -
> /*
> * Hash page table alignment on newer cpus(CPU_FTR_ARCH_206)
> * should be power of 2.
> @@ -196,16 +181,9 @@ int kvmppc_hwrng_present(void)
> }
> EXPORT_SYMBOL_GPL(kvmppc_hwrng_present);
>
> -long kvmppc_h_random(struct kvm_vcpu *vcpu)
> +long kvmppc_rm_h_random(struct kvm_vcpu *vcpu)
> {
> - int r;
> -
> - /* Only need to do the expensive mfmsr() on radix */
> - if (kvm_is_radix(vcpu->kvm) && (mfmsr() & MSR_IR))
> - r = powernv_get_random_long(&vcpu->arch.regs.gpr[4]);
> - else
> - r = powernv_get_random_real_mode(&vcpu->arch.regs.gpr[4]);
> - if (r)
> + if (powernv_get_random_real_mode(&vcpu->arch.regs.gpr[4]))
> return H_SUCCESS;
>
> return H_HARDWARE;
> @@ -541,22 +519,13 @@ static long kvmppc_read_one_intr(bool *again)
> }
>
> #ifdef CONFIG_KVM_XICS
> -static inline bool is_rm(void)
> -{
> - return !(mfmsr() & MSR_DR);
> -}
> -
> unsigned long kvmppc_rm_h_xirr(struct kvm_vcpu *vcpu)
> {
> if (!kvmppc_xics_enabled(vcpu))
> return H_TOO_HARD;
> - if (xics_on_xive()) {
> - if (is_rm())
> - return xive_rm_h_xirr(vcpu);
> - if (unlikely(!__xive_vm_h_xirr))
> - return H_NOT_AVAILABLE;
> - return __xive_vm_h_xirr(vcpu);
> - } else
> + if (xics_on_xive())
> + return xive_rm_h_xirr(vcpu);
> + else
> return xics_rm_h_xirr(vcpu);
> }
>
> @@ -565,13 +534,9 @@ unsigned long kvmppc_rm_h_xirr_x(struct kvm_vcpu *vcpu)
> if (!kvmppc_xics_enabled(vcpu))
> return H_TOO_HARD;
> vcpu->arch.regs.gpr[5] = get_tb();
> - if (xics_on_xive()) {
> - if (is_rm())
> - return xive_rm_h_xirr(vcpu);
> - if (unlikely(!__xive_vm_h_xirr))
> - return H_NOT_AVAILABLE;
> - return __xive_vm_h_xirr(vcpu);
> - } else
> + if (xics_on_xive())
> + return xive_rm_h_xirr(vcpu);
> + else
> return xics_rm_h_xirr(vcpu);
> }
>
> @@ -579,13 +544,9 @@ unsigned long kvmppc_rm_h_ipoll(struct kvm_vcpu *vcpu, unsigned long server)
> {
> if (!kvmppc_xics_enabled(vcpu))
> return H_TOO_HARD;
> - if (xics_on_xive()) {
> - if (is_rm())
> - return xive_rm_h_ipoll(vcpu, server);
> - if (unlikely(!__xive_vm_h_ipoll))
> - return H_NOT_AVAILABLE;
> - return __xive_vm_h_ipoll(vcpu, server);
> - } else
> + if (xics_on_xive())
> + return xive_rm_h_ipoll(vcpu, server);
> + else
> return H_TOO_HARD;
> }
>
> @@ -594,13 +555,9 @@ int kvmppc_rm_h_ipi(struct kvm_vcpu *vcpu, unsigned long server,
> {
> if (!kvmppc_xics_enabled(vcpu))
> return H_TOO_HARD;
> - if (xics_on_xive()) {
> - if (is_rm())
> - return xive_rm_h_ipi(vcpu, server, mfrr);
> - if (unlikely(!__xive_vm_h_ipi))
> - return H_NOT_AVAILABLE;
> - return __xive_vm_h_ipi(vcpu, server, mfrr);
> - } else
> + if (xics_on_xive())
> + return xive_rm_h_ipi(vcpu, server, mfrr);
> + else
> return xics_rm_h_ipi(vcpu, server, mfrr);
> }
>
> @@ -608,13 +565,9 @@ int kvmppc_rm_h_cppr(struct kvm_vcpu *vcpu, unsigned long cppr)
> {
> if (!kvmppc_xics_enabled(vcpu))
> return H_TOO_HARD;
> - if (xics_on_xive()) {
> - if (is_rm())
> - return xive_rm_h_cppr(vcpu, cppr);
> - if (unlikely(!__xive_vm_h_cppr))
> - return H_NOT_AVAILABLE;
> - return __xive_vm_h_cppr(vcpu, cppr);
> - } else
> + if (xics_on_xive())
> + return xive_rm_h_cppr(vcpu, cppr);
> + else
> return xics_rm_h_cppr(vcpu, cppr);
> }
>
> @@ -622,13 +575,9 @@ int kvmppc_rm_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr)
> {
> if (!kvmppc_xics_enabled(vcpu))
> return H_TOO_HARD;
> - if (xics_on_xive()) {
> - if (is_rm())
> - return xive_rm_h_eoi(vcpu, xirr);
> - if (unlikely(!__xive_vm_h_eoi))
> - return H_NOT_AVAILABLE;
> - return __xive_vm_h_eoi(vcpu, xirr);
> - } else
> + if (xics_on_xive())
> + return xive_rm_h_eoi(vcpu, xirr);
> + else
> return xics_rm_h_eoi(vcpu, xirr);
> }
> #endif /* CONFIG_KVM_XICS */
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index b1f3ee16fd84..564ca9feef35 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -2323,7 +2323,7 @@ hcall_real_table:
> #else
> .long 0 /* 0x2fc - H_XIRR_X*/
> #endif
> - .long DOTSYM(kvmppc_h_random) - hcall_real_table
> + .long DOTSYM(kvmppc_rm_h_random) - hcall_real_table
> .globl hcall_real_table_end
> hcall_real_table_end:
>
> diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
> index dcc07ceaf5ca..80d32b4eb898 100644
> --- a/arch/powerpc/kvm/book3s_xive.c
> +++ b/arch/powerpc/kvm/book3s_xive.c
> @@ -2358,21 +2358,3 @@ struct kvm_device_ops kvm_xive_ops = {
> .get_attr = xive_get_attr,
> .has_attr = xive_has_attr,
> };
> -
> -void kvmppc_xive_init_module(void)
> -{
> - __xive_vm_h_xirr = xive_vm_h_xirr;
> - __xive_vm_h_ipoll = xive_vm_h_ipoll;
> - __xive_vm_h_ipi = xive_vm_h_ipi;
> - __xive_vm_h_cppr = xive_vm_h_cppr;
> - __xive_vm_h_eoi = xive_vm_h_eoi;
> -}
> -
> -void kvmppc_xive_exit_module(void)
> -{
> - __xive_vm_h_xirr = NULL;
> - __xive_vm_h_ipoll = NULL;
> - __xive_vm_h_ipi = NULL;
> - __xive_vm_h_cppr = NULL;
> - __xive_vm_h_eoi = NULL;
> -}
> diff --git a/arch/powerpc/kvm/book3s_xive.h b/arch/powerpc/kvm/book3s_xive.h
> index 86c24a4ad809..afe9eeac6d56 100644
> --- a/arch/powerpc/kvm/book3s_xive.h
> +++ b/arch/powerpc/kvm/book3s_xive.h
> @@ -289,13 +289,6 @@ extern int xive_rm_h_ipi(struct kvm_vcpu *vcpu, unsigned long server,
> extern int xive_rm_h_cppr(struct kvm_vcpu *vcpu, unsigned long cppr);
> extern int xive_rm_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr);
>
> -extern unsigned long (*__xive_vm_h_xirr)(struct kvm_vcpu *vcpu);
> -extern unsigned long (*__xive_vm_h_ipoll)(struct kvm_vcpu *vcpu, unsigned long server);
> -extern int (*__xive_vm_h_ipi)(struct kvm_vcpu *vcpu, unsigned long server,
> - unsigned long mfrr);
> -extern int (*__xive_vm_h_cppr)(struct kvm_vcpu *vcpu, unsigned long cppr);
> -extern int (*__xive_vm_h_eoi)(struct kvm_vcpu *vcpu, unsigned long xirr);
> -
> /*
> * Common Xive routines for XICS-over-XIVE and XIVE native
> */
> diff --git a/arch/powerpc/kvm/book3s_xive_native.c b/arch/powerpc/kvm/book3s_xive_native.c
> index 76800c84f2a3..1253666dd4d8 100644
> --- a/arch/powerpc/kvm/book3s_xive_native.c
> +++ b/arch/powerpc/kvm/book3s_xive_native.c
> @@ -1281,13 +1281,3 @@ struct kvm_device_ops kvm_xive_native_ops = {
> .has_attr = kvmppc_xive_native_has_attr,
> .mmap = kvmppc_xive_native_mmap,
> };
> -
> -void kvmppc_xive_native_init_module(void)
> -{
> - ;
> -}
> -
> -void kvmppc_xive_native_exit_module(void)
> -{
> - ;
> -}
>
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 40/46] KVM: PPC: Book3S HV: Remove unused nested HV tests in XICS emulation
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:02 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Cédric Le Goater, Nicholas Piggin
Commit f3c18e9342a44 ("KVM: PPC: Book3S HV: Use XICS hypercalls when
running as a nested hypervisor") added nested HV tests in XICS
hypercalls, but not all are required.
* icp_eoi is only called by kvmppc_deliver_irq_passthru which is only
called by kvmppc_check_passthru which is only caled by
kvmppc_read_one_intr.
* kvmppc_read_one_intr is only called by kvmppc_read_intr which is only
called by the L0 HV rmhandlers code.
* kvmhv_rm_send_ipi is called by:
- kvmhv_interrupt_vcore which is only called by kvmhv_commence_exit
which is only called by the L0 HV rmhandlers code.
- icp_send_hcore_msg which is only called by icp_rm_set_vcpu_irq.
- icp_rm_set_vcpu_irq which is only called by icp_rm_try_update
- icp_rm_set_vcpu_irq is not nested HV safe because it writes to
LPCR directly without a kvmhv_on_pseries test. Nested handlers
should not in general be using the rm handlers.
The important test seems to be in kvmppc_ipi_thread, which sends the
virt-mode H_IPI handler kick to use smp_call_function rather than
msgsnd.
Cc: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_builtin.c | 44 +++++-----------------------
arch/powerpc/kvm/book3s_hv_rm_xics.c | 15 ----------
2 files changed, 8 insertions(+), 51 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_builtin.c b/arch/powerpc/kvm/book3s_hv_builtin.c
index 8d669a0e15f8..259492bb4153 100644
--- a/arch/powerpc/kvm/book3s_hv_builtin.c
+++ b/arch/powerpc/kvm/book3s_hv_builtin.c
@@ -199,15 +199,6 @@ void kvmhv_rm_send_ipi(int cpu)
void __iomem *xics_phys;
unsigned long msg = PPC_DBELL_TYPE(PPC_DBELL_SERVER);
- /* For a nested hypervisor, use the XICS via hcall */
- if (kvmhv_on_pseries()) {
- unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
-
- plpar_hcall_raw(H_IPI, retbuf, get_hard_smp_processor_id(cpu),
- IPI_PRIORITY);
- return;
- }
-
/* On POWER9 we can use msgsnd for any destination cpu. */
if (cpu_has_feature(CPU_FTR_ARCH_300)) {
msg |= get_hard_smp_processor_id(cpu);
@@ -420,19 +411,12 @@ static long kvmppc_read_one_intr(bool *again)
return 1;
/* Now read the interrupt from the ICP */
- if (kvmhv_on_pseries()) {
- unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
-
- rc = plpar_hcall_raw(H_XIRR, retbuf, 0xFF);
- xirr = cpu_to_be32(retbuf[0]);
- } else {
- xics_phys = local_paca->kvm_hstate.xics_phys;
- rc = 0;
- if (!xics_phys)
- rc = opal_int_get_xirr(&xirr, false);
- else
- xirr = __raw_rm_readl(xics_phys + XICS_XIRR);
- }
+ xics_phys = local_paca->kvm_hstate.xics_phys;
+ rc = 0;
+ if (!xics_phys)
+ rc = opal_int_get_xirr(&xirr, false);
+ else
+ xirr = __raw_rm_readl(xics_phys + XICS_XIRR);
if (rc < 0)
return 1;
@@ -461,13 +445,7 @@ static long kvmppc_read_one_intr(bool *again)
*/
if (xisr == XICS_IPI) {
rc = 0;
- if (kvmhv_on_pseries()) {
- unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
-
- plpar_hcall_raw(H_IPI, retbuf,
- hard_smp_processor_id(), 0xff);
- plpar_hcall_raw(H_EOI, retbuf, h_xirr);
- } else if (xics_phys) {
+ if (xics_phys) {
__raw_rm_writeb(0xff, xics_phys + XICS_MFRR);
__raw_rm_writel(xirr, xics_phys + XICS_XIRR);
} else {
@@ -493,13 +471,7 @@ static long kvmppc_read_one_intr(bool *again)
/* We raced with the host,
* we need to resend that IPI, bummer
*/
- if (kvmhv_on_pseries()) {
- unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
-
- plpar_hcall_raw(H_IPI, retbuf,
- hard_smp_processor_id(),
- IPI_PRIORITY);
- } else if (xics_phys)
+ if (xics_phys)
__raw_rm_writeb(IPI_PRIORITY,
xics_phys + XICS_MFRR);
else
diff --git a/arch/powerpc/kvm/book3s_hv_rm_xics.c b/arch/powerpc/kvm/book3s_hv_rm_xics.c
index c2c9c733f359..0a11ec88a0ae 100644
--- a/arch/powerpc/kvm/book3s_hv_rm_xics.c
+++ b/arch/powerpc/kvm/book3s_hv_rm_xics.c
@@ -141,13 +141,6 @@ static void icp_rm_set_vcpu_irq(struct kvm_vcpu *vcpu,
return;
}
- if (xive_enabled() && kvmhv_on_pseries()) {
- /* No XICS access or hypercalls available, too hard */
- this_icp->rm_action |= XICS_RM_KICK_VCPU;
- this_icp->rm_kick_target = vcpu;
- return;
- }
-
/*
* Check if the core is loaded,
* if not, find an available host core to post to wake the VCPU,
@@ -771,14 +764,6 @@ static void icp_eoi(struct irq_chip *c, u32 hwirq, __be32 xirr, bool *again)
void __iomem *xics_phys;
int64_t rc;
- if (kvmhv_on_pseries()) {
- unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
-
- iosync();
- plpar_hcall_raw(H_EOI, retbuf, hwirq);
- return;
- }
-
rc = pnv_opal_pci_msi_eoi(c, hwirq);
if (rc)
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 40/46] KVM: PPC: Book3S HV: Remove unused nested HV tests in XICS emulation
@ 2021-03-23 1:02 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:02 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Cédric Le Goater, Nicholas Piggin
Commit f3c18e9342a44 ("KVM: PPC: Book3S HV: Use XICS hypercalls when
running as a nested hypervisor") added nested HV tests in XICS
hypercalls, but not all are required.
* icp_eoi is only called by kvmppc_deliver_irq_passthru which is only
called by kvmppc_check_passthru which is only caled by
kvmppc_read_one_intr.
* kvmppc_read_one_intr is only called by kvmppc_read_intr which is only
called by the L0 HV rmhandlers code.
* kvmhv_rm_send_ipi is called by:
- kvmhv_interrupt_vcore which is only called by kvmhv_commence_exit
which is only called by the L0 HV rmhandlers code.
- icp_send_hcore_msg which is only called by icp_rm_set_vcpu_irq.
- icp_rm_set_vcpu_irq which is only called by icp_rm_try_update
- icp_rm_set_vcpu_irq is not nested HV safe because it writes to
LPCR directly without a kvmhv_on_pseries test. Nested handlers
should not in general be using the rm handlers.
The important test seems to be in kvmppc_ipi_thread, which sends the
virt-mode H_IPI handler kick to use smp_call_function rather than
msgsnd.
Cc: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv_builtin.c | 44 +++++-----------------------
arch/powerpc/kvm/book3s_hv_rm_xics.c | 15 ----------
2 files changed, 8 insertions(+), 51 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_builtin.c b/arch/powerpc/kvm/book3s_hv_builtin.c
index 8d669a0e15f8..259492bb4153 100644
--- a/arch/powerpc/kvm/book3s_hv_builtin.c
+++ b/arch/powerpc/kvm/book3s_hv_builtin.c
@@ -199,15 +199,6 @@ void kvmhv_rm_send_ipi(int cpu)
void __iomem *xics_phys;
unsigned long msg = PPC_DBELL_TYPE(PPC_DBELL_SERVER);
- /* For a nested hypervisor, use the XICS via hcall */
- if (kvmhv_on_pseries()) {
- unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
-
- plpar_hcall_raw(H_IPI, retbuf, get_hard_smp_processor_id(cpu),
- IPI_PRIORITY);
- return;
- }
-
/* On POWER9 we can use msgsnd for any destination cpu. */
if (cpu_has_feature(CPU_FTR_ARCH_300)) {
msg |= get_hard_smp_processor_id(cpu);
@@ -420,19 +411,12 @@ static long kvmppc_read_one_intr(bool *again)
return 1;
/* Now read the interrupt from the ICP */
- if (kvmhv_on_pseries()) {
- unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
-
- rc = plpar_hcall_raw(H_XIRR, retbuf, 0xFF);
- xirr = cpu_to_be32(retbuf[0]);
- } else {
- xics_phys = local_paca->kvm_hstate.xics_phys;
- rc = 0;
- if (!xics_phys)
- rc = opal_int_get_xirr(&xirr, false);
- else
- xirr = __raw_rm_readl(xics_phys + XICS_XIRR);
- }
+ xics_phys = local_paca->kvm_hstate.xics_phys;
+ rc = 0;
+ if (!xics_phys)
+ rc = opal_int_get_xirr(&xirr, false);
+ else
+ xirr = __raw_rm_readl(xics_phys + XICS_XIRR);
if (rc < 0)
return 1;
@@ -461,13 +445,7 @@ static long kvmppc_read_one_intr(bool *again)
*/
if (xisr = XICS_IPI) {
rc = 0;
- if (kvmhv_on_pseries()) {
- unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
-
- plpar_hcall_raw(H_IPI, retbuf,
- hard_smp_processor_id(), 0xff);
- plpar_hcall_raw(H_EOI, retbuf, h_xirr);
- } else if (xics_phys) {
+ if (xics_phys) {
__raw_rm_writeb(0xff, xics_phys + XICS_MFRR);
__raw_rm_writel(xirr, xics_phys + XICS_XIRR);
} else {
@@ -493,13 +471,7 @@ static long kvmppc_read_one_intr(bool *again)
/* We raced with the host,
* we need to resend that IPI, bummer
*/
- if (kvmhv_on_pseries()) {
- unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
-
- plpar_hcall_raw(H_IPI, retbuf,
- hard_smp_processor_id(),
- IPI_PRIORITY);
- } else if (xics_phys)
+ if (xics_phys)
__raw_rm_writeb(IPI_PRIORITY,
xics_phys + XICS_MFRR);
else
diff --git a/arch/powerpc/kvm/book3s_hv_rm_xics.c b/arch/powerpc/kvm/book3s_hv_rm_xics.c
index c2c9c733f359..0a11ec88a0ae 100644
--- a/arch/powerpc/kvm/book3s_hv_rm_xics.c
+++ b/arch/powerpc/kvm/book3s_hv_rm_xics.c
@@ -141,13 +141,6 @@ static void icp_rm_set_vcpu_irq(struct kvm_vcpu *vcpu,
return;
}
- if (xive_enabled() && kvmhv_on_pseries()) {
- /* No XICS access or hypercalls available, too hard */
- this_icp->rm_action |= XICS_RM_KICK_VCPU;
- this_icp->rm_kick_target = vcpu;
- return;
- }
-
/*
* Check if the core is loaded,
* if not, find an available host core to post to wake the VCPU,
@@ -771,14 +764,6 @@ static void icp_eoi(struct irq_chip *c, u32 hwirq, __be32 xirr, bool *again)
void __iomem *xics_phys;
int64_t rc;
- if (kvmhv_on_pseries()) {
- unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
-
- iosync();
- plpar_hcall_raw(H_EOI, retbuf, hwirq);
- return;
- }
-
rc = pnv_opal_pci_msi_eoi(c, hwirq);
if (rc)
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 41/46] KVM: PPC: Book3S HV P9: Allow all P9 processors to enable nested HV
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:03 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:03 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
All radix guests go via the P9 path now, so there is no need to limit
nested HV to processors that support "mixed mode" MMU. Remove the
restriction.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 85a14ce0ea0e..0f3c1792ab86 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -5420,7 +5420,7 @@ static int kvmhv_enable_nested(struct kvm *kvm)
{
if (!nested)
return -EPERM;
- if (!cpu_has_feature(CPU_FTR_ARCH_300) || no_mixing_hpt_and_radix)
+ if (!cpu_has_feature(CPU_FTR_ARCH_300))
return -ENODEV;
/* kvm == NULL means the caller is testing if the capability exists */
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 41/46] KVM: PPC: Book3S HV P9: Allow all P9 processors to enable nested HV
@ 2021-03-23 1:03 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:03 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
All radix guests go via the P9 path now, so there is no need to limit
nested HV to processors that support "mixed mode" MMU. Remove the
restriction.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 85a14ce0ea0e..0f3c1792ab86 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -5420,7 +5420,7 @@ static int kvmhv_enable_nested(struct kvm *kvm)
{
if (!nested)
return -EPERM;
- if (!cpu_has_feature(CPU_FTR_ARCH_300) || no_mixing_hpt_and_radix)
+ if (!cpu_has_feature(CPU_FTR_ARCH_300))
return -ENODEV;
/* kvm = NULL means the caller is testing if the capability exists */
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 42/46] KVM: PPC: Book3S HV: small pseries_do_hcall cleanup
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:03 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:03 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Functionality should not be changed.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 29 +++++++++++++++--------------
1 file changed, 15 insertions(+), 14 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 0f3c1792ab86..e6d113dbc076 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -925,6 +925,7 @@ static int kvmppc_get_yield_count(struct kvm_vcpu *vcpu)
int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
{
+ struct kvm *kvm = vcpu->kvm;
unsigned long req = kvmppc_get_gpr(vcpu, 3);
unsigned long target, ret = H_SUCCESS;
int yield_count;
@@ -940,7 +941,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
break;
case H_PROD:
target = kvmppc_get_gpr(vcpu, 4);
- tvcpu = kvmppc_find_vcpu(vcpu->kvm, target);
+ tvcpu = kvmppc_find_vcpu(kvm, target);
if (!tvcpu) {
ret = H_PARAMETER;
break;
@@ -954,7 +955,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
target = kvmppc_get_gpr(vcpu, 4);
if (target == -1)
break;
- tvcpu = kvmppc_find_vcpu(vcpu->kvm, target);
+ tvcpu = kvmppc_find_vcpu(kvm, target);
if (!tvcpu) {
ret = H_PARAMETER;
break;
@@ -970,12 +971,12 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
kvmppc_get_gpr(vcpu, 6));
break;
case H_RTAS:
- if (list_empty(&vcpu->kvm->arch.rtas_tokens))
+ if (list_empty(&kvm->arch.rtas_tokens))
return RESUME_HOST;
- idx = srcu_read_lock(&vcpu->kvm->srcu);
+ idx = srcu_read_lock(&kvm->srcu);
rc = kvmppc_rtas_hcall(vcpu);
- srcu_read_unlock(&vcpu->kvm->srcu, idx);
+ srcu_read_unlock(&kvm->srcu, idx);
if (rc == -ENOENT)
return RESUME_HOST;
@@ -1062,12 +1063,12 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
case H_SET_PARTITION_TABLE:
ret = H_FUNCTION;
- if (nesting_enabled(vcpu->kvm))
+ if (nesting_enabled(kvm))
ret = kvmhv_set_partition_table(vcpu);
break;
case H_ENTER_NESTED:
ret = H_FUNCTION;
- if (!nesting_enabled(vcpu->kvm))
+ if (!nesting_enabled(kvm))
break;
ret = kvmhv_enter_nested_guest(vcpu);
if (ret == H_INTERRUPT) {
@@ -1082,12 +1083,12 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
break;
case H_TLB_INVALIDATE:
ret = H_FUNCTION;
- if (nesting_enabled(vcpu->kvm))
+ if (nesting_enabled(kvm))
ret = kvmhv_do_nested_tlbie(vcpu);
break;
case H_COPY_TOFROM_GUEST:
ret = H_FUNCTION;
- if (nesting_enabled(vcpu->kvm))
+ if (nesting_enabled(kvm))
ret = kvmhv_copy_tofrom_guest_nested(vcpu);
break;
case H_PAGE_INIT:
@@ -1098,7 +1099,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
case H_SVM_PAGE_IN:
ret = H_UNSUPPORTED;
if (kvmppc_get_srr1(vcpu) & MSR_S)
- ret = kvmppc_h_svm_page_in(vcpu->kvm,
+ ret = kvmppc_h_svm_page_in(kvm,
kvmppc_get_gpr(vcpu, 4),
kvmppc_get_gpr(vcpu, 5),
kvmppc_get_gpr(vcpu, 6));
@@ -1106,7 +1107,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
case H_SVM_PAGE_OUT:
ret = H_UNSUPPORTED;
if (kvmppc_get_srr1(vcpu) & MSR_S)
- ret = kvmppc_h_svm_page_out(vcpu->kvm,
+ ret = kvmppc_h_svm_page_out(kvm,
kvmppc_get_gpr(vcpu, 4),
kvmppc_get_gpr(vcpu, 5),
kvmppc_get_gpr(vcpu, 6));
@@ -1114,12 +1115,12 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
case H_SVM_INIT_START:
ret = H_UNSUPPORTED;
if (kvmppc_get_srr1(vcpu) & MSR_S)
- ret = kvmppc_h_svm_init_start(vcpu->kvm);
+ ret = kvmppc_h_svm_init_start(kvm);
break;
case H_SVM_INIT_DONE:
ret = H_UNSUPPORTED;
if (kvmppc_get_srr1(vcpu) & MSR_S)
- ret = kvmppc_h_svm_init_done(vcpu->kvm);
+ ret = kvmppc_h_svm_init_done(kvm);
break;
case H_SVM_INIT_ABORT:
/*
@@ -1129,7 +1130,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
* Instead the kvm->arch.secure_guest flag is checked inside
* kvmppc_h_svm_init_abort().
*/
- ret = kvmppc_h_svm_init_abort(vcpu->kvm);
+ ret = kvmppc_h_svm_init_abort(kvm);
break;
default:
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 42/46] KVM: PPC: Book3S HV: small pseries_do_hcall cleanup
@ 2021-03-23 1:03 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:03 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Functionality should not be changed.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 29 +++++++++++++++--------------
1 file changed, 15 insertions(+), 14 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 0f3c1792ab86..e6d113dbc076 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -925,6 +925,7 @@ static int kvmppc_get_yield_count(struct kvm_vcpu *vcpu)
int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
{
+ struct kvm *kvm = vcpu->kvm;
unsigned long req = kvmppc_get_gpr(vcpu, 3);
unsigned long target, ret = H_SUCCESS;
int yield_count;
@@ -940,7 +941,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
break;
case H_PROD:
target = kvmppc_get_gpr(vcpu, 4);
- tvcpu = kvmppc_find_vcpu(vcpu->kvm, target);
+ tvcpu = kvmppc_find_vcpu(kvm, target);
if (!tvcpu) {
ret = H_PARAMETER;
break;
@@ -954,7 +955,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
target = kvmppc_get_gpr(vcpu, 4);
if (target = -1)
break;
- tvcpu = kvmppc_find_vcpu(vcpu->kvm, target);
+ tvcpu = kvmppc_find_vcpu(kvm, target);
if (!tvcpu) {
ret = H_PARAMETER;
break;
@@ -970,12 +971,12 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
kvmppc_get_gpr(vcpu, 6));
break;
case H_RTAS:
- if (list_empty(&vcpu->kvm->arch.rtas_tokens))
+ if (list_empty(&kvm->arch.rtas_tokens))
return RESUME_HOST;
- idx = srcu_read_lock(&vcpu->kvm->srcu);
+ idx = srcu_read_lock(&kvm->srcu);
rc = kvmppc_rtas_hcall(vcpu);
- srcu_read_unlock(&vcpu->kvm->srcu, idx);
+ srcu_read_unlock(&kvm->srcu, idx);
if (rc = -ENOENT)
return RESUME_HOST;
@@ -1062,12 +1063,12 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
case H_SET_PARTITION_TABLE:
ret = H_FUNCTION;
- if (nesting_enabled(vcpu->kvm))
+ if (nesting_enabled(kvm))
ret = kvmhv_set_partition_table(vcpu);
break;
case H_ENTER_NESTED:
ret = H_FUNCTION;
- if (!nesting_enabled(vcpu->kvm))
+ if (!nesting_enabled(kvm))
break;
ret = kvmhv_enter_nested_guest(vcpu);
if (ret = H_INTERRUPT) {
@@ -1082,12 +1083,12 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
break;
case H_TLB_INVALIDATE:
ret = H_FUNCTION;
- if (nesting_enabled(vcpu->kvm))
+ if (nesting_enabled(kvm))
ret = kvmhv_do_nested_tlbie(vcpu);
break;
case H_COPY_TOFROM_GUEST:
ret = H_FUNCTION;
- if (nesting_enabled(vcpu->kvm))
+ if (nesting_enabled(kvm))
ret = kvmhv_copy_tofrom_guest_nested(vcpu);
break;
case H_PAGE_INIT:
@@ -1098,7 +1099,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
case H_SVM_PAGE_IN:
ret = H_UNSUPPORTED;
if (kvmppc_get_srr1(vcpu) & MSR_S)
- ret = kvmppc_h_svm_page_in(vcpu->kvm,
+ ret = kvmppc_h_svm_page_in(kvm,
kvmppc_get_gpr(vcpu, 4),
kvmppc_get_gpr(vcpu, 5),
kvmppc_get_gpr(vcpu, 6));
@@ -1106,7 +1107,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
case H_SVM_PAGE_OUT:
ret = H_UNSUPPORTED;
if (kvmppc_get_srr1(vcpu) & MSR_S)
- ret = kvmppc_h_svm_page_out(vcpu->kvm,
+ ret = kvmppc_h_svm_page_out(kvm,
kvmppc_get_gpr(vcpu, 4),
kvmppc_get_gpr(vcpu, 5),
kvmppc_get_gpr(vcpu, 6));
@@ -1114,12 +1115,12 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
case H_SVM_INIT_START:
ret = H_UNSUPPORTED;
if (kvmppc_get_srr1(vcpu) & MSR_S)
- ret = kvmppc_h_svm_init_start(vcpu->kvm);
+ ret = kvmppc_h_svm_init_start(kvm);
break;
case H_SVM_INIT_DONE:
ret = H_UNSUPPORTED;
if (kvmppc_get_srr1(vcpu) & MSR_S)
- ret = kvmppc_h_svm_init_done(vcpu->kvm);
+ ret = kvmppc_h_svm_init_done(kvm);
break;
case H_SVM_INIT_ABORT:
/*
@@ -1129,7 +1130,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
* Instead the kvm->arch.secure_guest flag is checked inside
* kvmppc_h_svm_init_abort().
*/
- ret = kvmppc_h_svm_init_abort(vcpu->kvm);
+ ret = kvmppc_h_svm_init_abort(kvm);
break;
default:
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 43/46] KVM: PPC: Book3S HV: add virtual mode handlers for HPT hcalls and page faults
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:03 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:03 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
In order to support hash guests in the P9 path (which does not do real
mode hcalls or page fault handling), these real-mode hash specific
interrupts need to be implemented in virt mode.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 126 ++++++++++++++++++++++++++--
arch/powerpc/kvm/book3s_hv_rm_mmu.c | 8 ++
2 files changed, 127 insertions(+), 7 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index e6d113dbc076..5b3b6842d6b9 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -937,6 +937,52 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
return RESUME_HOST;
switch (req) {
+ case H_REMOVE:
+ ret = kvmppc_h_remove(vcpu, kvmppc_get_gpr(vcpu, 4),
+ kvmppc_get_gpr(vcpu, 5),
+ kvmppc_get_gpr(vcpu, 6));
+ if (ret == H_TOO_HARD)
+ return RESUME_HOST;
+ break;
+ case H_ENTER:
+ ret = kvmppc_h_enter(vcpu, kvmppc_get_gpr(vcpu, 4),
+ kvmppc_get_gpr(vcpu, 5),
+ kvmppc_get_gpr(vcpu, 6),
+ kvmppc_get_gpr(vcpu, 7));
+ if (ret == H_TOO_HARD)
+ return RESUME_HOST;
+ break;
+ case H_READ:
+ ret = kvmppc_h_read(vcpu, kvmppc_get_gpr(vcpu, 4),
+ kvmppc_get_gpr(vcpu, 5));
+ if (ret == H_TOO_HARD)
+ return RESUME_HOST;
+ break;
+ case H_CLEAR_MOD:
+ ret = kvmppc_h_clear_mod(vcpu, kvmppc_get_gpr(vcpu, 4),
+ kvmppc_get_gpr(vcpu, 5));
+ if (ret == H_TOO_HARD)
+ return RESUME_HOST;
+ break;
+ case H_CLEAR_REF:
+ ret = kvmppc_h_clear_ref(vcpu, kvmppc_get_gpr(vcpu, 4),
+ kvmppc_get_gpr(vcpu, 5));
+ if (ret == H_TOO_HARD)
+ return RESUME_HOST;
+ break;
+ case H_PROTECT:
+ ret = kvmppc_h_protect(vcpu, kvmppc_get_gpr(vcpu, 4),
+ kvmppc_get_gpr(vcpu, 5),
+ kvmppc_get_gpr(vcpu, 6));
+ if (ret == H_TOO_HARD)
+ return RESUME_HOST;
+ break;
+ case H_BULK_REMOVE:
+ ret = kvmppc_h_bulk_remove(vcpu);
+ if (ret == H_TOO_HARD)
+ return RESUME_HOST;
+ break;
+
case H_CEDE:
break;
case H_PROD:
@@ -1136,6 +1182,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
default:
return RESUME_HOST;
}
+ WARN_ON_ONCE(ret == H_TOO_HARD);
kvmppc_set_gpr(vcpu, 3, ret);
vcpu->arch.hcall_needed = 0;
return RESUME_GUEST;
@@ -1429,19 +1476,84 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
* host page has been paged out. Any other HDSI/HISI interrupts
* have been handled already.
*/
- case BOOK3S_INTERRUPT_H_DATA_STORAGE:
- r = RESUME_PAGE_FAULT;
- if (vcpu->arch.fault_dsisr == HDSISR_CANARY)
+ case BOOK3S_INTERRUPT_H_DATA_STORAGE: {
+ unsigned long vsid;
+ long err;
+
+ if (vcpu->arch.fault_dsisr == HDSISR_CANARY) {
r = RESUME_GUEST; /* Just retry if it's the canary */
+ break;
+ }
+
+ if (kvm_is_radix(vcpu->kvm)) {
+ r = RESUME_PAGE_FAULT;
+ break;
+ }
+
+ if (!(vcpu->arch.fault_dsisr & (DSISR_NOHPTE | DSISR_PROTFAULT))) {
+ kvmppc_core_queue_data_storage(vcpu,
+ vcpu->arch.fault_dar, vcpu->arch.fault_dsisr);
+ r = RESUME_GUEST;
+ break;
+ }
+
+ if (!(vcpu->arch.shregs.msr & MSR_DR))
+ vsid = vcpu->kvm->arch.vrma_slb_v;
+ else
+ vsid = vcpu->arch.fault_gpa;
+
+ err = kvmppc_hpte_hv_fault(vcpu, vcpu->arch.fault_dar,
+ vsid, vcpu->arch.fault_dsisr, true);
+ if (err == 0) {
+ r = RESUME_GUEST;
+ } else if (err == -1 || err == -2) {
+ r = RESUME_PAGE_FAULT;
+ } else {
+ kvmppc_core_queue_data_storage(vcpu,
+ vcpu->arch.fault_dar, err);
+ r = RESUME_GUEST;
+ }
break;
- case BOOK3S_INTERRUPT_H_INST_STORAGE:
+ }
+ case BOOK3S_INTERRUPT_H_INST_STORAGE: {
+ unsigned long vsid;
+ long err;
+
vcpu->arch.fault_dar = kvmppc_get_pc(vcpu);
vcpu->arch.fault_dsisr = vcpu->arch.shregs.msr &
DSISR_SRR1_MATCH_64S;
- if (vcpu->arch.shregs.msr & HSRR1_HISI_WRITE)
- vcpu->arch.fault_dsisr |= DSISR_ISSTORE;
- r = RESUME_PAGE_FAULT;
+ if (kvm_is_radix(vcpu->kvm)) {
+ if (vcpu->arch.shregs.msr & HSRR1_HISI_WRITE)
+ vcpu->arch.fault_dsisr |= DSISR_ISSTORE;
+ r = RESUME_PAGE_FAULT;
+ break;
+ }
+
+ if (!(vcpu->arch.fault_dsisr & SRR1_ISI_NOPT)) {
+ kvmppc_core_queue_inst_storage(vcpu,
+ vcpu->arch.fault_dsisr);
+ r = RESUME_GUEST;
+ break;
+ }
+
+ if (!(vcpu->arch.shregs.msr & MSR_DR))
+ vsid = vcpu->kvm->arch.vrma_slb_v;
+ else
+ vsid = vcpu->arch.fault_gpa;
+
+ err = kvmppc_hpte_hv_fault(vcpu, vcpu->arch.fault_dar,
+ vsid, vcpu->arch.fault_dsisr, false);
+ if (err == 0) {
+ r = RESUME_GUEST;
+ } else if (err == -1) {
+ r = RESUME_PAGE_FAULT;
+ } else {
+ kvmppc_core_queue_inst_storage(vcpu, err);
+ r = RESUME_GUEST;
+ }
break;
+ }
+
/*
* This occurs if the guest executes an illegal instruction.
* If the guest debug is disabled, generate a program interrupt
diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
index 7af7c70f1468..8cc73abbf42b 100644
--- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c
+++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
@@ -409,6 +409,7 @@ long kvmppc_h_enter(struct kvm_vcpu *vcpu, unsigned long flags,
vcpu->arch.pgdir, true,
&vcpu->arch.regs.gpr[4]);
}
+EXPORT_SYMBOL_GPL(kvmppc_h_enter);
#ifdef __BIG_ENDIAN__
#define LOCK_TOKEN (*(u32 *)(&get_paca()->lock_token))
@@ -553,6 +554,7 @@ long kvmppc_h_remove(struct kvm_vcpu *vcpu, unsigned long flags,
return kvmppc_do_h_remove(vcpu->kvm, flags, pte_index, avpn,
&vcpu->arch.regs.gpr[4]);
}
+EXPORT_SYMBOL_GPL(kvmppc_h_remove);
long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
{
@@ -671,6 +673,7 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
return ret;
}
+EXPORT_SYMBOL_GPL(kvmppc_h_bulk_remove);
long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long pte_index, unsigned long avpn)
@@ -741,6 +744,7 @@ long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags,
return H_SUCCESS;
}
+EXPORT_SYMBOL_GPL(kvmppc_h_protect);
long kvmppc_h_read(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long pte_index)
@@ -781,6 +785,7 @@ long kvmppc_h_read(struct kvm_vcpu *vcpu, unsigned long flags,
}
return H_SUCCESS;
}
+EXPORT_SYMBOL_GPL(kvmppc_h_read);
long kvmppc_h_clear_ref(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long pte_index)
@@ -829,6 +834,7 @@ long kvmppc_h_clear_ref(struct kvm_vcpu *vcpu, unsigned long flags,
unlock_hpte(hpte, v & ~HPTE_V_HVLOCK);
return ret;
}
+EXPORT_SYMBOL_GPL(kvmppc_h_clear_ref);
long kvmppc_h_clear_mod(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long pte_index)
@@ -876,6 +882,7 @@ long kvmppc_h_clear_mod(struct kvm_vcpu *vcpu, unsigned long flags,
unlock_hpte(hpte, v & ~HPTE_V_HVLOCK);
return ret;
}
+EXPORT_SYMBOL_GPL(kvmppc_h_clear_mod);
static int kvmppc_get_hpa(struct kvm_vcpu *vcpu, unsigned long mmu_seq,
unsigned long gpa, int writing, unsigned long *hpa,
@@ -1294,3 +1301,4 @@ long kvmppc_hpte_hv_fault(struct kvm_vcpu *vcpu, unsigned long addr,
return -1; /* send fault up to host kernel mode */
}
+EXPORT_SYMBOL_GPL(kvmppc_hpte_hv_fault);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 43/46] KVM: PPC: Book3S HV: add virtual mode handlers for HPT hcalls and page faults
@ 2021-03-23 1:03 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:03 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
In order to support hash guests in the P9 path (which does not do real
mode hcalls or page fault handling), these real-mode hash specific
interrupts need to be implemented in virt mode.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 126 ++++++++++++++++++++++++++--
arch/powerpc/kvm/book3s_hv_rm_mmu.c | 8 ++
2 files changed, 127 insertions(+), 7 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index e6d113dbc076..5b3b6842d6b9 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -937,6 +937,52 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
return RESUME_HOST;
switch (req) {
+ case H_REMOVE:
+ ret = kvmppc_h_remove(vcpu, kvmppc_get_gpr(vcpu, 4),
+ kvmppc_get_gpr(vcpu, 5),
+ kvmppc_get_gpr(vcpu, 6));
+ if (ret = H_TOO_HARD)
+ return RESUME_HOST;
+ break;
+ case H_ENTER:
+ ret = kvmppc_h_enter(vcpu, kvmppc_get_gpr(vcpu, 4),
+ kvmppc_get_gpr(vcpu, 5),
+ kvmppc_get_gpr(vcpu, 6),
+ kvmppc_get_gpr(vcpu, 7));
+ if (ret = H_TOO_HARD)
+ return RESUME_HOST;
+ break;
+ case H_READ:
+ ret = kvmppc_h_read(vcpu, kvmppc_get_gpr(vcpu, 4),
+ kvmppc_get_gpr(vcpu, 5));
+ if (ret = H_TOO_HARD)
+ return RESUME_HOST;
+ break;
+ case H_CLEAR_MOD:
+ ret = kvmppc_h_clear_mod(vcpu, kvmppc_get_gpr(vcpu, 4),
+ kvmppc_get_gpr(vcpu, 5));
+ if (ret = H_TOO_HARD)
+ return RESUME_HOST;
+ break;
+ case H_CLEAR_REF:
+ ret = kvmppc_h_clear_ref(vcpu, kvmppc_get_gpr(vcpu, 4),
+ kvmppc_get_gpr(vcpu, 5));
+ if (ret = H_TOO_HARD)
+ return RESUME_HOST;
+ break;
+ case H_PROTECT:
+ ret = kvmppc_h_protect(vcpu, kvmppc_get_gpr(vcpu, 4),
+ kvmppc_get_gpr(vcpu, 5),
+ kvmppc_get_gpr(vcpu, 6));
+ if (ret = H_TOO_HARD)
+ return RESUME_HOST;
+ break;
+ case H_BULK_REMOVE:
+ ret = kvmppc_h_bulk_remove(vcpu);
+ if (ret = H_TOO_HARD)
+ return RESUME_HOST;
+ break;
+
case H_CEDE:
break;
case H_PROD:
@@ -1136,6 +1182,7 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu)
default:
return RESUME_HOST;
}
+ WARN_ON_ONCE(ret = H_TOO_HARD);
kvmppc_set_gpr(vcpu, 3, ret);
vcpu->arch.hcall_needed = 0;
return RESUME_GUEST;
@@ -1429,19 +1476,84 @@ static int kvmppc_handle_exit_hv(struct kvm_vcpu *vcpu,
* host page has been paged out. Any other HDSI/HISI interrupts
* have been handled already.
*/
- case BOOK3S_INTERRUPT_H_DATA_STORAGE:
- r = RESUME_PAGE_FAULT;
- if (vcpu->arch.fault_dsisr = HDSISR_CANARY)
+ case BOOK3S_INTERRUPT_H_DATA_STORAGE: {
+ unsigned long vsid;
+ long err;
+
+ if (vcpu->arch.fault_dsisr = HDSISR_CANARY) {
r = RESUME_GUEST; /* Just retry if it's the canary */
+ break;
+ }
+
+ if (kvm_is_radix(vcpu->kvm)) {
+ r = RESUME_PAGE_FAULT;
+ break;
+ }
+
+ if (!(vcpu->arch.fault_dsisr & (DSISR_NOHPTE | DSISR_PROTFAULT))) {
+ kvmppc_core_queue_data_storage(vcpu,
+ vcpu->arch.fault_dar, vcpu->arch.fault_dsisr);
+ r = RESUME_GUEST;
+ break;
+ }
+
+ if (!(vcpu->arch.shregs.msr & MSR_DR))
+ vsid = vcpu->kvm->arch.vrma_slb_v;
+ else
+ vsid = vcpu->arch.fault_gpa;
+
+ err = kvmppc_hpte_hv_fault(vcpu, vcpu->arch.fault_dar,
+ vsid, vcpu->arch.fault_dsisr, true);
+ if (err = 0) {
+ r = RESUME_GUEST;
+ } else if (err = -1 || err = -2) {
+ r = RESUME_PAGE_FAULT;
+ } else {
+ kvmppc_core_queue_data_storage(vcpu,
+ vcpu->arch.fault_dar, err);
+ r = RESUME_GUEST;
+ }
break;
- case BOOK3S_INTERRUPT_H_INST_STORAGE:
+ }
+ case BOOK3S_INTERRUPT_H_INST_STORAGE: {
+ unsigned long vsid;
+ long err;
+
vcpu->arch.fault_dar = kvmppc_get_pc(vcpu);
vcpu->arch.fault_dsisr = vcpu->arch.shregs.msr &
DSISR_SRR1_MATCH_64S;
- if (vcpu->arch.shregs.msr & HSRR1_HISI_WRITE)
- vcpu->arch.fault_dsisr |= DSISR_ISSTORE;
- r = RESUME_PAGE_FAULT;
+ if (kvm_is_radix(vcpu->kvm)) {
+ if (vcpu->arch.shregs.msr & HSRR1_HISI_WRITE)
+ vcpu->arch.fault_dsisr |= DSISR_ISSTORE;
+ r = RESUME_PAGE_FAULT;
+ break;
+ }
+
+ if (!(vcpu->arch.fault_dsisr & SRR1_ISI_NOPT)) {
+ kvmppc_core_queue_inst_storage(vcpu,
+ vcpu->arch.fault_dsisr);
+ r = RESUME_GUEST;
+ break;
+ }
+
+ if (!(vcpu->arch.shregs.msr & MSR_DR))
+ vsid = vcpu->kvm->arch.vrma_slb_v;
+ else
+ vsid = vcpu->arch.fault_gpa;
+
+ err = kvmppc_hpte_hv_fault(vcpu, vcpu->arch.fault_dar,
+ vsid, vcpu->arch.fault_dsisr, false);
+ if (err = 0) {
+ r = RESUME_GUEST;
+ } else if (err = -1) {
+ r = RESUME_PAGE_FAULT;
+ } else {
+ kvmppc_core_queue_inst_storage(vcpu, err);
+ r = RESUME_GUEST;
+ }
break;
+ }
+
/*
* This occurs if the guest executes an illegal instruction.
* If the guest debug is disabled, generate a program interrupt
diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
index 7af7c70f1468..8cc73abbf42b 100644
--- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c
+++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
@@ -409,6 +409,7 @@ long kvmppc_h_enter(struct kvm_vcpu *vcpu, unsigned long flags,
vcpu->arch.pgdir, true,
&vcpu->arch.regs.gpr[4]);
}
+EXPORT_SYMBOL_GPL(kvmppc_h_enter);
#ifdef __BIG_ENDIAN__
#define LOCK_TOKEN (*(u32 *)(&get_paca()->lock_token))
@@ -553,6 +554,7 @@ long kvmppc_h_remove(struct kvm_vcpu *vcpu, unsigned long flags,
return kvmppc_do_h_remove(vcpu->kvm, flags, pte_index, avpn,
&vcpu->arch.regs.gpr[4]);
}
+EXPORT_SYMBOL_GPL(kvmppc_h_remove);
long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
{
@@ -671,6 +673,7 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
return ret;
}
+EXPORT_SYMBOL_GPL(kvmppc_h_bulk_remove);
long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long pte_index, unsigned long avpn)
@@ -741,6 +744,7 @@ long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags,
return H_SUCCESS;
}
+EXPORT_SYMBOL_GPL(kvmppc_h_protect);
long kvmppc_h_read(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long pte_index)
@@ -781,6 +785,7 @@ long kvmppc_h_read(struct kvm_vcpu *vcpu, unsigned long flags,
}
return H_SUCCESS;
}
+EXPORT_SYMBOL_GPL(kvmppc_h_read);
long kvmppc_h_clear_ref(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long pte_index)
@@ -829,6 +834,7 @@ long kvmppc_h_clear_ref(struct kvm_vcpu *vcpu, unsigned long flags,
unlock_hpte(hpte, v & ~HPTE_V_HVLOCK);
return ret;
}
+EXPORT_SYMBOL_GPL(kvmppc_h_clear_ref);
long kvmppc_h_clear_mod(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long pte_index)
@@ -876,6 +882,7 @@ long kvmppc_h_clear_mod(struct kvm_vcpu *vcpu, unsigned long flags,
unlock_hpte(hpte, v & ~HPTE_V_HVLOCK);
return ret;
}
+EXPORT_SYMBOL_GPL(kvmppc_h_clear_mod);
static int kvmppc_get_hpa(struct kvm_vcpu *vcpu, unsigned long mmu_seq,
unsigned long gpa, int writing, unsigned long *hpa,
@@ -1294,3 +1301,4 @@ long kvmppc_hpte_hv_fault(struct kvm_vcpu *vcpu, unsigned long addr,
return -1; /* send fault up to host kernel mode */
}
+EXPORT_SYMBOL_GPL(kvmppc_hpte_hv_fault);
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 44/46] KVM: PPC: Book3S HV P9: implement hash guest support
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:03 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:03 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Guest entry/exit has to restore and save/clear the SLB, plus several
other bits to accommodate hash guests in the P9 path.
Radix host, hash guest support is removed from the P7/8 path.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 20 ++-
arch/powerpc/kvm/book3s_hv_interrupt.c | 160 ++++++++++++++++--------
arch/powerpc/kvm/book3s_hv_rm_mmu.c | 4 +
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 14 +--
4 files changed, 131 insertions(+), 67 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 5b3b6842d6b9..c151a60c0daa 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3828,7 +3828,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
}
kvmppc_xive_pull_vcpu(vcpu);
- vcpu->arch.slb_max = 0;
+ if (kvm_is_radix(vcpu->kvm))
+ vcpu->arch.slb_max = 0;
}
dec = mfspr(SPRN_DEC);
@@ -4061,7 +4062,6 @@ static void kvmppc_vcore_blocked(struct kvmppc_vcore *vc)
/*
* This never fails for a radix guest, as none of the operations it does
* for a radix guest can fail or have a way to report failure.
- * kvmhv_run_single_vcpu() relies on this fact.
*/
static int kvmhv_setup_mmu(struct kvm_vcpu *vcpu)
{
@@ -4240,8 +4240,15 @@ int kvmhv_run_single_vcpu(struct kvm_vcpu *vcpu, u64 time_limit,
vc->runner = vcpu;
/* See if the MMU is ready to go */
- if (!kvm->arch.mmu_ready)
- kvmhv_setup_mmu(vcpu);
+ if (!kvm->arch.mmu_ready) {
+ r = kvmhv_setup_mmu(vcpu);
+ if (r) {
+ run->exit_reason = KVM_EXIT_FAIL_ENTRY;
+ run->fail_entry.hardware_entry_failure_reason = 0;
+ vcpu->arch.ret = r;
+ return r;
+ }
+ }
if (need_resched())
cond_resched();
@@ -4254,7 +4261,8 @@ int kvmhv_run_single_vcpu(struct kvm_vcpu *vcpu, u64 time_limit,
preempt_disable();
pcpu = smp_processor_id();
vc->pcpu = pcpu;
- kvmppc_prepare_radix_vcpu(vcpu, pcpu);
+ if (kvm_is_radix(kvm))
+ kvmppc_prepare_radix_vcpu(vcpu, pcpu);
local_irq_disable();
hard_irq_disable();
@@ -4454,7 +4462,7 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
vcpu->arch.state = KVMPPC_VCPU_BUSY_IN_HOST;
do {
- if (kvm_is_radix(kvm))
+ if (radix_enabled())
r = kvmhv_run_single_vcpu(vcpu, ~(u64)0,
vcpu->arch.vcore->lpcr);
else
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index cd84d2c37632..03fbfef708a8 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -55,6 +55,50 @@ static void __accumulate_time(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator
#define accumulate_time(vcpu, next) do {} while (0)
#endif
+static inline void mfslb(unsigned int idx, u64 *slbee, u64 *slbev)
+{
+ asm volatile("slbmfev %0,%1" : "=r" (*slbev) : "r" (idx));
+ asm volatile("slbmfee %0,%1" : "=r" (*slbee) : "r" (idx));
+}
+
+static inline void __mtslb(u64 slbee, u64 slbev)
+{
+ asm volatile("slbmte %0,%1" :: "r" (slbev), "r" (slbee));
+}
+
+static inline void mtslb(unsigned int idx, u64 slbee, u64 slbev)
+{
+ BUG_ON((slbee & 0xfff) != idx);
+
+ __mtslb(slbee, slbev);
+}
+
+static inline void slb_invalidate(unsigned int ih)
+{
+ asm volatile("slbia %0" :: "i"(ih));
+}
+
+/*
+ * Malicious or buggy radix guests may have inserted SLB entries
+ * (only 0..3 because radix always runs with UPRT=1), so these must
+ * be cleared here to avoid side-channels. slbmte is used rather
+ * than slbia, as it won't clear cached translations.
+ */
+static void radix_clear_slb(void)
+{
+ u64 slbee, slbev;
+ int i;
+
+ for (i = 0; i < 4; i++) {
+ mfslb(i, &slbee, &slbev);
+ if (unlikely(slbee || slbev)) {
+ slbee = i;
+ slbev = 0;
+ mtslb(i, slbee, slbev);
+ }
+ }
+}
+
static void switch_mmu_to_guest_radix(struct kvm *kvm, struct kvm_vcpu *vcpu, u64 lpcr)
{
struct kvmppc_vcore *vc = vcpu->arch.vcore;
@@ -80,6 +124,31 @@ static void switch_mmu_to_guest_radix(struct kvm *kvm, struct kvm_vcpu *vcpu, u6
kvmppc_check_need_tlb_flush(kvm, vc->pcpu, nested);
}
+static void switch_mmu_to_guest_hpt(struct kvm *kvm, struct kvm_vcpu *vcpu, u64 lpcr)
+{
+ struct kvm_nested_guest *nested = vcpu->arch.nested;
+ u32 lpid;
+ int i;
+
+ BUG_ON(nested);
+
+ lpid = kvm->arch.lpid;
+
+ mtspr(SPRN_LPID, lpid);
+ mtspr(SPRN_LPCR, lpcr);
+ mtspr(SPRN_PID, vcpu->arch.pid);
+
+ for (i = 0; i < vcpu->arch.slb_max; i++)
+ __mtslb(vcpu->arch.slb[i].orige, vcpu->arch.slb[i].origv);
+
+ isync();
+
+ /*
+ * TLBIEL is not virtualised for HPT guests, so check_need_tlb_flush
+ * is not required here.
+ */
+}
+
static void switch_mmu_to_host_radix(struct kvm *kvm, u32 pid)
{
isync();
@@ -91,42 +160,30 @@ static void switch_mmu_to_host_radix(struct kvm *kvm, u32 pid)
isync();
}
-static inline void mfslb(unsigned int idx, u64 *slbee, u64 *slbev)
-{
- asm volatile("slbmfev %0,%1" : "=r" (*slbev) : "r" (idx));
- asm volatile("slbmfee %0,%1" : "=r" (*slbee) : "r" (idx));
-}
-
-static inline void mtslb(unsigned int idx, u64 slbee, u64 slbev)
-{
- BUG_ON((slbee & 0xfff) != idx);
-
- asm volatile("slbmte %0,%1" :: "r" (slbev), "r" (slbee));
-}
-
-static inline void slb_invalidate(unsigned int ih)
-{
- asm volatile("slbia %0" :: "i"(ih));
-}
-
-/*
- * Malicious or buggy radix guests may have inserted SLB entries
- * (only 0..3 because radix always runs with UPRT=1), so these must
- * be cleared here to avoid side-channels. slbmte is used rather
- * than slbia, as it won't clear cached translations.
- */
-static void radix_clear_slb(void)
+static void save_clear_guest_mmu(struct kvm *kvm, struct kvm_vcpu *vcpu)
{
- u64 slbee, slbev;
- int i;
+ if (kvm_is_radix(kvm)) {
+ radix_clear_slb();
+ } else {
+ int i;
+ int nr = 0;
- for (i = 0; i < 4; i++) {
- mfslb(i, &slbee, &slbev);
- if (unlikely(slbee || slbev)) {
- slbee = i;
- slbev = 0;
- mtslb(i, slbee, slbev);
+ /*
+ * This must run before switching to host (radix host can't
+ * access all SLBs).
+ */
+ for (i = 0; i < vcpu->arch.slb_nr; i++) {
+ u64 slbee, slbev;
+ mfslb(i, &slbee, &slbev);
+ if (slbee & SLB_ESID_V) {
+ vcpu->arch.slb[nr].orige = slbee | i;
+ vcpu->arch.slb[nr].origv = slbev;
+ nr++;
+ }
}
+ vcpu->arch.slb_max = nr;
+ mtslb(0, 0, 0);
+ slb_invalidate(6);
}
}
@@ -234,10 +291,18 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
mtspr(SPRN_AMOR, ~0UL);
- if (cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
- __mtmsrd(msr & ~(MSR_IR|MSR_DR|MSR_RI), 0);
+ local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
+ if (kvm_is_radix(kvm)) {
+ if (cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
+ __mtmsrd(msr & ~(MSR_IR|MSR_DR|MSR_RI), 0);
+ switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
+ if (!cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
+ __mtmsrd(0, 1); /* clear RI */
- switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
+ } else {
+ __mtmsrd(msr & ~(MSR_IR|MSR_DR|MSR_RI), 0);
+ switch_mmu_to_guest_hpt(kvm, vcpu, lpcr);
+ }
/*
* P9 suppresses the HDEC exception when LPCR[HDICE] = 0,
@@ -245,9 +310,6 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
*/
mtspr(SPRN_HDEC, hdec);
- if (!cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
- __mtmsrd(0, 1); /* clear RI */
-
mtspr(SPRN_DAR, vcpu->arch.shregs.dar);
mtspr(SPRN_DSISR, vcpu->arch.shregs.dsisr);
mtspr(SPRN_SRR0, vcpu->arch.shregs.srr0);
@@ -255,10 +317,7 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
accumulate_time(vcpu, &vcpu->arch.guest_time);
- local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
kvmppc_p9_enter_guest(vcpu);
- // Radix host and guest means host never runs with guest MMU state
- local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
accumulate_time(vcpu, &vcpu->arch.rm_intr);
@@ -359,8 +418,6 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
#endif
}
- radix_clear_slb();
-
accumulate_time(vcpu, &vcpu->arch.rm_exit);
/* Advance host PURR/SPURR by the amount used by guest */
@@ -394,11 +451,14 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
mtspr(SPRN_DAWRX1, host_dawrx1);
}
- /*
- * Since this is radix, do a eieio; tlbsync; ptesync sequence in
- * case we interrupted the guest between a tlbie and a ptesync.
- */
- asm volatile("eieio; tlbsync; ptesync");
+ if (kvm_is_radix(kvm)) {
+ /*
+ * Since this is radix, do a eieio; tlbsync; ptesync sequence
+ * in case we interrupted the guest between a tlbie and a
+ * ptesync.
+ */
+ asm volatile("eieio; tlbsync; ptesync");
+ }
/*
* cp_abort is required if the processor supports local copy-paste
@@ -425,7 +485,9 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
/* HDEC must be at least as large as DEC, so decrementer_max fits */
mtspr(SPRN_HDEC, decrementer_max);
+ save_clear_guest_mmu(kvm, vcpu);
switch_mmu_to_host_radix(kvm, host_pidr);
+ local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
/*
* If we are in real mode, only switch MMU on after the MMU is
diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
index 8cc73abbf42b..f487ebb3a70a 100644
--- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c
+++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
@@ -57,6 +57,10 @@ static int global_invalidates(struct kvm *kvm)
else
global = 1;
+ /* LPID has been switched to host if in virt mode so can't do local */
+ if (!global && (mfmsr() & (MSR_IR|MSR_DR)))
+ global = 1;
+
if (!global) {
/* any other core might now have stale TLB entries... */
smp_wmb();
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 564ca9feef35..5cbc26ebcf8e 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -885,14 +885,11 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
cmpdi r3, 512 /* 1 microsecond */
blt hdec_soon
- /* For hash guest, clear out and reload the SLB */
-BEGIN_MMU_FTR_SECTION
- /* Radix host won't have populated the SLB, so no need to clear */
+ /* Clear out and reload the SLB */
li r6, 0
slbmte r6, r6
PPC_SLBIA(6)
ptesync
-END_MMU_FTR_SECTION_IFCLR(MMU_FTR_TYPE_RADIX)
/* Load up guest SLB entries (N.B. slb_max will be 0 for radix) */
lwz r5,VCPU_SLB_MAX(r4)
@@ -1394,9 +1391,6 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
stw r5,VCPU_SLB_MAX(r9)
/* load host SLB entries */
-BEGIN_MMU_FTR_SECTION
- b guest_bypass
-END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
ld r8,PACA_SLBSHADOWPTR(r13)
.rept SLB_NUM_BOLTED
@@ -3148,10 +3142,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
PPC_SLBIA(6)
ptesync
-BEGIN_MMU_FTR_SECTION
- b 4f
-END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
-
/* load host SLB entries */
ld r8, PACA_SLBSHADOWPTR(r13)
.rept SLB_NUM_BOLTED
@@ -3165,7 +3155,7 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
3: addi r8, r8, 16
.endr
-4: lwz r7, KVM_HOST_LPID(r10)
+ lwz r7, KVM_HOST_LPID(r10)
mtspr SPRN_LPID, r7
mtspr SPRN_PID, r0
ld r8, KVM_HOST_LPCR(r10)
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 44/46] KVM: PPC: Book3S HV P9: implement hash guest support
@ 2021-03-23 1:03 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:03 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Guest entry/exit has to restore and save/clear the SLB, plus several
other bits to accommodate hash guests in the P9 path.
Radix host, hash guest support is removed from the P7/8 path.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 20 ++-
arch/powerpc/kvm/book3s_hv_interrupt.c | 160 ++++++++++++++++--------
arch/powerpc/kvm/book3s_hv_rm_mmu.c | 4 +
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 14 +--
4 files changed, 131 insertions(+), 67 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 5b3b6842d6b9..c151a60c0daa 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -3828,7 +3828,8 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
}
kvmppc_xive_pull_vcpu(vcpu);
- vcpu->arch.slb_max = 0;
+ if (kvm_is_radix(vcpu->kvm))
+ vcpu->arch.slb_max = 0;
}
dec = mfspr(SPRN_DEC);
@@ -4061,7 +4062,6 @@ static void kvmppc_vcore_blocked(struct kvmppc_vcore *vc)
/*
* This never fails for a radix guest, as none of the operations it does
* for a radix guest can fail or have a way to report failure.
- * kvmhv_run_single_vcpu() relies on this fact.
*/
static int kvmhv_setup_mmu(struct kvm_vcpu *vcpu)
{
@@ -4240,8 +4240,15 @@ int kvmhv_run_single_vcpu(struct kvm_vcpu *vcpu, u64 time_limit,
vc->runner = vcpu;
/* See if the MMU is ready to go */
- if (!kvm->arch.mmu_ready)
- kvmhv_setup_mmu(vcpu);
+ if (!kvm->arch.mmu_ready) {
+ r = kvmhv_setup_mmu(vcpu);
+ if (r) {
+ run->exit_reason = KVM_EXIT_FAIL_ENTRY;
+ run->fail_entry.hardware_entry_failure_reason = 0;
+ vcpu->arch.ret = r;
+ return r;
+ }
+ }
if (need_resched())
cond_resched();
@@ -4254,7 +4261,8 @@ int kvmhv_run_single_vcpu(struct kvm_vcpu *vcpu, u64 time_limit,
preempt_disable();
pcpu = smp_processor_id();
vc->pcpu = pcpu;
- kvmppc_prepare_radix_vcpu(vcpu, pcpu);
+ if (kvm_is_radix(kvm))
+ kvmppc_prepare_radix_vcpu(vcpu, pcpu);
local_irq_disable();
hard_irq_disable();
@@ -4454,7 +4462,7 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
vcpu->arch.state = KVMPPC_VCPU_BUSY_IN_HOST;
do {
- if (kvm_is_radix(kvm))
+ if (radix_enabled())
r = kvmhv_run_single_vcpu(vcpu, ~(u64)0,
vcpu->arch.vcore->lpcr);
else
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index cd84d2c37632..03fbfef708a8 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -55,6 +55,50 @@ static void __accumulate_time(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator
#define accumulate_time(vcpu, next) do {} while (0)
#endif
+static inline void mfslb(unsigned int idx, u64 *slbee, u64 *slbev)
+{
+ asm volatile("slbmfev %0,%1" : "=r" (*slbev) : "r" (idx));
+ asm volatile("slbmfee %0,%1" : "=r" (*slbee) : "r" (idx));
+}
+
+static inline void __mtslb(u64 slbee, u64 slbev)
+{
+ asm volatile("slbmte %0,%1" :: "r" (slbev), "r" (slbee));
+}
+
+static inline void mtslb(unsigned int idx, u64 slbee, u64 slbev)
+{
+ BUG_ON((slbee & 0xfff) != idx);
+
+ __mtslb(slbee, slbev);
+}
+
+static inline void slb_invalidate(unsigned int ih)
+{
+ asm volatile("slbia %0" :: "i"(ih));
+}
+
+/*
+ * Malicious or buggy radix guests may have inserted SLB entries
+ * (only 0..3 because radix always runs with UPRT=1), so these must
+ * be cleared here to avoid side-channels. slbmte is used rather
+ * than slbia, as it won't clear cached translations.
+ */
+static void radix_clear_slb(void)
+{
+ u64 slbee, slbev;
+ int i;
+
+ for (i = 0; i < 4; i++) {
+ mfslb(i, &slbee, &slbev);
+ if (unlikely(slbee || slbev)) {
+ slbee = i;
+ slbev = 0;
+ mtslb(i, slbee, slbev);
+ }
+ }
+}
+
static void switch_mmu_to_guest_radix(struct kvm *kvm, struct kvm_vcpu *vcpu, u64 lpcr)
{
struct kvmppc_vcore *vc = vcpu->arch.vcore;
@@ -80,6 +124,31 @@ static void switch_mmu_to_guest_radix(struct kvm *kvm, struct kvm_vcpu *vcpu, u6
kvmppc_check_need_tlb_flush(kvm, vc->pcpu, nested);
}
+static void switch_mmu_to_guest_hpt(struct kvm *kvm, struct kvm_vcpu *vcpu, u64 lpcr)
+{
+ struct kvm_nested_guest *nested = vcpu->arch.nested;
+ u32 lpid;
+ int i;
+
+ BUG_ON(nested);
+
+ lpid = kvm->arch.lpid;
+
+ mtspr(SPRN_LPID, lpid);
+ mtspr(SPRN_LPCR, lpcr);
+ mtspr(SPRN_PID, vcpu->arch.pid);
+
+ for (i = 0; i < vcpu->arch.slb_max; i++)
+ __mtslb(vcpu->arch.slb[i].orige, vcpu->arch.slb[i].origv);
+
+ isync();
+
+ /*
+ * TLBIEL is not virtualised for HPT guests, so check_need_tlb_flush
+ * is not required here.
+ */
+}
+
static void switch_mmu_to_host_radix(struct kvm *kvm, u32 pid)
{
isync();
@@ -91,42 +160,30 @@ static void switch_mmu_to_host_radix(struct kvm *kvm, u32 pid)
isync();
}
-static inline void mfslb(unsigned int idx, u64 *slbee, u64 *slbev)
-{
- asm volatile("slbmfev %0,%1" : "=r" (*slbev) : "r" (idx));
- asm volatile("slbmfee %0,%1" : "=r" (*slbee) : "r" (idx));
-}
-
-static inline void mtslb(unsigned int idx, u64 slbee, u64 slbev)
-{
- BUG_ON((slbee & 0xfff) != idx);
-
- asm volatile("slbmte %0,%1" :: "r" (slbev), "r" (slbee));
-}
-
-static inline void slb_invalidate(unsigned int ih)
-{
- asm volatile("slbia %0" :: "i"(ih));
-}
-
-/*
- * Malicious or buggy radix guests may have inserted SLB entries
- * (only 0..3 because radix always runs with UPRT=1), so these must
- * be cleared here to avoid side-channels. slbmte is used rather
- * than slbia, as it won't clear cached translations.
- */
-static void radix_clear_slb(void)
+static void save_clear_guest_mmu(struct kvm *kvm, struct kvm_vcpu *vcpu)
{
- u64 slbee, slbev;
- int i;
+ if (kvm_is_radix(kvm)) {
+ radix_clear_slb();
+ } else {
+ int i;
+ int nr = 0;
- for (i = 0; i < 4; i++) {
- mfslb(i, &slbee, &slbev);
- if (unlikely(slbee || slbev)) {
- slbee = i;
- slbev = 0;
- mtslb(i, slbee, slbev);
+ /*
+ * This must run before switching to host (radix host can't
+ * access all SLBs).
+ */
+ for (i = 0; i < vcpu->arch.slb_nr; i++) {
+ u64 slbee, slbev;
+ mfslb(i, &slbee, &slbev);
+ if (slbee & SLB_ESID_V) {
+ vcpu->arch.slb[nr].orige = slbee | i;
+ vcpu->arch.slb[nr].origv = slbev;
+ nr++;
+ }
}
+ vcpu->arch.slb_max = nr;
+ mtslb(0, 0, 0);
+ slb_invalidate(6);
}
}
@@ -234,10 +291,18 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
mtspr(SPRN_AMOR, ~0UL);
- if (cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
- __mtmsrd(msr & ~(MSR_IR|MSR_DR|MSR_RI), 0);
+ local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
+ if (kvm_is_radix(kvm)) {
+ if (cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
+ __mtmsrd(msr & ~(MSR_IR|MSR_DR|MSR_RI), 0);
+ switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
+ if (!cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
+ __mtmsrd(0, 1); /* clear RI */
- switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
+ } else {
+ __mtmsrd(msr & ~(MSR_IR|MSR_DR|MSR_RI), 0);
+ switch_mmu_to_guest_hpt(kvm, vcpu, lpcr);
+ }
/*
* P9 suppresses the HDEC exception when LPCR[HDICE] = 0,
@@ -245,9 +310,6 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
*/
mtspr(SPRN_HDEC, hdec);
- if (!cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
- __mtmsrd(0, 1); /* clear RI */
-
mtspr(SPRN_DAR, vcpu->arch.shregs.dar);
mtspr(SPRN_DSISR, vcpu->arch.shregs.dsisr);
mtspr(SPRN_SRR0, vcpu->arch.shregs.srr0);
@@ -255,10 +317,7 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
accumulate_time(vcpu, &vcpu->arch.guest_time);
- local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
kvmppc_p9_enter_guest(vcpu);
- // Radix host and guest means host never runs with guest MMU state
- local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
accumulate_time(vcpu, &vcpu->arch.rm_intr);
@@ -359,8 +418,6 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
#endif
}
- radix_clear_slb();
-
accumulate_time(vcpu, &vcpu->arch.rm_exit);
/* Advance host PURR/SPURR by the amount used by guest */
@@ -394,11 +451,14 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
mtspr(SPRN_DAWRX1, host_dawrx1);
}
- /*
- * Since this is radix, do a eieio; tlbsync; ptesync sequence in
- * case we interrupted the guest between a tlbie and a ptesync.
- */
- asm volatile("eieio; tlbsync; ptesync");
+ if (kvm_is_radix(kvm)) {
+ /*
+ * Since this is radix, do a eieio; tlbsync; ptesync sequence
+ * in case we interrupted the guest between a tlbie and a
+ * ptesync.
+ */
+ asm volatile("eieio; tlbsync; ptesync");
+ }
/*
* cp_abort is required if the processor supports local copy-paste
@@ -425,7 +485,9 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
/* HDEC must be at least as large as DEC, so decrementer_max fits */
mtspr(SPRN_HDEC, decrementer_max);
+ save_clear_guest_mmu(kvm, vcpu);
switch_mmu_to_host_radix(kvm, host_pidr);
+ local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
/*
* If we are in real mode, only switch MMU on after the MMU is
diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
index 8cc73abbf42b..f487ebb3a70a 100644
--- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c
+++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
@@ -57,6 +57,10 @@ static int global_invalidates(struct kvm *kvm)
else
global = 1;
+ /* LPID has been switched to host if in virt mode so can't do local */
+ if (!global && (mfmsr() & (MSR_IR|MSR_DR)))
+ global = 1;
+
if (!global) {
/* any other core might now have stale TLB entries... */
smp_wmb();
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 564ca9feef35..5cbc26ebcf8e 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -885,14 +885,11 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
cmpdi r3, 512 /* 1 microsecond */
blt hdec_soon
- /* For hash guest, clear out and reload the SLB */
-BEGIN_MMU_FTR_SECTION
- /* Radix host won't have populated the SLB, so no need to clear */
+ /* Clear out and reload the SLB */
li r6, 0
slbmte r6, r6
PPC_SLBIA(6)
ptesync
-END_MMU_FTR_SECTION_IFCLR(MMU_FTR_TYPE_RADIX)
/* Load up guest SLB entries (N.B. slb_max will be 0 for radix) */
lwz r5,VCPU_SLB_MAX(r4)
@@ -1394,9 +1391,6 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
stw r5,VCPU_SLB_MAX(r9)
/* load host SLB entries */
-BEGIN_MMU_FTR_SECTION
- b guest_bypass
-END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
ld r8,PACA_SLBSHADOWPTR(r13)
.rept SLB_NUM_BOLTED
@@ -3148,10 +3142,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
PPC_SLBIA(6)
ptesync
-BEGIN_MMU_FTR_SECTION
- b 4f
-END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
-
/* load host SLB entries */
ld r8, PACA_SLBSHADOWPTR(r13)
.rept SLB_NUM_BOLTED
@@ -3165,7 +3155,7 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
3: addi r8, r8, 16
.endr
-4: lwz r7, KVM_HOST_LPID(r10)
+ lwz r7, KVM_HOST_LPID(r10)
mtspr SPRN_LPID, r7
mtspr SPRN_PID, r0
ld r8, KVM_HOST_LPCR(r10)
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* Re: [PATCH v4 44/46] KVM: PPC: Book3S HV P9: implement hash guest support
2021-03-23 1:03 ` Nicholas Piggin
@ 2021-03-23 15:53 ` Fabiano Rosas
-1 siblings, 0 replies; 188+ messages in thread
From: Fabiano Rosas @ 2021-03-23 15:53 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Nicholas Piggin <npiggin@gmail.com> writes:
> Guest entry/exit has to restore and save/clear the SLB, plus several
> other bits to accommodate hash guests in the P9 path.
>
> Radix host, hash guest support is removed from the P7/8 path.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
<snip>
> diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
> index cd84d2c37632..03fbfef708a8 100644
> --- a/arch/powerpc/kvm/book3s_hv_interrupt.c
> +++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
> @@ -55,6 +55,50 @@ static void __accumulate_time(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator
> #define accumulate_time(vcpu, next) do {} while (0)
> #endif
>
> +static inline void mfslb(unsigned int idx, u64 *slbee, u64 *slbev)
> +{
> + asm volatile("slbmfev %0,%1" : "=r" (*slbev) : "r" (idx));
> + asm volatile("slbmfee %0,%1" : "=r" (*slbee) : "r" (idx));
> +}
> +
> +static inline void __mtslb(u64 slbee, u64 slbev)
> +{
> + asm volatile("slbmte %0,%1" :: "r" (slbev), "r" (slbee));
> +}
> +
> +static inline void mtslb(unsigned int idx, u64 slbee, u64 slbev)
> +{
> + BUG_ON((slbee & 0xfff) != idx);
> +
> + __mtslb(slbee, slbev);
> +}
> +
> +static inline void slb_invalidate(unsigned int ih)
> +{
> + asm volatile("slbia %0" :: "i"(ih));
> +}
Fyi, in my environment the assembler complains:
{standard input}: Assembler messages:
{standard input}:1293: Error: junk at end of line: `6'
{standard input}:2138: Error: junk at end of line: `6'
make[3]: *** [../scripts/Makefile.build:271:
arch/powerpc/kvm/book3s_hv_interrupt.o] Error 1
This works:
- asm volatile("slbia %0" :: "i"(ih));
+ asm volatile(PPC_SLBIA(%0) :: "i"(ih));
But I don't know what is going on.
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 44/46] KVM: PPC: Book3S HV P9: implement hash guest support
@ 2021-03-23 15:53 ` Fabiano Rosas
0 siblings, 0 replies; 188+ messages in thread
From: Fabiano Rosas @ 2021-03-23 15:53 UTC (permalink / raw)
To: Nicholas Piggin, kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
Nicholas Piggin <npiggin@gmail.com> writes:
> Guest entry/exit has to restore and save/clear the SLB, plus several
> other bits to accommodate hash guests in the P9 path.
>
> Radix host, hash guest support is removed from the P7/8 path.
>
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
<snip>
> diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
> index cd84d2c37632..03fbfef708a8 100644
> --- a/arch/powerpc/kvm/book3s_hv_interrupt.c
> +++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
> @@ -55,6 +55,50 @@ static void __accumulate_time(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator
> #define accumulate_time(vcpu, next) do {} while (0)
> #endif
>
> +static inline void mfslb(unsigned int idx, u64 *slbee, u64 *slbev)
> +{
> + asm volatile("slbmfev %0,%1" : "=r" (*slbev) : "r" (idx));
> + asm volatile("slbmfee %0,%1" : "=r" (*slbee) : "r" (idx));
> +}
> +
> +static inline void __mtslb(u64 slbee, u64 slbev)
> +{
> + asm volatile("slbmte %0,%1" :: "r" (slbev), "r" (slbee));
> +}
> +
> +static inline void mtslb(unsigned int idx, u64 slbee, u64 slbev)
> +{
> + BUG_ON((slbee & 0xfff) != idx);
> +
> + __mtslb(slbee, slbev);
> +}
> +
> +static inline void slb_invalidate(unsigned int ih)
> +{
> + asm volatile("slbia %0" :: "i"(ih));
> +}
Fyi, in my environment the assembler complains:
{standard input}: Assembler messages:
{standard input}:1293: Error: junk at end of line: `6'
{standard input}:2138: Error: junk at end of line: `6'
make[3]: *** [../scripts/Makefile.build:271:
arch/powerpc/kvm/book3s_hv_interrupt.o] Error 1
This works:
- asm volatile("slbia %0" :: "i"(ih));
+ asm volatile(PPC_SLBIA(%0) :: "i"(ih));
But I don't know what is going on.
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 44/46] KVM: PPC: Book3S HV P9: implement hash guest support
2021-03-23 15:53 ` Fabiano Rosas
@ 2021-03-24 1:34 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-24 1:34 UTC (permalink / raw)
To: Fabiano Rosas, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Fabiano Rosas's message of March 24, 2021 1:53 am:
> Nicholas Piggin <npiggin@gmail.com> writes:
>
>> Guest entry/exit has to restore and save/clear the SLB, plus several
>> other bits to accommodate hash guests in the P9 path.
>>
>> Radix host, hash guest support is removed from the P7/8 path.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>> ---
>
> <snip>
>
>> diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
>> index cd84d2c37632..03fbfef708a8 100644
>> --- a/arch/powerpc/kvm/book3s_hv_interrupt.c
>> +++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
>> @@ -55,6 +55,50 @@ static void __accumulate_time(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator
>> #define accumulate_time(vcpu, next) do {} while (0)
>> #endif
>>
>> +static inline void mfslb(unsigned int idx, u64 *slbee, u64 *slbev)
>> +{
>> + asm volatile("slbmfev %0,%1" : "=r" (*slbev) : "r" (idx));
>> + asm volatile("slbmfee %0,%1" : "=r" (*slbee) : "r" (idx));
>> +}
>> +
>> +static inline void __mtslb(u64 slbee, u64 slbev)
>> +{
>> + asm volatile("slbmte %0,%1" :: "r" (slbev), "r" (slbee));
>> +}
>> +
>> +static inline void mtslb(unsigned int idx, u64 slbee, u64 slbev)
>> +{
>> + BUG_ON((slbee & 0xfff) != idx);
>> +
>> + __mtslb(slbee, slbev);
>> +}
>> +
>> +static inline void slb_invalidate(unsigned int ih)
>> +{
>> + asm volatile("slbia %0" :: "i"(ih));
>> +}
>
> Fyi, in my environment the assembler complains:
>
> {standard input}: Assembler messages:
> {standard input}:1293: Error: junk at end of line: `6'
> {standard input}:2138: Error: junk at end of line: `6'
> make[3]: *** [../scripts/Makefile.build:271:
> arch/powerpc/kvm/book3s_hv_interrupt.o] Error 1
>
> This works:
>
> - asm volatile("slbia %0" :: "i"(ih));
> + asm volatile(PPC_SLBIA(%0) :: "i"(ih));
>
> But I don't know what is going on.
Ah yes, we still need to use PPC_SLBIA. IH parameter to slbia was only
added in binutils 2.27 and we support down to 2.23.
Thanks for the fix I'll add it.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread* Re: [PATCH v4 44/46] KVM: PPC: Book3S HV P9: implement hash guest support
@ 2021-03-24 1:34 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-24 1:34 UTC (permalink / raw)
To: Fabiano Rosas, kvm-ppc; +Cc: linuxppc-dev
Excerpts from Fabiano Rosas's message of March 24, 2021 1:53 am:
> Nicholas Piggin <npiggin@gmail.com> writes:
>
>> Guest entry/exit has to restore and save/clear the SLB, plus several
>> other bits to accommodate hash guests in the P9 path.
>>
>> Radix host, hash guest support is removed from the P7/8 path.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>> ---
>
> <snip>
>
>> diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
>> index cd84d2c37632..03fbfef708a8 100644
>> --- a/arch/powerpc/kvm/book3s_hv_interrupt.c
>> +++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
>> @@ -55,6 +55,50 @@ static void __accumulate_time(struct kvm_vcpu *vcpu, struct kvmhv_tb_accumulator
>> #define accumulate_time(vcpu, next) do {} while (0)
>> #endif
>>
>> +static inline void mfslb(unsigned int idx, u64 *slbee, u64 *slbev)
>> +{
>> + asm volatile("slbmfev %0,%1" : "=r" (*slbev) : "r" (idx));
>> + asm volatile("slbmfee %0,%1" : "=r" (*slbee) : "r" (idx));
>> +}
>> +
>> +static inline void __mtslb(u64 slbee, u64 slbev)
>> +{
>> + asm volatile("slbmte %0,%1" :: "r" (slbev), "r" (slbee));
>> +}
>> +
>> +static inline void mtslb(unsigned int idx, u64 slbee, u64 slbev)
>> +{
>> + BUG_ON((slbee & 0xfff) != idx);
>> +
>> + __mtslb(slbee, slbev);
>> +}
>> +
>> +static inline void slb_invalidate(unsigned int ih)
>> +{
>> + asm volatile("slbia %0" :: "i"(ih));
>> +}
>
> Fyi, in my environment the assembler complains:
>
> {standard input}: Assembler messages:
> {standard input}:1293: Error: junk at end of line: `6'
> {standard input}:2138: Error: junk at end of line: `6'
> make[3]: *** [../scripts/Makefile.build:271:
> arch/powerpc/kvm/book3s_hv_interrupt.o] Error 1
>
> This works:
>
> - asm volatile("slbia %0" :: "i"(ih));
> + asm volatile(PPC_SLBIA(%0) :: "i"(ih));
>
> But I don't know what is going on.
Ah yes, we still need to use PPC_SLBIA. IH parameter to slbia was only
added in binutils 2.27 and we support down to 2.23.
Thanks for the fix I'll add it.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread
* [PATCH v4 45/46] KVM: PPC: Book3S HV P9: implement hash host / hash guest support
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:03 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:03 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
This additionally has to save and restore the host SLB, and also
ensure that the MMU is off while switching into the guest SLB.
P9 and later CPUs now always go via the P9 path.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_64_entry.S | 6 +++++
arch/powerpc/kvm/book3s_hv.c | 4 ++-
arch/powerpc/kvm/book3s_hv_interrupt.c | 35 ++++++++++++++++++++++----
3 files changed, 39 insertions(+), 6 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index 845df5fefdbd..7cef6f5212d8 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -339,6 +339,12 @@ kvmppc_p9_exit_interrupt:
* effort for a small bit of code. Lots of other things to do first.
*/
kvmppc_p9_bad_interrupt:
+BEGIN_MMU_FTR_SECTION
+ /*
+ * Hash host doesn't try to recover MMU (requires host SLB reload)
+ */
+ b .
+END_MMU_FTR_SECTION_IFCLR(MMU_FTR_TYPE_RADIX)
/*
* Set GUEST_MODE_NONE so the handler won't branch to KVM, and clear
* MSR_RI in r12 ([H]SRR1) so the handler won't try to return.
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index c151a60c0daa..e732d65772b1 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -4462,7 +4462,7 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
vcpu->arch.state = KVMPPC_VCPU_BUSY_IN_HOST;
do {
- if (radix_enabled())
+ if (cpu_has_feature(CPU_FTR_ARCH_300))
r = kvmhv_run_single_vcpu(vcpu, ~(u64)0,
vcpu->arch.vcore->lpcr);
else
@@ -5543,6 +5543,8 @@ static int kvmhv_enable_nested(struct kvm *kvm)
return -EPERM;
if (!cpu_has_feature(CPU_FTR_ARCH_300))
return -ENODEV;
+ if (!radix_enabled())
+ return -ENODEV;
/* kvm == NULL means the caller is testing if the capability exists */
if (kvm)
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index 03fbfef708a8..bd95d05219b7 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -149,7 +149,7 @@ static void switch_mmu_to_guest_hpt(struct kvm *kvm, struct kvm_vcpu *vcpu, u64
*/
}
-static void switch_mmu_to_host_radix(struct kvm *kvm, u32 pid)
+static void switch_mmu_to_host(struct kvm *kvm, u32 pid)
{
isync();
mtspr(SPRN_PID, pid);
@@ -158,6 +158,23 @@ static void switch_mmu_to_host_radix(struct kvm *kvm, u32 pid)
isync();
mtspr(SPRN_LPCR, kvm->arch.host_lpcr);
isync();
+
+ if (!radix_enabled())
+ slb_restore_bolted_realmode();
+}
+
+static void save_clear_host_mmu(struct kvm *kvm)
+{
+ if (!radix_enabled()) {
+ /*
+ * Hash host could save and restore host SLB entries to
+ * reduce SLB fault overheads of VM exits, but for now the
+ * existing code clears all entries and restores just the
+ * bolted ones when switching back to host.
+ */
+ mtslb(0, 0, 0);
+ slb_invalidate(6);
+ }
}
static void save_clear_guest_mmu(struct kvm *kvm, struct kvm_vcpu *vcpu)
@@ -292,15 +309,23 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
mtspr(SPRN_AMOR, ~0UL);
local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
+
+ /*
+ * Hash host, hash guest, or radix guest with prefetch bug, all have
+ * to disable the MMU before switching to guest MMU state.
+ */
+ if (!radix_enabled() || !kvm_is_radix(kvm) ||
+ cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
+ __mtmsrd(msr & ~(MSR_IR|MSR_DR|MSR_RI), 0);
+
+ save_clear_host_mmu(kvm);
+
if (kvm_is_radix(kvm)) {
- if (cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
- __mtmsrd(msr & ~(MSR_IR|MSR_DR|MSR_RI), 0);
switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
if (!cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
__mtmsrd(0, 1); /* clear RI */
} else {
- __mtmsrd(msr & ~(MSR_IR|MSR_DR|MSR_RI), 0);
switch_mmu_to_guest_hpt(kvm, vcpu, lpcr);
}
@@ -486,7 +511,7 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
mtspr(SPRN_HDEC, decrementer_max);
save_clear_guest_mmu(kvm, vcpu);
- switch_mmu_to_host_radix(kvm, host_pidr);
+ switch_mmu_to_host(kvm, host_pidr);
local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
/*
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 45/46] KVM: PPC: Book3S HV P9: implement hash host / hash guest support
@ 2021-03-23 1:03 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:03 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
This additionally has to save and restore the host SLB, and also
ensure that the MMU is off while switching into the guest SLB.
P9 and later CPUs now always go via the P9 path.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_64_entry.S | 6 +++++
arch/powerpc/kvm/book3s_hv.c | 4 ++-
arch/powerpc/kvm/book3s_hv_interrupt.c | 35 ++++++++++++++++++++++----
3 files changed, 39 insertions(+), 6 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index 845df5fefdbd..7cef6f5212d8 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -339,6 +339,12 @@ kvmppc_p9_exit_interrupt:
* effort for a small bit of code. Lots of other things to do first.
*/
kvmppc_p9_bad_interrupt:
+BEGIN_MMU_FTR_SECTION
+ /*
+ * Hash host doesn't try to recover MMU (requires host SLB reload)
+ */
+ b .
+END_MMU_FTR_SECTION_IFCLR(MMU_FTR_TYPE_RADIX)
/*
* Set GUEST_MODE_NONE so the handler won't branch to KVM, and clear
* MSR_RI in r12 ([H]SRR1) so the handler won't try to return.
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index c151a60c0daa..e732d65772b1 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -4462,7 +4462,7 @@ static int kvmppc_vcpu_run_hv(struct kvm_vcpu *vcpu)
vcpu->arch.state = KVMPPC_VCPU_BUSY_IN_HOST;
do {
- if (radix_enabled())
+ if (cpu_has_feature(CPU_FTR_ARCH_300))
r = kvmhv_run_single_vcpu(vcpu, ~(u64)0,
vcpu->arch.vcore->lpcr);
else
@@ -5543,6 +5543,8 @@ static int kvmhv_enable_nested(struct kvm *kvm)
return -EPERM;
if (!cpu_has_feature(CPU_FTR_ARCH_300))
return -ENODEV;
+ if (!radix_enabled())
+ return -ENODEV;
/* kvm = NULL means the caller is testing if the capability exists */
if (kvm)
diff --git a/arch/powerpc/kvm/book3s_hv_interrupt.c b/arch/powerpc/kvm/book3s_hv_interrupt.c
index 03fbfef708a8..bd95d05219b7 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupt.c
+++ b/arch/powerpc/kvm/book3s_hv_interrupt.c
@@ -149,7 +149,7 @@ static void switch_mmu_to_guest_hpt(struct kvm *kvm, struct kvm_vcpu *vcpu, u64
*/
}
-static void switch_mmu_to_host_radix(struct kvm *kvm, u32 pid)
+static void switch_mmu_to_host(struct kvm *kvm, u32 pid)
{
isync();
mtspr(SPRN_PID, pid);
@@ -158,6 +158,23 @@ static void switch_mmu_to_host_radix(struct kvm *kvm, u32 pid)
isync();
mtspr(SPRN_LPCR, kvm->arch.host_lpcr);
isync();
+
+ if (!radix_enabled())
+ slb_restore_bolted_realmode();
+}
+
+static void save_clear_host_mmu(struct kvm *kvm)
+{
+ if (!radix_enabled()) {
+ /*
+ * Hash host could save and restore host SLB entries to
+ * reduce SLB fault overheads of VM exits, but for now the
+ * existing code clears all entries and restores just the
+ * bolted ones when switching back to host.
+ */
+ mtslb(0, 0, 0);
+ slb_invalidate(6);
+ }
}
static void save_clear_guest_mmu(struct kvm *kvm, struct kvm_vcpu *vcpu)
@@ -292,15 +309,23 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
mtspr(SPRN_AMOR, ~0UL);
local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_GUEST_HV_FAST;
+
+ /*
+ * Hash host, hash guest, or radix guest with prefetch bug, all have
+ * to disable the MMU before switching to guest MMU state.
+ */
+ if (!radix_enabled() || !kvm_is_radix(kvm) ||
+ cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
+ __mtmsrd(msr & ~(MSR_IR|MSR_DR|MSR_RI), 0);
+
+ save_clear_host_mmu(kvm);
+
if (kvm_is_radix(kvm)) {
- if (cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
- __mtmsrd(msr & ~(MSR_IR|MSR_DR|MSR_RI), 0);
switch_mmu_to_guest_radix(kvm, vcpu, lpcr);
if (!cpu_has_feature(CPU_FTR_P9_RADIX_PREFETCH_BUG))
__mtmsrd(0, 1); /* clear RI */
} else {
- __mtmsrd(msr & ~(MSR_IR|MSR_DR|MSR_RI), 0);
switch_mmu_to_guest_hpt(kvm, vcpu, lpcr);
}
@@ -486,7 +511,7 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
mtspr(SPRN_HDEC, decrementer_max);
save_clear_guest_mmu(kvm, vcpu);
- switch_mmu_to_host_radix(kvm, host_pidr);
+ switch_mmu_to_host(kvm, host_pidr);
local_paca->kvm_hstate.in_guest = KVM_GUEST_MODE_NONE;
/*
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* [PATCH v4 46/46] KVM: PPC: Book3S HV: remove ISA v3.0 and v3.1 support from P7/8 path
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-23 1:03 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:03 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
POWER9 and later processors always go via the P9 guest entry path now.
Remove the remaining support from the P7/8 path.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 62 ++--
arch/powerpc/kvm/book3s_hv_interrupts.S | 9 +-
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 404 +-----------------------
arch/powerpc/platforms/powernv/idle.c | 52 +--
4 files changed, 39 insertions(+), 488 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index e732d65772b1..ad6a2501ab0f 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -130,9 +130,6 @@ static inline bool nesting_enabled(struct kvm *kvm)
return kvm->arch.nested_enable && kvm_is_radix(kvm);
}
-/* If set, the threads on each CPU core have to be in the same MMU mode */
-static bool no_mixing_hpt_and_radix __read_mostly;
-
static int kvmppc_hv_setup_htab_rma(struct kvm_vcpu *vcpu);
/*
@@ -3090,9 +3087,6 @@ static void prepare_threads(struct kvmppc_vcore *vc)
for_each_runnable_thread(i, vcpu, vc) {
if (signal_pending(vcpu->arch.run_task))
vcpu->arch.ret = -EINTR;
- else if (no_mixing_hpt_and_radix &&
- kvm_is_radix(vc->kvm) != radix_enabled())
- vcpu->arch.ret = -EINVAL;
else if (vcpu->arch.vpa.update_pending ||
vcpu->arch.slb_shadow.update_pending ||
vcpu->arch.dtl.update_pending)
@@ -3299,6 +3293,9 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
int trap;
bool is_power8;
+ if (WARN_ON_ONCE(cpu_has_feature(CPU_FTR_ARCH_300)))
+ return;
+
/*
* Remove from the list any threads that have a signal pending
* or need a VPA update done
@@ -3326,9 +3323,6 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
* Make sure we are running on primary threads, and that secondary
* threads are offline. Also check if the number of threads in this
* guest are greater than the current system threads per guest.
- * On POWER9, we need to be not in independent-threads mode if
- * this is a HPT guest on a radix host machine where the
- * CPU threads may not be in different MMU modes.
*/
if ((controlled_threads > 1) &&
((vc->num_threads > threads_per_subcore) || !on_primary_thread())) {
@@ -3352,18 +3346,6 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
if (vc->num_threads < target_threads)
collect_piggybacks(&core_info, target_threads);
- /*
- * On radix, arrange for TLB flushing if necessary.
- * This has to be done before disabling interrupts since
- * it uses smp_call_function().
- */
- pcpu = smp_processor_id();
- if (kvm_is_radix(vc->kvm)) {
- for (sub = 0; sub < core_info.n_subcores; ++sub)
- for_each_runnable_thread(i, vcpu, core_info.vc[sub])
- kvmppc_prepare_radix_vcpu(vcpu, pcpu);
- }
-
/*
* Hard-disable interrupts, and check resched flag and signals.
* If we need to reschedule or deliver a signal, clean up
@@ -3396,8 +3378,7 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
cmd_bit = stat_bit = 0;
split = core_info.n_subcores;
sip = NULL;
- is_power8 = cpu_has_feature(CPU_FTR_ARCH_207S)
- && !cpu_has_feature(CPU_FTR_ARCH_300);
+ is_power8 = cpu_has_feature(CPU_FTR_ARCH_207S);
if (split > 1) {
sip = &split_info;
@@ -3691,8 +3672,7 @@ static inline bool hcall_is_xics(unsigned long req)
}
/*
- * Virtual-mode guest entry for POWER9 and later when the host and
- * guest are both using the radix MMU. The LPIDR has already been set.
+ * Guest entry for POWER9 and later CPUs.
*/
static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
unsigned long lpcr)
@@ -5706,11 +5686,25 @@ static int kvmhv_enable_dawr1(struct kvm *kvm)
static bool kvmppc_hash_v3_possible(void)
{
- if (radix_enabled() && no_mixing_hpt_and_radix)
+ if (!cpu_has_feature(CPU_FTR_ARCH_300))
return false;
- return cpu_has_feature(CPU_FTR_ARCH_300) &&
- cpu_has_feature(CPU_FTR_HVMODE);
+ if (!cpu_has_feature(CPU_FTR_HVMODE))
+ return false;
+
+ /*
+ * POWER9 chips before version 2.02 can't have some threads in
+ * HPT mode and some in radix mode on the same core.
+ */
+ if (radix_enabled()) {
+ unsigned int pvr = mfspr(SPRN_PVR);
+ if ((pvr >> 16) == PVR_POWER9 &&
+ (((pvr & 0xe000) == 0 && (pvr & 0xfff) < 0x202) ||
+ ((pvr & 0xe000) == 0x2000 && (pvr & 0xfff) < 0x101)))
+ return false;
+ }
+
+ return true;
}
static struct kvmppc_ops kvm_ops_hv = {
@@ -5854,18 +5848,6 @@ static int kvmppc_book3s_init_hv(void)
if (kvmppc_radix_possible())
r = kvmppc_radix_init();
- /*
- * POWER9 chips before version 2.02 can't have some threads in
- * HPT mode and some in radix mode on the same core.
- */
- if (cpu_has_feature(CPU_FTR_ARCH_300)) {
- unsigned int pvr = mfspr(SPRN_PVR);
- if ((pvr >> 16) == PVR_POWER9 &&
- (((pvr & 0xe000) == 0 && (pvr & 0xfff) < 0x202) ||
- ((pvr & 0xe000) == 0x2000 && (pvr & 0xfff) < 0x101)))
- no_mixing_hpt_and_radix = true;
- }
-
r = kvmppc_uvmem_init();
if (r < 0)
pr_err("KVM-HV: kvmppc_uvmem_init failed %d\n", r);
diff --git a/arch/powerpc/kvm/book3s_hv_interrupts.S b/arch/powerpc/kvm/book3s_hv_interrupts.S
index 327417d79eac..4444f83cb133 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupts.S
+++ b/arch/powerpc/kvm/book3s_hv_interrupts.S
@@ -58,7 +58,7 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
/*
* Put whatever is in the decrementer into the
* hypervisor decrementer.
- * Because of a hardware deviation in P8 and P9,
+ * Because of a hardware deviation in P8,
* we need to set LPCR[HDICE] before writing HDEC.
*/
ld r5, HSTATE_KVM_VCORE(r13)
@@ -67,15 +67,10 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
ori r8, r9, LPCR_HDICE
mtspr SPRN_LPCR, r8
isync
- andis. r0, r9, LPCR_LD@h
mfspr r8,SPRN_DEC
mftb r7
-BEGIN_FTR_SECTION
- /* On POWER9, don't sign-extend if host LPCR[LD] bit is set */
- bne 32f
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
extsw r8,r8
-32: mtspr SPRN_HDEC,r8
+ mtspr SPRN_HDEC,r8
add r8,r8,r7
std r8,HSTATE_DECEXP(r13)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 5cbc26ebcf8e..b2a71a1b48fe 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -25,7 +25,6 @@
#include <asm/export.h>
#include <asm/tm.h>
#include <asm/opal.h>
-#include <asm/xive-regs.h>
#include <asm/thread_info.h>
#include <asm/asm-compat.h>
#include <asm/feature-fixups.h>
@@ -34,9 +33,7 @@
/* Sign-extend HDEC if not on POWER9 */
#define EXTEND_HDEC(reg) \
-BEGIN_FTR_SECTION; \
- extsw reg, reg; \
-END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
+ extsw reg, reg
/* Values in HSTATE_NAPPING(r13) */
#define NAPPING_CEDE 1
@@ -56,8 +53,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
#define STACK_SLOT_HFSCR (SFS-72)
#define STACK_SLOT_AMR (SFS-80)
#define STACK_SLOT_UAMOR (SFS-88)
-#define STACK_SLOT_DAWR1 (SFS-96)
-#define STACK_SLOT_DAWRX1 (SFS-104)
/*
* Call kvmppc_hv_entry in real mode.
@@ -333,7 +328,6 @@ kvm_secondary_got_guest:
LOAD_REG_ADDR(r6, decrementer_max)
ld r6, 0(r6)
mtspr SPRN_HDEC, r6
-BEGIN_FTR_SECTION
/* and set per-LPAR registers, if doing dynamic micro-threading */
ld r6, HSTATE_SPLIT_MODE(r13)
cmpdi r6, 0
@@ -345,7 +339,6 @@ BEGIN_FTR_SECTION
ld r0, KVM_SPLIT_LDBAR(r6)
mtspr SPRN_LDBAR, r0
isync
-END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
63:
/* Order load of vcpu after load of vcore */
lwsync
@@ -416,7 +409,6 @@ kvm_no_guest:
blr
53:
-BEGIN_FTR_SECTION
HMT_LOW
ld r5, HSTATE_KVM_VCORE(r13)
cmpdi r5, 0
@@ -431,14 +423,6 @@ BEGIN_FTR_SECTION
b kvm_unsplit_nap
60: HMT_MEDIUM
b kvm_secondary_got_guest
-FTR_SECTION_ELSE
- HMT_LOW
- ld r5, HSTATE_KVM_VCORE(r13)
- cmpdi r5, 0
- beq kvm_no_guest
- HMT_MEDIUM
- b kvm_secondary_got_guest
-ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
54: li r0, KVM_HWTHREAD_IN_KVM
stb r0, HSTATE_HWTHREAD_STATE(r13)
@@ -564,13 +548,11 @@ kvmppc_hv_entry:
bne 10f
lwz r7,KVM_LPID(r9)
-BEGIN_FTR_SECTION
ld r6,KVM_SDR1(r9)
li r0,LPID_RSVD /* switch to reserved LPID */
mtspr SPRN_LPID,r0
ptesync
mtspr SPRN_SDR1,r6 /* switch to partition page table */
-END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
mtspr SPRN_LPID,r7
isync
@@ -650,16 +632,6 @@ kvmppc_got_guest:
mtspr SPRN_SPURR,r8
/* Save host values of some registers */
-BEGIN_FTR_SECTION
- mfspr r5, SPRN_TIDR
- mfspr r6, SPRN_PSSCR
- mfspr r7, SPRN_PID
- std r5, STACK_SLOT_TID(r1)
- std r6, STACK_SLOT_PSSCR(r1)
- std r7, STACK_SLOT_PID(r1)
- mfspr r5, SPRN_HFSCR
- std r5, STACK_SLOT_HFSCR(r1)
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
BEGIN_FTR_SECTION
mfspr r5, SPRN_CIABR
mfspr r6, SPRN_DAWR0
@@ -670,12 +642,6 @@ BEGIN_FTR_SECTION
std r7, STACK_SLOT_DAWRX0(r1)
std r8, STACK_SLOT_IAMR(r1)
END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
-BEGIN_FTR_SECTION
- mfspr r6, SPRN_DAWR1
- mfspr r7, SPRN_DAWRX1
- std r6, STACK_SLOT_DAWR1(r1)
- std r7, STACK_SLOT_DAWRX1(r1)
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S | CPU_FTR_DAWR1)
mfspr r5, SPRN_AMR
std r5, STACK_SLOT_AMR(r1)
@@ -693,13 +659,9 @@ BEGIN_FTR_SECTION
END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
-/*
- * Branch around the call if both CPU_FTR_TM and
- * CPU_FTR_P9_TM_HV_ASSIST are off.
- */
BEGIN_FTR_SECTION
b 91f
-END_FTR_SECTION(CPU_FTR_TM | CPU_FTR_P9_TM_HV_ASSIST, 0)
+END_FTR_SECTION_IFCLR(CPU_FTR_TM)
/*
* NOTE THAT THIS TRASHES ALL NON-VOLATILE REGISTERS (but not CR)
*/
@@ -766,12 +728,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
ld r6, VCPU_DAWRX0(r4)
mtspr SPRN_DAWR0, r5
mtspr SPRN_DAWRX0, r6
-BEGIN_FTR_SECTION
- ld r5, VCPU_DAWR1(r4)
- ld r6, VCPU_DAWRX1(r4)
- mtspr SPRN_DAWR1, r5
- mtspr SPRN_DAWRX1, r6
-END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
1:
ld r7, VCPU_CIABR(r4)
ld r8, VCPU_TAR(r4)
@@ -789,7 +745,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
mtspr SPRN_BESCR, r6
mtspr SPRN_PID, r7
mtspr SPRN_WORT, r8
-BEGIN_FTR_SECTION
/* POWER8-only registers */
ld r5, VCPU_TCSCR(r4)
ld r6, VCPU_ACOP(r4)
@@ -800,18 +755,6 @@ BEGIN_FTR_SECTION
mtspr SPRN_CSIGR, r7
mtspr SPRN_TACR, r8
nop
-FTR_SECTION_ELSE
- /* POWER9-only registers */
- ld r5, VCPU_TID(r4)
- ld r6, VCPU_PSSCR(r4)
- lbz r8, HSTATE_FAKE_SUSPEND(r13)
- oris r6, r6, PSSCR_EC@h /* This makes stop trap to HV */
- rldimi r6, r8, PSSCR_FAKE_SUSPEND_LG, 63 - PSSCR_FAKE_SUSPEND_LG
- ld r7, VCPU_HFSCR(r4)
- mtspr SPRN_TIDR, r5
- mtspr SPRN_PSSCR, r6
- mtspr SPRN_HFSCR, r7
-ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
8:
ld r5, VCPU_SPRG0(r4)
@@ -904,93 +847,9 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
bdnz 1b
9:
-#ifdef CONFIG_KVM_XICS
- /* We are entering the guest on that thread, push VCPU to XIVE */
- ld r11, VCPU_XIVE_SAVED_STATE(r4)
- li r9, TM_QW1_OS
- lwz r8, VCPU_XIVE_CAM_WORD(r4)
- cmpwi r8, 0
- beq no_xive
- li r7, TM_QW1_OS + TM_WORD2
- mfmsr r0
- andi. r0, r0, MSR_DR /* in real mode? */
- beq 2f
- ld r10, HSTATE_XIVE_TIMA_VIRT(r13)
- cmpldi cr1, r10, 0
- beq cr1, no_xive
- eieio
- stdx r11,r9,r10
- stwx r8,r7,r10
- b 3f
-2: ld r10, HSTATE_XIVE_TIMA_PHYS(r13)
- cmpldi cr1, r10, 0
- beq cr1, no_xive
- eieio
- stdcix r11,r9,r10
- stwcix r8,r7,r10
-3: li r9, 1
- stb r9, VCPU_XIVE_PUSHED(r4)
- eieio
-
- /*
- * We clear the irq_pending flag. There is a small chance of a
- * race vs. the escalation interrupt happening on another
- * processor setting it again, but the only consequence is to
- * cause a spurrious wakeup on the next H_CEDE which is not an
- * issue.
- */
- li r0,0
- stb r0, VCPU_IRQ_PENDING(r4)
-
- /*
- * In single escalation mode, if the escalation interrupt is
- * on, we mask it.
- */
- lbz r0, VCPU_XIVE_ESC_ON(r4)
- cmpwi cr1, r0,0
- beq cr1, 1f
- li r9, XIVE_ESB_SET_PQ_01
- beq 4f /* in real mode? */
- ld r10, VCPU_XIVE_ESC_VADDR(r4)
- ldx r0, r10, r9
- b 5f
-4: ld r10, VCPU_XIVE_ESC_RADDR(r4)
- ldcix r0, r10, r9
-5: sync
-
- /* We have a possible subtle race here: The escalation interrupt might
- * have fired and be on its way to the host queue while we mask it,
- * and if we unmask it early enough (re-cede right away), there is
- * a theorical possibility that it fires again, thus landing in the
- * target queue more than once which is a big no-no.
- *
- * Fortunately, solving this is rather easy. If the above load setting
- * PQ to 01 returns a previous value where P is set, then we know the
- * escalation interrupt is somewhere on its way to the host. In that
- * case we simply don't clear the xive_esc_on flag below. It will be
- * eventually cleared by the handler for the escalation interrupt.
- *
- * Then, when doing a cede, we check that flag again before re-enabling
- * the escalation interrupt, and if set, we abort the cede.
- */
- andi. r0, r0, XIVE_ESB_VAL_P
- bne- 1f
-
- /* Now P is 0, we can clear the flag */
- li r0, 0
- stb r0, VCPU_XIVE_ESC_ON(r4)
-1:
-no_xive:
-#endif /* CONFIG_KVM_XICS */
-
deliver_guest_interrupt: /* r4 = vcpu, r13 = paca */
/* Check if we can deliver an external or decrementer interrupt now */
ld r0, VCPU_PENDING_EXC(r4)
-BEGIN_FTR_SECTION
- /* On POWER9, also check for emulated doorbell interrupt */
- lbz r3, VCPU_DBELL_REQ(r4)
- or r0, r0, r3
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
cmpdi r0, 0
beq 71f
mr r3, r4
@@ -1063,12 +922,6 @@ BEGIN_FTR_SECTION
mtspr SPRN_PPR, r0
END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
-/* Move canary into DSISR to check for later */
-BEGIN_FTR_SECTION
- li r0, 0x7fff
- mtspr SPRN_HDSISR, r0
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
-
ld r6, VCPU_KVM(r4)
lbz r7, KVM_SECURE_GUEST(r6)
cmpdi r7, 0
@@ -1284,14 +1137,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
/* Hypervisor doorbell - exit only if host IPI flag set */
cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
bne 3f
-BEGIN_FTR_SECTION
- PPC_MSGSYNC
- lwsync
- /* always exit if we're running a nested guest */
- ld r0, VCPU_NESTED(r9)
- cmpdi r0, 0
- bne guest_exit_cont
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
lbz r0, HSTATE_HOST_IPI(r13)
cmpwi r0, 0
beq maybe_reenter_guest
@@ -1321,43 +1166,6 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
mr r4, r9
bl kvmhv_accumulate_time
#endif
-#ifdef CONFIG_KVM_XICS
- /* We are exiting, pull the VP from the XIVE */
- lbz r0, VCPU_XIVE_PUSHED(r9)
- cmpwi cr0, r0, 0
- beq 1f
- li r7, TM_SPC_PULL_OS_CTX
- li r6, TM_QW1_OS
- mfmsr r0
- andi. r0, r0, MSR_DR /* in real mode? */
- beq 2f
- ld r10, HSTATE_XIVE_TIMA_VIRT(r13)
- cmpldi cr0, r10, 0
- beq 1f
- /* First load to pull the context, we ignore the value */
- eieio
- lwzx r11, r7, r10
- /* Second load to recover the context state (Words 0 and 1) */
- ldx r11, r6, r10
- b 3f
-2: ld r10, HSTATE_XIVE_TIMA_PHYS(r13)
- cmpldi cr0, r10, 0
- beq 1f
- /* First load to pull the context, we ignore the value */
- eieio
- lwzcix r11, r7, r10
- /* Second load to recover the context state (Words 0 and 1) */
- ldcix r11, r6, r10
-3: std r11, VCPU_XIVE_SAVED_STATE(r9)
- /* Fixup some of the state for the next load */
- li r10, 0
- li r0, 0xff
- stb r10, VCPU_XIVE_PUSHED(r9)
- stb r10, (VCPU_XIVE_SAVED_STATE+3)(r9)
- stb r0, (VCPU_XIVE_SAVED_STATE+4)(r9)
- eieio
-1:
-#endif /* CONFIG_KVM_XICS */
/*
* Possibly flush the link stack here, before we do a blr in
@@ -1412,12 +1220,6 @@ guest_bypass:
ld r3, HSTATE_KVM_VCORE(r13)
mfspr r5,SPRN_DEC
mftb r6
- /* On P9, if the guest has large decr enabled, don't sign extend */
-BEGIN_FTR_SECTION
- ld r4, VCORE_LPCR(r3)
- andis. r4, r4, LPCR_LD@h
- bne 16f
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
extsw r5,r5
16: add r5,r5,r6
/* r5 is a guest timebase value here, convert to host TB */
@@ -1491,7 +1293,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
std r6, VCPU_BESCR(r9)
stw r7, VCPU_GUEST_PID(r9)
std r8, VCPU_WORT(r9)
-BEGIN_FTR_SECTION
mfspr r5, SPRN_TCSCR
mfspr r6, SPRN_ACOP
mfspr r7, SPRN_CSIGR
@@ -1500,17 +1301,6 @@ BEGIN_FTR_SECTION
std r6, VCPU_ACOP(r9)
std r7, VCPU_CSIGR(r9)
std r8, VCPU_TACR(r9)
-FTR_SECTION_ELSE
- mfspr r5, SPRN_TIDR
- mfspr r6, SPRN_PSSCR
- std r5, VCPU_TID(r9)
- rldicl r6, r6, 4, 50 /* r6 &= PSSCR_GUEST_VIS */
- rotldi r6, r6, 60
- std r6, VCPU_PSSCR(r9)
- /* Restore host HFSCR value */
- ld r7, STACK_SLOT_HFSCR(r1)
- mtspr SPRN_HFSCR, r7
-ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
/*
* Restore various registers to 0, where non-zero values
* set by the guest could disrupt the host.
@@ -1518,13 +1308,11 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
li r0, 0
mtspr SPRN_PSPB, r0
mtspr SPRN_WORT, r0
-BEGIN_FTR_SECTION
mtspr SPRN_TCSCR, r0
/* Set MMCRS to 1<<31 to freeze and disable the SPMC counters */
li r0, 1
sldi r0, r0, 31
mtspr SPRN_MMCRS, r0
-END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
/* Save and restore AMR, IAMR and UAMOR before turning on the MMU */
ld r8, STACK_SLOT_IAMR(r1)
@@ -1581,13 +1369,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
bl kvmppc_save_fp
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
-/*
- * Branch around the call if both CPU_FTR_TM and
- * CPU_FTR_P9_TM_HV_ASSIST are off.
- */
BEGIN_FTR_SECTION
b 91f
-END_FTR_SECTION(CPU_FTR_TM | CPU_FTR_P9_TM_HV_ASSIST, 0)
+END_FTR_SECTION_IFCLR(CPU_FTR_TM)
/*
* NOTE THAT THIS TRASHES ALL NON-VOLATILE REGISTERS (but not CR)
*/
@@ -1633,28 +1417,6 @@ BEGIN_FTR_SECTION
mtspr SPRN_DAWR0, r6
mtspr SPRN_DAWRX0, r7
END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
-BEGIN_FTR_SECTION
- ld r6, STACK_SLOT_DAWR1(r1)
- ld r7, STACK_SLOT_DAWRX1(r1)
- mtspr SPRN_DAWR1, r6
- mtspr SPRN_DAWRX1, r7
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S | CPU_FTR_DAWR1)
-BEGIN_FTR_SECTION
- ld r5, STACK_SLOT_TID(r1)
- ld r6, STACK_SLOT_PSSCR(r1)
- ld r7, STACK_SLOT_PID(r1)
- mtspr SPRN_TIDR, r5
- mtspr SPRN_PSSCR, r6
- mtspr SPRN_PID, r7
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
-
- /*
- * cp_abort is required if the processor supports local copy-paste
- * to clear the copy buffer that was under control of the guest.
- */
-BEGIN_FTR_SECTION
- PPC_CP_ABORT
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_31)
/*
* POWER7/POWER8 guest -> host partition switch code.
@@ -1691,13 +1453,11 @@ kvmhv_switch_to_host:
/* Primary thread switches back to host partition */
lwz r7,KVM_HOST_LPID(r4)
-BEGIN_FTR_SECTION
ld r6,KVM_HOST_SDR1(r4)
li r8,LPID_RSVD /* switch to reserved LPID */
mtspr SPRN_LPID,r8
ptesync
mtspr SPRN_SDR1,r6 /* switch to host page table */
-END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
mtspr SPRN_LPID,r7
isync
@@ -1908,20 +1668,11 @@ kvmppc_tm_emul:
kvmppc_hdsi:
mfspr r4, SPRN_HDAR
mfspr r6, SPRN_HDSISR
-BEGIN_FTR_SECTION
- /* Look for DSISR canary. If we find it, retry instruction */
- cmpdi r6, 0x7fff
- beq 6f
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
/* HPTE not found fault or protection fault? */
andis. r0, r6, (DSISR_NOHPTE | DSISR_PROTFAULT)@h
beq 1f /* if not, send it to the guest */
andi. r0, r11, MSR_DR /* data relocation enabled? */
beq 3f
-BEGIN_FTR_SECTION
- mfspr r5, SPRN_ASDR /* on POWER9, use ASDR to get VSID */
- b 4f
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
clrrdi r0, r4, 28
PPC_SLBFEE_DOT(R5, R0) /* if so, look up SLB */
li r0, BOOK3S_INTERRUPT_DATA_SEGMENT
@@ -1998,10 +1749,6 @@ kvmppc_hisi:
beq 1f
andi. r0, r11, MSR_IR /* instruction relocation enabled? */
beq 3f
-BEGIN_FTR_SECTION
- mfspr r5, SPRN_ASDR /* on POWER9, use ASDR to get VSID */
- b 4f
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
clrrdi r0, r10, 28
PPC_SLBFEE_DOT(R5, R0) /* if so, look up SLB */
li r0, BOOK3S_INTERRUPT_INST_SEGMENT
@@ -2049,10 +1796,6 @@ hcall_try_real_mode:
andi. r0,r11,MSR_PR
/* sc 1 from userspace - reflect to guest syscall */
bne sc_1_fast_return
- /* sc 1 from nested guest - give it to L1 to handle */
- ld r0, VCPU_NESTED(r9)
- cmpdi r0, 0
- bne guest_exit_cont
clrrdi r3,r3,2
cmpldi r3,hcall_real_table_end - hcall_real_table
bge guest_exit_cont
@@ -2448,13 +2191,9 @@ _GLOBAL(kvmppc_h_cede) /* r3 = vcpu pointer, r11 = msr, r13 = paca */
bl kvmppc_save_fp
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
-/*
- * Branch around the call if both CPU_FTR_TM and
- * CPU_FTR_P9_TM_HV_ASSIST are off.
- */
BEGIN_FTR_SECTION
b 91f
-END_FTR_SECTION(CPU_FTR_TM | CPU_FTR_P9_TM_HV_ASSIST, 0)
+END_FTR_SECTION_IFCLR(CPU_FTR_TM)
/*
* NOTE THAT THIS TRASHES ALL NON-VOLATILE REGISTERS (but not CR)
*/
@@ -2474,15 +2213,8 @@ END_FTR_SECTION(CPU_FTR_TM | CPU_FTR_P9_TM_HV_ASSIST, 0)
mfspr r3, SPRN_DEC
mfspr r4, SPRN_HDEC
mftb r5
-BEGIN_FTR_SECTION
- /* On P9 check whether the guest has large decrementer mode enabled */
- ld r6, HSTATE_KVM_VCORE(r13)
- ld r6, VCORE_LPCR(r6)
- andis. r6, r6, LPCR_LD@h
- bne 68f
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
extsw r3, r3
-68: EXTEND_HDEC(r4)
+ EXTEND_HDEC(r4)
cmpd r3, r4
ble 67f
mtspr SPRN_DEC, r4
@@ -2527,28 +2259,14 @@ BEGIN_FTR_SECTION
END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
kvm_nap_sequence: /* desired LPCR value in r5 */
-BEGIN_FTR_SECTION
- /*
- * PSSCR bits: exit criterion = 1 (wakeup based on LPCR at sreset)
- * enable state loss = 1 (allow SMT mode switch)
- * requested level = 0 (just stop dispatching)
- */
- lis r3, (PSSCR_EC | PSSCR_ESL)@h
/* Set LPCR_PECE_HVEE bit to enable wakeup by HV interrupts */
li r4, LPCR_PECE_HVEE@higher
sldi r4, r4, 32
or r5, r5, r4
-FTR_SECTION_ELSE
- li r3, PNV_THREAD_NAP
-ALT_FTR_SECTION_END_IFSET(CPU_FTR_ARCH_300)
mtspr SPRN_LPCR,r5
isync
-BEGIN_FTR_SECTION
- bl isa300_idle_stop_mayloss
-FTR_SECTION_ELSE
bl isa206_idle_insn_mayloss
-ALT_FTR_SECTION_END_IFSET(CPU_FTR_ARCH_300)
mfspr r0, SPRN_CTRLF
ori r0, r0, 1
@@ -2567,10 +2285,8 @@ ALT_FTR_SECTION_END_IFSET(CPU_FTR_ARCH_300)
beq kvm_end_cede
cmpwi r0, NAPPING_NOVCPU
beq kvm_novcpu_wakeup
-BEGIN_FTR_SECTION
cmpwi r0, NAPPING_UNSPLIT
beq kvm_unsplit_wakeup
-END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
twi 31,0,0 /* Nap state must not be zero */
33: mr r4, r3
@@ -2590,13 +2306,9 @@ kvm_end_cede:
#endif
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
-/*
- * Branch around the call if both CPU_FTR_TM and
- * CPU_FTR_P9_TM_HV_ASSIST are off.
- */
BEGIN_FTR_SECTION
b 91f
-END_FTR_SECTION(CPU_FTR_TM | CPU_FTR_P9_TM_HV_ASSIST, 0)
+END_FTR_SECTION_IFCLR(CPU_FTR_TM)
/*
* NOTE THAT THIS TRASHES ALL NON-VOLATILE REGISTERS (but not CR)
*/
@@ -2686,47 +2398,7 @@ kvm_cede_prodded:
/* we've ceded but we want to give control to the host */
kvm_cede_exit:
ld r9, HSTATE_KVM_VCPU(r13)
-#ifdef CONFIG_KVM_XICS
- /* are we using XIVE with single escalation? */
- ld r10, VCPU_XIVE_ESC_VADDR(r9)
- cmpdi r10, 0
- beq 3f
- li r6, XIVE_ESB_SET_PQ_00
- /*
- * If we still have a pending escalation, abort the cede,
- * and we must set PQ to 10 rather than 00 so that we don't
- * potentially end up with two entries for the escalation
- * interrupt in the XIVE interrupt queue. In that case
- * we also don't want to set xive_esc_on to 1 here in
- * case we race with xive_esc_irq().
- */
- lbz r5, VCPU_XIVE_ESC_ON(r9)
- cmpwi r5, 0
- beq 4f
- li r0, 0
- stb r0, VCPU_CEDED(r9)
- /*
- * The escalation interrupts are special as we don't EOI them.
- * There is no need to use the load-after-store ordering offset
- * to set PQ to 10 as we won't use StoreEOI.
- */
- li r6, XIVE_ESB_SET_PQ_10
- b 5f
-4: li r0, 1
- stb r0, VCPU_XIVE_ESC_ON(r9)
- /* make sure store to xive_esc_on is seen before xive_esc_irq runs */
- sync
-5: /* Enable XIVE escalation */
- mfmsr r0
- andi. r0, r0, MSR_DR /* in real mode? */
- beq 1f
- ldx r0, r10, r6
- b 2f
-1: ld r10, VCPU_XIVE_ESC_RADDR(r9)
- ldcix r0, r10, r6
-2: sync
-#endif /* CONFIG_KVM_XICS */
-3: b guest_exit_cont
+ b guest_exit_cont
/* Try to do machine check recovery in real mode */
machine_check_realmode:
@@ -2803,10 +2475,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
PPC_MSGCLR(6)
/* see if it's a host IPI */
li r3, 1
-BEGIN_FTR_SECTION
- PPC_MSGSYNC
- lwsync
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
lbz r0, HSTATE_HOST_IPI(r13)
cmpwi r0, 0
bnelr
@@ -3115,70 +2783,12 @@ kvmppc_bad_host_intr:
std r3, STACK_FRAME_OVERHEAD-16(r1)
/*
- * On POWER9 do a minimal restore of the MMU and call C code,
- * which will print a message and panic.
* XXX On POWER7 and POWER8, we just spin here since we don't
* know what the other threads are doing (and we don't want to
* coordinate with them) - but at least we now have register state
* in memory that we might be able to look at from another CPU.
*/
-BEGIN_FTR_SECTION
b .
-END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
- ld r9, HSTATE_KVM_VCPU(r13)
- ld r10, VCPU_KVM(r9)
-
- li r0, 0
- mtspr SPRN_AMR, r0
- mtspr SPRN_IAMR, r0
- mtspr SPRN_CIABR, r0
- mtspr SPRN_DAWRX0, r0
-BEGIN_FTR_SECTION
- mtspr SPRN_DAWRX1, r0
-END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
-
- /* Clear guest SLB. */
- slbmte r0, r0
- PPC_SLBIA(6)
- ptesync
-
- /* load host SLB entries */
- ld r8, PACA_SLBSHADOWPTR(r13)
- .rept SLB_NUM_BOLTED
- li r3, SLBSHADOW_SAVEAREA
- LDX_BE r5, r8, r3
- addi r3, r3, 8
- LDX_BE r6, r8, r3
- andis. r7, r5, SLB_ESID_V@h
- beq 3f
- slbmte r6, r5
-3: addi r8, r8, 16
- .endr
-
- lwz r7, KVM_HOST_LPID(r10)
- mtspr SPRN_LPID, r7
- mtspr SPRN_PID, r0
- ld r8, KVM_HOST_LPCR(r10)
- mtspr SPRN_LPCR, r8
- isync
- li r0, KVM_GUEST_MODE_NONE
- stb r0, HSTATE_IN_GUEST(r13)
-
- /*
- * Turn on the MMU and jump to C code
- */
- bcl 20, 31, .+4
-5: mflr r3
- addi r3, r3, 9f - 5b
- li r4, -1
- rldimi r3, r4, 62, 0 /* ensure 0xc000000000000000 bits are set */
- ld r4, PACAKMSR(r13)
- mtspr SPRN_SRR0, r3
- mtspr SPRN_SRR1, r4
- RFI_TO_KERNEL
-9: addi r3, r1, STACK_FRAME_OVERHEAD
- bl kvmppc_bad_interrupt
- b 9b
/*
* This mimics the MSR transition on IRQ delivery. The new guest MSR is taken
diff --git a/arch/powerpc/platforms/powernv/idle.c b/arch/powerpc/platforms/powernv/idle.c
index 999997d9e9a9..528a7e0cf83a 100644
--- a/arch/powerpc/platforms/powernv/idle.c
+++ b/arch/powerpc/platforms/powernv/idle.c
@@ -604,7 +604,7 @@ struct p9_sprs {
u64 uamor;
};
-static unsigned long power9_idle_stop(unsigned long psscr, bool mmu_on)
+static unsigned long power9_idle_stop(unsigned long psscr)
{
int cpu = raw_smp_processor_id();
int first = cpu_first_thread_sibling(cpu);
@@ -620,8 +620,6 @@ static unsigned long power9_idle_stop(unsigned long psscr, bool mmu_on)
if (!(psscr & (PSSCR_EC|PSSCR_ESL))) {
/* EC=ESL=0 case */
- BUG_ON(!mmu_on);
-
/*
* Wake synchronously. SRESET via xscom may still cause
* a 0x100 powersave wakeup with SRR1 reason!
@@ -803,8 +801,7 @@ static unsigned long power9_idle_stop(unsigned long psscr, bool mmu_on)
__slb_restore_bolted_realmode();
out:
- if (mmu_on)
- mtmsr(MSR_KERNEL);
+ mtmsr(MSR_KERNEL);
return srr1;
}
@@ -895,7 +892,7 @@ struct p10_sprs {
*/
};
-static unsigned long power10_idle_stop(unsigned long psscr, bool mmu_on)
+static unsigned long power10_idle_stop(unsigned long psscr)
{
int cpu = raw_smp_processor_id();
int first = cpu_first_thread_sibling(cpu);
@@ -909,8 +906,6 @@ static unsigned long power10_idle_stop(unsigned long psscr, bool mmu_on)
if (!(psscr & (PSSCR_EC|PSSCR_ESL))) {
/* EC=ESL=0 case */
- BUG_ON(!mmu_on);
-
/*
* Wake synchronously. SRESET via xscom may still cause
* a 0x100 powersave wakeup with SRR1 reason!
@@ -991,8 +986,7 @@ static unsigned long power10_idle_stop(unsigned long psscr, bool mmu_on)
__slb_restore_bolted_realmode();
out:
- if (mmu_on)
- mtmsr(MSR_KERNEL);
+ mtmsr(MSR_KERNEL);
return srr1;
}
@@ -1002,40 +996,10 @@ static unsigned long arch300_offline_stop(unsigned long psscr)
{
unsigned long srr1;
-#ifndef CONFIG_KVM_BOOK3S_HV_POSSIBLE
- __ppc64_runlatch_off();
if (cpu_has_feature(CPU_FTR_ARCH_31))
- srr1 = power10_idle_stop(psscr, true);
+ srr1 = power10_idle_stop(psscr);
else
- srr1 = power9_idle_stop(psscr, true);
- __ppc64_runlatch_on();
-#else
- /*
- * Tell KVM we're entering idle.
- * This does not have to be done in real mode because the P9 MMU
- * is independent per-thread. Some steppings share radix/hash mode
- * between threads, but in that case KVM has a barrier sync in real
- * mode before and after switching between radix and hash.
- *
- * kvm_start_guest must still be called in real mode though, hence
- * the false argument.
- */
- local_paca->kvm_hstate.hwthread_state = KVM_HWTHREAD_IN_IDLE;
-
- __ppc64_runlatch_off();
- if (cpu_has_feature(CPU_FTR_ARCH_31))
- srr1 = power10_idle_stop(psscr, false);
- else
- srr1 = power9_idle_stop(psscr, false);
- __ppc64_runlatch_on();
-
- local_paca->kvm_hstate.hwthread_state = KVM_HWTHREAD_IN_KERNEL;
- /* Order setting hwthread_state vs. testing hwthread_req */
- smp_mb();
- if (local_paca->kvm_hstate.hwthread_req)
- srr1 = idle_kvm_start_guest(srr1);
- mtmsr(MSR_KERNEL);
-#endif
+ srr1 = power9_idle_stop(psscr);
return srr1;
}
@@ -1055,9 +1019,9 @@ void arch300_idle_type(unsigned long stop_psscr_val,
__ppc64_runlatch_off();
if (cpu_has_feature(CPU_FTR_ARCH_31))
- srr1 = power10_idle_stop(psscr, true);
+ srr1 = power10_idle_stop(psscr);
else
- srr1 = power9_idle_stop(psscr, true);
+ srr1 = power9_idle_stop(psscr);
__ppc64_runlatch_on();
fini_irq_for_idle_irqsoff();
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread* [PATCH v4 46/46] KVM: PPC: Book3S HV: remove ISA v3.0 and v3.1 support from P7/8 path
@ 2021-03-23 1:03 ` Nicholas Piggin
0 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-23 1:03 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev, Nicholas Piggin
POWER9 and later processors always go via the P9 guest entry path now.
Remove the remaining support from the P7/8 path.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/kvm/book3s_hv.c | 62 ++--
arch/powerpc/kvm/book3s_hv_interrupts.S | 9 +-
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 404 +-----------------------
arch/powerpc/platforms/powernv/idle.c | 52 +--
4 files changed, 39 insertions(+), 488 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index e732d65772b1..ad6a2501ab0f 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -130,9 +130,6 @@ static inline bool nesting_enabled(struct kvm *kvm)
return kvm->arch.nested_enable && kvm_is_radix(kvm);
}
-/* If set, the threads on each CPU core have to be in the same MMU mode */
-static bool no_mixing_hpt_and_radix __read_mostly;
-
static int kvmppc_hv_setup_htab_rma(struct kvm_vcpu *vcpu);
/*
@@ -3090,9 +3087,6 @@ static void prepare_threads(struct kvmppc_vcore *vc)
for_each_runnable_thread(i, vcpu, vc) {
if (signal_pending(vcpu->arch.run_task))
vcpu->arch.ret = -EINTR;
- else if (no_mixing_hpt_and_radix &&
- kvm_is_radix(vc->kvm) != radix_enabled())
- vcpu->arch.ret = -EINVAL;
else if (vcpu->arch.vpa.update_pending ||
vcpu->arch.slb_shadow.update_pending ||
vcpu->arch.dtl.update_pending)
@@ -3299,6 +3293,9 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
int trap;
bool is_power8;
+ if (WARN_ON_ONCE(cpu_has_feature(CPU_FTR_ARCH_300)))
+ return;
+
/*
* Remove from the list any threads that have a signal pending
* or need a VPA update done
@@ -3326,9 +3323,6 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
* Make sure we are running on primary threads, and that secondary
* threads are offline. Also check if the number of threads in this
* guest are greater than the current system threads per guest.
- * On POWER9, we need to be not in independent-threads mode if
- * this is a HPT guest on a radix host machine where the
- * CPU threads may not be in different MMU modes.
*/
if ((controlled_threads > 1) &&
((vc->num_threads > threads_per_subcore) || !on_primary_thread())) {
@@ -3352,18 +3346,6 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
if (vc->num_threads < target_threads)
collect_piggybacks(&core_info, target_threads);
- /*
- * On radix, arrange for TLB flushing if necessary.
- * This has to be done before disabling interrupts since
- * it uses smp_call_function().
- */
- pcpu = smp_processor_id();
- if (kvm_is_radix(vc->kvm)) {
- for (sub = 0; sub < core_info.n_subcores; ++sub)
- for_each_runnable_thread(i, vcpu, core_info.vc[sub])
- kvmppc_prepare_radix_vcpu(vcpu, pcpu);
- }
-
/*
* Hard-disable interrupts, and check resched flag and signals.
* If we need to reschedule or deliver a signal, clean up
@@ -3396,8 +3378,7 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
cmd_bit = stat_bit = 0;
split = core_info.n_subcores;
sip = NULL;
- is_power8 = cpu_has_feature(CPU_FTR_ARCH_207S)
- && !cpu_has_feature(CPU_FTR_ARCH_300);
+ is_power8 = cpu_has_feature(CPU_FTR_ARCH_207S);
if (split > 1) {
sip = &split_info;
@@ -3691,8 +3672,7 @@ static inline bool hcall_is_xics(unsigned long req)
}
/*
- * Virtual-mode guest entry for POWER9 and later when the host and
- * guest are both using the radix MMU. The LPIDR has already been set.
+ * Guest entry for POWER9 and later CPUs.
*/
static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
unsigned long lpcr)
@@ -5706,11 +5686,25 @@ static int kvmhv_enable_dawr1(struct kvm *kvm)
static bool kvmppc_hash_v3_possible(void)
{
- if (radix_enabled() && no_mixing_hpt_and_radix)
+ if (!cpu_has_feature(CPU_FTR_ARCH_300))
return false;
- return cpu_has_feature(CPU_FTR_ARCH_300) &&
- cpu_has_feature(CPU_FTR_HVMODE);
+ if (!cpu_has_feature(CPU_FTR_HVMODE))
+ return false;
+
+ /*
+ * POWER9 chips before version 2.02 can't have some threads in
+ * HPT mode and some in radix mode on the same core.
+ */
+ if (radix_enabled()) {
+ unsigned int pvr = mfspr(SPRN_PVR);
+ if ((pvr >> 16) = PVR_POWER9 &&
+ (((pvr & 0xe000) = 0 && (pvr & 0xfff) < 0x202) ||
+ ((pvr & 0xe000) = 0x2000 && (pvr & 0xfff) < 0x101)))
+ return false;
+ }
+
+ return true;
}
static struct kvmppc_ops kvm_ops_hv = {
@@ -5854,18 +5848,6 @@ static int kvmppc_book3s_init_hv(void)
if (kvmppc_radix_possible())
r = kvmppc_radix_init();
- /*
- * POWER9 chips before version 2.02 can't have some threads in
- * HPT mode and some in radix mode on the same core.
- */
- if (cpu_has_feature(CPU_FTR_ARCH_300)) {
- unsigned int pvr = mfspr(SPRN_PVR);
- if ((pvr >> 16) = PVR_POWER9 &&
- (((pvr & 0xe000) = 0 && (pvr & 0xfff) < 0x202) ||
- ((pvr & 0xe000) = 0x2000 && (pvr & 0xfff) < 0x101)))
- no_mixing_hpt_and_radix = true;
- }
-
r = kvmppc_uvmem_init();
if (r < 0)
pr_err("KVM-HV: kvmppc_uvmem_init failed %d\n", r);
diff --git a/arch/powerpc/kvm/book3s_hv_interrupts.S b/arch/powerpc/kvm/book3s_hv_interrupts.S
index 327417d79eac..4444f83cb133 100644
--- a/arch/powerpc/kvm/book3s_hv_interrupts.S
+++ b/arch/powerpc/kvm/book3s_hv_interrupts.S
@@ -58,7 +58,7 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
/*
* Put whatever is in the decrementer into the
* hypervisor decrementer.
- * Because of a hardware deviation in P8 and P9,
+ * Because of a hardware deviation in P8,
* we need to set LPCR[HDICE] before writing HDEC.
*/
ld r5, HSTATE_KVM_VCORE(r13)
@@ -67,15 +67,10 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
ori r8, r9, LPCR_HDICE
mtspr SPRN_LPCR, r8
isync
- andis. r0, r9, LPCR_LD@h
mfspr r8,SPRN_DEC
mftb r7
-BEGIN_FTR_SECTION
- /* On POWER9, don't sign-extend if host LPCR[LD] bit is set */
- bne 32f
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
extsw r8,r8
-32: mtspr SPRN_HDEC,r8
+ mtspr SPRN_HDEC,r8
add r8,r8,r7
std r8,HSTATE_DECEXP(r13)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 5cbc26ebcf8e..b2a71a1b48fe 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -25,7 +25,6 @@
#include <asm/export.h>
#include <asm/tm.h>
#include <asm/opal.h>
-#include <asm/xive-regs.h>
#include <asm/thread_info.h>
#include <asm/asm-compat.h>
#include <asm/feature-fixups.h>
@@ -34,9 +33,7 @@
/* Sign-extend HDEC if not on POWER9 */
#define EXTEND_HDEC(reg) \
-BEGIN_FTR_SECTION; \
- extsw reg, reg; \
-END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
+ extsw reg, reg
/* Values in HSTATE_NAPPING(r13) */
#define NAPPING_CEDE 1
@@ -56,8 +53,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
#define STACK_SLOT_HFSCR (SFS-72)
#define STACK_SLOT_AMR (SFS-80)
#define STACK_SLOT_UAMOR (SFS-88)
-#define STACK_SLOT_DAWR1 (SFS-96)
-#define STACK_SLOT_DAWRX1 (SFS-104)
/*
* Call kvmppc_hv_entry in real mode.
@@ -333,7 +328,6 @@ kvm_secondary_got_guest:
LOAD_REG_ADDR(r6, decrementer_max)
ld r6, 0(r6)
mtspr SPRN_HDEC, r6
-BEGIN_FTR_SECTION
/* and set per-LPAR registers, if doing dynamic micro-threading */
ld r6, HSTATE_SPLIT_MODE(r13)
cmpdi r6, 0
@@ -345,7 +339,6 @@ BEGIN_FTR_SECTION
ld r0, KVM_SPLIT_LDBAR(r6)
mtspr SPRN_LDBAR, r0
isync
-END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
63:
/* Order load of vcpu after load of vcore */
lwsync
@@ -416,7 +409,6 @@ kvm_no_guest:
blr
53:
-BEGIN_FTR_SECTION
HMT_LOW
ld r5, HSTATE_KVM_VCORE(r13)
cmpdi r5, 0
@@ -431,14 +423,6 @@ BEGIN_FTR_SECTION
b kvm_unsplit_nap
60: HMT_MEDIUM
b kvm_secondary_got_guest
-FTR_SECTION_ELSE
- HMT_LOW
- ld r5, HSTATE_KVM_VCORE(r13)
- cmpdi r5, 0
- beq kvm_no_guest
- HMT_MEDIUM
- b kvm_secondary_got_guest
-ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
54: li r0, KVM_HWTHREAD_IN_KVM
stb r0, HSTATE_HWTHREAD_STATE(r13)
@@ -564,13 +548,11 @@ kvmppc_hv_entry:
bne 10f
lwz r7,KVM_LPID(r9)
-BEGIN_FTR_SECTION
ld r6,KVM_SDR1(r9)
li r0,LPID_RSVD /* switch to reserved LPID */
mtspr SPRN_LPID,r0
ptesync
mtspr SPRN_SDR1,r6 /* switch to partition page table */
-END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
mtspr SPRN_LPID,r7
isync
@@ -650,16 +632,6 @@ kvmppc_got_guest:
mtspr SPRN_SPURR,r8
/* Save host values of some registers */
-BEGIN_FTR_SECTION
- mfspr r5, SPRN_TIDR
- mfspr r6, SPRN_PSSCR
- mfspr r7, SPRN_PID
- std r5, STACK_SLOT_TID(r1)
- std r6, STACK_SLOT_PSSCR(r1)
- std r7, STACK_SLOT_PID(r1)
- mfspr r5, SPRN_HFSCR
- std r5, STACK_SLOT_HFSCR(r1)
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
BEGIN_FTR_SECTION
mfspr r5, SPRN_CIABR
mfspr r6, SPRN_DAWR0
@@ -670,12 +642,6 @@ BEGIN_FTR_SECTION
std r7, STACK_SLOT_DAWRX0(r1)
std r8, STACK_SLOT_IAMR(r1)
END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
-BEGIN_FTR_SECTION
- mfspr r6, SPRN_DAWR1
- mfspr r7, SPRN_DAWRX1
- std r6, STACK_SLOT_DAWR1(r1)
- std r7, STACK_SLOT_DAWRX1(r1)
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S | CPU_FTR_DAWR1)
mfspr r5, SPRN_AMR
std r5, STACK_SLOT_AMR(r1)
@@ -693,13 +659,9 @@ BEGIN_FTR_SECTION
END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
-/*
- * Branch around the call if both CPU_FTR_TM and
- * CPU_FTR_P9_TM_HV_ASSIST are off.
- */
BEGIN_FTR_SECTION
b 91f
-END_FTR_SECTION(CPU_FTR_TM | CPU_FTR_P9_TM_HV_ASSIST, 0)
+END_FTR_SECTION_IFCLR(CPU_FTR_TM)
/*
* NOTE THAT THIS TRASHES ALL NON-VOLATILE REGISTERS (but not CR)
*/
@@ -766,12 +728,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
ld r6, VCPU_DAWRX0(r4)
mtspr SPRN_DAWR0, r5
mtspr SPRN_DAWRX0, r6
-BEGIN_FTR_SECTION
- ld r5, VCPU_DAWR1(r4)
- ld r6, VCPU_DAWRX1(r4)
- mtspr SPRN_DAWR1, r5
- mtspr SPRN_DAWRX1, r6
-END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
1:
ld r7, VCPU_CIABR(r4)
ld r8, VCPU_TAR(r4)
@@ -789,7 +745,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
mtspr SPRN_BESCR, r6
mtspr SPRN_PID, r7
mtspr SPRN_WORT, r8
-BEGIN_FTR_SECTION
/* POWER8-only registers */
ld r5, VCPU_TCSCR(r4)
ld r6, VCPU_ACOP(r4)
@@ -800,18 +755,6 @@ BEGIN_FTR_SECTION
mtspr SPRN_CSIGR, r7
mtspr SPRN_TACR, r8
nop
-FTR_SECTION_ELSE
- /* POWER9-only registers */
- ld r5, VCPU_TID(r4)
- ld r6, VCPU_PSSCR(r4)
- lbz r8, HSTATE_FAKE_SUSPEND(r13)
- oris r6, r6, PSSCR_EC@h /* This makes stop trap to HV */
- rldimi r6, r8, PSSCR_FAKE_SUSPEND_LG, 63 - PSSCR_FAKE_SUSPEND_LG
- ld r7, VCPU_HFSCR(r4)
- mtspr SPRN_TIDR, r5
- mtspr SPRN_PSSCR, r6
- mtspr SPRN_HFSCR, r7
-ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
8:
ld r5, VCPU_SPRG0(r4)
@@ -904,93 +847,9 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
bdnz 1b
9:
-#ifdef CONFIG_KVM_XICS
- /* We are entering the guest on that thread, push VCPU to XIVE */
- ld r11, VCPU_XIVE_SAVED_STATE(r4)
- li r9, TM_QW1_OS
- lwz r8, VCPU_XIVE_CAM_WORD(r4)
- cmpwi r8, 0
- beq no_xive
- li r7, TM_QW1_OS + TM_WORD2
- mfmsr r0
- andi. r0, r0, MSR_DR /* in real mode? */
- beq 2f
- ld r10, HSTATE_XIVE_TIMA_VIRT(r13)
- cmpldi cr1, r10, 0
- beq cr1, no_xive
- eieio
- stdx r11,r9,r10
- stwx r8,r7,r10
- b 3f
-2: ld r10, HSTATE_XIVE_TIMA_PHYS(r13)
- cmpldi cr1, r10, 0
- beq cr1, no_xive
- eieio
- stdcix r11,r9,r10
- stwcix r8,r7,r10
-3: li r9, 1
- stb r9, VCPU_XIVE_PUSHED(r4)
- eieio
-
- /*
- * We clear the irq_pending flag. There is a small chance of a
- * race vs. the escalation interrupt happening on another
- * processor setting it again, but the only consequence is to
- * cause a spurrious wakeup on the next H_CEDE which is not an
- * issue.
- */
- li r0,0
- stb r0, VCPU_IRQ_PENDING(r4)
-
- /*
- * In single escalation mode, if the escalation interrupt is
- * on, we mask it.
- */
- lbz r0, VCPU_XIVE_ESC_ON(r4)
- cmpwi cr1, r0,0
- beq cr1, 1f
- li r9, XIVE_ESB_SET_PQ_01
- beq 4f /* in real mode? */
- ld r10, VCPU_XIVE_ESC_VADDR(r4)
- ldx r0, r10, r9
- b 5f
-4: ld r10, VCPU_XIVE_ESC_RADDR(r4)
- ldcix r0, r10, r9
-5: sync
-
- /* We have a possible subtle race here: The escalation interrupt might
- * have fired and be on its way to the host queue while we mask it,
- * and if we unmask it early enough (re-cede right away), there is
- * a theorical possibility that it fires again, thus landing in the
- * target queue more than once which is a big no-no.
- *
- * Fortunately, solving this is rather easy. If the above load setting
- * PQ to 01 returns a previous value where P is set, then we know the
- * escalation interrupt is somewhere on its way to the host. In that
- * case we simply don't clear the xive_esc_on flag below. It will be
- * eventually cleared by the handler for the escalation interrupt.
- *
- * Then, when doing a cede, we check that flag again before re-enabling
- * the escalation interrupt, and if set, we abort the cede.
- */
- andi. r0, r0, XIVE_ESB_VAL_P
- bne- 1f
-
- /* Now P is 0, we can clear the flag */
- li r0, 0
- stb r0, VCPU_XIVE_ESC_ON(r4)
-1:
-no_xive:
-#endif /* CONFIG_KVM_XICS */
-
deliver_guest_interrupt: /* r4 = vcpu, r13 = paca */
/* Check if we can deliver an external or decrementer interrupt now */
ld r0, VCPU_PENDING_EXC(r4)
-BEGIN_FTR_SECTION
- /* On POWER9, also check for emulated doorbell interrupt */
- lbz r3, VCPU_DBELL_REQ(r4)
- or r0, r0, r3
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
cmpdi r0, 0
beq 71f
mr r3, r4
@@ -1063,12 +922,6 @@ BEGIN_FTR_SECTION
mtspr SPRN_PPR, r0
END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
-/* Move canary into DSISR to check for later */
-BEGIN_FTR_SECTION
- li r0, 0x7fff
- mtspr SPRN_HDSISR, r0
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
-
ld r6, VCPU_KVM(r4)
lbz r7, KVM_SECURE_GUEST(r6)
cmpdi r7, 0
@@ -1284,14 +1137,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
/* Hypervisor doorbell - exit only if host IPI flag set */
cmpwi r12, BOOK3S_INTERRUPT_H_DOORBELL
bne 3f
-BEGIN_FTR_SECTION
- PPC_MSGSYNC
- lwsync
- /* always exit if we're running a nested guest */
- ld r0, VCPU_NESTED(r9)
- cmpdi r0, 0
- bne guest_exit_cont
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
lbz r0, HSTATE_HOST_IPI(r13)
cmpwi r0, 0
beq maybe_reenter_guest
@@ -1321,43 +1166,6 @@ guest_exit_cont: /* r9 = vcpu, r12 = trap, r13 = paca */
mr r4, r9
bl kvmhv_accumulate_time
#endif
-#ifdef CONFIG_KVM_XICS
- /* We are exiting, pull the VP from the XIVE */
- lbz r0, VCPU_XIVE_PUSHED(r9)
- cmpwi cr0, r0, 0
- beq 1f
- li r7, TM_SPC_PULL_OS_CTX
- li r6, TM_QW1_OS
- mfmsr r0
- andi. r0, r0, MSR_DR /* in real mode? */
- beq 2f
- ld r10, HSTATE_XIVE_TIMA_VIRT(r13)
- cmpldi cr0, r10, 0
- beq 1f
- /* First load to pull the context, we ignore the value */
- eieio
- lwzx r11, r7, r10
- /* Second load to recover the context state (Words 0 and 1) */
- ldx r11, r6, r10
- b 3f
-2: ld r10, HSTATE_XIVE_TIMA_PHYS(r13)
- cmpldi cr0, r10, 0
- beq 1f
- /* First load to pull the context, we ignore the value */
- eieio
- lwzcix r11, r7, r10
- /* Second load to recover the context state (Words 0 and 1) */
- ldcix r11, r6, r10
-3: std r11, VCPU_XIVE_SAVED_STATE(r9)
- /* Fixup some of the state for the next load */
- li r10, 0
- li r0, 0xff
- stb r10, VCPU_XIVE_PUSHED(r9)
- stb r10, (VCPU_XIVE_SAVED_STATE+3)(r9)
- stb r0, (VCPU_XIVE_SAVED_STATE+4)(r9)
- eieio
-1:
-#endif /* CONFIG_KVM_XICS */
/*
* Possibly flush the link stack here, before we do a blr in
@@ -1412,12 +1220,6 @@ guest_bypass:
ld r3, HSTATE_KVM_VCORE(r13)
mfspr r5,SPRN_DEC
mftb r6
- /* On P9, if the guest has large decr enabled, don't sign extend */
-BEGIN_FTR_SECTION
- ld r4, VCORE_LPCR(r3)
- andis. r4, r4, LPCR_LD@h
- bne 16f
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
extsw r5,r5
16: add r5,r5,r6
/* r5 is a guest timebase value here, convert to host TB */
@@ -1491,7 +1293,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
std r6, VCPU_BESCR(r9)
stw r7, VCPU_GUEST_PID(r9)
std r8, VCPU_WORT(r9)
-BEGIN_FTR_SECTION
mfspr r5, SPRN_TCSCR
mfspr r6, SPRN_ACOP
mfspr r7, SPRN_CSIGR
@@ -1500,17 +1301,6 @@ BEGIN_FTR_SECTION
std r6, VCPU_ACOP(r9)
std r7, VCPU_CSIGR(r9)
std r8, VCPU_TACR(r9)
-FTR_SECTION_ELSE
- mfspr r5, SPRN_TIDR
- mfspr r6, SPRN_PSSCR
- std r5, VCPU_TID(r9)
- rldicl r6, r6, 4, 50 /* r6 &= PSSCR_GUEST_VIS */
- rotldi r6, r6, 60
- std r6, VCPU_PSSCR(r9)
- /* Restore host HFSCR value */
- ld r7, STACK_SLOT_HFSCR(r1)
- mtspr SPRN_HFSCR, r7
-ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
/*
* Restore various registers to 0, where non-zero values
* set by the guest could disrupt the host.
@@ -1518,13 +1308,11 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
li r0, 0
mtspr SPRN_PSPB, r0
mtspr SPRN_WORT, r0
-BEGIN_FTR_SECTION
mtspr SPRN_TCSCR, r0
/* Set MMCRS to 1<<31 to freeze and disable the SPMC counters */
li r0, 1
sldi r0, r0, 31
mtspr SPRN_MMCRS, r0
-END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
/* Save and restore AMR, IAMR and UAMOR before turning on the MMU */
ld r8, STACK_SLOT_IAMR(r1)
@@ -1581,13 +1369,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
bl kvmppc_save_fp
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
-/*
- * Branch around the call if both CPU_FTR_TM and
- * CPU_FTR_P9_TM_HV_ASSIST are off.
- */
BEGIN_FTR_SECTION
b 91f
-END_FTR_SECTION(CPU_FTR_TM | CPU_FTR_P9_TM_HV_ASSIST, 0)
+END_FTR_SECTION_IFCLR(CPU_FTR_TM)
/*
* NOTE THAT THIS TRASHES ALL NON-VOLATILE REGISTERS (but not CR)
*/
@@ -1633,28 +1417,6 @@ BEGIN_FTR_SECTION
mtspr SPRN_DAWR0, r6
mtspr SPRN_DAWRX0, r7
END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
-BEGIN_FTR_SECTION
- ld r6, STACK_SLOT_DAWR1(r1)
- ld r7, STACK_SLOT_DAWRX1(r1)
- mtspr SPRN_DAWR1, r6
- mtspr SPRN_DAWRX1, r7
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S | CPU_FTR_DAWR1)
-BEGIN_FTR_SECTION
- ld r5, STACK_SLOT_TID(r1)
- ld r6, STACK_SLOT_PSSCR(r1)
- ld r7, STACK_SLOT_PID(r1)
- mtspr SPRN_TIDR, r5
- mtspr SPRN_PSSCR, r6
- mtspr SPRN_PID, r7
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
-
- /*
- * cp_abort is required if the processor supports local copy-paste
- * to clear the copy buffer that was under control of the guest.
- */
-BEGIN_FTR_SECTION
- PPC_CP_ABORT
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_31)
/*
* POWER7/POWER8 guest -> host partition switch code.
@@ -1691,13 +1453,11 @@ kvmhv_switch_to_host:
/* Primary thread switches back to host partition */
lwz r7,KVM_HOST_LPID(r4)
-BEGIN_FTR_SECTION
ld r6,KVM_HOST_SDR1(r4)
li r8,LPID_RSVD /* switch to reserved LPID */
mtspr SPRN_LPID,r8
ptesync
mtspr SPRN_SDR1,r6 /* switch to host page table */
-END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
mtspr SPRN_LPID,r7
isync
@@ -1908,20 +1668,11 @@ kvmppc_tm_emul:
kvmppc_hdsi:
mfspr r4, SPRN_HDAR
mfspr r6, SPRN_HDSISR
-BEGIN_FTR_SECTION
- /* Look for DSISR canary. If we find it, retry instruction */
- cmpdi r6, 0x7fff
- beq 6f
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
/* HPTE not found fault or protection fault? */
andis. r0, r6, (DSISR_NOHPTE | DSISR_PROTFAULT)@h
beq 1f /* if not, send it to the guest */
andi. r0, r11, MSR_DR /* data relocation enabled? */
beq 3f
-BEGIN_FTR_SECTION
- mfspr r5, SPRN_ASDR /* on POWER9, use ASDR to get VSID */
- b 4f
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
clrrdi r0, r4, 28
PPC_SLBFEE_DOT(R5, R0) /* if so, look up SLB */
li r0, BOOK3S_INTERRUPT_DATA_SEGMENT
@@ -1998,10 +1749,6 @@ kvmppc_hisi:
beq 1f
andi. r0, r11, MSR_IR /* instruction relocation enabled? */
beq 3f
-BEGIN_FTR_SECTION
- mfspr r5, SPRN_ASDR /* on POWER9, use ASDR to get VSID */
- b 4f
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
clrrdi r0, r10, 28
PPC_SLBFEE_DOT(R5, R0) /* if so, look up SLB */
li r0, BOOK3S_INTERRUPT_INST_SEGMENT
@@ -2049,10 +1796,6 @@ hcall_try_real_mode:
andi. r0,r11,MSR_PR
/* sc 1 from userspace - reflect to guest syscall */
bne sc_1_fast_return
- /* sc 1 from nested guest - give it to L1 to handle */
- ld r0, VCPU_NESTED(r9)
- cmpdi r0, 0
- bne guest_exit_cont
clrrdi r3,r3,2
cmpldi r3,hcall_real_table_end - hcall_real_table
bge guest_exit_cont
@@ -2448,13 +2191,9 @@ _GLOBAL(kvmppc_h_cede) /* r3 = vcpu pointer, r11 = msr, r13 = paca */
bl kvmppc_save_fp
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
-/*
- * Branch around the call if both CPU_FTR_TM and
- * CPU_FTR_P9_TM_HV_ASSIST are off.
- */
BEGIN_FTR_SECTION
b 91f
-END_FTR_SECTION(CPU_FTR_TM | CPU_FTR_P9_TM_HV_ASSIST, 0)
+END_FTR_SECTION_IFCLR(CPU_FTR_TM)
/*
* NOTE THAT THIS TRASHES ALL NON-VOLATILE REGISTERS (but not CR)
*/
@@ -2474,15 +2213,8 @@ END_FTR_SECTION(CPU_FTR_TM | CPU_FTR_P9_TM_HV_ASSIST, 0)
mfspr r3, SPRN_DEC
mfspr r4, SPRN_HDEC
mftb r5
-BEGIN_FTR_SECTION
- /* On P9 check whether the guest has large decrementer mode enabled */
- ld r6, HSTATE_KVM_VCORE(r13)
- ld r6, VCORE_LPCR(r6)
- andis. r6, r6, LPCR_LD@h
- bne 68f
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
extsw r3, r3
-68: EXTEND_HDEC(r4)
+ EXTEND_HDEC(r4)
cmpd r3, r4
ble 67f
mtspr SPRN_DEC, r4
@@ -2527,28 +2259,14 @@ BEGIN_FTR_SECTION
END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
kvm_nap_sequence: /* desired LPCR value in r5 */
-BEGIN_FTR_SECTION
- /*
- * PSSCR bits: exit criterion = 1 (wakeup based on LPCR at sreset)
- * enable state loss = 1 (allow SMT mode switch)
- * requested level = 0 (just stop dispatching)
- */
- lis r3, (PSSCR_EC | PSSCR_ESL)@h
/* Set LPCR_PECE_HVEE bit to enable wakeup by HV interrupts */
li r4, LPCR_PECE_HVEE@higher
sldi r4, r4, 32
or r5, r5, r4
-FTR_SECTION_ELSE
- li r3, PNV_THREAD_NAP
-ALT_FTR_SECTION_END_IFSET(CPU_FTR_ARCH_300)
mtspr SPRN_LPCR,r5
isync
-BEGIN_FTR_SECTION
- bl isa300_idle_stop_mayloss
-FTR_SECTION_ELSE
bl isa206_idle_insn_mayloss
-ALT_FTR_SECTION_END_IFSET(CPU_FTR_ARCH_300)
mfspr r0, SPRN_CTRLF
ori r0, r0, 1
@@ -2567,10 +2285,8 @@ ALT_FTR_SECTION_END_IFSET(CPU_FTR_ARCH_300)
beq kvm_end_cede
cmpwi r0, NAPPING_NOVCPU
beq kvm_novcpu_wakeup
-BEGIN_FTR_SECTION
cmpwi r0, NAPPING_UNSPLIT
beq kvm_unsplit_wakeup
-END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
twi 31,0,0 /* Nap state must not be zero */
33: mr r4, r3
@@ -2590,13 +2306,9 @@ kvm_end_cede:
#endif
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
-/*
- * Branch around the call if both CPU_FTR_TM and
- * CPU_FTR_P9_TM_HV_ASSIST are off.
- */
BEGIN_FTR_SECTION
b 91f
-END_FTR_SECTION(CPU_FTR_TM | CPU_FTR_P9_TM_HV_ASSIST, 0)
+END_FTR_SECTION_IFCLR(CPU_FTR_TM)
/*
* NOTE THAT THIS TRASHES ALL NON-VOLATILE REGISTERS (but not CR)
*/
@@ -2686,47 +2398,7 @@ kvm_cede_prodded:
/* we've ceded but we want to give control to the host */
kvm_cede_exit:
ld r9, HSTATE_KVM_VCPU(r13)
-#ifdef CONFIG_KVM_XICS
- /* are we using XIVE with single escalation? */
- ld r10, VCPU_XIVE_ESC_VADDR(r9)
- cmpdi r10, 0
- beq 3f
- li r6, XIVE_ESB_SET_PQ_00
- /*
- * If we still have a pending escalation, abort the cede,
- * and we must set PQ to 10 rather than 00 so that we don't
- * potentially end up with two entries for the escalation
- * interrupt in the XIVE interrupt queue. In that case
- * we also don't want to set xive_esc_on to 1 here in
- * case we race with xive_esc_irq().
- */
- lbz r5, VCPU_XIVE_ESC_ON(r9)
- cmpwi r5, 0
- beq 4f
- li r0, 0
- stb r0, VCPU_CEDED(r9)
- /*
- * The escalation interrupts are special as we don't EOI them.
- * There is no need to use the load-after-store ordering offset
- * to set PQ to 10 as we won't use StoreEOI.
- */
- li r6, XIVE_ESB_SET_PQ_10
- b 5f
-4: li r0, 1
- stb r0, VCPU_XIVE_ESC_ON(r9)
- /* make sure store to xive_esc_on is seen before xive_esc_irq runs */
- sync
-5: /* Enable XIVE escalation */
- mfmsr r0
- andi. r0, r0, MSR_DR /* in real mode? */
- beq 1f
- ldx r0, r10, r6
- b 2f
-1: ld r10, VCPU_XIVE_ESC_RADDR(r9)
- ldcix r0, r10, r6
-2: sync
-#endif /* CONFIG_KVM_XICS */
-3: b guest_exit_cont
+ b guest_exit_cont
/* Try to do machine check recovery in real mode */
machine_check_realmode:
@@ -2803,10 +2475,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
PPC_MSGCLR(6)
/* see if it's a host IPI */
li r3, 1
-BEGIN_FTR_SECTION
- PPC_MSGSYNC
- lwsync
-END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
lbz r0, HSTATE_HOST_IPI(r13)
cmpwi r0, 0
bnelr
@@ -3115,70 +2783,12 @@ kvmppc_bad_host_intr:
std r3, STACK_FRAME_OVERHEAD-16(r1)
/*
- * On POWER9 do a minimal restore of the MMU and call C code,
- * which will print a message and panic.
* XXX On POWER7 and POWER8, we just spin here since we don't
* know what the other threads are doing (and we don't want to
* coordinate with them) - but at least we now have register state
* in memory that we might be able to look at from another CPU.
*/
-BEGIN_FTR_SECTION
b .
-END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
- ld r9, HSTATE_KVM_VCPU(r13)
- ld r10, VCPU_KVM(r9)
-
- li r0, 0
- mtspr SPRN_AMR, r0
- mtspr SPRN_IAMR, r0
- mtspr SPRN_CIABR, r0
- mtspr SPRN_DAWRX0, r0
-BEGIN_FTR_SECTION
- mtspr SPRN_DAWRX1, r0
-END_FTR_SECTION_IFSET(CPU_FTR_DAWR1)
-
- /* Clear guest SLB. */
- slbmte r0, r0
- PPC_SLBIA(6)
- ptesync
-
- /* load host SLB entries */
- ld r8, PACA_SLBSHADOWPTR(r13)
- .rept SLB_NUM_BOLTED
- li r3, SLBSHADOW_SAVEAREA
- LDX_BE r5, r8, r3
- addi r3, r3, 8
- LDX_BE r6, r8, r3
- andis. r7, r5, SLB_ESID_V@h
- beq 3f
- slbmte r6, r5
-3: addi r8, r8, 16
- .endr
-
- lwz r7, KVM_HOST_LPID(r10)
- mtspr SPRN_LPID, r7
- mtspr SPRN_PID, r0
- ld r8, KVM_HOST_LPCR(r10)
- mtspr SPRN_LPCR, r8
- isync
- li r0, KVM_GUEST_MODE_NONE
- stb r0, HSTATE_IN_GUEST(r13)
-
- /*
- * Turn on the MMU and jump to C code
- */
- bcl 20, 31, .+4
-5: mflr r3
- addi r3, r3, 9f - 5b
- li r4, -1
- rldimi r3, r4, 62, 0 /* ensure 0xc000000000000000 bits are set */
- ld r4, PACAKMSR(r13)
- mtspr SPRN_SRR0, r3
- mtspr SPRN_SRR1, r4
- RFI_TO_KERNEL
-9: addi r3, r1, STACK_FRAME_OVERHEAD
- bl kvmppc_bad_interrupt
- b 9b
/*
* This mimics the MSR transition on IRQ delivery. The new guest MSR is taken
diff --git a/arch/powerpc/platforms/powernv/idle.c b/arch/powerpc/platforms/powernv/idle.c
index 999997d9e9a9..528a7e0cf83a 100644
--- a/arch/powerpc/platforms/powernv/idle.c
+++ b/arch/powerpc/platforms/powernv/idle.c
@@ -604,7 +604,7 @@ struct p9_sprs {
u64 uamor;
};
-static unsigned long power9_idle_stop(unsigned long psscr, bool mmu_on)
+static unsigned long power9_idle_stop(unsigned long psscr)
{
int cpu = raw_smp_processor_id();
int first = cpu_first_thread_sibling(cpu);
@@ -620,8 +620,6 @@ static unsigned long power9_idle_stop(unsigned long psscr, bool mmu_on)
if (!(psscr & (PSSCR_EC|PSSCR_ESL))) {
/* EC=ESL=0 case */
- BUG_ON(!mmu_on);
-
/*
* Wake synchronously. SRESET via xscom may still cause
* a 0x100 powersave wakeup with SRR1 reason!
@@ -803,8 +801,7 @@ static unsigned long power9_idle_stop(unsigned long psscr, bool mmu_on)
__slb_restore_bolted_realmode();
out:
- if (mmu_on)
- mtmsr(MSR_KERNEL);
+ mtmsr(MSR_KERNEL);
return srr1;
}
@@ -895,7 +892,7 @@ struct p10_sprs {
*/
};
-static unsigned long power10_idle_stop(unsigned long psscr, bool mmu_on)
+static unsigned long power10_idle_stop(unsigned long psscr)
{
int cpu = raw_smp_processor_id();
int first = cpu_first_thread_sibling(cpu);
@@ -909,8 +906,6 @@ static unsigned long power10_idle_stop(unsigned long psscr, bool mmu_on)
if (!(psscr & (PSSCR_EC|PSSCR_ESL))) {
/* EC=ESL=0 case */
- BUG_ON(!mmu_on);
-
/*
* Wake synchronously. SRESET via xscom may still cause
* a 0x100 powersave wakeup with SRR1 reason!
@@ -991,8 +986,7 @@ static unsigned long power10_idle_stop(unsigned long psscr, bool mmu_on)
__slb_restore_bolted_realmode();
out:
- if (mmu_on)
- mtmsr(MSR_KERNEL);
+ mtmsr(MSR_KERNEL);
return srr1;
}
@@ -1002,40 +996,10 @@ static unsigned long arch300_offline_stop(unsigned long psscr)
{
unsigned long srr1;
-#ifndef CONFIG_KVM_BOOK3S_HV_POSSIBLE
- __ppc64_runlatch_off();
if (cpu_has_feature(CPU_FTR_ARCH_31))
- srr1 = power10_idle_stop(psscr, true);
+ srr1 = power10_idle_stop(psscr);
else
- srr1 = power9_idle_stop(psscr, true);
- __ppc64_runlatch_on();
-#else
- /*
- * Tell KVM we're entering idle.
- * This does not have to be done in real mode because the P9 MMU
- * is independent per-thread. Some steppings share radix/hash mode
- * between threads, but in that case KVM has a barrier sync in real
- * mode before and after switching between radix and hash.
- *
- * kvm_start_guest must still be called in real mode though, hence
- * the false argument.
- */
- local_paca->kvm_hstate.hwthread_state = KVM_HWTHREAD_IN_IDLE;
-
- __ppc64_runlatch_off();
- if (cpu_has_feature(CPU_FTR_ARCH_31))
- srr1 = power10_idle_stop(psscr, false);
- else
- srr1 = power9_idle_stop(psscr, false);
- __ppc64_runlatch_on();
-
- local_paca->kvm_hstate.hwthread_state = KVM_HWTHREAD_IN_KERNEL;
- /* Order setting hwthread_state vs. testing hwthread_req */
- smp_mb();
- if (local_paca->kvm_hstate.hwthread_req)
- srr1 = idle_kvm_start_guest(srr1);
- mtmsr(MSR_KERNEL);
-#endif
+ srr1 = power9_idle_stop(psscr);
return srr1;
}
@@ -1055,9 +1019,9 @@ void arch300_idle_type(unsigned long stop_psscr_val,
__ppc64_runlatch_off();
if (cpu_has_feature(CPU_FTR_ARCH_31))
- srr1 = power10_idle_stop(psscr, true);
+ srr1 = power10_idle_stop(psscr);
else
- srr1 = power9_idle_stop(psscr, true);
+ srr1 = power9_idle_stop(psscr);
__ppc64_runlatch_on();
fini_irq_for_idle_irqsoff();
--
2.23.0
^ permalink raw reply related [flat|nested] 188+ messages in thread
* Re: [PATCH v4 00/46] KVM: PPC: Book3S: C-ify the P9 entry/exit code
2021-03-23 1:02 ` Nicholas Piggin
@ 2021-03-24 16:25 ` Nicholas Piggin
-1 siblings, 0 replies; 188+ messages in thread
From: Nicholas Piggin @ 2021-03-24 16:25 UTC (permalink / raw)
To: kvm-ppc; +Cc: linuxppc-dev
Excerpts from Nicholas Piggin's message of March 23, 2021 11:02 am:
> I think enough changes and fixes have gone in since last round
> to repost.
>
> I put a git tree here to make things easier to get.
>
> https://github.com/npiggin/linux/tree/kvm-in-c-2
I just pushed a new tree https://github.com/npiggin/linux/tree/kvm-in-c-3
This mainly fixes up the compile error with older binutils and PR=1
hcall reflection into guest for PR KVM. Haven't been able to completely
test PR KVM because of unrelated connectivity issues but at least the
PR=1 syscall reflection is working again.
Thanks,
Nick
^ permalink raw reply [flat|nested] 188+ messages in thread