* [PATCH 0/7] KVM: PPC: Book3S HV: Enable on little endian hosts
@ 2014-06-17 16:17 ` Alexander Graf
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
So far we've been able to successfully run HV KVM on big endian hosts, but
once you dive into little endian land things start to fall apart.
This patch set enables HV KVM for little endian hosts. This should be the
final piece left missing to get little endian systems fully en par with big
endian ones in the KVM world - modulo bugs.
For now guest threading support is still slightly flaky, but I'm sure that's
only a minor breakage somewhere that we'll find soon.
v1 -> v2:
- fix typo in STWX_BE
- Add __be hints
- Fix H_REMOVE
- Fix dtl_idx
- Make XICS code easier to follow and use memory for bswap
Alex
Alexander Graf (7):
PPC: Add asm helpers for BE 32bit load/store
KVM: PPC: Book3S HV: Make HTAB code LE host aware
KVM: PPC: Book3S HV: Access guest VPA in BE
KVM: PPC: Book3S HV: Access host lppaca and shadow slb in BE
KVM: PPC: Book3S HV: Access XICS in BE
KVM: PPC: Book3S HV: Fix ABIv2 on LE
KVM: PPC: Book3S HV: Enable for little endian hosts
arch/powerpc/include/asm/asm-compat.h | 4 +
arch/powerpc/include/asm/kvm_book3s.h | 4 +-
arch/powerpc/include/asm/kvm_book3s_64.h | 15 +++-
arch/powerpc/kvm/Kconfig | 1 -
arch/powerpc/kvm/book3s_64_mmu_hv.c | 128 ++++++++++++++-------------
arch/powerpc/kvm/book3s_hv.c | 22 ++---
arch/powerpc/kvm/book3s_hv_ras.c | 6 +-
arch/powerpc/kvm/book3s_hv_rm_mmu.c | 146 ++++++++++++++++++-------------
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 60 ++++++++-----
9 files changed, 220 insertions(+), 166 deletions(-)
--
1.8.1.4
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 0/7] KVM: PPC: Book3S HV: Enable on little endian hosts
@ 2014-06-17 16:17 ` Alexander Graf
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
So far we've been able to successfully run HV KVM on big endian hosts, but
once you dive into little endian land things start to fall apart.
This patch set enables HV KVM for little endian hosts. This should be the
final piece left missing to get little endian systems fully en par with big
endian ones in the KVM world - modulo bugs.
For now guest threading support is still slightly flaky, but I'm sure that's
only a minor breakage somewhere that we'll find soon.
v1 -> v2:
- fix typo in STWX_BE
- Add __be hints
- Fix H_REMOVE
- Fix dtl_idx
- Make XICS code easier to follow and use memory for bswap
Alex
Alexander Graf (7):
PPC: Add asm helpers for BE 32bit load/store
KVM: PPC: Book3S HV: Make HTAB code LE host aware
KVM: PPC: Book3S HV: Access guest VPA in BE
KVM: PPC: Book3S HV: Access host lppaca and shadow slb in BE
KVM: PPC: Book3S HV: Access XICS in BE
KVM: PPC: Book3S HV: Fix ABIv2 on LE
KVM: PPC: Book3S HV: Enable for little endian hosts
arch/powerpc/include/asm/asm-compat.h | 4 +
arch/powerpc/include/asm/kvm_book3s.h | 4 +-
arch/powerpc/include/asm/kvm_book3s_64.h | 15 +++-
arch/powerpc/kvm/Kconfig | 1 -
arch/powerpc/kvm/book3s_64_mmu_hv.c | 128 ++++++++++++++-------------
arch/powerpc/kvm/book3s_hv.c | 22 ++---
arch/powerpc/kvm/book3s_hv_ras.c | 6 +-
arch/powerpc/kvm/book3s_hv_rm_mmu.c | 146 ++++++++++++++++++-------------
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 60 ++++++++-----
9 files changed, 220 insertions(+), 166 deletions(-)
--
1.8.1.4
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 1/7] PPC: Add asm helpers for BE 32bit load/store
2014-06-17 16:17 ` Alexander Graf
@ 2014-06-17 16:17 ` Alexander Graf
-1 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm, Benjamin Herrenschmidt
>From assembly code we might not only have to explicitly BE access 64bit values,
but sometimes also 32bit ones. Add helpers that allow for easy use of lwzx/stwx
in their respective byte-reverse or native form.
Signed-off-by: Alexander Graf <agraf@suse.de>
CC: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
v1 -> v2:
- fix typo in STWX_BE
---
arch/powerpc/include/asm/asm-compat.h | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/arch/powerpc/include/asm/asm-compat.h b/arch/powerpc/include/asm/asm-compat.h
index 4b237aa..21be8ae 100644
--- a/arch/powerpc/include/asm/asm-compat.h
+++ b/arch/powerpc/include/asm/asm-compat.h
@@ -34,10 +34,14 @@
#define PPC_MIN_STKFRM 112
#ifdef __BIG_ENDIAN__
+#define LWZX_BE stringify_in_c(lwzx)
#define LDX_BE stringify_in_c(ldx)
+#define STWX_BE stringify_in_c(stwx)
#define STDX_BE stringify_in_c(stdx)
#else
+#define LWZX_BE stringify_in_c(lwbrx)
#define LDX_BE stringify_in_c(ldbrx)
+#define STWX_BE stringify_in_c(stwbrx)
#define STDX_BE stringify_in_c(stdbrx)
#endif
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* [PATCH 1/7] PPC: Add asm helpers for BE 32bit load/store
@ 2014-06-17 16:17 ` Alexander Graf
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm, Benjamin Herrenschmidt
From assembly code we might not only have to explicitly BE access 64bit values,
but sometimes also 32bit ones. Add helpers that allow for easy use of lwzx/stwx
in their respective byte-reverse or native form.
Signed-off-by: Alexander Graf <agraf@suse.de>
CC: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
v1 -> v2:
- fix typo in STWX_BE
---
arch/powerpc/include/asm/asm-compat.h | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/arch/powerpc/include/asm/asm-compat.h b/arch/powerpc/include/asm/asm-compat.h
index 4b237aa..21be8ae 100644
--- a/arch/powerpc/include/asm/asm-compat.h
+++ b/arch/powerpc/include/asm/asm-compat.h
@@ -34,10 +34,14 @@
#define PPC_MIN_STKFRM 112
#ifdef __BIG_ENDIAN__
+#define LWZX_BE stringify_in_c(lwzx)
#define LDX_BE stringify_in_c(ldx)
+#define STWX_BE stringify_in_c(stwx)
#define STDX_BE stringify_in_c(stdx)
#else
+#define LWZX_BE stringify_in_c(lwbrx)
#define LDX_BE stringify_in_c(ldbrx)
+#define STWX_BE stringify_in_c(stwbrx)
#define STDX_BE stringify_in_c(stdbrx)
#endif
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* [PATCH 2/7] KVM: PPC: Book3S HV: Make HTAB code LE host aware
2014-06-17 16:17 ` Alexander Graf
@ 2014-06-17 16:17 ` Alexander Graf
-1 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
When running on an LE host all data structures are kept in little endian
byte order. However, the HTAB still needs to be maintained in big endian.
So every time we access any HTAB we need to make sure we do so in the right
byte order. Fix up all accesses to manually byte swap.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
v1 -> v2:
- Add __be32 hints
- Fix H_REMOVE
---
arch/powerpc/include/asm/kvm_book3s.h | 4 +-
arch/powerpc/include/asm/kvm_book3s_64.h | 15 +++-
arch/powerpc/kvm/book3s_64_mmu_hv.c | 128 ++++++++++++++-------------
arch/powerpc/kvm/book3s_hv_rm_mmu.c | 146 ++++++++++++++++++-------------
4 files changed, 164 insertions(+), 129 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_book3s.h b/arch/powerpc/include/asm/kvm_book3s.h
index a20cc0b..028f4b1 100644
--- a/arch/powerpc/include/asm/kvm_book3s.h
+++ b/arch/powerpc/include/asm/kvm_book3s.h
@@ -161,9 +161,9 @@ extern pfn_t kvmppc_gfn_to_pfn(struct kvm_vcpu *vcpu, gfn_t gfn, bool writing,
bool *writable);
extern void kvmppc_add_revmap_chain(struct kvm *kvm, struct revmap_entry *rev,
unsigned long *rmap, long pte_index, int realmode);
-extern void kvmppc_invalidate_hpte(struct kvm *kvm, unsigned long *hptep,
+extern void kvmppc_invalidate_hpte(struct kvm *kvm, __be64 *hptep,
unsigned long pte_index);
-void kvmppc_clear_ref_hpte(struct kvm *kvm, unsigned long *hptep,
+void kvmppc_clear_ref_hpte(struct kvm *kvm, __be64 *hptep,
unsigned long pte_index);
extern void *kvmppc_pin_guest_page(struct kvm *kvm, unsigned long addr,
unsigned long *nb_ret);
diff --git a/arch/powerpc/include/asm/kvm_book3s_64.h b/arch/powerpc/include/asm/kvm_book3s_64.h
index c7871f3..e504f88 100644
--- a/arch/powerpc/include/asm/kvm_book3s_64.h
+++ b/arch/powerpc/include/asm/kvm_book3s_64.h
@@ -59,20 +59,29 @@ extern unsigned long kvm_rma_pages;
/* These bits are reserved in the guest view of the HPTE */
#define HPTE_GR_RESERVED HPTE_GR_MODIFIED
-static inline long try_lock_hpte(unsigned long *hpte, unsigned long bits)
+static inline long try_lock_hpte(__be64 *hpte, unsigned long bits)
{
unsigned long tmp, old;
+ __be64 be_lockbit, be_bits;
+
+ /*
+ * We load/store in native endian, but the HTAB is in big endian. If
+ * we byte swap all data we apply on the PTE we're implicitly correct
+ * again.
+ */
+ be_lockbit = cpu_to_be64(HPTE_V_HVLOCK);
+ be_bits = cpu_to_be64(bits);
asm volatile(" ldarx %0,0,%2\n"
" and. %1,%0,%3\n"
" bne 2f\n"
- " ori %0,%0,%4\n"
+ " or %0,%0,%4\n"
" stdcx. %0,0,%2\n"
" beq+ 2f\n"
" mr %1,%3\n"
"2: isync"
: "=&r" (tmp), "=&r" (old)
- : "r" (hpte), "r" (bits), "i" (HPTE_V_HVLOCK)
+ : "r" (hpte), "r" (be_bits), "r" (be_lockbit)
: "cc", "memory");
return old == 0;
}
diff --git a/arch/powerpc/kvm/book3s_64_mmu_hv.c b/arch/powerpc/kvm/book3s_64_mmu_hv.c
index 8056107..2d154d9 100644
--- a/arch/powerpc/kvm/book3s_64_mmu_hv.c
+++ b/arch/powerpc/kvm/book3s_64_mmu_hv.c
@@ -450,7 +450,7 @@ static int kvmppc_mmu_book3s_64_hv_xlate(struct kvm_vcpu *vcpu, gva_t eaddr,
unsigned long slb_v;
unsigned long pp, key;
unsigned long v, gr;
- unsigned long *hptep;
+ __be64 *hptep;
int index;
int virtmode = vcpu->arch.shregs.msr & (data ? MSR_DR : MSR_IR);
@@ -473,13 +473,13 @@ static int kvmppc_mmu_book3s_64_hv_xlate(struct kvm_vcpu *vcpu, gva_t eaddr,
preempt_enable();
return -ENOENT;
}
- hptep = (unsigned long *)(kvm->arch.hpt_virt + (index << 4));
- v = hptep[0] & ~HPTE_V_HVLOCK;
+ hptep = (__be64 *)(kvm->arch.hpt_virt + (index << 4));
+ v = be64_to_cpu(hptep[0]) & ~HPTE_V_HVLOCK;
gr = kvm->arch.revmap[index].guest_rpte;
/* Unlock the HPTE */
asm volatile("lwsync" : : : "memory");
- hptep[0] = v;
+ hptep[0] = cpu_to_be64(v);
preempt_enable();
gpte->eaddr = eaddr;
@@ -583,7 +583,8 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
unsigned long ea, unsigned long dsisr)
{
struct kvm *kvm = vcpu->kvm;
- unsigned long *hptep, hpte[3], r;
+ unsigned long hpte[3], r;
+ __be64 *hptep;
unsigned long mmu_seq, psize, pte_size;
unsigned long gpa_base, gfn_base;
unsigned long gpa, gfn, hva, pfn;
@@ -606,16 +607,16 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
if (ea != vcpu->arch.pgfault_addr)
return RESUME_GUEST;
index = vcpu->arch.pgfault_index;
- hptep = (unsigned long *)(kvm->arch.hpt_virt + (index << 4));
+ hptep = (__be64 *)(kvm->arch.hpt_virt + (index << 4));
rev = &kvm->arch.revmap[index];
preempt_disable();
while (!try_lock_hpte(hptep, HPTE_V_HVLOCK))
cpu_relax();
- hpte[0] = hptep[0] & ~HPTE_V_HVLOCK;
- hpte[1] = hptep[1];
+ hpte[0] = be64_to_cpu(hptep[0]) & ~HPTE_V_HVLOCK;
+ hpte[1] = be64_to_cpu(hptep[1]);
hpte[2] = r = rev->guest_rpte;
asm volatile("lwsync" : : : "memory");
- hptep[0] = hpte[0];
+ hptep[0] = cpu_to_be64(hpte[0]);
preempt_enable();
if (hpte[0] != vcpu->arch.pgfault_hpte[0] ||
@@ -731,8 +732,9 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
preempt_disable();
while (!try_lock_hpte(hptep, HPTE_V_HVLOCK))
cpu_relax();
- if ((hptep[0] & ~HPTE_V_HVLOCK) != hpte[0] || hptep[1] != hpte[1] ||
- rev->guest_rpte != hpte[2])
+ if ((be64_to_cpu(hptep[0]) & ~HPTE_V_HVLOCK) != hpte[0] ||
+ be64_to_cpu(hptep[1]) != hpte[1] ||
+ rev->guest_rpte != hpte[2])
/* HPTE has been changed under us; let the guest retry */
goto out_unlock;
hpte[0] = (hpte[0] & ~HPTE_V_ABSENT) | HPTE_V_VALID;
@@ -752,20 +754,20 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
rcbits = *rmap >> KVMPPC_RMAP_RC_SHIFT;
r &= rcbits | ~(HPTE_R_R | HPTE_R_C);
- if (hptep[0] & HPTE_V_VALID) {
+ if (be64_to_cpu(hptep[0]) & HPTE_V_VALID) {
/* HPTE was previously valid, so we need to invalidate it */
unlock_rmap(rmap);
- hptep[0] |= HPTE_V_ABSENT;
+ hptep[0] |= cpu_to_be64(HPTE_V_ABSENT);
kvmppc_invalidate_hpte(kvm, hptep, index);
/* don't lose previous R and C bits */
- r |= hptep[1] & (HPTE_R_R | HPTE_R_C);
+ r |= be64_to_cpu(hptep[1]) & (HPTE_R_R | HPTE_R_C);
} else {
kvmppc_add_revmap_chain(kvm, rev, rmap, index, 0);
}
- hptep[1] = r;
+ hptep[1] = cpu_to_be64(r);
eieio();
- hptep[0] = hpte[0];
+ hptep[0] = cpu_to_be64(hpte[0]);
asm volatile("ptesync" : : : "memory");
preempt_enable();
if (page && hpte_is_writable(r))
@@ -784,7 +786,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
return ret;
out_unlock:
- hptep[0] &= ~HPTE_V_HVLOCK;
+ hptep[0] &= ~cpu_to_be64(HPTE_V_HVLOCK);
preempt_enable();
goto out_put;
}
@@ -860,7 +862,7 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
{
struct revmap_entry *rev = kvm->arch.revmap;
unsigned long h, i, j;
- unsigned long *hptep;
+ __be64 *hptep;
unsigned long ptel, psize, rcbits;
for (;;) {
@@ -876,11 +878,11 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
* rmap chain lock.
*/
i = *rmapp & KVMPPC_RMAP_INDEX;
- hptep = (unsigned long *) (kvm->arch.hpt_virt + (i << 4));
+ hptep = (__be64 *) (kvm->arch.hpt_virt + (i << 4));
if (!try_lock_hpte(hptep, HPTE_V_HVLOCK)) {
/* unlock rmap before spinning on the HPTE lock */
unlock_rmap(rmapp);
- while (hptep[0] & HPTE_V_HVLOCK)
+ while (be64_to_cpu(hptep[0]) & HPTE_V_HVLOCK)
cpu_relax();
continue;
}
@@ -899,14 +901,14 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
/* Now check and modify the HPTE */
ptel = rev[i].guest_rpte;
- psize = hpte_page_size(hptep[0], ptel);
- if ((hptep[0] & HPTE_V_VALID) &&
+ psize = hpte_page_size(be64_to_cpu(hptep[0]), ptel);
+ if ((be64_to_cpu(hptep[0]) & HPTE_V_VALID) &&
hpte_rpn(ptel, psize) == gfn) {
if (kvm->arch.using_mmu_notifiers)
- hptep[0] |= HPTE_V_ABSENT;
+ hptep[0] |= cpu_to_be64(HPTE_V_ABSENT);
kvmppc_invalidate_hpte(kvm, hptep, i);
/* Harvest R and C */
- rcbits = hptep[1] & (HPTE_R_R | HPTE_R_C);
+ rcbits = be64_to_cpu(hptep[1]) & (HPTE_R_R | HPTE_R_C);
*rmapp |= rcbits << KVMPPC_RMAP_RC_SHIFT;
if (rcbits & ~rev[i].guest_rpte) {
rev[i].guest_rpte = ptel | rcbits;
@@ -914,7 +916,7 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
}
}
unlock_rmap(rmapp);
- hptep[0] &= ~HPTE_V_HVLOCK;
+ hptep[0] &= ~cpu_to_be64(HPTE_V_HVLOCK);
}
return 0;
}
@@ -961,7 +963,7 @@ static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
{
struct revmap_entry *rev = kvm->arch.revmap;
unsigned long head, i, j;
- unsigned long *hptep;
+ __be64 *hptep;
int ret = 0;
retry:
@@ -977,23 +979,24 @@ static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
i = head = *rmapp & KVMPPC_RMAP_INDEX;
do {
- hptep = (unsigned long *) (kvm->arch.hpt_virt + (i << 4));
+ hptep = (__be64 *) (kvm->arch.hpt_virt + (i << 4));
j = rev[i].forw;
/* If this HPTE isn't referenced, ignore it */
- if (!(hptep[1] & HPTE_R_R))
+ if (!(be64_to_cpu(hptep[1]) & HPTE_R_R))
continue;
if (!try_lock_hpte(hptep, HPTE_V_HVLOCK)) {
/* unlock rmap before spinning on the HPTE lock */
unlock_rmap(rmapp);
- while (hptep[0] & HPTE_V_HVLOCK)
+ while (be64_to_cpu(hptep[0]) & HPTE_V_HVLOCK)
cpu_relax();
goto retry;
}
/* Now check and modify the HPTE */
- if ((hptep[0] & HPTE_V_VALID) && (hptep[1] & HPTE_R_R)) {
+ if ((be64_to_cpu(hptep[0]) & HPTE_V_VALID) &&
+ (be64_to_cpu(hptep[1]) & HPTE_R_R)) {
kvmppc_clear_ref_hpte(kvm, hptep, i);
if (!(rev[i].guest_rpte & HPTE_R_R)) {
rev[i].guest_rpte |= HPTE_R_R;
@@ -1001,7 +1004,7 @@ static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
}
ret = 1;
}
- hptep[0] &= ~HPTE_V_HVLOCK;
+ hptep[0] &= ~cpu_to_be64(HPTE_V_HVLOCK);
} while ((i = j) != head);
unlock_rmap(rmapp);
@@ -1035,7 +1038,7 @@ static int kvm_test_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
do {
hp = (unsigned long *)(kvm->arch.hpt_virt + (i << 4));
j = rev[i].forw;
- if (hp[1] & HPTE_R_R)
+ if (be64_to_cpu(hp[1]) & HPTE_R_R)
goto out;
} while ((i = j) != head);
}
@@ -1075,7 +1078,7 @@ static int kvm_test_clear_dirty_npages(struct kvm *kvm, unsigned long *rmapp)
unsigned long head, i, j;
unsigned long n;
unsigned long v, r;
- unsigned long *hptep;
+ __be64 *hptep;
int npages_dirty = 0;
retry:
@@ -1091,7 +1094,8 @@ static int kvm_test_clear_dirty_npages(struct kvm *kvm, unsigned long *rmapp)
i = head = *rmapp & KVMPPC_RMAP_INDEX;
do {
- hptep = (unsigned long *) (kvm->arch.hpt_virt + (i << 4));
+ unsigned long hptep1;
+ hptep = (__be64 *) (kvm->arch.hpt_virt + (i << 4));
j = rev[i].forw;
/*
@@ -1108,29 +1112,30 @@ static int kvm_test_clear_dirty_npages(struct kvm *kvm, unsigned long *rmapp)
* Otherwise we need to do the tlbie even if C==0 in
* order to pick up any delayed writeback of C.
*/
- if (!(hptep[1] & HPTE_R_C) &&
- (!hpte_is_writable(hptep[1]) || vcpus_running(kvm)))
+ hptep1 = be64_to_cpu(hptep[1]);
+ if (!(hptep1 & HPTE_R_C) &&
+ (!hpte_is_writable(hptep1) || vcpus_running(kvm)))
continue;
if (!try_lock_hpte(hptep, HPTE_V_HVLOCK)) {
/* unlock rmap before spinning on the HPTE lock */
unlock_rmap(rmapp);
- while (hptep[0] & HPTE_V_HVLOCK)
+ while (hptep[0] & cpu_to_be64(HPTE_V_HVLOCK))
cpu_relax();
goto retry;
}
/* Now check and modify the HPTE */
- if (!(hptep[0] & HPTE_V_VALID))
+ if (!(hptep[0] & cpu_to_be64(HPTE_V_VALID)))
continue;
/* need to make it temporarily absent so C is stable */
- hptep[0] |= HPTE_V_ABSENT;
+ hptep[0] |= cpu_to_be64(HPTE_V_ABSENT);
kvmppc_invalidate_hpte(kvm, hptep, i);
- v = hptep[0];
- r = hptep[1];
+ v = be64_to_cpu(hptep[0]);
+ r = be64_to_cpu(hptep[1]);
if (r & HPTE_R_C) {
- hptep[1] = r & ~HPTE_R_C;
+ hptep[1] = cpu_to_be64(r & ~HPTE_R_C);
if (!(rev[i].guest_rpte & HPTE_R_C)) {
rev[i].guest_rpte |= HPTE_R_C;
note_hpte_modification(kvm, &rev[i]);
@@ -1143,7 +1148,7 @@ static int kvm_test_clear_dirty_npages(struct kvm *kvm, unsigned long *rmapp)
}
v &= ~(HPTE_V_ABSENT | HPTE_V_HVLOCK);
v |= HPTE_V_VALID;
- hptep[0] = v;
+ hptep[0] = cpu_to_be64(v);
} while ((i = j) != head);
unlock_rmap(rmapp);
@@ -1307,7 +1312,7 @@ struct kvm_htab_ctx {
* Returns 1 if this HPT entry has been modified or has pending
* R/C bit changes.
*/
-static int hpte_dirty(struct revmap_entry *revp, unsigned long *hptp)
+static int hpte_dirty(struct revmap_entry *revp, __be64 *hptp)
{
unsigned long rcbits_unset;
@@ -1316,13 +1321,14 @@ static int hpte_dirty(struct revmap_entry *revp, unsigned long *hptp)
/* Also need to consider changes in reference and changed bits */
rcbits_unset = ~revp->guest_rpte & (HPTE_R_R | HPTE_R_C);
- if ((hptp[0] & HPTE_V_VALID) && (hptp[1] & rcbits_unset))
+ if ((be64_to_cpu(hptp[0]) & HPTE_V_VALID) &&
+ (be64_to_cpu(hptp[1]) & rcbits_unset))
return 1;
return 0;
}
-static long record_hpte(unsigned long flags, unsigned long *hptp,
+static long record_hpte(unsigned long flags, __be64 *hptp,
unsigned long *hpte, struct revmap_entry *revp,
int want_valid, int first_pass)
{
@@ -1337,10 +1343,10 @@ static long record_hpte(unsigned long flags, unsigned long *hptp,
return 0;
valid = 0;
- if (hptp[0] & (HPTE_V_VALID | HPTE_V_ABSENT)) {
+ if (be64_to_cpu(hptp[0]) & (HPTE_V_VALID | HPTE_V_ABSENT)) {
valid = 1;
if ((flags & KVM_GET_HTAB_BOLTED_ONLY) &&
- !(hptp[0] & HPTE_V_BOLTED))
+ !(be64_to_cpu(hptp[0]) & HPTE_V_BOLTED))
valid = 0;
}
if (valid != want_valid)
@@ -1352,7 +1358,7 @@ static long record_hpte(unsigned long flags, unsigned long *hptp,
preempt_disable();
while (!try_lock_hpte(hptp, HPTE_V_HVLOCK))
cpu_relax();
- v = hptp[0];
+ v = be64_to_cpu(hptp[0]);
/* re-evaluate valid and dirty from synchronized HPTE value */
valid = !!(v & HPTE_V_VALID);
@@ -1360,9 +1366,9 @@ static long record_hpte(unsigned long flags, unsigned long *hptp,
/* Harvest R and C into guest view if necessary */
rcbits_unset = ~revp->guest_rpte & (HPTE_R_R | HPTE_R_C);
- if (valid && (rcbits_unset & hptp[1])) {
- revp->guest_rpte |= (hptp[1] & (HPTE_R_R | HPTE_R_C)) |
- HPTE_GR_MODIFIED;
+ if (valid && (rcbits_unset & be64_to_cpu(hptp[1]))) {
+ revp->guest_rpte |= (be64_to_cpu(hptp[1]) &
+ (HPTE_R_R | HPTE_R_C)) | HPTE_GR_MODIFIED;
dirty = 1;
}
@@ -1381,13 +1387,13 @@ static long record_hpte(unsigned long flags, unsigned long *hptp,
revp->guest_rpte = r;
}
asm volatile(PPC_RELEASE_BARRIER "" : : : "memory");
- hptp[0] &= ~HPTE_V_HVLOCK;
+ hptp[0] &= ~cpu_to_be64(HPTE_V_HVLOCK);
preempt_enable();
if (!(valid == want_valid && (first_pass || dirty)))
ok = 0;
}
- hpte[0] = v;
- hpte[1] = r;
+ hpte[0] = cpu_to_be64(v);
+ hpte[1] = cpu_to_be64(r);
return ok;
}
@@ -1397,7 +1403,7 @@ static ssize_t kvm_htab_read(struct file *file, char __user *buf,
struct kvm_htab_ctx *ctx = file->private_data;
struct kvm *kvm = ctx->kvm;
struct kvm_get_htab_header hdr;
- unsigned long *hptp;
+ __be64 *hptp;
struct revmap_entry *revp;
unsigned long i, nb, nw;
unsigned long __user *lbuf;
@@ -1413,7 +1419,7 @@ static ssize_t kvm_htab_read(struct file *file, char __user *buf,
flags = ctx->flags;
i = ctx->index;
- hptp = (unsigned long *)(kvm->arch.hpt_virt + (i * HPTE_SIZE));
+ hptp = (__be64 *)(kvm->arch.hpt_virt + (i * HPTE_SIZE));
revp = kvm->arch.revmap + i;
lbuf = (unsigned long __user *)buf;
@@ -1497,7 +1503,7 @@ static ssize_t kvm_htab_write(struct file *file, const char __user *buf,
unsigned long i, j;
unsigned long v, r;
unsigned long __user *lbuf;
- unsigned long *hptp;
+ __be64 *hptp;
unsigned long tmp[2];
ssize_t nb;
long int err, ret;
@@ -1539,7 +1545,7 @@ static ssize_t kvm_htab_write(struct file *file, const char __user *buf,
i + hdr.n_valid + hdr.n_invalid > kvm->arch.hpt_npte)
break;
- hptp = (unsigned long *)(kvm->arch.hpt_virt + (i * HPTE_SIZE));
+ hptp = (__be64 *)(kvm->arch.hpt_virt + (i * HPTE_SIZE));
lbuf = (unsigned long __user *)buf;
for (j = 0; j < hdr.n_valid; ++j) {
err = -EFAULT;
@@ -1551,7 +1557,7 @@ static ssize_t kvm_htab_write(struct file *file, const char __user *buf,
lbuf += 2;
nb += HPTE_SIZE;
- if (hptp[0] & (HPTE_V_VALID | HPTE_V_ABSENT))
+ if (be64_to_cpu(hptp[0]) & (HPTE_V_VALID | HPTE_V_ABSENT))
kvmppc_do_h_remove(kvm, 0, i, 0, tmp);
err = -EIO;
ret = kvmppc_virtmode_do_h_enter(kvm, H_EXACT, i, v, r,
@@ -1577,7 +1583,7 @@ static ssize_t kvm_htab_write(struct file *file, const char __user *buf,
}
for (j = 0; j < hdr.n_invalid; ++j) {
- if (hptp[0] & (HPTE_V_VALID | HPTE_V_ABSENT))
+ if (be64_to_cpu(hptp[0]) & (HPTE_V_VALID | HPTE_V_ABSENT))
kvmppc_do_h_remove(kvm, 0, i, 0, tmp);
++i;
hptp += 2;
diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
index 6e62243..e5c6063 100644
--- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c
+++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
@@ -154,10 +154,10 @@ static pte_t lookup_linux_pte_and_update(pgd_t *pgdir, unsigned long hva,
return kvmppc_read_update_linux_pte(ptep, writing, hugepage_shift);
}
-static inline void unlock_hpte(unsigned long *hpte, unsigned long hpte_v)
+static inline void unlock_hpte(__be64 *hpte, unsigned long hpte_v)
{
asm volatile(PPC_RELEASE_BARRIER "" : : : "memory");
- hpte[0] = hpte_v;
+ hpte[0] = cpu_to_be64(hpte_v);
}
long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
@@ -166,7 +166,7 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
{
unsigned long i, pa, gpa, gfn, psize;
unsigned long slot_fn, hva;
- unsigned long *hpte;
+ __be64 *hpte;
struct revmap_entry *rev;
unsigned long g_ptel;
struct kvm_memory_slot *memslot;
@@ -275,9 +275,9 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
return H_PARAMETER;
if (likely((flags & H_EXACT) == 0)) {
pte_index &= ~7UL;
- hpte = (unsigned long *)(kvm->arch.hpt_virt + (pte_index << 4));
+ hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4));
for (i = 0; i < 8; ++i) {
- if ((*hpte & HPTE_V_VALID) == 0 &&
+ if ((be64_to_cpu(*hpte) & HPTE_V_VALID) == 0 &&
try_lock_hpte(hpte, HPTE_V_HVLOCK | HPTE_V_VALID |
HPTE_V_ABSENT))
break;
@@ -292,11 +292,13 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
*/
hpte -= 16;
for (i = 0; i < 8; ++i) {
+ u64 pte;
while (!try_lock_hpte(hpte, HPTE_V_HVLOCK))
cpu_relax();
- if (!(*hpte & (HPTE_V_VALID | HPTE_V_ABSENT)))
+ pte = be64_to_cpu(*hpte);
+ if (!(pte & (HPTE_V_VALID | HPTE_V_ABSENT)))
break;
- *hpte &= ~HPTE_V_HVLOCK;
+ *hpte &= ~cpu_to_be64(HPTE_V_HVLOCK);
hpte += 2;
}
if (i == 8)
@@ -304,14 +306,17 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
}
pte_index += i;
} else {
- hpte = (unsigned long *)(kvm->arch.hpt_virt + (pte_index << 4));
+ hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4));
if (!try_lock_hpte(hpte, HPTE_V_HVLOCK | HPTE_V_VALID |
HPTE_V_ABSENT)) {
/* Lock the slot and check again */
+ u64 pte;
+
while (!try_lock_hpte(hpte, HPTE_V_HVLOCK))
cpu_relax();
- if (*hpte & (HPTE_V_VALID | HPTE_V_ABSENT)) {
- *hpte &= ~HPTE_V_HVLOCK;
+ pte = be64_to_cpu(*hpte);
+ if (pte & (HPTE_V_VALID | HPTE_V_ABSENT)) {
+ *hpte &= ~cpu_to_be64(HPTE_V_HVLOCK);
return H_PTEG_FULL;
}
}
@@ -347,11 +352,11 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
}
}
- hpte[1] = ptel;
+ hpte[1] = cpu_to_be64(ptel);
/* Write the first HPTE dword, unlocking the HPTE and making it valid */
eieio();
- hpte[0] = pteh;
+ hpte[0] = cpu_to_be64(pteh);
asm volatile("ptesync" : : : "memory");
*pte_idx_ret = pte_index;
@@ -468,30 +473,35 @@ long kvmppc_do_h_remove(struct kvm *kvm, unsigned long flags,
unsigned long pte_index, unsigned long avpn,
unsigned long *hpret)
{
- unsigned long *hpte;
+ __be64 *hpte;
unsigned long v, r, rb;
struct revmap_entry *rev;
+ u64 pte;
if (pte_index >= kvm->arch.hpt_npte)
return H_PARAMETER;
- hpte = (unsigned long *)(kvm->arch.hpt_virt + (pte_index << 4));
+ hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4));
while (!try_lock_hpte(hpte, HPTE_V_HVLOCK))
cpu_relax();
- if ((hpte[0] & (HPTE_V_ABSENT | HPTE_V_VALID)) == 0 ||
- ((flags & H_AVPN) && (hpte[0] & ~0x7fUL) != avpn) ||
- ((flags & H_ANDCOND) && (hpte[0] & avpn) != 0)) {
- hpte[0] &= ~HPTE_V_HVLOCK;
+ pte = be64_to_cpu(hpte[0]);
+ if ((pte & (HPTE_V_ABSENT | HPTE_V_VALID)) == 0 ||
+ ((flags & H_AVPN) && (pte & ~0x7fUL) != avpn) ||
+ ((flags & H_ANDCOND) && (pte & avpn) != 0)) {
+ hpte[0] &= ~cpu_to_be64(HPTE_V_HVLOCK);
return H_NOT_FOUND;
}
rev = real_vmalloc_addr(&kvm->arch.revmap[pte_index]);
- v = hpte[0] & ~HPTE_V_HVLOCK;
+ v = pte & ~HPTE_V_HVLOCK;
if (v & HPTE_V_VALID) {
- hpte[0] &= ~HPTE_V_VALID;
- rb = compute_tlbie_rb(v, hpte[1], pte_index);
+ u64 pte1;
+
+ pte1 = be64_to_cpu(hpte[1]);
+ hpte[0] &= ~cpu_to_be64(HPTE_V_VALID);
+ rb = compute_tlbie_rb(v, pte1, pte_index);
do_tlbies(kvm, &rb, 1, global_invalidates(kvm, flags), true);
/* Read PTE low word after tlbie to get final R/C values */
- remove_revmap_chain(kvm, pte_index, rev, v, hpte[1]);
+ remove_revmap_chain(kvm, pte_index, rev, v, pte1);
}
r = rev->guest_rpte & ~HPTE_GR_RESERVED;
note_hpte_modification(kvm, rev);
@@ -514,12 +524,14 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
{
struct kvm *kvm = vcpu->kvm;
unsigned long *args = &vcpu->arch.gpr[4];
- unsigned long *hp, *hptes[4], tlbrb[4];
+ __be64 *hp, *hptes[4];
+ unsigned long tlbrb[4];
long int i, j, k, n, found, indexes[4];
unsigned long flags, req, pte_index, rcbits;
int global;
long int ret = H_SUCCESS;
struct revmap_entry *rev, *revs[4];
+ u64 hp0;
global = global_invalidates(kvm, 0);
for (i = 0; i < 4 && ret == H_SUCCESS; ) {
@@ -542,8 +554,7 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
ret = H_PARAMETER;
break;
}
- hp = (unsigned long *)
- (kvm->arch.hpt_virt + (pte_index << 4));
+ hp = (__be64 *) (kvm->arch.hpt_virt + (pte_index << 4));
/* to avoid deadlock, don't spin except for first */
if (!try_lock_hpte(hp, HPTE_V_HVLOCK)) {
if (n)
@@ -552,23 +563,24 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
cpu_relax();
}
found = 0;
- if (hp[0] & (HPTE_V_ABSENT | HPTE_V_VALID)) {
+ hp0 = be64_to_cpu(hp[0]);
+ if (hp0 & (HPTE_V_ABSENT | HPTE_V_VALID)) {
switch (flags & 3) {
case 0: /* absolute */
found = 1;
break;
case 1: /* andcond */
- if (!(hp[0] & args[j + 1]))
+ if (!(hp0 & args[j + 1]))
found = 1;
break;
case 2: /* AVPN */
- if ((hp[0] & ~0x7fUL) == args[j + 1])
+ if ((hp0 & ~0x7fUL) == args[j + 1])
found = 1;
break;
}
}
if (!found) {
- hp[0] &= ~HPTE_V_HVLOCK;
+ hp[0] &= ~cpu_to_be64(HPTE_V_HVLOCK);
args[j] = ((0x90 | flags) << 56) + pte_index;
continue;
}
@@ -577,7 +589,7 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
rev = real_vmalloc_addr(&kvm->arch.revmap[pte_index]);
note_hpte_modification(kvm, rev);
- if (!(hp[0] & HPTE_V_VALID)) {
+ if (!(hp0 & HPTE_V_VALID)) {
/* insert R and C bits from PTE */
rcbits = rev->guest_rpte & (HPTE_R_R|HPTE_R_C);
args[j] |= rcbits << (56 - 5);
@@ -585,8 +597,10 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
continue;
}
- hp[0] &= ~HPTE_V_VALID; /* leave it locked */
- tlbrb[n] = compute_tlbie_rb(hp[0], hp[1], pte_index);
+ /* leave it locked */
+ hp[0] &= ~cpu_to_be64(HPTE_V_VALID);
+ tlbrb[n] = compute_tlbie_rb(be64_to_cpu(hp[0]),
+ be64_to_cpu(hp[1]), pte_index);
indexes[n] = j;
hptes[n] = hp;
revs[n] = rev;
@@ -605,7 +619,8 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
pte_index = args[j] & ((1ul << 56) - 1);
hp = hptes[k];
rev = revs[k];
- remove_revmap_chain(kvm, pte_index, rev, hp[0], hp[1]);
+ remove_revmap_chain(kvm, pte_index, rev,
+ be64_to_cpu(hp[0]), be64_to_cpu(hp[1]));
rcbits = rev->guest_rpte & (HPTE_R_R|HPTE_R_C);
args[j] |= rcbits << (56 - 5);
hp[0] = 0;
@@ -620,23 +635,25 @@ long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long va)
{
struct kvm *kvm = vcpu->kvm;
- unsigned long *hpte;
+ __be64 *hpte;
struct revmap_entry *rev;
unsigned long v, r, rb, mask, bits;
+ u64 pte;
if (pte_index >= kvm->arch.hpt_npte)
return H_PARAMETER;
- hpte = (unsigned long *)(kvm->arch.hpt_virt + (pte_index << 4));
+ hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4));
while (!try_lock_hpte(hpte, HPTE_V_HVLOCK))
cpu_relax();
- if ((hpte[0] & (HPTE_V_ABSENT | HPTE_V_VALID)) == 0 ||
- ((flags & H_AVPN) && (hpte[0] & ~0x7fUL) != avpn)) {
- hpte[0] &= ~HPTE_V_HVLOCK;
+ pte = be64_to_cpu(hpte[0]);
+ if ((pte & (HPTE_V_ABSENT | HPTE_V_VALID)) == 0 ||
+ ((flags & H_AVPN) && (pte & ~0x7fUL) != avpn)) {
+ hpte[0] &= ~cpu_to_be64(HPTE_V_HVLOCK);
return H_NOT_FOUND;
}
- v = hpte[0];
+ v = pte;
bits = (flags << 55) & HPTE_R_PP0;
bits |= (flags << 48) & HPTE_R_KEY_HI;
bits |= flags & (HPTE_R_PP | HPTE_R_N | HPTE_R_KEY_LO);
@@ -650,12 +667,12 @@ long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags,
rev->guest_rpte = r;
note_hpte_modification(kvm, rev);
}
- r = (hpte[1] & ~mask) | bits;
+ r = (be64_to_cpu(hpte[1]) & ~mask) | bits;
/* Update HPTE */
if (v & HPTE_V_VALID) {
rb = compute_tlbie_rb(v, r, pte_index);
- hpte[0] = v & ~HPTE_V_VALID;
+ hpte[0] = cpu_to_be64(v & ~HPTE_V_VALID);
do_tlbies(kvm, &rb, 1, global_invalidates(kvm, flags), true);
/*
* If the host has this page as readonly but the guest
@@ -681,9 +698,9 @@ long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags,
}
}
}
- hpte[1] = r;
+ hpte[1] = cpu_to_be64(r);
eieio();
- hpte[0] = v & ~HPTE_V_HVLOCK;
+ hpte[0] = cpu_to_be64(v & ~HPTE_V_HVLOCK);
asm volatile("ptesync" : : : "memory");
return H_SUCCESS;
}
@@ -692,7 +709,8 @@ long kvmppc_h_read(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long pte_index)
{
struct kvm *kvm = vcpu->kvm;
- unsigned long *hpte, v, r;
+ __be64 *hpte;
+ unsigned long v, r;
int i, n = 1;
struct revmap_entry *rev = NULL;
@@ -704,9 +722,9 @@ long kvmppc_h_read(struct kvm_vcpu *vcpu, unsigned long flags,
}
rev = real_vmalloc_addr(&kvm->arch.revmap[pte_index]);
for (i = 0; i < n; ++i, ++pte_index) {
- hpte = (unsigned long *)(kvm->arch.hpt_virt + (pte_index << 4));
- v = hpte[0] & ~HPTE_V_HVLOCK;
- r = hpte[1];
+ hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4));
+ v = be64_to_cpu(hpte[0]) & ~HPTE_V_HVLOCK;
+ r = be64_to_cpu(hpte[1]);
if (v & HPTE_V_ABSENT) {
v &= ~HPTE_V_ABSENT;
v |= HPTE_V_VALID;
@@ -721,25 +739,27 @@ long kvmppc_h_read(struct kvm_vcpu *vcpu, unsigned long flags,
return H_SUCCESS;
}
-void kvmppc_invalidate_hpte(struct kvm *kvm, unsigned long *hptep,
+void kvmppc_invalidate_hpte(struct kvm *kvm, __be64 *hptep,
unsigned long pte_index)
{
unsigned long rb;
- hptep[0] &= ~HPTE_V_VALID;
- rb = compute_tlbie_rb(hptep[0], hptep[1], pte_index);
+ hptep[0] &= ~cpu_to_be64(HPTE_V_VALID);
+ rb = compute_tlbie_rb(be64_to_cpu(hptep[0]), be64_to_cpu(hptep[1]),
+ pte_index);
do_tlbies(kvm, &rb, 1, 1, true);
}
EXPORT_SYMBOL_GPL(kvmppc_invalidate_hpte);
-void kvmppc_clear_ref_hpte(struct kvm *kvm, unsigned long *hptep,
+void kvmppc_clear_ref_hpte(struct kvm *kvm, __be64 *hptep,
unsigned long pte_index)
{
unsigned long rb;
unsigned char rbyte;
- rb = compute_tlbie_rb(hptep[0], hptep[1], pte_index);
- rbyte = (hptep[1] & ~HPTE_R_R) >> 8;
+ rb = compute_tlbie_rb(be64_to_cpu(hptep[0]), be64_to_cpu(hptep[1]),
+ pte_index);
+ rbyte = (be64_to_cpu(hptep[1]) & ~HPTE_R_R) >> 8;
/* modify only the second-last byte, which contains the ref bit */
*((char *)hptep + 14) = rbyte;
do_tlbies(kvm, &rb, 1, 1, false);
@@ -765,7 +785,7 @@ long kvmppc_hv_find_lock_hpte(struct kvm *kvm, gva_t eaddr, unsigned long slb_v,
unsigned long somask;
unsigned long vsid, hash;
unsigned long avpn;
- unsigned long *hpte;
+ __be64 *hpte;
unsigned long mask, val;
unsigned long v, r;
@@ -797,11 +817,11 @@ long kvmppc_hv_find_lock_hpte(struct kvm *kvm, gva_t eaddr, unsigned long slb_v,
val |= avpn;
for (;;) {
- hpte = (unsigned long *)(kvm->arch.hpt_virt + (hash << 7));
+ hpte = (__be64 *)(kvm->arch.hpt_virt + (hash << 7));
for (i = 0; i < 16; i += 2) {
/* Read the PTE racily */
- v = hpte[i] & ~HPTE_V_HVLOCK;
+ v = be64_to_cpu(hpte[i]) & ~HPTE_V_HVLOCK;
/* Check valid/absent, hash, segment size and AVPN */
if (!(v & valid) || (v & mask) != val)
@@ -810,8 +830,8 @@ long kvmppc_hv_find_lock_hpte(struct kvm *kvm, gva_t eaddr, unsigned long slb_v,
/* Lock the PTE and read it under the lock */
while (!try_lock_hpte(&hpte[i], HPTE_V_HVLOCK))
cpu_relax();
- v = hpte[i] & ~HPTE_V_HVLOCK;
- r = hpte[i+1];
+ v = be64_to_cpu(hpte[i]) & ~HPTE_V_HVLOCK;
+ r = be64_to_cpu(hpte[i+1]);
/*
* Check the HPTE again, including large page size
@@ -825,7 +845,7 @@ long kvmppc_hv_find_lock_hpte(struct kvm *kvm, gva_t eaddr, unsigned long slb_v,
return (hash << 3) + (i >> 1);
/* Unlock and move on */
- hpte[i] = v;
+ hpte[i] = cpu_to_be64(v);
}
if (val & HPTE_V_SECONDARY)
@@ -854,7 +874,7 @@ long kvmppc_hpte_hv_fault(struct kvm_vcpu *vcpu, unsigned long addr,
struct kvm *kvm = vcpu->kvm;
long int index;
unsigned long v, r, gr;
- unsigned long *hpte;
+ __be64 *hpte;
unsigned long valid;
struct revmap_entry *rev;
unsigned long pp, key;
@@ -870,9 +890,9 @@ long kvmppc_hpte_hv_fault(struct kvm_vcpu *vcpu, unsigned long addr,
return status; /* there really was no HPTE */
return 0; /* for prot fault, HPTE disappeared */
}
- hpte = (unsigned long *)(kvm->arch.hpt_virt + (index << 4));
- v = hpte[0] & ~HPTE_V_HVLOCK;
- r = hpte[1];
+ hpte = (__be64 *)(kvm->arch.hpt_virt + (index << 4));
+ v = be64_to_cpu(hpte[0]) & ~HPTE_V_HVLOCK;
+ r = be64_to_cpu(hpte[1]);
rev = real_vmalloc_addr(&kvm->arch.revmap[index]);
gr = rev->guest_rpte;
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* [PATCH 2/7] KVM: PPC: Book3S HV: Make HTAB code LE host aware
@ 2014-06-17 16:17 ` Alexander Graf
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
When running on an LE host all data structures are kept in little endian
byte order. However, the HTAB still needs to be maintained in big endian.
So every time we access any HTAB we need to make sure we do so in the right
byte order. Fix up all accesses to manually byte swap.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
v1 -> v2:
- Add __be32 hints
- Fix H_REMOVE
---
arch/powerpc/include/asm/kvm_book3s.h | 4 +-
arch/powerpc/include/asm/kvm_book3s_64.h | 15 +++-
arch/powerpc/kvm/book3s_64_mmu_hv.c | 128 ++++++++++++++-------------
arch/powerpc/kvm/book3s_hv_rm_mmu.c | 146 ++++++++++++++++++-------------
4 files changed, 164 insertions(+), 129 deletions(-)
diff --git a/arch/powerpc/include/asm/kvm_book3s.h b/arch/powerpc/include/asm/kvm_book3s.h
index a20cc0b..028f4b1 100644
--- a/arch/powerpc/include/asm/kvm_book3s.h
+++ b/arch/powerpc/include/asm/kvm_book3s.h
@@ -161,9 +161,9 @@ extern pfn_t kvmppc_gfn_to_pfn(struct kvm_vcpu *vcpu, gfn_t gfn, bool writing,
bool *writable);
extern void kvmppc_add_revmap_chain(struct kvm *kvm, struct revmap_entry *rev,
unsigned long *rmap, long pte_index, int realmode);
-extern void kvmppc_invalidate_hpte(struct kvm *kvm, unsigned long *hptep,
+extern void kvmppc_invalidate_hpte(struct kvm *kvm, __be64 *hptep,
unsigned long pte_index);
-void kvmppc_clear_ref_hpte(struct kvm *kvm, unsigned long *hptep,
+void kvmppc_clear_ref_hpte(struct kvm *kvm, __be64 *hptep,
unsigned long pte_index);
extern void *kvmppc_pin_guest_page(struct kvm *kvm, unsigned long addr,
unsigned long *nb_ret);
diff --git a/arch/powerpc/include/asm/kvm_book3s_64.h b/arch/powerpc/include/asm/kvm_book3s_64.h
index c7871f3..e504f88 100644
--- a/arch/powerpc/include/asm/kvm_book3s_64.h
+++ b/arch/powerpc/include/asm/kvm_book3s_64.h
@@ -59,20 +59,29 @@ extern unsigned long kvm_rma_pages;
/* These bits are reserved in the guest view of the HPTE */
#define HPTE_GR_RESERVED HPTE_GR_MODIFIED
-static inline long try_lock_hpte(unsigned long *hpte, unsigned long bits)
+static inline long try_lock_hpte(__be64 *hpte, unsigned long bits)
{
unsigned long tmp, old;
+ __be64 be_lockbit, be_bits;
+
+ /*
+ * We load/store in native endian, but the HTAB is in big endian. If
+ * we byte swap all data we apply on the PTE we're implicitly correct
+ * again.
+ */
+ be_lockbit = cpu_to_be64(HPTE_V_HVLOCK);
+ be_bits = cpu_to_be64(bits);
asm volatile(" ldarx %0,0,%2\n"
" and. %1,%0,%3\n"
" bne 2f\n"
- " ori %0,%0,%4\n"
+ " or %0,%0,%4\n"
" stdcx. %0,0,%2\n"
" beq+ 2f\n"
" mr %1,%3\n"
"2: isync"
: "=&r" (tmp), "=&r" (old)
- : "r" (hpte), "r" (bits), "i" (HPTE_V_HVLOCK)
+ : "r" (hpte), "r" (be_bits), "r" (be_lockbit)
: "cc", "memory");
return old = 0;
}
diff --git a/arch/powerpc/kvm/book3s_64_mmu_hv.c b/arch/powerpc/kvm/book3s_64_mmu_hv.c
index 8056107..2d154d9 100644
--- a/arch/powerpc/kvm/book3s_64_mmu_hv.c
+++ b/arch/powerpc/kvm/book3s_64_mmu_hv.c
@@ -450,7 +450,7 @@ static int kvmppc_mmu_book3s_64_hv_xlate(struct kvm_vcpu *vcpu, gva_t eaddr,
unsigned long slb_v;
unsigned long pp, key;
unsigned long v, gr;
- unsigned long *hptep;
+ __be64 *hptep;
int index;
int virtmode = vcpu->arch.shregs.msr & (data ? MSR_DR : MSR_IR);
@@ -473,13 +473,13 @@ static int kvmppc_mmu_book3s_64_hv_xlate(struct kvm_vcpu *vcpu, gva_t eaddr,
preempt_enable();
return -ENOENT;
}
- hptep = (unsigned long *)(kvm->arch.hpt_virt + (index << 4));
- v = hptep[0] & ~HPTE_V_HVLOCK;
+ hptep = (__be64 *)(kvm->arch.hpt_virt + (index << 4));
+ v = be64_to_cpu(hptep[0]) & ~HPTE_V_HVLOCK;
gr = kvm->arch.revmap[index].guest_rpte;
/* Unlock the HPTE */
asm volatile("lwsync" : : : "memory");
- hptep[0] = v;
+ hptep[0] = cpu_to_be64(v);
preempt_enable();
gpte->eaddr = eaddr;
@@ -583,7 +583,8 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
unsigned long ea, unsigned long dsisr)
{
struct kvm *kvm = vcpu->kvm;
- unsigned long *hptep, hpte[3], r;
+ unsigned long hpte[3], r;
+ __be64 *hptep;
unsigned long mmu_seq, psize, pte_size;
unsigned long gpa_base, gfn_base;
unsigned long gpa, gfn, hva, pfn;
@@ -606,16 +607,16 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
if (ea != vcpu->arch.pgfault_addr)
return RESUME_GUEST;
index = vcpu->arch.pgfault_index;
- hptep = (unsigned long *)(kvm->arch.hpt_virt + (index << 4));
+ hptep = (__be64 *)(kvm->arch.hpt_virt + (index << 4));
rev = &kvm->arch.revmap[index];
preempt_disable();
while (!try_lock_hpte(hptep, HPTE_V_HVLOCK))
cpu_relax();
- hpte[0] = hptep[0] & ~HPTE_V_HVLOCK;
- hpte[1] = hptep[1];
+ hpte[0] = be64_to_cpu(hptep[0]) & ~HPTE_V_HVLOCK;
+ hpte[1] = be64_to_cpu(hptep[1]);
hpte[2] = r = rev->guest_rpte;
asm volatile("lwsync" : : : "memory");
- hptep[0] = hpte[0];
+ hptep[0] = cpu_to_be64(hpte[0]);
preempt_enable();
if (hpte[0] != vcpu->arch.pgfault_hpte[0] ||
@@ -731,8 +732,9 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
preempt_disable();
while (!try_lock_hpte(hptep, HPTE_V_HVLOCK))
cpu_relax();
- if ((hptep[0] & ~HPTE_V_HVLOCK) != hpte[0] || hptep[1] != hpte[1] ||
- rev->guest_rpte != hpte[2])
+ if ((be64_to_cpu(hptep[0]) & ~HPTE_V_HVLOCK) != hpte[0] ||
+ be64_to_cpu(hptep[1]) != hpte[1] ||
+ rev->guest_rpte != hpte[2])
/* HPTE has been changed under us; let the guest retry */
goto out_unlock;
hpte[0] = (hpte[0] & ~HPTE_V_ABSENT) | HPTE_V_VALID;
@@ -752,20 +754,20 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
rcbits = *rmap >> KVMPPC_RMAP_RC_SHIFT;
r &= rcbits | ~(HPTE_R_R | HPTE_R_C);
- if (hptep[0] & HPTE_V_VALID) {
+ if (be64_to_cpu(hptep[0]) & HPTE_V_VALID) {
/* HPTE was previously valid, so we need to invalidate it */
unlock_rmap(rmap);
- hptep[0] |= HPTE_V_ABSENT;
+ hptep[0] |= cpu_to_be64(HPTE_V_ABSENT);
kvmppc_invalidate_hpte(kvm, hptep, index);
/* don't lose previous R and C bits */
- r |= hptep[1] & (HPTE_R_R | HPTE_R_C);
+ r |= be64_to_cpu(hptep[1]) & (HPTE_R_R | HPTE_R_C);
} else {
kvmppc_add_revmap_chain(kvm, rev, rmap, index, 0);
}
- hptep[1] = r;
+ hptep[1] = cpu_to_be64(r);
eieio();
- hptep[0] = hpte[0];
+ hptep[0] = cpu_to_be64(hpte[0]);
asm volatile("ptesync" : : : "memory");
preempt_enable();
if (page && hpte_is_writable(r))
@@ -784,7 +786,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
return ret;
out_unlock:
- hptep[0] &= ~HPTE_V_HVLOCK;
+ hptep[0] &= ~cpu_to_be64(HPTE_V_HVLOCK);
preempt_enable();
goto out_put;
}
@@ -860,7 +862,7 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
{
struct revmap_entry *rev = kvm->arch.revmap;
unsigned long h, i, j;
- unsigned long *hptep;
+ __be64 *hptep;
unsigned long ptel, psize, rcbits;
for (;;) {
@@ -876,11 +878,11 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
* rmap chain lock.
*/
i = *rmapp & KVMPPC_RMAP_INDEX;
- hptep = (unsigned long *) (kvm->arch.hpt_virt + (i << 4));
+ hptep = (__be64 *) (kvm->arch.hpt_virt + (i << 4));
if (!try_lock_hpte(hptep, HPTE_V_HVLOCK)) {
/* unlock rmap before spinning on the HPTE lock */
unlock_rmap(rmapp);
- while (hptep[0] & HPTE_V_HVLOCK)
+ while (be64_to_cpu(hptep[0]) & HPTE_V_HVLOCK)
cpu_relax();
continue;
}
@@ -899,14 +901,14 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
/* Now check and modify the HPTE */
ptel = rev[i].guest_rpte;
- psize = hpte_page_size(hptep[0], ptel);
- if ((hptep[0] & HPTE_V_VALID) &&
+ psize = hpte_page_size(be64_to_cpu(hptep[0]), ptel);
+ if ((be64_to_cpu(hptep[0]) & HPTE_V_VALID) &&
hpte_rpn(ptel, psize) = gfn) {
if (kvm->arch.using_mmu_notifiers)
- hptep[0] |= HPTE_V_ABSENT;
+ hptep[0] |= cpu_to_be64(HPTE_V_ABSENT);
kvmppc_invalidate_hpte(kvm, hptep, i);
/* Harvest R and C */
- rcbits = hptep[1] & (HPTE_R_R | HPTE_R_C);
+ rcbits = be64_to_cpu(hptep[1]) & (HPTE_R_R | HPTE_R_C);
*rmapp |= rcbits << KVMPPC_RMAP_RC_SHIFT;
if (rcbits & ~rev[i].guest_rpte) {
rev[i].guest_rpte = ptel | rcbits;
@@ -914,7 +916,7 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
}
}
unlock_rmap(rmapp);
- hptep[0] &= ~HPTE_V_HVLOCK;
+ hptep[0] &= ~cpu_to_be64(HPTE_V_HVLOCK);
}
return 0;
}
@@ -961,7 +963,7 @@ static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
{
struct revmap_entry *rev = kvm->arch.revmap;
unsigned long head, i, j;
- unsigned long *hptep;
+ __be64 *hptep;
int ret = 0;
retry:
@@ -977,23 +979,24 @@ static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
i = head = *rmapp & KVMPPC_RMAP_INDEX;
do {
- hptep = (unsigned long *) (kvm->arch.hpt_virt + (i << 4));
+ hptep = (__be64 *) (kvm->arch.hpt_virt + (i << 4));
j = rev[i].forw;
/* If this HPTE isn't referenced, ignore it */
- if (!(hptep[1] & HPTE_R_R))
+ if (!(be64_to_cpu(hptep[1]) & HPTE_R_R))
continue;
if (!try_lock_hpte(hptep, HPTE_V_HVLOCK)) {
/* unlock rmap before spinning on the HPTE lock */
unlock_rmap(rmapp);
- while (hptep[0] & HPTE_V_HVLOCK)
+ while (be64_to_cpu(hptep[0]) & HPTE_V_HVLOCK)
cpu_relax();
goto retry;
}
/* Now check and modify the HPTE */
- if ((hptep[0] & HPTE_V_VALID) && (hptep[1] & HPTE_R_R)) {
+ if ((be64_to_cpu(hptep[0]) & HPTE_V_VALID) &&
+ (be64_to_cpu(hptep[1]) & HPTE_R_R)) {
kvmppc_clear_ref_hpte(kvm, hptep, i);
if (!(rev[i].guest_rpte & HPTE_R_R)) {
rev[i].guest_rpte |= HPTE_R_R;
@@ -1001,7 +1004,7 @@ static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
}
ret = 1;
}
- hptep[0] &= ~HPTE_V_HVLOCK;
+ hptep[0] &= ~cpu_to_be64(HPTE_V_HVLOCK);
} while ((i = j) != head);
unlock_rmap(rmapp);
@@ -1035,7 +1038,7 @@ static int kvm_test_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
do {
hp = (unsigned long *)(kvm->arch.hpt_virt + (i << 4));
j = rev[i].forw;
- if (hp[1] & HPTE_R_R)
+ if (be64_to_cpu(hp[1]) & HPTE_R_R)
goto out;
} while ((i = j) != head);
}
@@ -1075,7 +1078,7 @@ static int kvm_test_clear_dirty_npages(struct kvm *kvm, unsigned long *rmapp)
unsigned long head, i, j;
unsigned long n;
unsigned long v, r;
- unsigned long *hptep;
+ __be64 *hptep;
int npages_dirty = 0;
retry:
@@ -1091,7 +1094,8 @@ static int kvm_test_clear_dirty_npages(struct kvm *kvm, unsigned long *rmapp)
i = head = *rmapp & KVMPPC_RMAP_INDEX;
do {
- hptep = (unsigned long *) (kvm->arch.hpt_virt + (i << 4));
+ unsigned long hptep1;
+ hptep = (__be64 *) (kvm->arch.hpt_virt + (i << 4));
j = rev[i].forw;
/*
@@ -1108,29 +1112,30 @@ static int kvm_test_clear_dirty_npages(struct kvm *kvm, unsigned long *rmapp)
* Otherwise we need to do the tlbie even if C=0 in
* order to pick up any delayed writeback of C.
*/
- if (!(hptep[1] & HPTE_R_C) &&
- (!hpte_is_writable(hptep[1]) || vcpus_running(kvm)))
+ hptep1 = be64_to_cpu(hptep[1]);
+ if (!(hptep1 & HPTE_R_C) &&
+ (!hpte_is_writable(hptep1) || vcpus_running(kvm)))
continue;
if (!try_lock_hpte(hptep, HPTE_V_HVLOCK)) {
/* unlock rmap before spinning on the HPTE lock */
unlock_rmap(rmapp);
- while (hptep[0] & HPTE_V_HVLOCK)
+ while (hptep[0] & cpu_to_be64(HPTE_V_HVLOCK))
cpu_relax();
goto retry;
}
/* Now check and modify the HPTE */
- if (!(hptep[0] & HPTE_V_VALID))
+ if (!(hptep[0] & cpu_to_be64(HPTE_V_VALID)))
continue;
/* need to make it temporarily absent so C is stable */
- hptep[0] |= HPTE_V_ABSENT;
+ hptep[0] |= cpu_to_be64(HPTE_V_ABSENT);
kvmppc_invalidate_hpte(kvm, hptep, i);
- v = hptep[0];
- r = hptep[1];
+ v = be64_to_cpu(hptep[0]);
+ r = be64_to_cpu(hptep[1]);
if (r & HPTE_R_C) {
- hptep[1] = r & ~HPTE_R_C;
+ hptep[1] = cpu_to_be64(r & ~HPTE_R_C);
if (!(rev[i].guest_rpte & HPTE_R_C)) {
rev[i].guest_rpte |= HPTE_R_C;
note_hpte_modification(kvm, &rev[i]);
@@ -1143,7 +1148,7 @@ static int kvm_test_clear_dirty_npages(struct kvm *kvm, unsigned long *rmapp)
}
v &= ~(HPTE_V_ABSENT | HPTE_V_HVLOCK);
v |= HPTE_V_VALID;
- hptep[0] = v;
+ hptep[0] = cpu_to_be64(v);
} while ((i = j) != head);
unlock_rmap(rmapp);
@@ -1307,7 +1312,7 @@ struct kvm_htab_ctx {
* Returns 1 if this HPT entry has been modified or has pending
* R/C bit changes.
*/
-static int hpte_dirty(struct revmap_entry *revp, unsigned long *hptp)
+static int hpte_dirty(struct revmap_entry *revp, __be64 *hptp)
{
unsigned long rcbits_unset;
@@ -1316,13 +1321,14 @@ static int hpte_dirty(struct revmap_entry *revp, unsigned long *hptp)
/* Also need to consider changes in reference and changed bits */
rcbits_unset = ~revp->guest_rpte & (HPTE_R_R | HPTE_R_C);
- if ((hptp[0] & HPTE_V_VALID) && (hptp[1] & rcbits_unset))
+ if ((be64_to_cpu(hptp[0]) & HPTE_V_VALID) &&
+ (be64_to_cpu(hptp[1]) & rcbits_unset))
return 1;
return 0;
}
-static long record_hpte(unsigned long flags, unsigned long *hptp,
+static long record_hpte(unsigned long flags, __be64 *hptp,
unsigned long *hpte, struct revmap_entry *revp,
int want_valid, int first_pass)
{
@@ -1337,10 +1343,10 @@ static long record_hpte(unsigned long flags, unsigned long *hptp,
return 0;
valid = 0;
- if (hptp[0] & (HPTE_V_VALID | HPTE_V_ABSENT)) {
+ if (be64_to_cpu(hptp[0]) & (HPTE_V_VALID | HPTE_V_ABSENT)) {
valid = 1;
if ((flags & KVM_GET_HTAB_BOLTED_ONLY) &&
- !(hptp[0] & HPTE_V_BOLTED))
+ !(be64_to_cpu(hptp[0]) & HPTE_V_BOLTED))
valid = 0;
}
if (valid != want_valid)
@@ -1352,7 +1358,7 @@ static long record_hpte(unsigned long flags, unsigned long *hptp,
preempt_disable();
while (!try_lock_hpte(hptp, HPTE_V_HVLOCK))
cpu_relax();
- v = hptp[0];
+ v = be64_to_cpu(hptp[0]);
/* re-evaluate valid and dirty from synchronized HPTE value */
valid = !!(v & HPTE_V_VALID);
@@ -1360,9 +1366,9 @@ static long record_hpte(unsigned long flags, unsigned long *hptp,
/* Harvest R and C into guest view if necessary */
rcbits_unset = ~revp->guest_rpte & (HPTE_R_R | HPTE_R_C);
- if (valid && (rcbits_unset & hptp[1])) {
- revp->guest_rpte |= (hptp[1] & (HPTE_R_R | HPTE_R_C)) |
- HPTE_GR_MODIFIED;
+ if (valid && (rcbits_unset & be64_to_cpu(hptp[1]))) {
+ revp->guest_rpte |= (be64_to_cpu(hptp[1]) &
+ (HPTE_R_R | HPTE_R_C)) | HPTE_GR_MODIFIED;
dirty = 1;
}
@@ -1381,13 +1387,13 @@ static long record_hpte(unsigned long flags, unsigned long *hptp,
revp->guest_rpte = r;
}
asm volatile(PPC_RELEASE_BARRIER "" : : : "memory");
- hptp[0] &= ~HPTE_V_HVLOCK;
+ hptp[0] &= ~cpu_to_be64(HPTE_V_HVLOCK);
preempt_enable();
if (!(valid = want_valid && (first_pass || dirty)))
ok = 0;
}
- hpte[0] = v;
- hpte[1] = r;
+ hpte[0] = cpu_to_be64(v);
+ hpte[1] = cpu_to_be64(r);
return ok;
}
@@ -1397,7 +1403,7 @@ static ssize_t kvm_htab_read(struct file *file, char __user *buf,
struct kvm_htab_ctx *ctx = file->private_data;
struct kvm *kvm = ctx->kvm;
struct kvm_get_htab_header hdr;
- unsigned long *hptp;
+ __be64 *hptp;
struct revmap_entry *revp;
unsigned long i, nb, nw;
unsigned long __user *lbuf;
@@ -1413,7 +1419,7 @@ static ssize_t kvm_htab_read(struct file *file, char __user *buf,
flags = ctx->flags;
i = ctx->index;
- hptp = (unsigned long *)(kvm->arch.hpt_virt + (i * HPTE_SIZE));
+ hptp = (__be64 *)(kvm->arch.hpt_virt + (i * HPTE_SIZE));
revp = kvm->arch.revmap + i;
lbuf = (unsigned long __user *)buf;
@@ -1497,7 +1503,7 @@ static ssize_t kvm_htab_write(struct file *file, const char __user *buf,
unsigned long i, j;
unsigned long v, r;
unsigned long __user *lbuf;
- unsigned long *hptp;
+ __be64 *hptp;
unsigned long tmp[2];
ssize_t nb;
long int err, ret;
@@ -1539,7 +1545,7 @@ static ssize_t kvm_htab_write(struct file *file, const char __user *buf,
i + hdr.n_valid + hdr.n_invalid > kvm->arch.hpt_npte)
break;
- hptp = (unsigned long *)(kvm->arch.hpt_virt + (i * HPTE_SIZE));
+ hptp = (__be64 *)(kvm->arch.hpt_virt + (i * HPTE_SIZE));
lbuf = (unsigned long __user *)buf;
for (j = 0; j < hdr.n_valid; ++j) {
err = -EFAULT;
@@ -1551,7 +1557,7 @@ static ssize_t kvm_htab_write(struct file *file, const char __user *buf,
lbuf += 2;
nb += HPTE_SIZE;
- if (hptp[0] & (HPTE_V_VALID | HPTE_V_ABSENT))
+ if (be64_to_cpu(hptp[0]) & (HPTE_V_VALID | HPTE_V_ABSENT))
kvmppc_do_h_remove(kvm, 0, i, 0, tmp);
err = -EIO;
ret = kvmppc_virtmode_do_h_enter(kvm, H_EXACT, i, v, r,
@@ -1577,7 +1583,7 @@ static ssize_t kvm_htab_write(struct file *file, const char __user *buf,
}
for (j = 0; j < hdr.n_invalid; ++j) {
- if (hptp[0] & (HPTE_V_VALID | HPTE_V_ABSENT))
+ if (be64_to_cpu(hptp[0]) & (HPTE_V_VALID | HPTE_V_ABSENT))
kvmppc_do_h_remove(kvm, 0, i, 0, tmp);
++i;
hptp += 2;
diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
index 6e62243..e5c6063 100644
--- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c
+++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c
@@ -154,10 +154,10 @@ static pte_t lookup_linux_pte_and_update(pgd_t *pgdir, unsigned long hva,
return kvmppc_read_update_linux_pte(ptep, writing, hugepage_shift);
}
-static inline void unlock_hpte(unsigned long *hpte, unsigned long hpte_v)
+static inline void unlock_hpte(__be64 *hpte, unsigned long hpte_v)
{
asm volatile(PPC_RELEASE_BARRIER "" : : : "memory");
- hpte[0] = hpte_v;
+ hpte[0] = cpu_to_be64(hpte_v);
}
long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
@@ -166,7 +166,7 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
{
unsigned long i, pa, gpa, gfn, psize;
unsigned long slot_fn, hva;
- unsigned long *hpte;
+ __be64 *hpte;
struct revmap_entry *rev;
unsigned long g_ptel;
struct kvm_memory_slot *memslot;
@@ -275,9 +275,9 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
return H_PARAMETER;
if (likely((flags & H_EXACT) = 0)) {
pte_index &= ~7UL;
- hpte = (unsigned long *)(kvm->arch.hpt_virt + (pte_index << 4));
+ hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4));
for (i = 0; i < 8; ++i) {
- if ((*hpte & HPTE_V_VALID) = 0 &&
+ if ((be64_to_cpu(*hpte) & HPTE_V_VALID) = 0 &&
try_lock_hpte(hpte, HPTE_V_HVLOCK | HPTE_V_VALID |
HPTE_V_ABSENT))
break;
@@ -292,11 +292,13 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
*/
hpte -= 16;
for (i = 0; i < 8; ++i) {
+ u64 pte;
while (!try_lock_hpte(hpte, HPTE_V_HVLOCK))
cpu_relax();
- if (!(*hpte & (HPTE_V_VALID | HPTE_V_ABSENT)))
+ pte = be64_to_cpu(*hpte);
+ if (!(pte & (HPTE_V_VALID | HPTE_V_ABSENT)))
break;
- *hpte &= ~HPTE_V_HVLOCK;
+ *hpte &= ~cpu_to_be64(HPTE_V_HVLOCK);
hpte += 2;
}
if (i = 8)
@@ -304,14 +306,17 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
}
pte_index += i;
} else {
- hpte = (unsigned long *)(kvm->arch.hpt_virt + (pte_index << 4));
+ hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4));
if (!try_lock_hpte(hpte, HPTE_V_HVLOCK | HPTE_V_VALID |
HPTE_V_ABSENT)) {
/* Lock the slot and check again */
+ u64 pte;
+
while (!try_lock_hpte(hpte, HPTE_V_HVLOCK))
cpu_relax();
- if (*hpte & (HPTE_V_VALID | HPTE_V_ABSENT)) {
- *hpte &= ~HPTE_V_HVLOCK;
+ pte = be64_to_cpu(*hpte);
+ if (pte & (HPTE_V_VALID | HPTE_V_ABSENT)) {
+ *hpte &= ~cpu_to_be64(HPTE_V_HVLOCK);
return H_PTEG_FULL;
}
}
@@ -347,11 +352,11 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
}
}
- hpte[1] = ptel;
+ hpte[1] = cpu_to_be64(ptel);
/* Write the first HPTE dword, unlocking the HPTE and making it valid */
eieio();
- hpte[0] = pteh;
+ hpte[0] = cpu_to_be64(pteh);
asm volatile("ptesync" : : : "memory");
*pte_idx_ret = pte_index;
@@ -468,30 +473,35 @@ long kvmppc_do_h_remove(struct kvm *kvm, unsigned long flags,
unsigned long pte_index, unsigned long avpn,
unsigned long *hpret)
{
- unsigned long *hpte;
+ __be64 *hpte;
unsigned long v, r, rb;
struct revmap_entry *rev;
+ u64 pte;
if (pte_index >= kvm->arch.hpt_npte)
return H_PARAMETER;
- hpte = (unsigned long *)(kvm->arch.hpt_virt + (pte_index << 4));
+ hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4));
while (!try_lock_hpte(hpte, HPTE_V_HVLOCK))
cpu_relax();
- if ((hpte[0] & (HPTE_V_ABSENT | HPTE_V_VALID)) = 0 ||
- ((flags & H_AVPN) && (hpte[0] & ~0x7fUL) != avpn) ||
- ((flags & H_ANDCOND) && (hpte[0] & avpn) != 0)) {
- hpte[0] &= ~HPTE_V_HVLOCK;
+ pte = be64_to_cpu(hpte[0]);
+ if ((pte & (HPTE_V_ABSENT | HPTE_V_VALID)) = 0 ||
+ ((flags & H_AVPN) && (pte & ~0x7fUL) != avpn) ||
+ ((flags & H_ANDCOND) && (pte & avpn) != 0)) {
+ hpte[0] &= ~cpu_to_be64(HPTE_V_HVLOCK);
return H_NOT_FOUND;
}
rev = real_vmalloc_addr(&kvm->arch.revmap[pte_index]);
- v = hpte[0] & ~HPTE_V_HVLOCK;
+ v = pte & ~HPTE_V_HVLOCK;
if (v & HPTE_V_VALID) {
- hpte[0] &= ~HPTE_V_VALID;
- rb = compute_tlbie_rb(v, hpte[1], pte_index);
+ u64 pte1;
+
+ pte1 = be64_to_cpu(hpte[1]);
+ hpte[0] &= ~cpu_to_be64(HPTE_V_VALID);
+ rb = compute_tlbie_rb(v, pte1, pte_index);
do_tlbies(kvm, &rb, 1, global_invalidates(kvm, flags), true);
/* Read PTE low word after tlbie to get final R/C values */
- remove_revmap_chain(kvm, pte_index, rev, v, hpte[1]);
+ remove_revmap_chain(kvm, pte_index, rev, v, pte1);
}
r = rev->guest_rpte & ~HPTE_GR_RESERVED;
note_hpte_modification(kvm, rev);
@@ -514,12 +524,14 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
{
struct kvm *kvm = vcpu->kvm;
unsigned long *args = &vcpu->arch.gpr[4];
- unsigned long *hp, *hptes[4], tlbrb[4];
+ __be64 *hp, *hptes[4];
+ unsigned long tlbrb[4];
long int i, j, k, n, found, indexes[4];
unsigned long flags, req, pte_index, rcbits;
int global;
long int ret = H_SUCCESS;
struct revmap_entry *rev, *revs[4];
+ u64 hp0;
global = global_invalidates(kvm, 0);
for (i = 0; i < 4 && ret = H_SUCCESS; ) {
@@ -542,8 +554,7 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
ret = H_PARAMETER;
break;
}
- hp = (unsigned long *)
- (kvm->arch.hpt_virt + (pte_index << 4));
+ hp = (__be64 *) (kvm->arch.hpt_virt + (pte_index << 4));
/* to avoid deadlock, don't spin except for first */
if (!try_lock_hpte(hp, HPTE_V_HVLOCK)) {
if (n)
@@ -552,23 +563,24 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
cpu_relax();
}
found = 0;
- if (hp[0] & (HPTE_V_ABSENT | HPTE_V_VALID)) {
+ hp0 = be64_to_cpu(hp[0]);
+ if (hp0 & (HPTE_V_ABSENT | HPTE_V_VALID)) {
switch (flags & 3) {
case 0: /* absolute */
found = 1;
break;
case 1: /* andcond */
- if (!(hp[0] & args[j + 1]))
+ if (!(hp0 & args[j + 1]))
found = 1;
break;
case 2: /* AVPN */
- if ((hp[0] & ~0x7fUL) = args[j + 1])
+ if ((hp0 & ~0x7fUL) = args[j + 1])
found = 1;
break;
}
}
if (!found) {
- hp[0] &= ~HPTE_V_HVLOCK;
+ hp[0] &= ~cpu_to_be64(HPTE_V_HVLOCK);
args[j] = ((0x90 | flags) << 56) + pte_index;
continue;
}
@@ -577,7 +589,7 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
rev = real_vmalloc_addr(&kvm->arch.revmap[pte_index]);
note_hpte_modification(kvm, rev);
- if (!(hp[0] & HPTE_V_VALID)) {
+ if (!(hp0 & HPTE_V_VALID)) {
/* insert R and C bits from PTE */
rcbits = rev->guest_rpte & (HPTE_R_R|HPTE_R_C);
args[j] |= rcbits << (56 - 5);
@@ -585,8 +597,10 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
continue;
}
- hp[0] &= ~HPTE_V_VALID; /* leave it locked */
- tlbrb[n] = compute_tlbie_rb(hp[0], hp[1], pte_index);
+ /* leave it locked */
+ hp[0] &= ~cpu_to_be64(HPTE_V_VALID);
+ tlbrb[n] = compute_tlbie_rb(be64_to_cpu(hp[0]),
+ be64_to_cpu(hp[1]), pte_index);
indexes[n] = j;
hptes[n] = hp;
revs[n] = rev;
@@ -605,7 +619,8 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu)
pte_index = args[j] & ((1ul << 56) - 1);
hp = hptes[k];
rev = revs[k];
- remove_revmap_chain(kvm, pte_index, rev, hp[0], hp[1]);
+ remove_revmap_chain(kvm, pte_index, rev,
+ be64_to_cpu(hp[0]), be64_to_cpu(hp[1]));
rcbits = rev->guest_rpte & (HPTE_R_R|HPTE_R_C);
args[j] |= rcbits << (56 - 5);
hp[0] = 0;
@@ -620,23 +635,25 @@ long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long va)
{
struct kvm *kvm = vcpu->kvm;
- unsigned long *hpte;
+ __be64 *hpte;
struct revmap_entry *rev;
unsigned long v, r, rb, mask, bits;
+ u64 pte;
if (pte_index >= kvm->arch.hpt_npte)
return H_PARAMETER;
- hpte = (unsigned long *)(kvm->arch.hpt_virt + (pte_index << 4));
+ hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4));
while (!try_lock_hpte(hpte, HPTE_V_HVLOCK))
cpu_relax();
- if ((hpte[0] & (HPTE_V_ABSENT | HPTE_V_VALID)) = 0 ||
- ((flags & H_AVPN) && (hpte[0] & ~0x7fUL) != avpn)) {
- hpte[0] &= ~HPTE_V_HVLOCK;
+ pte = be64_to_cpu(hpte[0]);
+ if ((pte & (HPTE_V_ABSENT | HPTE_V_VALID)) = 0 ||
+ ((flags & H_AVPN) && (pte & ~0x7fUL) != avpn)) {
+ hpte[0] &= ~cpu_to_be64(HPTE_V_HVLOCK);
return H_NOT_FOUND;
}
- v = hpte[0];
+ v = pte;
bits = (flags << 55) & HPTE_R_PP0;
bits |= (flags << 48) & HPTE_R_KEY_HI;
bits |= flags & (HPTE_R_PP | HPTE_R_N | HPTE_R_KEY_LO);
@@ -650,12 +667,12 @@ long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags,
rev->guest_rpte = r;
note_hpte_modification(kvm, rev);
}
- r = (hpte[1] & ~mask) | bits;
+ r = (be64_to_cpu(hpte[1]) & ~mask) | bits;
/* Update HPTE */
if (v & HPTE_V_VALID) {
rb = compute_tlbie_rb(v, r, pte_index);
- hpte[0] = v & ~HPTE_V_VALID;
+ hpte[0] = cpu_to_be64(v & ~HPTE_V_VALID);
do_tlbies(kvm, &rb, 1, global_invalidates(kvm, flags), true);
/*
* If the host has this page as readonly but the guest
@@ -681,9 +698,9 @@ long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags,
}
}
}
- hpte[1] = r;
+ hpte[1] = cpu_to_be64(r);
eieio();
- hpte[0] = v & ~HPTE_V_HVLOCK;
+ hpte[0] = cpu_to_be64(v & ~HPTE_V_HVLOCK);
asm volatile("ptesync" : : : "memory");
return H_SUCCESS;
}
@@ -692,7 +709,8 @@ long kvmppc_h_read(struct kvm_vcpu *vcpu, unsigned long flags,
unsigned long pte_index)
{
struct kvm *kvm = vcpu->kvm;
- unsigned long *hpte, v, r;
+ __be64 *hpte;
+ unsigned long v, r;
int i, n = 1;
struct revmap_entry *rev = NULL;
@@ -704,9 +722,9 @@ long kvmppc_h_read(struct kvm_vcpu *vcpu, unsigned long flags,
}
rev = real_vmalloc_addr(&kvm->arch.revmap[pte_index]);
for (i = 0; i < n; ++i, ++pte_index) {
- hpte = (unsigned long *)(kvm->arch.hpt_virt + (pte_index << 4));
- v = hpte[0] & ~HPTE_V_HVLOCK;
- r = hpte[1];
+ hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4));
+ v = be64_to_cpu(hpte[0]) & ~HPTE_V_HVLOCK;
+ r = be64_to_cpu(hpte[1]);
if (v & HPTE_V_ABSENT) {
v &= ~HPTE_V_ABSENT;
v |= HPTE_V_VALID;
@@ -721,25 +739,27 @@ long kvmppc_h_read(struct kvm_vcpu *vcpu, unsigned long flags,
return H_SUCCESS;
}
-void kvmppc_invalidate_hpte(struct kvm *kvm, unsigned long *hptep,
+void kvmppc_invalidate_hpte(struct kvm *kvm, __be64 *hptep,
unsigned long pte_index)
{
unsigned long rb;
- hptep[0] &= ~HPTE_V_VALID;
- rb = compute_tlbie_rb(hptep[0], hptep[1], pte_index);
+ hptep[0] &= ~cpu_to_be64(HPTE_V_VALID);
+ rb = compute_tlbie_rb(be64_to_cpu(hptep[0]), be64_to_cpu(hptep[1]),
+ pte_index);
do_tlbies(kvm, &rb, 1, 1, true);
}
EXPORT_SYMBOL_GPL(kvmppc_invalidate_hpte);
-void kvmppc_clear_ref_hpte(struct kvm *kvm, unsigned long *hptep,
+void kvmppc_clear_ref_hpte(struct kvm *kvm, __be64 *hptep,
unsigned long pte_index)
{
unsigned long rb;
unsigned char rbyte;
- rb = compute_tlbie_rb(hptep[0], hptep[1], pte_index);
- rbyte = (hptep[1] & ~HPTE_R_R) >> 8;
+ rb = compute_tlbie_rb(be64_to_cpu(hptep[0]), be64_to_cpu(hptep[1]),
+ pte_index);
+ rbyte = (be64_to_cpu(hptep[1]) & ~HPTE_R_R) >> 8;
/* modify only the second-last byte, which contains the ref bit */
*((char *)hptep + 14) = rbyte;
do_tlbies(kvm, &rb, 1, 1, false);
@@ -765,7 +785,7 @@ long kvmppc_hv_find_lock_hpte(struct kvm *kvm, gva_t eaddr, unsigned long slb_v,
unsigned long somask;
unsigned long vsid, hash;
unsigned long avpn;
- unsigned long *hpte;
+ __be64 *hpte;
unsigned long mask, val;
unsigned long v, r;
@@ -797,11 +817,11 @@ long kvmppc_hv_find_lock_hpte(struct kvm *kvm, gva_t eaddr, unsigned long slb_v,
val |= avpn;
for (;;) {
- hpte = (unsigned long *)(kvm->arch.hpt_virt + (hash << 7));
+ hpte = (__be64 *)(kvm->arch.hpt_virt + (hash << 7));
for (i = 0; i < 16; i += 2) {
/* Read the PTE racily */
- v = hpte[i] & ~HPTE_V_HVLOCK;
+ v = be64_to_cpu(hpte[i]) & ~HPTE_V_HVLOCK;
/* Check valid/absent, hash, segment size and AVPN */
if (!(v & valid) || (v & mask) != val)
@@ -810,8 +830,8 @@ long kvmppc_hv_find_lock_hpte(struct kvm *kvm, gva_t eaddr, unsigned long slb_v,
/* Lock the PTE and read it under the lock */
while (!try_lock_hpte(&hpte[i], HPTE_V_HVLOCK))
cpu_relax();
- v = hpte[i] & ~HPTE_V_HVLOCK;
- r = hpte[i+1];
+ v = be64_to_cpu(hpte[i]) & ~HPTE_V_HVLOCK;
+ r = be64_to_cpu(hpte[i+1]);
/*
* Check the HPTE again, including large page size
@@ -825,7 +845,7 @@ long kvmppc_hv_find_lock_hpte(struct kvm *kvm, gva_t eaddr, unsigned long slb_v,
return (hash << 3) + (i >> 1);
/* Unlock and move on */
- hpte[i] = v;
+ hpte[i] = cpu_to_be64(v);
}
if (val & HPTE_V_SECONDARY)
@@ -854,7 +874,7 @@ long kvmppc_hpte_hv_fault(struct kvm_vcpu *vcpu, unsigned long addr,
struct kvm *kvm = vcpu->kvm;
long int index;
unsigned long v, r, gr;
- unsigned long *hpte;
+ __be64 *hpte;
unsigned long valid;
struct revmap_entry *rev;
unsigned long pp, key;
@@ -870,9 +890,9 @@ long kvmppc_hpte_hv_fault(struct kvm_vcpu *vcpu, unsigned long addr,
return status; /* there really was no HPTE */
return 0; /* for prot fault, HPTE disappeared */
}
- hpte = (unsigned long *)(kvm->arch.hpt_virt + (index << 4));
- v = hpte[0] & ~HPTE_V_HVLOCK;
- r = hpte[1];
+ hpte = (__be64 *)(kvm->arch.hpt_virt + (index << 4));
+ v = be64_to_cpu(hpte[0]) & ~HPTE_V_HVLOCK;
+ r = be64_to_cpu(hpte[1]);
rev = real_vmalloc_addr(&kvm->arch.revmap[index]);
gr = rev->guest_rpte;
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* [PATCH 3/7] KVM: PPC: Book3S HV: Access guest VPA in BE
2014-06-17 16:17 ` Alexander Graf
@ 2014-06-17 16:17 ` Alexander Graf
-1 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
There are a few shared data structures between the host and the guest. Most
of them get registered through the VPA interface.
These data structures are defined to always be in big endian byte order, so
let's make sure we always access them in big endian.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
v1 -> v2:
- Add __be hints
- Fix dtl_idx
---
arch/powerpc/kvm/book3s_hv.c | 22 +++++++++++-----------
arch/powerpc/kvm/book3s_hv_ras.c | 6 +++---
2 files changed, 14 insertions(+), 14 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 1562acf..f7bcc72 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -270,7 +270,7 @@ struct kvm_vcpu *kvmppc_find_vcpu(struct kvm *kvm, int id)
static void init_vpa(struct kvm_vcpu *vcpu, struct lppaca *vpa)
{
vpa->__old_status |= LPPACA_OLD_SHARED_PROC;
- vpa->yield_count = 1;
+ vpa->yield_count = cpu_to_be32(1);
}
static int set_vpa(struct kvm_vcpu *vcpu, struct kvmppc_vpa *v,
@@ -293,8 +293,8 @@ static int set_vpa(struct kvm_vcpu *vcpu, struct kvmppc_vpa *v,
struct reg_vpa {
u32 dummy;
union {
- u16 hword;
- u32 word;
+ __be16 hword;
+ __be32 word;
} length;
};
@@ -333,9 +333,9 @@ static unsigned long do_h_register_vpa(struct kvm_vcpu *vcpu,
if (va == NULL)
return H_PARAMETER;
if (subfunc == H_VPA_REG_VPA)
- len = ((struct reg_vpa *)va)->length.hword;
+ len = be16_to_cpu(((struct reg_vpa *)va)->length.hword);
else
- len = ((struct reg_vpa *)va)->length.word;
+ len = be32_to_cpu(((struct reg_vpa *)va)->length.word);
kvmppc_unpin_guest_page(kvm, va, vpa, false);
/* Check length */
@@ -540,18 +540,18 @@ static void kvmppc_create_dtl_entry(struct kvm_vcpu *vcpu,
return;
memset(dt, 0, sizeof(struct dtl_entry));
dt->dispatch_reason = 7;
- dt->processor_id = vc->pcpu + vcpu->arch.ptid;
- dt->timebase = now + vc->tb_offset;
- dt->enqueue_to_dispatch_time = stolen;
- dt->srr0 = kvmppc_get_pc(vcpu);
- dt->srr1 = vcpu->arch.shregs.msr;
+ dt->processor_id = cpu_to_be16(vc->pcpu + vcpu->arch.ptid);
+ dt->timebase = cpu_to_be64(now + vc->tb_offset);
+ dt->enqueue_to_dispatch_time = cpu_to_be32(stolen);
+ dt->srr0 = cpu_to_be64(kvmppc_get_pc(vcpu));
+ dt->srr1 = cpu_to_be64(vcpu->arch.shregs.msr);
++dt;
if (dt == vcpu->arch.dtl.pinned_end)
dt = vcpu->arch.dtl.pinned_addr;
vcpu->arch.dtl_ptr = dt;
/* order writing *dt vs. writing vpa->dtl_idx */
smp_wmb();
- vpa->dtl_idx = ++vcpu->arch.dtl_index;
+ vpa->dtl_idx = cpu_to_be64(++vcpu->arch.dtl_index);
vcpu->arch.dtl.dirty = true;
}
diff --git a/arch/powerpc/kvm/book3s_hv_ras.c b/arch/powerpc/kvm/book3s_hv_ras.c
index 768a9f9..6a3ba8a 100644
--- a/arch/powerpc/kvm/book3s_hv_ras.c
+++ b/arch/powerpc/kvm/book3s_hv_ras.c
@@ -45,14 +45,14 @@ static void reload_slb(struct kvm_vcpu *vcpu)
return;
/* Sanity check */
- n = min_t(u32, slb->persistent, SLB_MIN_SIZE);
+ n = min_t(u32, be32_to_cpu(slb->persistent), SLB_MIN_SIZE);
if ((void *) &slb->save_area[n] > vcpu->arch.slb_shadow.pinned_end)
return;
/* Load up the SLB from that */
for (i = 0; i < n; ++i) {
- unsigned long rb = slb->save_area[i].esid;
- unsigned long rs = slb->save_area[i].vsid;
+ unsigned long rb = be64_to_cpu(slb->save_area[i].esid);
+ unsigned long rs = be64_to_cpu(slb->save_area[i].vsid);
rb = (rb & ~0xFFFul) | i; /* insert entry number */
asm volatile("slbmte %0,%1" : : "r" (rs), "r" (rb));
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* [PATCH 3/7] KVM: PPC: Book3S HV: Access guest VPA in BE
@ 2014-06-17 16:17 ` Alexander Graf
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
There are a few shared data structures between the host and the guest. Most
of them get registered through the VPA interface.
These data structures are defined to always be in big endian byte order, so
let's make sure we always access them in big endian.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
v1 -> v2:
- Add __be hints
- Fix dtl_idx
---
arch/powerpc/kvm/book3s_hv.c | 22 +++++++++++-----------
arch/powerpc/kvm/book3s_hv_ras.c | 6 +++---
2 files changed, 14 insertions(+), 14 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 1562acf..f7bcc72 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -270,7 +270,7 @@ struct kvm_vcpu *kvmppc_find_vcpu(struct kvm *kvm, int id)
static void init_vpa(struct kvm_vcpu *vcpu, struct lppaca *vpa)
{
vpa->__old_status |= LPPACA_OLD_SHARED_PROC;
- vpa->yield_count = 1;
+ vpa->yield_count = cpu_to_be32(1);
}
static int set_vpa(struct kvm_vcpu *vcpu, struct kvmppc_vpa *v,
@@ -293,8 +293,8 @@ static int set_vpa(struct kvm_vcpu *vcpu, struct kvmppc_vpa *v,
struct reg_vpa {
u32 dummy;
union {
- u16 hword;
- u32 word;
+ __be16 hword;
+ __be32 word;
} length;
};
@@ -333,9 +333,9 @@ static unsigned long do_h_register_vpa(struct kvm_vcpu *vcpu,
if (va = NULL)
return H_PARAMETER;
if (subfunc = H_VPA_REG_VPA)
- len = ((struct reg_vpa *)va)->length.hword;
+ len = be16_to_cpu(((struct reg_vpa *)va)->length.hword);
else
- len = ((struct reg_vpa *)va)->length.word;
+ len = be32_to_cpu(((struct reg_vpa *)va)->length.word);
kvmppc_unpin_guest_page(kvm, va, vpa, false);
/* Check length */
@@ -540,18 +540,18 @@ static void kvmppc_create_dtl_entry(struct kvm_vcpu *vcpu,
return;
memset(dt, 0, sizeof(struct dtl_entry));
dt->dispatch_reason = 7;
- dt->processor_id = vc->pcpu + vcpu->arch.ptid;
- dt->timebase = now + vc->tb_offset;
- dt->enqueue_to_dispatch_time = stolen;
- dt->srr0 = kvmppc_get_pc(vcpu);
- dt->srr1 = vcpu->arch.shregs.msr;
+ dt->processor_id = cpu_to_be16(vc->pcpu + vcpu->arch.ptid);
+ dt->timebase = cpu_to_be64(now + vc->tb_offset);
+ dt->enqueue_to_dispatch_time = cpu_to_be32(stolen);
+ dt->srr0 = cpu_to_be64(kvmppc_get_pc(vcpu));
+ dt->srr1 = cpu_to_be64(vcpu->arch.shregs.msr);
++dt;
if (dt = vcpu->arch.dtl.pinned_end)
dt = vcpu->arch.dtl.pinned_addr;
vcpu->arch.dtl_ptr = dt;
/* order writing *dt vs. writing vpa->dtl_idx */
smp_wmb();
- vpa->dtl_idx = ++vcpu->arch.dtl_index;
+ vpa->dtl_idx = cpu_to_be64(++vcpu->arch.dtl_index);
vcpu->arch.dtl.dirty = true;
}
diff --git a/arch/powerpc/kvm/book3s_hv_ras.c b/arch/powerpc/kvm/book3s_hv_ras.c
index 768a9f9..6a3ba8a 100644
--- a/arch/powerpc/kvm/book3s_hv_ras.c
+++ b/arch/powerpc/kvm/book3s_hv_ras.c
@@ -45,14 +45,14 @@ static void reload_slb(struct kvm_vcpu *vcpu)
return;
/* Sanity check */
- n = min_t(u32, slb->persistent, SLB_MIN_SIZE);
+ n = min_t(u32, be32_to_cpu(slb->persistent), SLB_MIN_SIZE);
if ((void *) &slb->save_area[n] > vcpu->arch.slb_shadow.pinned_end)
return;
/* Load up the SLB from that */
for (i = 0; i < n; ++i) {
- unsigned long rb = slb->save_area[i].esid;
- unsigned long rs = slb->save_area[i].vsid;
+ unsigned long rb = be64_to_cpu(slb->save_area[i].esid);
+ unsigned long rs = be64_to_cpu(slb->save_area[i].vsid);
rb = (rb & ~0xFFFul) | i; /* insert entry number */
asm volatile("slbmte %0,%1" : : "r" (rs), "r" (rb));
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* [PATCH 4/7] KVM: PPC: Book3S HV: Access host lppaca and shadow slb in BE
2014-06-17 16:17 ` Alexander Graf
@ 2014-06-17 16:17 ` Alexander Graf
-1 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
Some data structures are always stored in big endian. Among those are the LPPACA
fields as well as the shadow slb. These structures might be shared with a
hypervisor.
So whenever we access those fields, make sure we do so in big endian byte order.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 20 ++++++++++----------
1 file changed, 10 insertions(+), 10 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 7e7fa01..75c7e22 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -32,10 +32,6 @@
#define VCPU_GPRS_TM(reg) (((reg) * ULONG_SIZE) + VCPU_GPR_TM)
-#ifdef __LITTLE_ENDIAN__
-#error Need to fix lppaca and SLB shadow accesses in little endian mode
-#endif
-
/* Values in HSTATE_NAPPING(r13) */
#define NAPPING_CEDE 1
#define NAPPING_NOVCPU 2
@@ -595,9 +591,10 @@ kvmppc_got_guest:
ld r3, VCPU_VPA(r4)
cmpdi r3, 0
beq 25f
- lwz r5, LPPACA_YIELDCOUNT(r3)
+ li r6, LPPACA_YIELDCOUNT
+ LWZX_BE r5, r3, r6
addi r5, r5, 1
- stw r5, LPPACA_YIELDCOUNT(r3)
+ STWX_BE r5, r3, r6
li r6, 1
stb r6, VCPU_VPA_DIRTY(r4)
25:
@@ -1442,9 +1439,10 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
ld r8, VCPU_VPA(r9) /* do they have a VPA? */
cmpdi r8, 0
beq 25f
- lwz r3, LPPACA_YIELDCOUNT(r8)
+ li r4, LPPACA_YIELDCOUNT
+ LWZX_BE r3, r8, r4
addi r3, r3, 1
- stw r3, LPPACA_YIELDCOUNT(r8)
+ STWX_BE r3, r8, r4
li r3, 1
stb r3, VCPU_VPA_DIRTY(r9)
25:
@@ -1757,8 +1755,10 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
33: ld r8,PACA_SLBSHADOWPTR(r13)
.rept SLB_NUM_BOLTED
- ld r5,SLBSHADOW_SAVEAREA(r8)
- ld r6,SLBSHADOW_SAVEAREA+8(r8)
+ 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 1f
slbmte r6,r5
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* [PATCH 4/7] KVM: PPC: Book3S HV: Access host lppaca and shadow slb in BE
@ 2014-06-17 16:17 ` Alexander Graf
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
Some data structures are always stored in big endian. Among those are the LPPACA
fields as well as the shadow slb. These structures might be shared with a
hypervisor.
So whenever we access those fields, make sure we do so in big endian byte order.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 20 ++++++++++----------
1 file changed, 10 insertions(+), 10 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 7e7fa01..75c7e22 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -32,10 +32,6 @@
#define VCPU_GPRS_TM(reg) (((reg) * ULONG_SIZE) + VCPU_GPR_TM)
-#ifdef __LITTLE_ENDIAN__
-#error Need to fix lppaca and SLB shadow accesses in little endian mode
-#endif
-
/* Values in HSTATE_NAPPING(r13) */
#define NAPPING_CEDE 1
#define NAPPING_NOVCPU 2
@@ -595,9 +591,10 @@ kvmppc_got_guest:
ld r3, VCPU_VPA(r4)
cmpdi r3, 0
beq 25f
- lwz r5, LPPACA_YIELDCOUNT(r3)
+ li r6, LPPACA_YIELDCOUNT
+ LWZX_BE r5, r3, r6
addi r5, r5, 1
- stw r5, LPPACA_YIELDCOUNT(r3)
+ STWX_BE r5, r3, r6
li r6, 1
stb r6, VCPU_VPA_DIRTY(r4)
25:
@@ -1442,9 +1439,10 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
ld r8, VCPU_VPA(r9) /* do they have a VPA? */
cmpdi r8, 0
beq 25f
- lwz r3, LPPACA_YIELDCOUNT(r8)
+ li r4, LPPACA_YIELDCOUNT
+ LWZX_BE r3, r8, r4
addi r3, r3, 1
- stw r3, LPPACA_YIELDCOUNT(r8)
+ STWX_BE r3, r8, r4
li r3, 1
stb r3, VCPU_VPA_DIRTY(r9)
25:
@@ -1757,8 +1755,10 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
33: ld r8,PACA_SLBSHADOWPTR(r13)
.rept SLB_NUM_BOLTED
- ld r5,SLBSHADOW_SAVEAREA(r8)
- ld r6,SLBSHADOW_SAVEAREA+8(r8)
+ 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 1f
slbmte r6,r5
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* [PATCH 5/7] KVM: PPC: Book3S HV: Access XICS in BE
2014-06-17 16:17 ` Alexander Graf
@ 2014-06-17 16:17 ` Alexander Graf
-1 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
On the exit path from the guest we check what type of interrupt we received
if we received one. This means we're doing hardware access to the XICS interrupt
controller.
However, when running on a little endian system, this access is byte reversed.
So let's make sure to swizzle the bytes back again and virtually make XICS
accesses big endian.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
v1 -> v2:
- Make code easier to follow and use hardware for bswap
---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 18 ++++++++++++++----
1 file changed, 14 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 75c7e22..1a71f60 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -2325,7 +2325,18 @@ kvmppc_read_intr:
cmpdi r6, 0
beq- 1f
lwzcix r0, r6, r7
- rlwinm. r3, r0, 0, 0xffffff
+ /*
+ * Save XIRR for later. Since we get in in reverse endian on LE
+ * systems, save it byte reversed and fetch it back in host endian.
+ */
+ li r3, HSTATE_SAVED_XIRR
+ STWX_BE r0, r3, r13
+#ifdef __LITTLE_ENDIAN__
+ lwz r3, HSTATE_SAVED_XIRR(r13)
+#else
+ mr r3, r0
+#endif
+ rlwinm. r3, r3, 0, 0xffffff
sync
beq 1f /* if nothing pending in the ICP */
@@ -2357,10 +2368,9 @@ kvmppc_read_intr:
li r3, -1
1: blr
-42: /* It's not an IPI and it's for the host, stash it in the PACA
- * before exit, it will be picked up by the host ICP driver
+42: /* It's not an IPI and it's for the host. We saved a copy of XIRR in
+ * the PACA earlier, it will be picked up by the host ICP driver
*/
- stw r0, HSTATE_SAVED_XIRR(r13)
li r3, 1
b 1b
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* [PATCH 5/7] KVM: PPC: Book3S HV: Access XICS in BE
@ 2014-06-17 16:17 ` Alexander Graf
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
On the exit path from the guest we check what type of interrupt we received
if we received one. This means we're doing hardware access to the XICS interrupt
controller.
However, when running on a little endian system, this access is byte reversed.
So let's make sure to swizzle the bytes back again and virtually make XICS
accesses big endian.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
v1 -> v2:
- Make code easier to follow and use hardware for bswap
---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 18 ++++++++++++++----
1 file changed, 14 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 75c7e22..1a71f60 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -2325,7 +2325,18 @@ kvmppc_read_intr:
cmpdi r6, 0
beq- 1f
lwzcix r0, r6, r7
- rlwinm. r3, r0, 0, 0xffffff
+ /*
+ * Save XIRR for later. Since we get in in reverse endian on LE
+ * systems, save it byte reversed and fetch it back in host endian.
+ */
+ li r3, HSTATE_SAVED_XIRR
+ STWX_BE r0, r3, r13
+#ifdef __LITTLE_ENDIAN__
+ lwz r3, HSTATE_SAVED_XIRR(r13)
+#else
+ mr r3, r0
+#endif
+ rlwinm. r3, r3, 0, 0xffffff
sync
beq 1f /* if nothing pending in the ICP */
@@ -2357,10 +2368,9 @@ kvmppc_read_intr:
li r3, -1
1: blr
-42: /* It's not an IPI and it's for the host, stash it in the PACA
- * before exit, it will be picked up by the host ICP driver
+42: /* It's not an IPI and it's for the host. We saved a copy of XIRR in
+ * the PACA earlier, it will be picked up by the host ICP driver
*/
- stw r0, HSTATE_SAVED_XIRR(r13)
li r3, 1
b 1b
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* [PATCH 6/7] KVM: PPC: Book3S HV: Fix ABIv2 on LE
2014-06-17 16:17 ` Alexander Graf
@ 2014-06-17 16:17 ` Alexander Graf
-1 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
We use ABIv2 on Little Endian systems which gets rid of the dotted function
names. Branch to the actual functions when we see such a system.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 22 ++++++++++++++--------
1 file changed, 14 insertions(+), 8 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 1a71f60..1ff3ebd 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -36,6 +36,12 @@
#define NAPPING_CEDE 1
#define NAPPING_NOVCPU 2
+#if defined(_CALL_ELF) && _CALL_ELF == 2
+#define FUNC(name) name
+#else
+#define FUNC(name) GLUE(.,name)
+#endif
+
/*
* Call kvmppc_hv_entry in real mode.
* Must be called with interrupts hard-disabled.
@@ -668,9 +674,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
mr r31, r4
addi r3, r31, VCPU_FPRS_TM
- bl .load_fp_state
+ bl FUNC(load_fp_state)
addi r3, r31, VCPU_VRS_TM
- bl .load_vr_state
+ bl FUNC(load_vr_state)
mr r4, r31
lwz r7, VCPU_VRSAVE_TM(r4)
mtspr SPRN_VRSAVE, r7
@@ -1414,9 +1420,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
/* Save FP/VSX. */
addi r3, r9, VCPU_FPRS_TM
- bl .store_fp_state
+ bl FUNC(store_fp_state)
addi r3, r9, VCPU_VRS_TM
- bl .store_vr_state
+ bl FUNC(store_vr_state)
mfspr r6, SPRN_VRSAVE
stw r6, VCPU_VRSAVE_TM(r9)
1:
@@ -2405,11 +2411,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
mtmsrd r8
isync
addi r3,r3,VCPU_FPRS
- bl .store_fp_state
+ bl FUNC(store_fp_state)
#ifdef CONFIG_ALTIVEC
BEGIN_FTR_SECTION
addi r3,r31,VCPU_VRS
- bl .store_vr_state
+ bl FUNC(store_vr_state)
END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
#endif
mfspr r6,SPRN_VRSAVE
@@ -2441,11 +2447,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
mtmsrd r8
isync
addi r3,r4,VCPU_FPRS
- bl .load_fp_state
+ bl FUNC(load_fp_state)
#ifdef CONFIG_ALTIVEC
BEGIN_FTR_SECTION
addi r3,r31,VCPU_VRS
- bl .load_vr_state
+ bl FUNC(load_vr_state)
END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
#endif
lwz r7,VCPU_VRSAVE(r31)
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* [PATCH 6/7] KVM: PPC: Book3S HV: Fix ABIv2 on LE
@ 2014-06-17 16:17 ` Alexander Graf
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
We use ABIv2 on Little Endian systems which gets rid of the dotted function
names. Branch to the actual functions when we see such a system.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 22 ++++++++++++++--------
1 file changed, 14 insertions(+), 8 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 1a71f60..1ff3ebd 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -36,6 +36,12 @@
#define NAPPING_CEDE 1
#define NAPPING_NOVCPU 2
+#if defined(_CALL_ELF) && _CALL_ELF = 2
+#define FUNC(name) name
+#else
+#define FUNC(name) GLUE(.,name)
+#endif
+
/*
* Call kvmppc_hv_entry in real mode.
* Must be called with interrupts hard-disabled.
@@ -668,9 +674,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
mr r31, r4
addi r3, r31, VCPU_FPRS_TM
- bl .load_fp_state
+ bl FUNC(load_fp_state)
addi r3, r31, VCPU_VRS_TM
- bl .load_vr_state
+ bl FUNC(load_vr_state)
mr r4, r31
lwz r7, VCPU_VRSAVE_TM(r4)
mtspr SPRN_VRSAVE, r7
@@ -1414,9 +1420,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
/* Save FP/VSX. */
addi r3, r9, VCPU_FPRS_TM
- bl .store_fp_state
+ bl FUNC(store_fp_state)
addi r3, r9, VCPU_VRS_TM
- bl .store_vr_state
+ bl FUNC(store_vr_state)
mfspr r6, SPRN_VRSAVE
stw r6, VCPU_VRSAVE_TM(r9)
1:
@@ -2405,11 +2411,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
mtmsrd r8
isync
addi r3,r3,VCPU_FPRS
- bl .store_fp_state
+ bl FUNC(store_fp_state)
#ifdef CONFIG_ALTIVEC
BEGIN_FTR_SECTION
addi r3,r31,VCPU_VRS
- bl .store_vr_state
+ bl FUNC(store_vr_state)
END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
#endif
mfspr r6,SPRN_VRSAVE
@@ -2441,11 +2447,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
mtmsrd r8
isync
addi r3,r4,VCPU_FPRS
- bl .load_fp_state
+ bl FUNC(load_fp_state)
#ifdef CONFIG_ALTIVEC
BEGIN_FTR_SECTION
addi r3,r31,VCPU_VRS
- bl .load_vr_state
+ bl FUNC(load_vr_state)
END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
#endif
lwz r7,VCPU_VRSAVE(r31)
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* [PATCH 7/7] KVM: PPC: Book3S HV: Enable for little endian hosts
2014-06-17 16:17 ` Alexander Graf
@ 2014-06-17 16:17 ` Alexander Graf
-1 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
Now that we've fixed all the issues that HV KVM code had on little endian
hosts, we can enable it in the kernel configuration for users to play with.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
arch/powerpc/kvm/Kconfig | 1 -
1 file changed, 1 deletion(-)
diff --git a/arch/powerpc/kvm/Kconfig b/arch/powerpc/kvm/Kconfig
index d6a53b9..8aeeda1 100644
--- a/arch/powerpc/kvm/Kconfig
+++ b/arch/powerpc/kvm/Kconfig
@@ -75,7 +75,6 @@ config KVM_BOOK3S_64
config KVM_BOOK3S_64_HV
tristate "KVM support for POWER7 and PPC970 using hypervisor mode in host"
depends on KVM_BOOK3S_64
- depends on !CPU_LITTLE_ENDIAN
select KVM_BOOK3S_HV_POSSIBLE
select MMU_NOTIFIER
select CMA
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* [PATCH 7/7] KVM: PPC: Book3S HV: Enable for little endian hosts
@ 2014-06-17 16:17 ` Alexander Graf
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-17 16:17 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
Now that we've fixed all the issues that HV KVM code had on little endian
hosts, we can enable it in the kernel configuration for users to play with.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
arch/powerpc/kvm/Kconfig | 1 -
1 file changed, 1 deletion(-)
diff --git a/arch/powerpc/kvm/Kconfig b/arch/powerpc/kvm/Kconfig
index d6a53b9..8aeeda1 100644
--- a/arch/powerpc/kvm/Kconfig
+++ b/arch/powerpc/kvm/Kconfig
@@ -75,7 +75,6 @@ config KVM_BOOK3S_64
config KVM_BOOK3S_64_HV
tristate "KVM support for POWER7 and PPC970 using hypervisor mode in host"
depends on KVM_BOOK3S_64
- depends on !CPU_LITTLE_ENDIAN
select KVM_BOOK3S_HV_POSSIBLE
select MMU_NOTIFIER
select CMA
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* Re: [PATCH 6/7] KVM: PPC: Book3S HV: Fix ABIv2 on LE
2014-06-17 16:17 ` Alexander Graf
@ 2014-06-18 15:33 ` Aneesh Kumar K.V
-1 siblings, 0 replies; 24+ messages in thread
From: Aneesh Kumar K.V @ 2014-06-18 15:21 UTC (permalink / raw)
To: Alexander Graf, kvm-ppc; +Cc: kvm
Alexander Graf <agraf@suse.de> writes:
> We use ABIv2 on Little Endian systems which gets rid of the dotted function
> names. Branch to the actual functions when we see such a system.
>
> Signed-off-by: Alexander Graf <agraf@suse.de>
As per patches sent by anton we don't need this. We can branch to the
function rathen than the dot symbol
http://article.gmane.org/gmane.linux.ports.ppc.embedded/68925
http://article.gmane.org/gmane.linux.ports.ppc.embedded/71005
-aneesh
> ---
> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 22 ++++++++++++++--------
> 1 file changed, 14 insertions(+), 8 deletions(-)
>
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index 1a71f60..1ff3ebd 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -36,6 +36,12 @@
> #define NAPPING_CEDE 1
> #define NAPPING_NOVCPU 2
>
> +#if defined(_CALL_ELF) && _CALL_ELF == 2
> +#define FUNC(name) name
> +#else
> +#define FUNC(name) GLUE(.,name)
> +#endif
> +
> /*
> * Call kvmppc_hv_entry in real mode.
> * Must be called with interrupts hard-disabled.
> @@ -668,9 +674,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
>
> mr r31, r4
> addi r3, r31, VCPU_FPRS_TM
> - bl .load_fp_state
> + bl FUNC(load_fp_state)
> addi r3, r31, VCPU_VRS_TM
> - bl .load_vr_state
> + bl FUNC(load_vr_state)
> mr r4, r31
> lwz r7, VCPU_VRSAVE_TM(r4)
> mtspr SPRN_VRSAVE, r7
> @@ -1414,9 +1420,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
>
> /* Save FP/VSX. */
> addi r3, r9, VCPU_FPRS_TM
> - bl .store_fp_state
> + bl FUNC(store_fp_state)
> addi r3, r9, VCPU_VRS_TM
> - bl .store_vr_state
> + bl FUNC(store_vr_state)
> mfspr r6, SPRN_VRSAVE
> stw r6, VCPU_VRSAVE_TM(r9)
> 1:
> @@ -2405,11 +2411,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
> mtmsrd r8
> isync
> addi r3,r3,VCPU_FPRS
> - bl .store_fp_state
> + bl FUNC(store_fp_state)
> #ifdef CONFIG_ALTIVEC
> BEGIN_FTR_SECTION
> addi r3,r31,VCPU_VRS
> - bl .store_vr_state
> + bl FUNC(store_vr_state)
> END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
> #endif
> mfspr r6,SPRN_VRSAVE
> @@ -2441,11 +2447,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
> mtmsrd r8
> isync
> addi r3,r4,VCPU_FPRS
> - bl .load_fp_state
> + bl FUNC(load_fp_state)
> #ifdef CONFIG_ALTIVEC
> BEGIN_FTR_SECTION
> addi r3,r31,VCPU_VRS
> - bl .load_vr_state
> + bl FUNC(load_vr_state)
> END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
> #endif
> lwz r7,VCPU_VRSAVE(r31)
-aneesh
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH 6/7] KVM: PPC: Book3S HV: Fix ABIv2 on LE
2014-06-18 15:33 ` Aneesh Kumar K.V
@ 2014-06-18 15:23 ` Alexander Graf
-1 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-18 15:23 UTC (permalink / raw)
To: Aneesh Kumar K.V, kvm-ppc; +Cc: kvm
On 18.06.14 17:21, Aneesh Kumar K.V wrote:
> Alexander Graf <agraf@suse.de> writes:
>
>> We use ABIv2 on Little Endian systems which gets rid of the dotted function
>> names. Branch to the actual functions when we see such a system.
>>
>> Signed-off-by: Alexander Graf <agraf@suse.de>
> As per patches sent by anton we don't need this. We can branch to the
> function rathen than the dot symbol
>
> http://article.gmane.org/gmane.linux.ports.ppc.embedded/68925
> http://article.gmane.org/gmane.linux.ports.ppc.embedded/71005
Ah, true in this case, because the file never gets compiled as a module.
Alex
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH 6/7] KVM: PPC: Book3S HV: Fix ABIv2 on LE
@ 2014-06-18 15:23 ` Alexander Graf
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-18 15:23 UTC (permalink / raw)
To: Aneesh Kumar K.V, kvm-ppc; +Cc: kvm
On 18.06.14 17:21, Aneesh Kumar K.V wrote:
> Alexander Graf <agraf@suse.de> writes:
>
>> We use ABIv2 on Little Endian systems which gets rid of the dotted function
>> names. Branch to the actual functions when we see such a system.
>>
>> Signed-off-by: Alexander Graf <agraf@suse.de>
> As per patches sent by anton we don't need this. We can branch to the
> function rathen than the dot symbol
>
> http://article.gmane.org/gmane.linux.ports.ppc.embedded/68925
> http://article.gmane.org/gmane.linux.ports.ppc.embedded/71005
Ah, true in this case, because the file never gets compiled as a module.
Alex
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH 6/7] KVM: PPC: Book3S HV: Fix ABIv2 on LE
@ 2014-06-18 15:33 ` Aneesh Kumar K.V
0 siblings, 0 replies; 24+ messages in thread
From: Aneesh Kumar K.V @ 2014-06-18 15:33 UTC (permalink / raw)
To: Alexander Graf, kvm-ppc; +Cc: kvm
Alexander Graf <agraf@suse.de> writes:
> We use ABIv2 on Little Endian systems which gets rid of the dotted function
> names. Branch to the actual functions when we see such a system.
>
> Signed-off-by: Alexander Graf <agraf@suse.de>
As per patches sent by anton we don't need this. We can branch to the
function rathen than the dot symbol
http://article.gmane.org/gmane.linux.ports.ppc.embedded/68925
http://article.gmane.org/gmane.linux.ports.ppc.embedded/71005
-aneesh
> ---
> arch/powerpc/kvm/book3s_hv_rmhandlers.S | 22 ++++++++++++++--------
> 1 file changed, 14 insertions(+), 8 deletions(-)
>
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index 1a71f60..1ff3ebd 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -36,6 +36,12 @@
> #define NAPPING_CEDE 1
> #define NAPPING_NOVCPU 2
>
> +#if defined(_CALL_ELF) && _CALL_ELF = 2
> +#define FUNC(name) name
> +#else
> +#define FUNC(name) GLUE(.,name)
> +#endif
> +
> /*
> * Call kvmppc_hv_entry in real mode.
> * Must be called with interrupts hard-disabled.
> @@ -668,9 +674,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
>
> mr r31, r4
> addi r3, r31, VCPU_FPRS_TM
> - bl .load_fp_state
> + bl FUNC(load_fp_state)
> addi r3, r31, VCPU_VRS_TM
> - bl .load_vr_state
> + bl FUNC(load_vr_state)
> mr r4, r31
> lwz r7, VCPU_VRSAVE_TM(r4)
> mtspr SPRN_VRSAVE, r7
> @@ -1414,9 +1420,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
>
> /* Save FP/VSX. */
> addi r3, r9, VCPU_FPRS_TM
> - bl .store_fp_state
> + bl FUNC(store_fp_state)
> addi r3, r9, VCPU_VRS_TM
> - bl .store_vr_state
> + bl FUNC(store_vr_state)
> mfspr r6, SPRN_VRSAVE
> stw r6, VCPU_VRSAVE_TM(r9)
> 1:
> @@ -2405,11 +2411,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
> mtmsrd r8
> isync
> addi r3,r3,VCPU_FPRS
> - bl .store_fp_state
> + bl FUNC(store_fp_state)
> #ifdef CONFIG_ALTIVEC
> BEGIN_FTR_SECTION
> addi r3,r31,VCPU_VRS
> - bl .store_vr_state
> + bl FUNC(store_vr_state)
> END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
> #endif
> mfspr r6,SPRN_VRSAVE
> @@ -2441,11 +2447,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
> mtmsrd r8
> isync
> addi r3,r4,VCPU_FPRS
> - bl .load_fp_state
> + bl FUNC(load_fp_state)
> #ifdef CONFIG_ALTIVEC
> BEGIN_FTR_SECTION
> addi r3,r31,VCPU_VRS
> - bl .load_vr_state
> + bl FUNC(load_vr_state)
> END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
> #endif
> lwz r7,VCPU_VRSAVE(r31)
-aneesh
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH v3 6/7] KVM: PPC: Book3S HV: Fix ABIv2 on LE
2014-06-18 15:23 ` Alexander Graf
@ 2014-06-24 13:34 ` Alexander Graf
-1 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-24 13:34 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm, aneesh.kumar
For code that doesn't live in modules we can just branch to the real function
names, giving us compatibility with ABIv1 and ABIv2.
Do this for the compiled-in code of HV KVM.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 1a71f60..a9de981 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -668,9 +668,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
mr r31, r4
addi r3, r31, VCPU_FPRS_TM
- bl .load_fp_state
+ bl load_fp_state
addi r3, r31, VCPU_VRS_TM
- bl .load_vr_state
+ bl load_vr_state
mr r4, r31
lwz r7, VCPU_VRSAVE_TM(r4)
mtspr SPRN_VRSAVE, r7
@@ -1414,9 +1414,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
/* Save FP/VSX. */
addi r3, r9, VCPU_FPRS_TM
- bl .store_fp_state
+ bl store_fp_state
addi r3, r9, VCPU_VRS_TM
- bl .store_vr_state
+ bl store_vr_state
mfspr r6, SPRN_VRSAVE
stw r6, VCPU_VRSAVE_TM(r9)
1:
@@ -2405,11 +2405,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
mtmsrd r8
isync
addi r3,r3,VCPU_FPRS
- bl .store_fp_state
+ bl store_fp_state
#ifdef CONFIG_ALTIVEC
BEGIN_FTR_SECTION
addi r3,r31,VCPU_VRS
- bl .store_vr_state
+ bl store_vr_state
END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
#endif
mfspr r6,SPRN_VRSAVE
@@ -2441,11 +2441,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
mtmsrd r8
isync
addi r3,r4,VCPU_FPRS
- bl .load_fp_state
+ bl load_fp_state
#ifdef CONFIG_ALTIVEC
BEGIN_FTR_SECTION
addi r3,r31,VCPU_VRS
- bl .load_vr_state
+ bl load_vr_state
END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
#endif
lwz r7,VCPU_VRSAVE(r31)
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* [PATCH v3 6/7] KVM: PPC: Book3S HV: Fix ABIv2 on LE
@ 2014-06-24 13:34 ` Alexander Graf
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-06-24 13:34 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm, aneesh.kumar
For code that doesn't live in modules we can just branch to the real function
names, giving us compatibility with ABIv1 and ABIv2.
Do this for the compiled-in code of HV KVM.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
arch/powerpc/kvm/book3s_hv_rmhandlers.S | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 1a71f60..a9de981 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -668,9 +668,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
mr r31, r4
addi r3, r31, VCPU_FPRS_TM
- bl .load_fp_state
+ bl load_fp_state
addi r3, r31, VCPU_VRS_TM
- bl .load_vr_state
+ bl load_vr_state
mr r4, r31
lwz r7, VCPU_VRSAVE_TM(r4)
mtspr SPRN_VRSAVE, r7
@@ -1414,9 +1414,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
/* Save FP/VSX. */
addi r3, r9, VCPU_FPRS_TM
- bl .store_fp_state
+ bl store_fp_state
addi r3, r9, VCPU_VRS_TM
- bl .store_vr_state
+ bl store_vr_state
mfspr r6, SPRN_VRSAVE
stw r6, VCPU_VRSAVE_TM(r9)
1:
@@ -2405,11 +2405,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
mtmsrd r8
isync
addi r3,r3,VCPU_FPRS
- bl .store_fp_state
+ bl store_fp_state
#ifdef CONFIG_ALTIVEC
BEGIN_FTR_SECTION
addi r3,r31,VCPU_VRS
- bl .store_vr_state
+ bl store_vr_state
END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
#endif
mfspr r6,SPRN_VRSAVE
@@ -2441,11 +2441,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
mtmsrd r8
isync
addi r3,r4,VCPU_FPRS
- bl .load_fp_state
+ bl load_fp_state
#ifdef CONFIG_ALTIVEC
BEGIN_FTR_SECTION
addi r3,r31,VCPU_VRS
- bl .load_vr_state
+ bl load_vr_state
END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
#endif
lwz r7,VCPU_VRSAVE(r31)
--
1.8.1.4
^ permalink raw reply related [flat|nested] 24+ messages in thread
* Re: [PATCH 0/7] KVM: PPC: Book3S HV: Enable on little endian hosts
2014-06-17 16:17 ` Alexander Graf
@ 2014-07-04 7:41 ` Alexander Graf
-1 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-07-04 7:41 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
On 17.06.14 18:17, Alexander Graf wrote:
> So far we've been able to successfully run HV KVM on big endian hosts, but
> once you dive into little endian land things start to fall apart.
>
> This patch set enables HV KVM for little endian hosts. This should be the
> final piece left missing to get little endian systems fully en par with big
> endian ones in the KVM world - modulo bugs.
>
> For now guest threading support is still slightly flaky, but I'm sure that's
> only a minor breakage somewhere that we'll find soon.
>
> v1 -> v2:
>
> - fix typo in STWX_BE
> - Add __be hints
> - Fix H_REMOVE
> - Fix dtl_idx
> - Make XICS code easier to follow and use memory for bswap
Applied to kvm-ppc-queue.
Alex
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH 0/7] KVM: PPC: Book3S HV: Enable on little endian hosts
@ 2014-07-04 7:41 ` Alexander Graf
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Graf @ 2014-07-04 7:41 UTC (permalink / raw)
To: kvm-ppc; +Cc: kvm
On 17.06.14 18:17, Alexander Graf wrote:
> So far we've been able to successfully run HV KVM on big endian hosts, but
> once you dive into little endian land things start to fall apart.
>
> This patch set enables HV KVM for little endian hosts. This should be the
> final piece left missing to get little endian systems fully en par with big
> endian ones in the KVM world - modulo bugs.
>
> For now guest threading support is still slightly flaky, but I'm sure that's
> only a minor breakage somewhere that we'll find soon.
>
> v1 -> v2:
>
> - fix typo in STWX_BE
> - Add __be hints
> - Fix H_REMOVE
> - Fix dtl_idx
> - Make XICS code easier to follow and use memory for bswap
Applied to kvm-ppc-queue.
Alex
^ permalink raw reply [flat|nested] 24+ messages in thread
end of thread, other threads:[~2014-07-04 7:41 UTC | newest]
Thread overview: 24+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-06-17 16:17 [PATCH 0/7] KVM: PPC: Book3S HV: Enable on little endian hosts Alexander Graf
2014-06-17 16:17 ` Alexander Graf
2014-06-17 16:17 ` [PATCH 1/7] PPC: Add asm helpers for BE 32bit load/store Alexander Graf
2014-06-17 16:17 ` Alexander Graf
2014-06-17 16:17 ` [PATCH 2/7] KVM: PPC: Book3S HV: Make HTAB code LE host aware Alexander Graf
2014-06-17 16:17 ` Alexander Graf
2014-06-17 16:17 ` [PATCH 3/7] KVM: PPC: Book3S HV: Access guest VPA in BE Alexander Graf
2014-06-17 16:17 ` Alexander Graf
2014-06-17 16:17 ` [PATCH 4/7] KVM: PPC: Book3S HV: Access host lppaca and shadow slb " Alexander Graf
2014-06-17 16:17 ` Alexander Graf
2014-06-17 16:17 ` [PATCH 5/7] KVM: PPC: Book3S HV: Access XICS " Alexander Graf
2014-06-17 16:17 ` Alexander Graf
2014-06-17 16:17 ` [PATCH 6/7] KVM: PPC: Book3S HV: Fix ABIv2 on LE Alexander Graf
2014-06-17 16:17 ` Alexander Graf
2014-06-18 15:21 ` Aneesh Kumar K.V
2014-06-18 15:33 ` Aneesh Kumar K.V
2014-06-18 15:23 ` Alexander Graf
2014-06-18 15:23 ` Alexander Graf
2014-06-24 13:34 ` [PATCH v3 " Alexander Graf
2014-06-24 13:34 ` Alexander Graf
2014-06-17 16:17 ` [PATCH 7/7] KVM: PPC: Book3S HV: Enable for little endian hosts Alexander Graf
2014-06-17 16:17 ` Alexander Graf
2014-07-04 7:41 ` [PATCH 0/7] KVM: PPC: Book3S HV: Enable on " Alexander Graf
2014-07-04 7:41 ` Alexander Graf
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.