* [PATCH 1/3] KVM: PPC: Book3S PR: Emulate PURR, SPURR and DSCR registers
2012-11-05 3:40 [PATCH 0/3] KVM: PPC: Fixes for PR-KVM on POWER7 Paul Mackerras
@ 2012-11-05 4:15 ` Paul Mackerras
2012-11-05 4:16 ` [PATCH 2/3] KVM: PPC: Book3S PR: Fix VSX handling Paul Mackerras
` (2 subsequent siblings)
3 siblings, 0 replies; 5+ messages in thread
From: Paul Mackerras @ 2012-11-05 4:15 UTC (permalink / raw)
To: Alexander Graf; +Cc: kvm, kvm-ppc
This adds basic emulation of the PURR and SPURR registers. We assume
we are emulating a single-threaded core, so these advance at the same
rate as the timebase. A Linux kernel running on a POWER7 expects to
be able to access these registers and is not prepared to handle a
program interrupt on accessing them.
This also adds a very minimal emulation of the DSCR (data stream
control register). Writes are ignored and reads return zero.
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
This adds these registers for all CPUs; currently we don't seem to
make the presence or absence of SPRs depend on the PVR, but maybe we
should.
arch/powerpc/include/asm/kvm_book3s.h | 2 ++
arch/powerpc/kvm/book3s_emulate.c | 16 +++++++++++++++-
2 files changed, 17 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/include/asm/kvm_book3s.h b/arch/powerpc/include/asm/kvm_book3s.h
index ab73800..c568d82 100644
--- a/arch/powerpc/include/asm/kvm_book3s.h
+++ b/arch/powerpc/include/asm/kvm_book3s.h
@@ -81,6 +81,8 @@ struct kvmppc_vcpu_book3s {
u64 sdr1;
u64 hior;
u64 msr_mask;
+ u64 purr_offset;
+ u64 spurr_offset;
#ifdef CONFIG_PPC_BOOK3S_32
u32 vsid_pool[VSID_POOL_SIZE];
u32 vsid_next;
diff --git a/arch/powerpc/kvm/book3s_emulate.c b/arch/powerpc/kvm/book3s_emulate.c
index b9a989d..d31a716 100644
--- a/arch/powerpc/kvm/book3s_emulate.c
+++ b/arch/powerpc/kvm/book3s_emulate.c
@@ -22,6 +22,7 @@
#include <asm/kvm_book3s.h>
#include <asm/reg.h>
#include <asm/switch_to.h>
+#include <asm/time.h>
#define OP_19_XOP_RFID 18
#define OP_19_XOP_RFI 50
@@ -395,6 +396,12 @@ int kvmppc_core_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, ulong spr_val)
(mfmsr() & MSR_HV))
vcpu->arch.hflags |= BOOK3S_HFLAG_DCBZ32;
break;
+ case SPRN_PURR:
+ to_book3s(vcpu)->purr_offset = spr_val - get_tb();
+ break;
+ case SPRN_SPURR:
+ to_book3s(vcpu)->spurr_offset = spr_val - get_tb();
+ break;
case SPRN_GQR0:
case SPRN_GQR1:
case SPRN_GQR2:
@@ -412,6 +419,7 @@ int kvmppc_core_emulate_mtspr(struct kvm_vcpu *vcpu, int sprn, ulong spr_val)
case SPRN_CTRLF:
case SPRN_CTRLT:
case SPRN_L2CR:
+ case SPRN_DSCR:
case SPRN_MMCR0_GEKKO:
case SPRN_MMCR1_GEKKO:
case SPRN_PMC1_GEKKO:
@@ -483,9 +491,15 @@ int kvmppc_core_emulate_mfspr(struct kvm_vcpu *vcpu, int sprn, ulong *spr_val)
*spr_val = to_book3s(vcpu)->hid[5];
break;
case SPRN_CFAR:
- case SPRN_PURR:
+ case SPRN_DSCR:
*spr_val = 0;
break;
+ case SPRN_PURR:
+ *spr_val = get_tb() + to_book3s(vcpu)->purr_offset;
+ break;
+ case SPRN_SPURR:
+ *spr_val = get_tb() + to_book3s(vcpu)->purr_offset;
+ break;
case SPRN_GQR0:
case SPRN_GQR1:
case SPRN_GQR2:
--
1.7.10.4
^ permalink raw reply related [flat|nested] 5+ messages in thread* [PATCH 2/3] KVM: PPC: Book3S PR: Fix VSX handling
2012-11-05 3:40 [PATCH 0/3] KVM: PPC: Fixes for PR-KVM on POWER7 Paul Mackerras
2012-11-05 4:15 ` [PATCH 1/3] KVM: PPC: Book3S PR: Emulate PURR, SPURR and DSCR registers Paul Mackerras
@ 2012-11-05 4:16 ` Paul Mackerras
2012-11-05 4:17 ` [PATCH 3/3] KVM: PPC: Book3S PR: MSR_DE doesn't exist on Book 3S Paul Mackerras
2012-11-26 13:30 ` [PATCH 0/3] KVM: PPC: Fixes for PR-KVM on POWER7 Alexander Graf
3 siblings, 0 replies; 5+ messages in thread
From: Paul Mackerras @ 2012-11-05 4:16 UTC (permalink / raw)
To: Alexander Graf; +Cc: kvm, kvm-ppc
This fixes various issues in how we were handling the VSX registers
that exist on POWER7 machines. First, we were running off the end
of the current->thread.fpr[] array. Ultimately this was because the
vcpu->arch.vsr[] array is sized to be able to store both the FP
registers and the extra VSX registers (i.e. 64 entries), but PR KVM
only uses it for the extra VSX registers (i.e. 32 entries).
Secondly, calling load_up_vsx() from C code is a really bad idea,
because it jumps to fast_exception_return at the end, rather than
returning with a blr instruction. This was causing it to jump off
to a random location with random register contents, since it was using
the largely uninitialized stack frame created by kvmppc_load_up_vsx.
In fact, it isn't necessary to call either __giveup_vsx or load_up_vsx,
since giveup_fpu and load_up_fpu handle the extra VSX registers as well
as the standard FP registers on machines with VSX. Also, since VSX
instructions can access the VMX registers and the FP registers as well
as the extra VSX registers, we have to load up the FP and VMX registers
before we can turn on the MSR_VSX bit for the guest. Conversely, if
we save away any of the VSX or FP registers, we have to turn off MSR_VSX
for the guest.
To handle all this, it is more convenient for a single call to
kvmppc_giveup_ext() to handle all the state saving that needs to be done,
so we make it take a set of MSR bits rather than just one, and the switch
statement becomes a series of if statements. Similarly kvmppc_handle_ext
needs to be able to load up more than one set of registers.
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
arch/powerpc/include/asm/reg.h | 1 +
arch/powerpc/kvm/book3s_exports.c | 3 -
arch/powerpc/kvm/book3s_pr.c | 112 ++++++++++++++++++----------------
arch/powerpc/kvm/book3s_rmhandlers.S | 3 -
4 files changed, 62 insertions(+), 57 deletions(-)
diff --git a/arch/powerpc/include/asm/reg.h b/arch/powerpc/include/asm/reg.h
index 6386086..5dc7014 100644
--- a/arch/powerpc/include/asm/reg.h
+++ b/arch/powerpc/include/asm/reg.h
@@ -515,6 +515,7 @@
#define SRR1_WS_DEEPER 0x00020000 /* Some resources not maintained */
#define SRR1_WS_DEEP 0x00010000 /* All resources maintained */
#define SRR1_PROGFPE 0x00100000 /* Floating Point Enabled */
+#define SRR1_PROGILL 0x00080000 /* Illegal instruction */
#define SRR1_PROGPRIV 0x00040000 /* Privileged instruction */
#define SRR1_PROGTRAP 0x00020000 /* Trap */
#define SRR1_PROGADDR 0x00010000 /* SRR0 contains subsequent addr */
diff --git a/arch/powerpc/kvm/book3s_exports.c b/arch/powerpc/kvm/book3s_exports.c
index a150817..7057a02 100644
--- a/arch/powerpc/kvm/book3s_exports.c
+++ b/arch/powerpc/kvm/book3s_exports.c
@@ -28,8 +28,5 @@ EXPORT_SYMBOL_GPL(kvmppc_load_up_fpu);
#ifdef CONFIG_ALTIVEC
EXPORT_SYMBOL_GPL(kvmppc_load_up_altivec);
#endif
-#ifdef CONFIG_VSX
-EXPORT_SYMBOL_GPL(kvmppc_load_up_vsx);
-#endif
#endif
diff --git a/arch/powerpc/kvm/book3s_pr.c b/arch/powerpc/kvm/book3s_pr.c
index b853696..15d74e5 100644
--- a/arch/powerpc/kvm/book3s_pr.c
+++ b/arch/powerpc/kvm/book3s_pr.c
@@ -81,9 +81,7 @@ void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu)
svcpu_put(svcpu);
#endif
- kvmppc_giveup_ext(vcpu, MSR_FP);
- kvmppc_giveup_ext(vcpu, MSR_VEC);
- kvmppc_giveup_ext(vcpu, MSR_VSX);
+ kvmppc_giveup_ext(vcpu, MSR_FP | MSR_VEC | MSR_VSX);
vcpu->cpu = -1;
}
@@ -433,10 +431,7 @@ int kvmppc_handle_pagefault(struct kvm_run *run, struct kvm_vcpu *vcpu,
static inline int get_fpr_index(int i)
{
-#ifdef CONFIG_VSX
- i *= 2;
-#endif
- return i;
+ return i * TS_FPRWIDTH;
}
/* Give up external provider (FPU, Altivec, VSX) */
@@ -450,41 +445,49 @@ void kvmppc_giveup_ext(struct kvm_vcpu *vcpu, ulong msr)
u64 *thread_fpr = (u64*)t->fpr;
int i;
- if (!(vcpu->arch.guest_owned_ext & msr))
+ /*
+ * VSX instructions can access FP and vector registers, so if
+ * we are giving up VSX, make sure we give up FP and VMX as well.
+ */
+ if (msr & MSR_VSX)
+ msr |= MSR_FP | MSR_VEC;
+
+ msr &= vcpu->arch.guest_owned_ext;
+ if (!msr)
return;
#ifdef DEBUG_EXT
printk(KERN_INFO "Giving up ext 0x%lx\n", msr);
#endif
- switch (msr) {
- case MSR_FP:
+ if (msr & MSR_FP) {
+ /*
+ * Note that on CPUs with VSX, giveup_fpu stores
+ * both the traditional FP registers and the added VSX
+ * registers into thread.fpr[].
+ */
giveup_fpu(current);
for (i = 0; i < ARRAY_SIZE(vcpu->arch.fpr); i++)
vcpu_fpr[i] = thread_fpr[get_fpr_index(i)];
vcpu->arch.fpscr = t->fpscr.val;
- break;
- case MSR_VEC:
+
+#ifdef CONFIG_VSX
+ if (cpu_has_feature(CPU_FTR_VSX))
+ for (i = 0; i < ARRAY_SIZE(vcpu->arch.vsr) / 2; i++)
+ vcpu_vsx[i] = thread_fpr[get_fpr_index(i) + 1];
+#endif
+ }
+
#ifdef CONFIG_ALTIVEC
+ if (msr & MSR_VEC) {
giveup_altivec(current);
memcpy(vcpu->arch.vr, t->vr, sizeof(vcpu->arch.vr));
vcpu->arch.vscr = t->vscr;
-#endif
- break;
- case MSR_VSX:
-#ifdef CONFIG_VSX
- __giveup_vsx(current);
- for (i = 0; i < ARRAY_SIZE(vcpu->arch.vsr); i++)
- vcpu_vsx[i] = thread_fpr[get_fpr_index(i) + 1];
-#endif
- break;
- default:
- BUG();
}
+#endif
- vcpu->arch.guest_owned_ext &= ~msr;
- current->thread.regs->msr &= ~msr;
+ vcpu->arch.guest_owned_ext &= ~(msr | MSR_VSX);
kvmppc_recalc_shadow_msr(vcpu);
}
@@ -544,47 +547,56 @@ static int kvmppc_handle_ext(struct kvm_vcpu *vcpu, unsigned int exit_nr,
return RESUME_GUEST;
}
- /* We already own the ext */
- if (vcpu->arch.guest_owned_ext & msr) {
- return RESUME_GUEST;
+ if (msr == MSR_VSX) {
+ /* No VSX? Give an illegal instruction interrupt */
+#ifdef CONFIG_VSX
+ if (!cpu_has_feature(CPU_FTR_VSX))
+#endif
+ {
+ kvmppc_core_queue_program(vcpu, SRR1_PROGILL);
+ return RESUME_GUEST;
+ }
+
+ /*
+ * We have to load up all the FP and VMX registers before
+ * we can let the guest use VSX instructions.
+ */
+ msr = MSR_FP | MSR_VEC | MSR_VSX;
}
+ /* See if we already own all the ext(s) needed */
+ msr &= ~vcpu->arch.guest_owned_ext;
+ if (!msr)
+ return RESUME_GUEST;
+
#ifdef DEBUG_EXT
printk(KERN_INFO "Loading up ext 0x%lx\n", msr);
#endif
current->thread.regs->msr |= msr;
- switch (msr) {
- case MSR_FP:
+ if (msr & MSR_FP) {
for (i = 0; i < ARRAY_SIZE(vcpu->arch.fpr); i++)
thread_fpr[get_fpr_index(i)] = vcpu_fpr[i];
-
+#ifdef CONFIG_VSX
+ for (i = 0; i < ARRAY_SIZE(vcpu->arch.vsr) / 2; i++)
+ thread_fpr[get_fpr_index(i) + 1] = vcpu_vsx[i];
+#endif
t->fpscr.val = vcpu->arch.fpscr;
t->fpexc_mode = 0;
kvmppc_load_up_fpu();
- break;
- case MSR_VEC:
+ }
+
+ if (msr & MSR_VEC) {
#ifdef CONFIG_ALTIVEC
memcpy(t->vr, vcpu->arch.vr, sizeof(vcpu->arch.vr));
t->vscr = vcpu->arch.vscr;
t->vrsave = -1;
kvmppc_load_up_altivec();
#endif
- break;
- case MSR_VSX:
-#ifdef CONFIG_VSX
- for (i = 0; i < ARRAY_SIZE(vcpu->arch.vsr); i++)
- thread_fpr[get_fpr_index(i) + 1] = vcpu_vsx[i];
- kvmppc_load_up_vsx();
-#endif
- break;
- default:
- BUG();
}
vcpu->arch.guest_owned_ext |= msr;
-
kvmppc_recalc_shadow_msr(vcpu);
return RESUME_GUEST;
@@ -1134,7 +1146,7 @@ int kvmppc_vcpu_run(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
/* Save VSX state in stack */
used_vsr = current->thread.used_vsr;
if (used_vsr && (current->thread.regs->msr & MSR_VSX))
- __giveup_vsx(current);
+ __giveup_vsx(current);
#endif
/* Remember the MSR with disabled extensions */
@@ -1151,14 +1163,12 @@ int kvmppc_vcpu_run(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
/* No need for kvm_guest_exit. It's done in handle_exit.
We also get here with interrupts enabled. */
- current->thread.regs->msr = ext_msr;
-
/* Make sure we save the guest FPU/Altivec/VSX state */
- kvmppc_giveup_ext(vcpu, MSR_FP);
- kvmppc_giveup_ext(vcpu, MSR_VEC);
- kvmppc_giveup_ext(vcpu, MSR_VSX);
+ kvmppc_giveup_ext(vcpu, MSR_FP | MSR_VEC | MSR_VSX);
+
+ current->thread.regs->msr = ext_msr;
- /* Restore FPU state from stack */
+ /* Restore FPU/VSX state from stack */
memcpy(current->thread.fpr, fpr, sizeof(current->thread.fpr));
current->thread.fpscr.val = fpscr;
current->thread.fpexc_mode = fpexc_mode;
diff --git a/arch/powerpc/kvm/book3s_rmhandlers.S b/arch/powerpc/kvm/book3s_rmhandlers.S
index b2f8258..8f7633e 100644
--- a/arch/powerpc/kvm/book3s_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_rmhandlers.S
@@ -234,8 +234,5 @@ define_load_up(fpu)
#ifdef CONFIG_ALTIVEC
define_load_up(altivec)
#endif
-#ifdef CONFIG_VSX
-define_load_up(vsx)
-#endif
#include "book3s_segment.S"
--
1.7.10.4
^ permalink raw reply related [flat|nested] 5+ messages in thread