KVM: PPC: book3s_pr: Simplify transitions between virtual and real mode
This simplifies the way that the book3s_pr makes the transition to
real mode when entering the guest. We now call kvmppc_entry_trampoline
(renamed from kvmppc_rmcall) in the base kernel using a normal function
call instead of doing an indirect call through a pointer in the vcpu.
If kvm is a module, the module loader takes care of generating a
trampoline as it does for other calls to functions outside the module.
kvmppc_entry_trampoline then disables interrupts and jumps to
kvmppc_handler_trampoline_enter in real mode using an rfi[d].
That then uses the link register as the address to return to
(potentially in module space) when the guest exits.
This also simplifies the way that we call the Linux interrupt handler
when we exit the guest due to an external, decrementer or performance
monitor interrupt. Instead of turning on the MMU, then deciding that
we need to call the Linux handler and turning the MMU back off again,
we now go straight to the handler at the point where we would turn the
MMU on. The handler will then return to the virtual-mode code
(potentially in the module).
Along the way, this moves the setting and clearing of the HID5 DCBZ32
bit into real-mode interrupts-off code, and also makes sure that
we clear the MSR[RI] bit before loading values into SRR0/1.
The net result is that we no longer need any code addresses to be
stored in vcpu->arch.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
diff --git a/arch/powerpc/kvm/book3s_segment.S b/arch/powerpc/kvm/book3s_segment.S
index 678b6be..0676ae2 100644
--- a/arch/powerpc/kvm/book3s_segment.S
+++ b/arch/powerpc/kvm/book3s_segment.S
@@ -23,6 +23,7 @@
#define GET_SHADOW_VCPU(reg) \
mr reg, r13
+#define MTMSR_EERI(reg) mtmsrd (reg),1
#elif defined(CONFIG_PPC_BOOK3S_32)
@@ -30,6 +31,7 @@
tophys(reg, r2); \
lwz reg, (THREAD + THREAD_KVM_SVCPU)(reg); \
tophys(reg, reg)
+#define MTMSR_EERI(reg) mtmsr (reg)
#endif
@@ -57,10 +59,12 @@
/* Required state:
*
* MSR = ~IR|DR
- * R13 = PACA
* R1 = host R1
* R2 = host R2
- * R10 = guest MSR
+ * R4 = guest shadow MSR
+ * R5 = normal host MSR
+ * R6 = current host MSR (EE, IR, DR off)
+ * LR = highmem guest exit code
* all other volatile GPRS = free
* SVCPU[CR] = guest CR
* SVCPU[XER] = guest XER
@@ -71,15 +75,15 @@
/* r3 = shadow vcpu */
GET_SHADOW_VCPU(r3)
+ /* Save guest exit handler address and MSR */
+ mflr r0
+ PPC_STL r0, HSTATE_VMHANDLER(r3)
+ PPC_STL r5, HSTATE_HOST_MSR(r3)
+
/* Save R1/R2 in the PACA (64-bit) or shadow_vcpu (32-bit) */
PPC_STL r1, HSTATE_HOST_R1(r3)
PPC_STL r2, HSTATE_HOST_R2(r3)
- /* Move SRR0 and SRR1 into the respective regs */
- PPC_LL r9, SVCPU_PC(r3)
- mtsrr0 r9
- mtsrr1 r10
-
/* Activate guest mode, so faults get handled by KVM */
li r11, KVM_GUEST_MODE_GUEST
stb r11, HSTATE_IN_GUEST(r3)
@@ -87,17 +91,46 @@
/* Switch to guest segment. This is subarch specific. */
LOAD_GUEST_SEGMENTS
+#ifdef CONFIG_PPC_BOOK3S_64
+ /* Some guests may need to have dcbz set to 32 byte length.
+ *
+ * Usually we ensure that by patching the guest's instructions
+ * to trap on dcbz and emulate it in the hypervisor.
+ *
+ * If we can, we should tell the CPU to use 32 byte dcbz though,
+ * because that's a lot faster.
+ */
+ lbz r0, HSTATE_RESTORE_HID5(r3)
+ cmpwi r0, 0
+ beq no_dcbz32_on
+
+ mfspr r0,SPRN_HID5
+ ori r0, r0, 0x80 /* XXX HID5_dcbz32 = 0x80 */
+ mtspr SPRN_HID5,r0
+no_dcbz32_on:
+
+#endif /* CONFIG_PPC_BOOK3S_64 */
+
/* Enter guest */
- PPC_LL r4, SVCPU_CTR(r3)
- PPC_LL r5, SVCPU_LR(r3)
- lwz r6, SVCPU_CR(r3)
- lwz r7, SVCPU_XER(r3)
+ PPC_LL r8, SVCPU_CTR(r3)
+ PPC_LL r9, SVCPU_LR(r3)
+ lwz r10, SVCPU_CR(r3)
+ lwz r11, SVCPU_XER(r3)
- mtctr r4
- mtlr r5
- mtcr r6
- mtxer r7
+ mtctr r8
+ mtlr r9
+ mtcr r10
+ mtxer r11
+
+ /* Move SRR0 and SRR1 into the respective regs */
+ PPC_LL r9, SVCPU_PC(r3)
+ /* First clear RI in our current MSR value */
+ li r0, MSR_RI
+ andc r6, r6, r0
+ MTMSR_EERI(r6)
+ mtsrr0 r9
+ mtsrr1 r4
PPC_LL r0, SVCPU_R0(r3)
PPC_LL r1, SVCPU_R1(r3)
@@ -259,6 +292,43 @@
/* Switch back to host MMU */
LOAD_HOST_SEGMENTS
+#ifdef CONFIG_PPC_BOOK3S_64
+
+ lbz r5, HSTATE_RESTORE_HID5(r13)
+ cmpwi r5, 0
+ beq no_dcbz32_off
+
+ li r4, 0
+ mfspr r5,SPRN_HID5
+ rldimi r5,r4,6,56
+ mtspr SPRN_HID5,r5
+
+no_dcbz32_off:
+
+#endif /* CONFIG_PPC_BOOK3S_64 */
+
+ /*
+ * For some interrupts, we need to call the real Linux
+ * handler, so it can do work for us. This has to happen
+ * as if the interrupt arrived from the kernel though,
+ * so let's fake it here where most state is restored.
+ *
+ * Having set up SRR0/1 with the address where we want
+ * to continue with relocation on (potentially in module
+ * space), we either just go straight there with rfi[d],
+ * or we jump to an interrupt handler with bctr if there
+ * is an interrupt to be handled first. In the latter
+ * case, the rfi[d] at the end of the interrupt handler
+ * will get us back to where we want to continue.
+ */
+
+ cmpwi r12, BOOK3S_INTERRUPT_EXTERNAL
+ beq 1f
+ cmpwi r12, BOOK3S_INTERRUPT_DECREMENTER
+ beq 1f
+ cmpwi r12, BOOK3S_INTERRUPT_PERFMON
+1: mtctr r12
+
/* Register usage at this point:
*
* R1 = host R1
@@ -269,13 +339,15 @@
*
*/
- /* RFI into the highmem handler */
- mfmsr r7
- ori r7, r7, MSR_IR|MSR_DR|MSR_RI|MSR_ME /* Enable paging */
- mtsrr1 r7
- /* Load highmem handler address */
+ PPC_LL r6, HSTATE_HOST_MSR(r13)
PPC_LL r8, HSTATE_VMHANDLER(r13)
+
+ /* Restore host msr -> SRR1 */
+ mtsrr1 r6
+ /* Load highmem handler address */
mtsrr0 r8
+ /* RFI into the highmem handler, or jump to interrupt handler */
+ beqctr
RFI
kvmppc_handler_trampoline_exit_end: