KVM: ppc: Refactor powerpc.c to relocate 440-specific code

This introduces a set of core-provided hooks. For 440, some of these are
implemented by booke.c, with the rest in (the new) 44x.c.

Note that these hooks are link-time, not run-time. Since it is not possible to
build a single kernel for both e500 and 440 (for example), using function
pointers would only add overhead.

Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@redhat.com>

7 files changed, 187 insertions, 113 deletions

diff --git a/arch/powerpc/kvm/44x.c b/arch/powerpc/kvm/44x.c
new file mode 100644
index 000000000000..fcf8c7d0af45
--- /dev/null
+++ b/arch/powerpc/kvm/44x.c
@@ -0,0 +1,123 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ * Copyright IBM Corp. 2008
+ *
+ * Authors: Hollis Blanchard <hollisb@us.ibm.com>
+ */
+
+#include <linux/kvm_host.h>
+#include <asm/reg.h>
+#include <asm/cputable.h>
+#include <asm/tlbflush.h>
+
+#include "44x_tlb.h"
+
+/* Note: clearing MSR[DE] just means that the debug interrupt will not be
+ * delivered *immediately*. Instead, it simply sets the appropriate DBSR bits.
+ * If those DBSR bits are still set when MSR[DE] is re-enabled, the interrupt
+ * will be delivered as an "imprecise debug event" (which is indicated by
+ * DBSR[IDE].
+ */
+static void kvm44x_disable_debug_interrupts(void)
+{
+        mtmsr(mfmsr() & ~MSR_DE);
+}
+
+void kvmppc_core_load_host_debugstate(struct kvm_vcpu *vcpu)
+{
+        kvm44x_disable_debug_interrupts();
+
+        mtspr(SPRN_IAC1, vcpu->arch.host_iac[0]);
+        mtspr(SPRN_IAC2, vcpu->arch.host_iac[1]);
+        mtspr(SPRN_IAC3, vcpu->arch.host_iac[2]);
+        mtspr(SPRN_IAC4, vcpu->arch.host_iac[3]);
+        mtspr(SPRN_DBCR1, vcpu->arch.host_dbcr1);
+        mtspr(SPRN_DBCR2, vcpu->arch.host_dbcr2);
+        mtspr(SPRN_DBCR0, vcpu->arch.host_dbcr0);
+        mtmsr(vcpu->arch.host_msr);
+}
+
+void kvmppc_core_load_guest_debugstate(struct kvm_vcpu *vcpu)
+{
+        struct kvm_guest_debug *dbg = &vcpu->guest_debug;
+        u32 dbcr0 = 0;
+
+        vcpu->arch.host_msr = mfmsr();
+        kvm44x_disable_debug_interrupts();
+
+        /* Save host debug register state. */
+        vcpu->arch.host_iac[0] = mfspr(SPRN_IAC1);
+        vcpu->arch.host_iac[1] = mfspr(SPRN_IAC2);
+        vcpu->arch.host_iac[2] = mfspr(SPRN_IAC3);
+        vcpu->arch.host_iac[3] = mfspr(SPRN_IAC4);
+        vcpu->arch.host_dbcr0 = mfspr(SPRN_DBCR0);
+        vcpu->arch.host_dbcr1 = mfspr(SPRN_DBCR1);
+        vcpu->arch.host_dbcr2 = mfspr(SPRN_DBCR2);
+
+        /* set registers up for guest */
+
+        if (dbg->bp[0]) {
+                mtspr(SPRN_IAC1, dbg->bp[0]);
+                dbcr0 |= DBCR0_IAC1 | DBCR0_IDM;
+        }
+        if (dbg->bp[1]) {
+                mtspr(SPRN_IAC2, dbg->bp[1]);
+                dbcr0 |= DBCR0_IAC2 | DBCR0_IDM;
+        }
+        if (dbg->bp[2]) {
+                mtspr(SPRN_IAC3, dbg->bp[2]);
+                dbcr0 |= DBCR0_IAC3 | DBCR0_IDM;
+        }
+        if (dbg->bp[3]) {
+                mtspr(SPRN_IAC4, dbg->bp[3]);
+                dbcr0 |= DBCR0_IAC4 | DBCR0_IDM;
+        }
+
+        mtspr(SPRN_DBCR0, dbcr0);
+        mtspr(SPRN_DBCR1, 0);
+        mtspr(SPRN_DBCR2, 0);
+}
+
+void kvmppc_core_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
+{
+        int i;
+
+        /* Mark every guest entry in the shadow TLB entry modified, so that they
+         * will all be reloaded on the next vcpu run (instead of being
+         * demand-faulted). */
+        for (i = 0; i <= tlb_44x_hwater; i++)
+                kvmppc_tlbe_set_modified(vcpu, i);
+}
+
+void kvmppc_core_vcpu_put(struct kvm_vcpu *vcpu)
+{
+        /* Don't leave guest TLB entries resident when being de-scheduled. */
+        /* XXX It would be nice to differentiate between heavyweight exit and
+         * sched_out here, since we could avoid the TLB flush for heavyweight
+         * exits. */
+        _tlbia();
+}
+
+int kvmppc_core_check_processor_compat(void)
+{
+        int r;
+
+        if (strcmp(cur_cpu_spec->platform, "ppc440") == 0)
+                r = 0;
+        else
+                r = -ENOTSUPP;
+
+        return r;
+}
diff --git a/arch/powerpc/kvm/44x_tlb.h b/arch/powerpc/kvm/44x_tlb.h
index e5b0a76798bd..357d79ae5493 100644
--- a/arch/powerpc/kvm/44x_tlb.h
+++ b/arch/powerpc/kvm/44x_tlb.h
@@ -29,6 +29,7 @@ extern struct kvmppc_44x_tlbe *kvmppc_44x_dtlb_search(struct kvm_vcpu *vcpu,
                                                       gva_t eaddr);
 extern struct kvmppc_44x_tlbe *kvmppc_44x_itlb_search(struct kvm_vcpu *vcpu,
                                                       gva_t eaddr);
+extern void kvmppc_tlbe_set_modified(struct kvm_vcpu *vcpu, unsigned int i);
 
 /* TLB helper functions */
 static inline unsigned int get_tlb_size(const struct kvmppc_44x_tlbe *tlbe)
diff --git a/arch/powerpc/kvm/Kconfig b/arch/powerpc/kvm/Kconfig
index ffed96f817f7..37e9b3c52a38 100644
--- a/arch/powerpc/kvm/Kconfig
+++ b/arch/powerpc/kvm/Kconfig
@@ -16,11 +16,9 @@ if VIRTUALIZATION
 
 config KVM
         bool "Kernel-based Virtual Machine (KVM) support"
-        depends on 44x && EXPERIMENTAL
+        depends on EXPERIMENTAL
         select PREEMPT_NOTIFIERS
         select ANON_INODES
-        # We can only run on Book E hosts so far
-        select KVM_BOOKE
         ---help---
           Support hosting virtualized guest machines. You will also
           need to select one or more of the processor modules below.
@@ -30,12 +28,11 @@ config KVM
 
           If unsure, say N.
 
-config KVM_BOOKE
-        bool "KVM support for Book E PowerPC processors"
+config KVM_440
+        bool "KVM support for PowerPC 440 processors"
         depends on KVM && 44x
         ---help---
-          Provides host support for KVM on Book E PowerPC processors. Currently
-          this works on 440 processors only.
+          KVM can run unmodified 440 guest kernels on 440 host processors.
 
 config KVM_TRACE
         bool "KVM trace support"
diff --git a/arch/powerpc/kvm/Makefile b/arch/powerpc/kvm/Makefile
index a7f857446c8a..f5e33756f318 100644
--- a/arch/powerpc/kvm/Makefile
+++ b/arch/powerpc/kvm/Makefile
@@ -13,5 +13,5 @@ obj-$(CONFIG_KVM) += kvm.o
 
 AFLAGS_booke_interrupts.o := -I$(obj)
 
-kvm-booke-objs := booke.o booke_interrupts.o 44x_tlb.o
-obj-$(CONFIG_KVM_BOOKE) += kvm-booke.o
+kvm-440-objs := booke.o booke_interrupts.o 44x.o 44x_tlb.o
+obj-$(CONFIG_KVM_440) += kvm-440.o
diff --git a/arch/powerpc/kvm/booke.c b/arch/powerpc/kvm/booke.c
index b1e90a15155a..138014acf3cf 100644
--- a/arch/powerpc/kvm/booke.c
+++ b/arch/powerpc/kvm/booke.c
@@ -134,6 +134,40 @@ void kvmppc_dump_vcpu(struct kvm_vcpu *vcpu)
         }
 }
 
+static void kvmppc_booke_queue_exception(struct kvm_vcpu *vcpu, int exception)
+{
+        unsigned int priority = exception_priority[exception];
+        set_bit(priority, &vcpu->arch.pending_exceptions);
+}
+
+static void kvmppc_booke_clear_exception(struct kvm_vcpu *vcpu, int exception)
+{
+        unsigned int priority = exception_priority[exception];
+        clear_bit(priority, &vcpu->arch.pending_exceptions);
+}
+
+void kvmppc_core_queue_program(struct kvm_vcpu *vcpu)
+{
+        kvmppc_booke_queue_exception(vcpu, BOOKE_INTERRUPT_PROGRAM);
+}
+
+void kvmppc_core_queue_dec(struct kvm_vcpu *vcpu)
+{
+        kvmppc_booke_queue_exception(vcpu, BOOKE_INTERRUPT_DECREMENTER);
+}
+
+int kvmppc_core_pending_dec(struct kvm_vcpu *vcpu)
+{
+        unsigned int priority = exception_priority[BOOKE_INTERRUPT_DECREMENTER];
+        return test_bit(priority, &vcpu->arch.pending_exceptions);
+}
+
+void kvmppc_core_queue_external(struct kvm_vcpu *vcpu,
+                                struct kvm_interrupt *irq)
+{
+        kvmppc_booke_queue_exception(vcpu, BOOKE_INTERRUPT_EXTERNAL);
+}
+
 /* Check if we are ready to deliver the interrupt */
 static int kvmppc_can_deliver_interrupt(struct kvm_vcpu *vcpu, int interrupt)
 {
@@ -168,7 +202,7 @@ static int kvmppc_can_deliver_interrupt(struct kvm_vcpu *vcpu, int interrupt)
         return r;
 }
 
-static void kvmppc_deliver_interrupt(struct kvm_vcpu *vcpu, int interrupt)
+static void kvmppc_booke_deliver_interrupt(struct kvm_vcpu *vcpu, int interrupt)
 {
         switch (interrupt) {
         case BOOKE_INTERRUPT_DECREMENTER:
@@ -183,7 +217,7 @@ static void kvmppc_deliver_interrupt(struct kvm_vcpu *vcpu, int interrupt)
 }
 
 /* Check pending exceptions and deliver one, if possible. */
-void kvmppc_check_and_deliver_interrupts(struct kvm_vcpu *vcpu)
+void kvmppc_core_deliver_interrupts(struct kvm_vcpu *vcpu)
 {
         unsigned long *pending = &vcpu->arch.pending_exceptions;
         unsigned int exception;
@@ -193,8 +227,8 @@ void kvmppc_check_and_deliver_interrupts(struct kvm_vcpu *vcpu)
         while (priority <= BOOKE_MAX_INTERRUPT) {
                 exception = priority_exception[priority];
                 if (kvmppc_can_deliver_interrupt(vcpu, exception)) {
-                        kvmppc_clear_exception(vcpu, exception);
-                        kvmppc_deliver_interrupt(vcpu, exception);
+                        kvmppc_booke_clear_exception(vcpu, exception);
+                        kvmppc_booke_deliver_interrupt(vcpu, exception);
                         break;
                 }
 
@@ -251,7 +285,7 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
                         /* Program traps generated by user-level software must be handled
                          * by the guest kernel. */
                         vcpu->arch.esr = vcpu->arch.fault_esr;
-                        kvmppc_queue_exception(vcpu, BOOKE_INTERRUPT_PROGRAM);
+                        kvmppc_booke_queue_exception(vcpu, BOOKE_INTERRUPT_PROGRAM);
                         r = RESUME_GUEST;
                         break;
                 }
@@ -284,27 +318,27 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
                 break;
 
         case BOOKE_INTERRUPT_FP_UNAVAIL:
-                kvmppc_queue_exception(vcpu, exit_nr);
+                kvmppc_booke_queue_exception(vcpu, exit_nr);
                 r = RESUME_GUEST;
                 break;
 
         case BOOKE_INTERRUPT_DATA_STORAGE:
                 vcpu->arch.dear = vcpu->arch.fault_dear;
                 vcpu->arch.esr = vcpu->arch.fault_esr;
-                kvmppc_queue_exception(vcpu, exit_nr);
+                kvmppc_booke_queue_exception(vcpu, exit_nr);
                 vcpu->stat.dsi_exits++;
                 r = RESUME_GUEST;
                 break;
 
         case BOOKE_INTERRUPT_INST_STORAGE:
                 vcpu->arch.esr = vcpu->arch.fault_esr;
-                kvmppc_queue_exception(vcpu, exit_nr);
+                kvmppc_booke_queue_exception(vcpu, exit_nr);
                 vcpu->stat.isi_exits++;
                 r = RESUME_GUEST;
                 break;
 
         case BOOKE_INTERRUPT_SYSCALL:
-                kvmppc_queue_exception(vcpu, exit_nr);
+                kvmppc_booke_queue_exception(vcpu, exit_nr);
                 vcpu->stat.syscall_exits++;
                 r = RESUME_GUEST;
                 break;
@@ -318,7 +352,7 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
                 gtlbe = kvmppc_44x_dtlb_search(vcpu, eaddr);
                 if (!gtlbe) {
                         /* The guest didn't have a mapping for it. */
-                        kvmppc_queue_exception(vcpu, exit_nr);
+                        kvmppc_booke_queue_exception(vcpu, exit_nr);
                         vcpu->arch.dear = vcpu->arch.fault_dear;
                         vcpu->arch.esr = vcpu->arch.fault_esr;
                         vcpu->stat.dtlb_real_miss_exits++;
@@ -360,7 +394,7 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
                 gtlbe = kvmppc_44x_itlb_search(vcpu, eaddr);
                 if (!gtlbe) {
                         /* The guest didn't have a mapping for it. */
-                        kvmppc_queue_exception(vcpu, exit_nr);
+                        kvmppc_booke_queue_exception(vcpu, exit_nr);
                         vcpu->stat.itlb_real_miss_exits++;
                         break;
                 }
@@ -380,8 +414,7 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
                                        gtlbe->word2);
                 } else {
                         /* Guest mapped and leaped at non-RAM! */
-                        kvmppc_queue_exception(vcpu,
-                                               BOOKE_INTERRUPT_MACHINE_CHECK);
+                        kvmppc_booke_queue_exception(vcpu, BOOKE_INTERRUPT_MACHINE_CHECK);
                 }
 
                 break;
@@ -409,7 +442,7 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu,
 
         local_irq_disable();
 
-        kvmppc_check_and_deliver_interrupts(vcpu);
+        kvmppc_core_deliver_interrupts(vcpu);
 
         /* Do some exit accounting. */
         vcpu->stat.sum_exits++;
diff --git a/arch/powerpc/kvm/emulate.c b/arch/powerpc/kvm/emulate.c
index 0ce8ed539bae..c5d2bfcf567a 100644
--- a/arch/powerpc/kvm/emulate.c
+++ b/arch/powerpc/kvm/emulate.c
@@ -139,7 +139,7 @@ int kvmppc_emulate_instruction(struct kvm_run *run, struct kvm_vcpu *vcpu)
         switch (get_op(inst)) {
         case 3:                                                 /* trap */
                 printk("trap!\n");
-                kvmppc_queue_exception(vcpu, BOOKE_INTERRUPT_PROGRAM);
+                kvmppc_core_queue_program(vcpu);
                 advance = 0;
                 break;
 
diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
index 8bef0efcdfe1..8d0aaf96d838 100644
--- a/arch/powerpc/kvm/powerpc.c
+++ b/arch/powerpc/kvm/powerpc.c
@@ -99,14 +99,7 @@ void kvm_arch_hardware_unsetup(void)
 
 void kvm_arch_check_processor_compat(void *rtn)
 {
-        int r;
-
-        if (strcmp(cur_cpu_spec->platform, "ppc440") == 0)
-                r = 0;
-        else
-                r = -ENOTSUPP;
-
-        *(int *)rtn = r;
+        *(int *)rtn = kvmppc_core_check_processor_compat();
 }
 
 struct kvm *kvm_arch_create_vm(void)
@@ -212,16 +205,14 @@ void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
 
 int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
 {
-        unsigned int priority = exception_priority[BOOKE_INTERRUPT_DECREMENTER];
-
-        return test_bit(priority, &vcpu->arch.pending_exceptions);
+        return kvmppc_core_pending_dec(vcpu);
 }
 
 static void kvmppc_decrementer_func(unsigned long data)
 {
         struct kvm_vcpu *vcpu = (struct kvm_vcpu *)data;
 
-        kvmppc_queue_exception(vcpu, BOOKE_INTERRUPT_DECREMENTER);
+        kvmppc_core_queue_dec(vcpu);
 
         if (waitqueue_active(&vcpu->wq)) {
                 wake_up_interruptible(&vcpu->wq);
@@ -242,96 +233,25 @@ void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
         kvmppc_core_destroy_mmu(vcpu);
 }
 
-/* Note: clearing MSR[DE] just means that the debug interrupt will not be
- * delivered *immediately*. Instead, it simply sets the appropriate DBSR bits.
- * If those DBSR bits are still set when MSR[DE] is re-enabled, the interrupt
- * will be delivered as an "imprecise debug event" (which is indicated by
- * DBSR[IDE].
- */
-static void kvmppc_disable_debug_interrupts(void)
-{
-        mtmsr(mfmsr() & ~MSR_DE);
-}
-
-static void kvmppc_restore_host_debug_state(struct kvm_vcpu *vcpu)
-{
-        kvmppc_disable_debug_interrupts();
-
-        mtspr(SPRN_IAC1, vcpu->arch.host_iac[0]);
-        mtspr(SPRN_IAC2, vcpu->arch.host_iac[1]);
-        mtspr(SPRN_IAC3, vcpu->arch.host_iac[2]);
-        mtspr(SPRN_IAC4, vcpu->arch.host_iac[3]);
-        mtspr(SPRN_DBCR1, vcpu->arch.host_dbcr1);
-        mtspr(SPRN_DBCR2, vcpu->arch.host_dbcr2);
-        mtspr(SPRN_DBCR0, vcpu->arch.host_dbcr0);
-        mtmsr(vcpu->arch.host_msr);
-}
-
-static void kvmppc_load_guest_debug_registers(struct kvm_vcpu *vcpu)
-{
-        struct kvm_guest_debug *dbg = &vcpu->guest_debug;
-        u32 dbcr0 = 0;
-
-        vcpu->arch.host_msr = mfmsr();
-        kvmppc_disable_debug_interrupts();
-
-        /* Save host debug register state. */
-        vcpu->arch.host_iac[0] = mfspr(SPRN_IAC1);
-        vcpu->arch.host_iac[1] = mfspr(SPRN_IAC2);
-        vcpu->arch.host_iac[2] = mfspr(SPRN_IAC3);
-        vcpu->arch.host_iac[3] = mfspr(SPRN_IAC4);
-        vcpu->arch.host_dbcr0 = mfspr(SPRN_DBCR0);
-        vcpu->arch.host_dbcr1 = mfspr(SPRN_DBCR1);
-        vcpu->arch.host_dbcr2 = mfspr(SPRN_DBCR2);
-
-        /* set registers up for guest */
-
-        if (dbg->bp[0]) {
-                mtspr(SPRN_IAC1, dbg->bp[0]);
-                dbcr0 |= DBCR0_IAC1 | DBCR0_IDM;
-        }
-        if (dbg->bp[1]) {
-                mtspr(SPRN_IAC2, dbg->bp[1]);
-                dbcr0 |= DBCR0_IAC2 | DBCR0_IDM;
-        }
-        if (dbg->bp[2]) {
-                mtspr(SPRN_IAC3, dbg->bp[2]);
-                dbcr0 |= DBCR0_IAC3 | DBCR0_IDM;
-        }
-        if (dbg->bp[3]) {
-                mtspr(SPRN_IAC4, dbg->bp[3]);
-                dbcr0 |= DBCR0_IAC4 | DBCR0_IDM;
-        }
-
-        mtspr(SPRN_DBCR0, dbcr0);
-        mtspr(SPRN_DBCR1, 0);
-        mtspr(SPRN_DBCR2, 0);
-}
-
 void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
-        int i;
-
         if (vcpu->guest_debug.enabled)
-                kvmppc_load_guest_debug_registers(vcpu);
+                kvmppc_core_load_guest_debugstate(vcpu);
 
-        /* Mark every guest entry in the shadow TLB entry modified, so that they
-         * will all be reloaded on the next vcpu run (instead of being
-         * demand-faulted). */
-        for (i = 0; i <= tlb_44x_hwater; i++)
-                kvmppc_tlbe_set_modified(vcpu, i);
+        kvmppc_core_vcpu_load(vcpu, cpu);
 }
 
 void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 {
         if (vcpu->guest_debug.enabled)
-                kvmppc_restore_host_debug_state(vcpu);
+                kvmppc_core_load_host_debugstate(vcpu);
 
         /* Don't leave guest TLB entries resident when being de-scheduled. */
         /* XXX It would be nice to differentiate between heavyweight exit and
          * sched_out here, since we could avoid the TLB flush for heavyweight
          * exits. */
         _tlbil_all();
+        kvmppc_core_vcpu_put(vcpu);
 }
 
 int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
@@ -460,7 +380,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
                 vcpu->arch.dcr_needed = 0;
         }
 
-        kvmppc_check_and_deliver_interrupts(vcpu);
+        kvmppc_core_deliver_interrupts(vcpu);
 
         local_irq_disable();
         kvm_guest_enter();
@@ -478,7 +398,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
 
 int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, struct kvm_interrupt *irq)
 {
-        kvmppc_queue_exception(vcpu, BOOKE_INTERRUPT_EXTERNAL);
+        kvmppc_core_queue_external(vcpu, irq);
 
         if (waitqueue_active(&vcpu->wq)) {
                 wake_up_interruptible(&vcpu->wq);