aboutsummaryrefslogtreecommitdiff
path: root/target-ppc/kvm.c
diff options
context:
space:
mode:
Diffstat (limited to 'target-ppc/kvm.c')
-rw-r--r--target-ppc/kvm.c281
1 files changed, 186 insertions, 95 deletions
diff --git a/target-ppc/kvm.c b/target-ppc/kvm.c
index 829e180f8b..436ca474ff 100644
--- a/target-ppc/kvm.c
+++ b/target-ppc/kvm.c
@@ -23,13 +23,13 @@
#include <linux/kvm.h>
#include "qemu-common.h"
-#include "qemu-timer.h"
-#include "sysemu.h"
-#include "kvm.h"
+#include "qemu/timer.h"
+#include "sysemu/sysemu.h"
+#include "sysemu/kvm.h"
#include "kvm_ppc.h"
#include "cpu.h"
-#include "cpus.h"
-#include "device_tree.h"
+#include "sysemu/cpus.h"
+#include "sysemu/device_tree.h"
#include "hw/sysbus.h"
#include "hw/spapr.h"
@@ -60,6 +60,7 @@ static int cap_booke_sregs;
static int cap_ppc_smt;
static int cap_ppc_rma;
static int cap_spapr_tce;
+static int cap_hior;
/* XXX We have a race condition where we actually have a level triggered
* interrupt, but the infrastructure can't expose that yet, so the guest
@@ -72,9 +73,11 @@ static int cap_spapr_tce;
*/
static QEMUTimer *idle_timer;
-static void kvm_kick_env(void *env)
+static void kvm_kick_cpu(void *opaque)
{
- qemu_cpu_kick(env);
+ PowerPCCPU *cpu = opaque;
+
+ qemu_cpu_kick(CPU(cpu));
}
int kvm_arch_init(KVMState *s)
@@ -86,6 +89,7 @@ int kvm_arch_init(KVMState *s)
cap_ppc_smt = kvm_check_extension(s, KVM_CAP_PPC_SMT);
cap_ppc_rma = kvm_check_extension(s, KVM_CAP_PPC_RMA);
cap_spapr_tce = kvm_check_extension(s, KVM_CAP_SPAPR_TCE);
+ cap_hior = kvm_check_extension(s, KVM_CAP_PPC_HIOR);
if (!cap_interrupt_level) {
fprintf(stderr, "KVM: Couldn't find level irq capability. Expect the "
@@ -95,8 +99,10 @@ int kvm_arch_init(KVMState *s)
return 0;
}
-static int kvm_arch_sync_sregs(CPUPPCState *cenv)
+static int kvm_arch_sync_sregs(PowerPCCPU *cpu)
{
+ CPUPPCState *cenv = &cpu->env;
+ CPUState *cs = CPU(cpu);
struct kvm_sregs sregs;
int ret;
@@ -113,18 +119,20 @@ static int kvm_arch_sync_sregs(CPUPPCState *cenv)
}
}
- ret = kvm_vcpu_ioctl(cenv, KVM_GET_SREGS, &sregs);
+ ret = kvm_vcpu_ioctl(cs, KVM_GET_SREGS, &sregs);
if (ret) {
return ret;
}
sregs.pvr = cenv->spr[SPR_PVR];
- return kvm_vcpu_ioctl(cenv, KVM_SET_SREGS, &sregs);
+ return kvm_vcpu_ioctl(cs, KVM_SET_SREGS, &sregs);
}
/* Set up a shared TLB array with KVM */
-static int kvm_booke206_tlb_init(CPUPPCState *env)
+static int kvm_booke206_tlb_init(PowerPCCPU *cpu)
{
+ CPUPPCState *env = &cpu->env;
+ CPUState *cs = CPU(cpu);
struct kvm_book3e_206_tlb_params params = {};
struct kvm_config_tlb cfg = {};
struct kvm_enable_cap encap = {};
@@ -132,7 +140,7 @@ static int kvm_booke206_tlb_init(CPUPPCState *env)
int ret, i;
if (!kvm_enabled() ||
- !kvm_check_extension(env->kvm_state, KVM_CAP_SW_TLB)) {
+ !kvm_check_extension(cs->kvm_state, KVM_CAP_SW_TLB)) {
return 0;
}
@@ -157,7 +165,7 @@ static int kvm_booke206_tlb_init(CPUPPCState *env)
encap.cap = KVM_CAP_SW_TLB;
encap.args[0] = (uintptr_t)&cfg;
- ret = kvm_vcpu_ioctl(env, KVM_ENABLE_CAP, &encap);
+ ret = kvm_vcpu_ioctl(cs, KVM_ENABLE_CAP, &encap);
if (ret < 0) {
fprintf(stderr, "%s: couldn't enable KVM_CAP_SW_TLB: %s\n",
__func__, strerror(-ret));
@@ -170,9 +178,12 @@ static int kvm_booke206_tlb_init(CPUPPCState *env)
#if defined(TARGET_PPC64)
-static void kvm_get_fallback_smmu_info(CPUPPCState *env,
+static void kvm_get_fallback_smmu_info(PowerPCCPU *cpu,
struct kvm_ppc_smmu_info *info)
{
+ CPUPPCState *env = &cpu->env;
+ CPUState *cs = CPU(cpu);
+
memset(info, 0, sizeof(*info));
/* We don't have the new KVM_PPC_GET_SMMU_INFO ioctl, so
@@ -198,7 +209,7 @@ static void kvm_get_fallback_smmu_info(CPUPPCState *env,
* implements KVM_CAP_PPC_GET_SMMU_INFO and thus doesn't hit
* this fallback.
*/
- if (kvm_check_extension(env->kvm_state, KVM_CAP_PPC_GET_PVINFO)) {
+ if (kvm_check_extension(cs->kvm_state, KVM_CAP_PPC_GET_PVINFO)) {
/* No flags */
info->flags = 0;
info->slb_size = 64;
@@ -254,18 +265,19 @@ static void kvm_get_fallback_smmu_info(CPUPPCState *env,
}
}
-static void kvm_get_smmu_info(CPUPPCState *env, struct kvm_ppc_smmu_info *info)
+static void kvm_get_smmu_info(PowerPCCPU *cpu, struct kvm_ppc_smmu_info *info)
{
+ CPUState *cs = CPU(cpu);
int ret;
- if (kvm_check_extension(env->kvm_state, KVM_CAP_PPC_GET_SMMU_INFO)) {
- ret = kvm_vm_ioctl(env->kvm_state, KVM_PPC_GET_SMMU_INFO, info);
+ if (kvm_check_extension(cs->kvm_state, KVM_CAP_PPC_GET_SMMU_INFO)) {
+ ret = kvm_vm_ioctl(cs->kvm_state, KVM_PPC_GET_SMMU_INFO, info);
if (ret == 0) {
return;
}
}
- kvm_get_fallback_smmu_info(env, info);
+ kvm_get_fallback_smmu_info(cpu, info);
}
static long getrampagesize(void)
@@ -308,10 +320,11 @@ static bool kvm_valid_page_size(uint32_t flags, long rampgsize, uint32_t shift)
return (1ul << shift) <= rampgsize;
}
-static void kvm_fixup_page_sizes(CPUPPCState *env)
+static void kvm_fixup_page_sizes(PowerPCCPU *cpu)
{
static struct kvm_ppc_smmu_info smmu_info;
static bool has_smmu_info;
+ CPUPPCState *env = &cpu->env;
long rampagesize;
int iq, ik, jq, jk;
@@ -322,7 +335,7 @@ static void kvm_fixup_page_sizes(CPUPPCState *env)
/* Collect MMU info from kernel if not already */
if (!has_smmu_info) {
- kvm_get_smmu_info(env, &smmu_info);
+ kvm_get_smmu_info(cpu, &smmu_info);
has_smmu_info = true;
}
@@ -365,31 +378,33 @@ static void kvm_fixup_page_sizes(CPUPPCState *env)
}
#else /* defined (TARGET_PPC64) */
-static inline void kvm_fixup_page_sizes(CPUPPCState *env)
+static inline void kvm_fixup_page_sizes(PowerPCCPU *cpu)
{
}
#endif /* !defined (TARGET_PPC64) */
-int kvm_arch_init_vcpu(CPUPPCState *cenv)
+int kvm_arch_init_vcpu(CPUState *cs)
{
+ PowerPCCPU *cpu = POWERPC_CPU(cs);
+ CPUPPCState *cenv = &cpu->env;
int ret;
/* Gather server mmu info from KVM and update the CPU state */
- kvm_fixup_page_sizes(cenv);
+ kvm_fixup_page_sizes(cpu);
/* Synchronize sregs with kvm */
- ret = kvm_arch_sync_sregs(cenv);
+ ret = kvm_arch_sync_sregs(cpu);
if (ret) {
return ret;
}
- idle_timer = qemu_new_timer_ns(vm_clock, kvm_kick_env, cenv);
+ idle_timer = qemu_new_timer_ns(vm_clock, kvm_kick_cpu, cpu);
/* Some targets support access to KVM's guest TLB. */
switch (cenv->mmu_model) {
case POWERPC_MMU_BOOKE206:
- ret = kvm_booke206_tlb_init(cenv);
+ ret = kvm_booke206_tlb_init(cpu);
break;
default:
break;
@@ -398,12 +413,14 @@ int kvm_arch_init_vcpu(CPUPPCState *cenv)
return ret;
}
-void kvm_arch_reset_vcpu(CPUPPCState *env)
+void kvm_arch_reset_vcpu(CPUState *cpu)
{
}
-static void kvm_sw_tlb_put(CPUPPCState *env)
+static void kvm_sw_tlb_put(PowerPCCPU *cpu)
{
+ CPUPPCState *env = &cpu->env;
+ CPUState *cs = CPU(cpu);
struct kvm_dirty_tlb dirty_tlb;
unsigned char *bitmap;
int ret;
@@ -418,7 +435,7 @@ static void kvm_sw_tlb_put(CPUPPCState *env)
dirty_tlb.bitmap = (uintptr_t)bitmap;
dirty_tlb.num_dirty = env->nb_tlb;
- ret = kvm_vcpu_ioctl(env, KVM_DIRTY_TLB, &dirty_tlb);
+ ret = kvm_vcpu_ioctl(cs, KVM_DIRTY_TLB, &dirty_tlb);
if (ret) {
fprintf(stderr, "%s: KVM_DIRTY_TLB: %s\n",
__func__, strerror(-ret));
@@ -427,15 +444,18 @@ static void kvm_sw_tlb_put(CPUPPCState *env)
g_free(bitmap);
}
-int kvm_arch_put_registers(CPUPPCState *env, int level)
+int kvm_arch_put_registers(CPUState *cs, int level)
{
+ PowerPCCPU *cpu = POWERPC_CPU(cs);
+ CPUPPCState *env = &cpu->env;
struct kvm_regs regs;
int ret;
int i;
- ret = kvm_vcpu_ioctl(env, KVM_GET_REGS, &regs);
- if (ret < 0)
+ ret = kvm_vcpu_ioctl(cs, KVM_GET_REGS, &regs);
+ if (ret < 0) {
return ret;
+ }
regs.ctr = env->ctr;
regs.lr = env->lr;
@@ -460,26 +480,76 @@ int kvm_arch_put_registers(CPUPPCState *env, int level)
for (i = 0;i < 32; i++)
regs.gpr[i] = env->gpr[i];
- ret = kvm_vcpu_ioctl(env, KVM_SET_REGS, &regs);
+ ret = kvm_vcpu_ioctl(cs, KVM_SET_REGS, &regs);
if (ret < 0)
return ret;
if (env->tlb_dirty) {
- kvm_sw_tlb_put(env);
+ kvm_sw_tlb_put(cpu);
env->tlb_dirty = false;
}
+ if (cap_segstate && (level >= KVM_PUT_RESET_STATE)) {
+ struct kvm_sregs sregs;
+
+ sregs.pvr = env->spr[SPR_PVR];
+
+ sregs.u.s.sdr1 = env->spr[SPR_SDR1];
+
+ /* Sync SLB */
+#ifdef TARGET_PPC64
+ for (i = 0; i < 64; i++) {
+ sregs.u.s.ppc64.slb[i].slbe = env->slb[i].esid;
+ sregs.u.s.ppc64.slb[i].slbv = env->slb[i].vsid;
+ }
+#endif
+
+ /* Sync SRs */
+ for (i = 0; i < 16; i++) {
+ sregs.u.s.ppc32.sr[i] = env->sr[i];
+ }
+
+ /* Sync BATs */
+ for (i = 0; i < 8; i++) {
+ /* Beware. We have to swap upper and lower bits here */
+ sregs.u.s.ppc32.dbat[i] = ((uint64_t)env->DBAT[0][i] << 32)
+ | env->DBAT[1][i];
+ sregs.u.s.ppc32.ibat[i] = ((uint64_t)env->IBAT[0][i] << 32)
+ | env->IBAT[1][i];
+ }
+
+ ret = kvm_vcpu_ioctl(cs, KVM_SET_SREGS, &sregs);
+ if (ret) {
+ return ret;
+ }
+ }
+
+ if (cap_hior && (level >= KVM_PUT_RESET_STATE)) {
+ uint64_t hior = env->spr[SPR_HIOR];
+ struct kvm_one_reg reg = {
+ .id = KVM_REG_PPC_HIOR,
+ .addr = (uintptr_t) &hior,
+ };
+
+ ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &reg);
+ if (ret) {
+ return ret;
+ }
+ }
+
return ret;
}
-int kvm_arch_get_registers(CPUPPCState *env)
+int kvm_arch_get_registers(CPUState *cs)
{
+ PowerPCCPU *cpu = POWERPC_CPU(cs);
+ CPUPPCState *env = &cpu->env;
struct kvm_regs regs;
struct kvm_sregs sregs;
uint32_t cr;
int i, ret;
- ret = kvm_vcpu_ioctl(env, KVM_GET_REGS, &regs);
+ ret = kvm_vcpu_ioctl(cs, KVM_GET_REGS, &regs);
if (ret < 0)
return ret;
@@ -513,7 +583,7 @@ int kvm_arch_get_registers(CPUPPCState *env)
env->gpr[i] = regs.gpr[i];
if (cap_booke_sregs) {
- ret = kvm_vcpu_ioctl(env, KVM_GET_SREGS, &sregs);
+ ret = kvm_vcpu_ioctl(cs, KVM_GET_SREGS, &sregs);
if (ret < 0) {
return ret;
}
@@ -617,7 +687,7 @@ int kvm_arch_get_registers(CPUPPCState *env)
}
if (cap_segstate) {
- ret = kvm_vcpu_ioctl(env, KVM_GET_SREGS, &sregs);
+ ret = kvm_vcpu_ioctl(cs, KVM_GET_SREGS, &sregs);
if (ret < 0) {
return ret;
}
@@ -649,7 +719,7 @@ int kvm_arch_get_registers(CPUPPCState *env)
return 0;
}
-int kvmppc_set_interrupt(CPUPPCState *env, int irq, int level)
+int kvmppc_set_interrupt(PowerPCCPU *cpu, int irq, int level)
{
unsigned virq = level ? KVM_INTERRUPT_SET_LEVEL : KVM_INTERRUPT_UNSET;
@@ -661,7 +731,7 @@ int kvmppc_set_interrupt(CPUPPCState *env, int irq, int level)
return 0;
}
- kvm_vcpu_ioctl(env, KVM_INTERRUPT, &virq);
+ kvm_vcpu_ioctl(CPU(cpu), KVM_INTERRUPT, &virq);
return 0;
}
@@ -674,8 +744,10 @@ int kvmppc_set_interrupt(CPUPPCState *env, int irq, int level)
#define PPC_INPUT_INT PPC6xx_INPUT_INT
#endif
-void kvm_arch_pre_run(CPUPPCState *env, struct kvm_run *run)
+void kvm_arch_pre_run(CPUState *cs, struct kvm_run *run)
{
+ PowerPCCPU *cpu = POWERPC_CPU(cs);
+ CPUPPCState *env = &cpu->env;
int r;
unsigned irq;
@@ -693,7 +765,7 @@ void kvm_arch_pre_run(CPUPPCState *env, struct kvm_run *run)
irq = KVM_INTERRUPT_SET;
dprintf("injected interrupt %d\n", irq);
- r = kvm_vcpu_ioctl(env, KVM_INTERRUPT, &irq);
+ r = kvm_vcpu_ioctl(cs, KVM_INTERRUPT, &irq);
if (r < 0)
printf("cpu %d fail inject %x\n", env->cpu_index, irq);
@@ -707,13 +779,14 @@ void kvm_arch_pre_run(CPUPPCState *env, struct kvm_run *run)
* anyways, so we will get a chance to deliver the rest. */
}
-void kvm_arch_post_run(CPUPPCState *env, struct kvm_run *run)
+void kvm_arch_post_run(CPUState *cpu, struct kvm_run *run)
{
}
-int kvm_arch_process_async_events(CPUPPCState *env)
+int kvm_arch_process_async_events(CPUState *cs)
{
- return env->halted;
+ PowerPCCPU *cpu = POWERPC_CPU(cs);
+ return cpu->env.halted;
}
static int kvmppc_handle_halt(CPUPPCState *env)
@@ -743,8 +816,10 @@ static int kvmppc_handle_dcr_write(CPUPPCState *env, uint32_t dcrn, uint32_t dat
return 0;
}
-int kvm_arch_handle_exit(CPUPPCState *env, struct kvm_run *run)
+int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
{
+ PowerPCCPU *cpu = POWERPC_CPU(cs);
+ CPUPPCState *env = &cpu->env;
int ret;
switch (run->exit_reason) {
@@ -764,9 +839,10 @@ int kvm_arch_handle_exit(CPUPPCState *env, struct kvm_run *run)
#ifdef CONFIG_PSERIES
case KVM_EXIT_PAPR_HCALL:
dprintf("handle PAPR hypercall\n");
- run->papr_hcall.ret = spapr_hypercall(env, run->papr_hcall.nr,
+ run->papr_hcall.ret = spapr_hypercall(cpu,
+ run->papr_hcall.nr,
run->papr_hcall.args);
- ret = 1;
+ ret = 0;
break;
#endif
default:
@@ -795,7 +871,7 @@ static int read_cpuinfo(const char *field, char *value, int len)
break;
}
if (!strncmp(line, field, field_len)) {
- strncpy(value, line, len);
+ pstrcpy(value, len, line);
ret = 0;
break;
}
@@ -915,12 +991,14 @@ uint32_t kvmppc_get_dfp(void)
int kvmppc_get_hypercall(CPUPPCState *env, uint8_t *buf, int buf_len)
{
+ PowerPCCPU *cpu = ppc_env_get_cpu(env);
+ CPUState *cs = CPU(cpu);
uint32_t *hc = (uint32_t*)buf;
struct kvm_ppc_pvinfo pvinfo;
- if (kvm_check_extension(env->kvm_state, KVM_CAP_PPC_GET_PVINFO) &&
- !kvm_vm_ioctl(env->kvm_state, KVM_PPC_GET_PVINFO, &pvinfo)) {
+ if (kvm_check_extension(cs->kvm_state, KVM_CAP_PPC_GET_PVINFO) &&
+ !kvm_vm_ioctl(cs->kvm_state, KVM_PPC_GET_PVINFO, &pvinfo)) {
memcpy(buf, pvinfo.hcall, buf_len);
return 0;
@@ -943,55 +1021,19 @@ int kvmppc_get_hypercall(CPUPPCState *env, uint8_t *buf, int buf_len)
return 0;
}
-void kvmppc_set_papr(CPUPPCState *env)
+void kvmppc_set_papr(PowerPCCPU *cpu)
{
+ CPUPPCState *env = &cpu->env;
+ CPUState *cs = CPU(cpu);
struct kvm_enable_cap cap = {};
- struct kvm_one_reg reg = {};
- struct kvm_sregs sregs = {};
int ret;
- uint64_t hior = env->spr[SPR_HIOR];
cap.cap = KVM_CAP_PPC_PAPR;
- ret = kvm_vcpu_ioctl(env, KVM_ENABLE_CAP, &cap);
+ ret = kvm_vcpu_ioctl(cs, KVM_ENABLE_CAP, &cap);
if (ret) {
- goto fail;
+ cpu_abort(env, "This KVM version does not support PAPR\n");
}
-
- /*
- * XXX We set HIOR here. It really should be a qdev property of
- * the CPU node, but we don't have CPUs converted to qdev yet.
- *
- * Once we have qdev CPUs, move HIOR to a qdev property and
- * remove this chunk.
- */
- reg.id = KVM_REG_PPC_HIOR;
- reg.addr = (uintptr_t)&hior;
- ret = kvm_vcpu_ioctl(env, KVM_SET_ONE_REG, &reg);
- if (ret) {
- fprintf(stderr, "Couldn't set HIOR. Maybe you're running an old \n"
- "kernel with support for HV KVM but no PAPR PR \n"
- "KVM in which case things will work. If they don't \n"
- "please update your host kernel!\n");
- }
-
- /* Set SDR1 so kernel space finds the HTAB */
- ret = kvm_vcpu_ioctl(env, KVM_GET_SREGS, &sregs);
- if (ret) {
- goto fail;
- }
-
- sregs.u.s.sdr1 = env->spr[SPR_SDR1];
-
- ret = kvm_vcpu_ioctl(env, KVM_SET_SREGS, &sregs);
- if (ret) {
- goto fail;
- }
-
- return;
-
-fail:
- cpu_abort(env, "This KVM version does not support PAPR\n");
}
int kvmppc_smt_threads(void)
@@ -999,6 +1041,7 @@ int kvmppc_smt_threads(void)
return cap_ppc_smt ? cap_ppc_smt : 1;
}
+#ifdef TARGET_PPC64
off_t kvmppc_alloc_rma(const char *name, MemoryRegion *sysmem)
{
void *rma;
@@ -1042,6 +1085,16 @@ off_t kvmppc_alloc_rma(const char *name, MemoryRegion *sysmem)
return size;
}
+uint64_t kvmppc_rma_size(uint64_t current_size, unsigned int hash_shift)
+{
+ if (cap_ppc_rma >= 2) {
+ return current_size;
+ }
+ return MIN(current_size,
+ getrampagesize() << (hash_shift - 7));
+}
+#endif
+
void *kvmppc_create_spapr_tce(uint32_t liobn, uint32_t window_size, int *pfd)
{
struct kvm_create_spapr_tce args = {
@@ -1101,6 +1154,44 @@ int kvmppc_remove_spapr_tce(void *table, int fd, uint32_t window_size)
return 0;
}
+int kvmppc_reset_htab(int shift_hint)
+{
+ uint32_t shift = shift_hint;
+
+ if (!kvm_enabled()) {
+ /* Full emulation, tell caller to allocate htab itself */
+ return 0;
+ }
+ if (kvm_check_extension(kvm_state, KVM_CAP_PPC_ALLOC_HTAB)) {
+ int ret;
+ ret = kvm_vm_ioctl(kvm_state, KVM_PPC_ALLOCATE_HTAB, &shift);
+ if (ret == -ENOTTY) {
+ /* At least some versions of PR KVM advertise the
+ * capability, but don't implement the ioctl(). Oops.
+ * Return 0 so that we allocate the htab in qemu, as is
+ * correct for PR. */
+ return 0;
+ } else if (ret < 0) {
+ return ret;
+ }
+ return shift;
+ }
+
+ /* We have a kernel that predates the htab reset calls. For PR
+ * KVM, we need to allocate the htab ourselves, for an HV KVM of
+ * this era, it has allocated a 16MB fixed size hash table
+ * already. Kernels of this era have the GET_PVINFO capability
+ * only on PR, so we use this hack to determine the right
+ * answer */
+ if (kvm_check_extension(kvm_state, KVM_CAP_PPC_GET_PVINFO)) {
+ /* PR - tell caller to allocate htab */
+ return 0;
+ } else {
+ /* HV - assume 16MB kernel allocated htab */
+ return 24;
+ }
+}
+
static inline uint32_t mfpvr(void)
{
uint32_t pvr;
@@ -1160,12 +1251,12 @@ int kvmppc_fixup_cpu(CPUPPCState *env)
}
-bool kvm_arch_stop_on_emulation_error(CPUPPCState *env)
+bool kvm_arch_stop_on_emulation_error(CPUState *cpu)
{
return true;
}
-int kvm_arch_on_sigbus_vcpu(CPUPPCState *env, int code, void *addr)
+int kvm_arch_on_sigbus_vcpu(CPUState *cpu, int code, void *addr)
{
return 1;
}
generated by cgit v1.2.3 (git 2.26.2) at 2024-12-04 16:58:20 +0000