*
*/
+#include <dirent.h>
#include <sys/types.h>
#include <sys/ioctl.h>
#include <sys/mman.h>
#include "kvm_ppc.h"
#include "cpu.h"
#include "device_tree.h"
+#include "hw/sysbus.h"
+#include "hw/spapr.h"
+
+#include "hw/sysbus.h"
+#include "hw/spapr.h"
+#include "hw/spapr_vio.h"
//#define DEBUG_KVM
do { } while (0)
#endif
+#define PROC_DEVTREE_CPU "/proc/device-tree/cpus/"
+
const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
KVM_CAP_LAST_INFO
};
static int cap_interrupt_level = false;
static int cap_segstate;
static int cap_booke_sregs;
+static int cap_ppc_smt;
+static int cap_ppc_rma;
+static int cap_spapr_tce;
/* 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
cap_interrupt_level = kvm_check_extension(s, KVM_CAP_PPC_IRQ_LEVEL);
cap_segstate = kvm_check_extension(s, KVM_CAP_PPC_SEGSTATE);
cap_booke_sregs = kvm_check_extension(s, KVM_CAP_PPC_BOOKE_SREGS);
+ 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);
if (!cap_interrupt_level) {
fprintf(stderr, "KVM: Couldn't find level irq capability. Expect the "
return kvm_vcpu_ioctl(cenv, KVM_SET_SREGS, &sregs);
}
+/* Set up a shared TLB array with KVM */
+static int kvm_booke206_tlb_init(CPUState *env)
+{
+ struct kvm_book3e_206_tlb_params params = {};
+ struct kvm_config_tlb cfg = {};
+ struct kvm_enable_cap encap = {};
+ unsigned int entries = 0;
+ int ret, i;
+
+ if (!kvm_enabled() ||
+ !kvm_check_extension(env->kvm_state, KVM_CAP_SW_TLB)) {
+ return 0;
+ }
+
+ assert(ARRAY_SIZE(params.tlb_sizes) == BOOKE206_MAX_TLBN);
+
+ for (i = 0; i < BOOKE206_MAX_TLBN; i++) {
+ params.tlb_sizes[i] = booke206_tlb_size(env, i);
+ params.tlb_ways[i] = booke206_tlb_ways(env, i);
+ entries += params.tlb_sizes[i];
+ }
+
+ assert(entries == env->nb_tlb);
+ assert(sizeof(struct kvm_book3e_206_tlb_entry) == sizeof(ppcmas_tlb_t));
+
+ env->tlb_dirty = true;
+
+ cfg.array = (uintptr_t)env->tlb.tlbm;
+ cfg.array_len = sizeof(ppcmas_tlb_t) * entries;
+ cfg.params = (uintptr_t)¶ms;
+ cfg.mmu_type = KVM_MMU_FSL_BOOKE_NOHV;
+
+ encap.cap = KVM_CAP_SW_TLB;
+ encap.args[0] = (uintptr_t)&cfg;
+
+ ret = kvm_vcpu_ioctl(env, KVM_ENABLE_CAP, &encap);
+ if (ret < 0) {
+ fprintf(stderr, "%s: couldn't enable KVM_CAP_SW_TLB: %s\n",
+ __func__, strerror(-ret));
+ return ret;
+ }
+
+ env->kvm_sw_tlb = true;
+ return 0;
+}
+
int kvm_arch_init_vcpu(CPUState *cenv)
{
int ret;
idle_timer = qemu_new_timer_ns(vm_clock, kvm_kick_env, cenv);
+ /* Some targets support access to KVM's guest TLB. */
+ switch (cenv->mmu_model) {
+ case POWERPC_MMU_BOOKE206:
+ ret = kvm_booke206_tlb_init(cenv);
+ break;
+ default:
+ break;
+ }
+
return ret;
}
{
}
+static void kvm_sw_tlb_put(CPUState *env)
+{
+ struct kvm_dirty_tlb dirty_tlb;
+ unsigned char *bitmap;
+ int ret;
+
+ if (!env->kvm_sw_tlb) {
+ return;
+ }
+
+ bitmap = g_malloc((env->nb_tlb + 7) / 8);
+ memset(bitmap, 0xFF, (env->nb_tlb + 7) / 8);
+
+ dirty_tlb.bitmap = (uintptr_t)bitmap;
+ dirty_tlb.num_dirty = env->nb_tlb;
+
+ ret = kvm_vcpu_ioctl(env, KVM_DIRTY_TLB, &dirty_tlb);
+ if (ret) {
+ fprintf(stderr, "%s: KVM_DIRTY_TLB: %s\n",
+ __func__, strerror(-ret));
+ }
+
+ g_free(bitmap);
+}
+
int kvm_arch_put_registers(CPUState *env, int level)
{
struct kvm_regs regs;
if (ret < 0)
return ret;
+ if (env->tlb_dirty) {
+ kvm_sw_tlb_put(env);
+ env->tlb_dirty = false;
+ }
+
return ret;
}
dprintf("handle halt\n");
ret = kvmppc_handle_halt(env);
break;
+#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.args);
+ ret = 1;
+ break;
+#endif
default:
fprintf(stderr, "KVM: unknown exit reason %d\n", run->exit_reason);
ret = -1;
return retval;
}
+/* Try to find a device tree node for a CPU with clock-frequency property */
+static int kvmppc_find_cpu_dt(char *buf, int buf_len)
+{
+ struct dirent *dirp;
+ DIR *dp;
+
+ if ((dp = opendir(PROC_DEVTREE_CPU)) == NULL) {
+ printf("Can't open directory " PROC_DEVTREE_CPU "\n");
+ return -1;
+ }
+
+ buf[0] = '\0';
+ while ((dirp = readdir(dp)) != NULL) {
+ FILE *f;
+ snprintf(buf, buf_len, "%s%s/clock-frequency", PROC_DEVTREE_CPU,
+ dirp->d_name);
+ f = fopen(buf, "r");
+ if (f) {
+ snprintf(buf, buf_len, "%s%s", PROC_DEVTREE_CPU, dirp->d_name);
+ fclose(f);
+ break;
+ }
+ buf[0] = '\0';
+ }
+ closedir(dp);
+ if (buf[0] == '\0') {
+ printf("Unknown host!\n");
+ return -1;
+ }
+
+ return 0;
+}
+
+/* Read a CPU node property from the host device tree that's a single
+ * integer (32-bit or 64-bit). Returns 0 if anything goes wrong
+ * (can't find or open the property, or doesn't understand the
+ * format) */
+static uint64_t kvmppc_read_int_cpu_dt(const char *propname)
+{
+ char buf[PATH_MAX];
+ union {
+ uint32_t v32;
+ uint64_t v64;
+ } u;
+ FILE *f;
+ int len;
+
+ if (kvmppc_find_cpu_dt(buf, sizeof(buf))) {
+ return -1;
+ }
+
+ strncat(buf, "/", sizeof(buf) - strlen(buf));
+ strncat(buf, propname, sizeof(buf) - strlen(buf));
+
+ f = fopen(buf, "rb");
+ if (!f) {
+ return -1;
+ }
+
+ len = fread(&u, 1, sizeof(u), f);
+ fclose(f);
+ switch (len) {
+ case 4:
+ /* property is a 32-bit quantity */
+ return be32_to_cpu(u.v32);
+ case 8:
+ return be64_to_cpu(u.v64);
+ }
+
+ return 0;
+}
+
+uint64_t kvmppc_get_clockfreq(void)
+{
+ return kvmppc_read_int_cpu_dt("clock-frequency");
+}
+
+uint32_t kvmppc_get_vmx(void)
+{
+ return kvmppc_read_int_cpu_dt("ibm,vmx");
+}
+
+uint32_t kvmppc_get_dfp(void)
+{
+ return kvmppc_read_int_cpu_dt("ibm,dfp");
+}
+
int kvmppc_get_hypercall(CPUState *env, uint8_t *buf, int buf_len)
{
uint32_t *hc = (uint32_t*)buf;
return 0;
}
+void kvmppc_set_papr(CPUState *env)
+{
+ struct kvm_enable_cap cap = {};
+ struct kvm_one_reg reg = {};
+ struct kvm_sregs sregs = {};
+ int ret;
+
+ cap.cap = KVM_CAP_PPC_PAPR;
+ ret = kvm_vcpu_ioctl(env, KVM_ENABLE_CAP, &cap);
+
+ if (ret) {
+ goto fail;
+ }
+
+ /*
+ * 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_ONE_REG_PPC_HIOR;
+ reg.u.reg64 = env->spr[SPR_HIOR];
+ ret = kvm_vcpu_ioctl(env, KVM_SET_ONE_REG, ®);
+ if (ret) {
+ goto fail;
+ }
+
+ /* 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)
+{
+ return cap_ppc_smt ? cap_ppc_smt : 1;
+}
+
+off_t kvmppc_alloc_rma(const char *name, MemoryRegion *sysmem)
+{
+ void *rma;
+ off_t size;
+ int fd;
+ struct kvm_allocate_rma ret;
+ MemoryRegion *rma_region;
+
+ /* If cap_ppc_rma == 0, contiguous RMA allocation is not supported
+ * if cap_ppc_rma == 1, contiguous RMA allocation is supported, but
+ * not necessary on this hardware
+ * if cap_ppc_rma == 2, contiguous RMA allocation is needed on this hardware
+ *
+ * FIXME: We should allow the user to force contiguous RMA
+ * allocation in the cap_ppc_rma==1 case.
+ */
+ if (cap_ppc_rma < 2) {
+ return 0;
+ }
+
+ fd = kvm_vm_ioctl(kvm_state, KVM_ALLOCATE_RMA, &ret);
+ if (fd < 0) {
+ fprintf(stderr, "KVM: Error on KVM_ALLOCATE_RMA: %s\n",
+ strerror(errno));
+ return -1;
+ }
+
+ size = MIN(ret.rma_size, 256ul << 20);
+
+ rma = mmap(NULL, size, PROT_READ|PROT_WRITE, MAP_SHARED, fd, 0);
+ if (rma == MAP_FAILED) {
+ fprintf(stderr, "KVM: Error mapping RMA: %s\n", strerror(errno));
+ return -1;
+ };
+
+ rma_region = g_new(MemoryRegion, 1);
+ memory_region_init_ram_ptr(rma_region, NULL, name, size, rma);
+ memory_region_add_subregion(sysmem, 0, rma_region);
+
+ return size;
+}
+
+void *kvmppc_create_spapr_tce(uint32_t liobn, uint32_t window_size, int *pfd)
+{
+ struct kvm_create_spapr_tce args = {
+ .liobn = liobn,
+ .window_size = window_size,
+ };
+ long len;
+ int fd;
+ void *table;
+
+ if (!cap_spapr_tce) {
+ return NULL;
+ }
+
+ fd = kvm_vm_ioctl(kvm_state, KVM_CREATE_SPAPR_TCE, &args);
+ if (fd < 0) {
+ return NULL;
+ }
+
+ len = (window_size / SPAPR_VIO_TCE_PAGE_SIZE) * sizeof(VIOsPAPR_RTCE);
+ /* FIXME: round this up to page size */
+
+ table = mmap(NULL, len, PROT_READ, MAP_SHARED, fd, 0);
+ if (table == MAP_FAILED) {
+ close(fd);
+ return NULL;
+ }
+
+ *pfd = fd;
+ return table;
+}
+
+int kvmppc_remove_spapr_tce(void *table, int fd, uint32_t window_size)
+{
+ long len;
+
+ if (fd < 0) {
+ return -1;
+ }
+
+ len = (window_size / SPAPR_VIO_TCE_PAGE_SIZE)*sizeof(VIOsPAPR_RTCE);
+ if ((munmap(table, len) < 0) ||
+ (close(fd) < 0)) {
+ fprintf(stderr, "KVM: Unexpected error removing KVM SPAPR TCE "
+ "table: %s", strerror(errno));
+ /* Leak the table */
+ }
+
+ return 0;
+}
+
+static inline uint32_t mfpvr(void)
+{
+ uint32_t pvr;
+
+ asm ("mfpvr %0"
+ : "=r"(pvr));
+ return pvr;
+}
+
+static void alter_insns(uint64_t *word, uint64_t flags, bool on)
+{
+ if (on) {
+ *word |= flags;
+ } else {
+ *word &= ~flags;
+ }
+}
+
+const ppc_def_t *kvmppc_host_cpu_def(void)
+{
+ uint32_t host_pvr = mfpvr();
+ const ppc_def_t *base_spec;
+ ppc_def_t *spec;
+ uint32_t vmx = kvmppc_get_vmx();
+ uint32_t dfp = kvmppc_get_dfp();
+
+ base_spec = ppc_find_by_pvr(host_pvr);
+
+ spec = g_malloc0(sizeof(*spec));
+ memcpy(spec, base_spec, sizeof(*spec));
+
+ /* Now fix up the spec with information we can query from the host */
+
+ if (vmx != -1) {
+ /* Only override when we know what the host supports */
+ alter_insns(&spec->insns_flags, PPC_ALTIVEC, vmx > 0);
+ alter_insns(&spec->insns_flags2, PPC2_VSX, vmx > 1);
+ }
+ if (dfp != -1) {
+ /* Only override when we know what the host supports */
+ alter_insns(&spec->insns_flags2, PPC2_DFP, dfp);
+ }
+
+ return spec;
+}
+
bool kvm_arch_stop_on_emulation_error(CPUState *env)
{
return true;
projects / qemu.git / blobdiff