qemu-e2k/target/ppc/kvm.c

2795 lines
77 KiB
C
Raw Normal View History

/*
* PowerPC implementation of KVM hooks
*
* Copyright IBM Corp. 2007
* Copyright (C) 2011 Freescale Semiconductor, Inc.
*
* Authors:
* Jerone Young <jyoung5@us.ibm.com>
* Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
* Hollis Blanchard <hollisb@us.ibm.com>
*
* This work is licensed under the terms of the GNU GPL, version 2 or later.
* See the COPYING file in the top-level directory.
*
*/
#include "qemu/osdep.h"
#include <dirent.h>
#include <sys/ioctl.h>
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
#include <sys/vfs.h>
#include <linux/kvm.h>
#include "qemu-common.h"
#include "qapi/error.h"
#include "qemu/error-report.h"
#include "cpu.h"
#include "cpu-models.h"
#include "qemu/timer.h"
#include "sysemu/sysemu.h"
#include "sysemu/hw_accel.h"
#include "kvm_ppc.h"
#include "sysemu/cpus.h"
#include "sysemu/device_tree.h"
#include "mmu-hash64.h"
#include "hw/sysbus.h"
#include "hw/ppc/spapr.h"
#include "hw/ppc/spapr_vio.h"
#include "hw/ppc/spapr_cpu_core.h"
#include "hw/ppc/ppc.h"
#include "sysemu/watchdog.h"
#include "trace.h"
#include "exec/gdbstub.h"
#include "exec/memattrs.h"
#include "exec/ram_addr.h"
#include "sysemu/hostmem.h"
#include "qemu/cutils.h"
#include "qemu/mmap-alloc.h"
#include "elf.h"
#include "sysemu/kvm_int.h"
//#define DEBUG_KVM
#ifdef DEBUG_KVM
#define DPRINTF(fmt, ...) \
do { fprintf(stderr, fmt, ## __VA_ARGS__); } while (0)
#else
#define DPRINTF(fmt, ...) \
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_unset = false;
static int cap_interrupt_level = false;
static int cap_segstate;
static int cap_booke_sregs;
static int cap_ppc_smt;
static int cap_ppc_smt_possible;
static int cap_spapr_tce;
static int cap_spapr_tce_64;
static int cap_spapr_multitce;
spapr_iommu: Make in-kernel TCE table optional POWER KVM supports an KVM_CAP_SPAPR_TCE capability which allows allocating TCE tables in the host kernel memory and handle H_PUT_TCE requests targeted to specific LIOBN (logical bus number) right in the host without switching to QEMU. At the moment this is used for emulated devices only and the handler only puts TCE to the table. If the in-kernel H_PUT_TCE handler finds a LIOBN and corresponding table, it will put a TCE to the table and complete hypercall execution. The user space will not be notified. Upcoming VFIO support is going to use the same sPAPRTCETable device class so KVM_CAP_SPAPR_TCE is going to be used as well. That means that TCE tables for VFIO are going to be allocated in the host as well. However VFIO operates with real IOMMU tables and simple copying of a TCE to the real hardware TCE table will not work as guest physical to host physical address translation is requited. So until the host kernel gets VFIO support for H_PUT_TCE, we better not to register VFIO's TCE in the host. This adds a place holder for KVM_CAP_SPAPR_TCE_VFIO capability. It is not in upstream yet and being discussed so now it is always false which means that in-kernel VFIO acceleration is not supported. This adds a bool @vfio_accel flag to the sPAPRTCETable device telling that sPAPRTCETable should not try allocating TCE table in the host kernel for VFIO. The flag is false now as at the moment there is no VFIO. This adds an vfio_accel parameter to spapr_tce_new_table(), the semantic is the same. Since there is only emulated PCI and VIO now, the flag is set to false. Upcoming VFIO support will set it to true. This is a preparation patch so no change in behaviour is expected Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> Signed-off-by: Alexander Graf <agraf@suse.de>
2014-06-10 07:39:21 +02:00
static int cap_spapr_vfio;
ppc: Make kvm_arch_put_registers() put *all* the registers At least when invoked with high enough 'level' arguments, kvm_arch_put_registers() is supposed to copy essentially all the cpu state as encoded in qemu's internal structures into the kvm state. Currently the ppc version does not do this - it never calls KVM_SET_SREGS, for example, and therefore never sets the SDR1 and various other important though rarely changed registers. Instead, the code paths which need to set these registers need to explicitly make (conditional) kvm calls which transfer the changes to kvm. This breaks the usual model of handling state updates in qemu, where code just changes the internal model and has it flushed out to kvm automatically at some later point. This patch fixes this for Book S ppc CPUs by adding a suitable call to KVM_SET_SREGS and als to KVM_SET_ONE_REG to set the HIOR (the only register that is set with that call so far). This lets us remove the hacks to explicitly set these registers from the kvmppc_set_papr() function. The problem still exists for Book E CPUs (which use a different version of the kvm_sregs structure). But fixing that has some complications of its own so can be left to another day. Lkewise, there is still some ugly code for setting the PVR through special calls to SET_SREGS which is left in for now. The PVR needs to be set especially early because it can affect what other features are available on the CPU, so I need to do more thinking to see if it can be integrated into the normal paths or not. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-09-12 18:57:09 +02:00
static int cap_hior;
static int cap_one_reg;
static int cap_epr;
static int cap_ppc_watchdog;
static int cap_papr;
static int cap_htab_fd;
static int cap_fixup_hcalls;
static int cap_htm; /* Hardware transactional memory support */
static int cap_mmu_radix;
static int cap_mmu_hash_v3;
static int cap_resize_hpt;
target/ppc: 'PVR != host PVR' in KVM_SET_SREGS workaround Commit d5fc133eed ("ppc: Rework CPU compatibility testing across migration") changed the way cpu_post_load behaves with the PVR setting, causing an unexpected bug in KVM-HV migrations between hosts that are compatible (POWER8 and POWER8E, for example). Even with pvr_match() returning true, the guest freezes right after cpu_post_load. The reason is that the guest kernel can't handle a different PVR value other that the running host in KVM_SET_SREGS. In [1] it was discussed the possibility of a new KVM capability that would indicate that the guest kernel can handle a different PVR in KVM_SET_SREGS. Even if such feature is implemented, there is still the problem with older kernels that will not have this capability and will fail to migrate. This patch implements a workaround for that scenario. If running with KVM, check if the guest kernel does not have the capability (named here as 'cap_ppc_pvr_compat'). If it doesn't, calls kvmppc_is_pr() to see if the guest is running in KVM-HV. If all this happens, set env->spr[SPR_PVR] to the same value as the current host PVR. This ensures that we allow migrations with 'close enough' PVRs to still work in KVM-HV but also makes the code ready for this new KVM capability when it is done. A new function called 'kvmppc_pvr_workaround_required' was created to encapsulate the conditions said above and to avoid calling too many kvm.c internals inside cpu_post_load. [1] https://lists.gnu.org/archive/html/qemu-ppc/2017-06/msg00503.html Signed-off-by: Daniel Henrique Barboza <danielhb@linux.vnet.ibm.com> [dwg: Fix for the case of using TCG on a PPC host] Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-08-09 22:43:46 +02:00
static int cap_ppc_pvr_compat;
static int cap_ppc_safe_cache;
static int cap_ppc_safe_bounds_check;
static int cap_ppc_safe_indirect_branch;
static uint32_t debug_inst_opcode;
/* 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
* takes but ignores it, goes to sleep and never gets notified that there's
* still an interrupt pending.
*
* As a quick workaround, let's just wake up again 20 ms after we injected
* an interrupt. That way we can assure that we're always reinjecting
* interrupts in case the guest swallowed them.
*/
static QEMUTimer *idle_timer;
static void kvm_kick_cpu(void *opaque)
{
PowerPCCPU *cpu = opaque;
qemu_cpu_kick(CPU(cpu));
}
/* Check whether we are running with KVM-PR (instead of KVM-HV). This
* should only be used for fallback tests - generally we should use
* explicit capabilities for the features we want, rather than
* assuming what is/isn't available depending on the KVM variant. */
static bool kvmppc_is_pr(KVMState *ks)
{
/* Assume KVM-PR if the GET_PVINFO capability is available */
return kvm_vm_check_extension(ks, KVM_CAP_PPC_GET_PVINFO) != 0;
}
static int kvm_ppc_register_host_cpu_type(MachineState *ms);
static void kvmppc_get_cpu_characteristics(KVMState *s);
int kvm_arch_init(MachineState *ms, KVMState *s)
{
cap_interrupt_unset = kvm_check_extension(s, KVM_CAP_PPC_UNSET_IRQ);
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_possible = kvm_vm_check_extension(s, KVM_CAP_PPC_SMT_POSSIBLE);
cap_spapr_tce = kvm_check_extension(s, KVM_CAP_SPAPR_TCE);
cap_spapr_tce_64 = kvm_check_extension(s, KVM_CAP_SPAPR_TCE_64);
cap_spapr_multitce = kvm_check_extension(s, KVM_CAP_SPAPR_MULTITCE);
cap_spapr_vfio = kvm_vm_check_extension(s, KVM_CAP_SPAPR_TCE_VFIO);
cap_one_reg = kvm_check_extension(s, KVM_CAP_ONE_REG);
ppc: Make kvm_arch_put_registers() put *all* the registers At least when invoked with high enough 'level' arguments, kvm_arch_put_registers() is supposed to copy essentially all the cpu state as encoded in qemu's internal structures into the kvm state. Currently the ppc version does not do this - it never calls KVM_SET_SREGS, for example, and therefore never sets the SDR1 and various other important though rarely changed registers. Instead, the code paths which need to set these registers need to explicitly make (conditional) kvm calls which transfer the changes to kvm. This breaks the usual model of handling state updates in qemu, where code just changes the internal model and has it flushed out to kvm automatically at some later point. This patch fixes this for Book S ppc CPUs by adding a suitable call to KVM_SET_SREGS and als to KVM_SET_ONE_REG to set the HIOR (the only register that is set with that call so far). This lets us remove the hacks to explicitly set these registers from the kvmppc_set_papr() function. The problem still exists for Book E CPUs (which use a different version of the kvm_sregs structure). But fixing that has some complications of its own so can be left to another day. Lkewise, there is still some ugly code for setting the PVR through special calls to SET_SREGS which is left in for now. The PVR needs to be set especially early because it can affect what other features are available on the CPU, so I need to do more thinking to see if it can be integrated into the normal paths or not. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-09-12 18:57:09 +02:00
cap_hior = kvm_check_extension(s, KVM_CAP_PPC_HIOR);
cap_epr = kvm_check_extension(s, KVM_CAP_PPC_EPR);
cap_ppc_watchdog = kvm_check_extension(s, KVM_CAP_PPC_BOOKE_WATCHDOG);
/* Note: we don't set cap_papr here, because this capability is
* only activated after this by kvmppc_set_papr() */
cap_htab_fd = kvm_vm_check_extension(s, KVM_CAP_PPC_HTAB_FD);
cap_fixup_hcalls = kvm_check_extension(s, KVM_CAP_PPC_FIXUP_HCALL);
cap_ppc_smt = kvm_vm_check_extension(s, KVM_CAP_PPC_SMT);
cap_htm = kvm_vm_check_extension(s, KVM_CAP_PPC_HTM);
cap_mmu_radix = kvm_vm_check_extension(s, KVM_CAP_PPC_MMU_RADIX);
cap_mmu_hash_v3 = kvm_vm_check_extension(s, KVM_CAP_PPC_MMU_HASH_V3);
cap_resize_hpt = kvm_vm_check_extension(s, KVM_CAP_SPAPR_RESIZE_HPT);
kvmppc_get_cpu_characteristics(s);
target/ppc: 'PVR != host PVR' in KVM_SET_SREGS workaround Commit d5fc133eed ("ppc: Rework CPU compatibility testing across migration") changed the way cpu_post_load behaves with the PVR setting, causing an unexpected bug in KVM-HV migrations between hosts that are compatible (POWER8 and POWER8E, for example). Even with pvr_match() returning true, the guest freezes right after cpu_post_load. The reason is that the guest kernel can't handle a different PVR value other that the running host in KVM_SET_SREGS. In [1] it was discussed the possibility of a new KVM capability that would indicate that the guest kernel can handle a different PVR in KVM_SET_SREGS. Even if such feature is implemented, there is still the problem with older kernels that will not have this capability and will fail to migrate. This patch implements a workaround for that scenario. If running with KVM, check if the guest kernel does not have the capability (named here as 'cap_ppc_pvr_compat'). If it doesn't, calls kvmppc_is_pr() to see if the guest is running in KVM-HV. If all this happens, set env->spr[SPR_PVR] to the same value as the current host PVR. This ensures that we allow migrations with 'close enough' PVRs to still work in KVM-HV but also makes the code ready for this new KVM capability when it is done. A new function called 'kvmppc_pvr_workaround_required' was created to encapsulate the conditions said above and to avoid calling too many kvm.c internals inside cpu_post_load. [1] https://lists.gnu.org/archive/html/qemu-ppc/2017-06/msg00503.html Signed-off-by: Daniel Henrique Barboza <danielhb@linux.vnet.ibm.com> [dwg: Fix for the case of using TCG on a PPC host] Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-08-09 22:43:46 +02:00
/*
* Note: setting it to false because there is not such capability
* in KVM at this moment.
*
* TODO: call kvm_vm_check_extension() with the right capability
* after the kernel starts implementing it.*/
cap_ppc_pvr_compat = false;
if (!cap_interrupt_level) {
fprintf(stderr, "KVM: Couldn't find level irq capability. Expect the "
"VM to stall at times!\n");
}
kvm_ppc_register_host_cpu_type(ms);
return 0;
}
int kvm_arch_irqchip_create(MachineState *ms, KVMState *s)
{
return 0;
}
static int kvm_arch_sync_sregs(PowerPCCPU *cpu)
{
CPUPPCState *cenv = &cpu->env;
CPUState *cs = CPU(cpu);
struct kvm_sregs sregs;
int ret;
if (cenv->excp_model == POWERPC_EXCP_BOOKE) {
/* What we're really trying to say is "if we're on BookE, we use
the native PVR for now". This is the only sane way to check
it though, so we potentially confuse users that they can run
BookE guests on BookS. Let's hope nobody dares enough :) */
return 0;
} else {
if (!cap_segstate) {
fprintf(stderr, "kvm error: missing PVR setting capability\n");
return -ENOSYS;
}
}
ret = kvm_vcpu_ioctl(cs, KVM_GET_SREGS, &sregs);
if (ret) {
return ret;
}
sregs.pvr = cenv->spr[SPR_PVR];
return kvm_vcpu_ioctl(cs, KVM_SET_SREGS, &sregs);
}
/* Set up a shared TLB array with KVM */
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 = {};
unsigned int entries = 0;
int ret, i;
if (!kvm_enabled() ||
!kvm_check_extension(cs->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)&params;
cfg.mmu_type = KVM_MMU_FSL_BOOKE_NOHV;
ret = kvm_vcpu_enable_cap(cs, KVM_CAP_SW_TLB, 0, (uintptr_t)&cfg);
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;
}
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
#if defined(TARGET_PPC64)
static void kvm_get_smmu_info(struct kvm_ppc_smmu_info *info, Error **errp)
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
{
int ret;
assert(kvm_state != NULL);
if (!kvm_check_extension(kvm_state, KVM_CAP_PPC_GET_SMMU_INFO)) {
error_setg(errp, "KVM doesn't expose the MMU features it supports");
error_append_hint(errp, "Consider switching to a newer KVM\n");
return;
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
}
ret = kvm_vm_ioctl(kvm_state, KVM_PPC_GET_SMMU_INFO, info);
if (ret == 0) {
return;
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
}
error_setg_errno(errp, -ret,
"KVM failed to provide the MMU features it supports");
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
}
struct ppc_radix_page_info *kvm_get_radix_page_info(void)
{
KVMState *s = KVM_STATE(current_machine->accelerator);
struct ppc_radix_page_info *radix_page_info;
struct kvm_ppc_rmmu_info rmmu_info;
int i;
if (!kvm_check_extension(s, KVM_CAP_PPC_MMU_RADIX)) {
return NULL;
}
if (kvm_vm_ioctl(s, KVM_PPC_GET_RMMU_INFO, &rmmu_info)) {
return NULL;
}
radix_page_info = g_malloc0(sizeof(*radix_page_info));
radix_page_info->count = 0;
for (i = 0; i < PPC_PAGE_SIZES_MAX_SZ; i++) {
if (rmmu_info.ap_encodings[i]) {
radix_page_info->entries[i] = rmmu_info.ap_encodings[i];
radix_page_info->count++;
}
}
return radix_page_info;
}
target_ulong kvmppc_configure_v3_mmu(PowerPCCPU *cpu,
bool radix, bool gtse,
uint64_t proc_tbl)
{
CPUState *cs = CPU(cpu);
int ret;
uint64_t flags = 0;
struct kvm_ppc_mmuv3_cfg cfg = {
.process_table = proc_tbl,
};
if (radix) {
flags |= KVM_PPC_MMUV3_RADIX;
}
if (gtse) {
flags |= KVM_PPC_MMUV3_GTSE;
}
cfg.flags = flags;
ret = kvm_vm_ioctl(cs->kvm_state, KVM_PPC_CONFIGURE_V3_MMU, &cfg);
switch (ret) {
case 0:
return H_SUCCESS;
case -EINVAL:
return H_PARAMETER;
case -ENODEV:
return H_NOT_AVAILABLE;
default:
return H_HARDWARE;
}
}
bool kvmppc_hpt_needs_host_contiguous_pages(void)
{
static struct kvm_ppc_smmu_info smmu_info;
if (!kvm_enabled()) {
return false;
}
kvm_get_smmu_info(&smmu_info, &error_fatal);
return !!(smmu_info.flags & KVM_PPC_PAGE_SIZES_REAL);
}
2018-04-16 08:19:52 +02:00
void kvm_check_mmu(PowerPCCPU *cpu, Error **errp)
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
{
2018-04-16 08:19:52 +02:00
struct kvm_ppc_smmu_info smmu_info;
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
int iq, ik, jq, jk;
Error *local_err = NULL;
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
2018-04-16 08:19:52 +02:00
/* For now, we only have anything to check on hash64 MMUs */
if (!cpu->hash64_opts || !kvm_enabled()) {
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
return;
}
kvm_get_smmu_info(&smmu_info, &local_err);
if (local_err) {
error_propagate(errp, local_err);
return;
}
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
2018-04-16 08:19:52 +02:00
if (ppc_hash64_has(cpu, PPC_HASH64_1TSEG)
&& !(smmu_info.flags & KVM_PPC_1T_SEGMENTS)) {
error_setg(errp,
"KVM does not support 1TiB segments which guest expects");
return;
}
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
2018-04-16 08:19:52 +02:00
if (smmu_info.slb_size < cpu->hash64_opts->slb_size) {
error_setg(errp, "KVM only supports %u SLB entries, but guest needs %u",
smmu_info.slb_size, cpu->hash64_opts->slb_size);
return;
}
/*
2018-04-16 08:19:52 +02:00
* Verify that every pagesize supported by the cpu model is
* supported by KVM with the same encodings
*/
2018-04-16 08:19:52 +02:00
for (iq = 0; iq < ARRAY_SIZE(cpu->hash64_opts->sps); iq++) {
PPCHash64SegmentPageSizes *qsps = &cpu->hash64_opts->sps[iq];
2018-04-16 08:19:52 +02:00
struct kvm_ppc_one_seg_page_size *ksps;
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
2018-04-16 08:19:52 +02:00
for (ik = 0; ik < ARRAY_SIZE(smmu_info.sps); ik++) {
if (qsps->page_shift == smmu_info.sps[ik].page_shift) {
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
break;
}
}
2018-04-16 08:19:52 +02:00
if (ik >= ARRAY_SIZE(smmu_info.sps)) {
error_setg(errp, "KVM doesn't support for base page shift %u",
qsps->page_shift);
return;
}
ksps = &smmu_info.sps[ik];
if (ksps->slb_enc != qsps->slb_enc) {
error_setg(errp,
"KVM uses SLB encoding 0x%x for page shift %u, but guest expects 0x%x",
ksps->slb_enc, ksps->page_shift, qsps->slb_enc);
return;
}
for (jq = 0; jq < ARRAY_SIZE(qsps->enc); jq++) {
for (jk = 0; jk < ARRAY_SIZE(ksps->enc); jk++) {
if (qsps->enc[jq].page_shift == ksps->enc[jk].page_shift) {
break;
}
}
if (jk >= ARRAY_SIZE(ksps->enc)) {
error_setg(errp, "KVM doesn't support page shift %u/%u",
qsps->enc[jq].page_shift, qsps->page_shift);
return;
}
if (qsps->enc[jq].pte_enc != ksps->enc[jk].pte_enc) {
error_setg(errp,
"KVM uses PTE encoding 0x%x for page shift %u/%u, but guest expects 0x%x",
ksps->enc[jk].pte_enc, qsps->enc[jq].page_shift,
qsps->page_shift, qsps->enc[jq].pte_enc);
return;
}
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
}
}
2018-04-16 08:19:52 +02:00
if (ppc_hash64_has(cpu, PPC_HASH64_CI_LARGEPAGE)) {
/* Mostly what guest pagesizes we can use are related to the
* host pages used to map guest RAM, which is handled in the
* platform code. Cache-Inhibited largepages (64k) however are
* used for I/O, so if they're mapped to the host at all it
* will be a normal mapping, not a special hugepage one used
* for RAM. */
if (getpagesize() < 0x10000) {
error_setg(errp,
"KVM can't supply 64kiB CI pages, which guest expects");
}
}
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
}
#endif /* !defined (TARGET_PPC64) */
unsigned long kvm_arch_vcpu_id(CPUState *cpu)
{
return POWERPC_CPU(cpu)->vcpu_id;
}
/* e500 supports 2 h/w breakpoint and 2 watchpoint.
* book3s supports only 1 watchpoint, so array size
* of 4 is sufficient for now.
*/
#define MAX_HW_BKPTS 4
static struct HWBreakpoint {
target_ulong addr;
int type;
} hw_debug_points[MAX_HW_BKPTS];
static CPUWatchpoint hw_watchpoint;
/* Default there is no breakpoint and watchpoint supported */
static int max_hw_breakpoint;
static int max_hw_watchpoint;
static int nb_hw_breakpoint;
static int nb_hw_watchpoint;
static void kvmppc_hw_debug_points_init(CPUPPCState *cenv)
{
if (cenv->excp_model == POWERPC_EXCP_BOOKE) {
max_hw_breakpoint = 2;
max_hw_watchpoint = 2;
}
if ((max_hw_breakpoint + max_hw_watchpoint) > MAX_HW_BKPTS) {
fprintf(stderr, "Error initializing h/w breakpoints\n");
return;
}
}
int kvm_arch_init_vcpu(CPUState *cs)
{
PowerPCCPU *cpu = POWERPC_CPU(cs);
CPUPPCState *cenv = &cpu->env;
int ret;
ppc64: Rudimentary Support for extra page sizes on server CPUs More recent Power server chips (i.e. based on the 64 bit hash MMU) support more than just the traditional 4k and 16M page sizes. This can get quite complicated, because which page sizes are supported, which combinations are supported within an MMU segment and how these page sizes are encoded both in the SLB entry and the hash PTE can vary depending on the CPU model (they are not specified by the architecture). In addition the firmware or hypervisor may not permit use of certain page sizes, for various reasons. Whether various page sizes are supported on KVM, for example, depends on whether the PR or HV variant of KVM is in use, and on the page size of the memory backing the guest's RAM. This patch adds information to the CPUState and cpu defs to describe the supported page sizes and encodings. Since TCG does not yet support any extended page sizes, we just set this to NULL in the static CPU definitions, expanding this to the default 4k and 16M page sizes when we initialize the cpu state. When using KVM, however, we instead determine available page sizes using the new KVM_PPC_GET_SMMU_INFO call. For old kernels without that call, we use some defaults, with some guesswork which should do the right thing for existing HV and PR implementations. The fallback might not be correct for future versions, but that's ok, because they'll have KVM_PPC_GET_SMMU_INFO. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-06-18 21:56:25 +02:00
/* Synchronize sregs with kvm */
ret = kvm_arch_sync_sregs(cpu);
if (ret) {
if (ret == -EINVAL) {
error_report("Register sync failed... If you're using kvm-hv.ko,"
" only \"-cpu host\" is possible");
}
return ret;
}
idle_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, kvm_kick_cpu, cpu);
switch (cenv->mmu_model) {
case POWERPC_MMU_BOOKE206:
/* This target supports access to KVM's guest TLB */
ret = kvm_booke206_tlb_init(cpu);
break;
case POWERPC_MMU_2_07:
if (!cap_htm && !kvmppc_is_pr(cs->kvm_state)) {
/* KVM-HV has transactional memory on POWER8 also without the
* KVM_CAP_PPC_HTM extension, so enable it here instead as
* long as it's availble to userspace on the host. */
if (qemu_getauxval(AT_HWCAP2) & PPC_FEATURE2_HAS_HTM) {
cap_htm = true;
}
}
break;
default:
break;
}
kvm_get_one_reg(cs, KVM_REG_PPC_DEBUG_INST, &debug_inst_opcode);
kvmppc_hw_debug_points_init(cenv);
return ret;
}
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;
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(cs, KVM_DIRTY_TLB, &dirty_tlb);
if (ret) {
fprintf(stderr, "%s: KVM_DIRTY_TLB: %s\n",
__func__, strerror(-ret));
}
g_free(bitmap);
}
static void kvm_get_one_spr(CPUState *cs, uint64_t id, int spr)
{
PowerPCCPU *cpu = POWERPC_CPU(cs);
CPUPPCState *env = &cpu->env;
union {
uint32_t u32;
uint64_t u64;
} val;
struct kvm_one_reg reg = {
.id = id,
.addr = (uintptr_t) &val,
};
int ret;
ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &reg);
if (ret != 0) {
trace_kvm_failed_spr_get(spr, strerror(errno));
} else {
switch (id & KVM_REG_SIZE_MASK) {
case KVM_REG_SIZE_U32:
env->spr[spr] = val.u32;
break;
case KVM_REG_SIZE_U64:
env->spr[spr] = val.u64;
break;
default:
/* Don't handle this size yet */
abort();
}
}
}
static void kvm_put_one_spr(CPUState *cs, uint64_t id, int spr)
{
PowerPCCPU *cpu = POWERPC_CPU(cs);
CPUPPCState *env = &cpu->env;
union {
uint32_t u32;
uint64_t u64;
} val;
struct kvm_one_reg reg = {
.id = id,
.addr = (uintptr_t) &val,
};
int ret;
switch (id & KVM_REG_SIZE_MASK) {
case KVM_REG_SIZE_U32:
val.u32 = env->spr[spr];
break;
case KVM_REG_SIZE_U64:
val.u64 = env->spr[spr];
break;
default:
/* Don't handle this size yet */
abort();
}
ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &reg);
if (ret != 0) {
trace_kvm_failed_spr_set(spr, strerror(errno));
}
}
static int kvm_put_fp(CPUState *cs)
{
PowerPCCPU *cpu = POWERPC_CPU(cs);
CPUPPCState *env = &cpu->env;
struct kvm_one_reg reg;
int i;
int ret;
if (env->insns_flags & PPC_FLOAT) {
uint64_t fpscr = env->fpscr;
bool vsx = !!(env->insns_flags2 & PPC2_VSX);
reg.id = KVM_REG_PPC_FPSCR;
reg.addr = (uintptr_t)&fpscr;
ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to set FPSCR to KVM: %s\n", strerror(errno));
return ret;
}
for (i = 0; i < 32; i++) {
uint64_t vsr[2];
#ifdef HOST_WORDS_BIGENDIAN
vsr[0] = float64_val(env->fpr[i]);
vsr[1] = env->vsr[i];
#else
vsr[0] = env->vsr[i];
vsr[1] = float64_val(env->fpr[i]);
#endif
reg.addr = (uintptr_t) &vsr;
reg.id = vsx ? KVM_REG_PPC_VSR(i) : KVM_REG_PPC_FPR(i);
ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to set %s%d to KVM: %s\n", vsx ? "VSR" : "FPR",
i, strerror(errno));
return ret;
}
}
}
if (env->insns_flags & PPC_ALTIVEC) {
reg.id = KVM_REG_PPC_VSCR;
reg.addr = (uintptr_t)&env->vscr;
ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to set VSCR to KVM: %s\n", strerror(errno));
return ret;
}
for (i = 0; i < 32; i++) {
reg.id = KVM_REG_PPC_VR(i);
reg.addr = (uintptr_t)&env->avr[i];
ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to set VR%d to KVM: %s\n", i, strerror(errno));
return ret;
}
}
}
return 0;
}
static int kvm_get_fp(CPUState *cs)
{
PowerPCCPU *cpu = POWERPC_CPU(cs);
CPUPPCState *env = &cpu->env;
struct kvm_one_reg reg;
int i;
int ret;
if (env->insns_flags & PPC_FLOAT) {
uint64_t fpscr;
bool vsx = !!(env->insns_flags2 & PPC2_VSX);
reg.id = KVM_REG_PPC_FPSCR;
reg.addr = (uintptr_t)&fpscr;
ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to get FPSCR from KVM: %s\n", strerror(errno));
return ret;
} else {
env->fpscr = fpscr;
}
for (i = 0; i < 32; i++) {
uint64_t vsr[2];
reg.addr = (uintptr_t) &vsr;
reg.id = vsx ? KVM_REG_PPC_VSR(i) : KVM_REG_PPC_FPR(i);
ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to get %s%d from KVM: %s\n",
vsx ? "VSR" : "FPR", i, strerror(errno));
return ret;
} else {
#ifdef HOST_WORDS_BIGENDIAN
env->fpr[i] = vsr[0];
if (vsx) {
env->vsr[i] = vsr[1];
}
#else
env->fpr[i] = vsr[1];
if (vsx) {
env->vsr[i] = vsr[0];
}
#endif
}
}
}
if (env->insns_flags & PPC_ALTIVEC) {
reg.id = KVM_REG_PPC_VSCR;
reg.addr = (uintptr_t)&env->vscr;
ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to get VSCR from KVM: %s\n", strerror(errno));
return ret;
}
for (i = 0; i < 32; i++) {
reg.id = KVM_REG_PPC_VR(i);
reg.addr = (uintptr_t)&env->avr[i];
ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to get VR%d from KVM: %s\n",
i, strerror(errno));
return ret;
}
}
}
return 0;
}
#if defined(TARGET_PPC64)
static int kvm_get_vpa(CPUState *cs)
{
PowerPCCPU *cpu = POWERPC_CPU(cs);
sPAPRCPUState *spapr_cpu = spapr_cpu_state(cpu);
struct kvm_one_reg reg;
int ret;
reg.id = KVM_REG_PPC_VPA_ADDR;
reg.addr = (uintptr_t)&spapr_cpu->vpa_addr;
ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to get VPA address from KVM: %s\n", strerror(errno));
return ret;
}
assert((uintptr_t)&spapr_cpu->slb_shadow_size
== ((uintptr_t)&spapr_cpu->slb_shadow_addr + 8));
reg.id = KVM_REG_PPC_VPA_SLB;
reg.addr = (uintptr_t)&spapr_cpu->slb_shadow_addr;
ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to get SLB shadow state from KVM: %s\n",
strerror(errno));
return ret;
}
assert((uintptr_t)&spapr_cpu->dtl_size
== ((uintptr_t)&spapr_cpu->dtl_addr + 8));
reg.id = KVM_REG_PPC_VPA_DTL;
reg.addr = (uintptr_t)&spapr_cpu->dtl_addr;
ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to get dispatch trace log state from KVM: %s\n",
strerror(errno));
return ret;
}
return 0;
}
static int kvm_put_vpa(CPUState *cs)
{
PowerPCCPU *cpu = POWERPC_CPU(cs);
sPAPRCPUState *spapr_cpu = spapr_cpu_state(cpu);
struct kvm_one_reg reg;
int ret;
/* SLB shadow or DTL can't be registered unless a master VPA is
* registered. That means when restoring state, if a VPA *is*
* registered, we need to set that up first. If not, we need to
* deregister the others before deregistering the master VPA */
assert(spapr_cpu->vpa_addr
|| !(spapr_cpu->slb_shadow_addr || spapr_cpu->dtl_addr));
if (spapr_cpu->vpa_addr) {
reg.id = KVM_REG_PPC_VPA_ADDR;
reg.addr = (uintptr_t)&spapr_cpu->vpa_addr;
ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to set VPA address to KVM: %s\n", strerror(errno));
return ret;
}
}
assert((uintptr_t)&spapr_cpu->slb_shadow_size
== ((uintptr_t)&spapr_cpu->slb_shadow_addr + 8));
reg.id = KVM_REG_PPC_VPA_SLB;
reg.addr = (uintptr_t)&spapr_cpu->slb_shadow_addr;
ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to set SLB shadow state to KVM: %s\n", strerror(errno));
return ret;
}
assert((uintptr_t)&spapr_cpu->dtl_size
== ((uintptr_t)&spapr_cpu->dtl_addr + 8));
reg.id = KVM_REG_PPC_VPA_DTL;
reg.addr = (uintptr_t)&spapr_cpu->dtl_addr;
ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to set dispatch trace log state to KVM: %s\n",
strerror(errno));
return ret;
}
if (!spapr_cpu->vpa_addr) {
reg.id = KVM_REG_PPC_VPA_ADDR;
reg.addr = (uintptr_t)&spapr_cpu->vpa_addr;
ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &reg);
if (ret < 0) {
DPRINTF("Unable to set VPA address to KVM: %s\n", strerror(errno));
return ret;
}
}
return 0;
}
#endif /* TARGET_PPC64 */
int kvmppc_put_books_sregs(PowerPCCPU *cpu)
{
CPUPPCState *env = &cpu->env;
struct kvm_sregs sregs;
int i;
sregs.pvr = env->spr[SPR_PVR];
if (cpu->vhyp) {
PPCVirtualHypervisorClass *vhc =
PPC_VIRTUAL_HYPERVISOR_GET_CLASS(cpu->vhyp);
sregs.u.s.sdr1 = vhc->encode_hpt_for_kvm_pr(cpu->vhyp);
} else {
sregs.u.s.sdr1 = env->spr[SPR_SDR1];
}
/* Sync SLB */
#ifdef TARGET_PPC64
for (i = 0; i < ARRAY_SIZE(env->slb); i++) {
sregs.u.s.ppc64.slb[i].slbe = env->slb[i].esid;
if (env->slb[i].esid & SLB_ESID_V) {
sregs.u.s.ppc64.slb[i].slbe |= i;
}
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];
}
return kvm_vcpu_ioctl(CPU(cpu), KVM_SET_SREGS, &sregs);
}
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(cs, KVM_GET_REGS, &regs);
if (ret < 0) {
return ret;
}
regs.ctr = env->ctr;
regs.lr = env->lr;
regs.xer = cpu_read_xer(env);
regs.msr = env->msr;
regs.pc = env->nip;
regs.srr0 = env->spr[SPR_SRR0];
regs.srr1 = env->spr[SPR_SRR1];
regs.sprg0 = env->spr[SPR_SPRG0];
regs.sprg1 = env->spr[SPR_SPRG1];
regs.sprg2 = env->spr[SPR_SPRG2];
regs.sprg3 = env->spr[SPR_SPRG3];
regs.sprg4 = env->spr[SPR_SPRG4];
regs.sprg5 = env->spr[SPR_SPRG5];
regs.sprg6 = env->spr[SPR_SPRG6];
regs.sprg7 = env->spr[SPR_SPRG7];
regs.pid = env->spr[SPR_BOOKE_PID];
for (i = 0;i < 32; i++)
regs.gpr[i] = env->gpr[i];
regs.cr = 0;
for (i = 0; i < 8; i++) {
regs.cr |= (env->crf[i] & 15) << (4 * (7 - i));
}
ret = kvm_vcpu_ioctl(cs, KVM_SET_REGS, &regs);
if (ret < 0)
return ret;
kvm_put_fp(cs);
if (env->tlb_dirty) {
kvm_sw_tlb_put(cpu);
env->tlb_dirty = false;
}
ppc: Make kvm_arch_put_registers() put *all* the registers At least when invoked with high enough 'level' arguments, kvm_arch_put_registers() is supposed to copy essentially all the cpu state as encoded in qemu's internal structures into the kvm state. Currently the ppc version does not do this - it never calls KVM_SET_SREGS, for example, and therefore never sets the SDR1 and various other important though rarely changed registers. Instead, the code paths which need to set these registers need to explicitly make (conditional) kvm calls which transfer the changes to kvm. This breaks the usual model of handling state updates in qemu, where code just changes the internal model and has it flushed out to kvm automatically at some later point. This patch fixes this for Book S ppc CPUs by adding a suitable call to KVM_SET_SREGS and als to KVM_SET_ONE_REG to set the HIOR (the only register that is set with that call so far). This lets us remove the hacks to explicitly set these registers from the kvmppc_set_papr() function. The problem still exists for Book E CPUs (which use a different version of the kvm_sregs structure). But fixing that has some complications of its own so can be left to another day. Lkewise, there is still some ugly code for setting the PVR through special calls to SET_SREGS which is left in for now. The PVR needs to be set especially early because it can affect what other features are available on the CPU, so I need to do more thinking to see if it can be integrated into the normal paths or not. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-09-12 18:57:09 +02:00
if (cap_segstate && (level >= KVM_PUT_RESET_STATE)) {
ret = kvmppc_put_books_sregs(cpu);
if (ret < 0) {
ppc: Make kvm_arch_put_registers() put *all* the registers At least when invoked with high enough 'level' arguments, kvm_arch_put_registers() is supposed to copy essentially all the cpu state as encoded in qemu's internal structures into the kvm state. Currently the ppc version does not do this - it never calls KVM_SET_SREGS, for example, and therefore never sets the SDR1 and various other important though rarely changed registers. Instead, the code paths which need to set these registers need to explicitly make (conditional) kvm calls which transfer the changes to kvm. This breaks the usual model of handling state updates in qemu, where code just changes the internal model and has it flushed out to kvm automatically at some later point. This patch fixes this for Book S ppc CPUs by adding a suitable call to KVM_SET_SREGS and als to KVM_SET_ONE_REG to set the HIOR (the only register that is set with that call so far). This lets us remove the hacks to explicitly set these registers from the kvmppc_set_papr() function. The problem still exists for Book E CPUs (which use a different version of the kvm_sregs structure). But fixing that has some complications of its own so can be left to another day. Lkewise, there is still some ugly code for setting the PVR through special calls to SET_SREGS which is left in for now. The PVR needs to be set especially early because it can affect what other features are available on the CPU, so I need to do more thinking to see if it can be integrated into the normal paths or not. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-09-12 18:57:09 +02:00
return ret;
}
}
if (cap_hior && (level >= KVM_PUT_RESET_STATE)) {
kvm_put_one_spr(cs, KVM_REG_PPC_HIOR, SPR_HIOR);
}
ppc: Make kvm_arch_put_registers() put *all* the registers At least when invoked with high enough 'level' arguments, kvm_arch_put_registers() is supposed to copy essentially all the cpu state as encoded in qemu's internal structures into the kvm state. Currently the ppc version does not do this - it never calls KVM_SET_SREGS, for example, and therefore never sets the SDR1 and various other important though rarely changed registers. Instead, the code paths which need to set these registers need to explicitly make (conditional) kvm calls which transfer the changes to kvm. This breaks the usual model of handling state updates in qemu, where code just changes the internal model and has it flushed out to kvm automatically at some later point. This patch fixes this for Book S ppc CPUs by adding a suitable call to KVM_SET_SREGS and als to KVM_SET_ONE_REG to set the HIOR (the only register that is set with that call so far). This lets us remove the hacks to explicitly set these registers from the kvmppc_set_papr() function. The problem still exists for Book E CPUs (which use a different version of the kvm_sregs structure). But fixing that has some complications of its own so can be left to another day. Lkewise, there is still some ugly code for setting the PVR through special calls to SET_SREGS which is left in for now. The PVR needs to be set especially early because it can affect what other features are available on the CPU, so I need to do more thinking to see if it can be integrated into the normal paths or not. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-09-12 18:57:09 +02:00
if (cap_one_reg) {
int i;
/* We deliberately ignore errors here, for kernels which have
* the ONE_REG calls, but don't support the specific
* registers, there's a reasonable chance things will still
* work, at least until we try to migrate. */
for (i = 0; i < 1024; i++) {
uint64_t id = env->spr_cb[i].one_reg_id;
if (id != 0) {
kvm_put_one_spr(cs, id, i);
}
ppc: Make kvm_arch_put_registers() put *all* the registers At least when invoked with high enough 'level' arguments, kvm_arch_put_registers() is supposed to copy essentially all the cpu state as encoded in qemu's internal structures into the kvm state. Currently the ppc version does not do this - it never calls KVM_SET_SREGS, for example, and therefore never sets the SDR1 and various other important though rarely changed registers. Instead, the code paths which need to set these registers need to explicitly make (conditional) kvm calls which transfer the changes to kvm. This breaks the usual model of handling state updates in qemu, where code just changes the internal model and has it flushed out to kvm automatically at some later point. This patch fixes this for Book S ppc CPUs by adding a suitable call to KVM_SET_SREGS and als to KVM_SET_ONE_REG to set the HIOR (the only register that is set with that call so far). This lets us remove the hacks to explicitly set these registers from the kvmppc_set_papr() function. The problem still exists for Book E CPUs (which use a different version of the kvm_sregs structure). But fixing that has some complications of its own so can be left to another day. Lkewise, there is still some ugly code for setting the PVR through special calls to SET_SREGS which is left in for now. The PVR needs to be set especially early because it can affect what other features are available on the CPU, so I need to do more thinking to see if it can be integrated into the normal paths or not. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-09-12 18:57:09 +02:00
}
#ifdef TARGET_PPC64
if (msr_ts) {
for (i = 0; i < ARRAY_SIZE(env->tm_gpr); i++) {
kvm_set_one_reg(cs, KVM_REG_PPC_TM_GPR(i), &env->tm_gpr[i]);
}
for (i = 0; i < ARRAY_SIZE(env->tm_vsr); i++) {
kvm_set_one_reg(cs, KVM_REG_PPC_TM_VSR(i), &env->tm_vsr[i]);
}
kvm_set_one_reg(cs, KVM_REG_PPC_TM_CR, &env->tm_cr);
kvm_set_one_reg(cs, KVM_REG_PPC_TM_LR, &env->tm_lr);
kvm_set_one_reg(cs, KVM_REG_PPC_TM_CTR, &env->tm_ctr);
kvm_set_one_reg(cs, KVM_REG_PPC_TM_FPSCR, &env->tm_fpscr);
kvm_set_one_reg(cs, KVM_REG_PPC_TM_AMR, &env->tm_amr);
kvm_set_one_reg(cs, KVM_REG_PPC_TM_PPR, &env->tm_ppr);
kvm_set_one_reg(cs, KVM_REG_PPC_TM_VRSAVE, &env->tm_vrsave);
kvm_set_one_reg(cs, KVM_REG_PPC_TM_VSCR, &env->tm_vscr);
kvm_set_one_reg(cs, KVM_REG_PPC_TM_DSCR, &env->tm_dscr);
kvm_set_one_reg(cs, KVM_REG_PPC_TM_TAR, &env->tm_tar);
}
if (cap_papr) {
if (kvm_put_vpa(cs) < 0) {
DPRINTF("Warning: Unable to set VPA information to KVM\n");
}
}
kvm_set_one_reg(cs, KVM_REG_PPC_TB_OFFSET, &env->tb_env->tb_offset);
#endif /* TARGET_PPC64 */
ppc: Make kvm_arch_put_registers() put *all* the registers At least when invoked with high enough 'level' arguments, kvm_arch_put_registers() is supposed to copy essentially all the cpu state as encoded in qemu's internal structures into the kvm state. Currently the ppc version does not do this - it never calls KVM_SET_SREGS, for example, and therefore never sets the SDR1 and various other important though rarely changed registers. Instead, the code paths which need to set these registers need to explicitly make (conditional) kvm calls which transfer the changes to kvm. This breaks the usual model of handling state updates in qemu, where code just changes the internal model and has it flushed out to kvm automatically at some later point. This patch fixes this for Book S ppc CPUs by adding a suitable call to KVM_SET_SREGS and als to KVM_SET_ONE_REG to set the HIOR (the only register that is set with that call so far). This lets us remove the hacks to explicitly set these registers from the kvmppc_set_papr() function. The problem still exists for Book E CPUs (which use a different version of the kvm_sregs structure). But fixing that has some complications of its own so can be left to another day. Lkewise, there is still some ugly code for setting the PVR through special calls to SET_SREGS which is left in for now. The PVR needs to be set especially early because it can affect what other features are available on the CPU, so I need to do more thinking to see if it can be integrated into the normal paths or not. Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2012-09-12 18:57:09 +02:00
}
return ret;
}
static void kvm_sync_excp(CPUPPCState *env, int vector, int ivor)
{
env->excp_vectors[vector] = env->spr[ivor] + env->spr[SPR_BOOKE_IVPR];
}
static int kvmppc_get_booke_sregs(PowerPCCPU *cpu)
{
CPUPPCState *env = &cpu->env;
struct kvm_sregs sregs;
int ret;
ret = kvm_vcpu_ioctl(CPU(cpu), KVM_GET_SREGS, &sregs);
if (ret < 0) {
return ret;
}
if (sregs.u.e.features & KVM_SREGS_E_BASE) {
env->spr[SPR_BOOKE_CSRR0] = sregs.u.e.csrr0;
env->spr[SPR_BOOKE_CSRR1] = sregs.u.e.csrr1;
env->spr[SPR_BOOKE_ESR] = sregs.u.e.esr;
env->spr[SPR_BOOKE_DEAR] = sregs.u.e.dear;
env->spr[SPR_BOOKE_MCSR] = sregs.u.e.mcsr;
env->spr[SPR_BOOKE_TSR] = sregs.u.e.tsr;
env->spr[SPR_BOOKE_TCR] = sregs.u.e.tcr;
env->spr[SPR_DECR] = sregs.u.e.dec;
env->spr[SPR_TBL] = sregs.u.e.tb & 0xffffffff;
env->spr[SPR_TBU] = sregs.u.e.tb >> 32;
env->spr[SPR_VRSAVE] = sregs.u.e.vrsave;
}
if (sregs.u.e.features & KVM_SREGS_E_ARCH206) {
env->spr[SPR_BOOKE_PIR] = sregs.u.e.pir;
env->spr[SPR_BOOKE_MCSRR0] = sregs.u.e.mcsrr0;
env->spr[SPR_BOOKE_MCSRR1] = sregs.u.e.mcsrr1;
env->spr[SPR_BOOKE_DECAR] = sregs.u.e.decar;
env->spr[SPR_BOOKE_IVPR] = sregs.u.e.ivpr;
}
if (sregs.u.e.features & KVM_SREGS_E_64) {
env->spr[SPR_BOOKE_EPCR] = sregs.u.e.epcr;
}
if (sregs.u.e.features & KVM_SREGS_E_SPRG8) {
env->spr[SPR_BOOKE_SPRG8] = sregs.u.e.sprg8;
}
if (sregs.u.e.features & KVM_SREGS_E_IVOR) {
env->spr[SPR_BOOKE_IVOR0] = sregs.u.e.ivor_low[0];
kvm_sync_excp(env, POWERPC_EXCP_CRITICAL, SPR_BOOKE_IVOR0);
env->spr[SPR_BOOKE_IVOR1] = sregs.u.e.ivor_low[1];
kvm_sync_excp(env, POWERPC_EXCP_MCHECK, SPR_BOOKE_IVOR1);
env->spr[SPR_BOOKE_IVOR2] = sregs.u.e.ivor_low[2];
kvm_sync_excp(env, POWERPC_EXCP_DSI, SPR_BOOKE_IVOR2);
env->spr[SPR_BOOKE_IVOR3] = sregs.u.e.ivor_low[3];
kvm_sync_excp(env, POWERPC_EXCP_ISI, SPR_BOOKE_IVOR3);
env->spr[SPR_BOOKE_IVOR4] = sregs.u.e.ivor_low[4];
kvm_sync_excp(env, POWERPC_EXCP_EXTERNAL, SPR_BOOKE_IVOR4);
env->spr[SPR_BOOKE_IVOR5] = sregs.u.e.ivor_low[5];
kvm_sync_excp(env, POWERPC_EXCP_ALIGN, SPR_BOOKE_IVOR5);
env->spr[SPR_BOOKE_IVOR6] = sregs.u.e.ivor_low[6];
kvm_sync_excp(env, POWERPC_EXCP_PROGRAM, SPR_BOOKE_IVOR6);
env->spr[SPR_BOOKE_IVOR7] = sregs.u.e.ivor_low[7];
kvm_sync_excp(env, POWERPC_EXCP_FPU, SPR_BOOKE_IVOR7);
env->spr[SPR_BOOKE_IVOR8] = sregs.u.e.ivor_low[8];
kvm_sync_excp(env, POWERPC_EXCP_SYSCALL, SPR_BOOKE_IVOR8);
env->spr[SPR_BOOKE_IVOR9] = sregs.u.e.ivor_low[9];
kvm_sync_excp(env, POWERPC_EXCP_APU, SPR_BOOKE_IVOR9);
env->spr[SPR_BOOKE_IVOR10] = sregs.u.e.ivor_low[10];
kvm_sync_excp(env, POWERPC_EXCP_DECR, SPR_BOOKE_IVOR10);
env->spr[SPR_BOOKE_IVOR11] = sregs.u.e.ivor_low[11];
kvm_sync_excp(env, POWERPC_EXCP_FIT, SPR_BOOKE_IVOR11);
env->spr[SPR_BOOKE_IVOR12] = sregs.u.e.ivor_low[12];
kvm_sync_excp(env, POWERPC_EXCP_WDT, SPR_BOOKE_IVOR12);
env->spr[SPR_BOOKE_IVOR13] = sregs.u.e.ivor_low[13];
kvm_sync_excp(env, POWERPC_EXCP_DTLB, SPR_BOOKE_IVOR13);
env->spr[SPR_BOOKE_IVOR14] = sregs.u.e.ivor_low[14];
kvm_sync_excp(env, POWERPC_EXCP_ITLB, SPR_BOOKE_IVOR14);
env->spr[SPR_BOOKE_IVOR15] = sregs.u.e.ivor_low[15];
kvm_sync_excp(env, POWERPC_EXCP_DEBUG, SPR_BOOKE_IVOR15);
if (sregs.u.e.features & KVM_SREGS_E_SPE) {
env->spr[SPR_BOOKE_IVOR32] = sregs.u.e.ivor_high[0];
kvm_sync_excp(env, POWERPC_EXCP_SPEU, SPR_BOOKE_IVOR32);
env->spr[SPR_BOOKE_IVOR33] = sregs.u.e.ivor_high[1];
kvm_sync_excp(env, POWERPC_EXCP_EFPDI, SPR_BOOKE_IVOR33);
env->spr[SPR_BOOKE_IVOR34] = sregs.u.e.ivor_high[2];
kvm_sync_excp(env, POWERPC_EXCP_EFPRI, SPR_BOOKE_IVOR34);
}
if (sregs.u.e.features & KVM_SREGS_E_PM) {
env->spr[SPR_BOOKE_IVOR35] = sregs.u.e.ivor_high[3];
kvm_sync_excp(env, POWERPC_EXCP_EPERFM, SPR_BOOKE_IVOR35);
}
if (sregs.u.e.features & KVM_SREGS_E_PC) {
env->spr[SPR_BOOKE_IVOR36] = sregs.u.e.ivor_high[4];
kvm_sync_excp(env, POWERPC_EXCP_DOORI, SPR_BOOKE_IVOR36);
env->spr[SPR_BOOKE_IVOR37] = sregs.u.e.ivor_high[5];
kvm_sync_excp(env, POWERPC_EXCP_DOORCI, SPR_BOOKE_IVOR37);
}
}
if (sregs.u.e.features & KVM_SREGS_E_ARCH206_MMU) {
env->spr[SPR_BOOKE_MAS0] = sregs.u.e.mas0;
env->spr[SPR_BOOKE_MAS1] = sregs.u.e.mas1;
env->spr[SPR_BOOKE_MAS2] = sregs.u.e.mas2;
env->spr[SPR_BOOKE_MAS3] = sregs.u.e.mas7_3 & 0xffffffff;
env->spr[SPR_BOOKE_MAS4] = sregs.u.e.mas4;
env->spr[SPR_BOOKE_MAS6] = sregs.u.e.mas6;
env->spr[SPR_BOOKE_MAS7] = sregs.u.e.mas7_3 >> 32;
env->spr[SPR_MMUCFG] = sregs.u.e.mmucfg;
env->spr[SPR_BOOKE_TLB0CFG] = sregs.u.e.tlbcfg[0];
env->spr[SPR_BOOKE_TLB1CFG] = sregs.u.e.tlbcfg[1];
}
if (sregs.u.e.features & KVM_SREGS_EXP) {
env->spr[SPR_BOOKE_EPR] = sregs.u.e.epr;
}
if (sregs.u.e.features & KVM_SREGS_E_PD) {
env->spr[SPR_BOOKE_EPLC] = sregs.u.e.eplc;
env->spr[SPR_BOOKE_EPSC] = sregs.u.e.epsc;
}
if (sregs.u.e.impl_id == KVM_SREGS_E_IMPL_FSL) {
env->spr[SPR_E500_SVR] = sregs.u.e.impl.fsl.svr;
env->spr[SPR_Exxx_MCAR] = sregs.u.e.impl.fsl.mcar;
env->spr[SPR_HID0] = sregs.u.e.impl.fsl.hid0;
if (sregs.u.e.impl.fsl.features & KVM_SREGS_E_FSL_PIDn) {
env->spr[SPR_BOOKE_PID1] = sregs.u.e.impl.fsl.pid1;
env->spr[SPR_BOOKE_PID2] = sregs.u.e.impl.fsl.pid2;
}
}
return 0;
}
static int kvmppc_get_books_sregs(PowerPCCPU *cpu)
{
CPUPPCState *env = &cpu->env;
struct kvm_sregs sregs;
int ret;
int i;
ret = kvm_vcpu_ioctl(CPU(cpu), KVM_GET_SREGS, &sregs);
if (ret < 0) {
return ret;
}
if (!cpu->vhyp) {
ppc_store_sdr1(env, sregs.u.s.sdr1);
}
/* Sync SLB */
#ifdef TARGET_PPC64
/*
* The packed SLB array we get from KVM_GET_SREGS only contains
* information about valid entries. So we flush our internal copy
* to get rid of stale ones, then put all valid SLB entries back
* in.
*/
memset(env->slb, 0, sizeof(env->slb));
for (i = 0; i < ARRAY_SIZE(env->slb); i++) {
target_ulong rb = sregs.u.s.ppc64.slb[i].slbe;
target_ulong rs = sregs.u.s.ppc64.slb[i].slbv;
/*
* Only restore valid entries
*/
if (rb & SLB_ESID_V) {
ppc_store_slb(cpu, rb & 0xfff, rb & ~0xfffULL, rs);
}
}
#endif
/* Sync SRs */
for (i = 0; i < 16; i++) {
env->sr[i] = sregs.u.s.ppc32.sr[i];
}
/* Sync BATs */
for (i = 0; i < 8; i++) {
env->DBAT[0][i] = sregs.u.s.ppc32.dbat[i] & 0xffffffff;
env->DBAT[1][i] = sregs.u.s.ppc32.dbat[i] >> 32;
env->IBAT[0][i] = sregs.u.s.ppc32.ibat[i] & 0xffffffff;
env->IBAT[1][i] = sregs.u.s.ppc32.ibat[i] >> 32;
}
return 0;
}
int kvm_arch_get_registers(CPUState *cs)
{
PowerPCCPU *cpu = POWERPC_CPU(cs);
CPUPPCState *env = &cpu->env;
struct kvm_regs regs;
uint32_t cr;
int i, ret;
ret = kvm_vcpu_ioctl(cs, KVM_GET_REGS, &regs);
if (ret < 0)
return ret;
cr = regs.cr;
for (i = 7; i >= 0; i--) {
env->crf[i] = cr & 15;
cr >>= 4;
}
env->ctr = regs.ctr;
env->lr = regs.lr;
cpu_write_xer(env, regs.xer);
env->msr = regs.msr;
env->nip = regs.pc;
env->spr[SPR_SRR0] = regs.srr0;
env->spr[SPR_SRR1] = regs.srr1;
env->spr[SPR_SPRG0] = regs.sprg0;
env->spr[SPR_SPRG1] = regs.sprg1;
env->spr[SPR_SPRG2] = regs.sprg2;
env->spr[SPR_SPRG3] = regs.sprg3;
env->spr[SPR_SPRG4] = regs.sprg4;
env->spr[SPR_SPRG5] = regs.sprg5;
env->spr[SPR_SPRG6] = regs.sprg6;
env->spr[SPR_SPRG7] = regs.sprg7;
env->spr[SPR_BOOKE_PID] = regs.pid;
for (i = 0;i < 32; i++)
env->gpr[i] = regs.gpr[i];
kvm_get_fp(cs);
if (cap_booke_sregs) {
ret = kvmppc_get_booke_sregs(cpu);
if (ret < 0) {
return ret;
}
}
if (cap_segstate) {
ret = kvmppc_get_books_sregs(cpu);
if (ret < 0) {
return ret;
}
}
if (cap_hior) {
kvm_get_one_spr(cs, KVM_REG_PPC_HIOR, SPR_HIOR);
}
if (cap_one_reg) {
int i;
/* We deliberately ignore errors here, for kernels which have
* the ONE_REG calls, but don't support the specific
* registers, there's a reasonable chance things will still
* work, at least until we try to migrate. */
for (i = 0; i < 1024; i++) {
uint64_t id = env->spr_cb[i].one_reg_id;
if (id != 0) {
kvm_get_one_spr(cs, id, i);
}
}
#ifdef TARGET_PPC64
if (msr_ts) {
for (i = 0; i < ARRAY_SIZE(env->tm_gpr); i++) {
kvm_get_one_reg(cs, KVM_REG_PPC_TM_GPR(i), &env->tm_gpr[i]);
}
for (i = 0; i < ARRAY_SIZE(env->tm_vsr); i++) {
kvm_get_one_reg(cs, KVM_REG_PPC_TM_VSR(i), &env->tm_vsr[i]);
}
kvm_get_one_reg(cs, KVM_REG_PPC_TM_CR, &env->tm_cr);
kvm_get_one_reg(cs, KVM_REG_PPC_TM_LR, &env->tm_lr);
kvm_get_one_reg(cs, KVM_REG_PPC_TM_CTR, &env->tm_ctr);
kvm_get_one_reg(cs, KVM_REG_PPC_TM_FPSCR, &env->tm_fpscr);
kvm_get_one_reg(cs, KVM_REG_PPC_TM_AMR, &env->tm_amr);
kvm_get_one_reg(cs, KVM_REG_PPC_TM_PPR, &env->tm_ppr);
kvm_get_one_reg(cs, KVM_REG_PPC_TM_VRSAVE, &env->tm_vrsave);
kvm_get_one_reg(cs, KVM_REG_PPC_TM_VSCR, &env->tm_vscr);
kvm_get_one_reg(cs, KVM_REG_PPC_TM_DSCR, &env->tm_dscr);
kvm_get_one_reg(cs, KVM_REG_PPC_TM_TAR, &env->tm_tar);
}
if (cap_papr) {
if (kvm_get_vpa(cs) < 0) {
DPRINTF("Warning: Unable to get VPA information from KVM\n");
}
}
kvm_get_one_reg(cs, KVM_REG_PPC_TB_OFFSET, &env->tb_env->tb_offset);
#endif
}
return 0;
}
int kvmppc_set_interrupt(PowerPCCPU *cpu, int irq, int level)
{
unsigned virq = level ? KVM_INTERRUPT_SET_LEVEL : KVM_INTERRUPT_UNSET;
if (irq != PPC_INTERRUPT_EXT) {
return 0;
}
if (!kvm_enabled() || !cap_interrupt_unset || !cap_interrupt_level) {
return 0;
}
kvm_vcpu_ioctl(CPU(cpu), KVM_INTERRUPT, &virq);
return 0;
}
#if defined(TARGET_PPC64)
#define PPC_INPUT_INT PPC970_INPUT_INT
#else
#define PPC_INPUT_INT PPC6xx_INPUT_INT
#endif
void kvm_arch_pre_run(CPUState *cs, struct kvm_run *run)
{
PowerPCCPU *cpu = POWERPC_CPU(cs);
CPUPPCState *env = &cpu->env;
int r;
unsigned irq;
qemu_mutex_lock_iothread();
/* PowerPC QEMU tracks the various core input pins (interrupt, critical
* interrupt, reset, etc) in PPC-specific env->irq_input_state. */
if (!cap_interrupt_level &&
run->ready_for_interrupt_injection &&
(cs->interrupt_request & CPU_INTERRUPT_HARD) &&
(env->irq_input_state & (1<<PPC_INPUT_INT)))
{
/* For now KVM disregards the 'irq' argument. However, in the
* future KVM could cache it in-kernel to avoid a heavyweight exit
* when reading the UIC.
*/
irq = KVM_INTERRUPT_SET;
DPRINTF("injected interrupt %d\n", irq);
r = kvm_vcpu_ioctl(cs, KVM_INTERRUPT, &irq);
if (r < 0) {
printf("cpu %d fail inject %x\n", cs->cpu_index, irq);
}
/* Always wake up soon in case the interrupt was level based */
timer_mod(idle_timer, qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) +
(NANOSECONDS_PER_SECOND / 50));
}
/* We don't know if there are more interrupts pending after this. However,
* the guest will return to userspace in the course of handling this one
* anyways, so we will get a chance to deliver the rest. */
qemu_mutex_unlock_iothread();
}
MemTxAttrs kvm_arch_post_run(CPUState *cs, struct kvm_run *run)
{
return MEMTXATTRS_UNSPECIFIED;
}
int kvm_arch_process_async_events(CPUState *cs)
{
return cs->halted;
}
static int kvmppc_handle_halt(PowerPCCPU *cpu)
{
CPUState *cs = CPU(cpu);
CPUPPCState *env = &cpu->env;
if (!(cs->interrupt_request & CPU_INTERRUPT_HARD) && (msr_ee)) {
cs->halted = 1;
cs->exception_index = EXCP_HLT;
}
return 0;
}
/* map dcr access to existing qemu dcr emulation */
static int kvmppc_handle_dcr_read(CPUPPCState *env, uint32_t dcrn, uint32_t *data)
{
if (ppc_dcr_read(env->dcr_env, dcrn, data) < 0)
fprintf(stderr, "Read to unhandled DCR (0x%x)\n", dcrn);
return 0;
}
static int kvmppc_handle_dcr_write(CPUPPCState *env, uint32_t dcrn, uint32_t data)
{
if (ppc_dcr_write(env->dcr_env, dcrn, data) < 0)
fprintf(stderr, "Write to unhandled DCR (0x%x)\n", dcrn);
return 0;
}
int kvm_arch_insert_sw_breakpoint(CPUState *cs, struct kvm_sw_breakpoint *bp)
{
/* Mixed endian case is not handled */
uint32_t sc = debug_inst_opcode;
if (cpu_memory_rw_debug(cs, bp->pc, (uint8_t *)&bp->saved_insn,
sizeof(sc), 0) ||
cpu_memory_rw_debug(cs, bp->pc, (uint8_t *)&sc, sizeof(sc), 1)) {
return -EINVAL;
}
return 0;
}
int kvm_arch_remove_sw_breakpoint(CPUState *cs, struct kvm_sw_breakpoint *bp)
{
uint32_t sc;
if (cpu_memory_rw_debug(cs, bp->pc, (uint8_t *)&sc, sizeof(sc), 0) ||
sc != debug_inst_opcode ||
cpu_memory_rw_debug(cs, bp->pc, (uint8_t *)&bp->saved_insn,
sizeof(sc), 1)) {
return -EINVAL;
}
return 0;
}
static int find_hw_breakpoint(target_ulong addr, int type)
{
int n;
assert((nb_hw_breakpoint + nb_hw_watchpoint)
<= ARRAY_SIZE(hw_debug_points));
for (n = 0; n < nb_hw_breakpoint + nb_hw_watchpoint; n++) {
if (hw_debug_points[n].addr == addr &&
hw_debug_points[n].type == type) {
return n;
}
}
return -1;
}
static int find_hw_watchpoint(target_ulong addr, int *flag)
{
int n;
n = find_hw_breakpoint(addr, GDB_WATCHPOINT_ACCESS);
if (n >= 0) {
*flag = BP_MEM_ACCESS;
return n;
}
n = find_hw_breakpoint(addr, GDB_WATCHPOINT_WRITE);
if (n >= 0) {
*flag = BP_MEM_WRITE;
return n;
}
n = find_hw_breakpoint(addr, GDB_WATCHPOINT_READ);
if (n >= 0) {
*flag = BP_MEM_READ;
return n;
}
return -1;
}
int kvm_arch_insert_hw_breakpoint(target_ulong addr,
target_ulong len, int type)
{
if ((nb_hw_breakpoint + nb_hw_watchpoint) >= ARRAY_SIZE(hw_debug_points)) {
return -ENOBUFS;
}
hw_debug_points[nb_hw_breakpoint + nb_hw_watchpoint].addr = addr;
hw_debug_points[nb_hw_breakpoint + nb_hw_watchpoint].type = type;
switch (type) {
case GDB_BREAKPOINT_HW:
if (nb_hw_breakpoint >= max_hw_breakpoint) {
return -ENOBUFS;
}
if (find_hw_breakpoint(addr, type) >= 0) {
return -EEXIST;
}
nb_hw_breakpoint++;
break;
case GDB_WATCHPOINT_WRITE:
case GDB_WATCHPOINT_READ:
case GDB_WATCHPOINT_ACCESS:
if (nb_hw_watchpoint >= max_hw_watchpoint) {
return -ENOBUFS;
}
if (find_hw_breakpoint(addr, type) >= 0) {
return -EEXIST;
}
nb_hw_watchpoint++;
break;
default:
return -ENOSYS;
}
return 0;
}
int kvm_arch_remove_hw_breakpoint(target_ulong addr,
target_ulong len, int type)
{
int n;
n = find_hw_breakpoint(addr, type);
if (n < 0) {
return -ENOENT;
}
switch (type) {
case GDB_BREAKPOINT_HW:
nb_hw_breakpoint--;
break;
case GDB_WATCHPOINT_WRITE:
case GDB_WATCHPOINT_READ:
case GDB_WATCHPOINT_ACCESS:
nb_hw_watchpoint--;
break;
default:
return -ENOSYS;
}
hw_debug_points[n] = hw_debug_points[nb_hw_breakpoint + nb_hw_watchpoint];
return 0;
}
void kvm_arch_remove_all_hw_breakpoints(void)
{
nb_hw_breakpoint = nb_hw_watchpoint = 0;
}
void kvm_arch_update_guest_debug(CPUState *cs, struct kvm_guest_debug *dbg)
{
int n;
/* Software Breakpoint updates */
if (kvm_sw_breakpoints_active(cs)) {
dbg->control |= KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_SW_BP;
}
assert((nb_hw_breakpoint + nb_hw_watchpoint)
<= ARRAY_SIZE(hw_debug_points));
assert((nb_hw_breakpoint + nb_hw_watchpoint) <= ARRAY_SIZE(dbg->arch.bp));
if (nb_hw_breakpoint + nb_hw_watchpoint > 0) {
dbg->control |= KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_HW_BP;
memset(dbg->arch.bp, 0, sizeof(dbg->arch.bp));
for (n = 0; n < nb_hw_breakpoint + nb_hw_watchpoint; n++) {
switch (hw_debug_points[n].type) {
case GDB_BREAKPOINT_HW:
dbg->arch.bp[n].type = KVMPPC_DEBUG_BREAKPOINT;
break;
case GDB_WATCHPOINT_WRITE:
dbg->arch.bp[n].type = KVMPPC_DEBUG_WATCH_WRITE;
break;
case GDB_WATCHPOINT_READ:
dbg->arch.bp[n].type = KVMPPC_DEBUG_WATCH_READ;
break;
case GDB_WATCHPOINT_ACCESS:
dbg->arch.bp[n].type = KVMPPC_DEBUG_WATCH_WRITE |
KVMPPC_DEBUG_WATCH_READ;
break;
default:
cpu_abort(cs, "Unsupported breakpoint type\n");
}
dbg->arch.bp[n].addr = hw_debug_points[n].addr;
}
}
}
static int kvm_handle_debug(PowerPCCPU *cpu, struct kvm_run *run)
{
CPUState *cs = CPU(cpu);
CPUPPCState *env = &cpu->env;
struct kvm_debug_exit_arch *arch_info = &run->debug.arch;
int handle = 0;
int n;
int flag = 0;
if (cs->singlestep_enabled) {
handle = 1;
} else if (arch_info->status) {
if (nb_hw_breakpoint + nb_hw_watchpoint > 0) {
if (arch_info->status & KVMPPC_DEBUG_BREAKPOINT) {
n = find_hw_breakpoint(arch_info->address, GDB_BREAKPOINT_HW);
if (n >= 0) {
handle = 1;
}
} else if (arch_info->status & (KVMPPC_DEBUG_WATCH_READ |
KVMPPC_DEBUG_WATCH_WRITE)) {
n = find_hw_watchpoint(arch_info->address, &flag);
if (n >= 0) {
handle = 1;
cs->watchpoint_hit = &hw_watchpoint;
hw_watchpoint.vaddr = hw_debug_points[n].addr;
hw_watchpoint.flags = flag;
}
}
}
} else if (kvm_find_sw_breakpoint(cs, arch_info->address)) {
handle = 1;
} else {
/* QEMU is not able to handle debug exception, so inject
* program exception to guest;
* Yes program exception NOT debug exception !!
* When QEMU is using debug resources then debug exception must
* be always set. To achieve this we set MSR_DE and also set
* MSRP_DEP so guest cannot change MSR_DE.
* When emulating debug resource for guest we want guest
* to control MSR_DE (enable/disable debug interrupt on need).
* Supporting both configurations are NOT possible.
* So the result is that we cannot share debug resources
* between QEMU and Guest on BOOKE architecture.
* In the current design QEMU gets the priority over guest,
* this means that if QEMU is using debug resources then guest
* cannot use them;
* For software breakpoint QEMU uses a privileged instruction;
* So there cannot be any reason that we are here for guest
* set debug exception, only possibility is guest executed a
* privileged / illegal instruction and that's why we are
* injecting a program interrupt.
*/
cpu_synchronize_state(cs);
/* env->nip is PC, so increment this by 4 to use
* ppc_cpu_do_interrupt(), which set srr0 = env->nip - 4.
*/
env->nip += 4;
cs->exception_index = POWERPC_EXCP_PROGRAM;
env->error_code = POWERPC_EXCP_INVAL;
ppc_cpu_do_interrupt(cs);
}
return handle;
}
int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
{
PowerPCCPU *cpu = POWERPC_CPU(cs);
CPUPPCState *env = &cpu->env;
int ret;
qemu_mutex_lock_iothread();
switch (run->exit_reason) {
case KVM_EXIT_DCR:
if (run->dcr.is_write) {
DPRINTF("handle dcr write\n");
ret = kvmppc_handle_dcr_write(env, run->dcr.dcrn, run->dcr.data);
} else {
DPRINTF("handle dcr read\n");
ret = kvmppc_handle_dcr_read(env, run->dcr.dcrn, &run->dcr.data);
}
break;
case KVM_EXIT_HLT:
DPRINTF("handle halt\n");
ret = kvmppc_handle_halt(cpu);
break;
#if defined(TARGET_PPC64)
case KVM_EXIT_PAPR_HCALL:
DPRINTF("handle PAPR hypercall\n");
run->papr_hcall.ret = spapr_hypercall(cpu,
run->papr_hcall.nr,
run->papr_hcall.args);
ret = 0;
break;
#endif
case KVM_EXIT_EPR:
DPRINTF("handle epr\n");
run->epr.epr = ldl_phys(cs->as, env->mpic_iack);
ret = 0;
break;
case KVM_EXIT_WATCHDOG:
DPRINTF("handle watchdog expiry\n");
watchdog_perform_action();
ret = 0;
break;
case KVM_EXIT_DEBUG:
DPRINTF("handle debug exception\n");
if (kvm_handle_debug(cpu, run)) {
ret = EXCP_DEBUG;
break;
}
/* re-enter, this exception was guest-internal */
ret = 0;
break;
default:
fprintf(stderr, "KVM: unknown exit reason %d\n", run->exit_reason);
ret = -1;
break;
}
qemu_mutex_unlock_iothread();
return ret;
}
int kvmppc_or_tsr_bits(PowerPCCPU *cpu, uint32_t tsr_bits)
{
CPUState *cs = CPU(cpu);
uint32_t bits = tsr_bits;
struct kvm_one_reg reg = {
.id = KVM_REG_PPC_OR_TSR,
.addr = (uintptr_t) &bits,
};
return kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &reg);
}
int kvmppc_clear_tsr_bits(PowerPCCPU *cpu, uint32_t tsr_bits)
{
CPUState *cs = CPU(cpu);
uint32_t bits = tsr_bits;
struct kvm_one_reg reg = {
.id = KVM_REG_PPC_CLEAR_TSR,
.addr = (uintptr_t) &bits,
};
return kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &reg);
}
int kvmppc_set_tcr(PowerPCCPU *cpu)
{
CPUState *cs = CPU(cpu);
CPUPPCState *env = &cpu->env;
uint32_t tcr = env->spr[SPR_BOOKE_TCR];
struct kvm_one_reg reg = {
.id = KVM_REG_PPC_TCR,
.addr = (uintptr_t) &tcr,
};
return kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &reg);
}
int kvmppc_booke_watchdog_enable(PowerPCCPU *cpu)
{
CPUState *cs = CPU(cpu);
int ret;
if (!kvm_enabled()) {
return -1;
}
if (!cap_ppc_watchdog) {
printf("warning: KVM does not support watchdog");
return -1;
}
ret = kvm_vcpu_enable_cap(cs, KVM_CAP_PPC_BOOKE_WATCHDOG, 0);
if (ret < 0) {
fprintf(stderr, "%s: couldn't enable KVM_CAP_PPC_BOOKE_WATCHDOG: %s\n",
__func__, strerror(-ret));
return ret;
}
return ret;
}
static int read_cpuinfo(const char *field, char *value, int len)
{
FILE *f;
int ret = -1;
int field_len = strlen(field);
char line[512];
f = fopen("/proc/cpuinfo", "r");
if (!f) {
return -1;
}
do {
if (!fgets(line, sizeof(line), f)) {
break;
}
if (!strncmp(line, field, field_len)) {
pstrcpy(value, len, line);
ret = 0;
break;
}
} while(*line);
fclose(f);
return ret;
}
uint32_t kvmppc_get_tbfreq(void)
{
char line[512];
char *ns;
uint32_t retval = NANOSECONDS_PER_SECOND;
if (read_cpuinfo("timebase", line, sizeof(line))) {
return retval;
}
if (!(ns = strchr(line, ':'))) {
return retval;
}
ns++;
return atoi(ns);
}
bool kvmppc_get_host_serial(char **value)
{
return g_file_get_contents("/proc/device-tree/system-id", value, NULL,
NULL);
}
bool kvmppc_get_host_model(char **value)
{
return g_file_get_contents("/proc/device-tree/model", value, NULL, NULL);
}
/* 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;
}
static uint64_t kvmppc_read_int_dt(const char *filename)
{
union {
uint32_t v32;
uint64_t v64;
} u;
FILE *f;
int len;
f = fopen(filename, "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;
}
/* 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], *tmp;
uint64_t val;
if (kvmppc_find_cpu_dt(buf, sizeof(buf))) {
return -1;
}
tmp = g_strdup_printf("%s/%s", buf, propname);
val = kvmppc_read_int_dt(tmp);
g_free(tmp);
return val;
}
uint64_t kvmppc_get_clockfreq(void)
{
return kvmppc_read_int_cpu_dt("clock-frequency");
}
static int kvmppc_get_pvinfo(CPUPPCState *env, struct kvm_ppc_pvinfo *pvinfo)
{
PowerPCCPU *cpu = ppc_env_get_cpu(env);
CPUState *cs = CPU(cpu);
if (kvm_vm_check_extension(cs->kvm_state, KVM_CAP_PPC_GET_PVINFO) &&
!kvm_vm_ioctl(cs->kvm_state, KVM_PPC_GET_PVINFO, pvinfo)) {
return 0;
}
return 1;
}
int kvmppc_get_hasidle(CPUPPCState *env)
{
struct kvm_ppc_pvinfo pvinfo;
if (!kvmppc_get_pvinfo(env, &pvinfo) &&
(pvinfo.flags & KVM_PPC_PVINFO_FLAGS_EV_IDLE)) {
return 1;
}
return 0;
}
int kvmppc_get_hypercall(CPUPPCState *env, uint8_t *buf, int buf_len)
{
uint32_t *hc = (uint32_t*)buf;
struct kvm_ppc_pvinfo pvinfo;
if (!kvmppc_get_pvinfo(env, &pvinfo)) {
memcpy(buf, pvinfo.hcall, buf_len);
return 0;
}
/*
* Fallback to always fail hypercalls regardless of endianness:
*
* tdi 0,r0,72 (becomes b .+8 in wrong endian, nop in good endian)
* li r3, -1
* b .+8 (becomes nop in wrong endian)
* bswap32(li r3, -1)
*/
hc[0] = cpu_to_be32(0x08000048);
hc[1] = cpu_to_be32(0x3860ffff);
hc[2] = cpu_to_be32(0x48000008);
hc[3] = cpu_to_be32(bswap32(0x3860ffff));
spapr/target-ppc/kvm: Only add hcall-instructions if KVM supports it ePAPR defines "hcall-instructions" device-tree property which contains code to call hypercalls in ePAPR paravirtualized guests. In general pseries guests won't use this property, instead using the PAPR defined hypercall interface. However, this property has been re-used to implement a hack to allow PR KVM to run (slightly modified) guests in some situations where it otherwise wouldn't be able to (because the system's L0 hypervisor doesn't forward the PAPR hypercalls to the PR KVM kernel). Hence, this property is always present in the device tree for pseries guests. All KVM guests use it at least to read features via the KVM_HC_FEATURES hypercall. The property is populated by the code returned from the KVM's KVM_PPC_GET_PVINFO ioctl; if not implemented in the KVM, QEMU supplies code which will fail all hypercall attempts. If QEMU does not create the property, and the guest kernel is compiled with CONFIG_EPAPR_PARAVIRT (which is normally the case), there is exactly the same stub at @epapr_hypercall_start already. Rather than maintaining this fairly useless stub implementation, it makes more sense not to create the property in the device tree in the first place if the host kernel does not implement it. This changes kvmppc_get_hypercall() to return 1 if the host kernel does not implement KVM_CAP_PPC_GET_PVINFO. The caller can use it to decide on whether to create the property or not. This changes the pseries machine to not create the property if KVM does not implement KVM_PPC_GET_PVINFO. In practice this means that from now on the property will not be created if either HV KVM or TCG is used. Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> [reworded commit message for clarity --dwg] Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2016-03-21 03:14:02 +01:00
return 1;
}
static inline int kvmppc_enable_hcall(KVMState *s, target_ulong hcall)
{
return kvm_vm_enable_cap(s, KVM_CAP_PPC_ENABLE_HCALL, 0, hcall, 1);
}
void kvmppc_enable_logical_ci_hcalls(void)
{
/*
* FIXME: it would be nice if we could detect the cases where
* we're using a device which requires the in kernel
* implementation of these hcalls, but the kernel lacks them and
* produce a warning.
*/
kvmppc_enable_hcall(kvm_state, H_LOGICAL_CI_LOAD);
kvmppc_enable_hcall(kvm_state, H_LOGICAL_CI_STORE);
}
void kvmppc_enable_set_mode_hcall(void)
{
kvmppc_enable_hcall(kvm_state, H_SET_MODE);
}
void kvmppc_enable_clear_ref_mod_hcalls(void)
{
kvmppc_enable_hcall(kvm_state, H_CLEAR_REF);
kvmppc_enable_hcall(kvm_state, H_CLEAR_MOD);
}
void kvmppc_set_papr(PowerPCCPU *cpu)
{
CPUState *cs = CPU(cpu);
int ret;
if (!kvm_enabled()) {
return;
}
ret = kvm_vcpu_enable_cap(cs, KVM_CAP_PPC_PAPR, 0);
if (ret) {
error_report("This vCPU type or KVM version does not support PAPR");
exit(1);
}
/* Update the capability flag so we sync the right information
* with kvm */
cap_papr = 1;
}
int kvmppc_set_compat(PowerPCCPU *cpu, uint32_t compat_pvr)
{
return kvm_set_one_reg(CPU(cpu), KVM_REG_PPC_ARCH_COMPAT, &compat_pvr);
}
void kvmppc_set_mpic_proxy(PowerPCCPU *cpu, int mpic_proxy)
{
CPUState *cs = CPU(cpu);
int ret;
ret = kvm_vcpu_enable_cap(cs, KVM_CAP_PPC_EPR, 0, mpic_proxy);
if (ret && mpic_proxy) {
error_report("This KVM version does not support EPR");
exit(1);
}
}
int kvmppc_smt_threads(void)
{
return cap_ppc_smt ? cap_ppc_smt : 1;
}
int kvmppc_set_smt_threads(int smt)
{
int ret;
ret = kvm_vm_enable_cap(kvm_state, KVM_CAP_PPC_SMT, 0, smt, 0);
if (!ret) {
cap_ppc_smt = smt;
}
return ret;
}
void kvmppc_hint_smt_possible(Error **errp)
{
int i;
GString *g;
char *s;
assert(kvm_enabled());
if (cap_ppc_smt_possible) {
g = g_string_new("Available VSMT modes:");
for (i = 63; i >= 0; i--) {
if ((1UL << i) & cap_ppc_smt_possible) {
g_string_append_printf(g, " %lu", (1UL << i));
}
}
s = g_string_free(g, false);
error_append_hint(errp, "%s.\n", s);
g_free(s);
} else {
error_append_hint(errp,
"This KVM seems to be too old to support VSMT.\n");
}
}
#ifdef TARGET_PPC64
uint64_t kvmppc_rma_size(uint64_t current_size, unsigned int hash_shift)
{
struct kvm_ppc_smmu_info info;
long rampagesize, best_page_shift;
int i;
/* Find the largest hardware supported page size that's less than
* or equal to the (logical) backing page size of guest RAM */
kvm_get_smmu_info(&info, &error_fatal);
rampagesize = qemu_getrampagesize();
best_page_shift = 0;
for (i = 0; i < KVM_PPC_PAGE_SIZES_MAX_SZ; i++) {
struct kvm_ppc_one_seg_page_size *sps = &info.sps[i];
if (!sps->page_shift) {
continue;
}
if ((sps->page_shift > best_page_shift)
&& ((1UL << sps->page_shift) <= rampagesize)) {
best_page_shift = sps->page_shift;
}
}
return MIN(current_size,
1ULL << (best_page_shift + hash_shift - 7));
}
#endif
bool kvmppc_spapr_use_multitce(void)
{
return cap_spapr_multitce;
}
int kvmppc_spapr_enable_inkernel_multitce(void)
{
int ret;
ret = kvm_vm_enable_cap(kvm_state, KVM_CAP_PPC_ENABLE_HCALL, 0,
H_PUT_TCE_INDIRECT, 1);
if (!ret) {
ret = kvm_vm_enable_cap(kvm_state, KVM_CAP_PPC_ENABLE_HCALL, 0,
H_STUFF_TCE, 1);
}
return ret;
}
void *kvmppc_create_spapr_tce(uint32_t liobn, uint32_t page_shift,
uint64_t bus_offset, uint32_t nb_table,
int *pfd, bool need_vfio)
{
long len;
int fd;
void *table;
/* Must set fd to -1 so we don't try to munmap when called for
* destroying the table, which the upper layers -will- do
*/
*pfd = -1;
if (!cap_spapr_tce || (need_vfio && !cap_spapr_vfio)) {
return NULL;
}
if (cap_spapr_tce_64) {
struct kvm_create_spapr_tce_64 args = {
.liobn = liobn,
.page_shift = page_shift,
.offset = bus_offset >> page_shift,
.size = nb_table,
.flags = 0
};
fd = kvm_vm_ioctl(kvm_state, KVM_CREATE_SPAPR_TCE_64, &args);
if (fd < 0) {
fprintf(stderr,
"KVM: Failed to create TCE64 table for liobn 0x%x\n",
liobn);
return NULL;
}
} else if (cap_spapr_tce) {
uint64_t window_size = (uint64_t) nb_table << page_shift;
struct kvm_create_spapr_tce args = {
.liobn = liobn,
.window_size = window_size,
};
if ((window_size != args.window_size) || bus_offset) {
return NULL;
}
fd = kvm_vm_ioctl(kvm_state, KVM_CREATE_SPAPR_TCE, &args);
if (fd < 0) {
fprintf(stderr, "KVM: Failed to create TCE table for liobn 0x%x\n",
liobn);
return NULL;
}
} else {
return NULL;
}
len = nb_table * sizeof(uint64_t);
/* FIXME: round this up to page size */
table = mmap(NULL, len, PROT_READ|PROT_WRITE, MAP_SHARED, fd, 0);
if (table == MAP_FAILED) {
fprintf(stderr, "KVM: Failed to map TCE table for liobn 0x%x\n",
liobn);
close(fd);
return NULL;
}
*pfd = fd;
return table;
}
int kvmppc_remove_spapr_tce(void *table, int fd, uint32_t nb_table)
{
long len;
if (fd < 0) {
return -1;
}
len = nb_table * sizeof(uint64_t);
if ((munmap(table, len) < 0) ||
(close(fd) < 0)) {
fprintf(stderr, "KVM: Unexpected error removing TCE table: %s",
strerror(errno));
/* Leak the table */
}
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_vm_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. */
if (kvmppc_is_pr(kvm_state)) {
/* 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;
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;
}
}
static void kvmppc_host_cpu_class_init(ObjectClass *oc, void *data)
{
PowerPCCPUClass *pcc = POWERPC_CPU_CLASS(oc);
uint32_t dcache_size = kvmppc_read_int_cpu_dt("d-cache-size");
uint32_t icache_size = kvmppc_read_int_cpu_dt("i-cache-size");
/* Now fix up the class with information we can query from the host */
pcc->pvr = mfpvr();
alter_insns(&pcc->insns_flags, PPC_ALTIVEC,
qemu_getauxval(AT_HWCAP) & PPC_FEATURE_HAS_ALTIVEC);
alter_insns(&pcc->insns_flags2, PPC2_VSX,
qemu_getauxval(AT_HWCAP) & PPC_FEATURE_HAS_VSX);
alter_insns(&pcc->insns_flags2, PPC2_DFP,
qemu_getauxval(AT_HWCAP) & PPC_FEATURE_HAS_DFP);
if (dcache_size != -1) {
pcc->l1_dcache_size = dcache_size;
}
if (icache_size != -1) {
pcc->l1_icache_size = icache_size;
}
#if defined(TARGET_PPC64)
pcc->radix_page_info = kvm_get_radix_page_info();
if ((pcc->pvr & 0xffffff00) == CPU_POWERPC_POWER9_DD1) {
/*
* POWER9 DD1 has some bugs which make it not really ISA 3.00
* compliant. More importantly, advertising ISA 3.00
* architected mode may prevent guests from activating
* necessary DD1 workarounds.
*/
pcc->pcr_supported &= ~(PCR_COMPAT_3_00 | PCR_COMPAT_2_07
| PCR_COMPAT_2_06 | PCR_COMPAT_2_05);
}
#endif /* defined(TARGET_PPC64) */
}
bool kvmppc_has_cap_epr(void)
{
return cap_epr;
}
bool kvmppc_has_cap_fixup_hcalls(void)
{
return cap_fixup_hcalls;
}
bool kvmppc_has_cap_htm(void)
{
return cap_htm;
}
bool kvmppc_has_cap_mmu_radix(void)
{
return cap_mmu_radix;
}
bool kvmppc_has_cap_mmu_hash_v3(void)
{
return cap_mmu_hash_v3;
}
static bool kvmppc_power8_host(void)
{
bool ret = false;
#ifdef TARGET_PPC64
{
uint32_t base_pvr = CPU_POWERPC_POWER_SERVER_MASK & mfpvr();
ret = (base_pvr == CPU_POWERPC_POWER8E_BASE) ||
(base_pvr == CPU_POWERPC_POWER8NVL_BASE) ||
(base_pvr == CPU_POWERPC_POWER8_BASE);
}
#endif /* TARGET_PPC64 */
return ret;
}
static int parse_cap_ppc_safe_cache(struct kvm_ppc_cpu_char c)
{
bool l1d_thread_priv_req = !kvmppc_power8_host();
if (~c.behaviour & c.behaviour_mask & H_CPU_BEHAV_L1D_FLUSH_PR) {
return 2;
} else if ((!l1d_thread_priv_req ||
c.character & c.character_mask & H_CPU_CHAR_L1D_THREAD_PRIV) &&
(c.character & c.character_mask
& (H_CPU_CHAR_L1D_FLUSH_ORI30 | H_CPU_CHAR_L1D_FLUSH_TRIG2))) {
return 1;
}
return 0;
}
static int parse_cap_ppc_safe_bounds_check(struct kvm_ppc_cpu_char c)
{
if (~c.behaviour & c.behaviour_mask & H_CPU_BEHAV_BNDS_CHK_SPEC_BAR) {
return 2;
} else if (c.character & c.character_mask & H_CPU_CHAR_SPEC_BAR_ORI31) {
return 1;
}
return 0;
}
static int parse_cap_ppc_safe_indirect_branch(struct kvm_ppc_cpu_char c)
{
if (c.character & c.character_mask & H_CPU_CHAR_CACHE_COUNT_DIS) {
return SPAPR_CAP_FIXED_CCD;
} else if (c.character & c.character_mask & H_CPU_CHAR_BCCTRL_SERIALISED) {
return SPAPR_CAP_FIXED_IBS;
}
return 0;
}
static void kvmppc_get_cpu_characteristics(KVMState *s)
{
struct kvm_ppc_cpu_char c;
int ret;
/* Assume broken */
cap_ppc_safe_cache = 0;
cap_ppc_safe_bounds_check = 0;
cap_ppc_safe_indirect_branch = 0;
ret = kvm_vm_check_extension(s, KVM_CAP_PPC_GET_CPU_CHAR);
if (!ret) {
return;
}
ret = kvm_vm_ioctl(s, KVM_PPC_GET_CPU_CHAR, &c);
if (ret < 0) {
return;
}
cap_ppc_safe_cache = parse_cap_ppc_safe_cache(c);
cap_ppc_safe_bounds_check = parse_cap_ppc_safe_bounds_check(c);
cap_ppc_safe_indirect_branch = parse_cap_ppc_safe_indirect_branch(c);
}
int kvmppc_get_cap_safe_cache(void)
{
return cap_ppc_safe_cache;
}
int kvmppc_get_cap_safe_bounds_check(void)
{
return cap_ppc_safe_bounds_check;
}
int kvmppc_get_cap_safe_indirect_branch(void)
{
return cap_ppc_safe_indirect_branch;
}
bool kvmppc_has_cap_spapr_vfio(void)
{
return cap_spapr_vfio;
}
PowerPCCPUClass *kvm_ppc_get_host_cpu_class(void)
{
uint32_t host_pvr = mfpvr();
PowerPCCPUClass *pvr_pcc;
pvr_pcc = ppc_cpu_class_by_pvr(host_pvr);
if (pvr_pcc == NULL) {
pvr_pcc = ppc_cpu_class_by_pvr_mask(host_pvr);
}
return pvr_pcc;
}
static int kvm_ppc_register_host_cpu_type(MachineState *ms)
{
TypeInfo type_info = {
.name = TYPE_HOST_POWERPC_CPU,
.class_init = kvmppc_host_cpu_class_init,
};
MachineClass *mc = MACHINE_GET_CLASS(ms);
PowerPCCPUClass *pvr_pcc;
ObjectClass *oc;
DeviceClass *dc;
int i;
pvr_pcc = kvm_ppc_get_host_cpu_class();
if (pvr_pcc == NULL) {
return -1;
}
type_info.parent = object_class_get_name(OBJECT_CLASS(pvr_pcc));
type_register(&type_info);
if (object_dynamic_cast(OBJECT(ms), TYPE_SPAPR_MACHINE)) {
/* override TCG default cpu type with 'host' cpu model */
mc->default_cpu_type = TYPE_HOST_POWERPC_CPU;
}
oc = object_class_by_name(type_info.name);
g_assert(oc);
/*
* Update generic CPU family class alias (e.g. on a POWER8NVL host,
* we want "POWER8" to be a "family" alias that points to the current
* host CPU type, too)
*/
dc = DEVICE_CLASS(ppc_cpu_get_family_class(pvr_pcc));
for (i = 0; ppc_cpu_aliases[i].alias != NULL; i++) {
if (strcasecmp(ppc_cpu_aliases[i].alias, dc->desc) == 0) {
char *suffix;
ppc_cpu_aliases[i].model = g_strdup(object_class_get_name(oc));
suffix = strstr(ppc_cpu_aliases[i].model, POWERPC_CPU_TYPE_SUFFIX);
if (suffix) {
*suffix = 0;
}
break;
}
}
return 0;
}
int kvmppc_define_rtas_kernel_token(uint32_t token, const char *function)
{
struct kvm_rtas_token_args args = {
.token = token,
};
if (!kvm_check_extension(kvm_state, KVM_CAP_PPC_RTAS)) {
return -ENOENT;
}
strncpy(args.name, function, sizeof(args.name));
return kvm_vm_ioctl(kvm_state, KVM_PPC_RTAS_DEFINE_TOKEN, &args);
}
int kvmppc_get_htab_fd(bool write, uint64_t index, Error **errp)
{
struct kvm_get_htab_fd s = {
.flags = write ? KVM_GET_HTAB_WRITE : 0,
.start_index = index,
};
int ret;
if (!cap_htab_fd) {
error_setg(errp, "KVM version doesn't support %s the HPT",
write ? "writing" : "reading");
return -ENOTSUP;
}
ret = kvm_vm_ioctl(kvm_state, KVM_PPC_GET_HTAB_FD, &s);
if (ret < 0) {
error_setg(errp, "Unable to open fd for %s HPT %s KVM: %s",
write ? "writing" : "reading", write ? "to" : "from",
strerror(errno));
return -errno;
}
return ret;
}
int kvmppc_save_htab(QEMUFile *f, int fd, size_t bufsize, int64_t max_ns)
{
int64_t starttime = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
uint8_t buf[bufsize];
ssize_t rc;
do {
rc = read(fd, buf, bufsize);
if (rc < 0) {
fprintf(stderr, "Error reading data from KVM HTAB fd: %s\n",
strerror(errno));
return rc;
} else if (rc) {
uint8_t *buffer = buf;
ssize_t n = rc;
while (n) {
struct kvm_get_htab_header *head =
(struct kvm_get_htab_header *) buffer;
size_t chunksize = sizeof(*head) +
HASH_PTE_SIZE_64 * head->n_valid;
qemu_put_be32(f, head->index);
qemu_put_be16(f, head->n_valid);
qemu_put_be16(f, head->n_invalid);
qemu_put_buffer(f, (void *)(head + 1),
HASH_PTE_SIZE_64 * head->n_valid);
buffer += chunksize;
n -= chunksize;
}
}
} while ((rc != 0)
&& ((max_ns < 0)
|| ((qemu_clock_get_ns(QEMU_CLOCK_REALTIME) - starttime) < max_ns)));
return (rc == 0) ? 1 : 0;
}
int kvmppc_load_htab_chunk(QEMUFile *f, int fd, uint32_t index,
uint16_t n_valid, uint16_t n_invalid)
{
struct kvm_get_htab_header *buf;
size_t chunksize = sizeof(*buf) + n_valid*HASH_PTE_SIZE_64;
ssize_t rc;
buf = alloca(chunksize);
buf->index = index;
buf->n_valid = n_valid;
buf->n_invalid = n_invalid;
qemu_get_buffer(f, (void *)(buf + 1), HASH_PTE_SIZE_64*n_valid);
rc = write(fd, buf, chunksize);
if (rc < 0) {
fprintf(stderr, "Error writing KVM hash table: %s\n",
strerror(errno));
return rc;
}
if (rc != chunksize) {
/* We should never get a short write on a single chunk */
fprintf(stderr, "Short write, restoring KVM hash table\n");
return -1;
}
return 0;
}
bool kvm_arch_stop_on_emulation_error(CPUState *cpu)
{
return true;
}
void kvm_arch_init_irq_routing(KVMState *s)
{
}
target/ppc: Fix KVM-HV HPTE accessors When a 'pseries' guest is running with KVM-HV, the guest's hashed page table (HPT) is stored within the host kernel, so it is not directly accessible to qemu. Most of the time, qemu doesn't need to access it: we're using the hardware MMU, and KVM itself implements the guest hypercalls for manipulating the HPT. However, qemu does need access to the in-KVM HPT to implement get_phys_page_debug() for the benefit of the gdbstub, and maybe for other debug operations. To allow this, 7c43bca "target-ppc: Fix page table lookup with kvm enabled" added kvmppc_hash64_read_pteg() to target/ppc/kvm.c to read in a batch of HPTEs from the KVM table. Unfortunately, there are a couple of problems with this: First, the name of the function implies it always reads a whole PTEG from the HPT, but in fact in some cases it's used to grab individual HPTEs (which ends up pulling 8 HPTEs, not aligned to a PTEG from the kernel). Second, and more importantly, the code to read the HPTEs from KVM is simply wrong, in general. The data from the fd that KVM provides is designed mostly for compact migration rather than this sort of one-off access, and so needs some decoding for this purpose. The current code will work in some cases, but if there are invalid HPTEs then it will not get sane results. This patch rewrite the HPTE reading function to have a simpler interface (just read n HPTEs into a caller provided buffer), and to correctly decode the stream from the kernel. For consistency we also clean up the similar function for altering HPTEs within KVM (introduced in c138593 "target-ppc: Update ppc_hash64_store_hpte to support updating in-kernel htab"). Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-02-27 05:34:19 +01:00
void kvmppc_read_hptes(ppc_hash_pte64_t *hptes, hwaddr ptex, int n)
{
target/ppc: Fix KVM-HV HPTE accessors When a 'pseries' guest is running with KVM-HV, the guest's hashed page table (HPT) is stored within the host kernel, so it is not directly accessible to qemu. Most of the time, qemu doesn't need to access it: we're using the hardware MMU, and KVM itself implements the guest hypercalls for manipulating the HPT. However, qemu does need access to the in-KVM HPT to implement get_phys_page_debug() for the benefit of the gdbstub, and maybe for other debug operations. To allow this, 7c43bca "target-ppc: Fix page table lookup with kvm enabled" added kvmppc_hash64_read_pteg() to target/ppc/kvm.c to read in a batch of HPTEs from the KVM table. Unfortunately, there are a couple of problems with this: First, the name of the function implies it always reads a whole PTEG from the HPT, but in fact in some cases it's used to grab individual HPTEs (which ends up pulling 8 HPTEs, not aligned to a PTEG from the kernel). Second, and more importantly, the code to read the HPTEs from KVM is simply wrong, in general. The data from the fd that KVM provides is designed mostly for compact migration rather than this sort of one-off access, and so needs some decoding for this purpose. The current code will work in some cases, but if there are invalid HPTEs then it will not get sane results. This patch rewrite the HPTE reading function to have a simpler interface (just read n HPTEs into a caller provided buffer), and to correctly decode the stream from the kernel. For consistency we also clean up the similar function for altering HPTEs within KVM (introduced in c138593 "target-ppc: Update ppc_hash64_store_hpte to support updating in-kernel htab"). Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-02-27 05:34:19 +01:00
int fd, rc;
int i;
fd = kvmppc_get_htab_fd(false, ptex, &error_abort);
target/ppc: Fix KVM-HV HPTE accessors When a 'pseries' guest is running with KVM-HV, the guest's hashed page table (HPT) is stored within the host kernel, so it is not directly accessible to qemu. Most of the time, qemu doesn't need to access it: we're using the hardware MMU, and KVM itself implements the guest hypercalls for manipulating the HPT. However, qemu does need access to the in-KVM HPT to implement get_phys_page_debug() for the benefit of the gdbstub, and maybe for other debug operations. To allow this, 7c43bca "target-ppc: Fix page table lookup with kvm enabled" added kvmppc_hash64_read_pteg() to target/ppc/kvm.c to read in a batch of HPTEs from the KVM table. Unfortunately, there are a couple of problems with this: First, the name of the function implies it always reads a whole PTEG from the HPT, but in fact in some cases it's used to grab individual HPTEs (which ends up pulling 8 HPTEs, not aligned to a PTEG from the kernel). Second, and more importantly, the code to read the HPTEs from KVM is simply wrong, in general. The data from the fd that KVM provides is designed mostly for compact migration rather than this sort of one-off access, and so needs some decoding for this purpose. The current code will work in some cases, but if there are invalid HPTEs then it will not get sane results. This patch rewrite the HPTE reading function to have a simpler interface (just read n HPTEs into a caller provided buffer), and to correctly decode the stream from the kernel. For consistency we also clean up the similar function for altering HPTEs within KVM (introduced in c138593 "target-ppc: Update ppc_hash64_store_hpte to support updating in-kernel htab"). Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-02-27 05:34:19 +01:00
i = 0;
while (i < n) {
struct kvm_get_htab_header *hdr;
int m = n < HPTES_PER_GROUP ? n : HPTES_PER_GROUP;
char buf[sizeof(*hdr) + m * HASH_PTE_SIZE_64];
target/ppc: Fix KVM-HV HPTE accessors When a 'pseries' guest is running with KVM-HV, the guest's hashed page table (HPT) is stored within the host kernel, so it is not directly accessible to qemu. Most of the time, qemu doesn't need to access it: we're using the hardware MMU, and KVM itself implements the guest hypercalls for manipulating the HPT. However, qemu does need access to the in-KVM HPT to implement get_phys_page_debug() for the benefit of the gdbstub, and maybe for other debug operations. To allow this, 7c43bca "target-ppc: Fix page table lookup with kvm enabled" added kvmppc_hash64_read_pteg() to target/ppc/kvm.c to read in a batch of HPTEs from the KVM table. Unfortunately, there are a couple of problems with this: First, the name of the function implies it always reads a whole PTEG from the HPT, but in fact in some cases it's used to grab individual HPTEs (which ends up pulling 8 HPTEs, not aligned to a PTEG from the kernel). Second, and more importantly, the code to read the HPTEs from KVM is simply wrong, in general. The data from the fd that KVM provides is designed mostly for compact migration rather than this sort of one-off access, and so needs some decoding for this purpose. The current code will work in some cases, but if there are invalid HPTEs then it will not get sane results. This patch rewrite the HPTE reading function to have a simpler interface (just read n HPTEs into a caller provided buffer), and to correctly decode the stream from the kernel. For consistency we also clean up the similar function for altering HPTEs within KVM (introduced in c138593 "target-ppc: Update ppc_hash64_store_hpte to support updating in-kernel htab"). Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-02-27 05:34:19 +01:00
rc = read(fd, buf, sizeof(buf));
if (rc < 0) {
hw_error("kvmppc_read_hptes: Unable to read HPTEs");
}
target/ppc: Fix KVM-HV HPTE accessors When a 'pseries' guest is running with KVM-HV, the guest's hashed page table (HPT) is stored within the host kernel, so it is not directly accessible to qemu. Most of the time, qemu doesn't need to access it: we're using the hardware MMU, and KVM itself implements the guest hypercalls for manipulating the HPT. However, qemu does need access to the in-KVM HPT to implement get_phys_page_debug() for the benefit of the gdbstub, and maybe for other debug operations. To allow this, 7c43bca "target-ppc: Fix page table lookup with kvm enabled" added kvmppc_hash64_read_pteg() to target/ppc/kvm.c to read in a batch of HPTEs from the KVM table. Unfortunately, there are a couple of problems with this: First, the name of the function implies it always reads a whole PTEG from the HPT, but in fact in some cases it's used to grab individual HPTEs (which ends up pulling 8 HPTEs, not aligned to a PTEG from the kernel). Second, and more importantly, the code to read the HPTEs from KVM is simply wrong, in general. The data from the fd that KVM provides is designed mostly for compact migration rather than this sort of one-off access, and so needs some decoding for this purpose. The current code will work in some cases, but if there are invalid HPTEs then it will not get sane results. This patch rewrite the HPTE reading function to have a simpler interface (just read n HPTEs into a caller provided buffer), and to correctly decode the stream from the kernel. For consistency we also clean up the similar function for altering HPTEs within KVM (introduced in c138593 "target-ppc: Update ppc_hash64_store_hpte to support updating in-kernel htab"). Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-02-27 05:34:19 +01:00
hdr = (struct kvm_get_htab_header *)buf;
while ((i < n) && ((char *)hdr < (buf + rc))) {
int invalid = hdr->n_invalid, valid = hdr->n_valid;
target/ppc: Fix KVM-HV HPTE accessors When a 'pseries' guest is running with KVM-HV, the guest's hashed page table (HPT) is stored within the host kernel, so it is not directly accessible to qemu. Most of the time, qemu doesn't need to access it: we're using the hardware MMU, and KVM itself implements the guest hypercalls for manipulating the HPT. However, qemu does need access to the in-KVM HPT to implement get_phys_page_debug() for the benefit of the gdbstub, and maybe for other debug operations. To allow this, 7c43bca "target-ppc: Fix page table lookup with kvm enabled" added kvmppc_hash64_read_pteg() to target/ppc/kvm.c to read in a batch of HPTEs from the KVM table. Unfortunately, there are a couple of problems with this: First, the name of the function implies it always reads a whole PTEG from the HPT, but in fact in some cases it's used to grab individual HPTEs (which ends up pulling 8 HPTEs, not aligned to a PTEG from the kernel). Second, and more importantly, the code to read the HPTEs from KVM is simply wrong, in general. The data from the fd that KVM provides is designed mostly for compact migration rather than this sort of one-off access, and so needs some decoding for this purpose. The current code will work in some cases, but if there are invalid HPTEs then it will not get sane results. This patch rewrite the HPTE reading function to have a simpler interface (just read n HPTEs into a caller provided buffer), and to correctly decode the stream from the kernel. For consistency we also clean up the similar function for altering HPTEs within KVM (introduced in c138593 "target-ppc: Update ppc_hash64_store_hpte to support updating in-kernel htab"). Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-02-27 05:34:19 +01:00
if (hdr->index != (ptex + i)) {
hw_error("kvmppc_read_hptes: Unexpected HPTE index %"PRIu32
" != (%"HWADDR_PRIu" + %d", hdr->index, ptex, i);
}
if (n - i < valid) {
valid = n - i;
}
memcpy(hptes + i, hdr + 1, HASH_PTE_SIZE_64 * valid);
i += valid;
target/ppc: Fix KVM-HV HPTE accessors When a 'pseries' guest is running with KVM-HV, the guest's hashed page table (HPT) is stored within the host kernel, so it is not directly accessible to qemu. Most of the time, qemu doesn't need to access it: we're using the hardware MMU, and KVM itself implements the guest hypercalls for manipulating the HPT. However, qemu does need access to the in-KVM HPT to implement get_phys_page_debug() for the benefit of the gdbstub, and maybe for other debug operations. To allow this, 7c43bca "target-ppc: Fix page table lookup with kvm enabled" added kvmppc_hash64_read_pteg() to target/ppc/kvm.c to read in a batch of HPTEs from the KVM table. Unfortunately, there are a couple of problems with this: First, the name of the function implies it always reads a whole PTEG from the HPT, but in fact in some cases it's used to grab individual HPTEs (which ends up pulling 8 HPTEs, not aligned to a PTEG from the kernel). Second, and more importantly, the code to read the HPTEs from KVM is simply wrong, in general. The data from the fd that KVM provides is designed mostly for compact migration rather than this sort of one-off access, and so needs some decoding for this purpose. The current code will work in some cases, but if there are invalid HPTEs then it will not get sane results. This patch rewrite the HPTE reading function to have a simpler interface (just read n HPTEs into a caller provided buffer), and to correctly decode the stream from the kernel. For consistency we also clean up the similar function for altering HPTEs within KVM (introduced in c138593 "target-ppc: Update ppc_hash64_store_hpte to support updating in-kernel htab"). Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-02-27 05:34:19 +01:00
if ((n - i) < invalid) {
invalid = n - i;
}
memset(hptes + i, 0, invalid * HASH_PTE_SIZE_64);
i += invalid;
target/ppc: Fix KVM-HV HPTE accessors When a 'pseries' guest is running with KVM-HV, the guest's hashed page table (HPT) is stored within the host kernel, so it is not directly accessible to qemu. Most of the time, qemu doesn't need to access it: we're using the hardware MMU, and KVM itself implements the guest hypercalls for manipulating the HPT. However, qemu does need access to the in-KVM HPT to implement get_phys_page_debug() for the benefit of the gdbstub, and maybe for other debug operations. To allow this, 7c43bca "target-ppc: Fix page table lookup with kvm enabled" added kvmppc_hash64_read_pteg() to target/ppc/kvm.c to read in a batch of HPTEs from the KVM table. Unfortunately, there are a couple of problems with this: First, the name of the function implies it always reads a whole PTEG from the HPT, but in fact in some cases it's used to grab individual HPTEs (which ends up pulling 8 HPTEs, not aligned to a PTEG from the kernel). Second, and more importantly, the code to read the HPTEs from KVM is simply wrong, in general. The data from the fd that KVM provides is designed mostly for compact migration rather than this sort of one-off access, and so needs some decoding for this purpose. The current code will work in some cases, but if there are invalid HPTEs then it will not get sane results. This patch rewrite the HPTE reading function to have a simpler interface (just read n HPTEs into a caller provided buffer), and to correctly decode the stream from the kernel. For consistency we also clean up the similar function for altering HPTEs within KVM (introduced in c138593 "target-ppc: Update ppc_hash64_store_hpte to support updating in-kernel htab"). Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-02-27 05:34:19 +01:00
hdr = (struct kvm_get_htab_header *)
((char *)(hdr + 1) + HASH_PTE_SIZE_64 * hdr->n_valid);
}
}
close(fd);
}
target/ppc: Fix KVM-HV HPTE accessors When a 'pseries' guest is running with KVM-HV, the guest's hashed page table (HPT) is stored within the host kernel, so it is not directly accessible to qemu. Most of the time, qemu doesn't need to access it: we're using the hardware MMU, and KVM itself implements the guest hypercalls for manipulating the HPT. However, qemu does need access to the in-KVM HPT to implement get_phys_page_debug() for the benefit of the gdbstub, and maybe for other debug operations. To allow this, 7c43bca "target-ppc: Fix page table lookup with kvm enabled" added kvmppc_hash64_read_pteg() to target/ppc/kvm.c to read in a batch of HPTEs from the KVM table. Unfortunately, there are a couple of problems with this: First, the name of the function implies it always reads a whole PTEG from the HPT, but in fact in some cases it's used to grab individual HPTEs (which ends up pulling 8 HPTEs, not aligned to a PTEG from the kernel). Second, and more importantly, the code to read the HPTEs from KVM is simply wrong, in general. The data from the fd that KVM provides is designed mostly for compact migration rather than this sort of one-off access, and so needs some decoding for this purpose. The current code will work in some cases, but if there are invalid HPTEs then it will not get sane results. This patch rewrite the HPTE reading function to have a simpler interface (just read n HPTEs into a caller provided buffer), and to correctly decode the stream from the kernel. For consistency we also clean up the similar function for altering HPTEs within KVM (introduced in c138593 "target-ppc: Update ppc_hash64_store_hpte to support updating in-kernel htab"). Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-02-27 05:34:19 +01:00
void kvmppc_write_hpte(hwaddr ptex, uint64_t pte0, uint64_t pte1)
{
target/ppc: Fix KVM-HV HPTE accessors When a 'pseries' guest is running with KVM-HV, the guest's hashed page table (HPT) is stored within the host kernel, so it is not directly accessible to qemu. Most of the time, qemu doesn't need to access it: we're using the hardware MMU, and KVM itself implements the guest hypercalls for manipulating the HPT. However, qemu does need access to the in-KVM HPT to implement get_phys_page_debug() for the benefit of the gdbstub, and maybe for other debug operations. To allow this, 7c43bca "target-ppc: Fix page table lookup with kvm enabled" added kvmppc_hash64_read_pteg() to target/ppc/kvm.c to read in a batch of HPTEs from the KVM table. Unfortunately, there are a couple of problems with this: First, the name of the function implies it always reads a whole PTEG from the HPT, but in fact in some cases it's used to grab individual HPTEs (which ends up pulling 8 HPTEs, not aligned to a PTEG from the kernel). Second, and more importantly, the code to read the HPTEs from KVM is simply wrong, in general. The data from the fd that KVM provides is designed mostly for compact migration rather than this sort of one-off access, and so needs some decoding for this purpose. The current code will work in some cases, but if there are invalid HPTEs then it will not get sane results. This patch rewrite the HPTE reading function to have a simpler interface (just read n HPTEs into a caller provided buffer), and to correctly decode the stream from the kernel. For consistency we also clean up the similar function for altering HPTEs within KVM (introduced in c138593 "target-ppc: Update ppc_hash64_store_hpte to support updating in-kernel htab"). Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-02-27 05:34:19 +01:00
int fd, rc;
struct {
struct kvm_get_htab_header hdr;
uint64_t pte0;
uint64_t pte1;
} buf;
fd = kvmppc_get_htab_fd(true, 0 /* Ignored */, &error_abort);
target/ppc: Fix KVM-HV HPTE accessors When a 'pseries' guest is running with KVM-HV, the guest's hashed page table (HPT) is stored within the host kernel, so it is not directly accessible to qemu. Most of the time, qemu doesn't need to access it: we're using the hardware MMU, and KVM itself implements the guest hypercalls for manipulating the HPT. However, qemu does need access to the in-KVM HPT to implement get_phys_page_debug() for the benefit of the gdbstub, and maybe for other debug operations. To allow this, 7c43bca "target-ppc: Fix page table lookup with kvm enabled" added kvmppc_hash64_read_pteg() to target/ppc/kvm.c to read in a batch of HPTEs from the KVM table. Unfortunately, there are a couple of problems with this: First, the name of the function implies it always reads a whole PTEG from the HPT, but in fact in some cases it's used to grab individual HPTEs (which ends up pulling 8 HPTEs, not aligned to a PTEG from the kernel). Second, and more importantly, the code to read the HPTEs from KVM is simply wrong, in general. The data from the fd that KVM provides is designed mostly for compact migration rather than this sort of one-off access, and so needs some decoding for this purpose. The current code will work in some cases, but if there are invalid HPTEs then it will not get sane results. This patch rewrite the HPTE reading function to have a simpler interface (just read n HPTEs into a caller provided buffer), and to correctly decode the stream from the kernel. For consistency we also clean up the similar function for altering HPTEs within KVM (introduced in c138593 "target-ppc: Update ppc_hash64_store_hpte to support updating in-kernel htab"). Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-02-27 05:34:19 +01:00
buf.hdr.n_valid = 1;
buf.hdr.n_invalid = 0;
buf.hdr.index = ptex;
buf.pte0 = cpu_to_be64(pte0);
buf.pte1 = cpu_to_be64(pte1);
target/ppc: Fix KVM-HV HPTE accessors When a 'pseries' guest is running with KVM-HV, the guest's hashed page table (HPT) is stored within the host kernel, so it is not directly accessible to qemu. Most of the time, qemu doesn't need to access it: we're using the hardware MMU, and KVM itself implements the guest hypercalls for manipulating the HPT. However, qemu does need access to the in-KVM HPT to implement get_phys_page_debug() for the benefit of the gdbstub, and maybe for other debug operations. To allow this, 7c43bca "target-ppc: Fix page table lookup with kvm enabled" added kvmppc_hash64_read_pteg() to target/ppc/kvm.c to read in a batch of HPTEs from the KVM table. Unfortunately, there are a couple of problems with this: First, the name of the function implies it always reads a whole PTEG from the HPT, but in fact in some cases it's used to grab individual HPTEs (which ends up pulling 8 HPTEs, not aligned to a PTEG from the kernel). Second, and more importantly, the code to read the HPTEs from KVM is simply wrong, in general. The data from the fd that KVM provides is designed mostly for compact migration rather than this sort of one-off access, and so needs some decoding for this purpose. The current code will work in some cases, but if there are invalid HPTEs then it will not get sane results. This patch rewrite the HPTE reading function to have a simpler interface (just read n HPTEs into a caller provided buffer), and to correctly decode the stream from the kernel. For consistency we also clean up the similar function for altering HPTEs within KVM (introduced in c138593 "target-ppc: Update ppc_hash64_store_hpte to support updating in-kernel htab"). Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-02-27 05:34:19 +01:00
rc = write(fd, &buf, sizeof(buf));
if (rc != sizeof(buf)) {
hw_error("kvmppc_write_hpte: Unable to update KVM HPT");
}
close(fd);
}
int kvm_arch_fixup_msi_route(struct kvm_irq_routing_entry *route,
uint64_t address, uint32_t data, PCIDevice *dev)
{
return 0;
}
int kvm_arch_add_msi_route_post(struct kvm_irq_routing_entry *route,
int vector, PCIDevice *dev)
{
return 0;
}
int kvm_arch_release_virq_post(int virq)
{
return 0;
}
int kvm_arch_msi_data_to_gsi(uint32_t data)
{
return data & 0xffff;
}
int kvmppc_enable_hwrng(void)
{
if (!kvm_enabled() || !kvm_check_extension(kvm_state, KVM_CAP_PPC_HWRNG)) {
return -1;
}
return kvmppc_enable_hcall(kvm_state, H_RANDOM);
}
void kvmppc_check_papr_resize_hpt(Error **errp)
{
if (!kvm_enabled()) {
return; /* No KVM, we're good */
}
if (cap_resize_hpt) {
return; /* Kernel has explicit support, we're good */
}
/* Otherwise fallback on looking for PR KVM */
if (kvmppc_is_pr(kvm_state)) {
return;
}
error_setg(errp,
"Hash page table resizing not available with this KVM version");
}
int kvmppc_resize_hpt_prepare(PowerPCCPU *cpu, target_ulong flags, int shift)
{
CPUState *cs = CPU(cpu);
struct kvm_ppc_resize_hpt rhpt = {
.flags = flags,
.shift = shift,
};
if (!cap_resize_hpt) {
return -ENOSYS;
}
return kvm_vm_ioctl(cs->kvm_state, KVM_PPC_RESIZE_HPT_PREPARE, &rhpt);
}
int kvmppc_resize_hpt_commit(PowerPCCPU *cpu, target_ulong flags, int shift)
{
CPUState *cs = CPU(cpu);
struct kvm_ppc_resize_hpt rhpt = {
.flags = flags,
.shift = shift,
};
if (!cap_resize_hpt) {
return -ENOSYS;
}
return kvm_vm_ioctl(cs->kvm_state, KVM_PPC_RESIZE_HPT_COMMIT, &rhpt);
}
target/ppc: 'PVR != host PVR' in KVM_SET_SREGS workaround Commit d5fc133eed ("ppc: Rework CPU compatibility testing across migration") changed the way cpu_post_load behaves with the PVR setting, causing an unexpected bug in KVM-HV migrations between hosts that are compatible (POWER8 and POWER8E, for example). Even with pvr_match() returning true, the guest freezes right after cpu_post_load. The reason is that the guest kernel can't handle a different PVR value other that the running host in KVM_SET_SREGS. In [1] it was discussed the possibility of a new KVM capability that would indicate that the guest kernel can handle a different PVR in KVM_SET_SREGS. Even if such feature is implemented, there is still the problem with older kernels that will not have this capability and will fail to migrate. This patch implements a workaround for that scenario. If running with KVM, check if the guest kernel does not have the capability (named here as 'cap_ppc_pvr_compat'). If it doesn't, calls kvmppc_is_pr() to see if the guest is running in KVM-HV. If all this happens, set env->spr[SPR_PVR] to the same value as the current host PVR. This ensures that we allow migrations with 'close enough' PVRs to still work in KVM-HV but also makes the code ready for this new KVM capability when it is done. A new function called 'kvmppc_pvr_workaround_required' was created to encapsulate the conditions said above and to avoid calling too many kvm.c internals inside cpu_post_load. [1] https://lists.gnu.org/archive/html/qemu-ppc/2017-06/msg00503.html Signed-off-by: Daniel Henrique Barboza <danielhb@linux.vnet.ibm.com> [dwg: Fix for the case of using TCG on a PPC host] Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
2017-08-09 22:43:46 +02:00
/*
* This is a helper function to detect a post migration scenario
* in which a guest, running as KVM-HV, freezes in cpu_post_load because
* the guest kernel can't handle a PVR value other than the actual host
* PVR in KVM_SET_SREGS, even if pvr_match() returns true.
*
* If we don't have cap_ppc_pvr_compat and we're not running in PR
* (so, we're HV), return true. The workaround itself is done in
* cpu_post_load.
*
* The order here is important: we'll only check for KVM PR as a
* fallback if the guest kernel can't handle the situation itself.
* We need to avoid as much as possible querying the running KVM type
* in QEMU level.
*/
bool kvmppc_pvr_workaround_required(PowerPCCPU *cpu)
{
CPUState *cs = CPU(cpu);
if (!kvm_enabled()) {
return false;
}
if (cap_ppc_pvr_compat) {
return false;
}
return !kvmppc_is_pr(cs->kvm_state);
}
void kvmppc_set_reg_ppc_online(PowerPCCPU *cpu, unsigned int online)
{
CPUState *cs = CPU(cpu);
if (kvm_enabled()) {
kvm_set_one_reg(cs, KVM_REG_PPC_ONLINE, &online);
}
}