linux/arch/mips/kvm/trap_emul.c

588 lines
16 KiB
C
Raw Normal View History

/*
* This file is subject to the terms and conditions of the GNU General Public
* License. See the file "COPYING" in the main directory of this archive
* for more details.
*
* KVM/MIPS: Deliver/Emulate exceptions to the guest kernel
*
* Copyright (C) 2012 MIPS Technologies, Inc. All rights reserved.
* Authors: Sanjay Lal <sanjayl@kymasys.com>
*/
#include <linux/errno.h>
#include <linux/err.h>
#include <linux/module.h>
#include <linux/vmalloc.h>
#include <linux/kvm_host.h>
#include "opcode.h"
#include "interrupt.h"
static gpa_t kvm_trap_emul_gva_to_gpa_cb(gva_t gva)
{
gpa_t gpa;
uint32_t kseg = KSEGX(gva);
if ((kseg == CKSEG0) || (kseg == CKSEG1))
gpa = CPHYSADDR(gva);
else {
kvm_err("%s: cannot find GPA for GVA: %#lx\n", __func__, gva);
kvm_mips_dump_host_tlbs();
gpa = KVM_INVALID_ADDR;
}
kvm_debug("%s: gva %#lx, gpa: %#llx\n", __func__, gva, gpa);
return gpa;
}
static int kvm_trap_emul_handle_cop_unusable(struct kvm_vcpu *vcpu)
{
struct kvm_run *run = vcpu->run;
uint32_t __user *opc = (uint32_t __user *) vcpu->arch.pc;
unsigned long cause = vcpu->arch.host_cp0_cause;
enum emulation_result er = EMULATE_DONE;
int ret = RESUME_GUEST;
if (((cause & CAUSEF_CE) >> CAUSEB_CE) == 1)
er = kvm_mips_emulate_fpu_exc(cause, opc, run, vcpu);
else
er = kvm_mips_emulate_inst(cause, opc, run, vcpu);
switch (er) {
case EMULATE_DONE:
ret = RESUME_GUEST;
break;
case EMULATE_FAIL:
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
break;
case EMULATE_WAIT:
run->exit_reason = KVM_EXIT_INTR;
ret = RESUME_HOST;
break;
default:
BUG();
}
return ret;
}
static int kvm_trap_emul_handle_tlb_mod(struct kvm_vcpu *vcpu)
{
struct kvm_run *run = vcpu->run;
uint32_t __user *opc = (uint32_t __user *) vcpu->arch.pc;
unsigned long badvaddr = vcpu->arch.host_cp0_badvaddr;
unsigned long cause = vcpu->arch.host_cp0_cause;
enum emulation_result er = EMULATE_DONE;
int ret = RESUME_GUEST;
if (KVM_GUEST_KSEGX(badvaddr) < KVM_GUEST_KSEG0
|| KVM_GUEST_KSEGX(badvaddr) == KVM_GUEST_KSEG23) {
kvm_debug("USER/KSEG23 ADDR TLB MOD fault: cause %#lx, PC: %p, BadVaddr: %#lx\n",
cause, opc, badvaddr);
er = kvm_mips_handle_tlbmod(cause, opc, run, vcpu);
if (er == EMULATE_DONE)
ret = RESUME_GUEST;
else {
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
} else if (KVM_GUEST_KSEGX(badvaddr) == KVM_GUEST_KSEG0) {
/*
* XXXKYMA: The guest kernel does not expect to get this fault
* when we are not using HIGHMEM. Need to address this in a
* HIGHMEM kernel
*/
kvm_err("TLB MOD fault not handled, cause %#lx, PC: %p, BadVaddr: %#lx\n",
cause, opc, badvaddr);
kvm_mips_dump_host_tlbs();
kvm_arch_vcpu_dump_regs(vcpu);
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
} else {
kvm_err("Illegal TLB Mod fault address , cause %#lx, PC: %p, BadVaddr: %#lx\n",
cause, opc, badvaddr);
kvm_mips_dump_host_tlbs();
kvm_arch_vcpu_dump_regs(vcpu);
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
return ret;
}
static int kvm_trap_emul_handle_tlb_st_miss(struct kvm_vcpu *vcpu)
{
struct kvm_run *run = vcpu->run;
uint32_t __user *opc = (uint32_t __user *) vcpu->arch.pc;
unsigned long badvaddr = vcpu->arch.host_cp0_badvaddr;
unsigned long cause = vcpu->arch.host_cp0_cause;
enum emulation_result er = EMULATE_DONE;
int ret = RESUME_GUEST;
if (((badvaddr & PAGE_MASK) == KVM_GUEST_COMMPAGE_ADDR)
&& KVM_GUEST_KERNEL_MODE(vcpu)) {
if (kvm_mips_handle_commpage_tlb_fault(badvaddr, vcpu) < 0) {
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
} else if (KVM_GUEST_KSEGX(badvaddr) < KVM_GUEST_KSEG0
|| KVM_GUEST_KSEGX(badvaddr) == KVM_GUEST_KSEG23) {
kvm_debug("USER ADDR TLB LD fault: cause %#lx, PC: %p, BadVaddr: %#lx\n",
cause, opc, badvaddr);
er = kvm_mips_handle_tlbmiss(cause, opc, run, vcpu);
if (er == EMULATE_DONE)
ret = RESUME_GUEST;
else {
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
} else if (KVM_GUEST_KSEGX(badvaddr) == KVM_GUEST_KSEG0) {
/*
* All KSEG0 faults are handled by KVM, as the guest kernel does
* not expect to ever get them
*/
if (kvm_mips_handle_kseg0_tlb_fault
(vcpu->arch.host_cp0_badvaddr, vcpu) < 0) {
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
} else {
kvm_err("Illegal TLB LD fault address , cause %#lx, PC: %p, BadVaddr: %#lx\n",
cause, opc, badvaddr);
kvm_mips_dump_host_tlbs();
kvm_arch_vcpu_dump_regs(vcpu);
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
return ret;
}
static int kvm_trap_emul_handle_tlb_ld_miss(struct kvm_vcpu *vcpu)
{
struct kvm_run *run = vcpu->run;
uint32_t __user *opc = (uint32_t __user *) vcpu->arch.pc;
unsigned long badvaddr = vcpu->arch.host_cp0_badvaddr;
unsigned long cause = vcpu->arch.host_cp0_cause;
enum emulation_result er = EMULATE_DONE;
int ret = RESUME_GUEST;
if (((badvaddr & PAGE_MASK) == KVM_GUEST_COMMPAGE_ADDR)
&& KVM_GUEST_KERNEL_MODE(vcpu)) {
if (kvm_mips_handle_commpage_tlb_fault(badvaddr, vcpu) < 0) {
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
} else if (KVM_GUEST_KSEGX(badvaddr) < KVM_GUEST_KSEG0
|| KVM_GUEST_KSEGX(badvaddr) == KVM_GUEST_KSEG23) {
kvm_debug("USER ADDR TLB ST fault: PC: %#lx, BadVaddr: %#lx\n",
vcpu->arch.pc, badvaddr);
/*
* User Address (UA) fault, this could happen if
* (1) TLB entry not present/valid in both Guest and shadow host
* TLBs, in this case we pass on the fault to the guest
* kernel and let it handle it.
* (2) TLB entry is present in the Guest TLB but not in the
* shadow, in this case we inject the TLB from the Guest TLB
* into the shadow host TLB
*/
er = kvm_mips_handle_tlbmiss(cause, opc, run, vcpu);
if (er == EMULATE_DONE)
ret = RESUME_GUEST;
else {
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
} else if (KVM_GUEST_KSEGX(badvaddr) == KVM_GUEST_KSEG0) {
if (kvm_mips_handle_kseg0_tlb_fault
(vcpu->arch.host_cp0_badvaddr, vcpu) < 0) {
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
} else {
kvm_err("Illegal TLB ST fault address , cause %#lx, PC: %p, BadVaddr: %#lx\n",
cause, opc, badvaddr);
kvm_mips_dump_host_tlbs();
kvm_arch_vcpu_dump_regs(vcpu);
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
return ret;
}
static int kvm_trap_emul_handle_addr_err_st(struct kvm_vcpu *vcpu)
{
struct kvm_run *run = vcpu->run;
uint32_t __user *opc = (uint32_t __user *) vcpu->arch.pc;
unsigned long badvaddr = vcpu->arch.host_cp0_badvaddr;
unsigned long cause = vcpu->arch.host_cp0_cause;
enum emulation_result er = EMULATE_DONE;
int ret = RESUME_GUEST;
if (KVM_GUEST_KERNEL_MODE(vcpu)
&& (KSEGX(badvaddr) == CKSEG0 || KSEGX(badvaddr) == CKSEG1)) {
kvm_debug("Emulate Store to MMIO space\n");
er = kvm_mips_emulate_inst(cause, opc, run, vcpu);
if (er == EMULATE_FAIL) {
kvm_err("Emulate Store to MMIO space failed\n");
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
} else {
run->exit_reason = KVM_EXIT_MMIO;
ret = RESUME_HOST;
}
} else {
kvm_err("Address Error (STORE): cause %#lx, PC: %p, BadVaddr: %#lx\n",
cause, opc, badvaddr);
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
return ret;
}
static int kvm_trap_emul_handle_addr_err_ld(struct kvm_vcpu *vcpu)
{
struct kvm_run *run = vcpu->run;
uint32_t __user *opc = (uint32_t __user *) vcpu->arch.pc;
unsigned long badvaddr = vcpu->arch.host_cp0_badvaddr;
unsigned long cause = vcpu->arch.host_cp0_cause;
enum emulation_result er = EMULATE_DONE;
int ret = RESUME_GUEST;
if (KSEGX(badvaddr) == CKSEG0 || KSEGX(badvaddr) == CKSEG1) {
kvm_debug("Emulate Load from MMIO space @ %#lx\n", badvaddr);
er = kvm_mips_emulate_inst(cause, opc, run, vcpu);
if (er == EMULATE_FAIL) {
kvm_err("Emulate Load from MMIO space failed\n");
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
} else {
run->exit_reason = KVM_EXIT_MMIO;
ret = RESUME_HOST;
}
} else {
kvm_err("Address Error (LOAD): cause %#lx, PC: %p, BadVaddr: %#lx\n",
cause, opc, badvaddr);
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
er = EMULATE_FAIL;
}
return ret;
}
static int kvm_trap_emul_handle_syscall(struct kvm_vcpu *vcpu)
{
struct kvm_run *run = vcpu->run;
uint32_t __user *opc = (uint32_t __user *) vcpu->arch.pc;
unsigned long cause = vcpu->arch.host_cp0_cause;
enum emulation_result er = EMULATE_DONE;
int ret = RESUME_GUEST;
er = kvm_mips_emulate_syscall(cause, opc, run, vcpu);
if (er == EMULATE_DONE)
ret = RESUME_GUEST;
else {
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
return ret;
}
static int kvm_trap_emul_handle_res_inst(struct kvm_vcpu *vcpu)
{
struct kvm_run *run = vcpu->run;
uint32_t __user *opc = (uint32_t __user *) vcpu->arch.pc;
unsigned long cause = vcpu->arch.host_cp0_cause;
enum emulation_result er = EMULATE_DONE;
int ret = RESUME_GUEST;
er = kvm_mips_handle_ri(cause, opc, run, vcpu);
if (er == EMULATE_DONE)
ret = RESUME_GUEST;
else {
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
return ret;
}
static int kvm_trap_emul_handle_break(struct kvm_vcpu *vcpu)
{
struct kvm_run *run = vcpu->run;
uint32_t __user *opc = (uint32_t __user *) vcpu->arch.pc;
unsigned long cause = vcpu->arch.host_cp0_cause;
enum emulation_result er = EMULATE_DONE;
int ret = RESUME_GUEST;
er = kvm_mips_emulate_bp_exc(cause, opc, run, vcpu);
if (er == EMULATE_DONE)
ret = RESUME_GUEST;
else {
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
return ret;
}
static int kvm_trap_emul_handle_trap(struct kvm_vcpu *vcpu)
{
struct kvm_run *run = vcpu->run;
uint32_t __user *opc = (uint32_t __user *)vcpu->arch.pc;
unsigned long cause = vcpu->arch.host_cp0_cause;
enum emulation_result er = EMULATE_DONE;
int ret = RESUME_GUEST;
er = kvm_mips_emulate_trap_exc(cause, opc, run, vcpu);
if (er == EMULATE_DONE) {
ret = RESUME_GUEST;
} else {
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
}
return ret;
}
static int kvm_trap_emul_handle_msa_disabled(struct kvm_vcpu *vcpu)
{
struct kvm_run *run = vcpu->run;
uint32_t __user *opc = (uint32_t __user *) vcpu->arch.pc;
unsigned long cause = vcpu->arch.host_cp0_cause;
enum emulation_result er = EMULATE_DONE;
int ret = RESUME_GUEST;
/* No MSA supported in guest, guest reserved instruction exception */
er = kvm_mips_emulate_ri_exc(cause, opc, run, vcpu);
switch (er) {
case EMULATE_DONE:
ret = RESUME_GUEST;
break;
case EMULATE_FAIL:
run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
ret = RESUME_HOST;
break;
default:
BUG();
}
return ret;
}
static int kvm_trap_emul_vm_init(struct kvm *kvm)
{
return 0;
}
static int kvm_trap_emul_vcpu_init(struct kvm_vcpu *vcpu)
{
return 0;
}
static int kvm_trap_emul_vcpu_setup(struct kvm_vcpu *vcpu)
{
struct mips_coproc *cop0 = vcpu->arch.cop0;
uint32_t config1;
int vcpu_id = vcpu->vcpu_id;
/*
* Arch specific stuff, set up config registers properly so that the
* guest will come up as expected, for now we simulate a MIPS 24kc
*/
kvm_write_c0_guest_prid(cop0, 0x00019300);
/* Have config1, Cacheable, noncoherent, write-back, write allocate */
kvm_write_c0_guest_config(cop0, MIPS_CONF_M | (0x3 << CP0C0_K0) |
(0x1 << CP0C0_AR) |
(MMU_TYPE_R4000 << CP0C0_MT));
/* Read the cache characteristics from the host Config1 Register */
config1 = (read_c0_config1() & ~0x7f);
/* Set up MMU size */
config1 &= ~(0x3f << 25);
config1 |= ((KVM_MIPS_GUEST_TLB_SIZE - 1) << 25);
/* We unset some bits that we aren't emulating */
config1 &=
~((1 << CP0C1_C2) | (1 << CP0C1_MD) | (1 << CP0C1_PC) |
(1 << CP0C1_WR) | (1 << CP0C1_CA));
kvm_write_c0_guest_config1(cop0, config1);
/* Have config3, no tertiary/secondary caches implemented */
kvm_write_c0_guest_config2(cop0, MIPS_CONF_M);
/* MIPS_CONF_M | (read_c0_config2() & 0xfff) */
/* Have config4, UserLocal */
kvm_write_c0_guest_config3(cop0, MIPS_CONF_M | MIPS_CONF3_ULRI);
/* Have config5 */
kvm_write_c0_guest_config4(cop0, MIPS_CONF_M);
/* No config6 */
kvm_write_c0_guest_config5(cop0, 0);
/* Set Wait IE/IXMT Ignore in Config7, IAR, AR */
kvm_write_c0_guest_config7(cop0, (MIPS_CONF7_WII) | (1 << 10));
/*
* Setup IntCtl defaults, compatibilty mode for timer interrupts (HW5)
*/
kvm_write_c0_guest_intctl(cop0, 0xFC000000);
/* Put in vcpu id as CPUNum into Ebase Reg to handle SMP Guests */
kvm_write_c0_guest_ebase(cop0, KVM_GUEST_KSEG0 | (vcpu_id & 0xFF));
return 0;
}
static int kvm_trap_emul_get_one_reg(struct kvm_vcpu *vcpu,
const struct kvm_one_reg *reg,
s64 *v)
{
switch (reg->id) {
case KVM_REG_MIPS_CP0_COUNT:
MIPS: KVM: Rewrite count/compare timer emulation Previously the emulation of the CPU timer was just enough to get a Linux guest running but some shortcuts were taken: - The guest timer interrupt was hard coded to always happen every 10 ms rather than being timed to when CP0_Count would match CP0_Compare. - The guest's CP0_Count register was based on the host's CP0_Count register. This isn't very portable and fails on cores without a CP_Count register implemented such as Ingenic XBurst. It also meant that the guest's CP0_Cause.DC bit to disable the CP0_Count register took no effect. - The guest's CP0_Count register was emulated by just dividing the host's CP0_Count register by 4. This resulted in continuity problems when used as a clock source, since when the host CP0_Count overflows from 0x7fffffff to 0x80000000, the guest CP0_Count transitions discontinuously from 0x1fffffff to 0xe0000000. Therefore rewrite & fix emulation of the guest timer based on the monotonic kernel time (i.e. ktime_get()). Internally a 32-bit count_bias value is added to the frequency scaled nanosecond monotonic time to get the guest's CP0_Count. The frequency of the timer is initialised to 100MHz and cannot yet be changed, but a later patch will allow the frequency to be configured via the KVM_{GET,SET}_ONE_REG ioctl interface. The timer can now be stopped via the CP0_Cause.DC bit (by the guest or via the KVM_SET_ONE_REG ioctl interface), at which point the current CP0_Count is stored and can be read directly. When it is restarted the bias is recalculated such that the CP0_Count value is continuous. Due to the nature of hrtimer interrupts any read of the guest's CP0_Count register while it is running triggers a check for whether the hrtimer has expired, so that the guest/userland cannot observe the CP0_Count passing CP0_Compare without queuing a timer interrupt. This is also taken advantage of when stopping the timer to ensure that a pending timer interrupt is queued. This replaces the implementation of: - Guest read of CP0_Count - Guest write of CP0_Count - Guest write of CP0_Compare - Guest write of CP0_Cause - Guest read of HWR 2 (CC) with RDHWR - Host read of CP0_Count via KVM_GET_ONE_REG ioctl interface - Host write of CP0_Count via KVM_SET_ONE_REG ioctl interface - Host write of CP0_Compare via KVM_SET_ONE_REG ioctl interface - Host write of CP0_Cause via KVM_SET_ONE_REG ioctl interface Signed-off-by: James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Gleb Natapov <gleb@kernel.org> Cc: kvm@vger.kernel.org Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: Sanjay Lal <sanjayl@kymasys.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-05-29 17:16:35 +08:00
*v = kvm_mips_read_count(vcpu);
break;
MIPS: KVM: Add master disable count interface Expose two new virtual registers to userland via the KVM_{GET,SET}_ONE_REG ioctls. KVM_REG_MIPS_COUNT_CTL is for timer configuration fields and just contains a master disable count bit. This can be used by userland to freeze the timer in order to read a consistent state from the timer count value and timer interrupt pending bit. This cannot be done with the CP0_Cause.DC bit because the timer interrupt pending bit (TI) is also in CP0_Cause so it would be impossible to stop the timer without also risking a race with an hrtimer interrupt and having to explicitly check whether an interrupt should have occurred. When the timer is re-enabled it resumes without losing time, i.e. the CP0_Count value jumps to what it would have been had the timer not been disabled, which would also be impossible to do from userland with CP0_Cause.DC. The timer interrupt also cannot be lost, i.e. if a timer interrupt would have occurred had the timer not been disabled it is queued when the timer is re-enabled. This works by storing the nanosecond monotonic time when the master disable is set, and using it for various operations instead of the current monotonic time (e.g. when recalculating the bias when the CP0_Count is set), until the master disable is cleared again, i.e. the timer state is read/written as it would have been at that time. This state is exposed to userland via the read-only KVM_REG_MIPS_COUNT_RESUME virtual register so that userland can determine the exact time the master disable took effect. This should allow userland to atomically save the state of the timer, and later restore it. Signed-off-by: James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Gleb Natapov <gleb@kernel.org> Cc: kvm@vger.kernel.org Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: David Daney <david.daney@cavium.com> Cc: Sanjay Lal <sanjayl@kymasys.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-05-29 17:16:37 +08:00
case KVM_REG_MIPS_COUNT_CTL:
*v = vcpu->arch.count_ctl;
break;
case KVM_REG_MIPS_COUNT_RESUME:
*v = ktime_to_ns(vcpu->arch.count_resume);
break;
case KVM_REG_MIPS_COUNT_HZ:
*v = vcpu->arch.count_hz;
break;
default:
return -EINVAL;
}
return 0;
}
static int kvm_trap_emul_set_one_reg(struct kvm_vcpu *vcpu,
const struct kvm_one_reg *reg,
s64 v)
{
struct mips_coproc *cop0 = vcpu->arch.cop0;
MIPS: KVM: Add master disable count interface Expose two new virtual registers to userland via the KVM_{GET,SET}_ONE_REG ioctls. KVM_REG_MIPS_COUNT_CTL is for timer configuration fields and just contains a master disable count bit. This can be used by userland to freeze the timer in order to read a consistent state from the timer count value and timer interrupt pending bit. This cannot be done with the CP0_Cause.DC bit because the timer interrupt pending bit (TI) is also in CP0_Cause so it would be impossible to stop the timer without also risking a race with an hrtimer interrupt and having to explicitly check whether an interrupt should have occurred. When the timer is re-enabled it resumes without losing time, i.e. the CP0_Count value jumps to what it would have been had the timer not been disabled, which would also be impossible to do from userland with CP0_Cause.DC. The timer interrupt also cannot be lost, i.e. if a timer interrupt would have occurred had the timer not been disabled it is queued when the timer is re-enabled. This works by storing the nanosecond monotonic time when the master disable is set, and using it for various operations instead of the current monotonic time (e.g. when recalculating the bias when the CP0_Count is set), until the master disable is cleared again, i.e. the timer state is read/written as it would have been at that time. This state is exposed to userland via the read-only KVM_REG_MIPS_COUNT_RESUME virtual register so that userland can determine the exact time the master disable took effect. This should allow userland to atomically save the state of the timer, and later restore it. Signed-off-by: James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Gleb Natapov <gleb@kernel.org> Cc: kvm@vger.kernel.org Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: David Daney <david.daney@cavium.com> Cc: Sanjay Lal <sanjayl@kymasys.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-05-29 17:16:37 +08:00
int ret = 0;
unsigned int cur, change;
switch (reg->id) {
case KVM_REG_MIPS_CP0_COUNT:
MIPS: KVM: Rewrite count/compare timer emulation Previously the emulation of the CPU timer was just enough to get a Linux guest running but some shortcuts were taken: - The guest timer interrupt was hard coded to always happen every 10 ms rather than being timed to when CP0_Count would match CP0_Compare. - The guest's CP0_Count register was based on the host's CP0_Count register. This isn't very portable and fails on cores without a CP_Count register implemented such as Ingenic XBurst. It also meant that the guest's CP0_Cause.DC bit to disable the CP0_Count register took no effect. - The guest's CP0_Count register was emulated by just dividing the host's CP0_Count register by 4. This resulted in continuity problems when used as a clock source, since when the host CP0_Count overflows from 0x7fffffff to 0x80000000, the guest CP0_Count transitions discontinuously from 0x1fffffff to 0xe0000000. Therefore rewrite & fix emulation of the guest timer based on the monotonic kernel time (i.e. ktime_get()). Internally a 32-bit count_bias value is added to the frequency scaled nanosecond monotonic time to get the guest's CP0_Count. The frequency of the timer is initialised to 100MHz and cannot yet be changed, but a later patch will allow the frequency to be configured via the KVM_{GET,SET}_ONE_REG ioctl interface. The timer can now be stopped via the CP0_Cause.DC bit (by the guest or via the KVM_SET_ONE_REG ioctl interface), at which point the current CP0_Count is stored and can be read directly. When it is restarted the bias is recalculated such that the CP0_Count value is continuous. Due to the nature of hrtimer interrupts any read of the guest's CP0_Count register while it is running triggers a check for whether the hrtimer has expired, so that the guest/userland cannot observe the CP0_Count passing CP0_Compare without queuing a timer interrupt. This is also taken advantage of when stopping the timer to ensure that a pending timer interrupt is queued. This replaces the implementation of: - Guest read of CP0_Count - Guest write of CP0_Count - Guest write of CP0_Compare - Guest write of CP0_Cause - Guest read of HWR 2 (CC) with RDHWR - Host read of CP0_Count via KVM_GET_ONE_REG ioctl interface - Host write of CP0_Count via KVM_SET_ONE_REG ioctl interface - Host write of CP0_Compare via KVM_SET_ONE_REG ioctl interface - Host write of CP0_Cause via KVM_SET_ONE_REG ioctl interface Signed-off-by: James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Gleb Natapov <gleb@kernel.org> Cc: kvm@vger.kernel.org Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: Sanjay Lal <sanjayl@kymasys.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-05-29 17:16:35 +08:00
kvm_mips_write_count(vcpu, v);
break;
case KVM_REG_MIPS_CP0_COMPARE:
MIPS: KVM: Rewrite count/compare timer emulation Previously the emulation of the CPU timer was just enough to get a Linux guest running but some shortcuts were taken: - The guest timer interrupt was hard coded to always happen every 10 ms rather than being timed to when CP0_Count would match CP0_Compare. - The guest's CP0_Count register was based on the host's CP0_Count register. This isn't very portable and fails on cores without a CP_Count register implemented such as Ingenic XBurst. It also meant that the guest's CP0_Cause.DC bit to disable the CP0_Count register took no effect. - The guest's CP0_Count register was emulated by just dividing the host's CP0_Count register by 4. This resulted in continuity problems when used as a clock source, since when the host CP0_Count overflows from 0x7fffffff to 0x80000000, the guest CP0_Count transitions discontinuously from 0x1fffffff to 0xe0000000. Therefore rewrite & fix emulation of the guest timer based on the monotonic kernel time (i.e. ktime_get()). Internally a 32-bit count_bias value is added to the frequency scaled nanosecond monotonic time to get the guest's CP0_Count. The frequency of the timer is initialised to 100MHz and cannot yet be changed, but a later patch will allow the frequency to be configured via the KVM_{GET,SET}_ONE_REG ioctl interface. The timer can now be stopped via the CP0_Cause.DC bit (by the guest or via the KVM_SET_ONE_REG ioctl interface), at which point the current CP0_Count is stored and can be read directly. When it is restarted the bias is recalculated such that the CP0_Count value is continuous. Due to the nature of hrtimer interrupts any read of the guest's CP0_Count register while it is running triggers a check for whether the hrtimer has expired, so that the guest/userland cannot observe the CP0_Count passing CP0_Compare without queuing a timer interrupt. This is also taken advantage of when stopping the timer to ensure that a pending timer interrupt is queued. This replaces the implementation of: - Guest read of CP0_Count - Guest write of CP0_Count - Guest write of CP0_Compare - Guest write of CP0_Cause - Guest read of HWR 2 (CC) with RDHWR - Host read of CP0_Count via KVM_GET_ONE_REG ioctl interface - Host write of CP0_Count via KVM_SET_ONE_REG ioctl interface - Host write of CP0_Compare via KVM_SET_ONE_REG ioctl interface - Host write of CP0_Cause via KVM_SET_ONE_REG ioctl interface Signed-off-by: James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Gleb Natapov <gleb@kernel.org> Cc: kvm@vger.kernel.org Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: Sanjay Lal <sanjayl@kymasys.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-05-29 17:16:35 +08:00
kvm_mips_write_compare(vcpu, v);
break;
case KVM_REG_MIPS_CP0_CAUSE:
/*
* If the timer is stopped or started (DC bit) it must look
* atomic with changes to the interrupt pending bits (TI, IRQ5).
* A timer interrupt should not happen in between.
*/
if ((kvm_read_c0_guest_cause(cop0) ^ v) & CAUSEF_DC) {
if (v & CAUSEF_DC) {
/* disable timer first */
kvm_mips_count_disable_cause(vcpu);
kvm_change_c0_guest_cause(cop0, ~CAUSEF_DC, v);
} else {
/* enable timer last */
kvm_change_c0_guest_cause(cop0, ~CAUSEF_DC, v);
kvm_mips_count_enable_cause(vcpu);
}
} else {
kvm_write_c0_guest_cause(cop0, v);
}
break;
case KVM_REG_MIPS_CP0_CONFIG:
/* read-only for now */
break;
case KVM_REG_MIPS_CP0_CONFIG1:
cur = kvm_read_c0_guest_config1(cop0);
change = (cur ^ v) & kvm_mips_config1_wrmask(vcpu);
if (change) {
v = cur ^ change;
kvm_write_c0_guest_config1(cop0, v);
}
break;
case KVM_REG_MIPS_CP0_CONFIG2:
/* read-only for now */
break;
case KVM_REG_MIPS_CP0_CONFIG3:
cur = kvm_read_c0_guest_config3(cop0);
change = (cur ^ v) & kvm_mips_config3_wrmask(vcpu);
if (change) {
v = cur ^ change;
kvm_write_c0_guest_config3(cop0, v);
}
break;
case KVM_REG_MIPS_CP0_CONFIG4:
cur = kvm_read_c0_guest_config4(cop0);
change = (cur ^ v) & kvm_mips_config4_wrmask(vcpu);
if (change) {
v = cur ^ change;
kvm_write_c0_guest_config4(cop0, v);
}
break;
case KVM_REG_MIPS_CP0_CONFIG5:
cur = kvm_read_c0_guest_config5(cop0);
change = (cur ^ v) & kvm_mips_config5_wrmask(vcpu);
if (change) {
v = cur ^ change;
kvm_write_c0_guest_config5(cop0, v);
}
break;
MIPS: KVM: Add master disable count interface Expose two new virtual registers to userland via the KVM_{GET,SET}_ONE_REG ioctls. KVM_REG_MIPS_COUNT_CTL is for timer configuration fields and just contains a master disable count bit. This can be used by userland to freeze the timer in order to read a consistent state from the timer count value and timer interrupt pending bit. This cannot be done with the CP0_Cause.DC bit because the timer interrupt pending bit (TI) is also in CP0_Cause so it would be impossible to stop the timer without also risking a race with an hrtimer interrupt and having to explicitly check whether an interrupt should have occurred. When the timer is re-enabled it resumes without losing time, i.e. the CP0_Count value jumps to what it would have been had the timer not been disabled, which would also be impossible to do from userland with CP0_Cause.DC. The timer interrupt also cannot be lost, i.e. if a timer interrupt would have occurred had the timer not been disabled it is queued when the timer is re-enabled. This works by storing the nanosecond monotonic time when the master disable is set, and using it for various operations instead of the current monotonic time (e.g. when recalculating the bias when the CP0_Count is set), until the master disable is cleared again, i.e. the timer state is read/written as it would have been at that time. This state is exposed to userland via the read-only KVM_REG_MIPS_COUNT_RESUME virtual register so that userland can determine the exact time the master disable took effect. This should allow userland to atomically save the state of the timer, and later restore it. Signed-off-by: James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Gleb Natapov <gleb@kernel.org> Cc: kvm@vger.kernel.org Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: David Daney <david.daney@cavium.com> Cc: Sanjay Lal <sanjayl@kymasys.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-05-29 17:16:37 +08:00
case KVM_REG_MIPS_COUNT_CTL:
ret = kvm_mips_set_count_ctl(vcpu, v);
break;
case KVM_REG_MIPS_COUNT_RESUME:
ret = kvm_mips_set_count_resume(vcpu, v);
break;
case KVM_REG_MIPS_COUNT_HZ:
ret = kvm_mips_set_count_hz(vcpu, v);
break;
default:
return -EINVAL;
}
MIPS: KVM: Add master disable count interface Expose two new virtual registers to userland via the KVM_{GET,SET}_ONE_REG ioctls. KVM_REG_MIPS_COUNT_CTL is for timer configuration fields and just contains a master disable count bit. This can be used by userland to freeze the timer in order to read a consistent state from the timer count value and timer interrupt pending bit. This cannot be done with the CP0_Cause.DC bit because the timer interrupt pending bit (TI) is also in CP0_Cause so it would be impossible to stop the timer without also risking a race with an hrtimer interrupt and having to explicitly check whether an interrupt should have occurred. When the timer is re-enabled it resumes without losing time, i.e. the CP0_Count value jumps to what it would have been had the timer not been disabled, which would also be impossible to do from userland with CP0_Cause.DC. The timer interrupt also cannot be lost, i.e. if a timer interrupt would have occurred had the timer not been disabled it is queued when the timer is re-enabled. This works by storing the nanosecond monotonic time when the master disable is set, and using it for various operations instead of the current monotonic time (e.g. when recalculating the bias when the CP0_Count is set), until the master disable is cleared again, i.e. the timer state is read/written as it would have been at that time. This state is exposed to userland via the read-only KVM_REG_MIPS_COUNT_RESUME virtual register so that userland can determine the exact time the master disable took effect. This should allow userland to atomically save the state of the timer, and later restore it. Signed-off-by: James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Gleb Natapov <gleb@kernel.org> Cc: kvm@vger.kernel.org Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: David Daney <david.daney@cavium.com> Cc: Sanjay Lal <sanjayl@kymasys.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-05-29 17:16:37 +08:00
return ret;
}
static struct kvm_mips_callbacks kvm_trap_emul_callbacks = {
/* exit handlers */
.handle_cop_unusable = kvm_trap_emul_handle_cop_unusable,
.handle_tlb_mod = kvm_trap_emul_handle_tlb_mod,
.handle_tlb_st_miss = kvm_trap_emul_handle_tlb_st_miss,
.handle_tlb_ld_miss = kvm_trap_emul_handle_tlb_ld_miss,
.handle_addr_err_st = kvm_trap_emul_handle_addr_err_st,
.handle_addr_err_ld = kvm_trap_emul_handle_addr_err_ld,
.handle_syscall = kvm_trap_emul_handle_syscall,
.handle_res_inst = kvm_trap_emul_handle_res_inst,
.handle_break = kvm_trap_emul_handle_break,
.handle_trap = kvm_trap_emul_handle_trap,
.handle_msa_disabled = kvm_trap_emul_handle_msa_disabled,
.vm_init = kvm_trap_emul_vm_init,
.vcpu_init = kvm_trap_emul_vcpu_init,
.vcpu_setup = kvm_trap_emul_vcpu_setup,
.gva_to_gpa = kvm_trap_emul_gva_to_gpa_cb,
.queue_timer_int = kvm_mips_queue_timer_int_cb,
.dequeue_timer_int = kvm_mips_dequeue_timer_int_cb,
.queue_io_int = kvm_mips_queue_io_int_cb,
.dequeue_io_int = kvm_mips_dequeue_io_int_cb,
.irq_deliver = kvm_mips_irq_deliver_cb,
.irq_clear = kvm_mips_irq_clear_cb,
.get_one_reg = kvm_trap_emul_get_one_reg,
.set_one_reg = kvm_trap_emul_set_one_reg,
};
int kvm_mips_emulation_init(struct kvm_mips_callbacks **install_callbacks)
{
*install_callbacks = &kvm_trap_emul_callbacks;
return 0;
}