staticint kvm_xen_shared_info_init(struct kvm *kvm)
{ struct gfn_to_pfn_cache *gpc = &kvm->arch.xen.shinfo_cache; struct pvclock_wall_clock *wc;
u32 *wc_sec_hi;
u32 wc_version;
u64 wall_nsec; int ret = 0; int idx = srcu_read_lock(&kvm->srcu);
read_lock_irq(&gpc->lock); while (!kvm_gpc_check(gpc, PAGE_SIZE)) {
read_unlock_irq(&gpc->lock);
ret = kvm_gpc_refresh(gpc, PAGE_SIZE); if (ret) goto out;
read_lock_irq(&gpc->lock);
}
/* * This code mirrors kvm_write_wall_clock() except that it writes * directly through the pfn cache and doesn't mark the page dirty.
*/
wall_nsec = kvm_get_wall_clock_epoch(kvm);
/* * Sanity check TSC shift+multiplier to verify the guest's view of time * is more or less consistent.
*/ if (hv_clock->tsc_shift != vcpu->arch.pvclock_tsc_shift ||
hv_clock->tsc_to_system_mul != vcpu->arch.pvclock_tsc_mul) return -EINVAL;
/* * The guest provides the requested timeout in absolute nanoseconds * of the KVM clock — as *it* sees it, based on the scaled TSC and * the pvclock information provided by KVM. * * The kernel doesn't support hrtimers based on CLOCK_MONOTONIC_RAW * so use CLOCK_MONOTONIC. In the timescales covered by timers, the * difference won't matter much as there is no cumulative effect. * * Calculate the time for some arbitrary point in time around "now" * in terms of both kvmclock and CLOCK_MONOTONIC. Calculate the * delta between the kvmclock "now" value and the guest's requested * timeout, apply the "Linux workaround" described below, and add * the resulting delta to the CLOCK_MONOTONIC "now" value, to get * the absolute CLOCK_MONOTONIC time at which the timer should * fire.
*/ do { struct pvclock_vcpu_time_info hv_clock;
uint64_t host_tsc, guest_tsc;
if (!static_cpu_has(X86_FEATURE_CONSTANT_TSC) ||
!vcpu->kvm->arch.use_master_clock) break;
/* * If both Xen PV clocks are active, arbitrarily try to use the * compat clock first, but also try to use the non-compat clock * if the compat clock is unusable. The two PV clocks hold the * same information, but it's possible one (or both) is stale * and/or currently unreachable.
*/ if (xen->vcpu_info_cache.active)
r = xen_get_guest_pvclock(vcpu, &hv_clock, &xen->vcpu_info_cache,
offsetof(struct compat_vcpu_info, time)); if (r && xen->vcpu_time_info_cache.active)
r = xen_get_guest_pvclock(vcpu, &hv_clock, &xen->vcpu_time_info_cache, 0); if (r) break;
if (!IS_ENABLED(CONFIG_64BIT) ||
!kvm_get_monotonic_and_clockread(&kernel_now, &host_tsc)) { /* * Don't fall back to get_kvmclock_ns() because it's * broken; it has a systemic error in its results * because it scales directly from host TSC to * nanoseconds, and doesn't scale first to guest TSC * and *then* to nanoseconds as the guest does. * * There is a small error introduced here because time * continues to elapse between the ktime_get() and the * subsequent rdtsc(). But not the systemic drift due * to get_kvmclock_ns().
*/
kernel_now = ktime_get(); /* This is CLOCK_MONOTONIC */
host_tsc = rdtsc();
}
/* Calculate the guest kvmclock as the guest would do it. */
guest_tsc = kvm_read_l1_tsc(vcpu, host_tsc);
guest_now = __pvclock_read_cycles(&hv_clock, guest_tsc);
} while (0);
if (r) { /* * Without CONSTANT_TSC, get_kvmclock_ns() is the only option. * * Also if the guest PV clock hasn't been set up yet, as is * likely to be the case during migration when the vCPU has * not been run yet. It would be possible to calculate the * scaling factors properly in that case but there's not much * point in doing so. The get_kvmclock_ns() drift accumulates * over time, so it's OK to use it at startup. Besides, on * migration there's going to be a little bit of skew in the * precise moment at which timers fire anyway. Often they'll * be in the "past" by the time the VM is running again after * migration.
*/
guest_now = get_kvmclock_ns(vcpu->kvm);
kernel_now = ktime_get();
}
delta = guest_abs - guest_now;
/* * Xen has a 'Linux workaround' in do_set_timer_op() which checks for * negative absolute timeout values (caused by integer overflow), and * for values about 13 days in the future (2^50ns) which would be * caused by jiffies overflow. For those cases, Xen sets the timeout * 100ms in the future (not *too* soon, since if a guest really did * set a long timeout on purpose we don't want to keep churning CPU * time by waking it up). Emulate Xen's workaround when starting the * timer in response to __HYPERVISOR_set_timer_op.
*/ if (linux_wa &&
unlikely((int64_t)guest_abs < 0 ||
(delta > 0 && (uint32_t) (delta >> 50) != 0))) {
delta = 100 * NSEC_PER_MSEC;
guest_abs = guest_now + delta;
}
/* * Avoid races with the old timer firing. Checking timer_expires * to avoid calling hrtimer_cancel() will only have false positives * so is fine.
*/ if (vcpu->arch.xen.timer_expires)
hrtimer_cancel(&vcpu->arch.xen.timer);
/* * The only difference between 32-bit and 64-bit versions of the * runstate struct is the alignment of uint64_t in 32-bit, which * means that the 64-bit version has an additional 4 bytes of * padding after the first field 'state'. Let's be really really * paranoid about that, and matching it with our internal data * structures that we memcpy into it...
*/
BUILD_BUG_ON(offsetof(struct vcpu_runstate_info, state) != 0);
BUILD_BUG_ON(offsetof(struct compat_vcpu_runstate_info, state) != 0);
BUILD_BUG_ON(sizeof(struct compat_vcpu_runstate_info) != 0x2c); #ifdef CONFIG_X86_64 /* * The 64-bit structure has 4 bytes of padding before 'state_entry_time' * so each subsequent field is shifted by 4, and it's 4 bytes longer.
*/
BUILD_BUG_ON(offsetof(struct vcpu_runstate_info, state_entry_time) !=
offsetof(struct compat_vcpu_runstate_info, state_entry_time) + 4);
BUILD_BUG_ON(offsetof(struct vcpu_runstate_info, time) !=
offsetof(struct compat_vcpu_runstate_info, time) + 4);
BUILD_BUG_ON(sizeof(struct vcpu_runstate_info) != 0x2c + 4); #endif /* * The state field is in the same place at the start of both structs, * and is the same size (int) as vx->current_runstate.
*/
BUILD_BUG_ON(offsetof(struct vcpu_runstate_info, state) !=
offsetof(struct compat_vcpu_runstate_info, state));
BUILD_BUG_ON(sizeof_field(struct vcpu_runstate_info, state) != sizeof(vx->current_runstate));
BUILD_BUG_ON(sizeof_field(struct compat_vcpu_runstate_info, state) != sizeof(vx->current_runstate));
/* * The state_entry_time field is 64 bits in both versions, and the * XEN_RUNSTATE_UPDATE flag is in the top bit, which given that x86 * is little-endian means that it's in the last *byte* of the word. * That detail is important later.
*/
BUILD_BUG_ON(sizeof_field(struct vcpu_runstate_info, state_entry_time) != sizeof(uint64_t));
BUILD_BUG_ON(sizeof_field(struct compat_vcpu_runstate_info, state_entry_time) != sizeof(uint64_t));
BUILD_BUG_ON((XEN_RUNSTATE_UPDATE >> 56) != 0x80);
/* * The time array is four 64-bit quantities in both versions, matching * the vx->runstate_times and immediately following state_entry_time.
*/
BUILD_BUG_ON(offsetof(struct vcpu_runstate_info, state_entry_time) !=
offsetof(struct vcpu_runstate_info, time) - sizeof(uint64_t));
BUILD_BUG_ON(offsetof(struct compat_vcpu_runstate_info, state_entry_time) !=
offsetof(struct compat_vcpu_runstate_info, time) - sizeof(uint64_t));
BUILD_BUG_ON(sizeof_field(struct vcpu_runstate_info, time) !=
sizeof_field(struct compat_vcpu_runstate_info, time));
BUILD_BUG_ON(sizeof_field(struct vcpu_runstate_info, time) != sizeof(vx->runstate_times));
/* * There are basically no alignment constraints. The guest can set it * up so it crosses from one page to the next, and at arbitrary byte * alignment (and the 32-bit ABI doesn't align the 64-bit integers * anyway, even if the overall struct had been 64-bit aligned).
*/ if ((gpc1->gpa & ~PAGE_MASK) + user_len >= PAGE_SIZE) {
user_len1 = PAGE_SIZE - (gpc1->gpa & ~PAGE_MASK);
user_len2 = user_len - user_len1;
} else {
user_len1 = user_len;
user_len2 = 0;
}
BUG_ON(user_len1 + user_len2 != user_len);
retry: /* * Attempt to obtain the GPC lock on *both* (if there are two) * gfn_to_pfn caches that cover the region.
*/ if (atomic) {
local_irq_save(flags); if (!read_trylock(&gpc1->lock)) {
local_irq_restore(flags); return;
}
} else {
read_lock_irqsave(&gpc1->lock, flags);
} while (!kvm_gpc_check(gpc1, user_len1)) {
read_unlock_irqrestore(&gpc1->lock, flags);
/* When invoked from kvm_sched_out() we cannot sleep */ if (atomic) return;
if (kvm_gpc_refresh(gpc1, user_len1)) return;
read_lock_irqsave(&gpc1->lock, flags);
}
if (likely(!user_len2)) { /* * Set up three pointers directly to the runstate_info * struct in the guest (via the GPC). * * • @rs_state → state field * • @rs_times → state_entry_time field. * • @update_bit → last byte of state_entry_time, which * contains the XEN_RUNSTATE_UPDATE bit.
*/
rs_state = gpc1->khva;
rs_times = gpc1->khva + times_ofs; if (v->kvm->arch.xen.runstate_update_flag)
update_bit = ((void *)(&rs_times[1])) - 1;
} else { /* * The guest's runstate_info is split across two pages and we * need to hold and validate both GPCs simultaneously. We can * declare a lock ordering GPC1 > GPC2 because nothing else * takes them more than one at a time. Set a subclass on the * gpc1 lock to make lockdep shut up about it.
*/
lock_set_subclass(&gpc1->lock.dep_map, 1, _THIS_IP_); if (atomic) { if (!read_trylock(&gpc2->lock)) {
read_unlock_irqrestore(&gpc1->lock, flags); return;
}
} else {
read_lock(&gpc2->lock);
}
if (!kvm_gpc_check(gpc2, user_len2)) {
read_unlock(&gpc2->lock);
read_unlock_irqrestore(&gpc1->lock, flags);
/* When invoked from kvm_sched_out() we cannot sleep */ if (atomic) return;
/* * Use kvm_gpc_activate() here because if the runstate * area was configured in 32-bit mode and only extends * to the second page now because the guest changed to * 64-bit mode, the second GPC won't have been set up.
*/ if (kvm_gpc_activate(gpc2, gpc1->gpa + user_len1,
user_len2)) return;
/* * We dropped the lock on GPC1 so we have to go all the * way back and revalidate that too.
*/ goto retry;
}
/* * In this case, the runstate_info struct will be assembled on * the kernel stack (compat or not as appropriate) and will * be copied to GPC1/GPC2 with a dual memcpy. Set up the three * rs pointers accordingly.
*/
rs_times = &rs.state_entry_time;
/* * The rs_state pointer points to the start of what we'll * copy to the guest, which in the case of a compat guest * is the 32-bit field that the compiler thinks is padding.
*/
rs_state = ((void *)rs_times) - times_ofs;
/* * The update_bit is still directly in the guest memory, * via one GPC or the other.
*/ if (v->kvm->arch.xen.runstate_update_flag) { if (user_len1 >= times_ofs + sizeof(uint64_t))
update_bit = gpc1->khva + times_ofs + sizeof(uint64_t) - 1; else
update_bit = gpc2->khva + times_ofs + sizeof(uint64_t) - 1 - user_len1;
}
#ifdef CONFIG_X86_64 /* * Don't leak kernel memory through the padding in the 64-bit * version of the struct.
*/
memset(&rs, 0, offsetof(struct vcpu_runstate_info, state_entry_time)); #endif
}
/* * First, set the XEN_RUNSTATE_UPDATE bit in the top bit of the * state_entry_time field, directly in the guest. We need to set * that (and write-barrier) before writing to the rest of the * structure, and clear it last. Just as Xen does, we address the * single *byte* in which it resides because it might be in a * different cache line to the rest of the 64-bit word, due to * the (lack of) alignment constraints.
*/
entry_time = vx->runstate_entry_time; if (update_bit) {
entry_time |= XEN_RUNSTATE_UPDATE;
*update_bit = (vx->runstate_entry_time | XEN_RUNSTATE_UPDATE) >> 56;
smp_wmb();
}
/* * Now assemble the actual structure, either on our kernel stack * or directly in the guest according to how the rs_state and * rs_times pointers were set up above.
*/
*rs_state = vx->current_runstate;
rs_times[0] = entry_time;
memcpy(rs_times + 1, vx->runstate_times, sizeof(vx->runstate_times));
/* For the split case, we have to then copy it to the guest. */ if (user_len2) {
memcpy(gpc1->khva, rs_state, user_len1);
memcpy(gpc2->khva, ((void *)rs_state) + user_len1, user_len2);
}
smp_wmb();
/* Finally, clear the XEN_RUNSTATE_UPDATE bit. */ if (update_bit) {
entry_time &= ~XEN_RUNSTATE_UPDATE;
*update_bit = entry_time >> 56;
smp_wmb();
}
if (user_len2) {
kvm_gpc_mark_dirty_in_slot(gpc2);
read_unlock(&gpc2->lock);
}
void kvm_xen_update_runstate(struct kvm_vcpu *v, int state)
{ struct kvm_vcpu_xen *vx = &v->arch.xen;
u64 now = get_kvmclock_ns(v->kvm);
u64 delta_ns = now - vx->runstate_entry_time;
u64 run_delay = current->sched_info.run_delay;
if (unlikely(!vx->runstate_entry_time))
vx->current_runstate = RUNSTATE_offline;
/* * Time waiting for the scheduler isn't "stolen" if the * vCPU wasn't running anyway.
*/ if (vx->current_runstate == RUNSTATE_running) {
u64 steal_ns = run_delay - vx->last_steal;
/* * On event channel delivery, the vcpu_info may not have been accessible. * In that case, there are bits in vcpu->arch.xen.evtchn_pending_sel which * need to be marked into the vcpu_info (and evtchn_upcall_pending set). * Do so now that we can sleep in the context of the vCPU to bring the * page in, and refresh the pfn cache for it.
*/ void kvm_xen_inject_pending_events(struct kvm_vcpu *v)
{ unsignedlong evtchn_pending_sel = READ_ONCE(v->arch.xen.evtchn_pending_sel); struct gfn_to_pfn_cache *gpc = &v->arch.xen.vcpu_info_cache; unsignedlong flags;
if (!evtchn_pending_sel) return;
/* * Yes, this is an open-coded loop. But that's just what put_user() * does anyway. Page it in and retry the instruction. We're just a * little more honest about it.
*/
read_lock_irqsave(&gpc->lock, flags); while (!kvm_gpc_check(gpc, sizeof(struct vcpu_info))) {
read_unlock_irqrestore(&gpc->lock, flags);
if (kvm_gpc_refresh(gpc, sizeof(struct vcpu_info))) return;
read_lock_irqsave(&gpc->lock, flags);
}
/* Now gpc->khva is a valid kernel address for the vcpu_info */ if (IS_ENABLED(CONFIG_64BIT) && v->kvm->arch.xen.long_mode) { struct vcpu_info *vi = gpc->khva;
/* * If the global upcall vector (HVMIRQ_callback_vector) is set and * the vCPU's evtchn_upcall_pending flag is set, the IRQ is pending.
*/
/* No need for compat handling here */
BUILD_BUG_ON(offsetof(struct vcpu_info, evtchn_upcall_pending) !=
offsetof(struct compat_vcpu_info, evtchn_upcall_pending));
BUILD_BUG_ON(sizeof(rc) !=
sizeof_field(struct vcpu_info, evtchn_upcall_pending));
BUILD_BUG_ON(sizeof(rc) !=
sizeof_field(struct compat_vcpu_info, evtchn_upcall_pending));
read_lock_irqsave(&gpc->lock, flags); while (!kvm_gpc_check(gpc, sizeof(struct vcpu_info))) {
read_unlock_irqrestore(&gpc->lock, flags);
/* * This function gets called from kvm_vcpu_block() after setting the * task to TASK_INTERRUPTIBLE, to see if it needs to wake immediately * from a HLT. So we really mustn't sleep. If the page ended up absent * at that point, just return 1 in order to trigger an immediate wake, * and we'll end up getting called again from a context where we *can* * fault in the page and wait for it.
*/ if (in_atomic() || !task_is_running(current)) return 1;
if (kvm_gpc_refresh(gpc, sizeof(struct vcpu_info))) { /* * If this failed, userspace has screwed up the * vcpu_info mapping. No interrupts for you.
*/ return 0;
}
read_lock_irqsave(&gpc->lock, flags);
}
int kvm_xen_hvm_set_attr(struct kvm *kvm, struct kvm_xen_hvm_attr *data)
{ int r = -ENOENT;
switch (data->type) { case KVM_XEN_ATTR_TYPE_LONG_MODE: if (!IS_ENABLED(CONFIG_64BIT) && data->u.long_mode) {
r = -EINVAL;
} else {
mutex_lock(&kvm->arch.xen.xen_lock);
kvm->arch.xen.long_mode = !!data->u.long_mode;
/* * Re-initialize shared_info to put the wallclock in the * correct place. Whilst it's not necessary to do this * unless the mode is actually changed, it does no harm * to make the call anyway.
*/
r = kvm->arch.xen.shinfo_cache.active ?
kvm_xen_shared_info_init(kvm) : 0;
mutex_unlock(&kvm->arch.xen.xen_lock);
} break;
case KVM_XEN_ATTR_TYPE_SHARED_INFO: case KVM_XEN_ATTR_TYPE_SHARED_INFO_HVA: { int idx;
mutex_lock(&kvm->arch.xen.xen_lock);
idx = srcu_read_lock(&kvm->srcu);
if (data->type == KVM_XEN_ATTR_TYPE_SHARED_INFO) {
gfn_t gfn = data->u.shared_info.gfn;
if (gfn == KVM_XEN_INVALID_GFN) {
kvm_gpc_deactivate(&kvm->arch.xen.shinfo_cache);
r = 0;
} else {
r = kvm_gpc_activate(&kvm->arch.xen.shinfo_cache,
gfn_to_gpa(gfn), PAGE_SIZE);
}
} else { void __user * hva = u64_to_user_ptr(data->u.shared_info.hva);
if (!PAGE_ALIGNED(hva)) {
r = -EINVAL;
} elseif (!hva) {
kvm_gpc_deactivate(&kvm->arch.xen.shinfo_cache);
r = 0;
} else {
r = kvm_gpc_activate_hva(&kvm->arch.xen.shinfo_cache,
(unsignedlong)hva, PAGE_SIZE);
}
}
srcu_read_unlock(&kvm->srcu, idx);
if (!r && kvm->arch.xen.shinfo_cache.active)
r = kvm_xen_shared_info_init(kvm);
mutex_unlock(&kvm->arch.xen.xen_lock); break;
} case KVM_XEN_ATTR_TYPE_UPCALL_VECTOR: if (data->u.vector && data->u.vector < 0x10)
r = -EINVAL; else {
mutex_lock(&kvm->arch.xen.xen_lock);
kvm->arch.xen.upcall_vector = data->u.vector;
mutex_unlock(&kvm->arch.xen.xen_lock);
r = 0;
} break;
case KVM_XEN_ATTR_TYPE_EVTCHN:
r = kvm_xen_setattr_evtchn(kvm, data); break;
case KVM_XEN_ATTR_TYPE_XEN_VERSION:
mutex_lock(&kvm->arch.xen.xen_lock);
kvm->arch.xen.xen_version = data->u.xen_version;
mutex_unlock(&kvm->arch.xen.xen_lock);
r = 0; break;
case KVM_XEN_ATTR_TYPE_RUNSTATE_UPDATE_FLAG: if (!sched_info_on()) {
r = -EOPNOTSUPP; break;
}
mutex_lock(&kvm->arch.xen.xen_lock);
kvm->arch.xen.runstate_update_flag = !!data->u.runstate_update_flag;
mutex_unlock(&kvm->arch.xen.xen_lock);
r = 0; break;
default: break;
}
return r;
}
int kvm_xen_hvm_get_attr(struct kvm *kvm, struct kvm_xen_hvm_attr *data)
{ int r = -ENOENT;
mutex_lock(&kvm->arch.xen.xen_lock);
switch (data->type) { case KVM_XEN_ATTR_TYPE_LONG_MODE:
data->u.long_mode = kvm->arch.xen.long_mode;
r = 0; break;
case KVM_XEN_ATTR_TYPE_SHARED_INFO: if (kvm_gpc_is_gpa_active(&kvm->arch.xen.shinfo_cache))
data->u.shared_info.gfn = gpa_to_gfn(kvm->arch.xen.shinfo_cache.gpa); else
data->u.shared_info.gfn = KVM_XEN_INVALID_GFN;
r = 0; break;
case KVM_XEN_ATTR_TYPE_SHARED_INFO_HVA: if (kvm_gpc_is_hva_active(&kvm->arch.xen.shinfo_cache))
data->u.shared_info.hva = kvm->arch.xen.shinfo_cache.uhva; else
data->u.shared_info.hva = 0;
r = 0; break;
case KVM_XEN_ATTR_TYPE_UPCALL_VECTOR:
data->u.vector = kvm->arch.xen.upcall_vector;
r = 0; break;
case KVM_XEN_ATTR_TYPE_XEN_VERSION:
data->u.xen_version = kvm->arch.xen.xen_version;
r = 0; break;
case KVM_XEN_ATTR_TYPE_RUNSTATE_UPDATE_FLAG: if (!sched_info_on()) {
r = -EOPNOTSUPP; break;
}
data->u.runstate_update_flag = kvm->arch.xen.runstate_update_flag;
r = 0; break;
switch (data->type) { case KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO: case KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO_HVA: /* No compat necessary here. */
BUILD_BUG_ON(sizeof(struct vcpu_info) != sizeof(struct compat_vcpu_info));
BUILD_BUG_ON(offsetof(struct vcpu_info, time) !=
offsetof(struct compat_vcpu_info, time));
if (data->type == KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO) { if (data->u.gpa == KVM_XEN_INVALID_GPA) {
kvm_gpc_deactivate(&vcpu->arch.xen.vcpu_info_cache);
r = 0; break;
}
r = kvm_gpc_activate(&vcpu->arch.xen.vcpu_info_cache,
data->u.gpa, sizeof(struct vcpu_info));
} else { if (data->u.hva == 0) {
kvm_gpc_deactivate(&vcpu->arch.xen.vcpu_info_cache);
r = 0; break;
}
r = kvm_gpc_activate_hva(&vcpu->arch.xen.vcpu_info_cache,
data->u.hva, sizeof(struct vcpu_info));
}
if (!r)
kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
break;
case KVM_XEN_VCPU_ATTR_TYPE_VCPU_TIME_INFO: if (data->u.gpa == KVM_XEN_INVALID_GPA) {
kvm_gpc_deactivate(&vcpu->arch.xen.vcpu_time_info_cache);
r = 0; break;
}
r = kvm_gpc_activate(&vcpu->arch.xen.vcpu_time_info_cache,
data->u.gpa, sizeof(struct pvclock_vcpu_time_info)); if (!r)
kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu); break;
case KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR: {
size_t sz, sz1, sz2;
if (!sched_info_on()) {
r = -EOPNOTSUPP; break;
} if (data->u.gpa == KVM_XEN_INVALID_GPA) {
r = 0;
deactivate_out:
kvm_gpc_deactivate(&vcpu->arch.xen.runstate_cache);
kvm_gpc_deactivate(&vcpu->arch.xen.runstate2_cache); break;
}
/* * If the guest switches to 64-bit mode after setting the runstate * address, that's actually OK. kvm_xen_update_runstate_guest() * will cope.
*/ if (IS_ENABLED(CONFIG_64BIT) && vcpu->kvm->arch.xen.long_mode)
sz = sizeof(struct vcpu_runstate_info); else
sz = sizeof(struct compat_vcpu_runstate_info);
/* How much fits in the (first) page? */
sz1 = PAGE_SIZE - (data->u.gpa & ~PAGE_MASK);
r = kvm_gpc_activate(&vcpu->arch.xen.runstate_cache,
data->u.gpa, sz1); if (r) goto deactivate_out;
/* Either map the second page, or deactivate the second GPC */ if (sz1 >= sz) {
kvm_gpc_deactivate(&vcpu->arch.xen.runstate2_cache);
} else {
sz2 = sz - sz1;
BUG_ON((data->u.gpa + sz1) & ~PAGE_MASK);
r = kvm_gpc_activate(&vcpu->arch.xen.runstate2_cache,
data->u.gpa + sz1, sz2); if (r) goto deactivate_out;
}
kvm_xen_update_runstate_guest(vcpu, false); break;
} case KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_CURRENT: if (!sched_info_on()) {
r = -EOPNOTSUPP; break;
} if (data->u.runstate.state > RUNSTATE_offline) {
r = -EINVAL; break;
}
kvm_xen_update_runstate(vcpu, data->u.runstate.state);
r = 0; break;
case KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_DATA: if (!sched_info_on()) {
r = -EOPNOTSUPP; break;
} if (data->u.runstate.state > RUNSTATE_offline) {
r = -EINVAL; break;
} if (data->u.runstate.state_entry_time !=
(data->u.runstate.time_running +
data->u.runstate.time_runnable +
data->u.runstate.time_blocked +
data->u.runstate.time_offline)) {
r = -EINVAL; break;
} if (get_kvmclock_ns(vcpu->kvm) <
data->u.runstate.state_entry_time) {
r = -EINVAL; break;
}
case KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST: if (!sched_info_on()) {
r = -EOPNOTSUPP; break;
} if (data->u.runstate.state > RUNSTATE_offline &&
data->u.runstate.state != (u64)-1) {
r = -EINVAL; break;
} /* The adjustment must add up */ if (data->u.runstate.state_entry_time !=
(data->u.runstate.time_running +
data->u.runstate.time_runnable +
data->u.runstate.time_blocked +
data->u.runstate.time_offline)) {
r = -EINVAL; break;
}
if (get_kvmclock_ns(vcpu->kvm) <
(vcpu->arch.xen.runstate_entry_time +
data->u.runstate.state_entry_time)) {
r = -EINVAL; break;
}
if (data->u.runstate.state <= RUNSTATE_offline)
kvm_xen_update_runstate(vcpu, data->u.runstate.state); elseif (vcpu->arch.xen.runstate_cache.active)
kvm_xen_update_runstate_guest(vcpu, false);
r = 0; break;
case KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID: if (data->u.vcpu_id >= KVM_MAX_VCPUS)
r = -EINVAL; else {
vcpu->arch.xen.vcpu_id = data->u.vcpu_id;
r = 0;
} break;
case KVM_XEN_VCPU_ATTR_TYPE_TIMER: if (data->u.timer.port &&
data->u.timer.priority != KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL) {
r = -EINVAL; break;
}
/* Stop the timer (if it's running) before changing the vector */
kvm_xen_stop_timer(vcpu);
vcpu->arch.xen.timer_virq = data->u.timer.port;
/* Start the timer if the new value has a valid vector+expiry. */ if (data->u.timer.port && data->u.timer.expires_ns)
kvm_xen_start_timer(vcpu, data->u.timer.expires_ns, false);
r = 0; break;
case KVM_XEN_VCPU_ATTR_TYPE_UPCALL_VECTOR: if (data->u.vector && data->u.vector < 0x10)
r = -EINVAL; else {
vcpu->arch.xen.upcall_vector = data->u.vector;
r = 0;
} break;
int kvm_xen_vcpu_get_attr(struct kvm_vcpu *vcpu, struct kvm_xen_vcpu_attr *data)
{ int r = -ENOENT;
mutex_lock(&vcpu->kvm->arch.xen.xen_lock);
switch (data->type) { case KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO: if (kvm_gpc_is_gpa_active(&vcpu->arch.xen.vcpu_info_cache))
data->u.gpa = vcpu->arch.xen.vcpu_info_cache.gpa; else
data->u.gpa = KVM_XEN_INVALID_GPA;
r = 0; break;
case KVM_XEN_VCPU_ATTR_TYPE_VCPU_INFO_HVA: if (kvm_gpc_is_hva_active(&vcpu->arch.xen.vcpu_info_cache))
data->u.hva = vcpu->arch.xen.vcpu_info_cache.uhva; else
data->u.hva = 0;
r = 0; break;
case KVM_XEN_VCPU_ATTR_TYPE_VCPU_TIME_INFO: if (vcpu->arch.xen.vcpu_time_info_cache.active)
data->u.gpa = vcpu->arch.xen.vcpu_time_info_cache.gpa; else
data->u.gpa = KVM_XEN_INVALID_GPA;
r = 0; break;
case KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADDR: if (!sched_info_on()) {
r = -EOPNOTSUPP; break;
} if (vcpu->arch.xen.runstate_cache.active) {
data->u.gpa = vcpu->arch.xen.runstate_cache.gpa;
r = 0;
} break;
case KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_CURRENT: if (!sched_info_on()) {
r = -EOPNOTSUPP; break;
}
data->u.runstate.state = vcpu->arch.xen.current_runstate;
r = 0; break;
case KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_DATA: if (!sched_info_on()) {
r = -EOPNOTSUPP; break;
}
data->u.runstate.state = vcpu->arch.xen.current_runstate;
data->u.runstate.state_entry_time =
vcpu->arch.xen.runstate_entry_time;
data->u.runstate.time_running =
vcpu->arch.xen.runstate_times[RUNSTATE_running];
data->u.runstate.time_runnable =
vcpu->arch.xen.runstate_times[RUNSTATE_runnable];
data->u.runstate.time_blocked =
vcpu->arch.xen.runstate_times[RUNSTATE_blocked];
data->u.runstate.time_offline =
vcpu->arch.xen.runstate_times[RUNSTATE_offline];
r = 0; break;
case KVM_XEN_VCPU_ATTR_TYPE_RUNSTATE_ADJUST:
r = -EINVAL; break;
case KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID:
data->u.vcpu_id = vcpu->arch.xen.vcpu_id;
r = 0; break;
case KVM_XEN_VCPU_ATTR_TYPE_TIMER: /* * Ensure a consistent snapshot of state is captured, with a * timer either being pending, or the event channel delivered * to the corresponding bit in the shared_info. Not still * lurking in the timer_pending flag for deferred delivery. * Purely as an optimisation, if the timer_expires field is * zero, that means the timer isn't active (or even in the * timer_pending flag) and there is no need to cancel it.
*/ if (vcpu->arch.xen.timer_expires) {
hrtimer_cancel(&vcpu->arch.xen.timer);
kvm_xen_inject_timer_irqs(vcpu);
}
/* * The hrtimer may trigger and raise the IRQ immediately, * while the returned state causes it to be set up and * raised again on the destination system after migration. * That's fine, as the guest won't even have had a chance * to run and handle the interrupt. Asserting an already * pending event channel is idempotent.
*/ if (vcpu->arch.xen.timer_expires)
hrtimer_start_expires(&vcpu->arch.xen.timer,
HRTIMER_MODE_ABS_HARD);
r = 0; break;
case KVM_XEN_VCPU_ATTR_TYPE_UPCALL_VECTOR:
data->u.vector = vcpu->arch.xen.upcall_vector;
r = 0; break;
int kvm_xen_write_hypercall_page(struct kvm_vcpu *vcpu, u64 data)
{ struct kvm *kvm = vcpu->kvm;
u32 page_num = data & ~PAGE_MASK;
u64 page_addr = data & PAGE_MASK; bool lm = is_long_mode(vcpu); int r = 0;
mutex_lock(&kvm->arch.xen.xen_lock); if (kvm->arch.xen.long_mode != lm) {
kvm->arch.xen.long_mode = lm;
/* * Re-initialize shared_info to put the wallclock in the * correct place.
*/ if (kvm->arch.xen.shinfo_cache.active &&
kvm_xen_shared_info_init(kvm))
r = 1;
}
mutex_unlock(&kvm->arch.xen.xen_lock);
if (r) return r;
/* * If Xen hypercall intercept is enabled, fill the hypercall * page with VMCALL/VMMCALL instructions since that's what * we catch. Else the VMM has provided the hypercall pages * with instructions of its own choosing, so use those.
*/ if (kvm_xen_hypercall_enabled(kvm)) {
u8 instructions[32]; int i;
/* int3 to pad */
memset(instructions + 9, 0xcc, sizeof(instructions) - 9);
for (i = 0; i < PAGE_SIZE / sizeof(instructions); i++) {
*(u32 *)&instructions[1] = i; if (kvm_vcpu_write_guest(vcpu,
page_addr + (i * sizeof(instructions)),
instructions, sizeof(instructions))) return 1;
}
} else { /* * Note, truncation is a non-issue as 'lm' is guaranteed to be * false for a 32-bit kernel, i.e. when hva_t is only 4 bytes.
*/
hva_t blob_addr = lm ? kvm->arch.xen.hvm_config.blob_addr_64
: kvm->arch.xen.hvm_config.blob_addr_32;
u8 blob_size = lm ? kvm->arch.xen.hvm_config.blob_size_64
: kvm->arch.xen.hvm_config.blob_size_32;
u8 *page; int ret;
if (page_num >= blob_size) return 1;
blob_addr += page_num * PAGE_SIZE;
page = memdup_user((u8 __user *)blob_addr, PAGE_SIZE); if (IS_ERR(page)) return PTR_ERR(page);
ret = kvm_vcpu_write_guest(vcpu, page_addr, page, PAGE_SIZE);
kfree(page); if (ret) return 1;
} return 0;
}
int kvm_xen_hvm_config(struct kvm *kvm, struct kvm_xen_hvm_config *xhc)
{ /* Only some feature flags need to be *enabled* by userspace */
u32 permitted_flags = KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL |
KVM_XEN_HVM_CONFIG_EVTCHN_SEND |
KVM_XEN_HVM_CONFIG_PVCLOCK_TSC_UNSTABLE;
u32 old_flags;
if (xhc->flags & ~permitted_flags) return -EINVAL;
/* * With hypercall interception the kernel generates its own * hypercall page so it must not be provided.
*/ if ((xhc->flags & KVM_XEN_HVM_CONFIG_INTERCEPT_HCALL) &&
(xhc->blob_addr_32 || xhc->blob_addr_64 ||
xhc->blob_size_32 || xhc->blob_size_64)) return -EINVAL;
/* * Restrict the MSR to the range that is unofficially reserved for * synthetic, virtualization-defined MSRs, e.g. to prevent confusing * KVM by colliding with a real MSR that requires special handling.
*/ if (xhc->msr &&
(xhc->msr < KVM_XEN_MSR_MIN_INDEX || xhc->msr > KVM_XEN_MSR_MAX_INDEX)) return -EINVAL;
mutex_lock(&kvm->arch.xen.xen_lock);
if (xhc->msr && !kvm->arch.xen.hvm_config.msr)
static_branch_inc(&kvm_xen_enabled.key); elseif (!xhc->msr && kvm->arch.xen.hvm_config.msr)
static_branch_slow_dec_deferred(&kvm_xen_enabled);
/* * This is a 32-bit pointer to an array of evtchn_port_t which * are uint32_t, so once it's converted no further compat * handling is needed.
*/
sched_poll.ports = (void *)(unsignedlong)(sp32.ports);
sched_poll.nr_ports = sp32.nr_ports;
sched_poll.timeout = sp32.timeout;
} else { if (kvm_read_guest_virt(vcpu, param, &sched_poll, sizeof(sched_poll), &e)) {
*r = -EFAULT; returntrue;
}
}
if (unlikely(sched_poll.nr_ports > 1)) { /* Xen (unofficially) limits number of pollers to 128 */ if (sched_poll.nr_ports > 128) {
*r = -EINVAL; returntrue;
}
if (!wait_pending_event(vcpu, sched_poll.nr_ports, ports)) {
kvm_set_mp_state(vcpu, KVM_MP_STATE_HALTED);
if (sched_poll.timeout)
mod_timer(&vcpu->arch.xen.poll_timer,
jiffies + nsecs_to_jiffies(sched_poll.timeout));
kvm_vcpu_halt(vcpu);
if (sched_poll.timeout)
timer_delete(&vcpu->arch.xen.poll_timer);
kvm_set_mp_state(vcpu, KVM_MP_STATE_RUNNABLE);
}
vcpu->arch.xen.poll_evtchn = 0;
*r = 0;
out: /* Really, this is only needed in case of timeout */
clear_bit(vcpu->vcpu_idx, vcpu->kvm->arch.xen.poll_mask);
if (unlikely(sched_poll.nr_ports > 1))
kfree(ports); returntrue;
}
switch (cmd) { case VCPUOP_set_singleshot_timer: if (vcpu->arch.xen.vcpu_id != vcpu_id) {
*r = -EINVAL; returntrue;
}
/* * The only difference for 32-bit compat is the 4 bytes of * padding after the interesting part of the structure. So * for a faithful emulation of Xen we have to *try* to copy * the padding and return -EFAULT if we can't. Otherwise we * might as well just have copied the 12-byte 32-bit struct.
*/
BUILD_BUG_ON(offsetof(struct compat_vcpu_set_singleshot_timer, timeout_abs_ns) !=
offsetof(struct vcpu_set_singleshot_timer, timeout_abs_ns));
BUILD_BUG_ON(sizeof_field(struct compat_vcpu_set_singleshot_timer, timeout_abs_ns) !=
sizeof_field(struct vcpu_set_singleshot_timer, timeout_abs_ns));
BUILD_BUG_ON(offsetof(struct compat_vcpu_set_singleshot_timer, flags) !=
offsetof(struct vcpu_set_singleshot_timer, flags));
BUILD_BUG_ON(sizeof_field(struct compat_vcpu_set_singleshot_timer, flags) !=
sizeof_field(struct vcpu_set_singleshot_timer, flags));
/* * Only allow hypercall acceleration for CPL0. The rare hypercalls that * are permitted in guest userspace can be handled by the VMM.
*/ if (unlikely(cpl > 0)) goto handle_in_userspace;
switch (input) { case __HYPERVISOR_xen_version: if (params[0] == XENVER_version && vcpu->kvm->arch.xen.xen_version) {
r = vcpu->kvm->arch.xen.xen_version;
handled = true;
} break; case __HYPERVISOR_event_channel_op: if (params[0] == EVTCHNOP_send)
handled = kvm_xen_hcall_evtchn_send(vcpu, params[1], &r); break; case __HYPERVISOR_sched_op:
handled = kvm_xen_hcall_sched_op(vcpu, longmode, params[0],
params[1], &r); break; case __HYPERVISOR_vcpu_op:
handled = kvm_xen_hcall_vcpu_op(vcpu, longmode, params[0], params[1],
params[2], &r); break; case __HYPERVISOR_set_timer_op: {
u64 timeout = params[0]; /* In 32-bit mode, the 64-bit timeout is in two 32-bit params. */ if (!longmode)
timeout |= params[1] << 32;
handled = kvm_xen_hcall_set_timer_op(vcpu, timeout, &r); break;
} default: break;
}
if (handled) return kvm_xen_hypercall_set_result(vcpu, r);
staticvoid kvm_xen_check_poller(struct kvm_vcpu *vcpu, int port)
{ int poll_evtchn = vcpu->arch.xen.poll_evtchn;
if ((poll_evtchn == port || poll_evtchn == -1) &&
test_and_clear_bit(vcpu->vcpu_idx, vcpu->kvm->arch.xen.poll_mask)) {
kvm_make_request(KVM_REQ_UNBLOCK, vcpu);
kvm_vcpu_kick(vcpu);
}
}
/* * The return value from this function is propagated to kvm_set_irq() API, * so it returns: * < 0 Interrupt was ignored (masked or not delivered for other reasons) * = 0 Interrupt was coalesced (previous irq is still pending) * > 0 Number of CPUs interrupt was delivered to * * It is also called directly from kvm_arch_set_irq_inatomic(), where the * only check on its return value is a comparison with -EWOULDBLOCK'.
*/ int kvm_xen_set_evtchn_fast(struct kvm_xen_evtchn *xe, struct kvm *kvm)
{ struct gfn_to_pfn_cache *gpc = &kvm->arch.xen.shinfo_cache; struct kvm_vcpu *vcpu; unsignedlong *pending_bits, *mask_bits; unsignedlong flags; int port_word_bit; bool kick_vcpu = false; int vcpu_idx, idx, rc;
/* * If this port wasn't already set, and if it isn't masked, then * we try to set the corresponding bit in the in-kernel shadow of * evtchn_pending_sel for the target vCPU. And if *that* wasn't * already set, then we kick the vCPU in question to write to the * *real* evtchn_pending_sel in its own guest vcpu_info struct.
*/ if (test_and_set_bit(xe->port, pending_bits)) {
rc = 0; /* It was already raised */
} elseif (test_bit(xe->port, mask_bits)) {
rc = -ENOTCONN; /* Masked */
kvm_xen_check_poller(vcpu, xe->port);
} else {
rc = 1; /* Delivered to the bitmap in shared_info. */ /* Now switch to the vCPU's vcpu_info to set the index and pending_sel */
read_unlock_irqrestore(&gpc->lock, flags);
gpc = &vcpu->arch.xen.vcpu_info_cache;
read_lock_irqsave(&gpc->lock, flags); if (!kvm_gpc_check(gpc, sizeof(struct vcpu_info))) { /* * Could not access the vcpu_info. Set the bit in-kernel * and prod the vCPU to deliver it for itself.
*/ if (!test_and_set_bit(port_word_bit, &vcpu->arch.xen.evtchn_pending_sel))
kick_vcpu = true; goto out_rcu;
}
/* For the per-vCPU lapic vector, deliver it as MSI. */ if (kick_vcpu && vcpu->arch.xen.upcall_vector) {
kvm_xen_inject_vcpu_vector(vcpu);
kick_vcpu = false;
}
}
rc = kvm_xen_set_evtchn_fast(xe, kvm); if (rc != -EWOULDBLOCK) return rc;
if (current->mm != kvm->mm) { /* * If not on a thread which already belongs to this KVM, * we'd better be in the irqfd workqueue.
*/ if (WARN_ON_ONCE(current->mm)) return -EINVAL;
kthread_use_mm(kvm->mm);
mm_borrowed = true;
}
/* * It is theoretically possible for the page to be unmapped * and the MMU notifier to invalidate the shared_info before * we even get to use it. In that case, this looks like an * infinite loop. It was tempting to do it via the userspace * HVA instead... but that just *hides* the fact that it's * an infinite loop, because if a fault occurs and it waits * for the page to come back, it can *still* immediately * fault and have to wait again, repeatedly. * * Conversely, the page could also have been reinstated by * another thread before we even obtain the mutex above, so * check again *first* before remapping it.
*/ do { struct gfn_to_pfn_cache *gpc = &kvm->arch.xen.shinfo_cache; int idx;
rc = kvm_xen_set_evtchn_fast(xe, kvm); if (rc != -EWOULDBLOCK) break;
/* This is the version called from kvm_set_irq() as the .set function */ staticint evtchn_set_fn(struct kvm_kernel_irq_routing_entry *e, struct kvm *kvm, int irq_source_id, int level, bool line_status)
{ if (!level) return -EINVAL;
return kvm_xen_set_evtchn(&e->xen_evtchn, kvm);
}
/* * Set up an event channel interrupt from the KVM IRQ routing table. * Used for e.g. PIRQ from passed through physical devices.
*/ int kvm_xen_setup_evtchn(struct kvm *kvm, struct kvm_kernel_irq_routing_entry *e, conststruct kvm_irq_routing_entry *ue)
{ struct kvm_vcpu *vcpu;
/* * Don't check for the port being within range of max_evtchn_port(). * Userspace can configure what ever targets it likes; events just won't * be delivered if/while the target is invalid, just like userspace can * configure MSIs which target non-existent APICs. * * This allow on Live Migration and Live Update, the IRQ routing table * can be restored *independently* of other things like creating vCPUs, * without imposing an ordering dependency on userspace. In this * particular case, the problematic ordering would be with setting the * Xen 'long mode' flag, which changes max_evtchn_port() to allow 4096 * instead of 1024 event channels.
*/
/* We only support 2 level event channels for now */ if (ue->u.xen_evtchn.priority != KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL) return -EINVAL;
/* * Xen gives us interesting mappings from vCPU index to APIC ID, * which means kvm_get_vcpu_by_id() has to iterate over all vCPUs * to find it. Do that once at setup time, instead of every time. * But beware that on live update / live migration, the routing * table might be reinstated before the vCPU threads have finished * recreating their vCPUs.
*/
vcpu = kvm_get_vcpu_by_id(kvm, ue->u.xen_evtchn.vcpu); if (vcpu)
e->xen_evtchn.vcpu_idx = vcpu->vcpu_idx; else
e->xen_evtchn.vcpu_idx = -1;
/* * None of that 'return 1 if it actually got delivered' nonsense. * We don't care if it was masked (-ENOTCONN) either.
*/ if (ret > 0 || ret == -ENOTCONN)
ret = 0;
return ret;
}
/* * Support for *outbound* event channel events via the EVTCHNOP_send hypercall.
*/ struct evtchnfd {
u32 send_port;
u32 type; union { struct kvm_xen_evtchn port; struct {
u32 port; /* zero */ struct eventfd_ctx *ctx;
} eventfd;
} deliver;
};
/* * Update target vCPU or priority for a registered sending channel.
*/ staticint kvm_xen_eventfd_update(struct kvm *kvm, struct kvm_xen_hvm_attr *data)
{
u32 port = data->u.evtchn.send_port; struct evtchnfd *evtchnfd; int ret;
/* Protect writes to evtchnfd as well as the idr lookup. */
mutex_lock(&kvm->arch.xen.xen_lock);
evtchnfd = idr_find(&kvm->arch.xen.evtchn_ports, port);
ret = -ENOENT; if (!evtchnfd) goto out_unlock;
/* For an UPDATE, nothing may change except the priority/vcpu */
ret = -EINVAL; if (evtchnfd->type != data->u.evtchn.type) goto out_unlock;
/* * Port cannot change, and if it's zero that was an eventfd * which can't be changed either.
*/ if (!evtchnfd->deliver.port.port ||
evtchnfd->deliver.port.port != data->u.evtchn.deliver.port.port) goto out_unlock;
/* We only support 2 level event channels for now */ if (data->u.evtchn.deliver.port.priority != KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL) goto out_unlock;
/* * Configure the target (eventfd or local port delivery) for sending on * a given event channel.
*/ staticint kvm_xen_eventfd_assign(struct kvm *kvm, struct kvm_xen_hvm_attr *data)
{
u32 port = data->u.evtchn.send_port; struct eventfd_ctx *eventfd = NULL; struct evtchnfd *evtchnfd; int ret = -EINVAL;
evtchnfd = kzalloc(sizeof(struct evtchnfd), GFP_KERNEL); if (!evtchnfd) return -ENOMEM;
switch(data->u.evtchn.type) { case EVTCHNSTAT_ipi: /* IPI must map back to the same port# */ if (data->u.evtchn.deliver.port.port != data->u.evtchn.send_port) goto out_noeventfd; /* -EINVAL */ break;
case EVTCHNSTAT_interdomain: if (data->u.evtchn.deliver.port.port) { if (data->u.evtchn.deliver.port.port >= max_evtchn_port(kvm)) goto out_noeventfd; /* -EINVAL */
} else {
eventfd = eventfd_ctx_fdget(data->u.evtchn.deliver.eventfd.fd); if (IS_ERR(eventfd)) {
ret = PTR_ERR(eventfd); goto out_noeventfd;
}
} break;
case EVTCHNSTAT_virq: case EVTCHNSTAT_closed: case EVTCHNSTAT_unbound: case EVTCHNSTAT_pirq: default: /* Unknown event channel type */ goto out; /* -EINVAL */
}
evtchnfd->send_port = data->u.evtchn.send_port;
evtchnfd->type = data->u.evtchn.type; if (eventfd) {
evtchnfd->deliver.eventfd.ctx = eventfd;
} else { /* We only support 2 level event channels for now */ if (data->u.evtchn.deliver.port.priority != KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL) goto out; /* -EINVAL; */
synchronize_srcu(&kvm->srcu); if (!evtchnfd->deliver.port.port)
eventfd_ctx_put(evtchnfd->deliver.eventfd.ctx);
kfree(evtchnfd); return 0;
}
staticint kvm_xen_eventfd_reset(struct kvm *kvm)
{ struct evtchnfd *evtchnfd, **all_evtchnfds; int i; int n = 0;
mutex_lock(&kvm->arch.xen.xen_lock);
/* * Because synchronize_srcu() cannot be called inside the * critical section, first collect all the evtchnfd objects * in an array as they are removed from evtchn_ports.
*/
idr_for_each_entry(&kvm->arch.xen.evtchn_ports, evtchnfd, i)
n++;
/* Sanity check: this structure is the same for 32-bit and 64-bit */
BUILD_BUG_ON(sizeof(send) != 4); if (kvm_read_guest_virt(vcpu, param, &send, sizeof(send), &e)) {
*r = -EFAULT; returntrue;
}
/* * evtchnfd is protected by kvm->srcu; the idr lookup instead * is protected by RCU.
*/
rcu_read_lock();
evtchnfd = idr_find(&vcpu->kvm->arch.xen.evtchn_ports, send.port);
rcu_read_unlock(); if (!evtchnfd) returnfalse;
if (evtchnfd->deliver.port.port) { int ret = kvm_xen_set_evtchn(&evtchnfd->deliver.port, vcpu->kvm); if (ret < 0 && ret != -ENOTCONN) returnfalse;
} else {
eventfd_signal(evtchnfd->deliver.eventfd.ctx);
}
Die Informationen auf dieser Webseite wurden
nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit,
noch Qualität der bereit gestellten Informationen zugesichert.
Bemerkung:
Die farbliche Syntaxdarstellung und die Messung sind noch experimentell.
