mirror of
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git
synced 2024-09-28 21:33:52 +00:00
KVM: x86/xen: Fix runstate updates to be atomic when preempting vCPU
There are circumstances whem kvm_xen_update_runstate_guest() should not
sleep because it ends up being called from __schedule() when the vCPU
is preempted:
[ 222.830825] kvm_xen_update_runstate_guest+0x24/0x100
[ 222.830878] kvm_arch_vcpu_put+0x14c/0x200
[ 222.830920] kvm_sched_out+0x30/0x40
[ 222.830960] __schedule+0x55c/0x9f0
To handle this, make it use the same trick as __kvm_xen_has_interrupt(),
of using the hva from the gfn_to_hva_cache directly. Then it can use
pagefault_disable() around the accesses and just bail out if the page
is absent (which is unlikely).
I almost switched to using a gfn_to_pfn_cache here and bailing out if
kvm_map_gfn() fails, like kvm_steal_time_set_preempted() does — but on
closer inspection it looks like kvm_map_gfn() will *always* fail in
atomic context for a page in IOMEM, which means it will silently fail
to make the update every single time for such guests, AFAICT. So I
didn't do it that way after all. And will probably fix that one too.
Cc: stable@vger.kernel.org
Fixes: 30b5c851af
("KVM: x86/xen: Add support for vCPU runstate information")
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Message-Id: <b17a93e5ff4561e57b1238e3e7ccd0b613eb827e.camel@infradead.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This commit is contained in:
parent
3915035282
commit
fcb732d8f8
1 changed files with 67 additions and 30 deletions
|
@ -133,32 +133,57 @@ static void kvm_xen_update_runstate(struct kvm_vcpu *v, int state)
|
|||
void kvm_xen_update_runstate_guest(struct kvm_vcpu *v, int state)
|
||||
{
|
||||
struct kvm_vcpu_xen *vx = &v->arch.xen;
|
||||
struct gfn_to_hva_cache *ghc = &vx->runstate_cache;
|
||||
struct kvm_memslots *slots = kvm_memslots(v->kvm);
|
||||
bool atomic = (state == RUNSTATE_runnable);
|
||||
uint64_t state_entry_time;
|
||||
unsigned int offset;
|
||||
int __user *user_state;
|
||||
uint64_t __user *user_times;
|
||||
|
||||
kvm_xen_update_runstate(v, state);
|
||||
|
||||
if (!vx->runstate_set)
|
||||
return;
|
||||
|
||||
if (unlikely(slots->generation != ghc->generation || kvm_is_error_hva(ghc->hva)) &&
|
||||
kvm_gfn_to_hva_cache_init(v->kvm, ghc, ghc->gpa, ghc->len))
|
||||
return;
|
||||
|
||||
/* We made sure it fits in a single page */
|
||||
BUG_ON(!ghc->memslot);
|
||||
|
||||
if (atomic)
|
||||
pagefault_disable();
|
||||
|
||||
/*
|
||||
* The only difference between 32-bit and 64-bit versions of the
|
||||
* runstate struct us 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'.
|
||||
*
|
||||
* So we use 'int __user *user_state' to point to the state field,
|
||||
* and 'uint64_t __user *user_times' for runstate_entry_time. So
|
||||
* the actual array of time[] in each state starts at user_times[1].
|
||||
*/
|
||||
BUILD_BUG_ON(offsetof(struct vcpu_runstate_info, state) != 0);
|
||||
BUILD_BUG_ON(offsetof(struct compat_vcpu_runstate_info, state) != 0);
|
||||
user_state = (int __user *)ghc->hva;
|
||||
|
||||
BUILD_BUG_ON(sizeof(struct compat_vcpu_runstate_info) != 0x2c);
|
||||
|
||||
offset = offsetof(struct compat_vcpu_runstate_info, state_entry_time);
|
||||
user_times = (uint64_t __user *)(ghc->hva +
|
||||
offsetof(struct compat_vcpu_runstate_info,
|
||||
state_entry_time));
|
||||
#ifdef CONFIG_X86_64
|
||||
/*
|
||||
* The only difference is alignment of uint64_t in 32-bit.
|
||||
* So the first field 'state' is accessed directly using
|
||||
* offsetof() (where its offset happens to be zero), while the
|
||||
* remaining fields which are all uint64_t, start at 'offset'
|
||||
* which we tweak here by adding 4.
|
||||
*/
|
||||
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);
|
||||
|
||||
if (v->kvm->arch.xen.long_mode)
|
||||
offset = offsetof(struct vcpu_runstate_info, state_entry_time);
|
||||
user_times = (uint64_t __user *)(ghc->hva +
|
||||
offsetof(struct vcpu_runstate_info,
|
||||
state_entry_time));
|
||||
#endif
|
||||
/*
|
||||
* First write the updated state_entry_time at the appropriate
|
||||
|
@ -172,10 +197,8 @@ void kvm_xen_update_runstate_guest(struct kvm_vcpu *v, int state)
|
|||
BUILD_BUG_ON(sizeof_field(struct compat_vcpu_runstate_info, state_entry_time) !=
|
||||
sizeof(state_entry_time));
|
||||
|
||||
if (kvm_write_guest_offset_cached(v->kvm, &v->arch.xen.runstate_cache,
|
||||
&state_entry_time, offset,
|
||||
sizeof(state_entry_time)))
|
||||
return;
|
||||
if (__put_user(state_entry_time, user_times))
|
||||
goto out;
|
||||
smp_wmb();
|
||||
|
||||
/*
|
||||
|
@ -189,11 +212,8 @@ void kvm_xen_update_runstate_guest(struct kvm_vcpu *v, int state)
|
|||
BUILD_BUG_ON(sizeof_field(struct compat_vcpu_runstate_info, state) !=
|
||||
sizeof(vx->current_runstate));
|
||||
|
||||
if (kvm_write_guest_offset_cached(v->kvm, &v->arch.xen.runstate_cache,
|
||||
&vx->current_runstate,
|
||||
offsetof(struct vcpu_runstate_info, state),
|
||||
sizeof(vx->current_runstate)))
|
||||
return;
|
||||
if (__put_user(vx->current_runstate, user_state))
|
||||
goto out;
|
||||
|
||||
/*
|
||||
* Write the actual runstate times immediately after the
|
||||
|
@ -208,24 +228,23 @@ void kvm_xen_update_runstate_guest(struct kvm_vcpu *v, int state)
|
|||
BUILD_BUG_ON(sizeof_field(struct vcpu_runstate_info, time) !=
|
||||
sizeof(vx->runstate_times));
|
||||
|
||||
if (kvm_write_guest_offset_cached(v->kvm, &v->arch.xen.runstate_cache,
|
||||
&vx->runstate_times[0],
|
||||
offset + sizeof(u64),
|
||||
sizeof(vx->runstate_times)))
|
||||
return;
|
||||
|
||||
if (__copy_to_user(user_times + 1, vx->runstate_times, sizeof(vx->runstate_times)))
|
||||
goto out;
|
||||
smp_wmb();
|
||||
|
||||
/*
|
||||
* Finally, clear the XEN_RUNSTATE_UPDATE bit in the guest's
|
||||
* runstate_entry_time field.
|
||||
*/
|
||||
|
||||
state_entry_time &= ~XEN_RUNSTATE_UPDATE;
|
||||
if (kvm_write_guest_offset_cached(v->kvm, &v->arch.xen.runstate_cache,
|
||||
&state_entry_time, offset,
|
||||
sizeof(state_entry_time)))
|
||||
return;
|
||||
__put_user(state_entry_time, user_times);
|
||||
smp_wmb();
|
||||
|
||||
out:
|
||||
mark_page_dirty_in_slot(v->kvm, ghc->memslot, ghc->gpa >> PAGE_SHIFT);
|
||||
|
||||
if (atomic)
|
||||
pagefault_enable();
|
||||
}
|
||||
|
||||
int __kvm_xen_has_interrupt(struct kvm_vcpu *v)
|
||||
|
@ -443,6 +462,12 @@ int kvm_xen_vcpu_set_attr(struct kvm_vcpu *vcpu, struct kvm_xen_vcpu_attr *data)
|
|||
break;
|
||||
}
|
||||
|
||||
/* It must fit within a single page */
|
||||
if ((data->u.gpa & ~PAGE_MASK) + sizeof(struct vcpu_info) > PAGE_SIZE) {
|
||||
r = -EINVAL;
|
||||
break;
|
||||
}
|
||||
|
||||
r = kvm_gfn_to_hva_cache_init(vcpu->kvm,
|
||||
&vcpu->arch.xen.vcpu_info_cache,
|
||||
data->u.gpa,
|
||||
|
@ -460,6 +485,12 @@ int kvm_xen_vcpu_set_attr(struct kvm_vcpu *vcpu, struct kvm_xen_vcpu_attr *data)
|
|||
break;
|
||||
}
|
||||
|
||||
/* It must fit within a single page */
|
||||
if ((data->u.gpa & ~PAGE_MASK) + sizeof(struct pvclock_vcpu_time_info) > PAGE_SIZE) {
|
||||
r = -EINVAL;
|
||||
break;
|
||||
}
|
||||
|
||||
r = kvm_gfn_to_hva_cache_init(vcpu->kvm,
|
||||
&vcpu->arch.xen.vcpu_time_info_cache,
|
||||
data->u.gpa,
|
||||
|
@ -481,6 +512,12 @@ int kvm_xen_vcpu_set_attr(struct kvm_vcpu *vcpu, struct kvm_xen_vcpu_attr *data)
|
|||
break;
|
||||
}
|
||||
|
||||
/* It must fit within a single page */
|
||||
if ((data->u.gpa & ~PAGE_MASK) + sizeof(struct vcpu_runstate_info) > PAGE_SIZE) {
|
||||
r = -EINVAL;
|
||||
break;
|
||||
}
|
||||
|
||||
r = kvm_gfn_to_hva_cache_init(vcpu->kvm,
|
||||
&vcpu->arch.xen.runstate_cache,
|
||||
data->u.gpa,
|
||||
|
|
Loading…
Reference in a new issue