From: Paul Mackerras <paulus@ozlabs.org>
To: kvm@vger.kernel.org, kvm-ppc@vger.kernel.org
Subject: [PATCH] KVM: PPC: Book3S HV: Drop locks before reading guest memory
Date: Mon, 29 Jan 2018 23:51:32 +0000 [thread overview]
Message-ID: <20180129235132.GA27928@fergus.ozlabs.ibm.com> (raw)
In-Reply-To: <20120601102024.GB11714@pale.ozlabs.ibm.com>
Running with CONFIG_DEBUG_ATOMIC_SLEEP reveals that HV KVM tries to
read guest memory, in order to emulate guest instructions, while
preempt is disabled and a vcore lock is held. This occurs in
kvmppc_handle_exit_hv(), called from post_guest_process(), when
emulating guest doorbell instructions on POWER9 systems, and also
when checking whether we have hit a hypervisor breakpoint.
Reading guest memory can cause a page fault and thus cause the
task to sleep, so we need to avoid reading guest memory while
holding a spinlock or when preempt is disabled.
To fix this, we move the preempt_enable() in kvmppc_run_core() to
before the loop that calls post_guest_process() for each vcore that
has just run, and we drop and re-take the vcore lock around the calls
to kvmppc_emulate_debug_inst() and kvmppc_emulate_doorbell_instr().
Dropping the lock is safe with respect to the iteration over the
runnable vcpus in post_guest_process(); for_each_runnable_thread
is actually safe to use locklessly. It is possible for a vcpu
to become runnable and add itself to the runnable_threads array
(code near the beginning of kvmppc_run_vcpu()) and then get included
in the iteration in post_guest_process despite the fact that it
has not just run. This is benign because vcpu->arch.trap and
vcpu->arch.ceded will be zero.
Cc: stable@vger.kernel.org # v4.13+
Fixes: 579006944e0d ("KVM: PPC: Book3S HV: Virtualize doorbell facility on POWER9")
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
---
arch/powerpc/kvm/book3s_hv.c | 16 ++++++++++++----
1 file changed, 12 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index e5f81fc..aa6130b 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -1008,8 +1008,6 @@ static int kvmppc_emulate_doorbell_instr(struct kvm_vcpu *vcpu)
struct kvm *kvm = vcpu->kvm;
struct kvm_vcpu *tvcpu;
- if (!cpu_has_feature(CPU_FTR_ARCH_300))
- return EMULATE_FAIL;
if (kvmppc_get_last_inst(vcpu, INST_GENERIC, &inst) != EMULATE_DONE)
return RESUME_GUEST;
if (get_op(inst) != 31)
@@ -1059,6 +1057,7 @@ static int kvmppc_emulate_doorbell_instr(struct kvm_vcpu *vcpu)
return RESUME_GUEST;
}
+/* Called with vcpu->arch.vcore->lock held */
static int kvmppc_handle_exit_hv(struct kvm_run *run, struct kvm_vcpu *vcpu,
struct task_struct *tsk)
{
@@ -1179,7 +1178,10 @@ static int kvmppc_handle_exit_hv(struct kvm_run *run, struct kvm_vcpu *vcpu,
swab32(vcpu->arch.emul_inst) :
vcpu->arch.emul_inst;
if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP) {
+ /* Need vcore unlocked to call kvmppc_get_last_inst */
+ spin_unlock(&vcpu->arch.vcore->lock);
r = kvmppc_emulate_debug_inst(run, vcpu);
+ spin_lock(&vcpu->arch.vcore->lock);
} else {
kvmppc_core_queue_program(vcpu, SRR1_PROGILL);
r = RESUME_GUEST;
@@ -1194,8 +1196,13 @@ static int kvmppc_handle_exit_hv(struct kvm_run *run, struct kvm_vcpu *vcpu,
*/
case BOOK3S_INTERRUPT_H_FAC_UNAVAIL:
r = EMULATE_FAIL;
- if ((vcpu->arch.hfscr >> 56) = FSCR_MSGP_LG)
+ if (((vcpu->arch.hfscr >> 56) = FSCR_MSGP_LG) &&
+ cpu_has_feature(CPU_FTR_ARCH_300)) {
+ /* Need vcore unlocked to call kvmppc_get_last_inst */
+ spin_unlock(&vcpu->arch.vcore->lock);
r = kvmppc_emulate_doorbell_instr(vcpu);
+ spin_lock(&vcpu->arch.vcore->lock);
+ }
if (r = EMULATE_FAIL) {
kvmppc_core_queue_program(vcpu, SRR1_PROGILL);
r = RESUME_GUEST;
@@ -2946,13 +2953,14 @@ static noinline void kvmppc_run_core(struct kvmppc_vcore *vc)
/* make sure updates to secondary vcpu structs are visible now */
smp_mb();
+ preempt_enable();
+
for (sub = 0; sub < core_info.n_subcores; ++sub) {
pvc = core_info.vc[sub];
post_guest_process(pvc, pvc = vc);
}
spin_lock(&vc->lock);
- preempt_enable();
out:
vc->vcore_state = VCORE_INACTIVE;
--
2.7.4
prev parent reply other threads:[~2018-01-29 23:51 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-01 10:20 [PATCH] KVM: PPC: Book3S HV: Drop locks around call to kvmppc_pin_guest_page Paul Mackerras
2012-06-06 12:28 ` Avi Kivity
2012-06-06 15:52 ` Alexander Graf
2012-06-19 10:50 ` Alexander Graf
2012-06-19 12:04 ` Avi Kivity
2018-01-29 23:51 ` Paul Mackerras [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20180129235132.GA27928@fergus.ozlabs.ibm.com \
--to=paulus@ozlabs.org \
--cc=kvm-ppc@vger.kernel.org \
--cc=kvm@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox