qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: Vitaly Kuznetsov <vkuznets@redhat.com>,
	Jan Richter <jarichte@redhat.com>
Subject: [PULL 09/25] target/i386: Make sure SynIC state is really updated before KVM_RUN
Date: Tue, 15 Oct 2024 16:16:55 +0200	[thread overview]
Message-ID: <20241015141711.528342-10-pbonzini@redhat.com> (raw)
In-Reply-To: <20241015141711.528342-1-pbonzini@redhat.com>

From: Vitaly Kuznetsov <vkuznets@redhat.com>

'hyperv_synic' test from KVM unittests was observed to be flaky on certain
hardware (hangs sometimes). Debugging shows that the problem happens in
hyperv_sint_route_new() when the test tries to set up a new SynIC
route. The function bails out on:

 if (!synic->sctl_enabled) {
         goto cleanup;
 }

but the test writes to HV_X64_MSR_SCONTROL just before it starts
establishing SINT routes. Further investigation shows that
synic_update() (called from async_synic_update()) happens after the SINT
setup attempt and not before. Apparently, the comment before
async_safe_run_on_cpu() in kvm_hv_handle_exit() does not correctly describe
the guarantees async_safe_run_on_cpu() gives. In particular, async worked
added to a CPU is actually processed from qemu_wait_io_event() which is not
always called before KVM_RUN, i.e. kvm_cpu_exec() checks whether an exit
request is pending for a CPU and if not, keeps running the vCPU until it
meets an exit it can't handle internally. Hyper-V specific MSR writes are
not automatically trigger an exit.

Fix the issue by simply raising an exit request for the vCPU where SynIC
update was queued. This is not a performance critical path as SynIC state
does not get updated so often (and async_safe_run_on_cpu() is a big hammer
anyways).

Reported-by: Jan Richter <jarichte@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Link: https://lore.kernel.org/r/20240917160051.2637594-4-vkuznets@redhat.com
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 target/i386/kvm/hyperv.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/target/i386/kvm/hyperv.c b/target/i386/kvm/hyperv.c
index b94f12acc2c..70b89cacf94 100644
--- a/target/i386/kvm/hyperv.c
+++ b/target/i386/kvm/hyperv.c
@@ -80,6 +80,7 @@ int kvm_hv_handle_exit(X86CPU *cpu, struct kvm_hyperv_exit *exit)
          * necessary because memory hierarchy is being changed
          */
         async_safe_run_on_cpu(CPU(cpu), async_synic_update, RUN_ON_CPU_NULL);
+        cpu_exit(CPU(cpu));
 
         return EXCP_INTERRUPT;
     case KVM_EXIT_HYPERV_HCALL: {
-- 
2.46.2



  parent reply	other threads:[~2024-10-15 14:21 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-15 14:16 [PULL 00/25] x86 and KVM patches for 2024-10-15 Paolo Bonzini
2024-10-15 14:16 ` [PULL 01/25] target/i386: Don't construct a all-zero entry for CPUID[0xD 0x3f] Paolo Bonzini
2024-10-15 14:16 ` [PULL 02/25] target/i386: Enable fdp-excptn-only and zero-fcs-fds Paolo Bonzini
2024-10-15 14:16 ` [PULL 03/25] target/i386: Construct CPUID 2 as stateful iff times > 1 Paolo Bonzini
2024-10-15 14:16 ` [PULL 04/25] target/i386: Make invtsc migratable when user sets tsc-khz explicitly Paolo Bonzini
2024-10-15 14:16 ` [PULL 05/25] target/i386: Add more features enumerated by CPUID.7.2.EDX Paolo Bonzini
2024-10-15 14:16 ` [PULL 06/25] target/i386: Add support save/load HWCR MSR Paolo Bonzini
2024-10-15 14:16 ` [PULL 07/25] target/i386: Fix conditional CONFIG_SYNDBG enablement Paolo Bonzini
2024-10-15 14:16 ` [PULL 08/25] target/i386: Exclude 'hv-syndbg' from 'hv-passthrough' Paolo Bonzini
2024-10-15 14:16 ` Paolo Bonzini [this message]
2024-10-15 14:16 ` [PULL 10/25] docs/system: Add recommendations to Hyper-V enlightenments doc Paolo Bonzini
2024-10-15 14:16 ` [PULL 11/25] target/i386: convert bit test instructions to new decoder Paolo Bonzini
2024-10-15 14:16 ` [PULL 12/25] target/i386: decode address before going back to translate.c Paolo Bonzini
2024-10-15 14:16 ` [PULL 13/25] target/i386: convert CMPXCHG8B/CMPXCHG16B to new decoder Paolo Bonzini
2024-10-16 16:37   ` Philippe Mathieu-Daudé
2024-10-17  9:14     ` Paolo Bonzini
2024-10-15 14:17 ` [PULL 14/25] target/i386: do not check PREFIX_LOCK in old-style decoder Paolo Bonzini
2024-10-15 14:17 ` [PULL 15/25] target/i386: list instructions still in translate.c Paolo Bonzini
2024-10-15 14:17 ` [PULL 16/25] target/i386: assert that cc_op* and pc_save are preserved Paolo Bonzini
2024-10-15 14:17 ` [PULL 17/25] KVM: Dynamic sized kvm memslots array Paolo Bonzini
2024-10-15 14:17 ` [PULL 18/25] KVM: Define KVM_MEMSLOTS_NUM_MAX_DEFAULT Paolo Bonzini
2024-10-15 14:17 ` [PULL 19/25] KVM: Rename KVMMemoryListener.nr_used_slots to nr_slots_used Paolo Bonzini
2024-10-15 14:17 ` [PULL 20/25] KVM: Rename KVMState->nr_slots to nr_slots_max Paolo Bonzini
2024-10-15 14:17 ` [PULL 21/25] target/i386/tcg: Use DPL-level accesses for interrupts and call gates Paolo Bonzini
2024-10-15 14:17 ` [PULL 22/25] accel/kvm: check for KVM_CAP_MULTI_ADDRESS_SPACE on vm Paolo Bonzini
2024-10-15 14:17 ` [PULL 23/25] accel/kvm: check for KVM_CAP_MEMORY_ATTRIBUTES " Paolo Bonzini
2024-10-15 14:17 ` [PULL 24/25] accel/kvm: check for KVM_CAP_READONLY_MEM on VM Paolo Bonzini
2024-10-15 14:17 ` [PULL 25/25] target/i386: Use only 16 and 32-bit operands for IN/OUT Paolo Bonzini
2024-10-17 10:32 ` [PULL 00/25] x86 and KVM patches for 2024-10-15 Peter Maydell

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=20241015141711.528342-10-pbonzini@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=jarichte@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=vkuznets@redhat.com \
    /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;
as well as URLs for NNTP newsgroup(s).