From: Paolo Bonzini <pbonzini@redhat.com>
To: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: wei.huang2@amd.com, cavery@redhat.com,
Sean Christopherson <sean.j.christopherson@intel.com>,
Oliver Upton <oupton@google.com>,
Jim Mattson <jmattson@google.com>,
linux-kernel@vger.kernel.org, kvm@vger.kernel.org
Subject: Re: [PATCH v2 03/22] KVM: SVM: immediately inject INTR vmexit
Date: Thu, 21 May 2020 23:04:10 +0200 [thread overview]
Message-ID: <73e55aba-dc71-ae36-d491-a6afed844c9a@redhat.com> (raw)
In-Reply-To: <8bc4c38a-1717-1e4f-b322-fdd51f614717@redhat.com>
On 21/05/20 16:08, Paolo Bonzini wrote:
> On 21/05/20 14:50, Vitaly Kuznetsov wrote:
>> Sorry for reporting this late but I just found out that this commit
>> breaks Hyper-V 2016 on KVM on SVM completely (always hangs on boot). I
>> haven't investigated it yet (well, this is Windows, you know...) but
>> what's usually different about Hyper-V is that unlike KVM/Linux it has
>> handlers for some hardware interrupts in the guest and not in the
>> hypervisor.
>
> "Always hangs on boot" is easy. :) At this point I think it's easiest
> to debug it on top of the whole pending SVM patches that remove
> exit_required completely (and exit_required is not coming back anyway).
Ok so there could be two bugs, as the hang seems to happens much earlier
later in the series (try "grep int_ctl:.0x.....1.." on the trace).
As one could guess from the grep, one thing that is certainly different
between KVM and Hyper-V is that Hyper-V injects interrupts using
int_ctl; sometimes it also uses eventinj but presumably it's just
copying it from exitintinfo).
This could cause problems: for example, when L1 wants to inject a
virtual interrupt into L2 that has interrupts disabled or V_TPR >=
V_INTR_PRIO, and KVM also wants to inject an interrupt to L1, then KVM
might end up stomping on Hyper-V's int_ctl. However I cannot think
off-hand of a scenario where this could happen in this case, because
Hyper-V does set EXIT_INTR and therefore we should never get into
enable_irq_window while L2 is running. Still, that's one place where
I'd start adding some trace_printk's.
Also, if a uniprocessor guest also fails, it might be easier to debug.
Paolo
next prev parent reply other threads:[~2020-05-21 21:04 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-24 17:23 [PATCH v2 00/22] KVM: Event fixes and cleanup Paolo Bonzini
2020-04-24 17:23 ` [PATCH v2 01/22] KVM: SVM: introduce nested_run_pending Paolo Bonzini
2020-04-24 17:23 ` [PATCH v2 02/22] KVM: SVM: leave halted state on vmexit Paolo Bonzini
2020-04-24 17:41 ` Oliver Upton
2020-04-24 17:23 ` [PATCH v2 03/22] KVM: SVM: immediately inject INTR vmexit Paolo Bonzini
2020-05-21 12:50 ` Vitaly Kuznetsov
2020-05-21 14:08 ` Paolo Bonzini
2020-05-21 21:04 ` Paolo Bonzini [this message]
2020-04-24 17:23 ` [PATCH v2 04/22] KVM: SVM: Implement check_nested_events for NMI Paolo Bonzini
2020-04-24 17:23 ` [PATCH v2 05/22] KVM: nVMX: Preserve exception priority irrespective of exiting behavior Paolo Bonzini
2020-04-24 17:24 ` [PATCH v2 06/22] KVM: nVMX: Open a window for pending nested VMX preemption timer Paolo Bonzini
2020-04-24 17:24 ` [PATCH v2 07/22] KVM: x86: Set KVM_REQ_EVENT if run is canceled with req_immediate_exit set Paolo Bonzini
2020-04-24 17:24 ` [PATCH v2 08/22] KVM: x86: Make return for {interrupt_nmi,smi}_allowed() a bool instead of int Paolo Bonzini
2020-04-24 17:24 ` [PATCH v2 09/22] KVM: x86: replace is_smm checks with kvm_x86_ops.smi_allowed Paolo Bonzini
2020-04-24 17:29 ` [PATCH v2 00/22] KVM: Event fixes and cleanup Sean Christopherson
2020-04-24 21:02 ` Oliver Upton
2020-04-24 21:05 ` Sean Christopherson
2020-04-25 7:21 ` Paolo Bonzini
2020-04-25 7:01 ` [PATCH v2 10/22] KVM: nVMX: Report NMIs as allowed when in L2 and Exit-on-NMI is set Paolo Bonzini
2020-04-25 7:01 ` [PATCH v2 11/22] KVM: nSVM: " Paolo Bonzini
2020-04-25 7:01 ` [PATCH v2 12/22] KVM: nSVM: Move SMI vmexit handling to svm_check_nested_events() Paolo Bonzini
2020-04-25 7:01 ` [PATCH v2 13/22] KVM: VMX: Split out architectural interrupt/NMI blocking checks Paolo Bonzini
2020-04-25 7:01 ` [PATCH v2 14/22] KVM: SVM: Split out architectural interrupt/NMI/SMI " Paolo Bonzini
2020-04-25 7:01 ` [PATCH v2 15/22] KVM: nVMX: Preserve IRQ/NMI priority irrespective of exiting behavior Paolo Bonzini
2020-04-25 7:01 ` [PATCH v2 16/22] KVM: nVMX: Prioritize SMI over nested IRQ/NMI Paolo Bonzini
2020-04-25 7:01 ` [PATCH v2 17/22] KVM: nSVM: Report interrupts as allowed when in L2 and exit-on-interrupt is set Paolo Bonzini
2020-04-25 7:01 ` [PATCH v2 18/22] KVM: nSVM: Preserve IRQ/NMI/SMI priority irrespective of exiting behavior Paolo Bonzini
2020-04-25 7:01 ` [PATCH v2 19/22] KVM: x86: WARN on injected+pending exception even in nested case Paolo Bonzini
2020-04-25 7:02 ` [PATCH v2 20/22] KVM: VMX: Use vmx_interrupt_blocked() directly from vmx_handle_exit() Paolo Bonzini
2020-04-25 7:02 ` [PATCH v2 21/22] KVM: VMX: Use vmx_get_rflags() to query RFLAGS in vmx_interrupt_blocked() Paolo Bonzini
2020-04-25 7:02 ` [PATCH v2 22/22] KVM: x86: Replace late check_nested_events() hack with more precise fix Paolo Bonzini
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=73e55aba-dc71-ae36-d491-a6afed844c9a@redhat.com \
--to=pbonzini@redhat.com \
--cc=cavery@redhat.com \
--cc=jmattson@google.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=oupton@google.com \
--cc=sean.j.christopherson@intel.com \
--cc=vkuznets@redhat.com \
--cc=wei.huang2@amd.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).