From: "Li, Xin3" <xin3.li@intel.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"x86@kernel.org" <x86@kernel.org>,
"kvm@vger.kernel.org" <kvm@vger.kernel.org>,
"tglx@linutronix.de" <tglx@linutronix.de>,
"mingo@redhat.com" <mingo@redhat.com>,
"bp@alien8.de" <bp@alien8.de>,
"dave.hansen@linux.intel.com" <dave.hansen@linux.intel.com>,
"hpa@zytor.com" <hpa@zytor.com>, "Christopherson,,
Sean" <seanjc@google.com>, "Tian, Kevin" <kevin.tian@intel.com>
Subject: RE: [RESEND PATCH 5/6] KVM: x86/VMX: add kvm_vmx_reinject_nmi_irq() for NMI/IRQ reinjection
Date: Sat, 12 Nov 2022 06:35:39 +0000 [thread overview]
Message-ID: <BN6PR1101MB216163B12055F02E8C54D2BFA8039@BN6PR1101MB2161.namprd11.prod.outlook.com> (raw)
In-Reply-To: <Y26jkHfK9INwU7Yy@hirez.programming.kicks-ass.net>
> > > So instead of fixing it, they made it worse :-(
> > >
> > > And now FRED is arguably making it worse again, and people wonder
> > > why I hate virt...
> >
> > Maybe I take it wrong, but FRED doesn't make anything worse. Fred
> > entry code will call external_interrupt() immediately for IRQs.
>
> But what about NMIs, afaict this is all horribly broken for NMIs.
NMIs are NOT handled by external_interrupt(), which is introduced in patch 4.
The NMI handling is added to kvm_vmx_reinject_nmi_irq() in patch 5 purely for VMX only.
>
> So the whole VMX thing latches the NMI (which stops NMI recursion), right?
>
> But then you drop out of noinstr code, which means any random exception can
> happen (kprobes #BP, hw_breakpoint #DB, or even #PF due to random
> nonsense like *SAN). This exception will do IRET and clear the NMI latch, all
> before you get to run any of the NMI code.
>
> Note how the normal NMI code is very careful to clear DR7 and both kprobes
> and hw_breakpoint know not to accept noinstr code as targets.
>
> You threw all that out the window.
>
> Also, NMI is IST, and with FRED it will run on a different stack as well, directly
> calling external_interrupt() doesn't honour that either.
>
> > You really really don't like the context how VMX dispatches NMI/IRQs
> > (which has been there for a long time), right?
>
> I really really hate this with a passion. The fact that it's been this way is no
> justification for keeping it. Crap is crap.
>
> Intel should have taken an example of SVM in this regard, and not doubled
> down and extended this NMI hole to regular IRQs. These are exactly the kind of
> exception delivery trainwrecks FRED is supposed to fix, except in this case it
> appears it doesn't :/
next prev parent reply other threads:[~2022-11-12 6:35 UTC|newest]
Thread overview: 56+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-11-10 6:15 [RESEND PATCH 0/6] x86/traps,VMX: implement software based NMI/IRQ dispatch for VMX NMI/IRQ reinjection Xin Li
2022-11-10 6:15 ` [RESEND PATCH 1/6] x86/traps: let common_interrupt() handle IRQ_MOVE_CLEANUP_VECTOR Xin Li
2022-11-10 18:59 ` Ashok Raj
2022-11-10 22:09 ` Li, Xin3
2022-11-10 6:15 ` [RESEND PATCH 2/6] x86/traps: add a system interrupt table for system interrupt dispatch Xin Li
2022-11-10 8:56 ` Peter Zijlstra
2022-11-10 19:55 ` Li, Xin3
2022-11-10 20:36 ` Li, Xin3
2022-11-10 21:12 ` Nathan Chancellor
2022-11-10 23:00 ` Li, Xin3
2022-11-11 0:08 ` Nathan Chancellor
2022-11-11 3:03 ` Li, Xin3
2022-11-11 8:58 ` Peter Zijlstra
2022-11-11 1:12 ` Tian, Kevin
2022-11-11 3:54 ` Li, Xin3
2022-11-11 8:55 ` Peter Zijlstra
2022-11-11 22:07 ` H. Peter Anvin
2022-11-12 9:47 ` Peter Zijlstra
2022-11-10 6:15 ` [RESEND PATCH 3/6] x86/traps: add install_system_interrupt_handler() Xin Li
2022-11-10 6:15 ` [RESEND PATCH 4/6] x86/traps: add external_interrupt() to dispatch external interrupts Xin Li
2022-11-10 16:24 ` Sean Christopherson
2022-11-10 18:02 ` Li, Xin3
2022-11-10 20:10 ` Sean Christopherson
2022-11-10 6:15 ` [RESEND PATCH 5/6] KVM: x86/VMX: add kvm_vmx_reinject_nmi_irq() for NMI/IRQ reinjection Xin Li
2022-11-10 9:03 ` Peter Zijlstra
2022-11-10 20:53 ` Li, Xin3
2022-11-11 9:15 ` Peter Zijlstra
2022-11-11 12:04 ` Paolo Bonzini
2022-11-11 12:19 ` Peter Zijlstra
2022-11-11 12:48 ` Paolo Bonzini
2022-11-11 14:23 ` Peter Zijlstra
2022-11-11 16:35 ` Andrew Cooper
2022-11-11 22:22 ` H. Peter Anvin
2022-11-12 0:08 ` Andrew Cooper
2022-11-11 18:06 ` Li, Xin3
2022-11-11 19:33 ` Peter Zijlstra
2022-11-12 6:35 ` Li, Xin3 [this message]
2022-11-14 4:39 ` Li, Xin3
2022-11-14 9:08 ` Peter Zijlstra
2022-11-15 7:50 ` Li, Xin3
2022-11-15 9:17 ` Peter Zijlstra
2022-11-17 3:37 ` Li, Xin3
2022-11-17 15:51 ` Sean Christopherson
2022-11-18 0:05 ` Li, Xin3
2022-11-22 13:00 ` Li, Xin3
2022-11-22 20:52 ` Sean Christopherson
2022-11-23 8:31 ` Li, Xin3
2022-11-23 20:42 ` Sean Christopherson
2022-11-24 3:40 ` Li, Xin3
2022-11-28 16:26 ` Sean Christopherson
2022-11-24 9:46 ` Peter Zijlstra
2022-11-28 19:05 ` Sean Christopherson
2022-11-23 9:16 ` Peter Zijlstra
2022-11-23 19:18 ` Sean Christopherson
2022-11-11 22:15 ` H. Peter Anvin
2022-11-10 6:15 ` [RESEND PATCH 6/6] x86/traps: remove unused NMI entry exc_nmi_noist() Xin Li
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=BN6PR1101MB216163B12055F02E8C54D2BFA8039@BN6PR1101MB2161.namprd11.prod.outlook.com \
--to=xin3.li@intel.com \
--cc=bp@alien8.de \
--cc=dave.hansen@linux.intel.com \
--cc=hpa@zytor.com \
--cc=kevin.tian@intel.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=pbonzini@redhat.com \
--cc=peterz@infradead.org \
--cc=seanjc@google.com \
--cc=tglx@linutronix.de \
--cc=x86@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;
as well as URLs for NNTP newsgroup(s).