From: Nadav Amit <namit@vmware.com>
To: Sean Christopherson <sean.j.christopherson@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>,
LKML <linux-kernel@vger.kernel.org>, x86 <x86@kernel.org>,
"Kenneth R. Crudup" <kenny@panix.com>,
Paolo Bonzini <pbonzini@redhat.com>,
Fenghua Yu <fenghua.yu@intel.com>,
Xiaoyao Li <xiaoyao.li@intel.com>,
Thomas Hellstrom <thellstrom@vmware.com>,
Tony Luck <tony.luck@intel.com>,
"Peter Zijlstra (Intel)" <peterz@infradead.org>,
Jessica Yu <jeyu@kernel.org>,
Steven Rostedt <rostedt@goodmis.org>
Subject: Re: [patch 2/2] x86/kvm/vmx: Prevent split lock detection induced #AC wreckage
Date: Thu, 2 Apr 2020 15:44:00 +0000 [thread overview]
Message-ID: <18758F52-BB97-4F47-9481-F66AF4465A06@vmware.com> (raw)
In-Reply-To: <20200402153035.GA13879@linux.intel.com>
> On Apr 2, 2020, at 8:30 AM, Sean Christopherson <sean.j.christopherson@intel.com> wrote:
>
> On Thu, Apr 02, 2020 at 02:33:00PM +0200, Thomas Gleixner wrote:
>> Without at least minimal handling for split lock detection induced #AC, VMX
>> will just run into the same problem as the VMWare hypervisor, which was
>> reported by Kenneth.
>>
>> It will inject the #AC blindly into the guest whether the guest is prepared
>> or not.
>>
>> Add the minimal required handling for it:
>>
>> - Check guest state whether CR0.AM is enabled and EFLAGS.AC is set. If
>> so, then the #AC originated from CPL3 and the guest has is prepared to
>> handle it. In this case it does not matter whether the #AC is due to a
>> split lock or a regular unaligned check.
>>
>> - Invoke a minimal split lock detection handler. If the host SLD mode is
>> sld_warn, then handle it in the same way as user space handling works:
>> Emit a warning, disable SLD and mark the current task with TIF_SLD.
>> With that resume the guest without injecting #AC.
>>
>> If the host mode is sld_fatal or sld_off, emit a warning and deliver
>> the exception to user space which can crash and burn itself.
>>
>> Mark the module with MOD_INFO(sld_safe, "Y") so the module loader does not
>> force SLD off.
>
> Some comments below. But, any objection to taking Xiaoyao's patches that
> do effectively the same things, minus the MOD_INFO()? I'll repost them in
> reply to this thread.
IIUC they also deal with emulated split-lock accesses in the host, during
instruction emulation [1]. This seems also to be required, although I am not
sure the approach that he took once emulation encounters a split-lock is
robust.
[1] https://lore.kernel.org/lkml/20200324151859.31068-5-xiaoyao.li@intel.com/
next prev parent reply other threads:[~2020-04-02 15:44 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-02 12:32 [patch 0/2] x86: Prevent Split-Lock-Detection wreckage on VMX hypervisors Thomas Gleixner
2020-04-02 12:32 ` [patch 1/2] x86,module: Detect VMX modules and disable Split-Lock-Detect Thomas Gleixner
2020-04-02 15:23 ` [patch v2 " Peter Zijlstra
2020-04-02 16:20 ` Xiaoyao Li
2020-04-02 16:25 ` Peter Zijlstra
2020-04-02 16:39 ` Nadav Amit
2020-04-02 16:41 ` Xiaoyao Li
2020-04-02 17:34 ` Thomas Gleixner
2020-04-02 17:51 ` Sean Christopherson
2020-04-02 18:51 ` Peter Zijlstra
2020-04-02 20:23 ` Sean Christopherson
2020-04-02 21:04 ` Thomas Gleixner
2020-04-02 21:16 ` Sean Christopherson
2020-04-03 8:09 ` David Laight
2020-04-03 14:33 ` Peter Zijlstra
2020-04-02 23:42 ` [patch " Rasmus Villemoes
2020-04-03 14:35 ` Jessica Yu
2020-04-03 15:21 ` Peter Zijlstra
2020-04-03 16:01 ` Sean Christopherson
2020-04-03 16:12 ` Peter Zijlstra
2020-04-03 16:16 ` David Laight
2020-04-03 16:39 ` Peter Zijlstra
2020-04-03 16:25 ` Sean Christopherson
2020-04-03 16:40 ` Peter Zijlstra
2020-04-03 16:48 ` Nadav Amit
2020-04-03 17:21 ` Sean Christopherson
2020-04-03 18:53 ` Thomas Gleixner
2020-04-03 20:58 ` Andy Lutomirski
2020-04-03 21:49 ` Thomas Gleixner
2020-04-03 11:29 ` kbuild test robot
2020-04-03 14:43 ` kbuild test robot
2020-04-03 16:36 ` Sean Christopherson
2020-04-03 16:41 ` Peter Zijlstra
2020-04-03 18:35 ` Jessica Yu
2020-04-06 12:23 ` Christoph Hellwig
2020-04-06 14:40 ` Peter Zijlstra
2020-04-06 15:18 ` Christoph Hellwig
2020-04-06 15:22 ` Peter Zijlstra
2020-04-06 18:27 ` Steven Rostedt
2020-04-02 12:33 ` [patch 2/2] x86/kvm/vmx: Prevent split lock detection induced #AC wreckage Thomas Gleixner
2020-04-02 15:30 ` Sean Christopherson
2020-04-02 15:44 ` Nadav Amit [this message]
2020-04-02 16:04 ` Sean Christopherson
2020-04-02 16:56 ` Thomas Gleixner
2020-04-02 15:55 ` [PATCH 0/3] x86: KVM: VMX: Add basic split-lock #AC handling Sean Christopherson
2020-04-02 15:55 ` [PATCH 1/3] KVM: x86: Emulate split-lock access as a write in emulator Sean Christopherson
2020-04-02 15:55 ` [PATCH 2/3] x86/split_lock: Refactor and export handle_user_split_lock() for KVM Sean Christopherson
2020-04-02 17:01 ` Thomas Gleixner
2020-04-02 17:19 ` Sean Christopherson
2020-04-02 19:06 ` Thomas Gleixner
2020-04-10 4:39 ` Xiaoyao Li
2020-04-10 10:21 ` Paolo Bonzini
2020-04-02 15:55 ` [PATCH 3/3] KVM: VMX: Extend VMX's #AC interceptor to handle split lock #AC in guest Sean Christopherson
2020-04-02 17:19 ` Thomas Gleixner
2020-04-02 17:40 ` Sean Christopherson
2020-04-02 20:07 ` Thomas Gleixner
2020-04-02 20:36 ` Andy Lutomirski
2020-04-02 20:48 ` Peter Zijlstra
2020-04-02 20:51 ` Sean Christopherson
2020-04-02 22:27 ` Thomas Gleixner
2020-04-02 22:40 ` Nadav Amit
2020-04-02 23:03 ` Thomas Gleixner
2020-04-02 23:08 ` Steven Rostedt
2020-04-02 23:16 ` Kenneth R. Crudup
2020-04-02 23:18 ` Jim Mattson
2020-04-03 12:16 ` Thomas Gleixner
2020-04-10 10:23 ` [PATCH 0/3] x86: KVM: VMX: Add basic split-lock #AC handling Paolo Bonzini
2020-04-10 11:14 ` Thomas Gleixner
2020-04-02 13:43 ` [patch 0/2] x86: Prevent Split-Lock-Detection wreckage on VMX hypervisors Kenneth R. Crudup
2020-04-02 14:32 ` Peter Zijlstra
2020-04-02 14:41 ` Kenneth R. Crudup
2020-04-02 14:46 ` Peter Zijlstra
2020-04-02 14:53 ` Kenneth R. Crudup
2020-04-02 14:37 ` Thomas Gleixner
2020-04-02 14:47 ` Nadav Amit
2020-04-02 15:11 ` Peter Zijlstra
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=18758F52-BB97-4F47-9481-F66AF4465A06@vmware.com \
--to=namit@vmware.com \
--cc=fenghua.yu@intel.com \
--cc=jeyu@kernel.org \
--cc=kenny@panix.com \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=sean.j.christopherson@intel.com \
--cc=tglx@linutronix.de \
--cc=thellstrom@vmware.com \
--cc=tony.luck@intel.com \
--cc=x86@kernel.org \
--cc=xiaoyao.li@intel.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).