From: "H. Peter Anvin" <hpa@zytor.com>
To: andrew.cooper3@citrix.com, Lai Jiangshan <jiangshanlai@gmail.com>,
Xin Li <xin3.li@intel.com>
Cc: linux-kernel@vger.kernel.org, x86@kernel.org,
kvm@vger.kernel.org, tglx@linutronix.de, mingo@redhat.com,
bp@alien8.de, dave.hansen@linux.intel.com, peterz@infradead.org,
seanjc@google.com, pbonzini@redhat.com, ravi.v.shankar@intel.com
Subject: Re: [PATCH v5 28/34] x86/fred: fixup fault on ERETU by jumping to fred_entrypoint_user
Date: Fri, 17 Mar 2023 14:00:16 -0700 [thread overview]
Message-ID: <D3DB8A8F-7293-4715-AAB6-213F00CB521C@zytor.com> (raw)
In-Reply-To: <ed318bd6-25b2-efcf-0cc4-c57699f6654a@citrix.com>
On March 17, 2023 2:55:44 AM PDT, andrew.cooper3@citrix.com wrote:
>On 17/03/2023 9:39 am, Lai Jiangshan wrote:
>>> +#ifdef CONFIG_X86_FRED
>>> +static bool ex_handler_eretu(const struct exception_table_entry *fixup,
>>> + struct pt_regs *regs, unsigned long error_code)
>>> +{
>>> + struct pt_regs *uregs = (struct pt_regs *)(regs->sp - offsetof(struct pt_regs, ip));
>>> + unsigned short ss = uregs->ss;
>>> + unsigned short cs = uregs->cs;
>>> +
>>> + fred_info(uregs)->edata = fred_event_data(regs);
>>> + uregs->ssx = regs->ssx;
>>> + uregs->ss = ss;
>>> + uregs->csx = regs->csx;
>>> + uregs->current_stack_level = 0;
>>> + uregs->cs = cs;
>> Hello
>>
>> If the ERETU instruction had tried to return from NMI to ring3 and just faulted,
>> is NMI still blocked?
>>
>> We know that IRET unconditionally enables NMI, but I can't find any clue in the
>> FRED's manual.
>>
>> In the pseudocode of ERETU in the manual, it seems that NMI is only enabled when
>> ERETU succeeds with bit28 in csx set. If so, this code will fail to reenable
>> NMI if bit28 is not explicitly re-set in csx.
>
>IRET clearing NMI blocking is the source of an immense amount of grief,
>and ultimately the reason why Linux and others can't use supervisor
>shadow stacks at the moment.
>
>Changing this property, so NMIs only get unblocked on successful
>execution of an ERET{S,U}, was a key demand of the FRED spec.
>
>i.e. until you have successfully ERET*'d, you're still logically in the
>NMI handler and NMIs need to remain blocked even when handling the #GP
>from a bad ERET.
>
>~Andrew
This is correct.
next prev parent reply other threads:[~2023-03-17 21:01 UTC|newest]
Thread overview: 80+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-07 2:39 [PATCH v5 00/34] x86: enable FRED for x86-64 Xin Li
2023-03-07 2:39 ` [PATCH v5 01/34] x86/traps: let common_interrupt() handle IRQ_MOVE_CLEANUP_VECTOR Xin Li
2023-03-07 2:39 ` [PATCH v5 02/34] x86/traps: add a system interrupt table for system interrupt dispatch Xin Li
2023-03-07 2:39 ` [PATCH v5 03/34] x86/traps: add install_system_interrupt_handler() Xin Li
2023-03-07 2:39 ` [PATCH v5 04/34] x86/traps: add external_interrupt() to dispatch external interrupts Xin Li
2023-03-20 15:36 ` Peter Zijlstra
2023-03-20 17:42 ` Peter Zijlstra
2023-03-20 23:47 ` Li, Xin3
2023-03-20 17:53 ` Li, Xin3
2023-03-07 2:39 ` [PATCH v5 05/34] x86/traps: export external_interrupt() for VMX IRQ reinjection Xin Li
2023-03-22 17:52 ` Sean Christopherson
2023-03-22 22:38 ` Li, Xin3
2023-03-07 2:39 ` [PATCH v5 06/34] x86/cpufeature: add the cpu feature bit for FRED Xin Li
2023-03-07 2:39 ` [PATCH v5 07/34] x86/opcode: add ERETU, ERETS instructions to x86-opcode-map Xin Li
2023-03-07 2:39 ` [PATCH v5 08/34] x86/objtool: teach objtool about ERETU and ERETS Xin Li
2023-03-07 2:39 ` [PATCH v5 09/34] x86/cpu: add X86_CR4_FRED macro Xin Li
2023-03-07 2:39 ` [PATCH v5 10/34] x86/fred: add Kconfig option for FRED (CONFIG_X86_FRED) Xin Li
2023-03-07 2:39 ` [PATCH v5 11/34] x86/fred: if CONFIG_X86_FRED is disabled, disable FRED support Xin Li
2023-03-07 2:39 ` [PATCH v5 12/34] x86/cpu: add MSR numbers for FRED configuration Xin Li
2023-03-07 2:39 ` [PATCH v5 13/34] x86/fred: header file for event types Xin Li
2023-03-07 2:39 ` [PATCH v5 14/34] x86/fred: header file with FRED definitions Xin Li
2023-03-07 2:39 ` [PATCH v5 15/34] x86/fred: make unions for the cs and ss fields in struct pt_regs Xin Li
2023-03-07 2:39 ` [PATCH v5 16/34] x86/fred: reserve space for the FRED stack frame Xin Li
2023-03-07 2:39 ` [PATCH v5 17/34] x86/fred: add a page fault entry stub for FRED Xin Li
2023-03-07 2:39 ` [PATCH v5 18/34] x86/fred: add a debug " Xin Li
2023-03-07 2:39 ` [PATCH v5 19/34] x86/fred: add a NMI " Xin Li
2023-03-07 2:39 ` [PATCH v5 20/34] x86/fred: add a machine check " Xin Li
2023-03-20 16:00 ` Peter Zijlstra
2023-03-21 0:04 ` Li, Xin3
2023-03-21 8:59 ` Peter Zijlstra
2023-03-21 16:38 ` Li, Xin3
2023-03-07 2:39 ` [PATCH v5 21/34] x86/fred: FRED entry/exit and dispatch code Xin Li
2023-03-07 2:39 ` [PATCH v5 22/34] x86/fred: FRED initialization code Xin Li
2023-03-17 13:35 ` Lai Jiangshan
2023-03-17 21:32 ` H. Peter Anvin
2023-03-18 6:33 ` Lai Jiangshan
2023-03-20 16:49 ` Peter Zijlstra
2023-03-21 0:12 ` Li, Xin3
2023-03-21 1:02 ` andrew.cooper3
2023-03-21 7:49 ` Li, Xin3
2023-03-22 16:29 ` Dave Hansen
2023-03-22 2:22 ` Li, Xin3
2023-03-22 4:01 ` Dave Hansen
2023-03-22 5:40 ` Li, Xin3
2023-03-22 18:25 ` andrew.cooper3
2023-03-20 16:44 ` Peter Zijlstra
2023-03-21 0:13 ` Li, Xin3
2023-03-07 2:39 ` [PATCH v5 23/34] x86/fred: update MSR_IA32_FRED_RSP0 during task switch Xin Li
2023-03-20 16:52 ` Peter Zijlstra
2023-03-20 23:54 ` Li, Xin3
2023-03-07 2:39 ` [PATCH v5 24/34] x86/fred: let ret_from_fork() jmp to fred_exit_user when FRED is enabled Xin Li
2023-03-07 2:39 ` [PATCH v5 25/34] x86/fred: disallow the swapgs instruction " Xin Li
2023-03-20 16:54 ` Peter Zijlstra
2023-03-20 17:58 ` Li, Xin3
2023-03-07 2:39 ` [PATCH v5 26/34] x86/fred: no ESPFIX needed " Xin Li
2023-03-07 2:39 ` [PATCH v5 27/34] x86/fred: allow single-step trap and NMI when starting a new thread Xin Li
2023-03-07 2:39 ` [PATCH v5 28/34] x86/fred: fixup fault on ERETU by jumping to fred_entrypoint_user Xin Li
2023-03-17 9:39 ` Lai Jiangshan
2023-03-17 9:55 ` andrew.cooper3
2023-03-17 13:02 ` Lai Jiangshan
2023-03-17 21:23 ` H. Peter Anvin
2023-03-17 21:00 ` H. Peter Anvin [this message]
2023-03-18 7:55 ` Li, Xin3
2023-03-07 2:39 ` [PATCH v5 29/34] x86/ia32: do not modify the DPL bits for a null selector Xin Li
2023-03-07 2:39 ` [PATCH v5 30/34] x86/fred: allow FRED systems to use interrupt vectors 0x10-0x1f Xin Li
2023-03-07 2:39 ` [PATCH v5 31/34] x86/fred: allow dynamic stack frame size Xin Li
2023-03-07 2:39 ` [PATCH v5 32/34] x86/fred: disable FRED by default in its early stage Xin Li
2023-03-07 2:39 ` [PATCH v5 33/34] KVM: x86/vmx: call external_interrupt() to handle IRQ in IRQ caused VM exits Xin Li
2023-03-22 17:57 ` Sean Christopherson
2023-03-07 2:39 ` [PATCH v5 34/34] KVM: x86/vmx: execute "int $2" to handle NMI in NMI caused VM exits when FRED is enabled Xin Li
2023-03-07 22:00 ` Li, Xin3
2023-03-22 17:49 ` Sean Christopherson
2023-03-22 23:03 ` andrew.cooper3
2023-03-22 23:42 ` Sean Christopherson
2023-03-23 0:26 ` Li, Xin3
2023-03-24 17:45 ` Li, Xin3
2023-03-22 23:43 ` Li, Xin3
2023-03-11 9:58 ` [PATCH v5 00/34] x86: enable FRED for x86-64 Kang, Shan
2023-03-11 21:29 ` Li, Xin3
2023-03-20 7:40 ` Kang, Shan
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=D3DB8A8F-7293-4715-AAB6-213F00CB521C@zytor.com \
--to=hpa@zytor.com \
--cc=andrew.cooper3@citrix.com \
--cc=bp@alien8.de \
--cc=dave.hansen@linux.intel.com \
--cc=jiangshanlai@gmail.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=ravi.v.shankar@intel.com \
--cc=seanjc@google.com \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
--cc=xin3.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).