From: marc.zyngier@arm•com (Marc Zyngier)
To: linux-arm-kernel@lists•infradead.org
Subject: [PATCH v7 07/16] arm64: kvm: allows kvm cpu hotplug
Date: Mon, 25 Apr 2016 10:28:43 +0100 [thread overview]
Message-ID: <571DE34B.1010300@arm.com> (raw)
In-Reply-To: <571DE077.8060601@arm.com>
On 25/04/16 10:16, James Morse wrote:
> Hi,
>
> On 25/04/16 09:41, AKASHI Takahiro wrote:
>> On Wed, Apr 20, 2016 at 12:19:45PM +0100, James Morse wrote:
>>> On 20/04/16 11:37, Marc Zyngier wrote:
>>>> On 19/04/16 18:37, James Morse wrote:
>>>>> It looks like x86 uses the extable to work around this, their vmx_vcpu_run() has:
>>>>>> __ex(ASM_VMX_VMLAUNCH) "\n\t"
>>>>> Where __ex ends up calling ____kvm_handle_fault_on_reboot(), with a nearby comment:
>>>>>> * Hardware virtualization extension instructions may fault if a
>>>>>> * reboot turns off virtualization while processes are running.
>>>>>> * Trap the fault and ignore the instruction if that happens.
>>>>
>>>> I very much like that approach, to be honest. Tearing down a CPU is
>>>> something exceptional, so let's make it an actual exception.
>>>>
>>>> It is now pretty easy to discriminate between KVM functions and stub
>>>> functions thanks to your earlier patch, so if we end up calling the
>>>> hyp-stub because we've torn down KVM's EL2, let's just return an
>>>> appropriate error code (ARM_EXCEPTION_HYP_GONE), and handle it at EL1.
>>>
>>> Okay. kexec uses kvm_call_hyp() against the hyp-stub to do the kernel-copy and
>>> hand over to purgatory, but we could change that to a new 'special' builtin
>>> call, something like HVC_KEXEC_CALL_HYP. It never calls it with kvm loaded, so
>>> there is no reason the calls have to be same.
>>>
>>> Given hibernate doesn't hit this issue, I will drop this hunk from this version
>>> of the patch, and repost hibernate incorporating the feedback so far. I will
>>> provide a patch for kexec to do the above.
>>
>> Thanks, but you don' have to.
>
> I was wrong with the 'hibernate doesn't hit this issue', with this patch we
> re-install the hyp-stub during system reboot, and race with the scheduler.
> ('reboot -f' while running a guest).
>
>
>> If the fix below is acceptable, we will merge it to our next kexec/kdump
>> patch series.
>
> I'm testing something that looks very similar at the moment.
>
>
>> diff --git a/arch/arm64/kvm/handle_exit.c b/arch/arm64/kvm/handle_exit.c
>> index eba89e4..31b5224 100644
>> --- a/arch/arm64/kvm/handle_exit.c
>> +++ b/arch/arm64/kvm/handle_exit.c
>> @@ -186,6 +186,10 @@ int handle_exit(struct kvm_vcpu *vcpu, struct kvm_run *run,
>> exit_handler = kvm_get_exit_handler(vcpu);
>>
>> return exit_handler(vcpu, run);
>> + case ARM_EXCEPTION_HYP_GONE:
>> + /* due to kexec reboot */
>> + run->exit_reason = KVM_EXIT_SHUTDOWN;
>> + return 0;
>
> Is it fair to throw this back out to user space? While the hypervisor doesn't
> have long to live, it may not be expecting this exit_reason. I couldn't see a
> value for 'suprise cpu removal', and it looks like the x86 code causes the vcpu
> to spin round enter guest...
Yeah, KVM_EXIT_SHUTDOWN is a vcpu reset (which on x86 is caused by a
triple fault). KVM_EXIT_FAIL_ENTRY seems slightly better.
As for getting back to userspace, I don't see that as a problem (though
the documentation on that part of the API is... sparse).
Thanks,
M.
--
Jazz is not dead. It just smells funny...
next prev parent reply other threads:[~2016-04-25 9:28 UTC|newest]
Thread overview: 60+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-04-01 16:53 [PATCH v7 00/16] arm64: kernel: Add support for hibernate/suspend-to-disk James Morse
2016-04-01 16:53 ` [PATCH v7 01/16] arm64: KVM: Register CPU notifiers when the kernel runs at HYP James Morse
2016-04-18 16:10 ` Catalin Marinas
2016-04-19 8:58 ` James Morse
2016-04-19 14:39 ` Marc Zyngier
2016-04-01 16:53 ` [PATCH v7 02/16] arm64: Fold proc-macros.S into assembler.h James Morse
2016-04-18 16:11 ` Catalin Marinas
2016-04-01 16:53 ` [PATCH v7 03/16] arm64: Cleanup SCTLR flags James Morse
2016-04-19 14:44 ` Marc Zyngier
2016-04-01 16:53 ` [PATCH v7 04/16] arm64: kvm: Move the do_el2_call macro to a header file James Morse
2016-04-19 15:02 ` Marc Zyngier
2016-04-19 15:05 ` James Morse
2016-04-19 15:10 ` Marc Zyngier
2016-04-01 16:53 ` [PATCH v7 05/16] arm64: kvm: Move lr save/restore from do_el2_call into EL1 James Morse
2016-04-19 15:11 ` Marc Zyngier
2016-04-01 16:53 ` [PATCH v7 06/16] arm64: hyp/kvm: Extend hyp-stub API to allow function calls at EL2 James Morse
2016-04-19 15:22 ` Marc Zyngier
2016-04-01 16:53 ` [PATCH v7 07/16] arm64: kvm: allows kvm cpu hotplug James Morse
2016-04-19 16:03 ` Marc Zyngier
2016-04-19 17:37 ` James Morse
2016-04-20 10:29 ` AKASHI Takahiro
2016-04-20 11:19 ` James Morse
2016-04-20 10:37 ` Marc Zyngier
2016-04-20 11:19 ` James Morse
2016-04-20 11:46 ` Marc Zyngier
2016-04-25 8:41 ` AKASHI Takahiro
2016-04-25 9:16 ` James Morse
2016-04-25 9:28 ` Marc Zyngier [this message]
2016-04-01 16:53 ` [PATCH v7 08/16] arm64: kernel: Rework finisher callback out of __cpu_suspend_enter() James Morse
2016-04-18 17:20 ` Catalin Marinas
2016-04-01 16:53 ` [PATCH v7 09/16] arm64: Change cpu_resume() to enable mmu early then access sleep_sp by va James Morse
2016-04-20 16:24 ` Catalin Marinas
2016-04-01 16:53 ` [PATCH v7 10/16] arm64: kernel: Include _AC definition in page.h James Morse
2016-04-20 16:25 ` Catalin Marinas
2016-04-01 16:53 ` [PATCH v7 11/16] arm64: Promote KERNEL_START/KERNEL_END definitions to a header file James Morse
2016-04-20 16:26 ` Catalin Marinas
2016-04-01 16:53 ` [PATCH v7 12/16] arm64: Add new asm macro copy_page James Morse
2016-04-20 16:38 ` Catalin Marinas
2016-04-20 16:56 ` James Morse
2016-04-01 16:53 ` [PATCH v7 13/16] arm64: head.S: el2_setup() to accept sctlr_el1 as an argument James Morse
2016-04-20 17:12 ` Catalin Marinas
2016-04-20 17:35 ` James Morse
2016-04-22 10:36 ` Catalin Marinas
2016-04-01 16:53 ` [PATCH v7 14/16] PM / Hibernate: Call flush_icache_range() on pages restored in-place James Morse
2016-04-20 17:16 ` Catalin Marinas
2016-04-01 16:53 ` [PATCH v7 15/16] arm64: kernel: Add support for hibernate/suspend-to-disk James Morse
2016-04-22 10:29 ` Catalin Marinas
2016-04-25 9:19 ` James Morse
2016-04-01 16:53 ` [PATCH v7 16/16] arm64: hibernate: Prevent resume from a different kernel version James Morse
2016-04-10 12:16 ` Ard Biesheuvel
2016-04-13 16:35 ` James Morse
2016-04-13 16:31 ` [PATCH v7 17/16] arm64: hibernate: Refuse to hibernate if the boot cpu is offline James Morse
2016-04-21 11:33 ` Lorenzo Pieralisi
2016-04-21 11:44 ` Mark Rutland
2016-04-21 12:33 ` Mark Rutland
2016-04-21 16:28 ` Lorenzo Pieralisi
2016-04-22 10:41 ` Mark Rutland
2016-04-22 15:32 ` James Morse
2016-04-22 10:41 ` Catalin Marinas
2016-04-22 15:32 ` James Morse
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=571DE34B.1010300@arm.com \
--to=marc.zyngier@arm$(echo .)com \
--cc=linux-arm-kernel@lists$(echo .)infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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