From: AKASHI Takahiro <takahiro.akashi@linaro.org>
To: Geoff Levand <geoff@infradead.org>
Cc: Mark Rutland <mark.rutland@arm.com>,
Marc Zyngier <marc.zyngier@arm.com>,
Catalin Marinas <Catalin.Marinas@arm.com>,
Will Deacon <Will.Deacon@arm.com>,
"linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>,
"grant.likely@linaro.org" <grant.likely@linaro.org>,
"kexec@lists.infradead.org" <kexec@lists.infradead.org>,
"christoffer.dall@linaro.org" <christoffer.dall@linaro.org>
Subject: Re: [PATCH 7/8] arm64/kexec: Add checks for KVM
Date: Fri, 06 Feb 2015 13:18:53 +0900 [thread overview]
Message-ID: <54D440AD.2030608@linaro.org> (raw)
In-Reply-To: <1423181494.16019.2.camel@infradead.org>
On 02/06/2015 09:11 AM, Geoff Levand wrote:
> Hi Takahiro,
>
> On Mon, 2015-02-02 at 17:18 +0900, AKASHI Takahiro wrote:
>> On 01/31/2015 04:48 AM, Geoff Levand wrote:
>>> As an initial implementation we can hook into the CPU_DYING_FROZEN
>>> notifier sent to hyp_init_cpu_notify(). The longer term solution
>>> should use kvm_arch_hardware_enable() and kvm_arch_hardware_disable().
>>
>> Are these two different approaches?
>
> Yes, these are two different solutions, One initial work-around, and a
> more involved proper solution. Hooking into the CPU_DYING_FROZEN
> notifier would be a initial fix. The proper solution would be to move
> the KVM setup to kvm_arch_hardware_enable(), and the shutdown to
> kvm_arch_hardware_disable().
>
>
>> kernel_exec() -> machine_shutdown() -> disable_nonboot_cpu()
>> -> _cpu_down() -> cpu_notify_nofail(CPU_DEAD|...)
>>
>> On the other hand, kvm already has a hook into kvm_arch_hardware_disable():
>> (ignoring kvm_usage_count here)
>> kvm_cpu_hotplug(CPU_DYING) -> hardware_disable()
>> -> hardware_disable_nolock() -> kvm_arch_hardware_disable()
>>
>> So it seems that we don't have to add a new hook at hyp_init_cpu_notify()
>> if kvm_arch_hardware_disable() is properly implemented.
>
> Yes, that is correct. But, as above, you would also need to update the
> KVM startup to use kvm_arch_hardware_enable().
>
>> disable_nonboot_cpu() will not inovke cpu hotplug on *boot* cpu, and
>> we should handle it in a separate way though.
>
> IIRC, the secondary cpus go through PSCI on shutdown, and that path
> is working OK. Maybe I am mistaken though.
If so, why should we add a hook at hyp_init_cpu_notify() as initial work-around?
> The primary cpu shutdown (hyp stubs restored) is what is missing. The
> primary cpu goes through cpu_soft_restart(), and that is what is
> currently failing.
Yeah, we will call teardown function manually in soft_restart();
-Takahiro AKASHI
>
> -Geoff
>
_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec
WARNING: multiple messages have this Message-ID (diff)
From: takahiro.akashi@linaro.org (AKASHI Takahiro)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 7/8] arm64/kexec: Add checks for KVM
Date: Fri, 06 Feb 2015 13:18:53 +0900 [thread overview]
Message-ID: <54D440AD.2030608@linaro.org> (raw)
In-Reply-To: <1423181494.16019.2.camel@infradead.org>
On 02/06/2015 09:11 AM, Geoff Levand wrote:
> Hi Takahiro,
>
> On Mon, 2015-02-02 at 17:18 +0900, AKASHI Takahiro wrote:
>> On 01/31/2015 04:48 AM, Geoff Levand wrote:
>>> As an initial implementation we can hook into the CPU_DYING_FROZEN
>>> notifier sent to hyp_init_cpu_notify(). The longer term solution
>>> should use kvm_arch_hardware_enable() and kvm_arch_hardware_disable().
>>
>> Are these two different approaches?
>
> Yes, these are two different solutions, One initial work-around, and a
> more involved proper solution. Hooking into the CPU_DYING_FROZEN
> notifier would be a initial fix. The proper solution would be to move
> the KVM setup to kvm_arch_hardware_enable(), and the shutdown to
> kvm_arch_hardware_disable().
>
>
>> kernel_exec() -> machine_shutdown() -> disable_nonboot_cpu()
>> -> _cpu_down() -> cpu_notify_nofail(CPU_DEAD|...)
>>
>> On the other hand, kvm already has a hook into kvm_arch_hardware_disable():
>> (ignoring kvm_usage_count here)
>> kvm_cpu_hotplug(CPU_DYING) -> hardware_disable()
>> -> hardware_disable_nolock() -> kvm_arch_hardware_disable()
>>
>> So it seems that we don't have to add a new hook at hyp_init_cpu_notify()
>> if kvm_arch_hardware_disable() is properly implemented.
>
> Yes, that is correct. But, as above, you would also need to update the
> KVM startup to use kvm_arch_hardware_enable().
>
>> disable_nonboot_cpu() will not inovke cpu hotplug on *boot* cpu, and
>> we should handle it in a separate way though.
>
> IIRC, the secondary cpus go through PSCI on shutdown, and that path
> is working OK. Maybe I am mistaken though.
If so, why should we add a hook at hyp_init_cpu_notify() as initial work-around?
> The primary cpu shutdown (hyp stubs restored) is what is missing. The
> primary cpu goes through cpu_soft_restart(), and that is what is
> currently failing.
Yeah, we will call teardown function manually in soft_restart();
-Takahiro AKASHI
>
> -Geoff
>
next prev parent reply other threads:[~2015-02-06 4:19 UTC|newest]
Thread overview: 100+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <cover.1415926876.git.geoff@infradead.orgg>
2015-01-17 0:23 ` [PATCH 0/8] arm64 kexec kernel patches V7 Geoff Levand
2015-01-17 0:23 ` Geoff Levand
2015-01-17 0:23 ` [PATCH 8/8] arm64/kexec: Enable kexec in the arm64 defconfig Geoff Levand
2015-01-17 0:23 ` Geoff Levand
2015-01-17 0:23 ` [PATCH 6/8] arm64/kexec: Add pr_devel output Geoff Levand
2015-01-17 0:23 ` Geoff Levand
2015-01-17 0:23 ` [PATCH 1/8] arm64: Move proc-macros.S to include/asm Geoff Levand
2015-01-17 0:23 ` Geoff Levand
2015-01-26 17:45 ` Catalin Marinas
2015-01-26 17:45 ` Catalin Marinas
2015-01-27 19:33 ` [PATCH V2 1/8] arm64: Fold proc-macros.S into assembler.h Geoff Levand
2015-01-27 19:33 ` Geoff Levand
2015-01-17 0:23 ` [PATCH 7/8] arm64/kexec: Add checks for KVM Geoff Levand
2015-01-17 0:23 ` Geoff Levand
2015-01-26 19:19 ` Mark Rutland
2015-01-26 19:19 ` Mark Rutland
2015-01-26 20:39 ` Christoffer Dall
2015-01-26 20:39 ` Christoffer Dall
2015-01-26 20:58 ` Geoff Levand
2015-01-26 20:58 ` Geoff Levand
2015-01-26 21:00 ` Geoff Levand
2015-01-26 21:00 ` Geoff Levand
2015-01-29 9:36 ` AKASHI Takahiro
2015-01-29 9:36 ` AKASHI Takahiro
2015-01-29 9:57 ` AKASHI Takahiro
2015-01-29 9:57 ` AKASHI Takahiro
2015-01-29 10:59 ` Marc Zyngier
2015-01-29 10:59 ` Marc Zyngier
2015-01-29 18:47 ` Mark Rutland
2015-01-29 18:47 ` Mark Rutland
2015-01-30 6:10 ` AKASHI Takahiro
2015-01-30 6:10 ` AKASHI Takahiro
2015-01-30 12:14 ` Mark Rutland
2015-01-30 12:14 ` Mark Rutland
2015-01-30 19:48 ` Geoff Levand
2015-01-30 19:48 ` Geoff Levand
2015-02-02 8:18 ` AKASHI Takahiro
2015-02-02 8:18 ` AKASHI Takahiro
2015-02-06 0:11 ` Geoff Levand
2015-02-06 0:11 ` Geoff Levand
2015-02-06 4:18 ` AKASHI Takahiro [this message]
2015-02-06 4:18 ` AKASHI Takahiro
2015-02-06 7:06 ` Geoff Levand
2015-02-06 7:06 ` Geoff Levand
2015-01-17 0:23 ` [PATCH 3/8] arm64: Add new hcall HVC_CALL_FUNC Geoff Levand
2015-01-17 0:23 ` Geoff Levand
2015-01-27 17:39 ` Catalin Marinas
2015-01-27 17:39 ` Catalin Marinas
2015-01-27 18:00 ` Mark Rutland
2015-01-27 18:00 ` Mark Rutland
2015-01-30 21:52 ` Geoff Levand
2015-01-30 21:52 ` Geoff Levand
2015-01-17 0:23 ` [PATCH 2/8] arm64: Convert hcalls to use ISS field Geoff Levand
2015-01-17 0:23 ` Geoff Levand
2015-01-26 18:26 ` Catalin Marinas
2015-01-26 18:26 ` Catalin Marinas
2015-01-30 23:31 ` Geoff Levand
2015-01-30 23:31 ` Geoff Levand
2015-02-02 16:04 ` Catalin Marinas
2015-02-02 16:04 ` Catalin Marinas
2015-01-30 23:33 ` [PATCH v2 " Geoff Levand
2015-01-30 23:33 ` Geoff Levand
2015-02-19 20:57 ` Christoffer Dall
2015-02-19 20:57 ` Christoffer Dall
2015-02-25 22:09 ` Geoff Levand
2015-02-25 22:09 ` Geoff Levand
2015-03-02 22:13 ` Christoffer Dall
2015-03-02 22:13 ` Christoffer Dall
2015-03-02 23:22 ` Geoff Levand
2015-03-02 23:22 ` Geoff Levand
2015-03-03 21:47 ` Christopher Covington
2015-03-03 21:47 ` Christopher Covington
2015-03-03 22:35 ` Geoff Levand
2015-03-03 22:35 ` Geoff Levand
2015-01-17 0:23 ` [PATCH 4/8] arm64: Add EL2 switch to soft_restart Geoff Levand
2015-01-17 0:23 ` Geoff Levand
2015-01-26 19:02 ` Mark Rutland
2015-01-26 19:02 ` Mark Rutland
2015-01-26 21:48 ` Geoff Levand
2015-01-26 21:48 ` Geoff Levand
2015-01-27 16:46 ` Mark Rutland
2015-01-27 16:46 ` Mark Rutland
2015-01-27 18:34 ` Geoff Levand
2015-01-27 18:34 ` Geoff Levand
2015-01-27 17:57 ` Catalin Marinas
2015-01-27 17:57 ` Catalin Marinas
2015-01-30 21:47 ` Geoff Levand
2015-01-30 21:47 ` Geoff Levand
2015-01-17 0:23 ` [PATCH 5/8] arm64/kexec: Add core kexec support Geoff Levand
2015-01-17 0:23 ` Geoff Levand
2015-01-26 19:16 ` Mark Rutland
2015-01-26 19:16 ` Mark Rutland
2015-01-26 17:44 ` [PATCH 0/8] arm64 kexec kernel patches V7 Catalin Marinas
2015-01-26 17:44 ` Catalin Marinas
2015-01-26 18:37 ` Grant Likely
2015-01-26 18:37 ` Grant Likely
2015-01-26 18:55 ` Mark Rutland
2015-01-26 18:55 ` Mark Rutland
2015-01-26 20:57 ` Geoff Levand
2015-01-26 20:57 ` Geoff Levand
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=54D440AD.2030608@linaro.org \
--to=takahiro.akashi@linaro.org \
--cc=Catalin.Marinas@arm.com \
--cc=Will.Deacon@arm.com \
--cc=christoffer.dall@linaro.org \
--cc=geoff@infradead.org \
--cc=grant.likely@linaro.org \
--cc=kexec@lists.infradead.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=marc.zyngier@arm.com \
--cc=mark.rutland@arm.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.