Re: [PATCH 4.9 V2 09/24] ARM: spectre-v2: add firmware based hardening

From: David Long <dave.long@linaro.org>
To: Russell King - ARM Linux <linux@armlinux.org.uk>,
	Florian Fainelli <f.fainelli@gmail.com>
Cc: Marc Zyngier <marc.zyngier@arm.com>,
	stable@vger.kernel.org, Tony Lindgren <tony@atomide.com>,
	Mark Rutland <mark.rutland@arm.com>,
	Greg KH <gregkh@linuxfoundation.org>,
	Mark Brown <broonie@kernel.org>
Subject: Re: [PATCH 4.9 V2 09/24] ARM: spectre-v2: add firmware based hardening
Date: Tue, 20 Nov 2018 11:24:22 -0500	[thread overview]
Message-ID: <51e1f3a1-d140-98f6-2db8-29153631cce5@linaro.org> (raw)
In-Reply-To: <20181120105943.GC30658@n2100.armlinux.org.uk>

On 11/20/18 5:59 AM, Russell King - ARM Linux wrote:
> All,
> 
> It is now almost two weeks since David posted the patches, and a week
> since the last message in this thread.
> 
> What is happening with these patches?  What about the 4.4-stable
> backport as well?  Does anyone care about these anymore?  Do we have
> any product customers using 32-bit Cortex CPUs anymore?

Please note that the v4.4 patches I sent out do not include the kvm 
changes from the original patch set, therefore they were never affected 
by either of the two issues we've been discussing.  The kvm-unit-tests 
for v4.4 look OK, with or without a thumb-2 host kernel.

> 
> David - FYI - everything in my Spectre branch is now in mainline as
> of 4.20-rc3, and I have nothing further planned for core 32-bit ARM
> Spectre workarounds.
> 
> On Tue, Nov 13, 2018 at 10:08:48AM -0800, Florian Fainelli wrote:
>> On 11/13/18 6:23 AM, Marc Zyngier wrote:
>>> Russell,
>>>
>>> On Mon, 12 Nov 2018 16:54:10 +0000,
>>> Russell King - ARM Linux <linux@armlinux.org.uk> wrote:
>>>>
>>>> Marc,
>>>>
>>>> Can you please ack this to say that you are now happy with it after
>>>> your comments on version 1, so we can move forward and have Greg
>>>> merge it.
>>>>
>>>> Thanks.
>>>>
>>>> On Wed, Nov 07, 2018 at 11:43:47AM -0500, David Long wrote:
>>>>> From: Russell King <rmk+kernel@armlinux.org.uk>
>>>>>
>>>>> Commit 10115105cb3aa17b5da1cb726ae8dd5f6854bd93 upstream.
>>>>> Commit 6282e916f774e37845c65d1eae9f8c649004f033 upstream.
>>>>>
>>>>> Add firmware based hardening for cores that require more complex
>>>>> handling in firmware.
>>>>>
>>>>> Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
>>>>> Boot-tested-by: Tony Lindgren <tony@atomide.com>
>>>>> Reviewed-by: Tony Lindgren <tony@atomide.com>
>>>>> Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
>>>>> Signed-off-by: David A. Long <dave.long@linaro.org>
>>>
>>> Sure. Feel free to add my
>>>
>>> Acked-by: Marc Zyngier <marc.zyngier@arm.com>
>>>
>>> I assume someone has tested these patches (I haven't, and I'm unlikely
>>> to do so in the near future as I'm travelling). I'm not sure Tony's
>>> "Boot-tested-by" is still valid, and Florian's earlier set of tests
>>> didn't show the issues of the initial backport.
>>
>> Correct, I was not testing any KVM path at all, which is why this did
>> not show up as a problem for me, I am not really well equipped to
>> perform KVM testing at the moment.
>> -- 
>> Florian
> 

-dl