All of lore.kernel.org
 help / color / mirror / Atom feed
From: Julien Grall <julien@xen.org>
To: Bertrand Marquis <Bertrand.Marquis@arm.com>
Cc: "xen-devel@lists.xenproject.org" <xen-devel@lists.xenproject.org>,
	Stefano Stabellini <sstabellini@kernel.org>,
	Volodymyr Babchuk <Volodymyr_Babchuk@epam.com>
Subject: Re: [PATCH 3/3] xen/arm: Add sb instruction support
Date: Mon, 9 May 2022 11:31:35 +0100	[thread overview]
Message-ID: <0ac6d6c0-4f34-ad1e-c473-2278dd7289c9@xen.org> (raw)
In-Reply-To: <15DEAB6E-3130-49E8-87BD-AC82C08C7698@arm.com>



On 09/05/2022 11:08, Bertrand Marquis wrote:
> Hi Julien,

Hi,

>> On 4 May 2022, at 09:06, Julien Grall <julien@xen.org> wrote:
>>
>>
>>
>> On 04/05/2022 08:24, Bertrand Marquis wrote:
>>> Hi Julien,
>>
>> Hi Bertrand,
>>
>>>> On 3 May 2022, at 19:47, Julien Grall <julien@xen.org> wrote:
>>>>> A new cpuerrata capability is introduced to enable the alternative
>>>>
>>>> 'sb' is definitely not an erratum. Errata are for stuff that are meant to be specific to one (or multiple) CPU and they are not part of the architecture.
>>>>
>>>> This is the first time we introduce a feature in Xen. So we need to add a new array in cpufeature.c that will cover 'SB' for now. In future we could add feature like pointer auth, LSE atomics...
>>> I am not quite sure why you would want to do that.
>>> Using the sb instruction is definitely something to do to solve erratas, if a CPU is not impacted by those erratas, why would you want to use this ?
>>
>> I agree that SB is used to solve errata but the instruction itself is not a workaround (it may be part of it though). Instead, this is a more efficient way to prevent speculation and will replace dsb/isb.
>>
>> Speculation is never going to disappear from processor. So, in the future, there might be valid reason for us to say "We don't want the processor to speculate". This would mean using SB.
> 
> If the need arise then we will check depending on that how we can support it but in the current status as there is no use case I don’t think we need that.

It is not clear how I should read this answer... If you add SB in 
cpuerrata.c, then a user will start to see message like:

"enabled workaround for Speculation Barrier".

Which is completely bogus. Replacing "dsb; isb" with "sb" is mostly an 
optimization and none of the current use will end up to be 
architecturaly executed.

I appreciate this is more work to add cpufeature.c. However, AFAIK, 
there are no rush to get this optimization in (see why above) and muddy 
(even temporarily) the logs.

Cheers,

-- 
Julien Grall


  reply	other threads:[~2022-05-09 10:32 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-03  9:38 [PATCH 0/3] Spectre BHB follow up Bertrand Marquis
2022-05-03  9:38 ` [PATCH 1/3] xen/arm: Sync sysregs and cpuinfo with Linux 5.18-rc3 Bertrand Marquis
2022-05-03 18:08   ` Julien Grall
2022-05-04  7:39     ` Bertrand Marquis
2022-05-04  8:20       ` Julien Grall
2022-05-04  9:49         ` Bertrand Marquis
2022-05-04 11:49           ` Julien Grall
2022-05-10  2:03             ` Stefano Stabellini
2022-05-11 14:41               ` Bertrand Marquis
2022-05-11 15:20                 ` Julien Grall
2022-05-11 15:40                   ` Bertrand Marquis
2022-05-11 15:47                     ` Julien Grall
2022-05-11 16:01                       ` Bertrand Marquis
2022-05-11 16:25                         ` Julien Grall
2022-05-11 20:06                     ` Stefano Stabellini
2022-05-12 12:34                       ` Bertrand Marquis
2022-05-10  2:04   ` Stefano Stabellini
2022-05-11 14:41     ` Bertrand Marquis
2022-05-03  9:38 ` [PATCH 2/3] xen/arm: Advertise workaround 1 if we apply 3 Bertrand Marquis
2022-05-03 18:17   ` Julien Grall
2022-05-04  7:25     ` Bertrand Marquis
2022-05-05 10:51       ` Julien Grall
2022-05-03  9:38 ` [PATCH 3/3] xen/arm: Add sb instruction support Bertrand Marquis
2022-05-03 18:47   ` Julien Grall
2022-05-04  7:24     ` Bertrand Marquis
2022-05-04  8:06       ` Julien Grall
2022-05-05 15:17         ` Julien Grall
2022-05-09 10:08         ` Bertrand Marquis
2022-05-09 10:31           ` Julien Grall [this message]
2022-05-09 10:49             ` Bertrand Marquis
2022-05-09 11:08               ` Julien Grall
2022-05-09 11:40                 ` Bertrand Marquis

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=0ac6d6c0-4f34-ad1e-c473-2278dd7289c9@xen.org \
    --to=julien@xen.org \
    --cc=Bertrand.Marquis@arm.com \
    --cc=Volodymyr_Babchuk@epam.com \
    --cc=sstabellini@kernel.org \
    --cc=xen-devel@lists.xenproject.org \
    /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.