From: Greg KH <gregkh@linuxfoundation.org>
To: speck@linutronix.de
Subject: [MODERATED] Re: [patch V4 06/11] x86/speculation/mds: Conditionally clear CPU buffers on idle entry
Date: Mon, 25 Feb 2019 22:09:16 +0100 [thread overview]
Message-ID: <20190225210915.GE12242@kroah.com> (raw)
In-Reply-To: <20190222224149.616531116@linutronix.de>
On Fri, Feb 22, 2019 at 11:24:24PM +0100, speck for Thomas Gleixner wrote:
> From: Thomas Gleixner <tglx@linutronix.de>
>
> Add a static key which controls the invocation of the CPU buffer clear
> mechanism on idle entry. This is independent of other MDS mitigations
> because the idle entry invocation to mitigate the potential leakage due to
> store buffer repartitioning is only necessary on SMT systems.
>
> Add the actual invocations to the different halt/mwait variants which
> covers all usage sites. mwaitx is not patched as it's not available on
> Intel CPUs.
>
> The buffer clear is only invoked before entering the C-State to prevent
> that stale data from the idling CPU is spilled to the Hyper-Thread sibling
> after the Store buffer got repartitioned and all entries are available to
> the non idle sibling.
>
> When coming out of idle the store buffer is partitioned again so each
> sibling has half of it available. Now CPU which returned from idle could be
> speculatively exposed to contents of the sibling, but the buffers are
> flushed either on exit to user space or on VMENTER.
>
> When later on conditional buffer clearing is implemented on top of this,
> then there is no action required either because before returning to user
> space the context switch will set the condition flag which causes a flush
> on the return to user path.
>
> This intentionaly does not handle the case in the acpi/processor_idle
> driver which uses the legacy IO port interface for C-State transitions for
> two reasons:
>
> - The acpi/processor_idle driver was replaced by the intel_idle driver
> almost a decade ago. Anything Nehalem upwards supports it and defaults
> to that new driver.
>
> - The legacy IO port interface is likely to be used on older and therefore
> unaffected CPUs or on systems which do not receive microcode updates
> anymore, so there is no point in adding that.
>
> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
> Reviewed-by: Borislav Petkov <bp@suse.de>
Comparing this to the Intel paper, I find this way more readable and
understandable. Things they "hint" at are actually spelled out here,
nice work.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
next prev parent reply other threads:[~2019-02-25 21:09 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-22 22:24 [patch V4 00/11] MDS basics Thomas Gleixner
2019-02-22 22:24 ` [patch V4 01/11] x86/msr-index: Cleanup bit defines Thomas Gleixner
2019-02-22 22:24 ` [patch V4 02/11] x86/speculation/mds: Add basic bug infrastructure for MDS Thomas Gleixner
2019-02-23 1:28 ` [MODERATED] " Linus Torvalds
2019-02-23 7:42 ` Thomas Gleixner
2019-02-27 13:04 ` Thomas Gleixner
2019-02-22 22:24 ` [patch V4 03/11] x86/kvm: Expose X86_FEATURE_MD_CLEAR to guests Thomas Gleixner
2019-02-22 22:24 ` [patch V4 04/11] x86/speculation/mds: Add mds_clear_cpu_buffer() Thomas Gleixner
2019-02-25 16:06 ` [MODERATED] " Frederic Weisbecker
2019-02-26 14:19 ` Josh Poimboeuf
2019-03-01 20:58 ` [MODERATED] Encrypted Message Jon Masters
2019-03-01 22:14 ` Jon Masters
2019-02-26 15:00 ` [MODERATED] Re: [patch V4 04/11] x86/speculation/mds: Add mds_clear_cpu_buffer() David Woodhouse
2019-02-22 22:24 ` [patch V4 05/11] x86/speculation/mds: Clear CPU buffers on exit to user Thomas Gleixner
2019-02-25 21:04 ` [MODERATED] " Greg KH
2019-02-26 15:20 ` Josh Poimboeuf
2019-02-26 20:26 ` Thomas Gleixner
2019-02-22 22:24 ` [patch V4 06/11] x86/speculation/mds: Conditionally clear CPU buffers on idle entry Thomas Gleixner
2019-02-25 21:09 ` Greg KH [this message]
2019-02-26 15:31 ` [MODERATED] " Josh Poimboeuf
2019-02-26 20:20 ` Thomas Gleixner
2019-02-22 22:24 ` [patch V4 07/11] x86/speculation/mds: Add mitigation control for MDS Thomas Gleixner
2019-02-25 20:17 ` [MODERATED] " mark gross
2019-02-26 15:50 ` Josh Poimboeuf
2019-02-26 20:16 ` Thomas Gleixner
2019-02-22 22:24 ` [patch V4 08/11] x86/speculation/mds: Add sysfs reporting " Thomas Gleixner
2019-02-22 22:24 ` [patch V4 09/11] x86/speculation/mds: Add mitigation mode VMWERV Thomas Gleixner
2019-02-23 9:52 ` [MODERATED] " Greg KH
2019-02-25 20:31 ` mark gross
2019-02-26 0:34 ` Andrew Cooper
2019-02-26 18:51 ` mark gross
2019-02-26 19:29 ` Thomas Gleixner
2019-02-22 22:24 ` [patch V4 10/11] Documentation: Move L1TF to separate directory Thomas Gleixner
2019-02-23 8:41 ` [MODERATED] " Greg KH
2019-02-22 22:24 ` [patch V4 11/11] Documentation: Add MDS vulnerability documentation Thomas Gleixner
2019-02-23 9:58 ` [MODERATED] " Greg KH
2019-02-26 20:11 ` Thomas Gleixner
2019-02-25 18:02 ` [MODERATED] " Dave Hansen
2019-02-26 20:10 ` Thomas Gleixner
2019-02-23 0:53 ` [MODERATED] Re: [patch V4 00/11] MDS basics Andrew Cooper
2019-02-23 14:12 ` Peter Zijlstra
2019-02-25 16:38 ` mark gross
2019-02-26 19:58 ` Thomas Gleixner
2019-02-26 16:28 ` [MODERATED] " Tyler Hicks
2019-02-26 19:58 ` Thomas Gleixner
2019-02-26 18:58 ` [MODERATED] " Kanth Ghatraju
2019-02-26 19:59 ` Thomas Gleixner
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=20190225210915.GE12242@kroah.com \
--to=gregkh@linuxfoundation.org \
--cc=speck@linutronix.de \
/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.