From: Andrew Cooper <andrew.cooper3@citrix.com>
To: <xen-devel@lists.xenproject.org>, <pdurrant@amazon.com>
Cc: roger.pau@citrix.com, luwei.kang@intel.com,
marmarek@invisiblethingslab.com, dwmw@amazon.com
Subject: Re: [Xen-devel] [ANNOUNCE] Xen 4.14 Development Update
Date: Wed, 29 Jan 2020 18:03:37 +0000 [thread overview]
Message-ID: <d6347fd0-19f7-f2ba-517d-4c7c794239c5@citrix.com> (raw)
In-Reply-To: <20200129123618.1202-1-pdurrant@amazon.com>
On 29/01/2020 12:36, Paul Durrant wrote:
> = Projects =
>
> == Hypervisor ==
>
> === x86 ===
>
> * Intel Processor Trace virtualization enabling (v1)
> - Luwei Kang
Hasn't seen any activity in several releases. Probably safe to drop.
Also at least somewhat entangled with CPUID/MSR support.
> * Fixes to #DB injection
> - Andrew Cooper
Parts of this manifested unexpected as XSA-308. Rest is still work in
progress, with a TODO of how not to break introspection while fixing
it. Unlikely to see any movement in the short term.
> * CPUID/MSR Xen/toolstack improvements
> - Andrew Cooper
Very much in progress.
https://lore.kernel.org/xen-devel/20200127143444.25538-1-andrew.cooper3@citrix.com/
is v2 of the "move data into the migration stream" aspect.
I'm working on "rework boot time CPUID handling" right now, which will
ultimately allow MSRs to be included in what gets configured.
All of this forms a base for "Virtualise MSR_ARCH_CAPS for guests" which
is (now) the logical goal of the work, and might be a better tracking name.
In reality, for now I mean only the various *_NO bits because ...
> * eIBRS
> - Andrew Cooper
... it is unfortunately quite tangled on top. Because of the overloaded
nature of IBRS, even having Xen try to use eIBRS while only supporting
the legacy IBRS interface for guests turns into a massive headache.
I'm fairly confident that eIBRS support shouldn't be attempted until
general MSR_ARCH_CAPS is in place.
> * Improvements to domain_crash()
> - Andrew Cooper
This is a fairly small piece of work, but I don't have time to progress
it. I've got a rebased branch if anyone else feels like spending the
day or so it will take to get up to quality.
Another task is AMD hardware mitigations for everything
SpeculativeStoreBypass (i.e. pre-L1TF) and later. There are crippling
performance problems caused by the lack of MSR_VIRT_SPEC_CTRL. Since
that series I posted for that, Rome has gained several other hardware
features which need exposing to guests suitably.
~Andrew
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
next prev parent reply other threads:[~2020-01-29 18:04 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-01-29 12:36 [Xen-devel] [ANNOUNCE] Xen 4.14 Development Update Paul Durrant
2020-01-29 12:43 ` Marek Marczykowski-Górecki
2020-01-29 12:57 ` Durrant, Paul
2020-01-29 12:47 ` Jan Beulich
2020-01-29 13:00 ` Durrant, Paul
2020-01-29 12:59 ` Jürgen Groß
2020-01-29 13:04 ` David Woodhouse
2020-01-29 14:12 ` Tamas K Lengyel
2020-01-29 18:03 ` Andrew Cooper [this message]
2020-02-26 10:03 Paul Durrant
2020-03-25 10:19 paul
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=d6347fd0-19f7-f2ba-517d-4c7c794239c5@citrix.com \
--to=andrew.cooper3@citrix.com \
--cc=dwmw@amazon.com \
--cc=luwei.kang@intel.com \
--cc=marmarek@invisiblethingslab.com \
--cc=pdurrant@amazon.com \
--cc=roger.pau@citrix.com \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).