All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Jürgen Groß" <jgross@suse.com>
To: George Dunlap <george.dunlap@citrix.com>,
	Steven Haigh <netwiz@crc.id.au>
Cc: Andrew Cooper <andrew.cooper3@citrix.com>,
	Andreas Kinzler <hfp@posteo.de>, Jan Beulich <jbeulich@suse.com>,
	Ian Jackson <ian.jackson@citrix.com>,
	Anthony Perard <anthony.perard@citrix.com>,
	xen-devel@lists.xenproject.org
Subject: Re: [Xen-devel] [PATCH RFC] x86: Add hack to disable "Fake HT" mode
Date: Fri, 15 Nov 2019 16:10:33 +0100	[thread overview]
Message-ID: <98e13c5d-fdc9-3805-3081-9a001d69898a@suse.com> (raw)
In-Reply-To: <c8ee22b9-5823-a362-4ca3-e7bcec105b99@citrix.com>

On 15.11.19 16:05, George Dunlap wrote:
> FTR, please avoid top-posting. :-)
> 
> On 11/15/19 2:31 PM, Steven Haigh wrote:
>> Just regarding the use of a system environment variable to turn this
>> feature / bugfix / hack on and off - this would probably break starting
>> the VM via the xendomains script.
>>
>> If the VM definition is in /etc/xen/auto/, then there would be nothing
>> to set the environment variable before the VM is launched - hence it
>> would not be applied and a guest crash would occur...
>>
>> Depending on the VM's settings, this would either continue to start &
>> crash - or just stop again until it could be started with the ENV variable.
> 
> Right.  So a couple of options:
> 
> 1. Users of xendomains could set the environment variable in their
> xendomains script
> 
> 2. We could add a xl.cfg option.  Unknown xl.cfg entries are ignored
> (for better or for worse); in the future, when the "fake ht" thing is
> replaced, we can either continue ignoring it, or give a useful error
> message saying how it should be changed.
> 
> 2a.  We could have the config option *replace* the environment variable;
> in which case we'd leave libvirt users high and dry
> 
> 2b. We could have the config option cause xl to *set* the environment
> variable, which should continue to allow other toolstacks (even those
> not using libxl) to potentially work around the issue.
> 
> Right now I'm leaning towards 2b, and having it be in a separate patch.

In which case we should consider having a way to set arbitrary
environment variables from the config file in order to avoid this kind
of discussion in future similar cases.


Juergen


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

  reply	other threads:[~2019-11-15 15:10 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-15 10:57 [Xen-devel] [PATCH RFC] x86: Add hack to disable "Fake HT" mode George Dunlap
2019-11-15 11:12 ` Jan Beulich
2019-11-15 11:58   ` George Dunlap
2019-11-15 12:39     ` Jan Beulich
2019-11-15 13:55       ` Andrew Cooper
2019-11-15 14:04         ` George Dunlap
2019-11-15 14:05           ` George Dunlap
2019-11-15 14:06           ` Andrew Cooper
2019-11-15 14:10             ` George Dunlap
2019-11-15 14:14               ` Andrew Cooper
2019-11-15 14:18               ` Jan Beulich
2019-11-15 14:29                 ` George Dunlap
2019-11-15 14:42                   ` Jan Beulich
2019-11-15 14:55                     ` George Dunlap
2019-11-15 14:59                       ` Andrew Cooper
2019-11-15 15:23                         ` George Dunlap
2019-11-15 15:33                           ` Jan Beulich
2019-11-15 15:35                           ` Andrew Cooper
2019-11-15 11:17 ` Andreas Kinzler
2019-11-15 11:29   ` George Dunlap
2019-11-15 11:39     ` Andreas Kinzler
2019-11-15 12:10       ` George Dunlap
2019-11-15 12:44         ` Andreas Kinzler
2019-11-15 14:00           ` Andrew Cooper
2019-11-15 12:23 ` Jürgen Groß
2019-11-15 12:42   ` George Dunlap
2019-11-15 14:31 ` Steven Haigh
2019-11-15 15:05   ` George Dunlap
2019-11-15 15:10     ` Jürgen Groß [this message]
2019-11-15 15:15       ` George Dunlap
2019-11-15 15:27     ` Jan Beulich
2019-11-15 15:30       ` George Dunlap
2019-11-15 15:34         ` Jan Beulich
2019-11-15 15:37           ` George Dunlap

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=98e13c5d-fdc9-3805-3081-9a001d69898a@suse.com \
    --to=jgross@suse.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=anthony.perard@citrix.com \
    --cc=george.dunlap@citrix.com \
    --cc=hfp@posteo.de \
    --cc=ian.jackson@citrix.com \
    --cc=jbeulich@suse.com \
    --cc=netwiz@crc.id.au \
    --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.