From: Dario Faggioli <dario.faggioli@citrix.com>
To: xen-devel@lists.xenproject.org
Cc: Euan Harris <euan.harris@citrix.com>,
	Ian Campbell <ian.campbell@citrix.com>,
	Wei Liu <wei.liu2@citrix.com>,
	Ian Jackson <ian.jackson@eu.citrix.com>,
	Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Subject: [PATCH 0/2] libxl: fix NUMA placement preventing domain config to be re-used
Date: Wed, 01 Jul 2015 16:02:58 +0200	[thread overview]
Message-ID: <20150701135059.15951.39938.stgit@Solace.station> (raw)
Hi,
This series fixes the issue reported here by Ian, about NUMA placement related
domain config fields preventing the config to be reused:
http://lists.xenproject.org/archives/html/xen-devel/2015-06/msg04454.html
The way it does it is by setting the b_info->numa_placement config switch to
false, after NUMA placement has actually happened. This way, the next time we
re-use the config, we won't run the placement again, we just will use the
result obtained from the original execution of it. This is, both IMO and for
Ian Jackson (as per an IRC conversation we had yesterday), the best behaviour,
as it yields consistent results across config re-uses.
BTW, with patch 2 applied, I've been able to run the test in here, and it
worked for me:
http://xenbits.xen.org/gitweb/?p=people/iwj/ring3-xl-test.git;a=shortlog;h=refs/heads/t.numa-bug
While I was there, though, I took the chance to put patch 1 together as well.
What it does it turning a couple of error conditions (still related to the NUMA
placement domain config switch) into warnings, as error was too much. This has
been discussed a while back, and I distinctly remember it mentioned in a
conversation on xen-devel, but I can't find it now. In any case, I think it's
the right thing to do.
Regards,
Dario
---
Dario Faggioli (2):
      libxl: turn NUMA placement misconfigs into warnings
      libxl: unset info->numa_placement upon successful placement
 tools/libxl/libxl_dom.c |   85 +++++++++++++++++++++++++----------------------
 1 file changed, 46 insertions(+), 39 deletions(-)
--
<<This happens because I choose it to happen!>> (Raistlin Majere)
-----------------------------------------------------------------
Dario Faggioli, Ph.D, http://about.me/dario.faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
next             reply	other threads:[~2015-07-01 14:03 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-01 14:02 Dario Faggioli [this message]
2015-07-01 14:03 ` [PATCH 1/2] libxl: turn NUMA placement misconfigs into warnings Dario Faggioli
2015-07-01 14:38   ` Ian Jackson
2015-07-03 10:35     ` Ian Campbell
2015-07-03 11:09       ` Ian Jackson
2015-07-03 14:52         ` Dario Faggioli
2015-07-01 14:03 ` [PATCH 2/2] libxl: unset info->numa_placement upon successful placement Dario Faggioli
2015-07-01 14:39   ` Ian Jackson
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=20150701135059.15951.39938.stgit@Solace.station \
    --to=dario.faggioli@citrix.com \
    --cc=euan.harris@citrix.com \
    --cc=ian.campbell@citrix.com \
    --cc=ian.jackson@eu.citrix.com \
    --cc=stefano.stabellini@eu.citrix.com \
    --cc=wei.liu2@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).