diff options
author | Gabor Pali <pgj@FreeBSD.org> | 2013-07-10 21:37:57 +0000 |
---|---|---|
committer | Gabor Pali <pgj@FreeBSD.org> | 2013-07-10 21:37:57 +0000 |
commit | 648661ea6393f6e94aebc0a9110d0b12d976c741 (patch) | |
tree | 2be255c9ccab5f1589f0000155ee3cd3589abd25 /en_US.ISO8859-1/htdocs/news/status/report-2013-04-2013-06.xml | |
parent | 2d80d1930a37df511bce625cd4c291dfcceb218a (diff) | |
download | doc-648661ea6393f6e94aebc0a9110d0b12d976c741.tar.gz doc-648661ea6393f6e94aebc0a9110d0b12d976c741.zip |
- Elaborate benefits of PVHM and PVH in the Xen entry
Submitted by: theraven
Notes
Notes:
svn path=/head/; revision=42238
Diffstat (limited to 'en_US.ISO8859-1/htdocs/news/status/report-2013-04-2013-06.xml')
-rw-r--r-- | en_US.ISO8859-1/htdocs/news/status/report-2013-04-2013-06.xml | 16 |
1 files changed, 16 insertions, 0 deletions
diff --git a/en_US.ISO8859-1/htdocs/news/status/report-2013-04-2013-06.xml b/en_US.ISO8859-1/htdocs/news/status/report-2013-04-2013-06.xml index 70715fd4ba..55fe912130 100644 --- a/en_US.ISO8859-1/htdocs/news/status/report-2013-04-2013-06.xml +++ b/en_US.ISO8859-1/htdocs/news/status/report-2013-04-2013-06.xml @@ -1512,6 +1512,22 @@ functionality through <tt>pkg(8)</tt>.</task> will also set the ground for a future PVH port (when PVH support is merged into Xen).</p> + <p>PVHVM allows a virtual machine that boots as a native guest to + be able to take full advantage of paravirtualized drivers, + giving a performance improvement in most I/O related tasks. PVH + allows a guest to take advantage of hardware assistance for + memory management, but uses fully paravirtualized events and + boot procedure, which brings two significant advantages beyond + performance. The first is that domain 0 does not have to run a + QEMU instance for emulated boot for PVH guests, which is a + common reason for hosting providers to charge more for Windows + and other HVM guests. The second is that PVH domains can be + used as domain 0, without requiring different pmap (memory + management) code from the conventional kernel. This will allow + us to ship a single kernel binary supporting bare metal + hardware, running as a Xen unprivileged guest, and eventually as + Xen domain 0.</p> + <p>Further improvements on blkfront and netfront have also been commited:</p> |