Signed-off-by: Jim Fehlig <jfehlig@suse.com>
---
docs/news.xml | 22 ++++++++++++++++++++++
1 file changed, 22 insertions(+)
diff --git a/docs/news.xml b/docs/news.xml
index 99c65447b0..074dc22e97 100644
--- a/docs/news.xml
+++ b/docs/news.xml
@@ -66,6 +66,28 @@
feedback from users on how well it works in practice.
</description>
</change>
+ <change>
+ <summary>
+ Xen: add support for e820_host hypervisor feature
+ </summary>
+ <description>
+ e820_host is a Xen-specific option only available for PV guests.
+ When enabled it provides the guest with a virtual e820 memory map
+ based on the host one. It must be enabled to allow hotplugging
+ PCI devices to PV guests, particularly when memory ballooning
+ is enabled.
+ </description>
+ </change>
+ <change>
+ <summary>
+ Xen: add support for passthrough hypervisor feature
+ </summary>
+ <description>
+ passthrough is a Xen-specific option new to Xen 4.13 that enables
+ PCI passthrough for guests. It must be enabled to allow hotplugging
+ PCI devices.
+ </description>
+ </change>
</section>
<section title="Improvements">
</section>
--
2.26.0
On Tue, 2020-04-21 at 16:06 -0600, Jim Fehlig wrote: > + <summary> > + Xen: add support for e820_host hypervisor feature s/Xen: add/xen: Add/ I suggest also putting either single or double quotes around the name of the feature for clarity. > + </summary> > + <description> > + e820_host is a Xen-specific option only available for PV guests. And here you could wrap it in a <code> element so that it will display nicely on the Web. These comments apply also to the second item. Thanks for taking care of the release notes! Reviewed-by: Andrea Bolognani <abologna@redhat.com> -- Andrea Bolognani / Red Hat / Virtualization
On 4/22/20 1:51 AM, Andrea Bolognani wrote: > On Tue, 2020-04-21 at 16:06 -0600, Jim Fehlig wrote: >> + <summary> >> + Xen: add support for e820_host hypervisor feature > > s/Xen: add/xen: Add/ There is only one 'xen:' in news, all other instances are 'Xen:'. I should stick with the majority. > I suggest also putting either single or double quotes around the name > of the feature for clarity. > >> + </summary> >> + <description> >> + e820_host is a Xen-specific option only available for PV guests. > > And here you could wrap it in a <code> element so that it will > display nicely on the Web. > > These comments apply also to the second item. Thanks! I'll address the comments before pushing. Regards, Jim
On Wed, 2020-04-22 at 08:48 -0600, Jim Fehlig wrote: > On 4/22/20 1:51 AM, Andrea Bolognani wrote: > > On Tue, 2020-04-21 at 16:06 -0600, Jim Fehlig wrote: > > > + <summary> > > > + Xen: add support for e820_host hypervisor feature > > > > s/Xen: add/xen: Add/ > > There is only one 'xen:' in news, all other instances are 'Xen:'. I should stick > with the majority. If you look at most (all?) other entries, they look like subsystem: Change which is the same style we use for commit messages. Xen seems to be the outlier when it comes to following this style, both in release notes and to some extent in commit messages. It would be great if it could fall in line with the rest of the code base. -- Andrea Bolognani / Red Hat / Virtualization
On 4/22/20 9:10 AM, Andrea Bolognani wrote: > On Wed, 2020-04-22 at 08:48 -0600, Jim Fehlig wrote: >> On 4/22/20 1:51 AM, Andrea Bolognani wrote: >>> On Tue, 2020-04-21 at 16:06 -0600, Jim Fehlig wrote: >>>> + <summary> >>>> + Xen: add support for e820_host hypervisor feature >>> >>> s/Xen: add/xen: Add/ >> >> There is only one 'xen:' in news, all other instances are 'Xen:'. I should stick >> with the majority. > > If you look at most (all?) other entries, they look like > > subsystem: Change > > which is the same style we use for commit messages. If you really want to split hairs, the subsytem is 'libxl'. But I don't think that is good for news since in general folks know of the 'Xen' project, but not 'libxl'. However, now that the old xen driver is removed libxl could be renamed to xen :-). > Xen seems to be the outlier when it comes to following this style, > both in release notes and to some extent in commit messages. It would > be great if it could fall in line with the rest of the code base. Sorry for being out of line. I'll try to be more conscious of that in the future. Regards, Jim
On Wed, 2020-04-22 at 10:43 -0600, Jim Fehlig wrote:
> On 4/22/20 9:10 AM, Andrea Bolognani wrote:
> > If you look at most (all?) other entries, they look like
> >
> > subsystem: Change
> >
> > which is the same style we use for commit messages.
>
> If you really want to split hairs, the subsytem is 'libxl'. But I don't think
> that is good for news since in general folks know of the 'Xen' project, but not
> 'libxl'. However, now that the old xen driver is removed libxl could be renamed
> to xen :-).
Yeah, I know of the history :) And I agree that we should probably
rename the driver to xen[1]; as you probably remember, the standalone
daemon is called virtxend and not virtlibxld already :)
> > Xen seems to be the outlier when it comes to following this style,
> > both in release notes and to some extent in commit messages. It would
> > be great if it could fall in line with the rest of the code base.
>
> Sorry for being out of line. I'll try to be more conscious of that in the future.
Nothing to apologize for! I just pointed it out so that we can, in
time, get closer to a state of consistency :)
[1] Although it will never be a full rename, because it still need to
handle libxl:// connection URLs.
--
Andrea Bolognani / Red Hat / Virtualization
On 4/22/20 11:06 AM, Andrea Bolognani wrote: > On Wed, 2020-04-22 at 10:43 -0600, Jim Fehlig wrote: >> On 4/22/20 9:10 AM, Andrea Bolognani wrote: >>> If you look at most (all?) other entries, they look like >>> >>> subsystem: Change >>> >>> which is the same style we use for commit messages. >> >> If you really want to split hairs, the subsytem is 'libxl'. But I don't think >> that is good for news since in general folks know of the 'Xen' project, but not >> 'libxl'. However, now that the old xen driver is removed libxl could be renamed >> to xen :-). > > Yeah, I know of the history :) And I agree that we should probably > rename the driver to xen[1]; as you probably remember, the standalone > daemon is called virtxend and not virtlibxld already :) > >>> Xen seems to be the outlier when it comes to following this style, >>> both in release notes and to some extent in commit messages. It would >>> be great if it could fall in line with the rest of the code base. >> >> Sorry for being out of line. I'll try to be more conscious of that in the future. > > Nothing to apologize for! I just pointed it out so that we can, in > time, get closer to a state of consistency :) > > > [1] Although it will never be a full rename, because it still need to > handle libxl:// connection URLs. libxl:// was never supported. DV rightly pointed out that introducing libxl:// as another way to connect to Xen violated one of libvirt's core principles of "minimize the change on the application stack as the lower layers of virtualization evolves". It took me a while to find that bit of history :-) https://www.redhat.com/archives/libvir-list/2011-March/msg00449.html You'll also notice libxl:// is not mentioned on the Connect URI page https://libvirt.org/uri.html Regards, Jim
On Wed, 2020-04-22 at 11:55 -0600, Jim Fehlig wrote: > On 4/22/20 11:06 AM, Andrea Bolognani wrote: > > [1] Although it will never be a full rename, because it still need to > > handle libxl:// connection URLs. > > libxl:// was never supported. DV rightly pointed out that introducing libxl:// > as another way to connect to Xen violated one of libvirt's core principles of > "minimize the change on the application stack as the lower layers of > virtualization evolves". It took me a while to find that bit of history :-) > > https://www.redhat.com/archives/libvir-list/2011-March/msg00449.html > > You'll also notice libxl:// is not mentioned on the Connect URI page > > https://libvirt.org/uri.html I stand corrected! I clearly don't know much about Xen O:-) -- Andrea Bolognani / Red Hat / Virtualization
© 2016 - 2024 Red Hat, Inc.