This bug was fixed in the package qemu - 1:2.11+dfsg-1ubuntu7.4
---
qemu (1:2.11+dfsg-1ubuntu7.4) bionic; urgency=medium
* d/p/ubuntu/machine-type-hpb.patch: add -hpb machine type
for host-phys-bits=true (LP: #1776189)
- add an info about this change in
In addition we had the same pushed through regression tests (on ppa with
same content).
** Tags removed: verification-needed verification-needed-bionic
** Tags added: verification-done verification-done-bionic
--
You received this bug notification because you are a member of Ubuntu
Bugs, which
Upgraded:
qemu-block-extra/bionic-proposed 1:2.11+dfsg-1ubuntu7.4 amd64 [upgradable from:
1:2.11+dfsg-1ubuntu7.3]
qemu-kvm/bionic-proposed 1:2.11+dfsg-1ubuntu7.4 amd64 [upgradable from:
1:2.11+dfsg-1ubuntu7.3]
qemu-system-common/bionic-proposed 1:2.11+dfsg-1ubuntu7.4 amd64 [upgradable
from:
SRU review for Bionic.
This looks reasonable to me. I'm not so sure about the hardware
enablement exception, but as a new feature it seems like something we'd
want for an LTS anyway, and complies with all the SRU requirements under
that exception. It's an entirely new machine type which isn't
- Final test build and test from PPA for target release
- Added SRU Template
- Uploaded to review/acceptance by the SRU Team
- Pushed to package repositories as we consider it proposed
Waiting on SRU Team now.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which
** Description changed:
+ [Impact]
+
+ * Qemu supports running guests >1TB but currently the virtualization
+stack above has no good way to control that (e.g. libvirt/openstack).
+
+ * Long term we'd want to see bug 1769053 (this is where all started)
+implemented in libvirt and
** Also affects: qemu (Ubuntu Bionic)
Importance: Undecided
Status: New
** Also affects: seabios (Ubuntu Bionic)
Importance: Undecided
Status: New
** No longer affects: seabios (Ubuntu Bionic)
--
You received this bug notification because you are a member of Ubuntu
Bugs,
This bug was fixed in the package qemu - 1:2.11+dfsg-1ubuntu11
---
qemu (1:2.11+dfsg-1ubuntu11) cosmic; urgency=medium
* d/p/ubuntu/machine-type-hpb.patch: add -hpb machine type
for host-phys-bits=true (LP: #1776189)
- add an info about this change in
** Merge proposal linked:
https://code.launchpad.net/~paelzer/ubuntu/+source/qemu/+git/qemu/+merge/347801
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1776189
Title:
implement a machine type
** Merge proposal linked:
https://code.launchpad.net/~paelzer/ubuntu/+source/qemu/+git/qemu/+merge/347796
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1776189
Title:
implement a machine type
We did a short brainstorm/bikeshed sesssion on the name and decided to:
- use "hpb" for host-phys-bits abbreviation as the former -large could be a lot
of other things
- acked on not start type proliferation in general (only in very special cases)
- if possible use a "+" instead of "-" in the
Yep, now all things fit together.
- My old (swap based) tests that worked without a change
- working on real >1TB machines without a change.
That said the change to seabios is really not needed.
Setting that task to invalid.
** Changed in: seabios (Ubuntu)
Status: Triaged => Invalid
--
RE: bikeshedding on the name; I don't think -large or -big maps well
the the feature, which is more like -cpu host; ie, give the host cpu
attributes to the guest vcpu.
In this case, we're asking to have host physical addressing bits
applied to the guest vcpus.
So maybe -hpb for HostPhysicalBits?
As assumed the seabios change really IS implemented differently.
That means I have the todo to restest without those.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1776189
Title:
implement a
FYI on the bug as I mentione dthis being controllable in higher stacks.
Libvirt is via type="" attribute.
Openstack can control this globally or per image
- global:
https://docs.openstack.org/mitaka/config-reference/compute/config-options.html
- per image:
So far all tests work, but they are still ongoing.
While that is testing I was submitting a request for discussion upstream why
the change is not integrated there.
See: https://mail.coreboot.org/pipermail/seabios/2018-June/012310.html
I hope that the feedback will help to evaluate if we want to
I'll run less case-bound tests in the form of a full pre-upload regression test
on these changes.
Then if safe IMHO can start pushing this to Cosmic and from there consider
Bionic SRU.
Note: due to the fact that Cosmic has no "own" qemu version yet the
change there will also just modify/extend
I already did some tests with these changes from ppa [1].
This also includes changes to seabios which seem safe for users with less
physical bits and required for those huge guests to work properly.
[1]: https://launchpad.net/~canonical-server/+archive/ubuntu/large-virt
** Also affects: seabios
18 matches
Mail list logo
