)]}'
{"doc/source/admin/configuration/hypervisor-kvm.rst":[{"author":{"_account_id":6962,"name":"Kashyap Chamarthy","email":"kchamart@redhat.com","username":"kashyapc"},"change_message_id":"c90297ce081163647ded038143edd4f0c2e5a08e","unresolved":false,"context_lines":[{"line_number":360,"context_line":".. code-block:: none"},{"line_number":361,"context_line":""},{"line_number":362,"context_line":"   options kvm_intel nested\u003d1"},{"line_number":363,"context_line":"   options kvm_amd nested\u003d1"},{"line_number":364,"context_line":""},{"line_number":365,"context_line":"A reboot may be required for the change to become effective."},{"line_number":366,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_4a250309","line":363,"updated":"2018-10-16 08:46:35.000000000","message":"Might want to separate it.  Something along the lines of:\n\nFor Intel CPUs:\n\n    options kvm_intel nested\u003d1\n\nFor AMD:\n\n    options kvm_amd nested\u003d1","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":2463,"name":"Florian Haas","email":"florian.haas@cleura.com","username":"fghaas"},"change_message_id":"93b776ae241e40e3eca93aaa865db1486249e5ac","unresolved":false,"context_lines":[{"line_number":360,"context_line":".. code-block:: none"},{"line_number":361,"context_line":""},{"line_number":362,"context_line":"   options kvm_intel nested\u003d1"},{"line_number":363,"context_line":"   options kvm_amd nested\u003d1"},{"line_number":364,"context_line":""},{"line_number":365,"context_line":"A reboot may be required for the change to become effective."},{"line_number":366,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_656cb09c","line":363,"in_reply_to":"3f79a3b5_4a250309","updated":"2018-10-16 09:01:43.000000000","message":"From an operator perspective, most people will be automating their deployment of this file. Since any compute node will always have either Intel or AMD CPUs, it will only ever load one of the modules, thus only one of the options will ever apply. Thus, rolling out just one file with both options isn\u0027t disadvantageous from a functionality perspective, but clearly facilitates operations.","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":6962,"name":"Kashyap Chamarthy","email":"kchamart@redhat.com","username":"kashyapc"},"change_message_id":"bdddc96d4a0156b43efc8bec9fea110a21c9bfad","unresolved":false,"context_lines":[{"line_number":360,"context_line":".. code-block:: none"},{"line_number":361,"context_line":""},{"line_number":362,"context_line":"   options kvm_intel nested\u003d1"},{"line_number":363,"context_line":"   options kvm_amd nested\u003d1"},{"line_number":364,"context_line":""},{"line_number":365,"context_line":"A reboot may be required for the change to become effective."},{"line_number":366,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_a595e832","line":363,"in_reply_to":"3f79a3b5_656cb09c","updated":"2018-10-16 09:15:10.000000000","message":"Good point, let\u0027s go with your existing `code-block`.","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":6962,"name":"Kashyap Chamarthy","email":"kchamart@redhat.com","username":"kashyapc"},"change_message_id":"c90297ce081163647ded038143edd4f0c2e5a08e","unresolved":false,"context_lines":[{"line_number":362,"context_line":"   options kvm_intel nested\u003d1"},{"line_number":363,"context_line":"   options kvm_amd nested\u003d1"},{"line_number":364,"context_line":""},{"line_number":365,"context_line":"A reboot may be required for the change to become effective."},{"line_number":366,"context_line":""},{"line_number":367,"context_line":"Nested guest support in ``nova.conf``"},{"line_number":368,"context_line":"-------------------------------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_caed938e","line":365,"updated":"2018-10-16 08:46:35.000000000","message":"Actually a reboot isn\u0027t required if you simply load the module:\n\n    $ sudo rmmod kvm-amd\n    $ sudo sh -c \"echo \u0027options kvm-amd nested\u003d1\u0027 \u003e\u003e /etc/modprobe.d/dist.conf\"\n    $ sudo modprobe kvm-amd\n\nHow about using something like the above, thus avoiding reboot.\n\nRelated: https://docs.openstack.org/devstack/latest/guides/devstack-with-nested-kvm.html","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":6962,"name":"Kashyap Chamarthy","email":"kchamart@redhat.com","username":"kashyapc"},"change_message_id":"bdddc96d4a0156b43efc8bec9fea110a21c9bfad","unresolved":false,"context_lines":[{"line_number":362,"context_line":"   options kvm_intel nested\u003d1"},{"line_number":363,"context_line":"   options kvm_amd nested\u003d1"},{"line_number":364,"context_line":""},{"line_number":365,"context_line":"A reboot may be required for the change to become effective."},{"line_number":366,"context_line":""},{"line_number":367,"context_line":"Nested guest support in ``nova.conf``"},{"line_number":368,"context_line":"-------------------------------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_85ab0c6f","line":365,"in_reply_to":"3f79a3b5_0515bce5","updated":"2018-10-16 09:15:10.000000000","message":"Hmm, that\u0027s true—about `rmmod` on a running guest.\n\nOkay, here too, let\u0027s stick to what you already wrote.","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":2463,"name":"Florian Haas","email":"florian.haas@cleura.com","username":"fghaas"},"change_message_id":"93b776ae241e40e3eca93aaa865db1486249e5ac","unresolved":false,"context_lines":[{"line_number":362,"context_line":"   options kvm_intel nested\u003d1"},{"line_number":363,"context_line":"   options kvm_amd nested\u003d1"},{"line_number":364,"context_line":""},{"line_number":365,"context_line":"A reboot may be required for the change to become effective."},{"line_number":366,"context_line":""},{"line_number":367,"context_line":"Nested guest support in ``nova.conf``"},{"line_number":368,"context_line":"-------------------------------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_0515bce5","line":365,"in_reply_to":"3f79a3b5_caed938e","updated":"2018-10-16 09:01:43.000000000","message":"Right, of course it\u0027s not. That\u0027s why it says \"may be required\". Meaning: if you reboot the box after making this change, it\u0027ll definitely work.\n\nIf we were to go with your suggestion, we\u0027d also need to go into detail as to why rmmod won\u0027t work if a VM is already running, and/or why you can\u0027t set this via /sys/module/kvm-*/parameters.","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":6962,"name":"Kashyap Chamarthy","email":"kchamart@redhat.com","username":"kashyapc"},"change_message_id":"c90297ce081163647ded038143edd4f0c2e5a08e","unresolved":false,"context_lines":[{"line_number":372,"context_line":""},{"line_number":373,"context_line":"Host pass through"},{"line_number":374,"context_line":"  In this mode, nested virtualization is automatically enabled once"},{"line_number":375,"context_line":"  the KVM kernel module is loaded with nesting support."},{"line_number":376,"context_line":""},{"line_number":377,"context_line":"  .. code-block:: ini"},{"line_number":378,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_ea63af17","line":375,"updated":"2018-10-16 08:46:35.000000000","message":"Might want to note the caveat about live migration --  libvirt and QEMU doesn\u0027t guarantee that a stable vCPU will be exposed to the guest, and your hardware _must_ be homogeneous (*and* you have the relevant software + microcode updates) to be able to live migrate.","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":6962,"name":"Kashyap Chamarthy","email":"kchamart@redhat.com","username":"kashyapc"},"change_message_id":"bdddc96d4a0156b43efc8bec9fea110a21c9bfad","unresolved":false,"context_lines":[{"line_number":372,"context_line":""},{"line_number":373,"context_line":"Host pass through"},{"line_number":374,"context_line":"  In this mode, nested virtualization is automatically enabled once"},{"line_number":375,"context_line":"  the KVM kernel module is loaded with nesting support."},{"line_number":376,"context_line":""},{"line_number":377,"context_line":"  .. code-block:: ini"},{"line_number":378,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_453c142f","line":375,"in_reply_to":"3f79a3b5_25b4781c","updated":"2018-10-16 09:15:10.000000000","message":"Hmm, I see your point.  I just want people to not go: \"ah, automatically enables; probably the \u0027simplest\u0027 approach\", while realizing only too late that they can\u0027t migrate their guest with \u0027host-passthrough\u0027, from Haswell host CPU to Skylake.\n\nMaybe just give a hint with a small sentence: \"bear in mind the live migration caveats of host pass through\".  \n\nFeel free to rephrase it suitably :-)","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":2463,"name":"Florian Haas","email":"florian.haas@cleura.com","username":"fghaas"},"change_message_id":"4d1544653aa4b5ded2ee81d6ee6b82842c1a1326","unresolved":false,"context_lines":[{"line_number":372,"context_line":""},{"line_number":373,"context_line":"Host pass through"},{"line_number":374,"context_line":"  In this mode, nested virtualization is automatically enabled once"},{"line_number":375,"context_line":"  the KVM kernel module is loaded with nesting support."},{"line_number":376,"context_line":""},{"line_number":377,"context_line":"  .. code-block:: ini"},{"line_number":378,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_a5e648ae","line":375,"in_reply_to":"3f79a3b5_453c142f","updated":"2018-10-30 15:13:56.000000000","message":"Yup, let me think of something and submit an updated patch.","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":2463,"name":"Florian Haas","email":"florian.haas@cleura.com","username":"fghaas"},"change_message_id":"93b776ae241e40e3eca93aaa865db1486249e5ac","unresolved":false,"context_lines":[{"line_number":372,"context_line":""},{"line_number":373,"context_line":"Host pass through"},{"line_number":374,"context_line":"  In this mode, nested virtualization is automatically enabled once"},{"line_number":375,"context_line":"  the KVM kernel module is loaded with nesting support."},{"line_number":376,"context_line":""},{"line_number":377,"context_line":"  .. code-block:: ini"},{"line_number":378,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_25b4781c","line":375,"in_reply_to":"3f79a3b5_ea63af17","updated":"2018-10-16 09:01:43.000000000","message":"Right, I thought about that too. Unfortunately the admin guide contains information about live migration in at least 4 different places, and I didn\u0027t want to add yet another one. What do you think?","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":6962,"name":"Kashyap Chamarthy","email":"kchamart@redhat.com","username":"kashyapc"},"change_message_id":"c90297ce081163647ded038143edd4f0c2e5a08e","unresolved":false,"context_lines":[{"line_number":424,"context_line":"Guest agent support"},{"line_number":425,"context_line":"~~~~~~~~~~~~~~~~~~~"},{"line_number":426,"context_line":""},{"line_number":427,"context_line":"Use guest agents to enable optional access between compute nodes and guests"},{"line_number":428,"context_line":"through a socket, using the QMP protocol."},{"line_number":429,"context_line":""},{"line_number":430,"context_line":"To enable this feature, you must set ``hw_qemu_guest_agent\u003dyes`` as a metadata"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_2a7527c9","line":427,"range":{"start_line":427,"start_character":4,"end_line":427,"end_character":16},"updated":"2018-10-16 08:46:35.000000000","message":"s/\"guest agents\"/\"QEMU guest agent\"/","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":2463,"name":"Florian Haas","email":"florian.haas@cleura.com","username":"fghaas"},"change_message_id":"93b776ae241e40e3eca93aaa865db1486249e5ac","unresolved":false,"context_lines":[{"line_number":424,"context_line":"Guest agent support"},{"line_number":425,"context_line":"~~~~~~~~~~~~~~~~~~~"},{"line_number":426,"context_line":""},{"line_number":427,"context_line":"Use guest agents to enable optional access between compute nodes and guests"},{"line_number":428,"context_line":"through a socket, using the QMP protocol."},{"line_number":429,"context_line":""},{"line_number":430,"context_line":"To enable this feature, you must set ``hw_qemu_guest_agent\u003dyes`` as a metadata"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_65a7d00a","line":427,"range":{"start_line":427,"start_character":4,"end_line":427,"end_character":16},"in_reply_to":"3f79a3b5_2a7527c9","updated":"2018-10-16 09:01:43.000000000","message":"Um, I didn\u0027t write this, nor did I change it (I just promoted the heading above).","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":6962,"name":"Kashyap Chamarthy","email":"kchamart@redhat.com","username":"kashyapc"},"change_message_id":"c90297ce081163647ded038143edd4f0c2e5a08e","unresolved":false,"context_lines":[{"line_number":425,"context_line":"~~~~~~~~~~~~~~~~~~~"},{"line_number":426,"context_line":""},{"line_number":427,"context_line":"Use guest agents to enable optional access between compute nodes and guests"},{"line_number":428,"context_line":"through a socket, using the QMP protocol."},{"line_number":429,"context_line":""},{"line_number":430,"context_line":"To enable this feature, you must set ``hw_qemu_guest_agent\u003dyes`` as a metadata"},{"line_number":431,"context_line":"parameter on the image you wish to use to create the guest-agent-capable"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_aa429765","line":428,"range":{"start_line":428,"start_character":18,"end_line":428,"end_character":40},"updated":"2018-10-16 08:46:35.000000000","message":"I would omit this.  QMP is an internal implementation detail from an OpenStack operator\u0027s point of view; and you\u0027d run into it largely when debugging issues between Nova --\u003e libvirt --\u003e QEMU :-)","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":6962,"name":"Kashyap Chamarthy","email":"kchamart@redhat.com","username":"kashyapc"},"change_message_id":"bdddc96d4a0156b43efc8bec9fea110a21c9bfad","unresolved":false,"context_lines":[{"line_number":425,"context_line":"~~~~~~~~~~~~~~~~~~~"},{"line_number":426,"context_line":""},{"line_number":427,"context_line":"Use guest agents to enable optional access between compute nodes and guests"},{"line_number":428,"context_line":"through a socket, using the QMP protocol."},{"line_number":429,"context_line":""},{"line_number":430,"context_line":"To enable this feature, you must set ``hw_qemu_guest_agent\u003dyes`` as a metadata"},{"line_number":431,"context_line":"parameter on the image you wish to use to create the guest-agent-capable"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_05713c45","line":428,"range":{"start_line":428,"start_character":18,"end_line":428,"end_character":40},"in_reply_to":"3f79a3b5_a5a168f8","updated":"2018-10-16 09:15:10.000000000","message":"Ah, this text was pre-existing.  My brain wrongly assumed this whole doc was new.  Should\u0027ve figured it out from the `diff`.  Sorry for the noise.","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":2463,"name":"Florian Haas","email":"florian.haas@cleura.com","username":"fghaas"},"change_message_id":"93b776ae241e40e3eca93aaa865db1486249e5ac","unresolved":false,"context_lines":[{"line_number":425,"context_line":"~~~~~~~~~~~~~~~~~~~"},{"line_number":426,"context_line":""},{"line_number":427,"context_line":"Use guest agents to enable optional access between compute nodes and guests"},{"line_number":428,"context_line":"through a socket, using the QMP protocol."},{"line_number":429,"context_line":""},{"line_number":430,"context_line":"To enable this feature, you must set ``hw_qemu_guest_agent\u003dyes`` as a metadata"},{"line_number":431,"context_line":"parameter on the image you wish to use to create the guest-agent-capable"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_a5a168f8","line":428,"range":{"start_line":428,"start_character":18,"end_line":428,"end_character":40},"in_reply_to":"3f79a3b5_aa429765","updated":"2018-10-16 09:01:43.000000000","message":"Again, not my writing, and since I am *definitely* no qemu-ga expert, I\u0027d prefer to leave any changes here to someone who is.","commit_id":"8cf5489a52a4a22b9f4f2f2ac795c0d2fb32353c"},{"author":{"_account_id":6962,"name":"Kashyap Chamarthy","email":"kchamart@redhat.com","username":"kashyapc"},"change_message_id":"a2e2c2c4524c6ae5312094ba532bd538a367f867","unresolved":false,"context_lines":[{"line_number":312,"context_line":""},{"line_number":313,"context_line":"In ``host-passthrough`` mode, the guest can only be migrated to an"},{"line_number":314,"context_line":"exactly matching host CPU. Use this option only if"},{"line_number":315,"context_line":""},{"line_number":316,"context_line":"* your compute nodes have a very large degree of homogeneity"},{"line_number":317,"context_line":"  (i.e. substantially all of your compute nodes use the exact same CPU"},{"line_number":318,"context_line":"  generation and model), `or`"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_03083823","line":315,"updated":"2018-10-30 14:54:33.000000000","message":"That\u0027s generic enough, but more strictly speaking: with `host-passthrough`, along with identical host CPUs, one also needs identical *microcode*, _and_ host kernels.  (I\u0027m not speculating here, saw a real live migration fail due to mismatch in microcode between source and destination hosts.)","commit_id":"5a5e40a87a4c160fe43e3d6daa866173fefa3eb7"},{"author":{"_account_id":2463,"name":"Florian Haas","email":"florian.haas@cleura.com","username":"fghaas"},"change_message_id":"4d1544653aa4b5ded2ee81d6ee6b82842c1a1326","unresolved":false,"context_lines":[{"line_number":312,"context_line":""},{"line_number":313,"context_line":"In ``host-passthrough`` mode, the guest can only be migrated to an"},{"line_number":314,"context_line":"exactly matching host CPU. Use this option only if"},{"line_number":315,"context_line":""},{"line_number":316,"context_line":"* your compute nodes have a very large degree of homogeneity"},{"line_number":317,"context_line":"  (i.e. substantially all of your compute nodes use the exact same CPU"},{"line_number":318,"context_line":"  generation and model), `or`"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_eea6430f","line":315,"in_reply_to":"3f79a3b5_03083823","updated":"2018-10-30 15:13:56.000000000","message":"Good point, will fix. Thanks!","commit_id":"5a5e40a87a4c160fe43e3d6daa866173fefa3eb7"},{"author":{"_account_id":6873,"name":"Matt Riedemann","email":"mriedem.os@gmail.com","username":"mriedem"},"change_message_id":"59d37b3d7368bc7238293a75b5351d2263040675","unresolved":false,"context_lines":[{"line_number":353,"context_line":"Set CPU feature flags"},{"line_number":354,"context_line":"~~~~~~~~~~~~~~~~~~~~~"},{"line_number":355,"context_line":""},{"line_number":356,"context_line":"When operating in ``host-model`` or ``custom`` mode, it is also"},{"line_number":357,"context_line":"possible to selectively enable specific feature flags. Suppose your"},{"line_number":358,"context_line":"selected ``custom`` CPU model is ``IvyBridge``, which normally does"},{"line_number":359,"context_line":"not enable the ``pcid`` feature flag --- but you do want to pass"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_a3f5c454","line":356,"range":{"start_line":356,"start_character":0,"end_line":356,"end_character":51},"updated":"2018-10-30 15:09:04.000000000","message":"From the docs on https://docs.openstack.org/nova/latest/configuration/config.html#libvirt.cpu_model_extra_flags host-passthrough is also supported. Is that intentionally omitted here?","commit_id":"5a5e40a87a4c160fe43e3d6daa866173fefa3eb7"},{"author":{"_account_id":2463,"name":"Florian Haas","email":"florian.haas@cleura.com","username":"fghaas"},"change_message_id":"4d1544653aa4b5ded2ee81d6ee6b82842c1a1326","unresolved":false,"context_lines":[{"line_number":353,"context_line":"Set CPU feature flags"},{"line_number":354,"context_line":"~~~~~~~~~~~~~~~~~~~~~"},{"line_number":355,"context_line":""},{"line_number":356,"context_line":"When operating in ``host-model`` or ``custom`` mode, it is also"},{"line_number":357,"context_line":"possible to selectively enable specific feature flags. Suppose your"},{"line_number":358,"context_line":"selected ``custom`` CPU model is ``IvyBridge``, which normally does"},{"line_number":359,"context_line":"not enable the ``pcid`` feature flag --- but you do want to pass"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_6e45534a","line":356,"range":{"start_line":356,"start_character":0,"end_line":356,"end_character":51},"in_reply_to":"3f79a3b5_a3f5c454","updated":"2018-10-30 15:13:56.000000000","message":"No, it\u0027s not. I only never thought of why that would be useful, so it never crossed my mind to check that bit of documentation. But what\u0027s stated there does make sense. Thanks, will fix.","commit_id":"5a5e40a87a4c160fe43e3d6daa866173fefa3eb7"},{"author":{"_account_id":6873,"name":"Matt Riedemann","email":"mriedem.os@gmail.com","username":"mriedem"},"change_message_id":"59d37b3d7368bc7238293a75b5351d2263040675","unresolved":false,"context_lines":[{"line_number":385,"context_line":""},{"line_number":386,"context_line":".. code-block:: none"},{"line_number":387,"context_line":""},{"line_number":388,"context_line":"   options kvm_intel nested\u003d1"},{"line_number":389,"context_line":"   options kvm_amd nested\u003d1"},{"line_number":390,"context_line":""},{"line_number":391,"context_line":"A reboot may be required for the change to become effective."}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_c37560cb","line":388,"range":{"start_line":388,"start_character":21,"end_line":388,"end_character":29},"updated":"2018-10-30 15:09:04.000000000","message":"The devstack docs on this use nested\u003dy for kvm-intel:\n\nhttps://docs.openstack.org/devstack/latest/guides/devstack-with-nested-kvm.html\n\nDoes it matter? Or are both 1 and y considered the same thing?","commit_id":"5a5e40a87a4c160fe43e3d6daa866173fefa3eb7"},{"author":{"_account_id":2463,"name":"Florian Haas","email":"florian.haas@cleura.com","username":"fghaas"},"change_message_id":"4d1544653aa4b5ded2ee81d6ee6b82842c1a1326","unresolved":false,"context_lines":[{"line_number":385,"context_line":""},{"line_number":386,"context_line":".. code-block:: none"},{"line_number":387,"context_line":""},{"line_number":388,"context_line":"   options kvm_intel nested\u003d1"},{"line_number":389,"context_line":"   options kvm_amd nested\u003d1"},{"line_number":390,"context_line":""},{"line_number":391,"context_line":"A reboot may be required for the change to become effective."}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_8e98efcb","line":388,"range":{"start_line":388,"start_character":21,"end_line":388,"end_character":29},"in_reply_to":"3f79a3b5_c37560cb","updated":"2018-10-30 15:13:56.000000000","message":"kvm_intel supports Y and 1; modinfo gives the parameter as bool.\nkvm_amd only supports 1; modinfo lists the parameter as int.\n\nI simply picked the option that works in both, to make it less confusing.","commit_id":"5a5e40a87a4c160fe43e3d6daa866173fefa3eb7"},{"author":{"_account_id":2463,"name":"Florian Haas","email":"florian.haas@cleura.com","username":"fghaas"},"change_message_id":"ff21803487758fd52af551e1be1de60a9014f9ee","unresolved":false,"context_lines":[{"line_number":313,"context_line":"In ``host-passthrough`` mode, the guest can only be live-migrated to a"},{"line_number":314,"context_line":"target host that matches the source host extremely closely. This"},{"line_number":315,"context_line":"definitely includes the physical CPU model and running microcode, and"},{"line_number":316,"context_line":"may even include the running kernel. Use this model only if"},{"line_number":317,"context_line":""},{"line_number":318,"context_line":"* your compute nodes have a very large degree of homogeneity"},{"line_number":319,"context_line":"  (i.e. substantially all of your compute nodes use the exact same CPU"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_a9d285fd","line":316,"updated":"2018-10-30 16:02:24.000000000","message":"Ugh! Typo.","commit_id":"39378a6d12b76ea5a4db526fa119a26cc829a214"}]}