)]}'
{"specs/rocky/approved/allow-disabling-ksm-via-extra-specs.rst":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"8b0d88e13e053bd35fa79e0682d6a77e9342acae","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"3f79a3b5_9cfead03","updated":"2018-08-20 12:06:47.000000000","message":"this needs to be re targeted to stein","commit_id":"3297412de407ddf06da4ad552ba83d22c3008e84"},{"author":{"_account_id":4690,"name":"melanie witt","display_name":"melwitt","email":"melwittt@gmail.com","username":"melwitt"},"change_message_id":"7833b4dd2f4d3981ca761548bc628a67e69d4194","unresolved":false,"context_lines":[{"line_number":18,"context_line":"Problem description"},{"line_number":19,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"Sometimes it\u0027s benefitial to be able to disable same page merging to improve VM\u0027s"},{"line_number":22,"context_line":"performance or avoid hypervisor induced issues caused by presense of similar VMs"},{"line_number":23,"context_line":"on the same hypervisor."},{"line_number":24,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_35d6a3f3","line":21,"updated":"2018-08-17 23:22:01.000000000","message":"This is the line the pep8 job failed on for line length \u003e 79 chars","commit_id":"3297412de407ddf06da4ad552ba83d22c3008e84"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"8b0d88e13e053bd35fa79e0682d6a77e9342acae","unresolved":false,"context_lines":[{"line_number":25,"context_line":"Use Cases"},{"line_number":26,"context_line":"---------"},{"line_number":27,"context_line":""},{"line_number":28,"context_line":"Cloud operators running into unexpected issues with VMs like freezing or stutter"},{"line_number":29,"context_line":"may try using this option to see if it resolves their issue."},{"line_number":30,"context_line":""},{"line_number":31,"context_line":"Proposed change"},{"line_number":32,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_b672a7e2","line":29,"range":{"start_line":28,"start_character":0,"end_line":29,"end_character":60},"updated":"2018-08-20 12:06:47.000000000","message":"kernel same page merging is basically a copy on write optimization that enables grater oversubscription of memory by merging read only or infrequently written pages between multipel process to refer to a single phyical page in memory.\n\nthe stuttering that is observed happens when a merged page is modified by the running guest in which case it needs to actually be copied by the kernel which intoduces more latancy then updating a non merged page.","commit_id":"3297412de407ddf06da4ad552ba83d22c3008e84"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"8b0d88e13e053bd35fa79e0682d6a77e9342acae","unresolved":false,"context_lines":[{"line_number":31,"context_line":"Proposed change"},{"line_number":32,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":33,"context_line":""},{"line_number":34,"context_line":"Add extra spec \"hw:ksm_disabled\" which will set \"nosharepages\" flag of Libvirt."},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"Alternatives"},{"line_number":37,"context_line":"------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_96b0cb8d","line":34,"range":{"start_line":34,"start_character":16,"end_line":34,"end_character":31},"updated":"2018-08-20 12:06:47.000000000","message":"i would just make this \"hw:kms\" and make it trisate boolean that if not set does nothing. if set to true will force kms and if set to false will disable kms.","commit_id":"3297412de407ddf06da4ad552ba83d22c3008e84"},{"author":{"_account_id":15140,"name":"Dmitry Sutyagin","email":"dsutyagin@mirantis.com","username":"dsutyagin"},"change_message_id":"fec27bb0a66bb6eda74e453e4a12efe28bd05265","unresolved":false,"context_lines":[{"line_number":31,"context_line":"Proposed change"},{"line_number":32,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":33,"context_line":""},{"line_number":34,"context_line":"Add extra spec \"hw:ksm_disabled\" which will set \"nosharepages\" flag of Libvirt."},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"Alternatives"},{"line_number":37,"context_line":"------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_71e4ebac","line":34,"range":{"start_line":34,"start_character":16,"end_line":34,"end_character":31},"in_reply_to":"3f79a3b5_96b0cb8d","updated":"2018-08-21 21:51:16.000000000","message":"Libvirt does not have such functionality (to force ksm). https://libvirt.org/formatdomain.html only \"\u003cnosharepages/\u003e\" to disable, and if not defined - determined by qemu, based on /etc/default/qemu-kvm and whether KSM was enabled in kernel config (afaik for major distros it is enabled).","commit_id":"3297412de407ddf06da4ad552ba83d22c3008e84"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"8b0d88e13e053bd35fa79e0682d6a77e9342acae","unresolved":false,"context_lines":[{"line_number":39,"context_line":"KSM may be disabled via \"hw:cpu_realtime\" but that adds additional flags"},{"line_number":40,"context_line":"which may not be required when the intent is to just disable KSM."},{"line_number":41,"context_line":""},{"line_number":42,"context_line":"KSM may be disabled via /etc/default/qemu-kvm.conf but this approach is"},{"line_number":43,"context_line":"not flexible, i.e. some hosts will always have KSM disabled."},{"line_number":44,"context_line":""},{"line_number":45,"context_line":"KSM may be disabled by recompiling the kernel, but this is even less flexible."},{"line_number":46,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_5606b322","line":43,"range":{"start_line":42,"start_character":0,"end_line":43,"end_character":60},"updated":"2018-08-20 12:06:47.000000000","message":"ksm on/off might be somting we should add to the host as a trait in placement.","commit_id":"3297412de407ddf06da4ad552ba83d22c3008e84"},{"author":{"_account_id":15140,"name":"Dmitry Sutyagin","email":"dsutyagin@mirantis.com","username":"dsutyagin"},"change_message_id":"fec27bb0a66bb6eda74e453e4a12efe28bd05265","unresolved":false,"context_lines":[{"line_number":39,"context_line":"KSM may be disabled via \"hw:cpu_realtime\" but that adds additional flags"},{"line_number":40,"context_line":"which may not be required when the intent is to just disable KSM."},{"line_number":41,"context_line":""},{"line_number":42,"context_line":"KSM may be disabled via /etc/default/qemu-kvm.conf but this approach is"},{"line_number":43,"context_line":"not flexible, i.e. some hosts will always have KSM disabled."},{"line_number":44,"context_line":""},{"line_number":45,"context_line":"KSM may be disabled by recompiling the kernel, but this is even less flexible."},{"line_number":46,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_f19ebb05","line":43,"range":{"start_line":42,"start_character":0,"end_line":43,"end_character":60},"in_reply_to":"3f79a3b5_5606b322","updated":"2018-08-21 21:51:16.000000000","message":"I really don\u0027t know that traits are so cannot elaborate on this. I was asked to create a spec and so I did, but my intention was only to add a simple flag to disable KSM, I\u0027m not a specialist in Nova architecture.","commit_id":"3297412de407ddf06da4ad552ba83d22c3008e84"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"80d3990cb1584895eb49459cf676026103c5cf62","unresolved":false,"context_lines":[{"line_number":39,"context_line":"KSM may be disabled via \"hw:cpu_realtime\" but that adds additional flags"},{"line_number":40,"context_line":"which may not be required when the intent is to just disable KSM."},{"line_number":41,"context_line":""},{"line_number":42,"context_line":"KSM may be disabled via /etc/default/qemu-kvm.conf but this approach is"},{"line_number":43,"context_line":"not flexible, i.e. some hosts will always have KSM disabled."},{"line_number":44,"context_line":""},{"line_number":45,"context_line":"KSM may be disabled by recompiling the kernel, but this is even less flexible."},{"line_number":46,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"3f79a3b5_6a57a659","line":43,"range":{"start_line":42,"start_character":0,"end_line":43,"end_character":60},"in_reply_to":"3f79a3b5_f19ebb05","updated":"2018-08-27 10:06:18.000000000","message":"traits are a set of capablities that we tag resouce providers in the placement api with do descibe the qualitaive aspects of a resouce.\n\nso ssd would be a trait for a block device.\na cpu instruct set would be a trait for a processor.\n\n\nso in this case i a suggsting that you would tag a host with and then schdule based on guest request. so to land on a host with kms disabeld you would set a fobinid trait of kms in the placement query which would be generated by setting hw:kms\u003doff in the flavor extra spec.","commit_id":"3297412de407ddf06da4ad552ba83d22c3008e84"}],"specs/stein/approved/allow-disabling-ksm-via-extra-specs.rst":[{"author":{"_account_id":27621,"name":"Vishakha Agarwal","email":"agarwalvishakha18@gmail.com","username":"Vishakha"},"change_message_id":"d3521f9958b68009290ef1fe2ecfa43161c088d8","unresolved":false,"context_lines":[{"line_number":18,"context_line":"Problem description"},{"line_number":19,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"Sometimes it\u0027s benefitial to be able to disable same page merging to improve"},{"line_number":22,"context_line":"VM\u0027s performance or avoid hypervisor induced issues caused by presense of"},{"line_number":23,"context_line":"similar VMs on the same hypervisor."},{"line_number":24,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_334cc640","line":21,"updated":"2018-08-27 11:21:25.000000000","message":"typo beneficial","commit_id":"ce22973c9c1a4115d0fcc01f06a6eefde1cc0fd5"},{"author":{"_account_id":15140,"name":"Dmitry Sutyagin","email":"dsutyagin@mirantis.com","username":"dsutyagin"},"change_message_id":"64fed7fc256d90bd3a8b2d2a30c415c28ed5b421","unresolved":false,"context_lines":[{"line_number":18,"context_line":"Problem description"},{"line_number":19,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"Sometimes it\u0027s benefitial to be able to disable same page merging to improve"},{"line_number":22,"context_line":"VM\u0027s performance or avoid hypervisor induced issues caused by presense of"},{"line_number":23,"context_line":"similar VMs on the same hypervisor."},{"line_number":24,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_e1566d2a","line":21,"in_reply_to":"3f79a3b5_334cc640","updated":"2018-08-27 20:26:18.000000000","message":"Done","commit_id":"ce22973c9c1a4115d0fcc01f06a6eefde1cc0fd5"},{"author":{"_account_id":27621,"name":"Vishakha Agarwal","email":"agarwalvishakha18@gmail.com","username":"Vishakha"},"change_message_id":"d3521f9958b68009290ef1fe2ecfa43161c088d8","unresolved":false,"context_lines":[{"line_number":29,"context_line":"may try using this option to see if it resolves their issue."},{"line_number":30,"context_line":"kernel same page merging is basically a copy on write optimization that enables"},{"line_number":31,"context_line":"grater oversubscription of memory by merging read only or infrequently written"},{"line_number":32,"context_line":"pages between multipel process to refer to a single phyical page in memory."},{"line_number":33,"context_line":"The stuttering that is observed happens when a merged page is modified by the"},{"line_number":34,"context_line":"running guest in which case it needs to actually be copied by the kernel which"},{"line_number":35,"context_line":"intoduces more latancy then updating a non merged page."}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_93411a0d","line":32,"updated":"2018-08-27 11:21:25.000000000","message":"typo multiple and physical","commit_id":"ce22973c9c1a4115d0fcc01f06a6eefde1cc0fd5"},{"author":{"_account_id":15140,"name":"Dmitry Sutyagin","email":"dsutyagin@mirantis.com","username":"dsutyagin"},"change_message_id":"64fed7fc256d90bd3a8b2d2a30c415c28ed5b421","unresolved":false,"context_lines":[{"line_number":29,"context_line":"may try using this option to see if it resolves their issue."},{"line_number":30,"context_line":"kernel same page merging is basically a copy on write optimization that enables"},{"line_number":31,"context_line":"grater oversubscription of memory by merging read only or infrequently written"},{"line_number":32,"context_line":"pages between multipel process to refer to a single phyical page in memory."},{"line_number":33,"context_line":"The stuttering that is observed happens when a merged page is modified by the"},{"line_number":34,"context_line":"running guest in which case it needs to actually be copied by the kernel which"},{"line_number":35,"context_line":"intoduces more latancy then updating a non merged page."}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_9c003e03","line":32,"in_reply_to":"3f79a3b5_93411a0d","updated":"2018-08-27 20:26:18.000000000","message":"Done","commit_id":"ce22973c9c1a4115d0fcc01f06a6eefde1cc0fd5"},{"author":{"_account_id":27621,"name":"Vishakha Agarwal","email":"agarwalvishakha18@gmail.com","username":"Vishakha"},"change_message_id":"d3521f9958b68009290ef1fe2ecfa43161c088d8","unresolved":false,"context_lines":[{"line_number":32,"context_line":"pages between multipel process to refer to a single phyical page in memory."},{"line_number":33,"context_line":"The stuttering that is observed happens when a merged page is modified by the"},{"line_number":34,"context_line":"running guest in which case it needs to actually be copied by the kernel which"},{"line_number":35,"context_line":"intoduces more latancy then updating a non merged page."},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"Proposed change"},{"line_number":38,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_d36472bb","line":35,"updated":"2018-08-27 11:21:25.000000000","message":"typo introduces","commit_id":"ce22973c9c1a4115d0fcc01f06a6eefde1cc0fd5"},{"author":{"_account_id":15140,"name":"Dmitry Sutyagin","email":"dsutyagin@mirantis.com","username":"dsutyagin"},"change_message_id":"64fed7fc256d90bd3a8b2d2a30c415c28ed5b421","unresolved":false,"context_lines":[{"line_number":32,"context_line":"pages between multipel process to refer to a single phyical page in memory."},{"line_number":33,"context_line":"The stuttering that is observed happens when a merged page is modified by the"},{"line_number":34,"context_line":"running guest in which case it needs to actually be copied by the kernel which"},{"line_number":35,"context_line":"intoduces more latancy then updating a non merged page."},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"Proposed change"},{"line_number":38,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_3c0faa2e","line":35,"in_reply_to":"3f79a3b5_d36472bb","updated":"2018-08-27 20:26:18.000000000","message":"Done","commit_id":"ce22973c9c1a4115d0fcc01f06a6eefde1cc0fd5"},{"author":{"_account_id":27621,"name":"Vishakha Agarwal","email":"agarwalvishakha18@gmail.com","username":"Vishakha"},"change_message_id":"d3521f9958b68009290ef1fe2ecfa43161c088d8","unresolved":false,"context_lines":[{"line_number":81,"context_line":""},{"line_number":82,"context_line":"Resolves guest OS hang in certain cases. May improve memory performance"},{"line_number":83,"context_line":"and decrease hypervisor\u0027s CPU usage at the cost of less free RAM due to"},{"line_number":84,"context_line":"disabled private page deduplication."},{"line_number":85,"context_line":""},{"line_number":86,"context_line":"Other deployer impact"},{"line_number":87,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_b30696cd","line":84,"updated":"2018-08-27 11:21:25.000000000","message":"I think u meant reduplication here?","commit_id":"ce22973c9c1a4115d0fcc01f06a6eefde1cc0fd5"},{"author":{"_account_id":15140,"name":"Dmitry Sutyagin","email":"dsutyagin@mirantis.com","username":"dsutyagin"},"change_message_id":"64fed7fc256d90bd3a8b2d2a30c415c28ed5b421","unresolved":false,"context_lines":[{"line_number":81,"context_line":""},{"line_number":82,"context_line":"Resolves guest OS hang in certain cases. May improve memory performance"},{"line_number":83,"context_line":"and decrease hypervisor\u0027s CPU usage at the cost of less free RAM due to"},{"line_number":84,"context_line":"disabled private page deduplication."},{"line_number":85,"context_line":""},{"line_number":86,"context_line":"Other deployer impact"},{"line_number":87,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"3f79a3b5_e1042d1a","line":84,"in_reply_to":"3f79a3b5_b30696cd","updated":"2018-08-27 20:26:18.000000000","message":"no, deduplication is correct, disabled deduplication \u003d duplicated data in memory is not merged.","commit_id":"ce22973c9c1a4115d0fcc01f06a6eefde1cc0fd5"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"92d8edb7672001beb2844d72b10854220c35f5cb","unresolved":false,"context_lines":[{"line_number":18,"context_line":"Problem description"},{"line_number":19,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":20,"context_line":""},{"line_number":21,"context_line":"Sometimes it\u0027s benefitial to be able to disable same page merging to improve"},{"line_number":22,"context_line":"VM\u0027s performance or avoid hypervisor induced issues caused by presense of"},{"line_number":23,"context_line":"similar VMs on the same hypervisor."},{"line_number":24,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_94c9d926","line":21,"range":{"start_line":21,"start_character":15,"end_line":21,"end_character":25},"updated":"2018-08-28 21:02:20.000000000","message":"beneficial","commit_id":"0171c56c9aac69e32758dd3955e236467fc0a44a"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"92d8edb7672001beb2844d72b10854220c35f5cb","unresolved":false,"context_lines":[{"line_number":27,"context_line":""},{"line_number":28,"context_line":"Cloud operators running into unexpected issues with VMs like freezing or stutter"},{"line_number":29,"context_line":"may try using this option to see if it resolves their issue."},{"line_number":30,"context_line":"kernel same page merging is basically a copy on write optimization that enables"},{"line_number":31,"context_line":"grater oversubscription of memory by merging read only or infrequently written"},{"line_number":32,"context_line":"pages between multipel process to refer to a single phyical page in memory."},{"line_number":33,"context_line":"The stuttering that is observed happens when a merged page is modified by the"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_34cec52d","line":30,"range":{"start_line":30,"start_character":0,"end_line":30,"end_character":6},"updated":"2018-08-28 21:02:20.000000000","message":"Kernel","commit_id":"0171c56c9aac69e32758dd3955e236467fc0a44a"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"92d8edb7672001beb2844d72b10854220c35f5cb","unresolved":false,"context_lines":[{"line_number":28,"context_line":"Cloud operators running into unexpected issues with VMs like freezing or stutter"},{"line_number":29,"context_line":"may try using this option to see if it resolves their issue."},{"line_number":30,"context_line":"kernel same page merging is basically a copy on write optimization that enables"},{"line_number":31,"context_line":"grater oversubscription of memory by merging read only or infrequently written"},{"line_number":32,"context_line":"pages between multipel process to refer to a single phyical page in memory."},{"line_number":33,"context_line":"The stuttering that is observed happens when a merged page is modified by the"},{"line_number":34,"context_line":"running guest in which case it needs to actually be copied by the kernel which"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_f4d7edbf","line":31,"range":{"start_line":31,"start_character":0,"end_line":31,"end_character":6},"updated":"2018-08-28 21:02:20.000000000","message":"greater","commit_id":"0171c56c9aac69e32758dd3955e236467fc0a44a"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"92d8edb7672001beb2844d72b10854220c35f5cb","unresolved":false,"context_lines":[{"line_number":29,"context_line":"may try using this option to see if it resolves their issue."},{"line_number":30,"context_line":"kernel same page merging is basically a copy on write optimization that enables"},{"line_number":31,"context_line":"grater oversubscription of memory by merging read only or infrequently written"},{"line_number":32,"context_line":"pages between multipel process to refer to a single phyical page in memory."},{"line_number":33,"context_line":"The stuttering that is observed happens when a merged page is modified by the"},{"line_number":34,"context_line":"running guest in which case it needs to actually be copied by the kernel which"},{"line_number":35,"context_line":"introduces more latancy then updating a non merged page."}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_b4e1759e","line":32,"range":{"start_line":32,"start_character":14,"end_line":32,"end_character":22},"updated":"2018-08-28 21:02:20.000000000","message":"multiple","commit_id":"0171c56c9aac69e32758dd3955e236467fc0a44a"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"92d8edb7672001beb2844d72b10854220c35f5cb","unresolved":false,"context_lines":[{"line_number":29,"context_line":"may try using this option to see if it resolves their issue."},{"line_number":30,"context_line":"kernel same page merging is basically a copy on write optimization that enables"},{"line_number":31,"context_line":"grater oversubscription of memory by merging read only or infrequently written"},{"line_number":32,"context_line":"pages between multipel process to refer to a single phyical page in memory."},{"line_number":33,"context_line":"The stuttering that is observed happens when a merged page is modified by the"},{"line_number":34,"context_line":"running guest in which case it needs to actually be copied by the kernel which"},{"line_number":35,"context_line":"introduces more latancy then updating a non merged page."}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_d4de71d9","line":32,"range":{"start_line":32,"start_character":23,"end_line":32,"end_character":30},"updated":"2018-08-28 21:02:20.000000000","message":"processes","commit_id":"0171c56c9aac69e32758dd3955e236467fc0a44a"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"92d8edb7672001beb2844d72b10854220c35f5cb","unresolved":false,"context_lines":[{"line_number":79,"context_line":"Performance Impact"},{"line_number":80,"context_line":"------------------"},{"line_number":81,"context_line":""},{"line_number":82,"context_line":"Resolves guest OS hang in certain cases. May improve memory performance"},{"line_number":83,"context_line":"and decrease hypervisor\u0027s CPU usage at the cost of less free RAM due to"},{"line_number":84,"context_line":"disabled private page deduplication."},{"line_number":85,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_d4a51149","line":82,"updated":"2018-08-28 21:02:20.000000000","message":"From what I\u0027ve read of the rest of comments, there seems to be a very big maybe associated with this statement. If the flag is an extra spec, yet it can only work if the host also happens to be configured in a few different ways where the flag could actually make any difference, then lots of things need to align for it to work.\n\nI don\u0027t think it is a good idea to introduce this kind of ambiguity. Either the requested action needs to always work or fail in predictable ways. It is very bad experience for things to be in a \"this might have fixed it, but it may also be the case that something else did.\"","commit_id":"0171c56c9aac69e32758dd3955e236467fc0a44a"},{"author":{"_account_id":12356,"name":"Vladyslav Drok","email":"vdrok@mirantis.com","username":"vdrok"},"change_message_id":"6ebff509d9b7d6cab27d17a1c8543a1c2867bb04","unresolved":false,"context_lines":[{"line_number":79,"context_line":"Performance Impact"},{"line_number":80,"context_line":"------------------"},{"line_number":81,"context_line":""},{"line_number":82,"context_line":"Resolves guest OS hang in certain cases. May improve memory performance"},{"line_number":83,"context_line":"and decrease hypervisor\u0027s CPU usage at the cost of less free RAM due to"},{"line_number":84,"context_line":"disabled private page deduplication."},{"line_number":85,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"3f79a3b5_8b75539a","line":82,"in_reply_to":"3f79a3b5_d4a51149","updated":"2018-09-21 11:18:48.000000000","message":"indeed this does not seem to be a good and clear reasoning for the change.","commit_id":"0171c56c9aac69e32758dd3955e236467fc0a44a"}]}