Difference between revisions of "Userfaultfd"

Revision as of 16:11, 26 February 2015

Collected on this page is the design notes about supporting userfaultfd in CRIU

Tasks after restore should have lazy VMAs being backed by userfaultfd, the fd itself should be sent before resume to CRIU daemon and closed
Only MAP_PRIVATE | MAP_ANONYMOUS will be supported in the 1st version due to kernel constraints
On source side the page server should be taught to get pages from tasks dumped and send them out on the destination node
Protocol should include out-of-order pages and background pages pushing (sending them before demand from the process)
The now dump-only --page-server option should be used for restore action to tell that the memory will be sucked in through page server

@@ Line 6: / Line 6: @@
 * Tasks after restore should have lazy VMAs being backed by userfaultfd, the fd itself should be sent before resume to CRIU daemon and closed
 * Only MAP_PRIVATE | MAP_ANONYMOUS will be supported in the 1st version due to kernel constraints
-* On source side the [[page server]] should be taught to use page-read to send pages out on the destination one
+* On source side the [[page server]] should be taught to get pages from tasks dumped and send them out on the destination node
 * Protocol should include out-of-order pages and background pages pushing (sending them before demand from the process)
 * The now dump-only <code>--page-server</code> option should be used for restore action to tell that the memory will be sucked in through page server