1 ===========================================
2 Automatically bind swap device to numa node
3 ===========================================
5 If the system has more than one swap device and swap device has the node
6 information, we can make use of this information to decide which swap
7 device to use in get_swap_pages() to get better performance.
10 How to use this feature
11 =======================
13 Swap device has priority and that decides the order of it to be used. To make
14 use of automatically binding, there is no need to manipulate priority settings
15 for swap devices. e.g. on a 2 node machine, assume 2 swap devices swapA and
16 swapB, with swapA attached to node 0 and swapB attached to node 1, are going
17 to be swapped on. Simply swapping them on by doing::
22 Then node 0 will use the two swap devices in the order of swapA then swapB and
23 node 1 will use the two swap devices in the order of swapB then swapA. Note
24 that the order of them being swapped on doesn't matter.
26 A more complex example on a 4 node machine. Assume 6 swap devices are going to
27 be swapped on: swapA and swapB are attached to node 0, swapC is attached to
28 node 1, swapD and swapE are attached to node 2 and swapF is attached to node3.
29 The way to swap them on is the same as above::
38 Then node 0 will use them in the order of::
40 swapA/swapB -> swapC -> swapD -> swapE -> swapF
42 swapA and swapB will be used in a round robin mode before any other swap device.
44 node 1 will use them in the order of::
46 swapC -> swapA -> swapB -> swapD -> swapE -> swapF
48 node 2 will use them in the order of::
50 swapD/swapE -> swapA -> swapB -> swapC -> swapF
52 Similaly, swapD and swapE will be used in a round robin mode before any
55 node 3 will use them in the order of::
57 swapF -> swapA -> swapB -> swapC -> swapD -> swapE
60 Implementation details
61 ======================
63 The current code uses a priority based list, swap_avail_list, to decide
64 which swap device to use and if multiple swap devices share the same
65 priority, they are used round robin. This change here replaces the single
66 global swap_avail_list with a per-numa-node list, i.e. for each numa node,
67 it sees its own priority based list of available swap devices. Swap
68 device's priority can be promoted on its matching node's swap_avail_list.
70 The current swap device's priority is set as: user can set a >=0 value,
71 or the system will pick one starting from -1 then downwards. The priority
72 value in the swap_avail_list is the negated value of the swap device's
73 due to plist being sorted from low to high. The new policy doesn't change
74 the semantics for priority >=0 cases, the previous starting from -1 then
75 downwards now becomes starting from -2 then downwards and -1 is reserved
76 as the promoted value. So if multiple swap devices are attached to the same
77 node, they will all be promoted to priority -1 on that node's plist and will
78 be used round robin before any other swap devices.