Commit | Line | Data |
---|---|---|
61b0d760 SJ |
1 | Overview: |
2 | ||
3 | Zswap is a lightweight compressed cache for swap pages. It takes pages that are | |
4 | in the process of being swapped out and attempts to compress them into a | |
5 | dynamically allocated RAM-based memory pool. zswap basically trades CPU cycles | |
6 | for potentially reduced swap I/O. This trade-off can also result in a | |
7 | significant performance improvement if reads from the compressed cache are | |
8 | faster than reads from a swap device. | |
9 | ||
10 | NOTE: Zswap is a new feature as of v3.11 and interacts heavily with memory | |
0151e3d6 | 11 | reclaim. This interaction has not been fully explored on the large set of |
61b0d760 SJ |
12 | potential configurations and workloads that exist. For this reason, zswap |
13 | is a work in progress and should be considered experimental. | |
14 | ||
15 | Some potential benefits: | |
16 | * Desktop/laptop users with limited RAM capacities can mitigate the | |
17 | performance impact of swapping. | |
18 | * Overcommitted guests that share a common I/O resource can | |
19 | dramatically reduce their swap I/O pressure, avoiding heavy handed I/O | |
20 | throttling by the hypervisor. This allows more work to get done with less | |
21 | impact to the guest workload and guests sharing the I/O subsystem | |
22 | * Users with SSDs as swap devices can extend the life of the device by | |
23 | drastically reducing life-shortening writes. | |
24 | ||
25 | Zswap evicts pages from compressed cache on an LRU basis to the backing swap | |
0151e3d6 | 26 | device when the compressed pool reaches its size limit. This requirement had |
61b0d760 SJ |
27 | been identified in prior community discussions. |
28 | ||
29 | To enabled zswap, the "enabled" attribute must be set to 1 at boot time. e.g. | |
30 | zswap.enabled=1 | |
31 | ||
32 | Design: | |
33 | ||
34 | Zswap receives pages for compression through the Frontswap API and is able to | |
35 | evict pages from its own compressed pool on an LRU basis and write them back to | |
36 | the backing swap device in the case that the compressed pool is full. | |
37 | ||
38 | Zswap makes use of zbud for the managing the compressed memory pool. Each | |
39 | allocation in zbud is not directly accessible by address. Rather, a handle is | |
0151e3d6 | 40 | returned by the allocation routine and that handle must be mapped before being |
61b0d760 SJ |
41 | accessed. The compressed memory pool grows on demand and shrinks as compressed |
42 | pages are freed. The pool is not preallocated. | |
43 | ||
44 | When a swap page is passed from frontswap to zswap, zswap maintains a mapping | |
45 | of the swap entry, a combination of the swap type and swap offset, to the zbud | |
46 | handle that references that compressed swap page. This mapping is achieved | |
47 | with a red-black tree per swap type. The swap offset is the search key for the | |
48 | tree nodes. | |
49 | ||
50 | During a page fault on a PTE that is a swap entry, frontswap calls the zswap | |
51 | load function to decompress the page into the page allocated by the page fault | |
52 | handler. | |
53 | ||
54 | Once there are no PTEs referencing a swap page stored in zswap (i.e. the count | |
55 | in the swap_map goes to 0) the swap code calls the zswap invalidate function, | |
56 | via frontswap, to free the compressed entry. | |
57 | ||
58 | Zswap seeks to be simple in its policies. Sysfs attributes allow for one user | |
0151e3d6 | 59 | controlled policy: |
61b0d760 SJ |
60 | * max_pool_percent - The maximum percentage of memory that the compressed |
61 | pool can occupy. | |
62 | ||
63 | Zswap allows the compressor to be selected at kernel boot time by setting the | |
64 | “compressor” attribute. The default compressor is lzo. e.g. | |
65 | zswap.compressor=deflate | |
66 | ||
67 | A debugfs interface is provided for various statistic about pool size, number | |
68 | of pages stored, and various counters for the reasons pages are rejected. |