NeilBrown | 4912147 | 2006-03-27 01:15:12 -0800 | [diff] [blame] | 1 | This document gives a brief introduction to the caching |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 2 | mechanisms in the sunrpc layer that is used, in particular, |
| 3 | for NFS authentication. |
| 4 | |
| 5 | CACHES |
| 6 | ====== |
| 7 | The caching replaces the old exports table and allows for |
| 8 | a wide variety of values to be caches. |
| 9 | |
| 10 | There are a number of caches that are similar in structure though |
| 11 | quite possibly very different in content and use. There is a corpus |
| 12 | of common code for managing these caches. |
| 13 | |
| 14 | Examples of caches that are likely to be needed are: |
| 15 | - mapping from IP address to client name |
| 16 | - mapping from client name and filesystem to export options |
| 17 | - mapping from UID to list of GIDs, to work around NFS's limitation |
| 18 | of 16 gids. |
| 19 | - mappings between local UID/GID and remote UID/GID for sites that |
| 20 | do not have uniform uid assignment |
| 21 | - mapping from network identify to public key for crypto authentication. |
| 22 | |
| 23 | The common code handles such things as: |
| 24 | - general cache lookup with correct locking |
| 25 | - supporting 'NEGATIVE' as well as positive entries |
| 26 | - allowing an EXPIRED time on cache items, and removing |
Matt LaPlante | 2fe0ae7 | 2006-10-03 22:50:39 +0200 | [diff] [blame] | 27 | items after they expire, and are no longer in-use. |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 28 | - making requests to user-space to fill in cache entries |
| 29 | - allowing user-space to directly set entries in the cache |
| 30 | - delaying RPC requests that depend on as-yet incomplete |
| 31 | cache entries, and replaying those requests when the cache entry |
| 32 | is complete. |
NeilBrown | 4912147 | 2006-03-27 01:15:12 -0800 | [diff] [blame] | 33 | - clean out old entries as they expire. |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 34 | |
| 35 | Creating a Cache |
| 36 | ---------------- |
| 37 | |
NeilBrown | 4912147 | 2006-03-27 01:15:12 -0800 | [diff] [blame] | 38 | 1/ A cache needs a datum to store. This is in the form of a |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 39 | structure definition that must contain a |
| 40 | struct cache_head |
| 41 | as an element, usually the first. |
| 42 | It will also contain a key and some content. |
| 43 | Each cache element is reference counted and contains |
| 44 | expiry and update times for use in cache management. |
| 45 | 2/ A cache needs a "cache_detail" structure that |
NeilBrown | 4912147 | 2006-03-27 01:15:12 -0800 | [diff] [blame] | 46 | describes the cache. This stores the hash table, some |
| 47 | parameters for cache management, and some operations detailing how |
| 48 | to work with particular cache items. |
| 49 | The operations requires are: |
| 50 | struct cache_head *alloc(void) |
| 51 | This simply allocates appropriate memory and returns |
| 52 | a pointer to the cache_detail embedded within the |
| 53 | structure |
| 54 | void cache_put(struct kref *) |
| 55 | This is called when the last reference to an item is |
Paolo Ornati | 670e9f3 | 2006-10-03 22:57:56 +0200 | [diff] [blame] | 56 | dropped. The pointer passed is to the 'ref' field |
NeilBrown | 4912147 | 2006-03-27 01:15:12 -0800 | [diff] [blame] | 57 | in the cache_head. cache_put should release any |
| 58 | references create by 'cache_init' and, if CACHE_VALID |
| 59 | is set, any references created by cache_update. |
| 60 | It should then release the memory allocated by |
| 61 | 'alloc'. |
| 62 | int match(struct cache_head *orig, struct cache_head *new) |
| 63 | test if the keys in the two structures match. Return |
| 64 | 1 if they do, 0 if they don't. |
| 65 | void init(struct cache_head *orig, struct cache_head *new) |
| 66 | Set the 'key' fields in 'new' from 'orig'. This may |
| 67 | include taking references to shared objects. |
| 68 | void update(struct cache_head *orig, struct cache_head *new) |
| 69 | Set the 'content' fileds in 'new' from 'orig'. |
| 70 | int cache_show(struct seq_file *m, struct cache_detail *cd, |
| 71 | struct cache_head *h) |
| 72 | Optional. Used to provide a /proc file that lists the |
| 73 | contents of a cache. This should show one item, |
| 74 | usually on just one line. |
| 75 | int cache_request(struct cache_detail *cd, struct cache_head *h, |
| 76 | char **bpp, int *blen) |
| 77 | Format a request to be send to user-space for an item |
| 78 | to be instantiated. *bpp is a buffer of size *blen. |
| 79 | bpp should be moved forward over the encoded message, |
| 80 | and *blen should be reduced to show how much free |
| 81 | space remains. Return 0 on success or <0 if not |
| 82 | enough room or other problem. |
| 83 | int cache_parse(struct cache_detail *cd, char *buf, int len) |
| 84 | A message from user space has arrived to fill out a |
| 85 | cache entry. It is in 'buf' of length 'len'. |
| 86 | cache_parse should parse this, find the item in the |
| 87 | cache with sunrpc_cache_lookup, and update the item |
| 88 | with sunrpc_cache_update. |
| 89 | |
| 90 | |
| 91 | 3/ A cache needs to be registered using cache_register(). This |
| 92 | includes it on a list of caches that will be regularly |
| 93 | cleaned to discard old data. |
| 94 | |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 95 | Using a cache |
| 96 | ------------- |
| 97 | |
NeilBrown | 4912147 | 2006-03-27 01:15:12 -0800 | [diff] [blame] | 98 | To find a value in a cache, call sunrpc_cache_lookup passing a pointer |
| 99 | to the cache_head in a sample item with the 'key' fields filled in. |
| 100 | This will be passed to ->match to identify the target entry. If no |
| 101 | entry is found, a new entry will be create, added to the cache, and |
| 102 | marked as not containing valid data. |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 103 | |
NeilBrown | 4912147 | 2006-03-27 01:15:12 -0800 | [diff] [blame] | 104 | The item returned is typically passed to cache_check which will check |
| 105 | if the data is valid, and may initiate an up-call to get fresh data. |
| 106 | cache_check will return -ENOENT in the entry is negative or if an up |
| 107 | call is needed but not possible, -EAGAIN if an upcall is pending, |
| 108 | or 0 if the data is valid; |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 109 | |
| 110 | cache_check can be passed a "struct cache_req *". This structure is |
| 111 | typically embedded in the actual request and can be used to create a |
| 112 | deferred copy of the request (struct cache_deferred_req). This is |
| 113 | done when the found cache item is not uptodate, but the is reason to |
| 114 | believe that userspace might provide information soon. When the cache |
| 115 | item does become valid, the deferred copy of the request will be |
| 116 | revisited (->revisit). It is expected that this method will |
| 117 | reschedule the request for processing. |
| 118 | |
NeilBrown | 4912147 | 2006-03-27 01:15:12 -0800 | [diff] [blame] | 119 | The value returned by sunrpc_cache_lookup can also be passed to |
| 120 | sunrpc_cache_update to set the content for the item. A second item is |
| 121 | passed which should hold the content. If the item found by _lookup |
| 122 | has valid data, then it is discarded and a new item is created. This |
| 123 | saves any user of an item from worrying about content changing while |
| 124 | it is being inspected. If the item found by _lookup does not contain |
| 125 | valid data, then the content is copied across and CACHE_VALID is set. |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 126 | |
| 127 | Populating a cache |
| 128 | ------------------ |
| 129 | |
| 130 | Each cache has a name, and when the cache is registered, a directory |
| 131 | with that name is created in /proc/net/rpc |
| 132 | |
| 133 | This directory contains a file called 'channel' which is a channel |
| 134 | for communicating between kernel and user for populating the cache. |
| 135 | This directory may later contain other files of interacting |
| 136 | with the cache. |
| 137 | |
| 138 | The 'channel' works a bit like a datagram socket. Each 'write' is |
| 139 | passed as a whole to the cache for parsing and interpretation. |
| 140 | Each cache can treat the write requests differently, but it is |
| 141 | expected that a message written will contain: |
| 142 | - a key |
| 143 | - an expiry time |
| 144 | - a content. |
| 145 | with the intention that an item in the cache with the give key |
| 146 | should be create or updated to have the given content, and the |
| 147 | expiry time should be set on that item. |
| 148 | |
| 149 | Reading from a channel is a bit more interesting. When a cache |
NeilBrown | 4912147 | 2006-03-27 01:15:12 -0800 | [diff] [blame] | 150 | lookup fails, or when it succeeds but finds an entry that may soon |
| 151 | expire, a request is lodged for that cache item to be updated by |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 152 | user-space. These requests appear in the channel file. |
| 153 | |
| 154 | Successive reads will return successive requests. |
| 155 | If there are no more requests to return, read will return EOF, but a |
| 156 | select or poll for read will block waiting for another request to be |
| 157 | added. |
| 158 | |
| 159 | Thus a user-space helper is likely to: |
| 160 | open the channel. |
| 161 | select for readable |
| 162 | read a request |
| 163 | write a response |
| 164 | loop. |
| 165 | |
NeilBrown | 4912147 | 2006-03-27 01:15:12 -0800 | [diff] [blame] | 166 | If it dies and needs to be restarted, any requests that have not been |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 167 | answered will still appear in the file and will be read by the new |
| 168 | instance of the helper. |
| 169 | |
| 170 | Each cache should define a "cache_parse" method which takes a message |
| 171 | written from user-space and processes it. It should return an error |
| 172 | (which propagates back to the write syscall) or 0. |
| 173 | |
| 174 | Each cache should also define a "cache_request" method which |
| 175 | takes a cache item and encodes a request into the buffer |
| 176 | provided. |
| 177 | |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 178 | Note: If a cache has no active readers on the channel, and has had not |
| 179 | active readers for more than 60 seconds, further requests will not be |
NeilBrown | 4912147 | 2006-03-27 01:15:12 -0800 | [diff] [blame] | 180 | added to the channel but instead all lookups that do not find a valid |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 181 | entry will fail. This is partly for backward compatibility: The |
| 182 | previous nfs exports table was deemed to be authoritative and a |
| 183 | failed lookup meant a definite 'no'. |
| 184 | |
| 185 | request/response format |
| 186 | ----------------------- |
| 187 | |
| 188 | While each cache is free to use it's own format for requests |
NeilBrown | 4912147 | 2006-03-27 01:15:12 -0800 | [diff] [blame] | 189 | and responses over channel, the following is recommended as |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 190 | appropriate and support routines are available to help: |
| 191 | Each request or response record should be printable ASCII |
| 192 | with precisely one newline character which should be at the end. |
| 193 | Fields within the record should be separated by spaces, normally one. |
| 194 | If spaces, newlines, or nul characters are needed in a field they |
NeilBrown | 4912147 | 2006-03-27 01:15:12 -0800 | [diff] [blame] | 195 | much be quoted. two mechanisms are available: |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 196 | 1/ If a field begins '\x' then it must contain an even number of |
| 197 | hex digits, and pairs of these digits provide the bytes in the |
| 198 | field. |
| 199 | 2/ otherwise a \ in the field must be followed by 3 octal digits |
| 200 | which give the code for a byte. Other characters are treated |
NeilBrown | 4912147 | 2006-03-27 01:15:12 -0800 | [diff] [blame] | 201 | as them selves. At the very least, space, newline, nul, and |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 202 | '\' must be quoted in this way. |