1 .. SPDX-License-Identifier: GPL-2.0 1 .. SPDX-License-Identifier: GPL-2.0
2 2
3 ============================ 3 ============================
4 Glock internal locking rules 4 Glock internal locking rules
5 ============================ 5 ============================
6 6
7 This documents the basic principles of the glo 7 This documents the basic principles of the glock state machine
8 internals. Each glock (struct gfs2_glock in fs 8 internals. Each glock (struct gfs2_glock in fs/gfs2/incore.h)
9 has two main (internal) locks: 9 has two main (internal) locks:
10 10
11 1. A spinlock (gl_lockref.lock) which protect 11 1. A spinlock (gl_lockref.lock) which protects the internal state such
12 as gl_state, gl_target and the list of hol 12 as gl_state, gl_target and the list of holders (gl_holders)
13 2. A non-blocking bit lock, GLF_LOCK, which i 13 2. A non-blocking bit lock, GLF_LOCK, which is used to prevent other
14 threads from making calls to the DLM, etc. 14 threads from making calls to the DLM, etc. at the same time. If a
15 thread takes this lock, it must then call 15 thread takes this lock, it must then call run_queue (usually via the
16 workqueue) when it releases it in order to 16 workqueue) when it releases it in order to ensure any pending tasks
17 are completed. 17 are completed.
18 18
19 The gl_holders list contains all the queued lo 19 The gl_holders list contains all the queued lock requests (not
20 just the holders) associated with the glock. I 20 just the holders) associated with the glock. If there are any
21 held locks, then they will be contiguous entri 21 held locks, then they will be contiguous entries at the head
22 of the list. Locks are granted in strictly the 22 of the list. Locks are granted in strictly the order that they
23 are queued. !! 23 are queued, except for those marked LM_FLAG_PRIORITY which are
>> 24 used only during recovery, and even then only for journal locks.
24 25
25 There are three lock states that users of the 26 There are three lock states that users of the glock layer can request,
26 namely shared (SH), deferred (DF) and exclusiv 27 namely shared (SH), deferred (DF) and exclusive (EX). Those translate
27 to the following DLM lock modes: 28 to the following DLM lock modes:
28 29
29 ========== ====== ======================= 30 ========== ====== =====================================================
30 Glock mode DLM lock mode 31 Glock mode DLM lock mode
31 ========== ====== ======================= 32 ========== ====== =====================================================
32 UN IV/NL Unlocked (no DLM lock a 33 UN IV/NL Unlocked (no DLM lock associated with glock) or NL
33 SH PR (Protected read) 34 SH PR (Protected read)
34 DF CW (Concurrent write) 35 DF CW (Concurrent write)
35 EX EX (Exclusive) 36 EX EX (Exclusive)
36 ========== ====== ======================= 37 ========== ====== =====================================================
37 38
38 Thus DF is basically a shared mode which is in 39 Thus DF is basically a shared mode which is incompatible with the "normal"
39 shared lock mode, SH. In GFS2 the DF mode is u 40 shared lock mode, SH. In GFS2 the DF mode is used exclusively for direct I/O
40 operations. The glocks are basically a lock pl 41 operations. The glocks are basically a lock plus some routines which deal
41 with cache management. The following rules app 42 with cache management. The following rules apply for the cache:
42 43
43 ========== ============== ========== !! 44 ========== ========== ============== ========== ==============
44 Glock mode Cache Metadata Cache data !! 45 Glock mode Cache data Cache Metadata Dirty Data Dirty Metadata
45 ========== ============== ========== !! 46 ========== ========== ============== ========== ==============
46 UN No No !! 47 UN No No No No
47 DF Yes No !! 48 SH Yes Yes No No
48 SH Yes Yes !! 49 DF No Yes No No
49 EX Yes Yes !! 50 EX Yes Yes Yes Yes
50 ========== ============== ========== !! 51 ========== ========== ============== ========== ==============
51 52
52 These rules are implemented using the various 53 These rules are implemented using the various glock operations which
53 are defined for each type of glock. Not all ty 54 are defined for each type of glock. Not all types of glocks use
54 all the modes. Only inode glocks use the DF mo 55 all the modes. Only inode glocks use the DF mode for example.
55 56
56 Table of glock operations and per type constan 57 Table of glock operations and per type constants:
57 58
58 ============== =========================== !! 59 ============= =============================================================
59 Field Purpose 60 Field Purpose
60 ============== =========================== !! 61 ============= =============================================================
61 go_sync Called before remote state !! 62 go_xmote_th Called before remote state change (e.g. to sync dirty data)
62 go_xmote_bh Called after remote state c 63 go_xmote_bh Called after remote state change (e.g. to refill cache)
63 go_inval Called if remote state chan 64 go_inval Called if remote state change requires invalidating the cache
64 go_instantiate Called when a glock has bee !! 65 go_demote_ok Returns boolean value of whether its ok to demote a glock
65 go_held Called every time a glock h !! 66 (e.g. checks timeout, and that there is no cached data)
>> 67 go_lock Called for the first local holder of a lock
>> 68 go_unlock Called on the final local unlock of a lock
66 go_dump Called to print content of 69 go_dump Called to print content of object for debugfs file, or on
67 error to dump glock to the 70 error to dump glock to the log.
68 go_callback Called if the DLM sends a c <<
69 go_unlocked Called when a glock is unlo <<
70 go_type The type of the glock, ``LM 71 go_type The type of the glock, ``LM_TYPE_*``
>> 72 go_callback Called if the DLM sends a callback to drop this lock
71 go_flags GLOF_ASPACE is set, if the 73 go_flags GLOF_ASPACE is set, if the glock has an address space
72 associated with it 74 associated with it
73 ============== =========================== !! 75 ============= =============================================================
74 76
75 The minimum hold time for each lock is the tim 77 The minimum hold time for each lock is the time after a remote lock
76 grant for which we ignore remote demote reques 78 grant for which we ignore remote demote requests. This is in order to
77 prevent a situation where locks are being boun 79 prevent a situation where locks are being bounced around the cluster
78 from node to node with none of the nodes makin 80 from node to node with none of the nodes making any progress. This
79 tends to show up most with shared mmapped file !! 81 tends to show up most with shared mmaped files which are being written
80 to by multiple nodes. By delaying the demotion 82 to by multiple nodes. By delaying the demotion in response to a
81 remote callback, that gives the userspace prog 83 remote callback, that gives the userspace program time to make
82 some progress before the pages are unmapped. 84 some progress before the pages are unmapped.
83 85
84 Eventually, we hope to make the glock "EX" mod !! 86 There is a plan to try and remove the go_lock and go_unlock callbacks
85 local locking will be done with the i_mutex as !! 87 if possible, in order to try and speed up the fast path though the locking.
86 glock. !! 88 Also, eventually we hope to make the glock "EX" mode locally shared
>> 89 such that any local locking will be done with the i_mutex as required
>> 90 rather than via the glock.
87 91
88 Locking rules for glock operations: 92 Locking rules for glock operations:
89 93
90 ============== ====================== === !! 94 ============= ====================== =============================
91 Operation GLF_LOCK bit lock held gl_ 95 Operation GLF_LOCK bit lock held gl_lockref.lock spinlock held
92 ============== ====================== === !! 96 ============= ====================== =============================
93 go_sync Yes !! 97 go_xmote_th Yes No
94 go_xmote_bh Yes 98 go_xmote_bh Yes No
95 go_inval Yes 99 go_inval Yes No
96 go_instantiate No !! 100 go_demote_ok Sometimes Yes
97 go_held No !! 101 go_lock Yes No
>> 102 go_unlock Yes No
98 go_dump Sometimes 103 go_dump Sometimes Yes
99 go_callback Sometimes (N/A) 104 go_callback Sometimes (N/A) Yes
100 go_unlocked Yes !! 105 ============= ====================== =============================
101 ============== ====================== === <<
102 106
103 .. Note:: 107 .. Note::
104 108
105 Operations must not drop either the bit loc 109 Operations must not drop either the bit lock or the spinlock
106 if its held on entry. go_dump and do_demote 110 if its held on entry. go_dump and do_demote_ok must never block.
107 Note that go_dump will only be called if th 111 Note that go_dump will only be called if the glock's state
108 indicates that it is caching uptodate data. 112 indicates that it is caching uptodate data.
109 113
110 Glock locking order within GFS2: 114 Glock locking order within GFS2:
111 115
112 1. i_rwsem (if required) 116 1. i_rwsem (if required)
113 2. Rename glock (for rename only) 117 2. Rename glock (for rename only)
114 3. Inode glock(s) 118 3. Inode glock(s)
115 (Parents before children, inodes at "same 119 (Parents before children, inodes at "same level" with same parent in
116 lock number order) 120 lock number order)
117 4. Rgrp glock(s) (for (de)allocation operatio 121 4. Rgrp glock(s) (for (de)allocation operations)
118 5. Transaction glock (via gfs2_trans_begin) f 122 5. Transaction glock (via gfs2_trans_begin) for non-read operations
119 6. i_rw_mutex (if required) 123 6. i_rw_mutex (if required)
120 7. Page lock (always last, very important!) 124 7. Page lock (always last, very important!)
121 125
122 There are two glocks per inode. One deals with 126 There are two glocks per inode. One deals with access to the inode
123 itself (locking order as above), and the other 127 itself (locking order as above), and the other, known as the iopen
124 glock is used in conjunction with the i_nlink 128 glock is used in conjunction with the i_nlink field in the inode to
125 determine the lifetime of the inode in questio 129 determine the lifetime of the inode in question. Locking of inodes
126 is on a per-inode basis. Locking of rgrps is o 130 is on a per-inode basis. Locking of rgrps is on a per rgrp basis.
127 In general we prefer to lock local locks prior 131 In general we prefer to lock local locks prior to cluster locks.
128 132
129 Glock Statistics 133 Glock Statistics
130 ---------------- 134 ----------------
131 135
132 The stats are divided into two sets: those rel 136 The stats are divided into two sets: those relating to the
133 super block and those relating to an individua 137 super block and those relating to an individual glock. The
134 super block stats are done on a per cpu basis 138 super block stats are done on a per cpu basis in order to
135 try and reduce the overhead of gathering them. 139 try and reduce the overhead of gathering them. They are also
136 further divided by glock type. All timings are 140 further divided by glock type. All timings are in nanoseconds.
137 141
138 In the case of both the super block and glock 142 In the case of both the super block and glock statistics,
139 the same information is gathered in each case. 143 the same information is gathered in each case. The super
140 block timing statistics are used to provide de 144 block timing statistics are used to provide default values for
141 the glock timing statistics, so that newly cre 145 the glock timing statistics, so that newly created glocks
142 should have, as far as possible, a sensible st 146 should have, as far as possible, a sensible starting point.
143 The per-glock counters are initialised to zero 147 The per-glock counters are initialised to zero when the
144 glock is created. The per-glock statistics are 148 glock is created. The per-glock statistics are lost when
145 the glock is ejected from memory. 149 the glock is ejected from memory.
146 150
147 The statistics are divided into three pairs of 151 The statistics are divided into three pairs of mean and
148 variance, plus two counters. The mean/variance 152 variance, plus two counters. The mean/variance pairs are
149 smoothed exponential estimates and the algorit 153 smoothed exponential estimates and the algorithm used is
150 one which will be very familiar to those used 154 one which will be very familiar to those used to calculation
151 of round trip times in network code. See "TCP/ 155 of round trip times in network code. See "TCP/IP Illustrated,
152 Volume 1", W. Richard Stevens, sect 21.3, "Rou 156 Volume 1", W. Richard Stevens, sect 21.3, "Round-Trip Time Measurement",
153 p. 299 and onwards. Also, Volume 2, Sect. 25.1 157 p. 299 and onwards. Also, Volume 2, Sect. 25.10, p. 838 and onwards.
154 Unlike the TCP/IP Illustrated case, the mean a 158 Unlike the TCP/IP Illustrated case, the mean and variance are
155 not scaled, but are in units of integer nanose 159 not scaled, but are in units of integer nanoseconds.
156 160
157 The three pairs of mean/variance measure the f 161 The three pairs of mean/variance measure the following
158 things: 162 things:
159 163
160 1. DLM lock time (non-blocking requests) 164 1. DLM lock time (non-blocking requests)
161 2. DLM lock time (blocking requests) 165 2. DLM lock time (blocking requests)
162 3. Inter-request time (again to the DLM) 166 3. Inter-request time (again to the DLM)
163 167
164 A non-blocking request is one which will compl 168 A non-blocking request is one which will complete right
165 away, whatever the state of the DLM lock in qu 169 away, whatever the state of the DLM lock in question. That
166 currently means any requests when (a) the curr 170 currently means any requests when (a) the current state of
167 the lock is exclusive, i.e. a lock demotion (b 171 the lock is exclusive, i.e. a lock demotion (b) the requested
168 state is either null or unlocked (again, a dem 172 state is either null or unlocked (again, a demotion) or (c) the
169 "try lock" flag is set. A blocking request cov 173 "try lock" flag is set. A blocking request covers all the other
170 lock requests. 174 lock requests.
171 175
172 There are two counters. The first is there pri 176 There are two counters. The first is there primarily to show
173 how many lock requests have been made, and thu 177 how many lock requests have been made, and thus how much data
174 has gone into the mean/variance calculations. 178 has gone into the mean/variance calculations. The other counter
175 is counting queuing of holders at the top laye 179 is counting queuing of holders at the top layer of the glock
176 code. Hopefully that number will be a lot larg 180 code. Hopefully that number will be a lot larger than the number
177 of dlm lock requests issued. 181 of dlm lock requests issued.
178 182
179 So why gather these statistics? There are seve 183 So why gather these statistics? There are several reasons
180 we'd like to get a better idea of these timing 184 we'd like to get a better idea of these timings:
181 185
182 1. To be able to better set the glock "min hol 186 1. To be able to better set the glock "min hold time"
183 2. To spot performance issues more easily 187 2. To spot performance issues more easily
184 3. To improve the algorithm for selecting reso 188 3. To improve the algorithm for selecting resource groups for
185 allocation (to base it on lock wait time, r 189 allocation (to base it on lock wait time, rather than blindly
186 using a "try lock") 190 using a "try lock")
187 191
188 Due to the smoothing action of the updates, a 192 Due to the smoothing action of the updates, a step change in
189 some input quantity being sampled will only fu 193 some input quantity being sampled will only fully be taken
190 into account after 8 samples (or 4 for the var 194 into account after 8 samples (or 4 for the variance) and this
191 needs to be carefully considered when interpre 195 needs to be carefully considered when interpreting the
192 results. 196 results.
193 197
194 Knowing both the time it takes a lock request 198 Knowing both the time it takes a lock request to complete and
195 the average time between lock requests for a g 199 the average time between lock requests for a glock means we
196 can compute the total percentage of the time f 200 can compute the total percentage of the time for which the
197 node is able to use a glock vs. time that the 201 node is able to use a glock vs. time that the rest of the
198 cluster has its share. That will be very usefu 202 cluster has its share. That will be very useful when setting
199 the lock min hold time. 203 the lock min hold time.
200 204
201 Great care has been taken to ensure that we 205 Great care has been taken to ensure that we
202 measure exactly the quantities that we want, a 206 measure exactly the quantities that we want, as accurately
203 as possible. There are always inaccuracies in 207 as possible. There are always inaccuracies in any
204 measuring system, but I hope this is as accura 208 measuring system, but I hope this is as accurate as we
205 can reasonably make it. 209 can reasonably make it.
206 210
207 Per sb stats can be found here:: 211 Per sb stats can be found here::
208 212
209 /sys/kernel/debug/gfs2/<fsname>/sbstats 213 /sys/kernel/debug/gfs2/<fsname>/sbstats
210 214
211 Per glock stats can be found here:: 215 Per glock stats can be found here::
212 216
213 /sys/kernel/debug/gfs2/<fsname>/glstats 217 /sys/kernel/debug/gfs2/<fsname>/glstats
214 218
215 Assuming that debugfs is mounted on /sys/kerne 219 Assuming that debugfs is mounted on /sys/kernel/debug and also
216 that <fsname> is replaced with the name of the 220 that <fsname> is replaced with the name of the gfs2 filesystem
217 in question. 221 in question.
218 222
219 The abbreviations used in the output as are fo 223 The abbreviations used in the output as are follows:
220 224
221 ========= =================================== 225 ========= ================================================================
222 srtt Smoothed round trip time for non bl 226 srtt Smoothed round trip time for non blocking dlm requests
223 srttvar Variance estimate for srtt 227 srttvar Variance estimate for srtt
224 srttb Smoothed round trip time for (poten 228 srttb Smoothed round trip time for (potentially) blocking dlm requests
225 srttvarb Variance estimate for srttb 229 srttvarb Variance estimate for srttb
226 sirt Smoothed inter request time (for dl 230 sirt Smoothed inter request time (for dlm requests)
227 sirtvar Variance estimate for sirt 231 sirtvar Variance estimate for sirt
228 dlm Number of dlm requests made (dcnt i 232 dlm Number of dlm requests made (dcnt in glstats file)
229 queue Number of glock requests queued (qc 233 queue Number of glock requests queued (qcnt in glstats file)
230 ========= =================================== 234 ========= ================================================================
231 235
232 The sbstats file contains a set of these stats 236 The sbstats file contains a set of these stats for each glock type (so 8 lines
233 for each type) and for each cpu (one column pe 237 for each type) and for each cpu (one column per cpu). The glstats file contains
234 a set of these stats for each glock in a simil 238 a set of these stats for each glock in a similar format to the glocks file, but
235 using the format mean/variance for each of the 239 using the format mean/variance for each of the timing stats.
236 240
237 The gfs2_glock_lock_time tracepoint prints out 241 The gfs2_glock_lock_time tracepoint prints out the current values of the stats
238 for the glock in question, along with some add 242 for the glock in question, along with some addition information on each dlm
239 reply that is received: 243 reply that is received:
240 244
241 ====== ===================================== 245 ====== =======================================
242 status The status of the dlm request 246 status The status of the dlm request
243 flags The dlm request flags 247 flags The dlm request flags
244 tdiff The time taken by this specific reque 248 tdiff The time taken by this specific request
245 ====== ===================================== 249 ====== =======================================
246 250
247 (remaining fields as per above list) 251 (remaining fields as per above list)
248 252
249 253
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.