summaryrefslogtreecommitdiff
path: root/db/concurrency.h
blob: 9b91b0f235eed1c90deed6bc0dac5fb0b0295f91 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
/*
 *    Copyright (C) 2010 10gen Inc.
 *
 *    This program is free software: you can redistribute it and/or  modify
 *    it under the terms of the GNU Affero General Public License, version 3,
 *    as published by the Free Software Foundation.
 *
 *    This program is distributed in the hope that it will be useful,
 *    but WITHOUT ANY WARRANTY; without even the implied warranty of
 *    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *    GNU Affero General Public License for more details.
 *
 *    You should have received a copy of the GNU Affero General Public License
 *    along with this program.  If not, see <http://www.gnu.org/licenses/>.
 */

/* concurrency.h

   mongod concurrency rules & notes will be placed here.

   Mutex heirarchy (1 = "leaf")
     name                   level
     Logstream::mutex       1
     ClientCursor::ccmutex  2
     dblock                 3

     End func name with _inlock to indicate "caller must lock before calling".
*/

#pragma once

#include "../util/concurrency/rwlock.h"
#include "../util/mmap.h"

namespace mongo {

    string sayClientState();
    bool haveClient();
    
    void curopWaitingForLock( int type );
    void curopGotLock();

    /* mutex time stats */
    class MutexInfo {
        unsigned long long start, enter, timeLocked; // all in microseconds
        int locked;

    public:
        MutexInfo() : timeLocked(0) , locked(0) {
            start = curTimeMicros64();
        }
        void entered() {
            if ( locked == 0 )
                enter = curTimeMicros64();
            locked++;
            assert( locked >= 1 );
        }
        void leaving() {
            locked--;
            assert( locked >= 0 );
            if ( locked == 0 )
                timeLocked += curTimeMicros64() - enter;
        }
        int isLocked() const {
            return locked;
        }
        void getTimingInfo(unsigned long long &s, unsigned long long &tl) const {
            s = start;
            tl = timeLocked;
        }
        unsigned long long getTimeLocked() const {
            return timeLocked;
        }
    };

    class MongoMutex {
        MutexInfo _minfo;
        RWLock _m;
        ThreadLocalValue<int> _state;

        /* we use a separate TLS value for releasedEarly - that is ok as 
           our normal/common code path, we never even touch it.
        */
        ThreadLocalValue<bool> _releasedEarly;
    public:
        MongoMutex(const char * name) : _m(name) { }

        /**
         * @return
         *    > 0  write lock
         *    = 0  no lock
         *    < 0  read lock
         */
        int getState() { return _state.get(); }
        bool isWriteLocked() { return getState() > 0; }
        void assertWriteLocked() { 
            assert( getState() > 0 ); 
            DEV assert( !_releasedEarly.get() );
        }
        bool atLeastReadLocked() { return _state.get() != 0; }
        void assertAtLeastReadLocked() { assert(atLeastReadLocked()); }

        bool _checkWriteLockAlready(){
            //DEV cout << "LOCK" << endl;
            DEV assert( haveClient() );
                
            int s = _state.get();
            if( s > 0 ) {
                _state.set(s+1);
                return true;
            }

            massert( 10293 , (string)"internal error: locks are not upgradeable: " + sayClientState() , s == 0 );

            return false;
        }

        void lock() { 
            if ( _checkWriteLockAlready() )
                return;
            
            _state.set(1);

            curopWaitingForLock( 1 );
            _m.lock(); 
            curopGotLock();

            _minfo.entered();

            MongoFile::lockAll();
        }

        bool lock_try( int millis ) { 
            if ( _checkWriteLockAlready() )
                return true;

            curopWaitingForLock( 1 );
            bool got = _m.lock_try( millis ); 
            curopGotLock();
            
            if ( got ){
                _minfo.entered();
                _state.set(1);
                MongoFile::lockAll();
            }                
            
            return got;
        }


        void unlock() { 
            //DEV cout << "UNLOCK" << endl;
            int s = _state.get();
            if( s > 1 ) { 
                _state.set(s-1);
                return;
            }
            if( s != 1 ) { 
                if( _releasedEarly.get() ) { 
                    _releasedEarly.set(false);
                    return;
                }
                massert( 12599, "internal error: attempt to unlock when wasn't in a write lock", false);
            }

            MongoFile::unlockAll();

            _state.set(0);
            _minfo.leaving();
            _m.unlock(); 
        }

        /* unlock (write lock), and when unlock() is called later, 
           be smart then and don't unlock it again.
           */
        void releaseEarly() {
            assert( getState() == 1 ); // must not be recursive
            assert( !_releasedEarly.get() );
            _releasedEarly.set(true);
            unlock();
        }

        void lock_shared() { 
            //DEV cout << " LOCKSHARED" << endl;
            int s = _state.get();
            if( s ) {
                if( s > 0 ) { 
                    // already in write lock - just be recursive and stay write locked
                    _state.set(s+1);
                    return;
                }
                else { 
                    // already in read lock - recurse
                    _state.set(s-1);
                    return;
                }
            }
            _state.set(-1);
            curopWaitingForLock( -1 );
            _m.lock_shared(); 
            curopGotLock();
        }
        
        bool lock_shared_try( int millis ) {
            int s = _state.get();
            if ( s ){
                // we already have a lock, so no need to try
                lock_shared();
                return true;
            }

            bool got = _m.lock_shared_try( millis );
            if ( got )
                _state.set(-1);
            return got;
        }
        
        void unlock_shared() { 
            //DEV cout << " UNLOCKSHARED" << endl;
            int s = _state.get();
            if( s > 0 ) { 
                assert( s > 1 ); /* we must have done a lock write first to have s > 1 */
                _state.set(s-1);
                return;
            }
            if( s < -1 ) { 
                _state.set(s+1);
                return;
            }
            assert( s == -1 );
            _state.set(0);
            _m.unlock_shared(); 
        }
        
        MutexInfo& info() { return _minfo; }
    };

    extern MongoMutex &dbMutex;

    inline void dbunlocking_write() { }
    inline void dbunlocking_read() { }

    struct writelock {
        writelock(const string& ns) {
            dbMutex.lock();
        }
        ~writelock() { 
            DESTRUCTOR_GUARD(
                dbunlocking_write();
                dbMutex.unlock();
            );
        }
    };
    
    struct readlock {
        readlock(const string& ns) {
            dbMutex.lock_shared();
        }
        ~readlock() { 
            DESTRUCTOR_GUARD(
                dbunlocking_read();
                dbMutex.unlock_shared();
            );
        }
    };	

    struct readlocktry {
        readlocktry( const string&ns , int tryms ){
            _got = dbMutex.lock_shared_try( tryms );
        }
        ~readlocktry() {
            if ( _got ){
                dbunlocking_read();
                dbMutex.unlock_shared();
            }
        }
        bool got() const { return _got; }
    private:
        bool _got;
    };

    struct writelocktry {
        writelocktry( const string&ns , int tryms ){
            _got = dbMutex.lock_try( tryms );
        }
        ~writelocktry() {
            if ( _got ){
                dbunlocking_read();
                dbMutex.unlock();
            }
        }
        bool got() const { return _got; }
    private:
        bool _got;
    };

    struct readlocktryassert : public readlocktry { 
        readlocktryassert(const string& ns, int tryms) : 
          readlocktry(ns,tryms) { 
              uassert(13142, "timeout getting readlock", got());
        }
    };

    /** assure we have at least a read lock - they key with this being
        if you have a write lock, that's ok too.
    */
    struct atleastreadlock {
        atleastreadlock( const string& ns ){
            _prev = dbMutex.getState();
            if ( _prev == 0 )
                dbMutex.lock_shared();
        }
        ~atleastreadlock(){
            if ( _prev == 0 )
                dbMutex.unlock_shared();
        }
    private:
        int _prev;
    };

    class mongolock {
        bool _writelock;
    public:
        mongolock(bool write) : _writelock(write) {
            if( _writelock ) {
                dbMutex.lock();
            }
            else
                dbMutex.lock_shared();
        }
        ~mongolock() { 
            DESTRUCTOR_GUARD(
                if( _writelock ) { 
                    dbunlocking_write();
                    dbMutex.unlock();
                } else {
                    dbunlocking_read();
                    dbMutex.unlock_shared();
                }
            );
        }
        /* this unlocks, does NOT upgrade. that works for our current usage */
        void releaseAndWriteLock();
    };
    
    /* use writelock and readlock instead */
    struct dblock : public writelock {
        dblock() : writelock("") { }
    };

    // eliminate
    inline void assertInWriteLock() { dbMutex.assertWriteLocked(); }

}