Documentation/RCU/trace.txt

   1 CONFIG_RCU_TRACE debugfs Files and Formats
   2
   3
   4 The rcutree and rcutiny implementations of RCU provide debugfs trace
   5 output that summarizes counters and state.  This information is useful for
   6 debugging RCU itself, and can sometimes also help to debug abuses of RCU.
   7 The following sections describe the debugfs files and formats, first
   8 for rcutree and next for rcutiny.
   9
  10
  11 CONFIG_TREE_RCU and CONFIG_TREE_PREEMPT_RCU debugfs Files and Formats
  12
  13 These implementations of RCU provides five debugfs files under the
  14 top-level directory RCU: rcu/rcudata (which displays fields in struct
  15 rcu_data), rcu/rcudata.csv (which is a .csv spreadsheet version of
  16 rcu/rcudata), rcu/rcugp (which displays grace-period counters),
  17 rcu/rcuhier (which displays the struct rcu_node hierarchy), and
  18 rcu/rcu_pending (which displays counts of the reasons that the
  19 rcu_pending() function decided that there was core RCU work to do).
  20
  21 The output of "cat rcu/rcudata" looks as follows:
  22
  23 rcu_sched:
  24   0 c=17829 g=17829 pq=1 pqc=17829 qp=0 dt=10951/1/0 df=1101 of=0 ri=36 ql=0 b=10
  25   1 c=17829 g=17829 pq=1 pqc=17829 qp=0 dt=16117/1/0 df=1015 of=0 ri=0 ql=0 b=10
  26   2 c=17829 g=17829 pq=1 pqc=17829 qp=0 dt=1445/1/0 df=1839 of=0 ri=0 ql=0 b=10
  27   3 c=17829 g=17829 pq=1 pqc=17829 qp=0 dt=6681/1/0 df=1545 of=0 ri=0 ql=0 b=10
  28   4 c=17829 g=17829 pq=1 pqc=17829 qp=0 dt=1003/1/0 df=1992 of=0 ri=0 ql=0 b=10
  29   5 c=17829 g=17830 pq=1 pqc=17829 qp=1 dt=3887/1/0 df=3331 of=0 ri=4 ql=2 b=10
  30   6 c=17829 g=17829 pq=1 pqc=17829 qp=0 dt=859/1/0 df=3224 of=0 ri=0 ql=0 b=10
  31   7 c=17829 g=17830 pq=0 pqc=17829 qp=1 dt=3761/1/0 df=1818 of=0 ri=0 ql=2 b=10
  32 rcu_bh:
  33   0 c=-275 g=-275 pq=1 pqc=-275 qp=0 dt=10951/1/0 df=0 of=0 ri=0 ql=0 b=10
  34   1 c=-275 g=-275 pq=1 pqc=-275 qp=0 dt=16117/1/0 df=13 of=0 ri=0 ql=0 b=10
  35   2 c=-275 g=-275 pq=1 pqc=-275 qp=0 dt=1445/1/0 df=15 of=0 ri=0 ql=0 b=10
  36   3 c=-275 g=-275 pq=1 pqc=-275 qp=0 dt=6681/1/0 df=9 of=0 ri=0 ql=0 b=10
  37   4 c=-275 g=-275 pq=1 pqc=-275 qp=0 dt=1003/1/0 df=15 of=0 ri=0 ql=0 b=10
  38   5 c=-275 g=-275 pq=1 pqc=-275 qp=0 dt=3887/1/0 df=15 of=0 ri=0 ql=0 b=10
  39   6 c=-275 g=-275 pq=1 pqc=-275 qp=0 dt=859/1/0 df=15 of=0 ri=0 ql=0 b=10
  40   7 c=-275 g=-275 pq=1 pqc=-275 qp=0 dt=3761/1/0 df=15 of=0 ri=0 ql=0 b=10
  41
  42 The first section lists the rcu_data structures for rcu_sched, the second
  43 for rcu_bh.  Note that CONFIG_TREE_PREEMPT_RCU kernels will have an
  44 additional section for rcu_preempt.  Each section has one line per CPU,
  45 or eight for this 8-CPU system.  The fields are as follows:
  46
  47 o       The number at the beginning of each line is the CPU number.
  48         CPUs numbers followed by an exclamation mark are offline,
  49         but have been online at least once since boot.  There will be
  50         no output for CPUs that have never been online, which can be
  51         a good thing in the surprisingly common case where NR_CPUS is
  52         substantially larger than the number of actual CPUs.
  53
  54 o       "c" is the count of grace periods that this CPU believes have
  55         completed.  CPUs in dynticks idle mode may lag quite a ways
  56         behind, for example, CPU 4 under "rcu_sched" above, which has
  57         slept through the past 25 RCU grace periods.  It is not unusual
  58         to see CPUs lagging by thousands of grace periods.
  59
  60 o       "g" is the count of grace periods that this CPU believes have
  61         started.  Again, CPUs in dynticks idle mode may lag behind.
  62         If the "c" and "g" values are equal, this CPU has already
  63         reported a quiescent state for the last RCU grace period that
  64         it is aware of, otherwise, the CPU believes that it owes RCU a
  65         quiescent state.
  66
  67 o       "pq" indicates that this CPU has passed through a quiescent state
  68         for the current grace period.  It is possible for "pq" to be
  69         "1" and "c" different than "g", which indicates that although
  70         the CPU has passed through a quiescent state, either (1) this
  71         CPU has not yet reported that fact, (2) some other CPU has not
  72         yet reported for this grace period, or (3) both.
  73
  74 o       "pqc" indicates which grace period the last-observed quiescent
  75         state for this CPU corresponds to.  This is important for handling
  76         the race between CPU 0 reporting an extended dynticks-idle
  77         quiescent state for CPU 1 and CPU 1 suddenly waking up and
  78         reporting its own quiescent state.  If CPU 1 was the last CPU
  79         for the current grace period, then the CPU that loses this race
  80         will attempt to incorrectly mark CPU 1 as having checked in for
  81         the next grace period!
  82
  83 o       "qp" indicates that RCU still expects a quiescent state from
  84         this CPU.
  85
  86 o       "dt" is the current value of the dyntick counter that is incremented
  87         when entering or leaving dynticks idle state, either by the
  88         scheduler or by irq.  The number after the first "/" is the
  89         interrupt nesting depth when in dyntick-idle state, or one
  90         greater than the interrupt-nesting depth otherwise.  The number
  91         after the second "/" is the NMI nesting depth.
  92
  93         This field is displayed only for CONFIG_NO_HZ kernels.
  94
  95 o       "df" is the number of times that some other CPU has forced a
  96         quiescent state on behalf of this CPU due to this CPU being in
  97         dynticks-idle state.
  98
  99         This field is displayed only for CONFIG_NO_HZ kernels.
 100
 101 o       "of" is the number of times that some other CPU has forced a
 102         quiescent state on behalf of this CPU due to this CPU being
 103         offline.  In a perfect world, this might neve happen, but it
 104         turns out that offlining and onlining a CPU can take several grace
 105         periods, and so there is likely to be an extended period of time
 106         when RCU believes that the CPU is online when it really is not.
 107         Please note that erring in the other direction (RCU believing a
 108         CPU is offline when it is really alive and kicking) is a fatal
 109         error, so it makes sense to err conservatively.
 110
 111 o       "ri" is the number of times that RCU has seen fit to send a
 112         reschedule IPI to this CPU in order to get it to report a
 113         quiescent state.
 114
 115 o       "ql" is the number of RCU callbacks currently residing on
 116         this CPU.  This is the total number of callbacks, regardless
 117         of what state they are in (new, waiting for grace period to
 118         start, waiting for grace period to end, ready to invoke).
 119
 120 o       "b" is the batch limit for this CPU.  If more than this number
 121         of RCU callbacks is ready to invoke, then the remainder will
 122         be deferred.
 123
 124 o       "ci" is the number of RCU callbacks that have been invoked for
 125         this CPU.  Note that ci+ql is the number of callbacks that have
 126         been registered in absence of CPU-hotplug activity.
 127
 128 o       "co" is the number of RCU callbacks that have been orphaned due to
 129         this CPU going offline.  These orphaned callbacks have been moved
 130         to an arbitrarily chosen online CPU.
 131
 132 o       "ca" is the number of RCU callbacks that have been adopted due to
 133         other CPUs going offline.  Note that ci+co-ca+ql is the number of
 134         RCU callbacks registered on this CPU.
 135
 136 There is also an rcu/rcudata.csv file with the same information in
 137 comma-separated-variable spreadsheet format.
 138
 139
 140 The output of "cat rcu/rcugp" looks as follows:
 141
 142 rcu_sched: completed=33062  gpnum=33063
 143 rcu_bh: completed=464  gpnum=464
 144
 145 Again, this output is for both "rcu_sched" and "rcu_bh".  Note that
 146 kernels built with CONFIG_TREE_PREEMPT_RCU will have an additional
 147 "rcu_preempt" line.  The fields are taken from the rcu_state structure,
 148 and are as follows:
 149
 150 o       "completed" is the number of grace periods that have completed.
 151         It is comparable to the "c" field from rcu/rcudata in that a
 152         CPU whose "c" field matches the value of "completed" is aware
 153         that the corresponding RCU grace period has completed.
 154
 155 o       "gpnum" is the number of grace periods that have started.  It is
 156         comparable to the "g" field from rcu/rcudata in that a CPU
 157         whose "g" field matches the value of "gpnum" is aware that the
 158         corresponding RCU grace period has started.
 159
 160         If these two fields are equal (as they are for "rcu_bh" above),
 161         then there is no grace period in progress, in other words, RCU
 162         is idle.  On the other hand, if the two fields differ (as they
 163         do for "rcu_sched" above), then an RCU grace period is in progress.
 164
 165
 166 The output of "cat rcu/rcuhier" looks as follows, with very long lines:
 167
 168 c=6902 g=6903 s=2 jfq=3 j=72c7 nfqs=13142/nfqsng=0(13142) fqlh=6
 169 1/1 .>. 0:127 ^0
 170 3/3 .>. 0:35 ^0    0/0 .>. 36:71 ^1    0/0 .>. 72:107 ^2    0/0 .>. 108:127 ^3
 171 3/3f .>. 0:5 ^0    2/3 .>. 6:11 ^1    0/0 .>. 12:17 ^2    0/0 .>. 18:23 ^3    0/0 .>. 24:29 ^4    0/0 .>. 30:35 ^5    0/0 .>. 36:41 ^0    0/0 .>. 42:47 ^1    0/0 .>. 48:53 ^2    0/0 .>. 54:59 ^3    0/0 .>. 60:65 ^4    0/0 .>. 66:71 ^5    0/0 .>. 72:77 ^0    0/0 .>. 78:83 ^1    0/0 .>. 84:89 ^2    0/0 .>. 90:95 ^3    0/0 .>. 96:101 ^4    0/0 .>. 102:107 ^5    0/0 .>. 108:113 ^0    0/0 .>. 114:119 ^1    0/0 .>. 120:125 ^2    0/0 .>. 126:127 ^3
 172 rcu_bh:
 173 c=-226 g=-226 s=1 jfq=-5701 j=72c7 nfqs=88/nfqsng=0(88) fqlh=0
 174 0/1 .>. 0:127 ^0
 175 0/3 .>. 0:35 ^0    0/0 .>. 36:71 ^1    0/0 .>. 72:107 ^2    0/0 .>. 108:127 ^3
 176 0/3f .>. 0:5 ^0    0/3 .>. 6:11 ^1    0/0 .>. 12:17 ^2    0/0 .>. 18:23 ^3    0/0 .>. 24:29 ^4    0/0 .>. 30:35 ^5    0/0 .>. 36:41 ^0    0/0 .>. 42:47 ^1    0/0 .>. 48:53 ^2    0/0 .>. 54:59 ^3    0/0 .>. 60:65 ^4    0/0 .>. 66:71 ^5    0/0 .>. 72:77 ^0    0/0 .>. 78:83 ^1    0/0 .>. 84:89 ^2    0/0 .>. 90:95 ^3    0/0 .>. 96:101 ^4    0/0 .>. 102:107 ^5    0/0 .>. 108:113 ^0    0/0 .>. 114:119 ^1    0/0 .>. 120:125 ^2    0/0 .>. 126:127 ^3
 177
 178 This is once again split into "rcu_sched" and "rcu_bh" portions,
 179 and CONFIG_TREE_PREEMPT_RCU kernels will again have an additional
 180 "rcu_preempt" section.  The fields are as follows:
 181
 182 o       "c" is exactly the same as "completed" under rcu/rcugp.
 183
 184 o       "g" is exactly the same as "gpnum" under rcu/rcugp.
 185
 186 o       "s" is the "signaled" state that drives force_quiescent_state()'s
 187         state machine.
 188
 189 o       "jfq" is the number of jiffies remaining for this grace period
 190         before force_quiescent_state() is invoked to help push things
 191         along.  Note that CPUs in dyntick-idle mode throughout the grace
 192         period will not report on their own, but rather must be check by
 193         some other CPU via force_quiescent_state().
 194
 195 o       "j" is the low-order four hex digits of the jiffies counter.
 196         Yes, Paul did run into a number of problems that turned out to
 197         be due to the jiffies counter no longer counting.  Why do you ask?
 198
 199 o       "nfqs" is the number of calls to force_quiescent_state() since
 200         boot.
 201
 202 o       "nfqsng" is the number of useless calls to force_quiescent_state(),
 203         where there wasn't actually a grace period active.  This can
 204         happen due to races.  The number in parentheses is the difference
 205         between "nfqs" and "nfqsng", or the number of times that
 206         force_quiescent_state() actually did some real work.
 207
 208 o       "fqlh" is the number of calls to force_quiescent_state() that
 209         exited immediately (without even being counted in nfqs above)
 210         due to contention on ->fqslock.
 211
 212 o       Each element of the form "1/1 0:127 ^0" represents one struct
 213         rcu_node.  Each line represents one level of the hierarchy, from
 214         root to leaves.  It is best to think of the rcu_data structures
 215         as forming yet another level after the leaves.  Note that there
 216         might be either one, two, or three levels of rcu_node structures,
 217         depending on the relationship between CONFIG_RCU_FANOUT and
 218         CONFIG_NR_CPUS.
 219
 220         o       The numbers separated by the "/" are the qsmask followed
 221                 by the qsmaskinit.  The qsmask will have one bit
 222                 set for each entity in the next lower level that
 223                 has not yet checked in for the current grace period.
 224                 The qsmaskinit will have one bit for each entity that is
 225                 currently expected to check in during each grace period.
 226                 The value of qsmaskinit is assigned to that of qsmask
 227                 at the beginning of each grace period.
 228
 229                 For example, for "rcu_sched", the qsmask of the first
 230                 entry of the lowest level is 0x14, meaning that we
 231                 are still waiting for CPUs 2 and 4 to check in for the
 232                 current grace period.
 233
 234         o       The characters separated by the ">" indicate the state
 235                 of the blocked-tasks lists.  A "T" preceding the ">"
 236                 indicates that at least one task blocked in an RCU
 237                 read-side critical section blocks the current grace
 238                 period, while a "." preceding the ">" indicates otherwise.
 239                 The character following the ">" indicates similarly for
 240                 the next grace period.  A "T" should appear in this
 241                 field only for rcu-preempt.
 242
 243         o       The numbers separated by the ":" are the range of CPUs
 244                 served by this struct rcu_node.  This can be helpful
 245                 in working out how the hierarchy is wired together.
 246
 247                 For example, the first entry at the lowest level shows
 248                 "0:5", indicating that it covers CPUs 0 through 5.
 249
 250         o       The number after the "^" indicates the bit in the
 251                 next higher level rcu_node structure that this
 252                 rcu_node structure corresponds to.
 253
 254                 For example, the first entry at the lowest level shows
 255                 "^0", indicating that it corresponds to bit zero in
 256                 the first entry at the middle level.
 257
 258
 259 The output of "cat rcu/rcu_pending" looks as follows:
 260
 261 rcu_sched:
 262   0 np=255892 qsp=53936 rpq=85 cbr=0 cng=14417 gpc=10033 gps=24320 nf=6445 nn=146741
 263   1 np=261224 qsp=54638 rpq=33 cbr=0 cng=25723 gpc=16310 gps=2849 nf=5912 nn=155792
 264   2 np=237496 qsp=49664 rpq=23 cbr=0 cng=2762 gpc=45478 gps=1762 nf=1201 nn=136629
 265   3 np=236249 qsp=48766 rpq=98 cbr=0 cng=286 gpc=48049 gps=1218 nf=207 nn=137723
 266   4 np=221310 qsp=46850 rpq=7 cbr=0 cng=26 gpc=43161 gps=4634 nf=3529 nn=123110
 267   5 np=237332 qsp=48449 rpq=9 cbr=0 cng=54 gpc=47920 gps=3252 nf=201 nn=137456
 268   6 np=219995 qsp=46718 rpq=12 cbr=0 cng=50 gpc=42098 gps=6093 nf=4202 nn=120834
 269   7 np=249893 qsp=49390 rpq=42 cbr=0 cng=72 gpc=38400 gps=17102 nf=41 nn=144888
 270 rcu_bh:
 271   0 np=146741 qsp=1419 rpq=6 cbr=0 cng=6 gpc=0 gps=0 nf=2 nn=145314
 272   1 np=155792 qsp=12597 rpq=3 cbr=0 cng=0 gpc=4 gps=8 nf=3 nn=143180
 273   2 np=136629 qsp=18680 rpq=1 cbr=0 cng=0 gpc=7 gps=6 nf=0 nn=117936
 274   3 np=137723 qsp=2843 rpq=0 cbr=0 cng=0 gpc=10 gps=7 nf=0 nn=134863
 275   4 np=123110 qsp=12433 rpq=0 cbr=0 cng=0 gpc=4 gps=2 nf=0 nn=110671
 276   5 np=137456 qsp=4210 rpq=1 cbr=0 cng=0 gpc=6 gps=5 nf=0 nn=133235
 277   6 np=120834 qsp=9902 rpq=2 cbr=0 cng=0 gpc=6 gps=3 nf=2 nn=110921
 278   7 np=144888 qsp=26336 rpq=0 cbr=0 cng=0 gpc=8 gps=2 nf=0 nn=118542
 279
 280 As always, this is once again split into "rcu_sched" and "rcu_bh"
 281 portions, with CONFIG_TREE_PREEMPT_RCU kernels having an additional
 282 "rcu_preempt" section.  The fields are as follows:
 283
 284 o       "np" is the number of times that __rcu_pending() has been invoked
 285         for the corresponding flavor of RCU.
 286
 287 o       "qsp" is the number of times that the RCU was waiting for a
 288         quiescent state from this CPU.
 289
 290 o       "rpq" is the number of times that the CPU had passed through
 291         a quiescent state, but not yet reported it to RCU.
 292
 293 o       "cbr" is the number of times that this CPU had RCU callbacks
 294         that had passed through a grace period, and were thus ready
 295         to be invoked.
 296
 297 o       "cng" is the number of times that this CPU needed another
 298         grace period while RCU was idle.
 299
 300 o       "gpc" is the number of times that an old grace period had
 301         completed, but this CPU was not yet aware of it.
 302
 303 o       "gps" is the number of times that a new grace period had started,
 304         but this CPU was not yet aware of it.
 305
 306 o       "nf" is the number of times that this CPU suspected that the
 307         current grace period had run for too long, and thus needed to
 308         be forced.
 309
 310         Please note that "forcing" consists of sending resched IPIs
 311         to holdout CPUs.  If that CPU really still is in an old RCU
 312         read-side critical section, then we really do have to wait for it.
 313         The assumption behing "forcing" is that the CPU is not still in
 314         an old RCU read-side critical section, but has not yet responded
 315         for some other reason.
 316
 317 o       "nn" is the number of times that this CPU needed nothing.  Alert
 318         readers will note that the rcu "nn" number for a given CPU very
 319         closely matches the rcu_bh "np" number for that same CPU.  This
 320         is due to short-circuit evaluation in rcu_pending().
 321
 322
 323 CONFIG_TINY_RCU and CONFIG_TINY_PREEMPT_RCU debugfs Files and Formats
 324
 325 These implementations of RCU provides a single debugfs file under the
 326 top-level directory RCU, namely rcu/rcudata, which displays fields in
 327 rcu_bh_ctrlblk, rcu_sched_ctrlblk and, for CONFIG_TINY_PREEMPT_RCU,
 328 rcu_preempt_ctrlblk.
 329
 330 The output of "cat rcu/rcudata" is as follows:
 331
 332 rcu_preempt: qlen=24 gp=1097669 g197/p197/c197 tasks=...
 333              ttb=. btg=no ntb=184 neb=0 nnb=183 j=01f7 bt=0274
 334              normal balk: nt=1097669 gt=0 bt=371 b=0 ny=25073378 nos=0
 335              exp balk: bt=0 nos=0
 336 rcu_sched: qlen: 0
 337 rcu_bh: qlen: 0
 338
 339 This is split into rcu_preempt, rcu_sched, and rcu_bh sections, with the
 340 rcu_preempt section appearing only in CONFIG_TINY_PREEMPT_RCU builds.
 341 The last three lines of the rcu_preempt section appear only in
 342 CONFIG_RCU_BOOST kernel builds.  The fields are as follows:
 343
 344 o       "qlen" is the number of RCU callbacks currently waiting either
 345         for an RCU grace period or waiting to be invoked.  This is the
 346         only field present for rcu_sched and rcu_bh, due to the
 347         short-circuiting of grace period in those two cases.
 348
 349 o       "gp" is the number of grace periods that have completed.
 350
 351 o       "g197/p197/c197" displays the grace-period state, with the
 352         "g" number being the number of grace periods that have started
 353         (mod 256), the "p" number being the number of grace periods
 354         that the CPU has responded to (also mod 256), and the "c"
 355         number being the number of grace periods that have completed
 356         (once again mode 256).
 357
 358         Why have both "gp" and "g"?  Because the data flowing into
 359         "gp" is only present in a CONFIG_RCU_TRACE kernel.
 360
 361 o       "tasks" is a set of bits.  The first bit is "T" if there are
 362         currently tasks that have recently blocked within an RCU
 363         read-side critical section, the second bit is "N" if any of the
 364         aforementioned tasks are blocking the current RCU grace period,
 365         and the third bit is "E" if any of the aforementioned tasks are
 366         blocking the current expedited grace period.  Each bit is "."
 367         if the corresponding condition does not hold.
 368
 369 o       "ttb" is a single bit.  It is "B" if any of the blocked tasks
 370         need to be priority boosted and "." otherwise.
 371
 372 o       "btg" indicates whether boosting has been carried out during
 373         the current grace period, with "exp" indicating that boosting
 374         is in progress for an expedited grace period, "no" indicating
 375         that boosting has not yet started for a normal grace period,
 376         "begun" indicating that boosting has bebug for a normal grace
 377         period, and "done" indicating that boosting has completed for
 378         a normal grace period.
 379
 380 o       "ntb" is the total number of tasks subjected to RCU priority boosting
 381         periods since boot.
 382
 383 o       "neb" is the number of expedited grace periods that have had
 384         to resort to RCU priority boosting since boot.
 385
 386 o       "nnb" is the number of normal grace periods that have had
 387         to resort to RCU priority boosting since boot.
 388
 389 o       "j" is the low-order 12 bits of the jiffies counter in hexadecimal.
 390
 391 o       "bt" is the low-order 12 bits of the value that the jiffies counter
 392         will have at the next time that boosting is scheduled to begin.
 393
 394 o       In the line beginning with "normal balk", the fields are as follows:
 395
 396         o       "nt" is the number of times that the system balked from
 397                 boosting because there were no blocked tasks to boost.
 398                 Note that the system will balk from boosting even if the
 399                 grace period is overdue when the currently running task
 400                 is looping within an RCU read-side critical section.
 401                 There is no point in boosting in this case, because
 402                 boosting a running task won't make it run any faster.
 403
 404         o       "gt" is the number of times that the system balked
 405                 from boosting because, although there were blocked tasks,
 406                 none of them were preventing the current grace period
 407                 from completing.
 408
 409         o       "bt" is the number of times that the system balked
 410                 from boosting because boosting was already in progress.
 411
 412         o       "b" is the number of times that the system balked from
 413                 boosting because boosting had already completed for
 414                 the grace period in question.
 415
 416         o       "ny" is the number of times that the system balked from
 417                 boosting because it was not yet time to start boosting
 418                 the grace period in question.
 419
 420         o       "nos" is the number of times that the system balked from
 421                 boosting for inexplicable ("not otherwise specified")
 422                 reasons.  This can actually happen due to races involving
 423                 increments of the jiffies counter.
 424
 425 o       In the line beginning with "exp balk", the fields are as follows:
 426
 427         o       "bt" is the number of times that the system balked from
 428                 boosting because there were no blocked tasks to boost.
 429
 430         o       "nos" is the number of times that the system balked from
 431                  boosting for inexplicable ("not otherwise specified")
 432                  reasons.