1 perf-c2c(1) 1 perf-c2c(1)
2 =========== 2 ===========
3 3
4 NAME 4 NAME
5 ---- 5 ----
6 perf-c2c - Shared Data C2C/HITM Analyzer. 6 perf-c2c - Shared Data C2C/HITM Analyzer.
7 7
8 SYNOPSIS 8 SYNOPSIS
9 -------- 9 --------
10 [verse] 10 [verse]
11 'perf c2c record' [<options>] <command> 11 'perf c2c record' [<options>] <command>
12 'perf c2c record' [<options>] \-- [<record com 12 'perf c2c record' [<options>] \-- [<record command options>] <command>
13 'perf c2c report' [<options>] 13 'perf c2c report' [<options>]
14 14
15 DESCRIPTION 15 DESCRIPTION
16 ----------- 16 -----------
17 C2C stands for Cache To Cache. 17 C2C stands for Cache To Cache.
18 18
19 The perf c2c tool provides means for Shared Da 19 The perf c2c tool provides means for Shared Data C2C/HITM analysis. It allows
20 you to track down the cacheline contentions. 20 you to track down the cacheline contentions.
21 21
22 On Intel, the tool is based on load latency an !! 22 On x86, the tool is based on load latency and precise store facility events
23 provided by Intel CPUs. On PowerPC, the tool u 23 provided by Intel CPUs. On PowerPC, the tool uses random instruction sampling
24 with thresholding feature. On AMD, the tool us !! 24 with thresholding feature.
25 limitations, perf c2c is not supported on Zen3 <<
26 sample load and store operations, therefore ha <<
27 required. See linkperf:perf-arm-spe[1] for a s <<
28 statistical nature of Arm SPE sampling, not ev <<
29 sampled. <<
30 25
31 These events provide: 26 These events provide:
32 - memory address of the access 27 - memory address of the access
33 - type of the access (load and store details 28 - type of the access (load and store details)
34 - latency (in cycles) of the load access 29 - latency (in cycles) of the load access
35 30
36 The c2c tool provide means to record this data 31 The c2c tool provide means to record this data and report back access details
37 for cachelines with highest contention - highe 32 for cachelines with highest contention - highest number of HITM accesses.
38 33
39 The basic workflow with this tool follows the 34 The basic workflow with this tool follows the standard record/report phase.
40 User uses the record command to record events 35 User uses the record command to record events data and report command to
41 display it. 36 display it.
42 37
43 38
44 RECORD OPTIONS 39 RECORD OPTIONS
45 -------------- 40 --------------
46 -e:: 41 -e::
47 --event=:: 42 --event=::
48 Select the PMU event. Use 'perf c2c re 43 Select the PMU event. Use 'perf c2c record -e list'
49 to list available events. 44 to list available events.
50 45
51 -v:: 46 -v::
52 --verbose:: 47 --verbose::
53 Be more verbose (show counter open err 48 Be more verbose (show counter open errors, etc).
54 49
55 -l:: 50 -l::
56 --ldlat:: 51 --ldlat::
57 Configure mem-loads latency. Supported !! 52 Configure mem-loads latency. (x86 only)
58 only. Ignored on other archs. <<
59 53
60 -k:: 54 -k::
61 --all-kernel:: 55 --all-kernel::
62 Configure all used events to run in ke 56 Configure all used events to run in kernel space.
63 57
64 -u:: 58 -u::
65 --all-user:: 59 --all-user::
66 Configure all used events to run in us 60 Configure all used events to run in user space.
67 61
68 REPORT OPTIONS 62 REPORT OPTIONS
69 -------------- 63 --------------
70 -k:: 64 -k::
71 --vmlinux=<file>:: 65 --vmlinux=<file>::
72 vmlinux pathname 66 vmlinux pathname
73 67
74 -v:: 68 -v::
75 --verbose:: 69 --verbose::
76 Be more verbose (show counter open err 70 Be more verbose (show counter open errors, etc).
77 71
78 -i:: 72 -i::
79 --input:: 73 --input::
80 Specify the input file to process. 74 Specify the input file to process.
81 75
82 -N:: 76 -N::
83 --node-info:: 77 --node-info::
84 Show extra node info in report (see NO 78 Show extra node info in report (see NODE INFO section)
85 79
86 -c:: 80 -c::
87 --coalesce:: 81 --coalesce::
88 Specify sorting fields for single cach 82 Specify sorting fields for single cacheline display.
89 Following fields are available: tid,pi 83 Following fields are available: tid,pid,iaddr,dso
90 (see COALESCE) 84 (see COALESCE)
91 85
92 -g:: 86 -g::
93 --call-graph:: 87 --call-graph::
94 Setup callchains parameters. 88 Setup callchains parameters.
95 Please refer to perf-report man page f 89 Please refer to perf-report man page for details.
96 90
97 --stdio:: 91 --stdio::
98 Force the stdio output (see STDIO OUTP 92 Force the stdio output (see STDIO OUTPUT)
99 93
100 --stats:: 94 --stats::
101 Display only statistic tables and forc 95 Display only statistic tables and force stdio mode.
102 96
103 --full-symbols:: 97 --full-symbols::
104 Display full length of symbols. 98 Display full length of symbols.
105 99
106 --no-source:: 100 --no-source::
107 Do not display Source:Line column. 101 Do not display Source:Line column.
108 102
109 --show-all:: 103 --show-all::
110 Show all captured HITM lines, with no 104 Show all captured HITM lines, with no regard to HITM % 0.0005 limit.
111 105
112 -f:: 106 -f::
113 --force:: 107 --force::
114 Don't do ownership validation. 108 Don't do ownership validation.
115 109
116 -d:: 110 -d::
117 --display:: 111 --display::
118 Switch to HITM type (rmt, lcl) or peer 112 Switch to HITM type (rmt, lcl) or peer snooping type (peer) to display
119 and sort on. Total HITMs (tot) as defa 113 and sort on. Total HITMs (tot) as default, except Arm64 uses peer mode
120 as default. 114 as default.
121 115
122 --stitch-lbr:: 116 --stitch-lbr::
123 Show callgraph with stitched LBRs, whi 117 Show callgraph with stitched LBRs, which may have more complete
124 callgraph. The perf.data file must hav 118 callgraph. The perf.data file must have been obtained using
125 perf c2c record --call-graph lbr. 119 perf c2c record --call-graph lbr.
126 Disabled by default. In common cases w 120 Disabled by default. In common cases with call stack overflows,
127 it can recreate better call stacks tha 121 it can recreate better call stacks than the default lbr call stack
128 output. But this approach is not foolp !! 122 output. But this approach is not full proof. There can be cases
129 where it creates incorrect call stacks 123 where it creates incorrect call stacks from incorrect matches.
130 The known limitations include exceptio 124 The known limitations include exception handing such as
131 setjmp/longjmp will have calls/returns 125 setjmp/longjmp will have calls/returns not match.
132 126
133 --double-cl:: <<
134 Group the detection of shared cachelin <<
135 granularity. Some architectures have a <<
136 feature, which causes cacheline sharin <<
137 size is doubled. <<
138 <<
139 C2C RECORD 127 C2C RECORD
140 ---------- 128 ----------
141 The perf c2c record command setup options rela 129 The perf c2c record command setup options related to HITM cacheline analysis
142 and calls standard perf record command. 130 and calls standard perf record command.
143 131
144 Following perf record options are configured b 132 Following perf record options are configured by default:
145 (check perf record man page for details) 133 (check perf record man page for details)
146 134
147 -W,-d,--phys-data,--sample-cpu 135 -W,-d,--phys-data,--sample-cpu
148 136
149 Unless specified otherwise with '-e' option, f 137 Unless specified otherwise with '-e' option, following events are monitored by
150 default on Intel: !! 138 default on x86:
151 139
152 cpu/mem-loads,ldlat=30/P 140 cpu/mem-loads,ldlat=30/P
153 cpu/mem-stores/P 141 cpu/mem-stores/P
154 142
155 following on AMD: <<
156 <<
157 ibs_op// <<
158 <<
159 and following on PowerPC: 143 and following on PowerPC:
160 144
161 cpu/mem-loads/ 145 cpu/mem-loads/
162 cpu/mem-stores/ 146 cpu/mem-stores/
163 147
164 User can pass any 'perf record' option behind 148 User can pass any 'perf record' option behind '--' mark, like (to enable
165 callchains and system wide monitoring): 149 callchains and system wide monitoring):
166 150
167 $ perf c2c record -- -g -a 151 $ perf c2c record -- -g -a
168 152
169 Please check RECORD OPTIONS section for specif 153 Please check RECORD OPTIONS section for specific c2c record options.
170 154
171 C2C REPORT 155 C2C REPORT
172 ---------- 156 ----------
173 The perf c2c report command displays shared da 157 The perf c2c report command displays shared data analysis. It comes in two
174 display modes: stdio and tui (default). 158 display modes: stdio and tui (default).
175 159
176 The report command workflow is following: 160 The report command workflow is following:
177 - sort all the data based on the cacheline a 161 - sort all the data based on the cacheline address
178 - store access details for each cacheline 162 - store access details for each cacheline
179 - sort all cachelines based on user settings 163 - sort all cachelines based on user settings
180 - display data 164 - display data
181 165
182 In general perf report output consist of 2 bas 166 In general perf report output consist of 2 basic views:
183 1) most expensive cachelines list 167 1) most expensive cachelines list
184 2) offsets details for each cacheline 168 2) offsets details for each cacheline
185 169
186 For each cacheline in the 1) list we display f 170 For each cacheline in the 1) list we display following data:
187 (Both stdio and TUI modes follow the same fiel 171 (Both stdio and TUI modes follow the same fields output)
188 172
189 Index 173 Index
190 - zero based index to identify the cacheline 174 - zero based index to identify the cacheline
191 175
192 Cacheline 176 Cacheline
193 - cacheline address (hex number) 177 - cacheline address (hex number)
194 178
195 Rmt/Lcl Hitm (Display with HITM types) 179 Rmt/Lcl Hitm (Display with HITM types)
196 - cacheline percentage of all Remote/Local H 180 - cacheline percentage of all Remote/Local HITM accesses
197 181
198 Peer Snoop (Display with peer type) 182 Peer Snoop (Display with peer type)
199 - cacheline percentage of all peer accesses 183 - cacheline percentage of all peer accesses
200 184
201 LLC Load Hitm - Total, LclHitm, RmtHitm (For 185 LLC Load Hitm - Total, LclHitm, RmtHitm (For display with HITM types)
202 - count of Total/Local/Remote load HITMs 186 - count of Total/Local/Remote load HITMs
203 187
204 Load Peer - Total, Local, Remote (For displa 188 Load Peer - Total, Local, Remote (For display with peer type)
205 - count of Total/Local/Remote load from peer 189 - count of Total/Local/Remote load from peer cache or DRAM
206 190
207 Total records 191 Total records
208 - sum of all cachelines accesses 192 - sum of all cachelines accesses
209 193
210 Total loads 194 Total loads
211 - sum of all load accesses 195 - sum of all load accesses
212 196
213 Total stores 197 Total stores
214 - sum of all store accesses 198 - sum of all store accesses
215 199
216 Store Reference - L1Hit, L1Miss, N/A 200 Store Reference - L1Hit, L1Miss, N/A
217 L1Hit - store accesses that hit L1 201 L1Hit - store accesses that hit L1
218 L1Miss - store accesses that missed L1 202 L1Miss - store accesses that missed L1
219 N/A - store accesses with memory level is 203 N/A - store accesses with memory level is not available
220 204
221 Core Load Hit - FB, L1, L2 205 Core Load Hit - FB, L1, L2
222 - count of load hits in FB (Fill Buffer), L1 206 - count of load hits in FB (Fill Buffer), L1 and L2 cache
223 207
224 LLC Load Hit - LlcHit, LclHitm 208 LLC Load Hit - LlcHit, LclHitm
225 - count of LLC load accesses, includes LLC h 209 - count of LLC load accesses, includes LLC hits and LLC HITMs
226 210
227 RMT Load Hit - RmtHit, RmtHitm 211 RMT Load Hit - RmtHit, RmtHitm
228 - count of remote load accesses, includes re 212 - count of remote load accesses, includes remote hits and remote HITMs;
229 on Arm neoverse cores, RmtHit is used to a 213 on Arm neoverse cores, RmtHit is used to account remote accesses,
230 includes remote DRAM or any upward cache l 214 includes remote DRAM or any upward cache level in remote node
231 215
232 Load Dram - Lcl, Rmt 216 Load Dram - Lcl, Rmt
233 - count of local and remote DRAM accesses 217 - count of local and remote DRAM accesses
234 218
235 For each offset in the 2) list we display foll 219 For each offset in the 2) list we display following data:
236 220
237 HITM - Rmt, Lcl (Display with HITM types) 221 HITM - Rmt, Lcl (Display with HITM types)
238 - % of Remote/Local HITM accesses for given 222 - % of Remote/Local HITM accesses for given offset within cacheline
239 223
240 Peer Snoop - Rmt, Lcl (Display with peer typ 224 Peer Snoop - Rmt, Lcl (Display with peer type)
241 - % of Remote/Local peer accesses for given 225 - % of Remote/Local peer accesses for given offset within cacheline
242 226
243 Store Refs - L1 Hit, L1 Miss, N/A 227 Store Refs - L1 Hit, L1 Miss, N/A
244 - % of store accesses that hit L1, missed L1 228 - % of store accesses that hit L1, missed L1 and N/A (no available) memory
245 level for given offset within cacheline 229 level for given offset within cacheline
246 230
247 Data address - Offset 231 Data address - Offset
248 - offset address 232 - offset address
249 233
250 Pid 234 Pid
251 - pid of the process responsible for the acc 235 - pid of the process responsible for the accesses
252 236
253 Tid 237 Tid
254 - tid of the process responsible for the acc 238 - tid of the process responsible for the accesses
255 239
256 Code address 240 Code address
257 - code address responsible for the accesses 241 - code address responsible for the accesses
258 242
259 cycles - rmt hitm, lcl hitm, load (Display w 243 cycles - rmt hitm, lcl hitm, load (Display with HITM types)
260 - sum of cycles for given accesses - Remot 244 - sum of cycles for given accesses - Remote/Local HITM and generic load
261 245
262 cycles - rmt peer, lcl peer, load (Display w 246 cycles - rmt peer, lcl peer, load (Display with peer type)
263 - sum of cycles for given accesses - Remot 247 - sum of cycles for given accesses - Remote/Local peer load and generic load
264 248
265 cpu cnt 249 cpu cnt
266 - number of cpus that participated on the 250 - number of cpus that participated on the access
267 251
268 Symbol 252 Symbol
269 - code symbol related to the 'Code address 253 - code symbol related to the 'Code address' value
270 254
271 Shared Object 255 Shared Object
272 - shared object name related to the 'Code 256 - shared object name related to the 'Code address' value
273 257
274 Source:Line 258 Source:Line
275 - source information related to the 'Code 259 - source information related to the 'Code address' value
276 260
277 Node 261 Node
278 - nodes participating on the access (see N 262 - nodes participating on the access (see NODE INFO section)
279 263
280 NODE INFO 264 NODE INFO
281 --------- 265 ---------
282 The 'Node' field displays nodes that accesses 266 The 'Node' field displays nodes that accesses given cacheline
283 offset. Its output comes in 3 flavors: 267 offset. Its output comes in 3 flavors:
284 - node IDs separated by ',' 268 - node IDs separated by ','
285 - node IDs with stats for each ID, in follow 269 - node IDs with stats for each ID, in following format:
286 Node{cpus %hitms %stores} (Display with 270 Node{cpus %hitms %stores} (Display with HITM types)
287 Node{cpus %peers %stores} (Display with 271 Node{cpus %peers %stores} (Display with peer type)
288 - node IDs with list of affected CPUs in fol 272 - node IDs with list of affected CPUs in following format:
289 Node{cpu list} 273 Node{cpu list}
290 274
291 User can switch between above flavors with -N 275 User can switch between above flavors with -N option or
292 use 'n' key to interactively switch in TUI mod 276 use 'n' key to interactively switch in TUI mode.
293 277
294 COALESCE 278 COALESCE
295 -------- 279 --------
296 User can specify how to sort offsets for cache 280 User can specify how to sort offsets for cacheline.
297 281
298 Following fields are available and governs the 282 Following fields are available and governs the final
299 output fields set for cacheline offsets output 283 output fields set for cacheline offsets output:
300 284
301 tid - coalesced by process TIDs 285 tid - coalesced by process TIDs
302 pid - coalesced by process PIDs 286 pid - coalesced by process PIDs
303 iaddr - coalesced by code address, following 287 iaddr - coalesced by code address, following fields are displayed:
304 Code address, Code symbol, Shared 288 Code address, Code symbol, Shared Object, Source line
305 dso - coalesced by shared object 289 dso - coalesced by shared object
306 290
307 By default the coalescing is setup with 'pid,i 291 By default the coalescing is setup with 'pid,iaddr'.
308 292
309 STDIO OUTPUT 293 STDIO OUTPUT
310 ------------ 294 ------------
311 The stdio output displays data on standard out 295 The stdio output displays data on standard output.
312 296
313 Following tables are displayed: 297 Following tables are displayed:
314 Trace Event Information 298 Trace Event Information
315 - overall statistics of memory accesses 299 - overall statistics of memory accesses
316 300
317 Global Shared Cache Line Event Information 301 Global Shared Cache Line Event Information
318 - overall statistics on shared cachelines 302 - overall statistics on shared cachelines
319 303
320 Shared Data Cache Line Table 304 Shared Data Cache Line Table
321 - list of most expensive cachelines 305 - list of most expensive cachelines
322 306
323 Shared Cache Line Distribution Pareto 307 Shared Cache Line Distribution Pareto
324 - list of all accessed offsets for each cach 308 - list of all accessed offsets for each cacheline
325 309
326 TUI OUTPUT 310 TUI OUTPUT
327 ---------- 311 ----------
328 The TUI output provides interactive interface 312 The TUI output provides interactive interface to navigate
329 through cachelines list and to display offset 313 through cachelines list and to display offset details.
330 314
331 For details please refer to the help window by 315 For details please refer to the help window by pressing '?' key.
332 316
333 CREDITS 317 CREDITS
334 ------- 318 -------
335 Although Don Zickus, Dick Fowles and Joe Mario 319 Although Don Zickus, Dick Fowles and Joe Mario worked together
336 to get this implemented, we got lots of early 320 to get this implemented, we got lots of early help from Arnaldo
337 Carvalho de Melo, Stephane Eranian, Jiri Olsa 321 Carvalho de Melo, Stephane Eranian, Jiri Olsa and Andi Kleen.
338 322
339 C2C BLOG 323 C2C BLOG
340 -------- 324 --------
341 Check Joe's blog on c2c tool for detailed use 325 Check Joe's blog on c2c tool for detailed use case explanation:
342 https://joemario.github.io/blog/2016/09/01/c 326 https://joemario.github.io/blog/2016/09/01/c2c-blog/
343 327
344 SEE ALSO 328 SEE ALSO
345 -------- 329 --------
346 linkperf:perf-record[1], linkperf:perf-mem[1], !! 330 linkperf:perf-record[1], linkperf:perf-mem[1]
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.