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