1 Using TopDown metrics !! 1 Using TopDown metrics in user space
2 --------------------- !! 2 -----------------------------------
3 3
4 TopDown metrics break apart performance bottle !! 4 Intel CPUs (since Sandy Bridge and Silvermont) support a TopDown
5 1 it is typical to get metrics on retiring, ba !! 5 methology to break down CPU pipeline execution into 4 bottlenecks:
6 bound, and backend bound. Higher levels provid !! 6 frontend bound, backend bound, bad speculation, retiring.
7 level 1 bottlenecks, such as at level 2: core <<
8 heavy operations, light operations, branch mis <<
9 clears, fetch latency and fetch bandwidth. For <<
10 7
11 perf stat --topdown implements this using avai !! 8 For more details on Topdown see [1][5]
12 per architecture. <<
13 9
14 % perf stat -a --topdown -I1000 !! 10 Traditionally this was implemented by events in generic counters
15 # time % tma_retiring % tma_b !! 11 and specific formulas to compute the bottlenecks.
16 1.001141351 11.5 !! 12
17 2.006141972 13.4 !! 13 perf stat --topdown implements this.
18 3.010162040 12.9 !! 14
19 4.014009311 12.5 !! 15 Full Top Down includes more levels that can break down the
20 5.017838554 11.8 !! 16 bottlenecks further. This is not directly implemented in perf,
21 5.704818971 14.0 !! 17 but available in other tools that can run on top of perf,
22 ... !! 18 such as toplev[2] or vtune[3]
23 19
24 New Topdown features in Intel Ice Lake !! 20 New Topdown features in Ice Lake
25 ====================================== !! 21 ===============================
26 22
27 With Ice Lake CPUs the TopDown metrics are dir 23 With Ice Lake CPUs the TopDown metrics are directly available as
28 fixed counters and do not require generic coun 24 fixed counters and do not require generic counters. This allows
29 to collect TopDown always in addition to other 25 to collect TopDown always in addition to other events.
30 26
31 Using TopDown through RDPMC in applications on !! 27 % perf stat -a --topdown -I1000
32 ============================================== !! 28 # time retiring bad speculation frontend bound backend bound
>> 29 1.001281330 23.0% 15.3% 29.6% 32.1%
>> 30 2.003009005 5.0% 6.8% 46.6% 41.6%
>> 31 3.004646182 6.7% 6.7% 46.0% 40.6%
>> 32 4.006326375 5.0% 6.4% 47.6% 41.0%
>> 33 5.007991804 5.1% 6.3% 46.3% 42.3%
>> 34 6.009626773 6.2% 7.1% 47.3% 39.3%
>> 35 7.011296356 4.7% 6.7% 46.2% 42.4%
>> 36 8.012951831 4.7% 6.7% 47.5% 41.1%
>> 37 ...
>> 38
>> 39 This also enables measuring TopDown per thread/process instead
>> 40 of only per core.
>> 41
>> 42 Using TopDown through RDPMC in applications on Ice Lake
>> 43 ======================================================
33 44
34 For more fine grained measurements it can be u 45 For more fine grained measurements it can be useful to
35 access the new directly from user space. This 46 access the new directly from user space. This is more complicated,
36 but drastically lowers overhead. 47 but drastically lowers overhead.
37 48
38 On Ice Lake, there is a new fixed counter 3: S 49 On Ice Lake, there is a new fixed counter 3: SLOTS, which reports
39 "pipeline SLOTS" (cycles multiplied by core is 50 "pipeline SLOTS" (cycles multiplied by core issue width) and a
40 metric register that reports slots ratios for 51 metric register that reports slots ratios for the different bottleneck
41 categories. 52 categories.
42 53
43 The metrics counter is CPU model specific and 54 The metrics counter is CPU model specific and is not available on older
44 CPUs. 55 CPUs.
45 56
46 Example code 57 Example code
47 ============ 58 ============
48 59
49 Library functions to do the functionality desc 60 Library functions to do the functionality described below
50 is also available in libjevents [4] 61 is also available in libjevents [4]
51 62
52 The application opens a group with fixed count 63 The application opens a group with fixed counter 3 (SLOTS) and any
53 metric event, and allow user programs to read 64 metric event, and allow user programs to read the performance counters.
54 65
55 Fixed counter 3 is mapped to a pseudo event ev 66 Fixed counter 3 is mapped to a pseudo event event=0x00, umask=04,
56 so the perf_event_attr structure should be ini 67 so the perf_event_attr structure should be initialized with
57 { .config = 0x0400, .type = PERF_TYPE_RAW } 68 { .config = 0x0400, .type = PERF_TYPE_RAW }
58 The metric events are mapped to the pseudo eve 69 The metric events are mapped to the pseudo event event=0x00, umask=0x8X.
59 For example, the perf_event_attr structure can 70 For example, the perf_event_attr structure can be initialized with
60 { .config = 0x8000, .type = PERF_TYPE_RAW } fo 71 { .config = 0x8000, .type = PERF_TYPE_RAW } for Retiring metric event
61 The Fixed counter 3 must be the leader of the 72 The Fixed counter 3 must be the leader of the group.
62 73
63 #include <linux/perf_event.h> 74 #include <linux/perf_event.h>
64 #include <sys/mman.h> <<
65 #include <sys/syscall.h> 75 #include <sys/syscall.h>
66 #include <unistd.h> 76 #include <unistd.h>
67 77
68 /* Provide own perf_event_open stub because gl 78 /* Provide own perf_event_open stub because glibc doesn't */
69 __attribute__((weak)) 79 __attribute__((weak))
70 int perf_event_open(struct perf_event_attr *at 80 int perf_event_open(struct perf_event_attr *attr, pid_t pid,
71 int cpu, int group_fd, uns 81 int cpu, int group_fd, unsigned long flags)
72 { 82 {
73 return syscall(__NR_perf_event_open, a 83 return syscall(__NR_perf_event_open, attr, pid, cpu, group_fd, flags);
74 } 84 }
75 85
76 /* Open slots counter file descriptor for curr 86 /* Open slots counter file descriptor for current task. */
77 struct perf_event_attr slots = { 87 struct perf_event_attr slots = {
78 .type = PERF_TYPE_RAW, 88 .type = PERF_TYPE_RAW,
79 .size = sizeof(struct perf_event_attr) 89 .size = sizeof(struct perf_event_attr),
80 .config = 0x400, 90 .config = 0x400,
81 .exclude_kernel = 1, 91 .exclude_kernel = 1,
82 }; 92 };
83 93
84 int slots_fd = perf_event_open(&slots, 0, -1, 94 int slots_fd = perf_event_open(&slots, 0, -1, -1, 0);
85 if (slots_fd < 0) 95 if (slots_fd < 0)
86 ... error ... 96 ... error ...
87 97
88 /* Memory mapping the fd permits _rdpmc calls <<
89 void *slots_p = mmap(0, getpagesize(), PROT_RE <<
90 if (!slot_p) <<
91 .... error ... <<
92 <<
93 /* 98 /*
94 * Open metrics event file descriptor for curr 99 * Open metrics event file descriptor for current task.
95 * Set slots event as the leader of the group. 100 * Set slots event as the leader of the group.
96 */ 101 */
97 struct perf_event_attr metrics = { 102 struct perf_event_attr metrics = {
98 .type = PERF_TYPE_RAW, 103 .type = PERF_TYPE_RAW,
99 .size = sizeof(struct perf_event_attr) 104 .size = sizeof(struct perf_event_attr),
100 .config = 0x8000, 105 .config = 0x8000,
101 .exclude_kernel = 1, 106 .exclude_kernel = 1,
102 }; 107 };
103 108
104 int metrics_fd = perf_event_open(&metrics, 0, 109 int metrics_fd = perf_event_open(&metrics, 0, -1, slots_fd, 0);
105 if (metrics_fd < 0) 110 if (metrics_fd < 0)
106 ... error ... 111 ... error ...
107 112
108 /* Memory mapping the fd permits _rdpmc calls <<
109 void *metrics_p = mmap(0, getpagesize(), PROT_ <<
110 if (!metrics_p) <<
111 ... error ... <<
112 <<
113 Note: the file descriptors returned by the per <<
114 mapped to permit calls to the _rdpmd instructi <<
115 by writing the /sys/devices/cpu/rdpmc sysfs no <<
116 113
117 The RDPMC instruction (or _rdpmc compiler intr 114 The RDPMC instruction (or _rdpmc compiler intrinsic) can now be used
118 to read slots and the topdown metrics at diffe 115 to read slots and the topdown metrics at different points of the program:
119 116
120 #include <stdint.h> 117 #include <stdint.h>
121 #include <x86intrin.h> 118 #include <x86intrin.h>
122 119
123 #define RDPMC_FIXED (1 << 30) /* ret 120 #define RDPMC_FIXED (1 << 30) /* return fixed counters */
124 #define RDPMC_METRIC (1 << 29) /* ret 121 #define RDPMC_METRIC (1 << 29) /* return metric counters */
125 122
126 #define FIXED_COUNTER_SLOTS 3 123 #define FIXED_COUNTER_SLOTS 3
127 #define METRIC_COUNTER_TOPDOWN_L1_L2 0 124 #define METRIC_COUNTER_TOPDOWN_L1_L2 0
128 125
129 static inline uint64_t read_slots(void) 126 static inline uint64_t read_slots(void)
130 { 127 {
131 return _rdpmc(RDPMC_FIXED | FIXED_COUN 128 return _rdpmc(RDPMC_FIXED | FIXED_COUNTER_SLOTS);
132 } 129 }
133 130
134 static inline uint64_t read_metrics(void) 131 static inline uint64_t read_metrics(void)
135 { 132 {
136 return _rdpmc(RDPMC_METRIC | METRIC_CO 133 return _rdpmc(RDPMC_METRIC | METRIC_COUNTER_TOPDOWN_L1_L2);
137 } 134 }
138 135
139 Then the program can be instrumented to read t 136 Then the program can be instrumented to read these metrics at different
140 points. 137 points.
141 138
142 It's not a good idea to do this with too short 139 It's not a good idea to do this with too short code regions,
143 as the parallelism and overlap in the CPU prog 140 as the parallelism and overlap in the CPU program execution will
144 cause too much measurement inaccuracy. For exa 141 cause too much measurement inaccuracy. For example instrumenting
145 individual basic blocks is definitely too fine 142 individual basic blocks is definitely too fine grained.
146 143
147 _rdpmc calls should not be mixed with reading <<
148 through system calls, as the kernel will reset <<
149 call. <<
150 <<
151 Decoding metrics values 144 Decoding metrics values
152 ======================= 145 =======================
153 146
154 The value reported by read_metrics() contains 147 The value reported by read_metrics() contains four 8 bit fields
155 that represent a scaled ratio that represent t 148 that represent a scaled ratio that represent the Level 1 bottleneck.
156 All four fields add up to 0xff (= 100%) 149 All four fields add up to 0xff (= 100%)
157 150
158 The binary ratios in the metric value can be c 151 The binary ratios in the metric value can be converted to float ratios:
159 152
160 #define GET_METRIC(m, i) (((m) >> (i*8)) & 0xf 153 #define GET_METRIC(m, i) (((m) >> (i*8)) & 0xff)
161 154
162 /* L1 Topdown metric events */ 155 /* L1 Topdown metric events */
163 #define TOPDOWN_RETIRING(val) ((float)GET_ME 156 #define TOPDOWN_RETIRING(val) ((float)GET_METRIC(val, 0) / 0xff)
164 #define TOPDOWN_BAD_SPEC(val) ((float)GET_ME 157 #define TOPDOWN_BAD_SPEC(val) ((float)GET_METRIC(val, 1) / 0xff)
165 #define TOPDOWN_FE_BOUND(val) ((float)GET_ME 158 #define TOPDOWN_FE_BOUND(val) ((float)GET_METRIC(val, 2) / 0xff)
166 #define TOPDOWN_BE_BOUND(val) ((float)GET_ME 159 #define TOPDOWN_BE_BOUND(val) ((float)GET_METRIC(val, 3) / 0xff)
167 160
168 /* 161 /*
169 * L2 Topdown metric events. 162 * L2 Topdown metric events.
170 * Available on Sapphire Rapids and later plat 163 * Available on Sapphire Rapids and later platforms.
171 */ 164 */
172 #define TOPDOWN_HEAVY_OPS(val) ((floa 165 #define TOPDOWN_HEAVY_OPS(val) ((float)GET_METRIC(val, 4) / 0xff)
173 #define TOPDOWN_BR_MISPREDICT(val) ((floa 166 #define TOPDOWN_BR_MISPREDICT(val) ((float)GET_METRIC(val, 5) / 0xff)
174 #define TOPDOWN_FETCH_LAT(val) ((floa 167 #define TOPDOWN_FETCH_LAT(val) ((float)GET_METRIC(val, 6) / 0xff)
175 #define TOPDOWN_MEM_BOUND(val) ((floa 168 #define TOPDOWN_MEM_BOUND(val) ((float)GET_METRIC(val, 7) / 0xff)
176 169
177 and then converted to percent for printing. 170 and then converted to percent for printing.
178 171
179 The ratios in the metric accumulate for the ti 172 The ratios in the metric accumulate for the time when the counter
180 is enabled. For measuring programs it is often 173 is enabled. For measuring programs it is often useful to measure
181 specific sections. For this it is needed to de 174 specific sections. For this it is needed to deltas on metrics.
182 175
183 This can be done by scaling the metrics with t 176 This can be done by scaling the metrics with the slots counter
184 read at the same time. 177 read at the same time.
185 178
186 Then it's possible to take deltas of these slo 179 Then it's possible to take deltas of these slots counts
187 measured at different points, and determine th 180 measured at different points, and determine the metrics
188 for that time period. 181 for that time period.
189 182
190 slots_a = read_slots(); 183 slots_a = read_slots();
191 metric_a = read_metrics(); 184 metric_a = read_metrics();
192 185
193 ... larger code region ... 186 ... larger code region ...
194 187
195 slots_b = read_slots() 188 slots_b = read_slots()
196 metric_b = read_metrics() 189 metric_b = read_metrics()
197 190
198 # compute scaled metrics for measureme 191 # compute scaled metrics for measurement a
199 retiring_slots_a = GET_METRIC(metric_a 192 retiring_slots_a = GET_METRIC(metric_a, 0) * slots_a
200 bad_spec_slots_a = GET_METRIC(metric_a 193 bad_spec_slots_a = GET_METRIC(metric_a, 1) * slots_a
201 fe_bound_slots_a = GET_METRIC(metric_a 194 fe_bound_slots_a = GET_METRIC(metric_a, 2) * slots_a
202 be_bound_slots_a = GET_METRIC(metric_a 195 be_bound_slots_a = GET_METRIC(metric_a, 3) * slots_a
203 196
204 # compute delta scaled metrics between 197 # compute delta scaled metrics between b and a
205 retiring_slots = GET_METRIC(metric_b, 198 retiring_slots = GET_METRIC(metric_b, 0) * slots_b - retiring_slots_a
206 bad_spec_slots = GET_METRIC(metric_b, 199 bad_spec_slots = GET_METRIC(metric_b, 1) * slots_b - bad_spec_slots_a
207 fe_bound_slots = GET_METRIC(metric_b, 200 fe_bound_slots = GET_METRIC(metric_b, 2) * slots_b - fe_bound_slots_a
208 be_bound_slots = GET_METRIC(metric_b, 201 be_bound_slots = GET_METRIC(metric_b, 3) * slots_b - be_bound_slots_a
209 202
210 Later the individual ratios of L1 metric event 203 Later the individual ratios of L1 metric events for the measurement period can
211 be recreated from these counts. 204 be recreated from these counts.
212 205
213 slots_delta = slots_b - slots_a 206 slots_delta = slots_b - slots_a
214 retiring_ratio = (float)retiring_slots 207 retiring_ratio = (float)retiring_slots / slots_delta
215 bad_spec_ratio = (float)bad_spec_slots 208 bad_spec_ratio = (float)bad_spec_slots / slots_delta
216 fe_bound_ratio = (float)fe_bound_slots 209 fe_bound_ratio = (float)fe_bound_slots / slots_delta
217 be_bound_ratio = (float)be_bound_slots 210 be_bound_ratio = (float)be_bound_slots / slota_delta
218 211
219 printf("Retiring %.2f%% Bad Speculatio 212 printf("Retiring %.2f%% Bad Speculation %.2f%% FE Bound %.2f%% BE Bound %.2f%%\n",
220 retiring_ratio * 100., 213 retiring_ratio * 100.,
221 bad_spec_ratio * 100., 214 bad_spec_ratio * 100.,
222 fe_bound_ratio * 100., 215 fe_bound_ratio * 100.,
223 be_bound_ratio * 100.); 216 be_bound_ratio * 100.);
224 217
225 The individual ratios of L2 metric events for 218 The individual ratios of L2 metric events for the measurement period can be
226 recreated from L1 and L2 metric counters. (Ava 219 recreated from L1 and L2 metric counters. (Available on Sapphire Rapids and
227 later platforms) 220 later platforms)
228 221
229 # compute scaled metrics for measureme 222 # compute scaled metrics for measurement a
230 heavy_ops_slots_a = GET_METRIC(metric_ 223 heavy_ops_slots_a = GET_METRIC(metric_a, 4) * slots_a
231 br_mispredict_slots_a = GET_METRIC(met 224 br_mispredict_slots_a = GET_METRIC(metric_a, 5) * slots_a
232 fetch_lat_slots_a = GET_METRIC(metric_ 225 fetch_lat_slots_a = GET_METRIC(metric_a, 6) * slots_a
233 mem_bound_slots_a = GET_METRIC(metric_ 226 mem_bound_slots_a = GET_METRIC(metric_a, 7) * slots_a
234 227
235 # compute delta scaled metrics between 228 # compute delta scaled metrics between b and a
236 heavy_ops_slots = GET_METRIC(metric_b, 229 heavy_ops_slots = GET_METRIC(metric_b, 4) * slots_b - heavy_ops_slots_a
237 br_mispredict_slots = GET_METRIC(metri 230 br_mispredict_slots = GET_METRIC(metric_b, 5) * slots_b - br_mispredict_slots_a
238 fetch_lat_slots = GET_METRIC(metric_b, 231 fetch_lat_slots = GET_METRIC(metric_b, 6) * slots_b - fetch_lat_slots_a
239 mem_bound_slots = GET_METRIC(metric_b, 232 mem_bound_slots = GET_METRIC(metric_b, 7) * slots_b - mem_bound_slots_a
240 233
241 slots_delta = slots_b - slots_a 234 slots_delta = slots_b - slots_a
242 heavy_ops_ratio = (float)heavy_ops_slo 235 heavy_ops_ratio = (float)heavy_ops_slots / slots_delta
243 light_ops_ratio = retiring_ratio - hea 236 light_ops_ratio = retiring_ratio - heavy_ops_ratio;
244 237
245 br_mispredict_ratio = (float)br_mispre 238 br_mispredict_ratio = (float)br_mispredict_slots / slots_delta
246 machine_clears_ratio = bad_spec_ratio 239 machine_clears_ratio = bad_spec_ratio - br_mispredict_ratio;
247 240
248 fetch_lat_ratio = (float)fetch_lat_slo 241 fetch_lat_ratio = (float)fetch_lat_slots / slots_delta
249 fetch_bw_ratio = fe_bound_ratio - fetc 242 fetch_bw_ratio = fe_bound_ratio - fetch_lat_ratio;
250 243
251 mem_bound_ratio = (float)mem_bound_slo 244 mem_bound_ratio = (float)mem_bound_slots / slota_delta
252 core_bound_ratio = be_bound_ratio - me 245 core_bound_ratio = be_bound_ratio - mem_bound_ratio;
253 246
254 printf("Heavy Operations %.2f%% Light 247 printf("Heavy Operations %.2f%% Light Operations %.2f%% "
255 "Branch Mispredict %.2f%% Machi 248 "Branch Mispredict %.2f%% Machine Clears %.2f%% "
256 "Fetch Latency %.2f%% Fetch Ban 249 "Fetch Latency %.2f%% Fetch Bandwidth %.2f%% "
257 "Mem Bound %.2f%% Core Bound %. 250 "Mem Bound %.2f%% Core Bound %.2f%%\n",
258 heavy_ops_ratio * 100., 251 heavy_ops_ratio * 100.,
259 light_ops_ratio * 100., 252 light_ops_ratio * 100.,
260 br_mispredict_ratio * 100., 253 br_mispredict_ratio * 100.,
261 machine_clears_ratio * 100., 254 machine_clears_ratio * 100.,
262 fetch_lat_ratio * 100., 255 fetch_lat_ratio * 100.,
263 fetch_bw_ratio * 100., 256 fetch_bw_ratio * 100.,
264 mem_bound_ratio * 100., 257 mem_bound_ratio * 100.,
265 core_bound_ratio * 100.); 258 core_bound_ratio * 100.);
266 259
267 Resetting metrics counters 260 Resetting metrics counters
268 ========================== 261 ==========================
269 262
270 Since the individual metrics are only 8bit the 263 Since the individual metrics are only 8bit they lose precision for
271 short regions over time because the number of 264 short regions over time because the number of cycles covered by each
272 fraction bit shrinks. So the counters need to 265 fraction bit shrinks. So the counters need to be reset regularly.
273 266
274 When using the kernel perf API the kernel rese 267 When using the kernel perf API the kernel resets on every read.
275 So as long as the reading is at reasonable int 268 So as long as the reading is at reasonable intervals (every few
276 seconds) the precision is good. 269 seconds) the precision is good.
277 270
278 When using perf stat it is recommended to alwa 271 When using perf stat it is recommended to always use the -I option,
279 with no longer interval than a few seconds 272 with no longer interval than a few seconds
280 273
281 perf stat -I 1000 --topdown ... 274 perf stat -I 1000 --topdown ...
282 275
283 For user programs using RDPMC directly the cou 276 For user programs using RDPMC directly the counter can
284 be reset explicitly using ioctl: 277 be reset explicitly using ioctl:
285 278
286 ioctl(perf_fd, PERF_EVENT_IOC_RESET, 0 279 ioctl(perf_fd, PERF_EVENT_IOC_RESET, 0);
287 280
288 This "opens" a new measurement period. 281 This "opens" a new measurement period.
289 282
290 A program using RDPMC for TopDown should sched 283 A program using RDPMC for TopDown should schedule such a reset
291 regularly, as in every few seconds. 284 regularly, as in every few seconds.
292 285
293 Limits on Intel Ice Lake !! 286 Limits on Ice Lake
294 ======================== !! 287 ==================
295 288
296 Four pseudo TopDown metric events are exposed 289 Four pseudo TopDown metric events are exposed for the end-users,
297 topdown-retiring, topdown-bad-spec, topdown-fe 290 topdown-retiring, topdown-bad-spec, topdown-fe-bound and topdown-be-bound.
298 They can be used to collect the TopDown value 291 They can be used to collect the TopDown value under the following
299 rules: 292 rules:
300 - All the TopDown metric events must be in a g 293 - All the TopDown metric events must be in a group with the SLOTS event.
301 - The SLOTS event must be the leader of the gr 294 - The SLOTS event must be the leader of the group.
302 - The PERF_FORMAT_GROUP flag must be applied f 295 - The PERF_FORMAT_GROUP flag must be applied for each TopDown metric
303 events 296 events
304 297
305 The SLOTS event and the TopDown metric events 298 The SLOTS event and the TopDown metric events can be counting members of
306 a sampling read group. Since the SLOTS event m 299 a sampling read group. Since the SLOTS event must be the leader of a TopDown
307 group, the second event of the group is the sa 300 group, the second event of the group is the sampling event.
308 For example, perf record -e '{slots, $sampling 301 For example, perf record -e '{slots, $sampling_event, topdown-retiring}:S'
309 302
310 Extension on Intel Sapphire Rapids Server !! 303 Extension on Sapphire Rapids Server
311 ========================================= !! 304 ===================================
312 The metrics counter is extended to support TMA 305 The metrics counter is extended to support TMA method level 2 metrics.
313 The lower half of the register is the TMA leve 306 The lower half of the register is the TMA level 1 metrics (legacy).
314 The upper half is also divided into four 8-bit 307 The upper half is also divided into four 8-bit fields for the new level 2
315 metrics. Four more TopDown metric events are e 308 metrics. Four more TopDown metric events are exposed for the end-users,
316 topdown-heavy-ops, topdown-br-mispredict, topd 309 topdown-heavy-ops, topdown-br-mispredict, topdown-fetch-lat and
317 topdown-mem-bound. 310 topdown-mem-bound.
318 311
319 Each of the new level 2 metrics in the upper h 312 Each of the new level 2 metrics in the upper half is a subset of the
320 corresponding level 1 metric in the lower half 313 corresponding level 1 metric in the lower half. Software can deduce the
321 other four level 2 metrics by subtracting corr 314 other four level 2 metrics by subtracting corresponding metrics as below.
322 315
323 Light_Operations = Retiring - Heavy_Operat 316 Light_Operations = Retiring - Heavy_Operations
324 Machine_Clears = Bad_Speculation - Branch_ 317 Machine_Clears = Bad_Speculation - Branch_Mispredicts
325 Fetch_Bandwidth = Frontend_Bound - Fetch_L 318 Fetch_Bandwidth = Frontend_Bound - Fetch_Latency
326 Core_Bound = Backend_Bound - Memory_Bound 319 Core_Bound = Backend_Bound - Memory_Bound
327 320
328 TPEBS in TopDown <<
329 ================ <<
330 <<
331 TPEBS (Timed PEBS) is one of the new Intel PMU <<
332 Rapids microarchitecture. The TPEBS feature ad <<
333 in the Basic Info group of the PEBS record. It <<
334 retirement of the previous instruction to the <<
335 Please refer to Section 8.4.1 of "Intel® Arch <<
336 Programming Reference" for more details about <<
337 extends PEBS record, sampling with weight opti <<
338 retire_latency value. <<
339 <<
340 perf record -e event_name -W ... <<
341 <<
342 In the most recent release of TMA, the metrics <<
343 values in some of the metrics’ formulas on p <<
344 For previous generations that do not support T <<
345 predefined per processor family by the hardwar <<
346 of workloads in execution environments, retire <<
347 time are more accurate. Therefore, new TMA met <<
348 more accurate performance analysis results. <<
349 <<
350 To support TPEBS in TMA metrics, a new modifie <<
351 capture retire_latency value of required event <<
352 with perf record. The retire_latency value wou <<
353 Currently, this feature is supported through p <<
354 <<
355 perf stat -M metric_name --record-tpeb <<
356 <<
357 <<
358 321
359 [1] https://software.intel.com/en-us/top-down- 322 [1] https://software.intel.com/en-us/top-down-microarchitecture-analysis-method-win
360 [2] https://sites.google.com/site/analysismeth !! 323 [2] https://github.com/andikleen/pmu-tools/wiki/toplev-manual
361 [3] https://perf.wiki.kernel.org/index.php/Top !! 324 [3] https://software.intel.com/en-us/intel-vtune-amplifier-xe
362 [4] https://github.com/andikleen/pmu-tools/tre 325 [4] https://github.com/andikleen/pmu-tools/tree/master/jevents
>> 326 [5] https://sites.google.com/site/analysismethods/yasin-pubs
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.