| /Documentation/trace/coresight/ |
| D | coresight-perf.rst | 4 CoreSight - Perf
10 Perf is able to locally access CoreSight trace data and store it to the
11 output perf data files. This data can then be later decoded to give the
13 can log such data with a perf record command like::
15 perf record -e cs_etm//u testbinary
18 a perf.data trace file. That file would have AUX sections if CoreSight
22 perf report --stdio --dump -i perf.data
43 To compile perf with CoreSight support in the tools/perf directory do::
53 For complete information on building perf with CoreSight support and
81 Perf test - Verify kernel and userspace perf CoreSight work
[all …]
|
| D | coresight-config.rst | 15 can then be easily enabled from sysfs or perf.
115 2) The configuration is enabled. For example 'perf' may select the
118 perf record -e cs_etm/autofdo/ myapp
122 3) perf starts tracing on the system. As each ETMv4 that perf uses for
130 5) At the end of the perf session, the configuration will be disabled.
209 Using Configurations in perf
213 also declared in the perf 'cs_etm' event infrastructure so that they can
214 be selected when running trace under perf::
220 The key directory here is 'events' - a generic perf directory which allows
221 selection on the perf command line. As with the sinks entries, this provides
[all …]
|
| D | coresight.rst | 335 1. using the perf cmd line tools.
450 Using perf framework
453 Coresight tracers are represented using the Perf framework's Performance
454 Monitoring Unit (PMU) abstraction. As such the perf framework takes charge of
457 queried by the perf command line tool:
459 linaro@linaro-nano:~$ ./perf list pmu
486 root@linaro-nano:~# perf record -e cs_etm/@tmc_etr0/u --per-thread program
497 the perf tools can be found in the "HOWTO.md" file of the openCSD gitHub
500 Advanced perf framework usage
503 AutoFDO analysis using the perf tools
[all …]
|
| /Documentation/admin-guide/perf/ |
| D | hns3-pmu.rst | 17 The HNS3 PMU driver registers a perf PMU with the name of its sicl id.::
25 shown in perf list.
66 Example usage of perf::
68 $# perf list
73 …$# perf stat -g -e hns3_pmu_sicl_0/bw_ssu_rpu_byte_num,global=1/ -e hns3_pmu_sicl_0/bw_ssu_rpu_tim…
75 …$# perf stat -g -e hns3_pmu_sicl_0/config=0x00002,global=1/ -e hns3_pmu_sicl_0/config=0x10002,glob…
84 Example usage of perf::
86 $# perf stat -a -e hns3_pmu_sicl_0/config=0x1020F,global=1/ -I 1000
93 Example usage of perf::
95 $# perf stat -a -e hns3_pmu_sicl_0/config=0x1020F,port=0,tc=0xF/ -I 1000
[all …]
|
| D | xgene-pmu.rst | 11 PMU (perf) driver
14 The xgene-pmu driver registers several perf PMU drivers. Each of the perf
21 can be used with perf tool. For example, "l3c0/bank-fifo-full/" is an
34 Each perf driver also provides a "cpumask" sysfs attribute, which contains a
37 Example for perf tool use::
39 / # perf list | grep -e l3c -e iob -e mcb -e mc
44 / # perf stat -a -e l3c0/read-miss/,mcb1/csw-write-request/ sleep 1
46 / # perf stat -a -e l3c0/read-miss,config1=0xfffffffffffffffe/ sleep 1
48 The driver does not support sampling, therefore "perf record" will
49 not work. Per-task (without "-a") perf sessions are not supported.
|
| D | hisi-pcie-pmu.rst | 15 The PCIe PMU driver registers a perf PMU with the name of its sicl-id and PCIe
25 describes all documented events shown in perf list.
35 Example usage of perf::
37 $# perf list
42 $# perf stat -e hisi_pcie0_core0/rx_mwr_latency,port=0xffff/
43 $# perf stat -e hisi_pcie0_core0/rx_mwr_cnt,port=0xffff/
54 Example usage of perf group::
56 …$# perf stat -e "{hisi_pcie0_core0/rx_mwr_latency,port=0xffff/,hisi_pcie0_core0/rx_mwr_cnt,port=0x…
58 The current driver does not support sampling. So "perf record" is unsupported.
87 Example usage of perf::
[all …]
|
| D | arm-ccn.rst | 10 PMU (perf) driver
13 The CCN driver registers a perf PMU driver, which provides
20 events, that can be used with perf tool. For example "xp_valid_flit"
49 Example of perf tool use::
51 / # perf list | grep ccn
57 / # perf stat -a -e ccn/cycles/,ccn/xp_valid_flit,xp=1,port=0,vc=1,dir=1/ \
60 The driver does not support sampling, therefore "perf record" will
61 not work. Per-task (without "-a") perf sessions are not supported.
|
| D | thunderx2-pmu.rst | 19 PMU UNCORE (perf) driver:
21 The thunderx2_pmu driver registers per-socket perf PMUs for the DMC and
27 The driver does not support sampling, therefore "perf record" will not
28 work. Per-task perf sessions are also not supported.
32 # perf stat -a -e uncore_dmc_0/cnt_cycles/ sleep 1
34 # perf stat -a -e \
40 # perf stat -a -e \
|
| D | cxl.rst | 19 The CPMU driver registers a perf PMU with the name pmu_mem<X>.<Y> on the CXL bus
37 describes all documented events show in perf list.
39 The events shown in perf list are the most fine grained events with a single
52 $#perf list
60 $# perf stat -a -e cxl_pmu_mem0.0/clock_ticks/ -e cxl_pmu_mem0.0/d2h_req_rdshared/
64 $# perf stat -a -e cxl_pmu_mem0.0/vid=VID,gid=GID,mask=MASK/
66 The driver does not support sampling so "perf record" is unsupported.
|
| D | hisi-pmu.rst | 19 interrupt, and the PMU driver shall register perf PMU drivers like L3C,
24 The "perf list" command shall list the available events from sysfs.
26 Each L3C, HHA and DDRC is registered as a separate PMU with perf. The PMU
40 Example usage of perf::
42 $# perf list
52 $# perf stat -a -e hisi_sccl3_l3c0/rd_hit_cpipe/ sleep 5
53 $# perf stat -a -e hisi_sccl3_l3c0/config=0x02/ sleep 5
61 $# perf stat -a -e hisi_sccl3_l3c0/config=0x02,tt_core=0x3/ sleep 5
66 operations via the tt_req parameeter in perf. The default value counts all
71 $# perf stat -a -e hisi_sccl3_l3c0/config=0x02,tt_req=0x4/ sleep 5
[all …]
|
| D | imx-ddr.rst | 8 control register which causes a large number of PERF signals to be generated.
19 hardware supported that can be used with perf tool, see /sys/bus/event_source/
25 perf stat -a -e imx8_ddr0/cycles/ cmd
26 perf stat -a -e imx8_ddr0/read/,imx8_ddr0/write/ cmd
45 incremented. Perf counter is incremented if::
54 perf stat -a -e imx8_ddr0/axid-read,axi_mask=0xMMMM,axi_id=0xDDDD/ cmd
55 perf stat -a -e imx8_ddr0/axid-write,axi_mask=0xMMMM,axi_id=0xDDDD/ cmd
65 perf stat -a -e imx8_ddr0/axid-read,axi_id=0x12/ cmd, which will monitor ARID=0x12
92 perf stat -a -e imx8_ddr0/axid-read,axi_mask=0xMMMM,axi_id=0xDDDD,axi_channel=0xH/ cmd
93 perf stat -a -e imx8_ddr0/axid-write,axi_mask=0xMMMM,axi_id=0xDDDD,axi_channel=0xH/ cmd
|
| D | qcom_l2_pmu.rst | 32 Examples for use with perf::
34 perf stat -e l2cache_0/config=0x001/,l2cache_0/config=0x042/ -a sleep 1
36 perf stat -e l2cache_0/config=0xfe/ -C 2 sleep 1
38 The driver does not support sampling, therefore "perf record" will
39 not work. Per-task perf sessions are not supported.
|
| D | meson-ddr-pmu.rst | 12 Currently, this driver supports the following 5 perf events:
59 perf stat -a -e meson_ddr_bw/total_rw_bytes/ -I 1000 sleep 10
67 perf stat -a -e meson_ddr_bw/chan_1_rw_bytes,arm=1/ -I 1000 sleep 10
68 perf stat -a -e meson_ddr_bw/chan_2_rw_bytes,gpu=1/ -I 1000 sleep 10
69 perf stat -a -e meson_ddr_bw/chan_3_rw_bytes,arm=1,gpu=1/ -I 1000 sleep 10
|
| D | dwc_pcie_pmu.rst | 56 The DWC PCIe PMU driver registers a perf PMU driver, which provides
65 The "perf list" command shall list the available events from sysfs, e.g.::
67 $# perf list | grep dwc_rootport
78 $# perf stat -a -e dwc_rootport_13018/Rx_PCIe_TLP_Data_Payload/
91 $# perf stat -a -e dwc_rootport_13018/rx_memory_read,lane=4/
93 The driver does not support sampling, therefore "perf record" will not
94 work. Per-task (without "-a") perf sessions are not supported.
|
| D | starfive_starlink_pmu.rst | 28 Example of perf usage::
30 $ perf list
43 $ perf stat -a -e /starfive_starlink_pmu/cycles/ sleep 1
45 Sampling is not supported. As a result, "perf record" is not supported.
|
| D | qcom_l3_pmu.rst | 7 by all cores within a socket. Each slice is exposed as a separate uncore perf
23 perf stat -e l3cache_0_0/read-miss,lc/
26 "perf record" will not work. Per-task perf sessions are not supported.
|
| D | ampere_cspmu.rst | 19 Example for perf tool use::
21 / # perf list ampere
28 …/ # perf stat -a -e ampere_mcu_pmu_0/act_sent,bank=5,rank=3,threshold=2/,ampere_mcu_pmu_1/rd_sent/…
|
| /Documentation/admin-guide/ |
| D | perf-security.rst | 3 Perf events and tool security
13 generated by Perf tool user mode utility (Perf) [3]_ , [4]_ . The risk
15 units (PMU) [2]_ and Perf collect and expose for performance analysis.
21 its topology, used kernel and Perf versions, performance monitoring
22 setup including experiment time, events configuration, Perf command
102 Privileged Perf users groups
107 dedicated groups of privileged Perf users who are permitted to execute
109 steps can be taken to create such groups of privileged Perf users.
111 1. Create perf_users group of privileged Perf users, assign perf_users
112 group to Perf tool executable and limit access to the executable for
[all …]
|
| /Documentation/trace/ |
| D | tracepoint-analysis.rst | 21 assumed that the PCL tool tools/perf has been installed and is in your path.
40 are available with the perf tool. Getting a list of available events is a
43 $ perf list 2>&1 | grep Tracepoint
97 $ perf stat -a \
125 $ perf stat -e kmem:mm_page_alloc -e kmem:mm_page_free \
150 occurrences are useful to the performance analyst, then perf can be used.
153 $ perf stat --repeat 5 -e kmem:mm_page_alloc -e kmem:mm_page_free
176 $ perf stat -e kmem:mm_page_alloc -e kmem:mm_page_free \
228 $ perf record -c 1 \
233 [ perf record: Captured and wrote 0.733 MB perf.data (~32010 samples) ]
[all …]
|
| /Documentation/arch/powerpc/ |
| D | imc.rst | 62 # perf list
77 # ./perf stat -e "nest_mcs01/PM_MCS01_64B_WR_DISP_PORT01/" -a --per-socket
83 # ./perf stat -e "core_imc/CPM_NON_IDLE_INST/" -C 0 -I 1000
89 # ./perf stat -e "thread_imc/CPM_NON_IDLE_PCYC/" make
159 # perf list
167 # perf record -e trace_imc/trace_cycles/ yes > /dev/null
168 [ perf record: Woken up 1 times to write data ]
169 [ perf record: Captured and wrote 0.012 MB perf.data (21 samples) ]
171 The `perf.data` generated, can be read using perf report.
181 Performance data using `perf top` with and without trace-imc event.
[all …]
|
| /Documentation/dev-tools/ |
| D | autofdo.rst | 21 the "perf" tool. It is crucial that the workload used to create these
22 perf files is representative; they must exhibit runtime
40 collect the perf profile. While it's possible to use a kernel built
56 collected using hardware sampling, via perf. AutoFDO is most
60 #. AutoFDO profile generation: Perf output file is converted to
120 3) Run the load tests. The '-c' option in perf specifies the sample
126 $ perf record -e BR_INST_RETIRED.NEAR_TAKEN:k -a -N -b -c <count> -o <perf_file> -- <loadtest>
140 The following command generated the perf data file::
142 …$ perf record --pfm-events RETIRED_TAKEN_BRANCH_INSTRUCTIONS:k -a -N -b -c <count> -o <perf_file> …
150 $ perf record -e cs_etm/@tmc_etr0/k -a -o <etm_perf_file> -- <loadtest>
[all …]
|
| /Documentation/userspace-api/ |
| D | perf_ring_buffer.rst | 4 Perf ring buffer
34 The ring buffer is a fundamental mechanism for data transfer. perf uses
41 challenging work. On the one hand, the kernel and perf tool in the user
47 This documentation dives into the details for perf ring buffer with two
48 parts: firstly it explains the perf ring buffer implementation, then the
72 Perf uses the same way to manage its ring buffer. In the implementation
84 the perf tool has the permission to update the tail pointer after consuming
100 Figure 2. Perf ring buffer
102 When using the ``perf record`` tool, we can specify the ring buffer size
106 to VMA area until the perf tool accesses the buffer from the user space.
[all …]
|
| /Documentation/arch/arm64/ |
| D | perf.rst | 6 Perf title
9 Perf Event Attributes
95 Perf Userspace PMU Hardware Counter Access
100 The perf userspace tool relies on the PMU to monitor events. It offers an
122 It is necessary to open the event using the perf tool interface with config1:1
133 and perf_evsel__read() functions. See `tools/lib/perf/tests/test-evsel.c`_ for
143 Have a look at `tools/perf/arch/arm64/tests/user-events.c`_ for an example. It
144 can be run using the perf tool to check that the access to the registers works
149 perf test -v user
163 .. _tools/perf/arch/arm64/tests/user-events.c:
[all …]
|
| /Documentation/ABI/testing/ |
| D | sysfs-bus-event_source-devices-events | 19 using the 'perf(1)' tool.
26 "raw code" for the perf event identified by the file's
41 insensitive in the perf tool, the perf tool only looks
76 Description: Perf event units
88 Description: Perf event scaling factors
|
| /Documentation/translations/zh_CN/mm/damon/ |
| D | faq.rst | 24 “闲置页面跟踪” 或 “perf mem” 可以替代DAMON吗？
27 闲置页跟踪是物理地址空间访问检查的一个低层次的原始方法。“perf mem”也是类似的，尽管它可以
29 存管理优化，并提供复杂的精度/开销处理机制。因此，“空闲页面跟踪” 和 “perf mem” 可以提供
|