1 ================================================
2 HiSilicon PCIe Performance Monitoring Unit (PMU)
3 ================================================
5 On Hip09, HiSilicon PCIe Performance Monitoring Unit (PMU) could monitor
6 bandwidth, latency, bus utilization and buffer occupancy data of PCIe.
8 Each PCIe Core has a PMU to monitor multi Root Ports of this PCIe Core and
9 all Endpoints downstream these Root Ports.
12 HiSilicon PCIe PMU driver
13 =========================
15 The PCIe PMU driver registers a perf PMU with the name of its sicl-id and PCIe
18 /sys/bus/event_source/hisi_pcie<sicl>_core<core>
20 PMU driver provides description of available events and filter options in sysfs,
21 see /sys/bus/event_source/devices/hisi_pcie<sicl>_core<core>.
23 The "format" directory describes all formats of the config (events) and config1
24 (filter options) fields of the perf_event_attr structure. The "events" directory
25 describes all documented events shown in perf list.
27 The "identifier" sysfs file allows users to identify the version of the
30 The "bus" sysfs file allows users to get the bus number of Root Ports
33 Example usage of perf::
36 hisi_pcie0_core0/rx_mwr_latency/ [kernel PMU event]
37 hisi_pcie0_core0/rx_mwr_cnt/ [kernel PMU event]
38 ------------------------------------------
40 $# perf stat -e hisi_pcie0_core0/rx_mwr_latency/
41 $# perf stat -e hisi_pcie0_core0/rx_mwr_cnt/
42 $# perf stat -g -e hisi_pcie0_core0/rx_mwr_latency/ -e hisi_pcie0_core0/rx_mwr_cnt/
44 The current driver does not support sampling. So "perf record" is unsupported.
45 Also attach to a task is unsupported for PCIe PMU.
52 PMU could only monitor the performance of traffic downstream target Root
53 Ports or downstream target Endpoint. PCIe PMU driver support "port" and
54 "bdf" interfaces for users, and these two interfaces aren't supported at the
59 "port" filter can be used in all PCIe PMU events, target Root Port can be
60 selected by configuring the 16-bits-bitmap "port". Multi ports can be
61 selected for AP-layer-events, and only one port can be selected for
64 For example, if target Root Port is 0000:00:00.0 (x8 lanes), bit0 of
65 bitmap should be set, port=0x1; if target Root Port is 0000:00:04.0 (x4
66 lanes), bit8 is set, port=0x100; if these two Root Ports are both
67 monitored, port=0x101.
69 Example usage of perf::
71 $# perf stat -e hisi_pcie0_core0/rx_mwr_latency,port=0x1/ sleep 5
75 "bdf" filter can only be used in bandwidth events, target Endpoint is
76 selected by configuring BDF to "bdf". Counter only counts the bandwidth of
77 message requested by target Endpoint.
79 For example, "bdf=0x3900" means BDF of target Endpoint is 0000:39:00.0.
81 Example usage of perf::
83 $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,bdf=0x3900/ sleep 5
87 Event statistics start when the first time TLP length is greater/smaller
88 than trigger condition. You can set the trigger condition by writing
89 "trig_len", and set the trigger mode by writing "trig_mode". This filter can
90 only be used in bandwidth events.
92 For example, "trig_len=4" means trigger condition is 2^4 DW, "trig_mode=0"
93 means statistics start when TLP length > trigger condition, "trig_mode=1"
94 means start when TLP length < condition.
96 Example usage of perf::
98 $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,trig_len=0x4,trig_mode=1/ sleep 5
102 Counter counts when TLP length within the specified range. You can set the
103 threshold by writing "thr_len", and set the threshold mode by writing
104 "thr_mode". This filter can only be used in bandwidth events.
106 For example, "thr_len=4" means threshold is 2^4 DW, "thr_mode=0" means
107 counter counts when TLP length >= threshold, and "thr_mode=1" means counts
108 when TLP length < threshold.
110 Example usage of perf::
112 $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,thr_len=0x4,thr_mode=1/ sleep 5
116 When counting bandwidth, the data can be composed of certain parts of TLP
117 packets. You can specify it through "len_mode":
119 - 2'b00: Reserved (Do not use this since the behaviour is undefined)
120 - 2'b01: Bandwidth of TLP payloads
121 - 2'b10: Bandwidth of TLP headers
122 - 2'b11: Bandwidth of both TLP payloads and headers
124 For example, "len_mode=2" means only counting the bandwidth of TLP headers
125 and "len_mode=3" means the final bandwidth data is composed of both TLP
126 headers and payloads. Default value if not specified is 2'b11.
128 Example usage of perf::
130 $# perf stat -e hisi_pcie0_core0/rx_mrd_flux,len_mode=0x1/ sleep 5