1PQOS(8)                     System Manager's Manual                    PQOS(8)
2
3
4

NAME

6       pqos,  pqos-msr,  pqos-os  -  Intel(R) Resource Director Technology/AMD
7       PQoS monitoring and control tool
8

SYNOPSIS

10       pqos [OPTIONS]...
11

DESCRIPTION

13       Intel(R) Resource Director Technology/AMD PQoS is designed  to  monitor
14       and  manage  cpu  resources and improve performance of applications and
15       virtual machines.
16
17       Intel(R) Resource Director Technology/AMD PQoS includes monitoring  and
18       control  technologies. Monitoring technologies include CMT (Cache Moni‐
19       toring Technology), which monitors occupancy of last level  cache,  and
20       MBM  (Memory  Bandwidth  Monitoring).  Control technologies include CAT
21       (Cache Allocation Technology), CDP (Code Data Prioritization)  and  MBA
22       (Memory Bandwidth Allocation).
23
24       pqos  supports  CMT and MBM on a per core or hardware thread basis. MBM
25       supports two types of events reporting local and  remote  memory  band‐
26       width.
27       pqos-msr and pqos-os are simple pqos wrapper scripts that automatically
28       select the MSR or OS/Kernel library interface to program the  technolo‐
29       gies.
30       Please see the -I option below for more information.
31
32       For  hardware  information  please  refer  to  the  README  located on:
33       https://github.com/01org/intel-cmt-cat/blob/master/README
34

OPTIONS

36       pqos options are as follow:
37
38       -h, --help
39              show help
40
41       -v, --verbose
42              verbose mode
43
44       -V, --super-verbose
45              super-verbose mode
46
47       -l FILE, --log-file=FILE
48              log messages into selected log FILE
49
50       -s, --show
51              show the current allocation and monitoring configuration
52
53       -d, --display
54              display supported Intel(R) Resource Director Technology/AMD PQoS
55              capabilities
56
57       -D, --display-verbose
58              display supported Intel(R) Resource Director Technology/AMD PQoS
59              capabilities in verbose mode
60
61       -f FILE, --config-file=FILE
62              load commands from selected configuration FILE
63
64       -e CLASSDEF, --alloc-class=CLASSDEF
65              define the allocation classes on all CPU sockets. CLASSDEF  for‐
66              mat is "TYPE:ID=DEFINITION;...".
67              define  classes  for  selected CPU resources. CLASSDEF format is
68              "TYPE[@RESOURCE_ID]:ID=DEFINITION;...".
69              For CAT, TYPE is "llc" for the last level  cache  (aka  l3)  and
70              "l2"  for level 2 cache, ID is a CLOS number and DEFINITION is a
71              bitmask.
72              For MBA, TYPE is "mba", ID is a CLOS number and DEFINITION is  a
73              value between 1 and 100 representing the percentage of available
74              memory bandwidth.
75              For MBA CTRL, TYPE is "mba_max", ID is a CLOS number and DEFINI‐
76              TION  is  a  value  representing  the requested memory bandwidth
77              specified in MBps.
78              RESOURCE_ID is a unique number that can represent  a  socket  or
79              l2/l3 cache identifier. The RESOURCE_ID for each logical CPU can
80              be found using "pqos -s"
81              Note: When L2/L3 CDP is on, ID can be  postfixed  with  'D'  for
82              data or 'C' for code.
83              Note: L2/L3 CDP is available on selected CPUs only.
84              Note:  MBA  CTRL  is  supported  only  by  the  OS interface and
85              requires Linux and kernel version 4.18 or newer.
86              Some examples:
87                     "-e llc:0=0xffff;llc:1=0x00ff"
88                     "-e llc@0-1:2=0xff00;l2:2=0x3f;l2@2:1=0xf"
89                     "-e llc:0d=0xfff;llc:0c=0xfff00"
90                     "-e l2:0d=0xf;l2:0c=0xc"
91                     "-e mba:1=30;mba@1:3=80"
92                     "-e mba_max:1=6000;mba_max@1:3=10000"
93              Note:
94                     "-e l2:2=0x3f" means that COS2 for all L2 cache  clusters
95                     is changed to 0x3f.
96                     "-e l2@2:1=0xf" means that COS1 for L2 cache cluster 2 is
97                     changed to 0xf.
98                     "-e mba:1=30" means that COS1, on all sockets,  can  uti‐
99                     lize up to 30% of available memory bandwidth.
100                     "-e  mba_max:1=6000" means that COS1, on all sockets, can
101                     utilize up to 6000 MBps of memory bandwidth.
102
103       -a CLASS2ID, --alloc-assoc=CLASS2ID
104              associate allocation classes  with  cores.  CLASS2ID  format  is
105              "TYPE:ID=CORE_LIST;..." or "TYPE:ID=TASK_LIST;...".
106              For  CAT,  TYPE is "llc", "core" or "pid" and ID is a class num‐
107              ber. CORE_LIST  is  comma  or  dash  separated  list  of  cores.
108              TASK_LIST is comma or dash separated list of process/task ID's.
109              For example:
110                     "-a  llc:0=0,2,4,6-10;llc:1=1;" associates cores 0, 2, 4,
111                     6, 7, 8, 9, 10 with CAT class 0 and core 1 with class 1.
112                     "-a core:0=0,2,4,6-10;core:1=1;" associates cores  0,  2,
113                     4,  6, 7, 8, 9, 10 with CAT class 0 and core 1 with class
114                     1.
115                     "-I   -a   pid:0=3543,7643,4556;pid:1=7644;"   associates
116                     process  ID 3543, 7643, 4556 with CAT class 0 and process
117                     ID 7644 with class 1.
118              Note:
119                     The -I option must be used for PID association.
120
121       -R [CONFIG[,CONFIG]], --alloc-reset[=CONFIG[,CONFIG]]
122              reset allocation setting (L3 CAT, L2 CAT, MBA)  and  reconfigure
123              L3 CDP. CONFIG is one of the following options:
124              l3cdp-on  sets L3 CDP on
125              l3cdp-off sets L3 CDP off
126              l3cdp-any keeps current L3 CDP setting (default)
127              l2cdp-on  sets L2 CDP on
128              l2cdp-off sets L2 CDP off
129              l2cdp-any keeps current L2 CDP setting (default)
130              mbaCtrl-on     sets MBA CTRL on
131              mbaCtrl-off    sets MBA CTRL off
132              mbaCtrl-any    keeps current MBA CTRL setting (default)
133
134       -m EVTCORES, --mon-core=EVTCORES
135              select  the  cores and events for monitoring, EVTCORES format is
136              "EVENT:CORE_LIST". Valid EVENT settings are:
137              - "llc" for CMT (LLC occupancy)
138              - "mbr" for MBR (remote memory bandwidth)
139              - "mbl" for MBL (local memory bandwidth)
140              - "mbt" for MBT (total memory bandwidth)
141              - "all" or ""  for all detected event types (except MBT)
142              CORE_LIST is comma or dash separated list of cores.
143              Example "-m all:0,2,4-10;llc:1,3;mbr:11-12".
144              Core statistics can be grouped by enclosing  the  core  list  in
145              square brackets.
146              Example "-m llc:[0-3];all:[4,5,6];mbr:[0-3],7,8".
147
148       -p [EVTPIDS], --mon-pid[=EVTPIDS]
149              select top 10 most active (CPU utilizing) process ids to monitor
150              or select the process ids and events to monitor, EVTPIDS  format
151              is "EVENT:PID_LIST".
152              See  -m option for valid EVENT settings. PID_LIST is comma sepa‐
153              rated list of process ids.
154              Examples:
155                     "-p llc:22,25673"
156                     "-p all:892,4588-4592"
157
158              Process's IDs can be grouped by enclosing them in square  brack‐
159              ets.
160              Examples:
161                     "-p llc:[22,25673]"
162                     "-p all:892,[4588-4592]"
163
164              Note:
165                     Requires Linux and kernel versions 4.10 and newer.
166                     The -I option must be used for PID monitoring.
167                     It  is  not possible to track both processes and cores at
168                     the same time.
169
170       -T, --mon-top
171              enable top like monitoring output sorted by  highest  LLC  occu‐
172              pancy
173
174       -o FILE, --mon-file FILE
175              select  output  FILE  to store monitored data in, the default is
176              'stdout'
177
178       -u TYPE, --mon-file-type=TYPE
179              select the output format TYPE for monitored data. Supported TYPE
180              settings are: "text" (default), "xml" and "csv".
181
182       -i INTERVAL, --mon-interval=INTERVAL
183              define  monitoring  sampling  INTERVAL  in 100ms units, 1=100ms,
184              default 10=10x100ms=1s
185
186       -t SECONDS, --mon-time=SECONDS
187              define monitoring time in seconds, use 'inf' or  'infinite'  for
188              infinite monitoring. Use CTRL+C to stop monitoring at any time.
189
190       -r, --mon-reset
191              reset monitoring and use all RMID's and cores in the system
192
193       --disable-mon-ipc
194              Disable IPC monitoring
195
196       --disable-mon-llc_miss
197              Disable LLC misses monitoring
198
199       -H, --profile-list
200              list supported allocation profiles
201
202       -c PROFILE, --profile-set=PROFILE
203              select  a  PROFILE from predefined allocation classes, use -H to
204              list available profiles
205
206       -I, --iface-os
207              set the library interface to use the kernel  implementation.  If
208              not  set  the  default  implementation  is  to program the MSR's
209              directly.
210

NOTES

212       CMT, MBM and CAT are configured using Model Specific Registers  (MSRs).
213       The  pqos  software  executes  in user space, and access to the MSRs is
214       obtained through a standard Linux* interface. The msr file interface is
215       protected  and  requires  root privileges.  The msr driver might not be
216       auto-loaded and on some modular kernels  the  driver  may  need  to  be
217       loaded manually:
218
219       For Linux:
220       sudo modprobe msr
221
222       For FreeBSD:
223       sudo kldload cpuctl
224
225       Interface enforcement:
226       If  you require system wide interface enforcement you can do so by set‐
227       ting the "RDT_IFACE" environment variable.
228

SEE ALSO

230       msr(4)
231

AUTHOR

233       pqos was written by Tomasz Kantecki <tomasz.kantecki@intel.com>, Marcel
234       Cornu  <marcel.d.cornu@intel.com>, Aaron Hetherington <aaron.hethering‐
235       ton@intel.com>
236
237       This is free software; see the source for copying conditions.  There is
238       NO  warranty;  not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR
239       PURPOSE.
240
241
242
243                                 May 08, 2020                          PQOS(8)
Impressum