Hi all, does someone could help me with this issue?
I use perf to record such performance counter:
sudo perf record --stat -fo perf_dacapo.data -e context-switches
-e major-faults java -jar dacapo-9.12-bach.jar xalan
But when I use perf report :
sudo perf report -i perf_dacapo_tomcat.data --stdio -n --threads|less
I get the result like this:
# ========
# captured on: Tue Jun 18 15:22:30 2013
# hostname : shashaa-OptiPlex-960
# os release : 3.2.0-48-generic
# perf version : 3.2.46
# arch : x86_64
# nrcpus online : 4
# nrcpus avail : 4
# cpudesc : Intel(R) Core(TM)2 Quad CPU Q9400 @ 2.66GHz
# cpuid : GenuineIntel,6,23,10
# total memory : 8046912 kB
# cmdline : /usr/bin/perf_3.2.0-48 record --stat -fo
perf_dacapo_tomcat.data -e context-switches -e major-faults java -jar
Dropbox/perf/dacapo-9.12-bach.jar xalan -t 3
# event : name = context-switches, type = 1, config = 0x3, config1 =
0x0, config2 = 0x0, excl_usr = 0, excl_kern = 0, id = { 3768, 3769,
3770, 3771 }
# event : name = major-faults, type = 1, config = 0x6, config1 = 0x0,
config2 = 0x0, excl_usr = 0, excl_kern = 0, id = { 3772, 3773, 3774,
3775 }
# HEADER_CPU_TOPOLOGY info available, use -I to display
# HEADER_NUMA_TOPOLOGY info available, use -I to display
# ========
#
# Events: 1K context-switches
#
# Overhead Command Shared Object Symbol
# ........ ....... ................. ..........
#
100.00% java [kernel.kallsyms] [k] __schedule
# Events: 0 major-faults
#
# Overhead Command Shared Object Symbol
# ........ ....... ............. ......
#
#
# (For a higher level overview, try: perf report --sort comm,dso)
#
# PID TID major-faults major-faults context-switches
context-switches context-switches context-switches major-faults
major-faults
7092 7108 218 218 431
445 206
212 61 30
7092 7107 175 310 58
70 95
501 283 366
7092 7106 66 26 343
154 244
394 283 194
7092 7105 16 0 4
8 16
36 16 22
7092 7100 3 4 26
27 7
52 2 15
7092 7098 5 1 3
32 68
129 22 25
7092 7102 107 204 441
617 101
426 9 7
7092 7099 0 3 4
18 0
13 2 3
7092 7104 0 0 0
0 3
0 0 0
7092 7095 0 0 5
25 12
12 26 7
7092 7097 2 10 34
14 22
9 0 0
7092 7096 1 7 26
16 25
13 0 2
7092 7103 216 11 1
236 368
339 302 281
7092 7093 24 38 40
16 0
0 0 0
7092 7094 4 7 4
11 22
17 22 3
7092 7101 0 0 0
0 0
6 0 0
My Question is why each thread has more than 2 events counted in the
table?? For each thread, there are 4 major-faults and 4
context-switches, but they have different number for each thread??
Thanks very much for your advice.
Best wishes.
Yours,
Wang Peipei
--
To unsubscribe from this list: send the line "unsubscribe linux-perf-users" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
