00:02
mjw has quit [Quit: Leaving]
00:40
mjw has joined #systemtap
00:43
mjw has quit [Client Quit]
01:25
hpt has joined #systemtap
02:09
sapatel has quit [Ping timeout: 255 seconds]
02:11
sapatel has joined #systemtap
02:55
sscox has joined #systemtap
03:22
sapatel has quit [Quit: Leaving]
04:22
orivej has joined #systemtap
04:37
orivej has quit [Ping timeout: 265 seconds]
07:47
yogananth has joined #systemtap
09:05
ChanServ has quit [*.net *.split]
09:05
DUKENUKEM has quit [*.net *.split]
09:05
xar- has quit [*.net *.split]
09:05
darvon has quit [*.net *.split]
09:05
fLiPr3VeRsE has quit [*.net *.split]
09:05
hpt has quit [*.net *.split]
09:05
agentzh has quit [*.net *.split]
09:05
ema has quit [*.net *.split]
09:05
zodbot has quit [*.net *.split]
09:05
jistone has quit [*.net *.split]
09:05
gavinguo_ has quit [*.net *.split]
09:10
_whitelogger has joined #systemtap
09:43
lindi- has joined #systemtap
10:43
hpt has quit [Ping timeout: 265 seconds]
10:48
tonyj has quit [Remote host closed the connection]
10:56
mjw has joined #systemtap
11:20
sscox has quit [Ping timeout: 256 seconds]
12:21
irker376 has joined #systemtap
12:21
<
irker376 >
systemtap: fche systemtap.git:refs/heads/master * release-4.2-49-gffb40c659 / runtime/linux/autoconf-trace-printk.c tapset/linux/logging.stp testsuite/buildok/logging-embedded.stp testsuite/buildok/logging-embedded2.stp: RHBZ1810216: fix ftrace() tapset function related autoconf test
http://tinyurl.com/qrnolra
13:34
wcohen has quit [Ping timeout: 256 seconds]
13:35
tromey has joined #systemtap
13:49
beauty has joined #systemtap
13:53
beauty has quit [Remote host closed the connection]
14:04
sscox has joined #systemtap
14:25
beauty has joined #systemtap
14:27
<
beauty >
How to use relative path when using process,such as probe process("./a.out") ?
14:27
<
fche >
that works - relative to the cwd of the stap program
14:28
<
beauty >
semantic error: while resolving probe point: identifier 'process' at strobe_variable.stp:4:7
14:29
<
fche >
does that program exist at that location ?
14:29
<
beauty >
but I failed.
14:29
<
beauty >
yeah, it is
14:30
<
beauty >
but when I use full path ,it passed.
14:30
<
fche >
hm, works for me here ... so
14:30
<
fche >
what stap version
14:30
<
fche >
do you have a complete command line
14:31
<
beauty >
Systemtap translator/driver (version 4.0/0.172/0.176, rpm 4.0-10.el7_7)
14:31
<
beauty >
complete command line?
14:31
<
beauty >
probe process("./a.out").function("main@./main.c")
14:32
<
fche >
tried without the @./main.c ?
14:32
<
beauty >
yeah, it passed
14:32
<
beauty >
so amazing
14:33
<
fche >
my guess is that your a.out program was not compiled in such a way that the compiler saw the path "./main.c"
14:33
<
fche >
maybe it was just "main.c" or something longer
14:33
wcohen has joined #systemtap
14:33
<
beauty >
gcc -O0 -g main.c
14:34
<
fche >
and actually the NAME@FILE syntax is for .statement probes
14:34
<
beauty >
I use this compile parameters
14:34
<
fche >
and there NAME@FILE:LINE# is the scheme
14:35
<
beauty >
Yeah, I first use statement
14:35
<
beauty >
When I met error, I use functions.
14:36
<
beauty >
And statement also failed if using func@./main.c
14:36
<
fche >
next time try using wildcards in your .statement probe, or stap -L 'process("./a.out").statement("*")' e.g.
14:36
<
beauty >
I want probe spectial file
14:38
<
beauty >
so "func@file path" must keep file path is full?
14:38
<
fche >
function("name") is the proper syntax
14:38
<
fche >
or statement("name@file:line") where many components can be wildcarded
14:42
beauty has quit [Ping timeout: 260 seconds]
14:46
beauty has joined #systemtap
14:46
<
beauty >
sorry, just now I offline
14:48
<
beauty >
so the file path in statement must be full path?
14:51
beauty95 has joined #systemtap
14:52
<
beauty95 >
Sorry, just now I'm offline.
14:53
<
beauty95 >
So the file path in statement must use full path?
14:54
<
fche >
that can use wildcards too
14:54
<
fche >
use stap -L (list) to see the data as stap recovers it from the debuginfo and your wildcards
14:54
<
beauty95 >
Yeah, I know that can use wildcards
14:55
<
beauty95 >
I want to know whether must use full path in statement(fun@file:line)
14:56
<
fche >
generally no
14:56
<
beauty95 >
beause the full path is too long and not relative for portability
14:57
<
fche >
try using the basename, and/or a partial wildcard
14:57
<
beauty95 >
So what should I do to if using relative path in statement
14:57
<
beauty95 >
what is basename?
14:57
<
fche >
see [man basename]
14:57
<
fche >
"what should I do" - use whatever works. stap will try to match up a partial name against the ones recorded in the debuginfo by the compiler
14:58
<
fche >
stap -L is a good way of seeing those full names,
14:58
<
fche >
you can then try writing wildcards or basename abbreviations in the stap code
14:58
amerey has joined #systemtap
14:58
<
beauty95 >
thank you
14:59
<
beauty95 >
I have another question
14:59
<
beauty95 >
stap -e 'global steps; probe process("/bin/ls").insn {steps++}
14:59
<
beauty95 >
semantic error: while resolving probe point: identifier 'process' at <input>:1:21
15:00
<
fche >
yes, kernels more modern than rhel6 era don't support this unfortunately
15:00
<
beauty95 >
don't support ".instn" ????
15:01
<
beauty95 >
So what's the difference between statement("*") and ".instn"?
15:01
<
beauty95 >
It seems the same.
15:02
<
fche >
statement operates at source level statement boundaries
15:02
<
fche >
insn operates (operated) at the machine instruction boundaries
15:02
<
fche >
statements are generally much larger than instructions
15:03
<
beauty95 >
It seems insn is aimed at assembly
15:03
<
beauty95 >
statement is aimed at source code.
15:03
<
fche >
yes, and it's broad/nonspecific
15:03
<
beauty95 >
It seems insn is more accurate
15:04
<
fche >
they are different
15:05
<
beauty95 >
which is more close to perf instruction ?
15:05
<
fche >
perf doesn't do anything down at the instruction level ... it can sample periodically and it can trace intermittently
15:05
<
fche >
but not down at the insn level
15:05
<
beauty95 >
Yeah, I know thi.
15:07
<
beauty95 >
If I use probe every instruction and count using insn or statement("*") by systemtap. Are these values close to perf instruction?
15:08
<
fche >
if you just want to -count- instructions, use a performance counter type widget; stap exposes these to some extent
15:08
<
fche >
probing instructions or statements is more than counting, it's stopping & inspecting & resuming
15:09
khaled has joined #systemtap
15:10
<
beauty95 >
systemtap will record more instructions than the probed program? The extra instructions is used and generated by systemtap.
15:25
<
fche >
not sure what you mean by 'record'
15:28
<
beauty95 >
I mean that more instruction will be printed when probed by systemtap that the program's actual instructions when not probed by systemtap.
15:28
<
beauty95 >
the second that is "than"
15:28
<
fche >
printed by what software?
15:29
<
beauty95 >
stap -e 'global steps; probe process("/bin/ls").insn {steps++} probe end {printf("Total instructions: %d\n", steps);}' -c /bin/ls
15:29
<
beauty95 >
this "printf"
15:30
<
beauty95 >
total instructions
15:31
<
fche >
in that case, the number of instructions executed under .insn probing (on a capable kernel) should be right about the actual number of instructions executed by that process (in userspace)
15:32
<
beauty95 >
yeah, the perf sample instructions is the same as the actual number of instructions
15:32
<
beauty95 >
I have tested it.
15:32
<
beauty95 >
So the systemtap and perf is the same.
15:33
<
fche >
well they measure things very differently, but in this particular case should have comparable results
15:33
<
beauty95 >
Year, I will test it.
15:35
<
beauty95 >
When using process("PATH").mark("LABEL"), I just insert "STAP_PROBE(xxx,"LABEL")" in the special source code position and compile as normal. right?
15:40
sapatel has joined #systemtap
15:42
beauty95 has quit [Ping timeout: 260 seconds]
15:42
Amy1 has joined #systemtap
15:42
<
Amy1 >
I am offline just now
15:42
<
fche >
no you're not :)
15:43
<
Amy1 >
so is it right?
15:46
beauty30 has joined #systemtap
15:56
<
beauty30 >
STAP_PROBE1(0,"xx",1);
15:57
<
beauty30 >
semantic error: while resolving probe point: identifier 'process' at strobe_variable.stp:6:7
15:58
<
fche >
use ordinary symbols for the first and second parameters
15:58
<
fche >
STAP_PROBE1(beauty30,xx,1);
15:59
<
fche >
and again listing with wildcards is useful: stap -L 'process("a.out").mark("*")'
16:04
beauty30 has quit [Ping timeout: 260 seconds]
16:05
<
Amy1 >
cool. it passed
16:06
<
Amy1 >
than you very much.
16:07
<
Amy1 >
Almost every question has solved.
16:07
<
fche >
I have not solved the life-universe-everything problem JUST YET
16:07
<
Amy1 >
I spent a week to learn stap.
16:07
<
fche >
hey glad you did!
16:07
<
fche >
hope it helps you
16:08
<
Amy1 >
It's so easy to learn even much easier than vb. :)
16:08
<
fche >
now that's scary
16:53
mjw has quit [Ping timeout: 256 seconds]
17:05
mjw has joined #systemtap
17:17
tonyj has joined #systemtap
17:55
beauty has quit [Quit: Connection closed for inactivity]
18:43
beauty has joined #systemtap
18:45
<
beauty >
I know systemtap many years ago.
18:46
<
beauty >
This week I learned it.
18:47
<
fche >
if you find it successful - or even unsuccessful - at helping with your problems, we'd love to hear
18:49
<
beauty >
bpf is developing nowadays
18:49
<
beauty >
how about bpf, fche
18:50
<
fche >
stap --bpf exists as an optional backend - many limitations (inherited from kernel bpf rules)
18:50
<
fche >
but you can try simple things - on new kernels & new systemtap at least
18:51
<
beauty >
many conferences have it
18:51
<
fche >
yes, we know :)
18:55
<
beauty >
fche, could you show me some performance stp.
18:55
<
beauty >
useful and popular
18:59
<
beauty >
yeah, I know this
19:14
mjw has quit [Quit: Leaving]
19:17
mjw has joined #systemtap
19:42
mjw has quit [Quit: Leaving]
20:03
sscox has quit [Ping timeout: 240 seconds]
20:17
sscox has joined #systemtap
20:23
orivej has joined #systemtap
21:15
beauty has quit [Quit: Connection closed for inactivity]
21:26
mjw has joined #systemtap
21:45
orivej has quit [Ping timeout: 265 seconds]
21:58
khaled has quit [Quit: Konversation terminated!]
22:01
orivej has joined #systemtap
22:17
tromey has quit [Quit: ERC (IRC client for Emacs 28.0.50)]
22:20
orivej has quit [Ping timeout: 260 seconds]
22:24
sscox has quit [Ping timeout: 258 seconds]
22:33
wcohen has quit [Ping timeout: 256 seconds]
22:57
mjw has quit [Quit: Leaving]
23:04
amerey has quit [Quit: Leaving]
23:06
irker376 has quit [Quit: transmission timeout]
23:21
gromero has joined #systemtap
23:33
wcohen has joined #systemtap