fche changed the topic of #systemtap to: http://sourceware.org/systemtap; email systemtap@sourceware.org if answers here not timely, conversations may be logged
hkshaw has joined #systemtap
gila has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]
hkshaw has quit [Ping timeout: 245 seconds]
gromero_ has quit [Ping timeout: 246 seconds]
baotiao has joined #systemtap
baotiao has quit [Ping timeout: 260 seconds]
sanoj has quit [Ping timeout: 246 seconds]
CME has quit [Ping timeout: 240 seconds]
CME has joined #systemtap
nkambo_ has quit []
baotiao has joined #systemtap
baotiao has quit [Ping timeout: 260 seconds]
hkshaw has joined #systemtap
humblec has quit [Ping timeout: 255 seconds]
ravi__ has joined #systemtap
sanoj has joined #systemtap
Humble has joined #systemtap
baotiao has joined #systemtap
baotiao has quit [Ping timeout: 255 seconds]
orivej has quit [Ping timeout: 246 seconds]
baotiao has joined #systemtap
baotiao has quit [Ping timeout: 240 seconds]
wielaard has joined #systemtap
wielaard has quit [Client Quit]
wielaard has joined #systemtap
nkambo has joined #systemtap
pwithnall___ has joined #systemtap
mjw has quit [Quit: Leaving]
gila has joined #systemtap
wielaard is now known as mjw
nkambo has quit [Ping timeout: 240 seconds]
gila has quit [*.net *.split]
wcohen has quit [*.net *.split]
ppetraki has quit [*.net *.split]
csanting has quit [*.net *.split]
drsmith_away has quit [*.net *.split]
scox has quit [Ping timeout: 246 seconds]
gila has joined #systemtap
nkambo has joined #systemtap
gila has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]
csanting has joined #systemtap
ppetraki has joined #systemtap
drsmith_away has joined #systemtap
wcohen has joined #systemtap
hkshaw has quit [Quit: Leaving.]
orivej has joined #systemtap
ravi__ has quit [Quit: Leaving]
gromero_ has joined #systemtap
hkshaw has joined #systemtap
gromero_ has quit [Quit: Leaving]
gromero_ has joined #systemtap
sanoj has quit [Ping timeout: 240 seconds]
orivej has quit [Ping timeout: 255 seconds]
baotiao has joined #systemtap
scox has joined #systemtap
orivej has joined #systemtap
wcohen has quit [Ping timeout: 260 seconds]
mbenitez has joined #systemtap
gila has joined #systemtap
drsmith_away is now known as drsmith
drsmith is now known as drsmith_away
drsmith_away is now known as drsmith
brolley has joined #systemtap
wcohen has joined #systemtap
Humble has quit [Ping timeout: 268 seconds]
mbenitez has quit [Ping timeout: 240 seconds]
mbenitez has joined #systemtap
Humble has joined #systemtap
hkshaw has quit [Ping timeout: 272 seconds]
tromey has joined #systemtap
mjw has quit [Quit: Leaving]
gila has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]
mbenitez has quit [Quit: To lunch / office]
orivej has quit [Ping timeout: 240 seconds]
orivej has joined #systemtap
orivej has quit [Read error: Connection reset by peer]
<fche>
baotiao, 'good quality' - created by modern gcc, not degraded
<fche>
see the tail notes on the systemtap release notes
<baotiao>
fche: since I use centos. I always download the debuginfo and ddebuginfo-comman with same kernel version from centos? is it 'good quality'
<fche>
yeah centos should be fine.
<fche>
well, much older gcc than e.g. fedora rawhide, so less good quality
<fche>
but not as bad as other distros
<baotiao>
fche: thank you. Another question, I want to print a stack variable in the function, but I can't get this variable throught stap -L kernel.function(""). Is it impossible to printf all the variable in some kernel function?
<fche>
$$vars
<fche>
see [man stapprobes]
<fche>
CONTEXT VARIABLES
<baotiao>
I know $$vars, but some variable isn't in $$vars
<fche>
that would mean that the compiler did not preserve even the existence of those variables in the context where the probe is placed
<fche>
e.g. if you probe a function entry (with a .function probe), then any local variables in that function won't exist (yet), so can't be shown.
<baotiao>
Yes
<fche>
If you use a .statement probe in the interior of a function, then you will (should) see those that are in scope there.
<baotiao>
I guess kernel do this way
<baotiao>
wow . let me try .statement probe
<baotiao>
thank you fche
<fche>
righto, .statement probes are an underappreciated gem
<fche>
esp. when combined with wildcards or line-number-ranges
<baotiao>
wow what a pity. I want use systemtap the trace the kernel function and see the local variable there
<fche>
so go for it
<fche>
so probe kernel.statement("function_name@*:*") { println($$vars) }
<fche>
we have samples that do this
<baotiao>
where is the sample? can you show me?
<baotiao>
I have try this way. However, it always return error probe kernel.statement("*@mm/madvise.c:482")
<baotiao>
the line numbers is right, the error is: semantic error: unable to find local 'len'
<baotiao>
since len is just a local variable. So I can't print it
<fche>
yeah, the compiler might not have preserved it at that location
<fche>
try , as per my example, a range of line numbers, and $$vars
<fche>
$len may be accessible somewhere nearby
<baotiao>
ok thank you
<baotiao>
let me have a try
<baotiao>
No. I think wo can only printf the variable the same throught stap -L
<baotiao>
fche: thank you. I an wrong, Throught the way you teach me no I can printf the len local variable
<baotiao>
probe kernel.statement("*@mm/madvise.c:*") { printf("%s\n", $$vars) } and every time kernel hit madvise.c file will show all local variable
<fche>
yeah
<fche>
and then you can see which of those lines includes a $len
<fche>
then, if that line is still good, you could change the script to limit itself to that line only
<fche>
and then 'bob's your uncle'
<baotiao>
how can I limit to that line
<fche>
replace :* with :nnnn in the .statement probe
<baotiao>
ok
<baotiao>
how can I get the line number?
<baotiao>
printf("%s\n", $$vars) doesn't show the line number
baotiao has quit [Quit: baotiao]
<fche>
pp()
<fche>
drsmith, btw. re. ways to spawn processes from multithreaded process, we have some experience with that
<fche>
e.g. in pcp pmmgr
<fche>
but methinks also here and there in systemtap
<fche>
stap-serverd iirc
<drsmith>
right, I've been looking at that
<drsmith>
(systemtap that is, not pcp)
<fche>
righto
gila has joined #systemtap
pwithnall___ has quit [Remote host closed the connection]
pwithnall___ has joined #systemtap
orivej has quit [Read error: Connection reset by peer]
pwithnall__ has joined #systemtap
scox has quit [Ping timeout: 260 seconds]
pwithnall__ has quit [Ping timeout: 240 seconds]
gila has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]
brolley has left #systemtap [#systemtap]
mbenitez has quit [Quit: Leaving]
wcohen has quit [Ping timeout: 260 seconds]
<irker446>
systemtap: dsmith systemtap.git:refs/heads/master * release-3.1-82-g197f573 / httpd/client.py httpd/main.cxx httpd/server.cxx httpd/server.h: The httpd server now expects the command line to be sent tokenized. http://tinyurl.com/yafabwls
drsmith is now known as drsmith_away
orivej has joined #systemtap
wcohen has joined #systemtap
pwithnall____ has joined #systemtap
pwithnall___ has quit [Read error: Connection reset by peer]
gromero_ has quit [Remote host closed the connection]
pwithnall____ has quit [Quit: pwithnall____]
tromey has quit [Quit: ERC (IRC client for Emacs 26.0.50)]