fche changed the topic of #systemtap to: http://sourceware.org/systemtap; email systemtap@sourceware.org if answers here not timely, conversations may be logged
scox has joined #systemtap
irker224 has quit [Quit: transmission timeout]
hkshaw has quit [Ping timeout: 255 seconds]
hkshaw has joined #systemtap
srikar_away is now known as srikar
ravi_ has joined #systemtap
nkambo has joined #systemtap
srikar is now known as srikar_away
srikar_away is now known as srikar
srikar is now known as srikar_away
ph7 has quit [Read error: Connection timed out]
ph7 has joined #systemtap
srikar_away is now known as srikar
srikar is now known as srikar_away
zodbot has quit [Read error: Connection reset by peer]
srikar_away has quit [Ping timeout: 250 seconds]
zodbot has joined #systemtap
wcohen has quit [Ping timeout: 276 seconds]
lorddoskias1 has joined #systemtap
<lorddoskias1> what's the issue here?
<lorddoskias1> i mean clearly the ioblock_request probe point doesn't know what type bdev is
<lorddoskias1> in bdevname there is already: bdev = & @cast(bdev, "block_device") so bdev should be of known type ?
nkambo has quit [Ping timeout: 248 seconds]
mjw has joined #systemtap
nkambo has joined #systemtap
srikar_away has joined #systemtap
srikar_away is now known as srikar
wcohen has joined #systemtap
<lorddoskias1> fche: any ideas?
<lorddoskias1> :)
wcohen has quit [Ping timeout: 260 seconds]
nkambo has quit [Ping timeout: 244 seconds]
scox has quit [Ping timeout: 265 seconds]
wcohen has joined #systemtap
hkshaw has quit [Quit: Leaving.]
nkambo has joined #systemtap
ravi_ has quit [Remote host closed the connection]
<fche> hi lord
<fche> sorry, it's not jumping out at me ... obviously some info is missing, but we're not helping well to find it
<fche> try a little more verbosity (stap --vp 03 ....) say
<fche> does that 4.1.14-clouder5 vmlinux image have debuginfo/etc. retained ?
<lorddoskias1> yes indeed, it does
<lorddoskias1> let me try it
<lorddoskias1> also one more thing
<lorddoskias1> if i want to set a probe on a trace point e.g. kernel.trace("block_issue_request")
<lorddoskias1> is there a way to access the various "args" that the tracepoint prints?
<lorddoskias1> or no way?
<fche> oh sure there's a way
<fche> [man stapprobes] search for KERNEL TRACEPOINTS
<lorddoskias1> right
<lorddoskias1> still, i'd like to use whatever is provided by stapruntime :)
<fche> the tapset? yeah
<fche> but if something's up with the debuginfo/symbol-tables, yeah the kernel tracepoints are a great alternative
<lorddoskias1> eyeah, but it's giving me error
<fche> and really the tapset should prefer it (and does in other spots) over time
<lorddoskias1> but the debug symbols should be good
<lorddoskias1> since doing stap -L 'kernel.statement("generic_make_request@*:*")'
<lorddoskias1> prints what's at the particular line
<fche> yup
<lorddoskias1> ops
<lorddoskias1> maybe i was too hasty
* fche will return shortly, breakfast beckons with an insistent voice
<lorddoskias1> so fche stap -dkernel -r4.4.14-clouder5 -BCC=/opt/rh/devtoolset-3/root/usr/bin/gcc -L 'kernel.statement("generic_make_request@*:*")' this command works like a charm
<lorddoskias1> this one doesn't stap -dkernel -r4.4.14-clouder5 -BCC=/opt/rh/devtoolset-3/root/usr/bin/gcc block-trace.stp
wcohen has quit [Ping timeout: 244 seconds]
hkshaw has joined #systemtap
nkambo has quit [Ping timeout: 276 seconds]
<fche> hey lorddoskias1
<fche> any progress/luck ?
<lorddoskias1> nope
<lorddoskias1> as i said listing symbols work
<lorddoskias1> but when i try to compile the script - error
<fche> -L statement stuff relies on debuginfo, it's good that this works
<fche> I wonder if perhaps our implementation of @cast() doesn't do the right thing with respect to cross-compiled (-rXXXX) runs
nkambo has joined #systemtap
<lorddoskias1> hm, maybe
mbenitez has joined #systemtap
mbenitez has joined #systemtap
mbenitez has quit [Changing host]
ravi_ has joined #systemtap
ravi_ has quit [Quit: Leaving]
brolley has joined #systemtap
srikar is now known as srikar_away
tromey has joined #systemtap
scox has joined #systemtap
drsmith has joined #systemtap
wcohen has joined #systemtap
ph7 has quit [Read error: Connection timed out]
ph7 has joined #systemtap
mbenitez has quit [Quit: To office]
przemoc has quit [Ping timeout: 265 seconds]
przemoc has joined #systemtap
mbenitez has joined #systemtap
mbenitez has joined #systemtap
mbenitez has quit [Changing host]
nkambo has quit [Ping timeout: 240 seconds]
ph7 has quit [Read error: Connection timed out]
ph7 has joined #systemtap
srikar_away has quit [Ping timeout: 244 seconds]
hkshaw has quit [Ping timeout: 248 seconds]
wcohen has quit [Ping timeout: 244 seconds]
mbenitez has quit [Quit: Leaving]
tromey has quit [Quit: ERC (IRC client for Emacs 25.1.3)]
irker289 has joined #systemtap
<irker289> systemtap: dsmith systemtap.git:refs/heads/dsmith/python * release-3.0-246-ge5d623b / Makefile.in configure configure.ac doc/Makefile.in doc/SystemTap_Tapset_Reference/Makefile.in doc/beginners/Makefile.in dtrace.in java/Makefile.in man/Makefile.in man/cs/Makefile.in python/Makefile.in stapdyn/Makefile.in staprun/Makefile.in: Added autoconf logic to search for python version 2 and 3. http://tinyurl.com/heybnre
_whitelogger has joined #systemtap
bendlas has joined #systemtap
alkgayua has joined #systemtap
<bendlas> i tried to use systemtap-3.0 against kernel 4.7.3, but I get compilation errors during runtime.
<fche> you'll need git systemtap to work against newer kernels
<bendlas> OIC
<bendlas> is master good to track?
<fche> yessir
CustosLimen has joined #systemtap
<bendlas> cool, I'll try that
hchiramm has joined #systemtap
<bendlas> woot, it works
<bendlas> now we can have a working systemtap in nixos again :-)
<bendlas> thanks, fche
<fche> ship it!
<fche> ship it NOW
<bendlas> I only have to cross-check the failure mode for when DEBUG_INFO is missing from the kernel
<fche> we don't call that failure
<fche> we call that success in a different way :-)
<bendlas> oh, so it works with reduced functionality?
<bendlas> that's even better
<fche> quite a bit works -- see [man stapprobes] for DWARF
<bendlas> let me just boot a vm w/o DEBUG_INFO, real quick
<fche> not everything
<bendlas> btw, do you think it's worth it, to separate out debug data, like fedora does?
<fche> debuginfo is a godsend if one needs to debug
<fche> it's a waste of disk space if one never does
<bendlas> IE, can a production kernel be compiled with DEBUG_INFO, and yet be the same as w/o after stripping?
<fche> that's what fedora/rhel-style debuginfo splitting does
<bendlas> can systemtap take advantage of more of the extra debug flags from rawhide?
<fche> you may be referring to kernel-devel-* builds, wherein more self-instrumentation is compiled into the kernel
<fche> that's an orthogonal thing, stap is not involved in that
<bendlas> ie, should we provide a separate debug kernel for some use cases of systemtap, or is just debug_info good enough?
<bendlas> oh, that's good
<fche> -debug- kernels usually cost you speed but give better diagnostics/checking
<fche> kernel -debuginfo lets tools like systemtap debug kernels (and kernel-debug- kernels)
<fche> orthogonal
<bendlas> I'll consider pushing for an additional `debugInfo` output in our default kernel build, if the regular user won't be affected
<fche> dwarf debuginfo carries no runtime overhead at all, by design, for a good compiler like gcc
wcohen has joined #systemtap
<bendlas> yeah, I remember having heard that about dwarf
<bendlas> so, fedora also provides extra debuginfo downloads for most (all?) of userland?
<fche> yup.
<bendlas> that's pretty cool, we should also provide that in nixos
<fche> if you build your own kernel, to be systemtap-friendly, consider also carrying
drsmith has left #systemtap [#systemtap]
<bendlas> we just got userland hardening merged, so it might be a good time to think about debuginfo
<fche> unfortunately linus was misguided about gcc -g -fvar-tracking-assignments two years ago, and it stuck :(
<bendlas> he didn't want the debuginfo bloat?
<fche> it wasn't bloat
<fche> that URL explains well
<bendlas> yep
<bendlas> but it was about the size?
<fche> the gcc VTA issue was about concern about a shortlived gcc bug wherein that DWARF option could break normal code generation
<fche> it was fixed in a day ... but VTA was not reenabled in the mainstream kernel :-(
<bendlas> i see
<fche> anyway
* fche must step out, someone else can probably help
<bendlas> thanks for the tips
<fche> np
<bendlas> btw, https://sourceware.org/systemtap/wiki/SystemTapWithSelfBuiltKernel states, that it cannot handle split debug info
<bendlas> I suppose, that's outdated info?
<jistone> bendlas, no, that's true
<jistone> split, as in *.dwo files
<jistone> it certainly supports split the way distros like fedora do it
<bendlas> i see
alkgayua has quit [Quit: Page closed]