#systemtap on 2018-03-02 — irc logs at freenode.irclog.whitequark.org

2015-11-12 23:18 fche changed the topic of #systemtap to: http://sourceware.org/systemtap; email systemtap@sourceware.org if answers here not timely, conversations may be logged

00:53 orivej has quit [Ping timeout: 256 seconds]

01:15 gromero has quit [Ping timeout: 245 seconds]

01:29 gromero has joined #systemtap

01:41 hpt has joined #systemtap

01:45 gromero has quit [Ping timeout: 240 seconds]

02:15 orivej has joined #systemtap

02:33 orivej has quit [Ping timeout: 240 seconds]

02:49 scox has joined #systemtap

03:56 sanoj has joined #systemtap

04:10 prefork has quit [Read error: Connection reset by peer]

04:11 ggherdov` has quit [Read error: Connection reset by peer]

04:11 DuncanT has quit [Ping timeout: 256 seconds]

04:12 prefork has joined #systemtap

04:13 gila has quit [Ping timeout: 260 seconds]

04:16 gila has joined #systemtap

04:40 ggherdov` has joined #systemtap

04:40 ggherdov` has quit [Excess Flood]

04:50 DuncanT has joined #systemtap

04:51 ggherdov` has joined #systemtap

04:51 ggherdov` has quit [Excess Flood]

05:02 ggherdov` has joined #systemtap

05:02 ggherdov` has quit [Excess Flood]

05:13 ggherdov` has joined #systemtap

05:14 ggherdov` has quit [Excess Flood]

05:22 naveen has joined #systemtap

05:30 slowfranklin has joined #systemtap

06:15 ggherdov` has joined #systemtap

06:15 ggherdov` has quit [Excess Flood]

06:44 naveen has quit [Quit: WeeChat 2.0.1]

06:52 slowfranklin has quit [Quit: slowfranklin]

07:03 slowfranklin has joined #systemtap

07:04 slowfranklin has quit [Client Quit]

07:07 slowfranklin has joined #systemtap

07:15 acaringi__ has joined #systemtap

07:16 ggherdov` has joined #systemtap

07:16 ggherdov` has quit [Excess Flood]

07:18 acaringi has quit [Ping timeout: 256 seconds]

07:18 acaringi has joined #systemtap

07:21 acaringi__ has quit [Ping timeout: 248 seconds]

08:17 ggherdov` has joined #systemtap

08:17 ggherdov` has quit [Excess Flood]

08:35 naveen has joined #systemtap

08:38 naveen_ has joined #systemtap

08:39 naveen has quit [Ping timeout: 255 seconds]

08:44 naveen_ has quit [Quit: WeeChat 2.0.1]

08:56 hpt has quit [Quit: Lost terminal]

09:05 naveen has joined #systemtap

09:18 ggherdov` has joined #systemtap

09:19 ggherdov` has quit [Excess Flood]

09:30 mjw has joined #systemtap

09:37 RenHoekNL has joined #systemtap

09:38 <RenHoekNL> Hi I have a question. Does the kernel keep track of how many cycles a process has used (user and kernel space) that I can print when a process ends? Or do I have to keep track of it myself with timer.profile? If so, isn't that a lot of overhead?

09:42 <izabera> RenHoekNL: did you try perf stat?

09:43 <RenHoekNL> I'm new to systemtap, what exactly should I be looking for?

09:43 <izabera> it's not a systemtap thing

09:44 <izabera> your distro probably has a package called perf-tools or similar

09:45 <RenHoekNL> Oh yes, I looked at that.. problem is I need to analyse a pipeline that spawns multiple other process and see where the bottlenecks are.. so it's not just the performance of one process.. I thought systemtap would be the way to go since I can hook process creating and termination

09:45 <RenHoekNL> I don't think perf would be very useful here.. or am I wrong?

09:56 gila has quit [Quit: Textual IRC Client: www.textualapp.com]

10:20 ggherdov` has joined #systemtap

10:20 ggherdov` has quit [Excess Flood]

11:19 RenHoekNL has quit [Ping timeout: 260 seconds]

11:32 sanoj has quit [Ping timeout: 245 seconds]

12:21 ggherdov` has joined #systemtap

12:21 ggherdov` has quit [Excess Flood]

12:53 orivej has joined #systemtap

12:58 gromero has joined #systemtap

13:28 mbenitez has joined #systemtap

13:28 mbenitez has quit [Changing host]

13:47 wcohen has quit [Remote host closed the connection]

14:22 ggherdov` has joined #systemtap

14:22 ggherdov` has quit [Excess Flood]

14:37 wcohen has joined #systemtap

14:43 naveen has quit [Ping timeout: 245 seconds]

14:51 naveen has joined #systemtap

15:05 tromey has joined #systemtap

15:32 naveen has quit [Quit: WeeChat 2.0.1]

15:45 naveen has joined #systemtap

16:23 ggherdov` has joined #systemtap

16:24 ggherdov` has quit [Excess Flood]

17:09 tromey has quit [Read error: Connection reset by peer]

17:13 ggherdov` has joined #systemtap

17:13 ggherdov` has quit [Excess Flood]

17:23 naveen has quit [Ping timeout: 276 seconds]

17:24 naveen has joined #systemtap

17:52 slowfranklin has quit [Quit: slowfranklin]

18:04 slowfranklin has joined #systemtap

18:25 ggherdov` has joined #systemtap

18:25 ggherdov` has quit [Excess Flood]

18:27 slowfranklin has quit [Quit: slowfranklin]

18:44 slowfranklin has joined #systemtap

18:46 slowfranklin has quit [Client Quit]

19:00 zodbot has quit [Disconnected by services]

19:11 zodbot has joined #systemtap

19:14 ggherdov` has joined #systemtap

19:15 ggherdov` has quit [Excess Flood]

19:23 tromey has joined #systemtap

19:25 naveen has quit [Quit: WeeChat 2.0.1]

19:33 naveen has joined #systemtap

19:35 mjw has quit [Quit: Leaving]

20:03 irker230 has joined #systemtap

20:03 <irker230> systemtap: dsmith systemtap.git:refs/heads/master * release-3.2-121-g2f6fcfc / tapset/linux/ipmib.stp tapset/linux/netfilter.stp tapset/linux/networking.stp tapset/linux/tcp.stp tapset/linux/tcpmib.stp: Fix BZ1546179 by being careful when accessing sk_buff structures. http://tinyurl.com/ya4nexv5

20:23 eklitzke has quit [Quit: bye]

20:24 eklitzke has joined #systemtap

20:26 ggherdov` has joined #systemtap

20:26 ggherdov` has quit [Excess Flood]

20:37 mjw has joined #systemtap

20:37 slowfranklin has joined #systemtap

20:39 naveen has quit [Quit: WeeChat 2.0.1]

20:51 slowfranklin has quit [Quit: slowfranklin]

21:00 slowfranklin has joined #systemtap

21:16 ggherdov` has joined #systemtap

21:16 ggherdov` has quit [Excess Flood]

22:12 mbenitez has quit [Quit: Leaving]

22:19 tromey has quit [Quit: ERC (IRC client for Emacs 26.0.91)]

22:20 wcohen has quit [Ping timeout: 265 seconds]

22:21 slowfranklin has quit [Quit: slowfranklin]

22:27 gromero has quit [Ping timeout: 240 seconds]

22:28 ggherdov` has joined #systemtap

22:28 ggherdov` has quit [Excess Flood]

23:13 wcohen has joined #systemtap

23:17 ggherdov` has joined #systemtap

23:17 ggherdov` has quit [Excess Flood]

23:19 amerey has quit [Quit: Leaving]

23:29 extra has joined #systemtap

23:43 <extra> SystemTap and gdb should be able to be used at the same time, right?

23:43 <fche> yup

23:44 <fche> one place where conflicts have occurred is when a systemtap userspace probe and a gdb session are targeting the same process

23:44 <fche> sometimes ISTR gdb getting confused about extra sigtraps

23:44 <eklitzke> yeah i've seen gdb get stuck

23:45 <extra> alright... yeah, gdb seems to bite the dust after using systemtap with -x <pid of process being debugged>

23:45 <eklitzke> when i used stap on a process gdb was already attached to it got stuck, but when i gdb attached to a process that was already stapped it's fine

23:47 <extra> https://pastebin.com/zKK94344

23:47 <extra> interesting... I wonder why that is - does SystemTap use ptrace behind the scenes?

23:48 <fche> nope

23:48 <fche> hmm. wait yes, very briefly, to synchronize startup

23:48 <eklitzke> you can't have two processes ptrace attached at the same time

23:48 <eklitzke> so how does that work?

23:49 <fche> hm, actually I don't think we do that any more;

23:50 <fche> WORKAROUND_BZ467568 ... ancient thing that's not turned o

23:50 <fche> on

23:52 <fche> hm, looking at that pastebin, can't think of a reason for that off the top of my head

23:52 <fche> no such process -- did it die?

23:55 <extra> no, it is still running

23:58 <extra> the process was stopped at a breakpoint when I started stap, though... I wonder what happens when stap is started while it's running