... rounds, I propose these markers for 2.6.26. They > include work done ... a 5 bytes callout that is patched to a NOP, and _maybe_ a ... tongue-in-cheek commit below. Ingo ---------------------> Subject: sched: reintroduce markers From: Ingo Molnar < ...

linux-kernel - Ingo Molnar - Mar 27 2008 - 11:40

... rounds, I propose these markers for 2.6.26. They > > include work done ... a 5 > bytes callout that is patched to a NOP, and _maybe_ a ... this is without immediate values) cmpb $0, __mark_ctx_switch.33881+8 jne .L2164 corresponding ...

linux-kernel - Mathieu Desnoyers - Mar 27 2008 - 16:39

... a few RFC rounds, I propose these markers for 2.6.26. They include > work done after comments from the ... have > been used by the LTTng project for about 2 years. Not in this patch, but I noticed: #define __trace_mark(name, call_private, format, args ...

linux-kernel - Denys Vlasenko - Mar 27 2008 - 20:01

... rounds, I propose these markers for 2.6.26. They >> include work done ... Most of them have been used by the LTTng project for about 2 years. > > [...] > so we optimistically used ... in ftrace (see sched-devel.git) for > the scheduler, and i was shocked ...

linux-kernel - Frank Ch. Eigler - Mar 27 2008 - 17:49

... /sched.o > > 224 bytes added for 6 ftrace_*(). This is partly ... the patch below, which removes the ad-hoc tracepoints. Ingo ------------------------> Subject: ... p->pid, p->se.avg_overlap, 0); -#endif -} - -void ftrace_all_fair_tasks(void *__rq ...

linux-kernel - Ingo Molnar - Mar 28 2008 - 05:43

* Ingo Molnar wrote: > if you want to compare apples to apples, try the patch below, which > removes the ad-hoc tracepoints. ... new version of that is in sched-devel.git. Ingo -- To unsubscribe from this list: send the line ...

linux-kernel - Ingo Molnar - Mar 28 2008 - 07:22

... /sched.o > > > > 224 bytes added for 6 ftrace_*(). This is partly due ... apples to apples, try the patch below, which > removes the ad- ... introduction you quoted. Mathieu > Ingo > > ------------------------> > Subject: no: ad hoc ftrace points ...

linux-kernel - Mathieu Desnoyers - Mar 28 2008 - 07:38

... 's trivial to use it, very powerful and we've only begun exposing it towards users. I see no good reason why we'd patch any marker into the kernel - it's a maintenance cost from that point on. so yes, my argument is: tens of thousands of lightweight ...

linux-kernel - Ingo Molnar - Mar 28 2008 - 09:33

... .before > 37930 7134 384 45448 b188 sched.o.after > > 288 bytes for only 3 markers in the scheduler - that's 96 bytes ... > A classic lose-lose situation, so let's apply this patch. Not! > > NACK-ed-by: Ingo Molnar I ...

linux-kernel - KOSAKI Motohiro - Mar 27 2008 - 13:08

... classic lose-lose situation, so let's apply this patch. Not! > > > > NACK-ed-by: Ingo Molnar for dependencies: kernel-debuginfo-common x86_64 2.6.25-0.163.rc7.git1.fc9 ...

linux-kernel - Ingo Molnar - Mar 28 2008 - 06:15

... git1.fc9 > development-debuginfo 198 M > Installing for dependencies: > kernel-debuginfo-common x86_64 2.6.25-0.163.rc7.git1. ... And many other compressed filesystem can be used for saving disk space. Best regards, -- Masami ...

linux-kernel - Masami Hiramatsu - Mar 28 2008 - 09:34

* Frank Ch. Eigler wrote: > > >> > NACK-ed-by: Ingo Molnar > >> [...] > >> I wonder why nobody doesn't fix this serious problem until now. [...] > > Frankly, the "problem" has not yet been established. Ingo

linux-kernel - Ingo Molnar - Mar 28 2008 - 10:18

* Frank Ch. Eigler wrote: > > [...] furthermore, and because it's Friday again, let me remind > > folks that SystemTap has an even more significant bloat problem: the > > fact that it needs a huge download: [...the dwarf

linux-kernel - Ingo Molnar - Mar 28 2008 - 10:41

... This complaint is particularly rich, seeing how you constantly try to shoot down patches that would reduce systemtap's need for this, such as markers. And no, dyn-ftrace is not a complete substitute. ...

linux-kernel - Frank Ch. Eigler - Mar 28 2008 - 11:31

... of the critical path. With -O2 : mov $0,%al movq %rsi, 1912(%rbx) movq ... fc ff ff jmpq 4f6 For an added 31 bytes. Total size ... no good reason why we'd patch any marker into > the kernel - it ...

linux-kernel - Mathieu Desnoyers - Mar 29 2008 - 13:16

... DWARF data to eliminate more duplication - support for inserting probes only on function entry/exit ... information will still need to be around for those who need to put probes ... and to pull approximately any data out. > for heaven's sake, we can have 3 ...

linux-kernel - Frank Ch. Eigler - Mar 28 2008 - 09:40

... an additional 850 MB. > > Puh-lease ... ] > > Thank for Moore's law and all storage engineers ... other compressed filesystem can be used for saving disk space. This is not a valid technical reason for creating bloatware. Bloatware's main problem ...

linux-kernel - Denys Vlasenko - Mar 31 2008 - 21:43

... many other compressed filesystem can be used for saving disk space. > > This is not a valid technical reason for creating bloatware. > > Bloatware's main problem is ... If it is a program code, you're right. However, the debuginfo is just a ...

linux-kernel - Masami Hiramatsu - Apr 1 2008 - 10:30