[Bug translator/16764] New: registration error (rc -22): probe kernel.function("context_switch@kernel/sched/core.c:2105")

[Bug translator/16764] New: registration error (rc -22): probe kernel.function("context_switch@kernel/sched/core.c:2105")

[fahadaliarshad at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16764

            Bug ID: 16764
           Summary: registration error (rc -22): probe
                    kernel.function("context_switch@kernel/sched/core.c:21
                    05")
           Product: systemtap
           Version: unspecified
            Status: NEW
          Severity: normal
          Priority: P2
         Component: translator
          Assignee: systemtap at sourceware dot org
          Reporter: fahadaliarshad at gmail dot com

Created attachment 7495
  --> https://sourceware.org/bugzilla/attachment.cgi?id=7495&action=edit
stap -vvv -p2 -e 'probe kernel.function("context_switch") {printf("hit ")}'

I get the registration error (rc -22) when probing a scheduler
function("context_switch") of the latest kernel(3.13.6-100.fc19.x86_64).
Looking at the stap -vvv -p2,-p3 and DDEBUG_SYMBOLS, I am not able to pin the
problem. I also tried reinstalling the elfutils(0.156) but no success. I am not
sure whether it an OFFSET/reloc problem or there is something much deeper.

$ sudo stap -v --poison-cache -e 'probe kernel.function("context_switch")
{printf("hit ")}'
Pass 1: parsed user script and 149 library script(s) using
374144virt/189036res/3088shr/189200data kb, in 1360usr/140sys/1504real ms.
Pass 2: analyzed script: 1 probe(s), 0 function(s), 0 embed(s), 0 global(s)
using 576788virt/337180res/105248shr/235180data kb, in 1010usr/110sys/1111real
ms.
Pass 3: translated to C into
"/tmp/stapaGyTsS/stap_d231a2ffb286405b384dfd0e8e6fb46e_950_src.c" using
576788virt/337316res/105384shr/235180data kb, in 0usr/0sys/4real ms.
Pass 4: compiled C into "stap_d231a2ffb286405b384dfd0e8e6fb46e_950.ko" in
6900usr/1100sys/8148real ms.
Pass 5: starting run.
WARNING: probe kernel.function("context_switch@kernel/sched/core.c:2105")
(address 0xffffffff81683d78) registration error (rc -22)
^CPass 5: run completed in 10usr/20sys/4223real ms.

Any quick help will be greatly appreciated.
Thanks
Fahad.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug translator/16764] registration error (rc -22): probe kernel.function("context_switch@kernel/sched/core.c:2105")

[fahadaliarshad at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16764

--- Comment #1 from fahadaliarshad at gmail dot com ---
Created attachment 7496
  --> https://sourceware.org/bugzilla/attachment.cgi?id=7496&action=edit
stap -vvv -p3 -e 'probe kernel.function("context_switch") {printf("hit ")}'

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug translator/16764] registration error (rc -22): probe kernel.function("context_switch@kernel/sched/core.c:2105")

[fahadaliarshad at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16764

--- Comment #2 from fahadaliarshad at gmail dot com ---
Created attachment 7497
  --> https://sourceware.org/bugzilla/attachment.cgi?id=7497&action=edit
stap -vvv -DDEBUG_SYMBOLS -e 'probe kernel.function("context_switch")
{printf("hit ")}'

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug translator/16764] registration error (rc -22): probe kernel.function("context_switch@kernel/sched/core.c:2105")

[fahadaliarshad at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16764

fahadaliarshad at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |fahadaliarshad at gmail dot com

--- Comment #3 from fahadaliarshad at gmail dot com ---
Created attachment 7498
  --> https://sourceware.org/bugzilla/attachment.cgi?id=7498&action=edit
stap-report.txt

And here is the stap-report.txt

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug translator/16764] registration error (rc -22): probe kernel.function("context_switch@kernel/sched/core.c:2105")

[fche at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16764

--- Comment #4 from Frank Ch. Eigler <fche at redhat dot com> ---
Thank you for your high-quality problem report.  I can reproduce it
immediately on my f19 x86-64 workstation.

The error value is coming from the kernel, and represents some
sort of problem the kprobes layer.  Disassembly of the region shows
this instruction within the inlined muck that makes up the
context_switch / __schedule clique.

ffffffff81683d78:       0f 1f 44 00 00          nopl   0x0(%rax,%rax,1)

I'm not sure why the kernel is rejecting a kprobe on that nop; it could
be a kernel-disassembler problem or some sort of blacklist.  It does not
appear to represent a systemtap bug, so closing as INVALID.


Luckily, approximately the same spot can be probed via tracepoints:

probe kernel.trace("sched_switch") { }

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug translator/16764] registration error (rc -22): probe kernel.function("context_switch@kernel/sched/core.c:2105")

[fche at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16764

Frank Ch. Eigler <fche at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |fche at redhat dot com
         Resolution|---                         |WONTFIX

--- Comment #5 from Frank Ch. Eigler <fche at redhat dot com> ---
(Does not appear to be a systemtap-induced problem.)

-- 
You are receiving this mail because:
You are the assignee for the bug.

[Bug translator/16764] registration error (rc -22): probe kernel.function("context_switch@kernel/sched/core.c:2105")

[fahadaliarshad at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=16764

--- Comment #6 from fahadaliarshad at gmail dot com ---
Thank you for the quick reply. 

For the time being, I will use 'probe kernel.trace("sched_switch")'.

-- 
You are receiving this mail because:
You are the assignee for the bug.