Skip to contentSkip to conversation

Trouble parsing trace events

DISCUSSION Arun Sharma 2 participants 3 messages Nov 23, 2011

lore
Share & tools
lore
#0 lore raw

I'm using a 3.2-rc1 based kernel and perf userspace built from tip/auto-latest (commit id e471db8) + Andrew Vagin's sleep profiling patches.

# perf record -agT -e sched:sched_switch -F 100 -- sleep 3
[ perf record: Woken up 9 times to write data ]
[ perf record: Captured and wrote 4.044 MB perf.data (~176664 samples) ]
# perf script
  Warning: Error: expected type 5 but read 4
  Warning: Error: expected type 4 but read 0
  Fatal: bad op token {

Is there a known incompatibility between kernel and userspace that could explain this?

I don't see this problem with other trace points (eg: kmem:kmalloc).

-Arun

#1 lore raw

Hello Arun,

This bug was fixed already, but the commit isn't in Linus's tree yet.
http://git.kernel.org/?p=linux/kernel/git/tip/tip.git;a=commit;h=49908a1b25d448d68fd26faca260e1850201575f

Quoted 17 lines · Arun Sharma

I'm using a 3.2-rc1 based kernel and perf userspace built from tip/auto-latest (commit id e471db8) + Andrew Vagin's sleep profiling patches.

# perf record -agT -e sched:sched_switch -F 100 -- sleep 3
[ perf record: Woken up 9 times to write data ]
[ perf record: Captured and wrote 4.044 MB perf.data (~176664 samples) ]
# perf script
  Warning: Error: expected type 5 but read 4
  Warning: Error: expected type 4 but read 0
  Fatal: bad op token {

Is there a known incompatibility between kernel and userspace that could explain this?

I don't see this problem with other trace points (eg: kmem:kmalloc).

-Arun