NetBSD-Bugs archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

port-amd64/50979: dtrace -c leads to 100% cpu usage



>Number:         50979
>Category:       port-amd64
>Synopsis:       dtrace -c leads to 100% cpu usage
>Confidential:   no
>Severity:       non-critical
>Priority:       medium
>Responsible:    port-amd64-maintainer
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sat Mar 19 11:15:00 +0000 2016
>Originator:     Patrick Welche
>Release:        7.99.26/amd64
>Organization:
>Environment:
>Description:
As per http://mail-index.netbsd.org/netbsd-users/2016/03/11/msg018156.html

# cat foo.d
proc:::exec
{
        trace("hello");
        exit(0);
}
# dtrace -s foo.d                           Then type "ls" in another xterm
dtrace: script 'foo.d' matched 1 probe
dtrace: buffer size lowered to 2m
CPU     ID                    FUNCTION:NAME
  2      6                            :exec   hello

# dtrace -s foo.d -c ls

and in other xterm top shows:

  PID USERNAME PRI NICE   SIZE   RES STATE      TIME   WCPU    CPU COMMAND
  738 root      29    0    52M   11M CPU/3      0:54 99.17% 92.53% dtrace

Running "ls" in another xterm just in case doesn't stop it. Finally:

^Csorry, pid 554 was killed: orphaned traced process

I was expecting "-c ls" to run /bin/ls after the proc:::exec probe was set...

>How-To-Repeat:
see above
>Fix:



Home | Main Index | Thread Index | Old Index