This is the mail archive of the systemtap@sourceware.org mailing list for the systemtap project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

How to get the detailed kernel stack trace if specified fuction takes too long to finish?


Hi all,

Probably it's a newbie question, to be honest, i am new to SystemTap,
currently i want to capture the kernel stack trace if "sys_write" runs
too long, if that i could know where it sucks.

My OS:2.6.32-220.el6.x86_64
# stap -V
Systemtap translator/driver (version 1.6/0.152 non-git sources)

My tried script:
global i=0
global tb=0
global te=0
probe kernel.function("sys_write") {
  tb=gettimeofday_us();
}
probe kernel.function("sys_write").return {
  te=gettimeofday_us();
  if(te-tb > 100000) {
   i++;
   printf("time_us: %d\n", te-tb);
   print_backtrace();
   if(i > 15) {
    exit();
   }
  }
}

# stap -x <pid> xxx.stp
time_us: 226795
Returning from:  0xffffffff81176f90 : sys_write+0x0/0x90 [kernel]
Returning to  :  0xffffffff8100b0f2 : system_call_fastpath+0x16/0x1b [kernel]
time_us: 101400
Returning from:  0xffffffff81176f90 : sys_write+0x0/0x90 [kernel]
Returning to  :  0xffffffff8100b0f2 : system_call_fastpath+0x16/0x1b [kernel]
time_us: 527853
Returning from:  0xffffffff81176f90 : sys_write+0x0/0x90 [kernel]
Returning to  :  0xffffffff8100b0f2 : system_call_fastpath+0x16/0x1b [kernel]
time_us: 133948
Returning from:  0xffffffff81176f90 : sys_write+0x0/0x90 [kernel]
Returning to  :  0xffffffff8100b0f2 : system_call_fastpath+0x16/0x1b [kernel]

It doesn't show my "expected" kernel stack trace,e.g. file system or
block layer related stack trace. How to archive that? Any pointers
will be highly appreciated!

Best,
Liang


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]