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]

[Bug translator/4066] New: hist_linear() with large H value crashes system


I was writing a simple script that uses hist_linear() and decided to pass it a
ridiculously large H value to see what happens.  To my surprise, it crashed the
system.  

Here's the script:

global reads
probe netdev.receive { reads <<< length }
probe end { print (@hist_linear(reads, 0, 300000000000, 50)) }

I tried the script on ppc64 and x86 running SLES10 SP1 and x86_64 running RHEL5
RC1 and the latest FC5 kernel.  I tried the SLES10 SP1, RHEL5 RC1 and CVS
versions of systemtap.  All crashed.  I suspect there's a problem in
_stp_stat_init(). More details to follow.

-- 
           Summary: hist_linear() with large H value crashes system
           Product: systemtap
           Version: unspecified
            Status: NEW
          Severity: critical
          Priority: P2
         Component: translator
        AssignedTo: systemtap at sources dot redhat dot com
        ReportedBy: mmlnx at us dot ibm dot com


http://sourceware.org/bugzilla/show_bug.cgi?id=4066

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


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