lkml.org 
[lkml]   [2008]   [Jan]   [19]   [last100]   RSS Feed
Views: [more markup]  [less markup]  [headers]  [forward] 
 
Messages in this thread
    /
    DateSat, 19 Jan 2008 06:33:30 +0100
    FromAndi Kleen <>
    SubjectRe: [Announce] Development release 0.1 of the LatencyTOP tool
    > another thing that the current profiling can't do, is to show what the 
    > system is doing
    > when it hits the latency.. so someone calling fsync() will show up in the 
    > waiting for
    > IO function, but not that it was due to an fsync().
    
    Hmm so how about extending oprofile to always log the syscall number
    in the event logs (can be gotten from top of stack). I think given
    that you could reconstruct that data in the userland at least
    for single threads (not for work done on behalf of them in other
    threads; but I'm not sure you tried to solve that problem at all)
    
    The advantage is that it would be an generic mechanism that would work
    for all types of profiling.
    
    -Andi
    > 
    
    
    
    \
     
     \ /
      Last update: 2008-01-19 06:33    [from the cache]
    ©2003-2008