Changes between Version 9 and Version 10 of RTSsummaryEvents

Show
Ignore:
Timestamp:
12/21/11 07:53:44 (2 years ago)
Author:
MikolajKonarski (IP: 95.160.111.162)
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • RTSsummaryEvents

    v9 v10  
    1313}}} 
    1414 
    15 An event with a summary of all copying done, after the end of each GC,  
     15An event with a summary of all copying done, after the end of each GC, probably. 
    1616 
    1717{{{ 
     
    1919}}} 
    2020 
    21 Probably a separate event for that, instead of trying to calculate it from all events. 
    22 Perhaps the event should be emitted during major GC (the docs explain the "n samples"  
    23 saying "only checked during major garbage collections"). 
     21Either a separate event for that, perhaps emitted only after major GC, when we know how much memory 
     22is really used by the program (the docs explain the "n samples" saying "only checked during major garbage collections").  
     23Or try to calculate it from all events. in particular memory deallocation events or GC memory freeing events. 
    2424 
    2525{{{ 
     
    2727}}} 
    2828              
    29 We also need an event for slop, probably emitted rarely. 
     29We also need an extra event for slop, probably emitted rarely. 
    3030 
    3131{{{ 
     
    3939}}} 
    4040                   
    41 Ask JaffaCake the following formula means: 
     41Ask JaffaCake what the following formula means: 
    4242 
    4343{{{ 
     
    5252  Gen  1         5 colls,     0 par    0.07s    0.07s     0.0149s    0.0386s 
    5353}}} 
    54                                
     54 
     55Ask JaffaCake if RequestParGC is enough to tell seq/par and why some StartGC 
     56follow neither RequestParGC nor RequestSeqGC (on the same capability, at least). 
    5557Split the current GC events into generations and into seq/par 
    56 (if RequestParGC is not enough to tell seq/par). We don't want to report 
    57 the CPU time, only the elapsed time, and that's fine. 
     58(if RequestParGC is not enough to tell seq/par). Note that we don't want  
     59to report the CPU time, only the elapsed time, and that's fine. 
    5860With the current GC events only summarized information is available. 
    5961