[Vm-dev] Is it possible to suspend the garbage collector?

[Max Leske]

> On 12 Jan 2016, at 19:48, vm-dev-request at lists.squeakfoundation.org wrote:
> 
>> 
>> On 12-01-2016, at 8:39 AM, Max Leske <maxleske at gmail.com <mailto:maxleske at gmail.com>> wrote:
>> 
>> Why do I see this behaviour? In Cog I see the expected outcome (free space big, young space small) even without triggering an incremental GC manually.
> 
> Not sure about details but do remember that Spur uses a fairly different object space arrangement, so if that’s what you’e using John’s advice may no longer be very relevant.

I’m aware of that. My problem (for the moment at least) only concerns V3 (Squeak VM 4.0.3).

> 
>> 
>> 
>> 2. There’s also still the question about where those 500ms disappear to (see last e-mail), which are unaccounted for in the MessageTally output.
> 
> The tally code has never been particularly good at handling long running primitives. I *think* there is better profiling code somewhere that Andreas developed for Qwaq but IIRC it is not part of the current system.

Interesting.

I wondered about the primitives and went to give MessageTally a closer look. I noticed that I was not taking into account other running processes and voilà:

Process: other processes
--------------------------------
84.5% {2389ms} InputEventPollingFetcher(InputEventFetcher)>>eventLoop
  84.5% {2389ms} InputEventPollingFetcher>>waitForInput
    84.5% {2389ms} Delay>>wait

That probably explains where those 500ms I was talking about went (I was loading a larger segment here, so the process took longer to run).

I’ve “fixed” that for now by running the loader in a higher priority process.

Thanks for the hint.

Cheers,
Max

> 
> tim


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.squeakfoundation.org/pipermail/vm-dev/attachments/20160114/b8258d70/attachment.htm