[racket] cpu usage of DrRacket

Matthew and Robby,

I just checked the memory usage of the windows 7 64 bit version of DrRacket
(version 5.2.0.7--2011-12-23(188f868/a) [3m].) Despite the obscurity of the
problem, the fix was made quickly. It works great. I checked both single and
twin CPU usage.

The quick responses of PLT keeps amazing me. I can remember times (long ago)
when it sometimes took half a year to convince and to prove to an expensive
company that its system had a bug, let alone the time required to fix it. I
know you are a team rather than a company, but nevertheless ...; I hope your
funds for maintenance and further development of Racket and more profound
scientific studies in informatics are not in danger. I am sure that,
whatever expenses may be involved, the benefit for research and the world
certainly exceeds the costs.

Thanks very much.
Jos

-----Original Message-----
From: Matthew Flatt [mailto:mflatt at cs.utah.edu] 
Sent: miércoles, 21 de diciembre de 2011 19:36
To: Jos Koot
Cc: users at racket-lang.org
Subject: Re: [racket] cpu usage of DrRacket

We tracked this down to a Win64-specific problem with the recent JIT
switch from x87 to SSE for floating point --- now fixed.

Thanks for the report!

At Wed, 14 Dec 2011 13:52:05 +0100, "Jos Koot" wrote:
> I see that after starting up, DrRacket constantly uses between 40 and 60 %
> of my two CPUs, even with empty definitions window and before doing
anything
> by myself.
>  
> I use:
> Welcome to DrRacket, version 5.2.0.6--2011-12-10(6fae172/a) [3m]
> without debugging and without preserving stacktrace.
> Language determined by #lang line.
> Disabling online expansion does not decrease CPU usage.
>  
> What may be the reason for DrRacket to use so much CPU at moments I would
> expect it to be idle?
> Thanks, Jos