C# Memory Leak

[Chris Mullins [MVP - C#]]

As someone who regularly gives presentations on Garbage Collection, and
application profiling and tuning, I found this Slashdot article the other
day to be really funny:

http://developers.slashdot.org/article.pl?sid=07/11/17/0552247

I especially loved the slashdot spin that it is a bug in C# or in the CLR.

(I know, I know, but I like the title of the thread. How else to make sure
everyone reads this? Plus, it's a slow day today. Everyone must be out on
vacation...)

 

[Peter Duniho]

As someone who regularly gives presentations on Garbage Collection, and
application profiling and tuning, I found this Slashdot article the other
day to be really funny:

http://developers.slashdot.org/article.pl?sid=07/11/17/0552247

I especially loved the slashdot spin that it is a bug in C# or in the CLR.

I thought it was funny too, but it's my recollection that, in a rare
moment of Slashdot insightfulness, the early comments mostly were of
the "RTFA" type, pointing out how the referenced blog entry isn't about
a bug in C# or .NET at all, but rather just an ad for a profiling tool
to help a developer find their _own_ bugs.

In fact, I was a little surprised at the general lack of anti-MS
sentiment. Granted, I only read the first dozen or so comments; it
wouldn't surprise me to find that the comments eventually degenerated
into the usual Microsoft bashing.

Pete

 

[Hilton]

In fact, I was a little surprised at the general lack of anti-MS
sentiment. Granted, I only read the first dozen or so comments; it
wouldn't surprise me to find that the comments eventually degenerated into
the usual Microsoft bashing.

Pete, I find the overall tone on slashdot to be anti-MS so this doesn't
surprise me. Pity, I really enjoy reading slashdot as they seem to have
tech news days or even weeks before the (other) media folks pick up a story.

Hilton

 

[Guest]

When you get to the article itself we find that this is a very old and common
problem (the cause of which is what is between the computer and the chair):

"We profiled the memory usage and saw the obstacle list blowing up. How
could this be? We called "delete" on those old obstacles! To our amazement,
it was only minutes before we realized that our list of detected obstacles
was never getting garbage collected. Though we thought we had cleared all
references to old entries in the list, because the objects were still
registered as subscribers to an event, they were never getting deleted."

----Peter
"Inside every large program, there is a small program trying to get out."
http://www.eggheadcafe.com
http://petesbloggerama.blogspot.com
http://www.blogmetafinder.com

 

[Vapordan]

I got to reading this and was asking myself why don't they just attach the
debugger to the car when it slowing down teheheheh

--
Regards,
Alvin Bruney [MVP ASP.NET]

[Shameless Author plug]
The O.W.C. Black Book, 2nd Edition
Exclusively on www.lulu.com/owc $19.99

 

[Chris Mullins [MVP - C#]]

Yea, that's the whole point.

It's not a platform bug at all - not in any possible way. It is, 100%, a bug
in the application code, and one that the development team was very open
about.

I just loved the /. spin, which implied it was the .Net platform that not
only the cause of this bug, but the reason the team failed the DARPA
challenge, and (quite possibly) the root of all evil in general.

I do alot of presentations on GC, and then spend lots of time with people
tracking down memory leaks in managed code. That really made this article
funny to me...

 

[Wingot]

-----Original Message-----
From: Hilton [mailto:[email protected]]
Posted At: Wednesday, 21 November 2007 9:14 AM
Posted To: microsoft.public.dotnet.languages.csharp
Conversation: C# Memory Leak
Subject: Re: C# Memory Leak

In fact, I was a little surprised at the general lack of anti-MS
sentiment. Granted, I only read the first dozen or so comments; it
wouldn't surprise me to find that the comments eventually

degenerated

into
the usual Microsoft bashing.

Pete, I find the overall tone on slashdot to be anti-MS so this doesn't
surprise me. Pity, I really enjoy reading slashdot as they seem to
have
tech news days or even weeks before the (other) media folks pick up a
story.

Hilton

As an example:

That makes sense: Since chairs are larger than darts, you have a much

greater chance to hit the bugs.

 

[Michael D. Ober]

Yea, that's the whole point.

It's not a platform bug at all - not in any possible way. It is, 100%, a
bug in the application code, and one that the development team was very
open about.

I just loved the /. spin, which implied it was the .Net platform that not
only the cause of this bug, but the reason the team failed the DARPA
challenge, and (quite possibly) the root of all evil in general.

I do alot of presentations on GC, and then spend lots of time with people
tracking down memory leaks in managed code. That really made this article
funny to me...

This article and the problems they had with their car go to show that you
can write crap in any language.

Mike Ober.