Yup, that totally worked on a Mac too. That's WEIRD. Thanks Kas! -Mike On Sun, Oct 12, 2008 at 4:41 PM, mike clemow wrote:

Kas,

...

I really think this used to work. To make matters worse this (assuming the same function) seems to work ok;

//============================= spork ~ dummy() @=> Shred foo; hour => now; //===============================

So, if we declare the Shred reference at the same time it doesn't cause an error?? This might be the work-around for the time being...

Thanks! -Mike

On Sun, Oct 12, 2008 at 4:17 PM, Kassen wrote:

...

Hey mike!

...

I've decided to give "bus error" a proper kenning: "Clemow's Bane."

Ow, come on, the bus errors like you, they just want to play.

...

Kas, do you think that this is related to the bus errors we were getting assigning objects in that GrainPattern class?

I wouldn't be surprised. Something seems to be rather wrong with assignment these days. This kind of structure (assigning a sporked function to a Shred object) did work with that syntax in the past so it could be the new reference counting being over-active again.

I just simplified your code to make the real culprit come out;

//================8<================== fun void dummy() { while(1) { second => now; } }

Shred foo; //this runs //spork ~ dummy();

//I get a segmentation fault at this on Ubuntu spork ~ dummy() @=> foo;

hour => now; //==============8<==================

I really think this used to work. To make matters worse this (assuming the same function) seems to work ok;

//============================= spork ~ dummy() @=> Shred foo; hour => now; //===============================

I wouldn't be surprised if it was at least related because I can't think of anything else that would break this.

At least it should help pinpoint the culprit. I suggest add it to the Wiki bug page and we try sprinkling some variation of the working version with some "@" signs. Sprinkling magical "@" signs seems to pacify this bug ;¬).

Yours, Kas.

_______________________________________________ chuck-users mailing list chuck-users@lists.cs.princeton.edu https://lists.cs.princeton.edu/mailman/listinfo/chuck-users

-- http://semiotech.org http://semiotech.org/michael

-- http://semiotech.org http://semiotech.org/michael

I'm glad it's the same bug cross platforms, I just think it's a weird bug, that's all. Do I add to the bottom of the bugs page or the top? -Mike On Sun, Oct 12, 2008 at 5:03 PM, Kassen wrote:

2008/10/12 mike clemow

...

Yup, that totally worked on a Mac too.

That's WEIRD.

No, that's good! I want as little difference between the versions as possible. I'm already juggling; Linux, commandline (latest greatest, canonical) Linux, Mini (my own build with the latest version of the CK source, seems to work... so far)

Windows (latest greatest but modded to use ASIO, needs to run in compatibility mode) Windows Mini (official version, so lagging one version of ChucK).

I need yet more differences between versions like I need a ice-pick to the ear. it's great that the core stuff is identical and we can trade and share code.

...

Thanks Kas!

You're welcome, pinpointing was hardly any effort here. If you'd like; add your catch to the Wiki for tracking. It's a nice one, will make a fine trophy. :¬)

Kas.

_______________________________________________ chuck-users mailing list chuck-users@lists.cs.princeton.edu https://lists.cs.princeton.edu/mailman/listinfo/chuck-users

-- http://semiotech.org http://semiotech.org/michael

Hi kas and mike, I run mike's chuck patch and didn't get a bus error... but did get it with kassen's code. However I could get rid of the error if the shred is declared but not "initilised", thus using a reference when declaring it. I think that's been always the case..., can't remember though. So adding Shred @ foo; to kassen's code let's assign to shred later. The resulting code would be: fun void dummy() { while(1) { second => now; } } Shred @ foo; spork ~ dummy() @=> foo; <<< foo.id()>>>; hour => now; eduard On Oct 12, 2008, at 11:25 PM, Kassen wrote:

Mike,

I'm glad it's the same bug cross platforms, I just think it's a weird bug, that's all.

Yeah, it's probably the garbage/reference stuff. I had no idea the problem was quite that bad.

Do I add to the bottom of the bugs page or the top?

I'd take the "hopefully fixed for the next release" page and somewhere in the "open" section? I think it's a pretty bad one, it must've been broken by the latest version.

Kas. _______________________________________________ chuck-users mailing list chuck-users@lists.cs.princeton.edu https://lists.cs.princeton.edu/mailman/listinfo/chuck-users

On Oct 13, 2008, at 12:08 AM, Kassen wrote:

Hi kas and mike,

Hey, Eduard.

I run mike's chuck patch and didn't get a bus error... but did get it with kassen's code.

I believe the proper term here is "WHAT?!".

weird... I do get a "bus error" now. I am on OS X (intel)