ruby-core

So, is it correct to assume that for language constructs that create built-in types like Range, Array, Hash etc that user-defined initialize methods are never called?

[#11549] Re: Proc initialize method not called under certain circumstances — "Berger, Daniel" <Daniel.Berger@...> 2007/06/28

> -----Original Message-----

[#11550] Re: Proc initialize method not called under certain circumstances — MenTaLguY <mental@...> 2007/06/28

On Fri, 29 Jun 2007 06:24:48 +0900, "John Lam (CLR)" <jflam@microsoft.com> wrote:

[#11553] Re: Proc initialize method not called under certain circumstances — "Marcel Molina Jr." <marcel@...> 2007/06/28

On Fri, Jun 29, 2007 at 06:54:49AM +0900, MenTaLguY wrote:

[#11554] Re: Proc initialize method not called under certain circumstances — murphy <murphy@...> 2007/06/28

> ... for language constructs that create

[#11560] Re: Proc initialize method not called under certain circumstances — Yukihiro Matsumoto <matz@...> 2007/06/29

Hi,

[#11548] NilClass, TrueClass, FalseClass special cased for singleton classes? — "John Lam (CLR)" <jflam@...>

If you do:

3 messages 2007/06/28

[#11557] Re: NilClass, TrueClass, FalseClass special cased for singleton classes? — Nobuyoshi Nakada <nobu@...> 2007/06/29

Hi,

[#11559] Re: NilClass, TrueClass, FalseClass special cased for singleton classes? — Joel VanderWerf <vjoel@...> 2007/06/29

John Lam (CLR) wrote:

[#11555] Name of Classes — =?ISO-8859-15?Q?Wolfgang_N=E1dasi-Donner?= <wonado@...>

Moin, moin!

2 messages 2007/06/28

[#11558] Re: Name of Classes — Nobuyoshi Nakada <nobu@...> 2007/06/29

Moin,

[#11556] delete_if with destructive methods in block — Florian Gross <florgro@...>

Hi,

3 messages 2007/06/29

[#11561] Re: [BUG] delete_if with destructive methods in block — Nobuyoshi Nakada <nobu@...> 2007/06/29

Hi,

[#11567] Re: delete_if with destructive methods in block — Florian Gross <florgro@...> 2007/06/30

That fixes it for me. I got no regressions from Rubinius' hash spec

[#11562] Can clear frozen array using initialize() — Florian Gross <florgro@...>

Hi,

2 messages 2007/06/29

[#11563] Re: [BUG] Can clear frozen array using initialize() — Yukihiro Matsumoto <matz@...> 2007/06/30

Hi,

Re: $2000 USD Reward for help fixing Segmentation Fault in GC

From: Brent Roman <brent@...>

Date: 2007-06-02 16:13:36 UTC

List: ruby-core #11395

Zimbatm,

SourceRef and ObjectSpace are pure ruby.  They are for debugging and
should not be
used in normal operation.  They are loaded in our application's
initialization
as a convenience.  The only 'C' extensions used are the "standard"
termios for
configuring serial ports and our own, very trivial, mbarilib.  mbarilib
defines
only one method, called "Kernel.doze", that is equivalent to
Kernel.sleep but
clears Thread.critical before sleeping (just as Thread.stop does).  This
lets one
stop a thread with a "timeout" while avoiding the race condition
that would otherwise result if Thread.critical were previously cleared
as a separate
ruby operation as in:

       Thread.critical = false
       sleep(time)  #race if this thread gets interrupted between these
two lines

The full core dump at the time of the crash is available on our FTP site at:

ftp://ftp.mbari.org/pub/brent

"The code" is about 1MB of ruby source.  If you can be more specific about
you want to see, I'll look into releasing those parts to you.

- brent

> Hi Brent,
>
> I've looked a bit around but it seems hard to do anything without the
> full running stack. As you said, the problem also occurs on an x86
> computer so if you are not bound by licencing you could share the code ?
>
> I've also looked at your ruby extensions. Do you use them a lot ? From
> the backtrace, you have a lot of evals. Are they comming from the
> SourceRef extension ?
>
> Also are you using ObjectSpace ?
>
> -- 
> Cheers,
>   zimbatm

-- 
 Brent Roman
 Software Engineer                 Tel: 831 775 1808
 425 Clinton St., Santa Cruz,      California, 95062
 mailto:brent@mbari.org  http://www.mbari.org/~brent

Thread

Prev Next

In This Thread

Prev Next