[#7286] Re: ruby-dev summary 28206-28273 — ara.t.howard@...

On Thu, 2 Feb 2006, Minero Aoki wrote:

23 messages 2006/02/02
[#7292] ANDCALL / iff? / &? (was Re: ruby-dev summary 28206-28273) — mathew <meta@...> 2006/02/02

[#7293] Re: ANDCALL / iff? / &? (was Re: ruby-dev summary 28206-28273) — mathew <meta@...> 2006/02/02

mathew wrote:

[#7298] Re: ANDCALL / iff? / &? (was Re: ruby-dev summary 28206-28273) — James Britt <ruby@...> 2006/02/03

mathew wrote:

[#7310] Re: ANDCALL / iff? / &? (was Re: ruby-dev summary 28206-28273) — Evan Webb <evanwebb@...> 2006/02/07

I'm not sure we even need the 'with' syntax. Even if we do, it breaks

[#7311] Re: ANDCALL / iff? / &? (was Re: ruby-dev summary 28206-28273) — Eero Saynatkari <ruby-ml@...> 2006/02/07

On 2006.02.07 10:03, Evan Webb wrote:

[#7313] Re: ANDCALL / iff? / &? (was Re: ruby-dev summary 28206-28273) — Evan Webb <evanwebb@...> 2006/02/07

Umm, on what version are you seeing a warning there? I don't and never

[#7315] Re: ANDCALL / iff? / &? (was Re: ruby-dev summary 28206-28273) — Eero Saynatkari <ruby-ml@...> 2006/02/07

On 2006.02.07 14:47, Evan Webb wrote:

[#7316] Re: ANDCALL / iff? / &? (was Re: ruby-dev summary 28206-28273) — Evan Webb <evanwebb@...> 2006/02/07

I'd by far prefer it never emit a warning. The warning is assumes you

[#7305] Re: Problem with weak references on OS X 10.3 — Mauricio Fernandez <mfp@...>

On Sun, Feb 05, 2006 at 08:33:40PM +0900, Christian Neukirchen wrote:

28 messages 2006/02/05
[#7401] Symbols overlap ordinary objects, especially on OS X (Was: Re: Problem with weak references on OS X 10.3) — Eric Hodel <drbrain@...7.net> 2006/02/22

On Feb 5, 2006, at 5:05 AM, Mauricio Fernandez wrote:

[#7414] Re: Symbols overlap ordinary objects, especially on OS X (Was: Re: Problem with weak references on OS X 10.3) — Mauricio Fernandez <mfp@...> 2006/02/23

On Wed, Feb 22, 2006 at 02:21:24PM +0900, Eric Hodel wrote:

[#7428] Re: Symbols overlap ordinary objects, especially on OS X (Was: Re: Problem with weak references on OS X 10.3) — Tanaka Akira <akr@...17n.org> 2006/02/26

In article <1140968746.321377.18843.nullmailer@x31.priv.netlab.jp>,

[#7444] Re: Symbols overlap ordinary objects, especially on OS X (Was: Re: Problem with weak references on OS X 10.3) — nobu@... 2006/02/28

Hi,

[#7445] Re: Symbols overlap ordinary objects, especially on OS X (Was: Re: Problem with weak references on OS X 10.3) — Tanaka Akira <akr@...17n.org> 2006/02/28

In article <m1FDshr-0006MNC@Knoppix>,

[#7447] Re: Symbols overlap ordinary objects, especially on OS X (Was: Re: Problem with weak references on OS X 10.3) — Tanaka Akira <akr@...17n.org> 2006/02/28

In article <87irr047sx.fsf@m17n.org>,

[#7448] Re: Symbols overlap ordinary objects, especially on OS X (Was: Re: Problem with weak references on OS X 10.3) — Tanaka Akira <akr@...17n.org> 2006/02/28

In article <87vev0hxu5.fsf@m17n.org>,

[#7465] Re: Symbols overlap ordinary objects, especially on OS X (Was: Re: Problem with weak references on OS X 10.3) — "Evan Webb" <evanwebb@...> 2006/03/01

Just my quick 2 cents...

[#7468] Re: Symbols overlap ordinary objects, especially on OS X (Was: Re: Problem with weak references on OS X 10.3) — Tanaka Akira <akr@...17n.org> 2006/03/01

In article <92f5f81d0602281855g27e78f4eua8bf20e0b8e47b68@mail.gmail.com>,

[#7403] Module#define_method "send hack" fails with Ruby 1.9 — Emiel van de Laar <emiel@...>

Hi List,

12 messages 2006/02/22
[#7404] Re: Module#define_method "send hack" fails with Ruby 1.9 — George Ogata <g_ogata@...> 2006/02/22

Emiel van de Laar <emiel@rednode.nl> writes:

[#7406] Re: Module#define_method "send hack" fails with Ruby 1.9 — dblack@... 2006/02/22

Hi --

[#7442] GC Question — zdennis <zdennis@...>

I have been posting to the ruby-talk mailing list about ruby memory and GC, and I think it's ready

17 messages 2006/02/27

Re: GC Question

From: "H.Yamamoto" <ocean@...2.ccsnet.ne.jp>
Date: 2006-02-28 07:24:21 UTC
List: ruby-core #7449
Hello.

>Because ruby's GC is conservative.

I found interesting phenomenon.

def format(num)
  num.to_s.gsub(/(\d{1,3})(?=\d{3}+$)/) { $1 + "," }
end

def count(type)
   count = 0
   capacity = 0
   ObjectSpace.each_object(type) do |o|
     count += 1
     capacity += o.capacity
   end
   puts "#{type}"
   puts "  count = #{format(count)}"
   puts "  capacity = #{format(capacity)}"
end

def pause
   GC.enable
   GC.start
   GC.disable
   count(String)
   count(Array)
   puts
   sleep 5
end

class A
  def run
    arr = []
    800000.times { arr << "d" * 7 }
    pause
   end
end

GC.disable
A.new.run
1.times { pause } # not freed
pause # but freed here

////////////////////////////////////////////////////////

I have dived into gc.c, and applied this patch.

Index: gc.c
===================================================================
RCS file: /src/ruby/gc.c,v
retrieving revision 1.168.2.37
diff -u -w -b -p -r1.168.2.37 gc.c
--- gc.c	13 Feb 2006 09:10:53 -0000	1.168.2.37
+++ gc.c	28 Feb 2006 07:04:44 -0000
@@ -1354,6 +1354,7 @@ garbage_collect()
     setjmp(save_regs_gc_mark);
     mark_locations_array((VALUE*)save_regs_gc_mark, sizeof(save_regs_gc_mark) / sizeof(VALUE *));
 #if STACK_GROW_DIRECTION < 0
+    printf("----------> %p %p\n", STACK_END, rb_gc_stack_start);
     rb_gc_mark_locations((VALUE*)STACK_END, rb_gc_stack_start);
 #elif STACK_GROW_DIRECTION > 0
     rb_gc_mark_locations(rb_gc_stack_start, (VALUE*)STACK_END + 1);

////////////////////////////////////////////

E:\ruby-cvs\win32_1_8>miniruby \a.rb
----------> 0012FD0C 0012FFFC
----------> 0012E4C0 0012FFFC
String
  count = 800,111
  capacity = 5,602,208
Array
  count = 7
  capacity = 806,581

----------> 0012DD10 0012FFFC
String
  count = 800,116
  capacity = 5,602,275
Array
  count = 7
  capacity = 806,581

----------> 0012ED48 0012FFFC
String
  count = 113
  capacity = 2,219
Array
  count = 6
  capacity = 96

////////////////////////////////////////////

STACK_END is more far from rb_gc_stack_start on 1.times { pause }
(same thing happens on loop { pause }) this means ruby uses more stack
on block execution.

Because rb_gc_mark_locations marks all objects in the range of STACK_END to rb_gc_stack_start.
if GC.start runs inside block, block needs more stack, so more objects can be wrongly
marked as alive. (As you can see, last `pause` is outside of block, so less stack is used,
huge array goes out of stack range, it is freed)

# But I was suprised of this result. I heard conservative GC can mark dead object wrongly
# but this was a little differenct from what I expected. Maybe I'm missing something...



In This Thread