26947-27920

26748-27013 subjects 27138-27901

[Bug #2376] Kernel.__method__ rubyspec failures for 1.8.*
26947 [redmine@ru y] Issue #2376 has been updated by Vladimir Sizikov.

[Bug #2411] String#encode fails but eval("#coding:") works
26948 [redmine@ru y] Issue #2411 has been updated by Suraj Kurapati.

[Feature #2408](Open) better error reporting in windows when you try to run a directory
26951 [redmine@ru y] Issue #2408 has been updated by Usaku NAKAMURA.

[Feature #2294] [PATCH] ruby_bind_stack() to embed Ruby in coroutine
26952 [redmine@ru y] Issue #2294 has been updated by Ammar Ali.

[Feature #2294] [PATCH] ruby_bind_stack() to embed Ruby in coroutine
26953 [redmine@ru y] Issue #2294 has been updated by Suraj Kurapati.

[Bug #2413] MiniTest::Assertions#assert_raises ignores message
26954 [redmine@ru y] Bug #2413: MiniTest::Assertions#assert_raises ignores message

[Bug #2414] make test-all does not work after gem update
26955 [redmine@ru y] Bug #2414: make test-all does not work after gem update

[Bug #2384](Closed) mkmf find_executable doesn't find batch files in doze
26956 [redmine@ru y] Issue #2384 has been updated by Nobuyoshi Nakada.

Re: Caching #to_s for immutables (and a possible future for constant-folding) [with patch]
26958 [ks@ku ts ep ] ...
26960 [rogerdpack@g] n the
26963 [ks@ku ts ep ] I just ran rubyspec against it; ~ 5% time improvement.
26966 [ks@ku ts ep ] object (0x9aed298 flags=0x900805 klass=0x0)
26968 [ks@ku ts ep ] Allocating a String before rb_cString is created. :)

[Bug:trunk] syntax errors in method name or argument inside do block causes SEGV ( Re: Methods with default params not at the end and with rest params)
26961 [nobu@ru y- a] Simpler examples.

[Bug #2415](Closed) syntax errors in method name or argument inside do block causes SEGV ( Re: Methods with default params not at the end and with rest params)
26962 [redmine@ru y] Issue #2415 has been updated by Nobuyoshi Nakada.

[Backport #1939] Ripper doesn't handle local variables
26964 [redmine@ru y] Issue #1939 has been updated by Nobuyoshi Nakada.

[Backport #1939] Ripper doesn't handle local variables
26965 [redmine@ru y] Issue #1939 has been updated by Nobuyoshi Nakada.

[ANN] Ruby draft specification
26969 [shugo@ne la ] The first version of the draft specification of Ruby has been released
26997 [shugo@ne la ] A hyperlinked version of the draft specification is now available at the si=

Re: Caching #to_s for immutables (and a possible future for constant-folding) [with patch v2]
26970 [ks@ku ts ep ] ...
26978 [mame@ts .n .] This is really interesting result, but the impact of the spec change seems
+ 26979 [now@bi wi se] Couldn”Ēt one delay duplication until mutation?
| 26987 [ks@ku ts ep ] If Ruby passed arguments by name, not value... :).
| 26989 [now@bi wi se] Silly me.  I was actually thinking of returning a proxy that would
| 26999 [murphy@ru yc] I'm actually glad that it doesn't.
+ 26988 [ks@ku ts ep ] Another alternative would be to leave #to_s alone and introduce

[Bug #2383](Open) windows - calling FreeEnvironmentStrings on envarea invalidates memory in use
26971 [redmine@ru y] Issue #2383 has been updated by Usaku NAKAMURA.

How to
26972 [hackergene@g] Seeing that there is a wrong typing mistake in the page
+ 26975 [rogerdpack@g] which basically just instructs you on how to submit a patch to core.
| 26977 [james@gr ys ] for the
+ 26998 [yugui@yu ui ] Here are howtos for developers.

[Bug #1482](Feedback) Kernel.exec doesn't respect COMSPEC environment variable on Windows
26973 [redmine@ru y] Issue #1482 has been updated by Usaku NAKAMURA.

[Bug #1308](Open) bigdecimal debug code prevents tests from running
26974 [redmine@ru y] Issue #1308 has been updated by Usaku NAKAMURA.

[Bug #2417] Time.asctime/ctime output differs from 1.8
26980 [redmine@ru y] Bug #2417: Time.asctime/ctime output differs from 1.8

[Bug #2418] method_missing (assignment) returns args instead of return value
26981 [redmine@ru y] Bug #2418: method_missing (assignment) returns args instead of return value
26982 [judofyr@gm i] Any method ending in = ignores the result and returns the right-side.
26993 [nobu@ru y- a] Accurately, assignment expressions return rhs value, not

[Bug #2418] method_missing (assignment) returns args instead of return value
26983 [redmine@ru y] Issue #2418 has been updated by Dave B.

[Bug #2419] super call may use wrong receiver object
26984 [redmine@ru y] Bug #2419: super call may use wrong receiver object

[Bug #2419](Assigned) super call may use wrong receiver object
26985 [redmine@ru y] Issue #2419 has been updated by ujihisa ..

[Feature #2294] [PATCH] ruby_bind_stack() to embed Ruby in coroutine
26986 [redmine@ru y] Issue #2294 has been updated by Ammar Ali.

[Bug #2417](Rejected) Time.asctime/ctime output differs from 1.8
26990 [redmine@ru y] Issue #2417 has been updated by Usaku NAKAMURA.

Unsubscribe
26991 [nbitspoken@c] Sent from my iPhone.

[Bug #2417] Time.asctime/ctime output differs from 1.8
26992 [redmine@ru y] Issue #2417 has been updated by Tomas Matousek.

[Bug #2417](Closed) Time.asctime/ctime output differs from 1.8
26994 [redmine@ru y] Issue #2417 has been updated by Usaku NAKAMURA.

[Bug #1583] Time + String no Longer Raises TypeError?
26995 [redmine@ru y] Issue #1583 has been updated by Yui NARUSE.

[Bug #1354](Closed) stacktrace on ppc
26996 [redmine@ru y] Issue #1354 has been updated by Yui NARUSE.

[Bug #2420] super call may use wrong receiver object
27000 [redmine@ru y] Bug #2420: super call may use wrong receiver object

[Bug #2419](Closed) super call may use wrong receiver object
27001 [redmine@ru y] Issue #2419 has been updated by Nobuyoshi Nakada.

[Bug #2355] Please update rubygems!
27002 [redmine@ru y] Issue #2355 has been updated by Yui NARUSE.

[Bug #2422] splat operator fails on array of 1 element
27003 [redmine@ru y] Bug #2422: splat operator fails on array of 1 element
27004 [akr@fs j. rg] The Ruby 1.9 behavior is simplified and more consistent.

[Bug #2394] [BUG] pthread_mutex_lock: 22 revisited
27005 [redmine@ru y] Issue #2394 has been updated by Roger Pack.

[Bug #2422] splat operator fails on array of 1 element
27006 [redmine@ru y] Issue #2422 has been updated by Dave B.
27007 [akr@fs j. rg] The example is not balanced.

[Bug #2423] REXML edge case with <a>A &gt; B</a>
27008 [redmine@ru y] Bug #2423: REXML edge case with <a>A &gt; B</a>

[Bug #2423] REXML edge case with <a>A &gt; B</a>
27009 [redmine@ru y] Issue #2423 has been updated by Aaron Patterson.

[Bug #2355] Please update rubygems!
27010 [redmine@ru y] Issue #2355 has been updated by Eric Hodel.

[Bug #2066] Merge Latest RDoc/Ri
27011 [redmine@ru y] Issue #2066 has been updated by Eric Hodel.

[Bug #2426] ruby.h fails requireing ruby/config.h
27012 [redmine@ru y] Bug #2426: ruby.h fails requireing ruby/config.h

possible bug in Method#source_location
27014 [rogerdpack@g] It appears that Method#source_location returns different values for
27016 [ujihisa@gm i] In my environment, both versions of ruby said same results.
27017 [luislavena@g] ruby 1.9.1p243 (2009-07-16 revision 24175) [i386-mingw32]

seg fault sighting
27015 [rogerdpack@g] As an FYI, I got this once...
27717 [ko1@at ot ne] any reproduce script?
27749 [rogerdpack2@] I haven't seen it since then, so...let's just hope it was a fluke :)

[Bug #2427] possible bug in Method#source_location (from )
27018 [redmine@ru y] Bug #2427: possible bug in Method#source_location (from [ruby-core:27014])

[Bug #2428] URI::Generic#eql? broken when passed nil
27019 [redmine@ru y] Bug #2428: URI::Generic#eql? broken when passed nil
27053 [monty.willia] We should contribute an mspec for this if one doesn't already exist.

[Bug #2427] possible bug in Method#source_location (from )
27020 [redmine@ru y] Issue #2427 has been updated by Usaku NAKAMURA.

[Bug #2428] URI::Generic#eql? broken when passed nil
27021 [redmine@ru y] Issue #2428 has been updated by ujihisa ..

[Bug #2427] possible bug in Method#source_location (from )
27022 [redmine@ru y] Issue #2427 has been updated by Usaku NAKAMURA.

[Bug #2429] powerpc-linux-gcc compiling openssl error
27023 [redmine@ru y] Bug #2429: powerpc-linux-gcc compiling openssl error
27033 [nobu@ru y- a] Other lines before it have also RSTRING_LEN and RSTRING_PTR,

[Backport #2430] No StringIO#{readpartial,read_nonblock}
27024 [redmine@ru y] Backport #2430: No StringIO#{readpartial,read_nonblock}

[Backport #2431] StringIO#{gets,readlines} with "" (paragraph mode) trims last "\n"
27025 [redmine@ru y] Backport #2431: StringIO#{gets,readlines} with "" (paragraph mode) trims last "\n"

[Backport #2432] StringIO#reopen incomplete initialization
27026 [redmine@ru y] Backport #2432: StringIO#reopen incomplete initialization

[Bug #2429] powerpc-linux-gcc compiling openssl error
27027 [redmine@ru y] Issue #2429 has been updated by lin fu.

[Bug #2429] powerpc-linux-gcc compiling openssl error
27028 [redmine@ru y] Issue #2429 has been updated by Nobuyoshi Nakada.

[Bug #2429] powerpc-linux-gcc compiling openssl error
27029 [redmine@ru y] Issue #2429 has been updated by lin fu.

[Bug #2427](Closed) possible bug in Method#source_location (from )
27030 [redmine@ru y] Issue #2427 has been updated by Nobuyoshi Nakada.
27043 [rogerdpack@g] Thanks for fixing that (works here).

[Bug #2433] Ruby gem update --system /gem install [any_gem_name] ERROR
27031 [redmine@ru y] Bug #2433: Ruby gem update --system /gem install [any_gem_name] ERROR
27041 [rogerdpack@g] -1.4.gem
27042 [luislavena@g] u-1.4.gem

[Bug #2434] Ruby gem update --system /gem install [any_gem_name] ERROR
27032 [redmine@ru y] Bug #2434: Ruby gem update --system /gem install [any_gem_name] ERROR

[Bug #2436] maybe lambda ruby1.9.1.p129
27034 [redmine@ru y] Bug #2436: maybe lambda ruby1.9.1.p129

[Bug #2436](Rejected) maybe lambda ruby1.9.1.p129
27035 [redmine@ru y] Issue #2436 has been updated by Nobuyoshi Nakada.

Ruby causes nearly as much cpu wakeups as firefox
27036 [unexist@do f] ...
+ 27040 [rogerdpack@g] timer_wakeup)
| 27044 [brabuhr@gm i] rtimer_wakeup)
| 27045 [luislavena@g] hrtimer_wakeup)
| 27046 [unexist@do f] ...
+ 27048 [nobu@ru y- a] A timer thread wakes up per 10 milliseconds, for thread
  27050 [unexist@do f] ...
  27051 [hongli@pl n9] That thread is responsible for timesliced scheduling of Ruby threads. However in
  27066 [unexist@do f] ...
  27071 [mame@ts .n .] Because the running thread is acquiring the GIL.

[Bug #2438] Segmentation Fault in YAML
27037 [redmine@ru y] Bug #2438: Segmentation Fault in YAML

[Bug #2438](Feedback) Segmentation Fault in YAML
27038 [redmine@ru y] Issue #2438 has been updated by Nobuyoshi Nakada.

[Bug #2438] Segmentation Fault in YAML
27039 [redmine@ru y] Issue #2438 has been updated by Luis Lavena.

[Bug #2441] irb win32 - command help broken inside irb from installer ( irb version 0.9.5(05/04/13))
27047 [redmine@ru y] Bug #2441: irb win32 - command help broken inside irb from installer ( irb version 0.9.5(05/04/13))

[Bug #2441] irb win32 - command help broken inside irb from installer ( irb version 0.9.5(05/04/13))
27049 [redmine@ru y] Issue #2441 has been updated by Nobuyoshi Nakada.

[Backport #1547](Rejected) Fiber#alive? Returns 0 for true on 1.9.1
27052 [redmine@ru y] Issue #1547 has been updated by Yuki Sonoda.

[Bug #1744] Error with Marshal dump/load on a delegated class.
27054 [redmine@ru y] Issue #1744 has been updated by Yuki Sonoda.

[Backport #1939](Closed) Ripper doesn't handle local variables
27055 [redmine@ru y] Issue #1939 has been updated by Yuki Sonoda.

[Backport #1063](Closed) in `write': Not enough space - <STDOUT> (Errno::ENOMEM) on Windows XP
27056 [redmine@ru y] Issue #1063 has been updated by Yuki Sonoda.

[Backport #1934] Segmentation fault
27057 [redmine@ru y] Issue #1934 has been updated by Yuki Sonoda.

[Backport #2275](Closed) Please backport r22308 from trunk to branches/ruby_1_9_1: hash.c”Ēs rb_hash() always returning a Fixnum
27058 [redmine@ru y] Issue #2275 has been updated by Yuki Sonoda.

[Backport #2412](Closed) CSV regression after revision 25952
27059 [redmine@ru y] Issue #2412 has been updated by Yuki Sonoda.

[Bug #2443] {Object,BasicObject}.clone
27060 [redmine@ru y] Bug #2443: {Object,BasicObject}.clone

[Bug #1744](Closed) Error with Marshal dump/load on a delegated class.
27061 [redmine@ru y] Issue #1744 has been updated by Nobuyoshi Nakada.

[Backport #2194](Closed) Document update for GDBM#select
27062 [redmine@ru y] Issue #2194 has been updated by Yuki Sonoda.

[Backport #2214](Closed) Segfault with Range of Symbols in IRB
27063 [redmine@ru y] Issue #2214 has been updated by Yuki Sonoda.

[Backport #2099](Closed) Error when getting binding for a method via converting it to a Proc
27064 [redmine@ru y] Issue #2099 has been updated by Yuki Sonoda.

[Backport #2415](Closed) syntax errors in method name or argument inside do block causes SEGV ( Re: Methods with default params not at the end and with rest params)
27065 [redmine@ru y] Issue #2415 has been updated by Yuki Sonoda.

[Backport #2441](Closed) irb win32 - command help broken inside irb from installer ( irb version 0.9.5(05/04/13))
27067 [redmine@ru y] Issue #2441 has been updated by Yuki Sonoda.

[Bug #2303] dl.so segfaults on mingw32
27068 [redmine@ru y] Issue #2303 has been updated by Vlad Why.

[Backport #2441] irb win32 - command help broken inside irb from installer ( irb version 0.9.5(05/04/13))
27069 [redmine@ru y] Issue #2441 has been updated by Vlad Why.

[Bug #2376] Kernel.__method__ rubyspec failures for 1.8.*
27070 [redmine@ru y] Issue #2376 has been updated by Shugo Maeda.

[Bug #2446] documentation fix for strftime %l
27072 [redmine@ru y] Bug #2446: documentation fix for strftime %l

[Bug #2446](Closed) documentation fix for strftime %l
27073 [redmine@ru y] Issue #2446 has been updated by Yui NARUSE.

[Bug #2449] StringIO#ungetc behavior is contrary to its spec
27074 [redmine@ru y] Bug #2449: StringIO#ungetc behavior is contrary to its spec
27079 [matz@ru y- a] strio.ungetc(string)   -> nil
+ 27114 [nakahiro@gm ] Yes.
+ 27115 [nakahiro@gm ] Yes.

[Bug #2450] Array#sample can loop forever [patched]
27075 [redmine@ru y] Bug #2450: Array#sample can loop forever [patched]

[Backport #2431] StringIO#{gets,readlines} with "" (paragraph mode) trims last "\n"
27076 [redmine@ru y] Issue #2431 has been updated by Vladimir Sizikov.
27078 [nakahiro@gm ] Would you please point the 6 tests? It must be helpful to discuss this issue.
27082 [vsizikov@gm ] #> mspec -t j library/stringio
27083 [nakahiro@gm ] All the failed tests are for StringIO, right? Then we are free to fix
27118 [vsizikov@gm ] Yes.
27123 [nakahiro@gm ] Author(nobu) assigned this ticket to mame so it's a bug to be fixed in

[Bug #2364](Closed) Float conversion of NaN in 1.8.x
27077 [redmine@ru y] Issue #2364 has been updated by Marc-Andre Lafortune.

[Feature #2451] BasicObject.initialize with variable number of argument
27080 [redmine@ru y] Feature #2451: BasicObject.initialize with variable number of argument
27231 [matz@ru y- a] I don't think so.  Please check in the fix.

[Backport #2441] irb win32 - command help broken inside irb from installer ( irb version 0.9.5(05/04/13))
27081 [redmine@ru y] Issue #2441 has been updated by Nobuyoshi Nakada.

[Bug #2453] matrix#rank causes inf loop
27084 [redmine@ru y] Bug #2453: matrix#rank causes inf loop

[Bug #2453](Rejected) matrix#rank causes inf loop
27085 [redmine@ru y] Issue #2453 has been updated by Yuki Sonoda.

[Feature #2454] OpenSSL has no maintainer
27086 [redmine@ru y] Feature #2454: OpenSSL has no maintainer

code for redmine.ruby-lang.org?
27087 [rogerdpack@g] Is there any way we could get the code for redmine.ruby-lang.org
+ 27088 [aaron@te de ] Aaron Patterson
+ 27094 [sakuro@22 8c] Check yugui's fork at http://github.com/yugui/redmine4ruby-lang
+ 27109 [naruse@ai em] its project is

[Bug #2377] update documentation for IO#eof?
27089 [redmine@ru y] Issue #2377 has been updated by Roger Pack.

[Feature #2454] OpenSSL has no maintainer
27090 [redmine@ru y] Issue #2454 has been updated by Chris Saunders.

[Feature #2454] OpenSSL has no maintainer
27091 [redmine@ru y] Issue #2454 has been updated by Dirkjan Bussink.

[Feature #2454] OpenSSL has no maintainer
27092 [redmine@ru y] Issue #2454 has been updated by Evan Phoenix.

[Feature #2456] It is better to let user to select particular ssl protocol easily.
27093 [redmine@ru y] Feature #2456: It is better to let user to select particular ssl protocol easily.

[Feature #2454](Assigned) OpenSSL has no maintainer
27095 [redmine@ru y] Issue #2454 has been updated by Yui NARUSE.
27102 [matz@ru y- a] If Evan wants committer privilege to check in his update, I'd love to
27105 [evan@fa li g] See below.
27110 [matz@ru y- a] Thank you!  Please send your GPG signed ssh2 public key to

[Bug #2459] Cannot statically link ext/digest/sha2
27096 [redmine@ru y] Bug #2459: Cannot statically link ext/digest/sha2

[Bug #2461] ARGF.rewind does not reset ARGF.lineno to zero
27097 [redmine@ru y] Bug #2461: ARGF.rewind does not reset ARGF.lineno to zero

[Bug #2461](Rejected) ARGF.rewind does not reset ARGF.lineno to zero
27098 [redmine@ru y] Issue #2461 has been updated by Hiroshi NAKAMURA.

[Bug #1693] ARGF.rewind Doesn't Reset ARGF.lineno
27099 [redmine@ru y] Issue #1693 has been updated by Hiroshi NAKAMURA.

[Bug #2462] Bignum comparison fail [ruby-dev:38672]
27100 [redmine@ru y] Bug #2462: Bignum comparison fail [ruby-dev:38672]

[Bug #2462] Bignum comparison fail [ruby-dev:38672]
27101 [redmine@ru y] Issue #2462 has been updated by Hiroshi NAKAMURA.

[Bug #2464] cross-compile Ruby patchlevel 376 fails
27103 [redmine@ru y] Bug #2464: cross-compile Ruby patchlevel 376 fails

[Bug #2459](Feedback) Cannot statically link ext/digest/sha2
27104 [redmine@ru y] Issue #2459 has been updated by Nobuyoshi Nakada.

[Backport #2462] Bignum comparison fail [ruby-dev:38672]
27106 [redmine@ru y] Issue #2462 has been updated by Nobuyoshi Nakada.

[Bug #2376] Kernel.__method__ rubyspec failures for 1.8.*
27107 [redmine@ru y] Issue #2376 has been updated by Brian Ford.
27108 [shugo@ru y- ] I guess Kernel#send doesn't push a frame in 1.9, but does in 1.8.

[Bug #2376] Kernel.__method__ rubyspec failures for 1.8.*
27111 [redmine@ru y] Issue #2376 has been updated by Brian Ford.
27137 [shugo@ru y- ] Matz said that he prefer the behavior of Ruby 1.8.

[Bug #2465] Couldn't compile ruby 1.8.7-p72 on HP-UX 11.23 ia64 with pthread
27112 [redmine@ru y] Bug #2465: Couldn't compile ruby 1.8.7-p72 on HP-UX 11.23 ia64 with pthread

[RubySpec #2466] Mutex#lock documentation is incorrect
27113 [redmine@ru y] RubySpec #2466: Mutex#lock documentation is incorrect

[Bug #2468] Array expansion inside case/when gives unexpected results
27116 [redmine@ru y] Bug #2468: Array expansion inside case/when gives unexpected results

[Bug #2469] Difference behavior of IO#read_nonblock and #read between 1.8 and 1.9
27117 [redmine@ru y] Bug #2469: Difference behavior of IO#read_nonblock and #read between 1.8 and 1.9
27121 [akr@fs j. rg] It is changed intentionally.

extra settings in rbconfig on mingw?
27119 [rogerdpack@g] I noticed that this is in the RbConfig, mingw ruby1.9.1p243

#to_enum ignores block?
27120 [rogerdpack@g] Is #to_enum ignoring its block expected?
27126 [m.fellinger@] The whole purpose of to_enum is to return an Enumerator instance.
27127 [rogerdpack@g] I would expect it to act like a filter.
+ 27128 [rick.denatal] As far as I can tell, standard Ruby Enumerable#to_enum never took a
+ 27129 [rick.denatal] Why not just
| 27130 [rogerdpack@g] Oops well assuming the operation is correct--it's because I want an
| 27133 [mame@ts .n .] I really understand what you want to do.  But unfortunately, current
| 27134 [rogerdpack@g] Excellent.  I was wondering where those lazy versions of select were...
| 27141 [mame@ts .n .] Yes.  It does not use fiber unless you do a #next call.
+ 27131 [m.fellinger@] As I said, it doesn't treat blocks.
  27132 [rogerdpack@g] True, though I'd still like some confirmation this is what was expected...

[Bug #2469](Closed) Difference behavior of IO#read_nonblock and #read between 1.8 and 1.9
27122 [redmine@ru y] Issue #2469 has been updated by ujihisa ..

[Bug #2472] ri generation problem
27124 [redmine@ru y] Bug #2472: ri generation problem

[Bug #2472] ri generation problem
27125 [redmine@ru y] Issue #2472 has been updated by Sergey Avseyev.

better GC?
27135 [rogerdpack@g] Could I put in a small plea for a better GC?
+ 27136 [matz@ru y- a] Yes, but unfortunately it's not small at all.  GC has a lot of
| + 27476 [pbrannan@at ] As I mentioned to you at the conference, GC is the primary reason I can
| | + 27477 [ruby-ml@ki t] Rubinius is also compatible with existing extensions*
| | | + 27478 [ks@ku ts ep ] FFIs complicate GC.  There are numerous papers that assert that a simple
| | | | 27547 [headius@he d] To be honest, I think this is the *only* path forward for MRI.
| | | | 27553 [rick.denatal] in a
| | | | 27607 [headius@he d] For what it's worth, we have nowhere near the resources nor expertise
| | | + 27494 [rogerdpack2@] I wonder if its GC could be merged into MRI :)
| | |   + 27497 [ruby-ml@ki t] Nope.
| | |   | 27555 [rogerdpack2@] Explanation? (Licensing problem?)
| | |   | + 27556 [goncalossilv] Among other thins (like the whole architecture of MRI), it would break most
| | |   | + 27569 [evan@fa li g] The Rubinius GC (there are actually 3 of them) require write barriers. =
| | |   |   27585 [ice799@gm il] port to MRI, you'd have to rewrite most of the internals to use write barr=
| | |   |   27586 [ice799@gm il] o port to MRI, you'd have to rewrite most of the internals to use write bar=
| | |   |   + 27589 [goncalossilv] So, we're getting the worse of the two options. GC will still be slow and
| | |   |   | 27590 [binary42@gm ] us
| | |   |   + 27593 [rogerdpack2@] I'd be down with that, though I'll admit I don't really know what that
| | |   + 27498 [headius@he d] I'd like to hear about what else isn't running faster for you in
| | + 27479 [murphy@ru yc] How do you do that? I'm curious.
| | | 27544 [pbrannan@at ] How do we do which part?  Disable the GC or write code that doesn't
| | + 27480 [ks@ku ts ep ] Adding reference counting to MRI would make existing Ruby code that is
| | | 27531 [pbrannan@at ] Avoiding garbage doesn't solve the problem; if there is a large number
| | | + 27533 [goncalossilv] That's an interesting idea, Paul. I love it but I think it would imply a lo=
| | | | 27540 [judofyr@gm i] Well, Rails itself is single-threaded and so are most of the servers.
| | | + 27534 [e@il um .o g] I think a pluggable/loadable GC scheme, as long as its really simple to =
| | | | + 27535 [goncalossilv] I think there's no need for it to be hot-pluggable, so there wouldn't be an=
| | | | + 27539 [rick.denatal] So would be a lasting world peace!
| | | |   + 27542 [pbrannan@at ] I think I agree with your logic, though not necessarily your wording.
| | | |   + 27543 [e@il um .o g] Free beer wouldn't be bad, either ;-)
| | | + 27538 [matz@ru y- a] It's not easy.  Most of performance-wise garbage collectors require
| | + 27496 [brent@mb ri ] Paul,
| |   + 27530 [pbrannan@at ] For one application where we deployed both JRuby and Ruby 1.8, JRuby was
| |   | + 27532 [murphy@ru yc] and is it faster/more responsive now?
| |   | | 27537 [pbrannan@at ] Yes, by two orders of magnitude.
| |   | + 27550 [headius@he d] Some clarification of the pluggable GC stuff in the JVM (at least what
| |   + 27609 [meta@po ox c] Does JRuby run under Java RTS? Regular Java runtimes aren't realtime any
| |     27610 [headius@he d] The current Java RTS from Sun is an implementation of Java SE 5.0,
| + 27563 [brent@mb ri ] Toshio Endo and Kenjiro Taura adapted the Boehm conservative GC
|   + 27565 [rick.denatal] adds
|   | 27574 [brent@mb ri ] Rick,
|   + 27567 [wyhaines@gm ] adds
|   | 27568 [ks@ku ts ep ] If that's the case, then no one would use multi-core processors at
|   | 27570 [wyhaines@gm ] s an
|   | + 27571 [judofyr@gm i] Which leads us to why we need pluggable GC: There simply isn't one GC
|   | + 27576 [brent@mb ri ] The main point of this particular GC algorithm was to reduce the maximum
|   + 27592 [rogerdpack2@] adds
|     27596 [binary42@gm ] I'm not sure what kinds of app code you've been running but I see
|     + 27604 [lourens@me h] I've had this idea to use threaded code =
|     | 27605 [rogerdpack2@] Benchmarks?
|     + 27615 [rogerdpack2@] Interesting.  I suppose a cord is pretty much functionally equivalent
|     | 27619 [ks@ku ts ep ] The Boehm (BDW) GC has a GC-aware implementation of string cords in C.
|     + 27920 [rogerdpack2@] I'd be happy to hack something up if it would be useful--contact me offline.
+ 27536 [pbrannan@at ] I modified this program slightly (imo cases n=1..6 aren't useful for a
threads.html
top