[#46049] [ruby-trunk - Feature #6590] Dealing with bigdecimal, etc gems in JRuby — "mrkn (Kenta Murata)" <muraken@...>
4 messages
2012/07/01
[#46078] [ruby-trunk - Feature #2565] adding hooks for better tracing — "mame (Yusuke Endoh)" <mame@...>
4 messages
2012/07/01
[#46108] Re: [ruby-trunk - Feature #2565] adding hooks for better tracing
— Aaron Patterson <tenderlove@...>
2012/07/02
On Mon, Jul 02, 2012 at 03:06:59AM +0900, mame (Yusuke Endoh) wrote:
[#46127] [ruby-trunk - Feature #2565] adding hooks for better tracing — "vo.x (Vit Ondruch)" <v.ondruch@...>
4 messages
2012/07/03
[#46160] [ruby-trunk - Feature #6693][Open] Don't warn for unused variables starting with _ — "marcandre (Marc-Andre Lafortune)" <ruby-core@...>
15 messages
2012/07/04
[#46163] [ruby-trunk - Feature #6695][Open] Configuration for Thread/Fiber creation — "ko1 (Koichi Sasada)" <redmine@...>
7 messages
2012/07/04
[#46172] [ruby-trunk - Feature #6697][Open] [PATCH] Add Kernel#Symbol conversion method like String(), Array() etc. — "madeofcode (Mark Dodwell)" <mark@...>
7 messages
2012/07/04
[#46236] [ruby-trunk - Bug #6704][Open] Random core dump — "trans (Thomas Sawyer)" <transfire@...>
7 messages
2012/07/07
[#46248] building ruby-1.9.3-p194 on AIX 6.1 TL05 SP06 — Perry Smith <pedzsan@...>
I am just now starting to debug this but hoped someone has already =
10 messages
2012/07/08
[#46251] Re: building ruby-1.9.3-p194 on AIX 6.1 TL05 SP06
— Yutaka Kanemoto <kinpoco@...>
2012/07/08
Hi Perry
[#46252] Re: building ruby-1.9.3-p194 on AIX 6.1 TL05 SP06
— Perry Smith <pedzsan@...>
2012/07/08
[#46254] Re: building ruby-1.9.3-p194 on AIX 6.1 TL05 SP06
— Perry Smith <pedzsan@...>
2012/07/08
[#46259] Re: building ruby-1.9.3-p194 on AIX 6.1 TL05 SP06
— Perry Smith <pedzsan@...>
2012/07/09
[#46266] Re: building ruby-1.9.3-p194 on AIX 6.1 TL05 SP06
— Yutaka Kanemoto <kinpoco@...>
2012/07/09
Hi Perry,
[#46262] [ruby-trunk - Feature #6710][Open] new special binding specifier :isolated — "ko1 (Koichi Sasada)" <redmine@...>
8 messages
2012/07/09
[#46276] Lambdaification of Method Calls — Robert Klemme <shortcutter@...>
Hi,
3 messages
2012/07/09
[#46320] [ruby-trunk - Feature #6721][Open] Object#yield_self — "alexeymuranov (Alexey Muranov)" <redmine@...>
25 messages
2012/07/11
[#46339] [ruby-trunk - Bug #6724][Open] waaaaaaant! ( — "zenspider (Ryan Davis)" <redmine@...>
11 messages
2012/07/11
[#46537] Re: [ruby-trunk - Bug #6724][Open] waaaaaaant! (
— Aaron Patterson <tenderlove@...>
2012/07/17
On Thu, Jul 12, 2012 at 08:58:36AM +0900, zenspider (Ryan Davis) wrote:
[#46540] Re: [ruby-trunk - Bug #6724][Open] waaaaaaant! (
— Luis Lavena <luislavena@...>
2012/07/18
On Tue, Jul 17, 2012 at 6:27 PM, Aaron Patterson
[#46377] [ruby-trunk - Feature #6727][Open] Add Array#rest (with implementation) — "duckinator (Nick Markwell)" <nick@...>
25 messages
2012/07/13
[#46420] [ruby-trunk - Feature #6731][Open] add new method "Object.present?" as a counter to #empty? — "rogerdpack (Roger Pack)" <rogerpack2005@...>
4 messages
2012/07/14
[#46500] [ruby-trunk - Feature #6739][Open] One-line rescue statement should support specifying an exception class — Quintus (Marvin Gülker) <sutniuq@...>
22 messages
2012/07/15
[#46535] [ruby-trunk - Bug #6749][Open] rdoc of Time class (incorrect explanation of leap seconds) — "stomar (Marcus Stollsteimer)" <redmine@...>
7 messages
2012/07/17
[#46687] [ruby-trunk - Bug #6749] rdoc of Time class (incorrect explanation of leap seconds)
— "drbrain (Eric Hodel)" <[email protected]>
2012/07/23
[#46713] Re: [ruby-trunk - Bug #6749] rdoc of Time class (incorrect explanation of leap seconds)
— sto.mar@...
2012/07/24
Hi Eric,
[#46734] Re: [ruby-trunk - Bug #6749] rdoc of Time class (incorrect explanation of leap seconds)
— Eric Hodel <[email protected]>
2012/07/24
On Jul 23, 2012, at 11:52 PM, [email protected] wrote:
[#46763] Re: [ruby-trunk - Bug #6749] rdoc of Time class (incorrect explanation of leap seconds)
— sto.mar@...
2012/07/25
Am 24.07.2012 19:44, schrieb Eric Hodel:
[#46546] Fwd: [ruby-cvs:43609] ko1:r36433 (trunk): * thread.c (rb_thread_call_without_gvl2): added. — SASADA Koichi <ko1@...>
Hi,
4 messages
2012/07/18
[#46547] Re: Fwd: [ruby-cvs:43609] ko1:r36433 (trunk): * thread.c (rb_thread_call_without_gvl2): added.
— Eric Wong <normalperson@...>
2012/07/18
SASADA Koichi <[email protected]> wrote:
[#46553] [ruby-trunk - Feature #2565] adding hooks for better tracing — "tenderlovemaking (Aaron Patterson)" <aaron@...>
4 messages
2012/07/18
[#46564] Ruby under CI - Windows — Luis Lavena <luislavena@...>
Hello,
6 messages
2012/07/20
[#46574] [ruby-trunk - Feature #6762][Open] Control interrupt timing — "ko1 (Koichi Sasada)" <redmine@...>
39 messages
2012/07/20
[#46575] [ruby-trunk - Feature #6762] Control interrupt timing
— "ko1 (Koichi Sasada)" <redmine@...>
2012/07/20
[#50542] Re: [ruby-trunk - Feature #6762][Open] Control interrupt timing
— Brent Roman <brent@...>
2012/12/03
I was suggesting "interruptible" as a better alternative for
[#46577] [ruby-trunk - Feature #6763][Open] Introduce Flonum technique to speedup floating computation on th 64bit environment — "ko1 (Koichi Sasada)" <redmine@...>
7 messages
2012/07/20
[#46586] [ruby-trunk - Bug #6764][Open] IO#read(size, buf) causes can't set length of shared string in trunk (2.0.0dev) — "nahi (Hiroshi Nakamura)" <nakahiro@...>
4 messages
2012/07/21
[#46641] [ruby-trunk - Bug #6780][Open] cannot compile zlib module, when cross-compiling. — "jinleileiking (lei king)" <jinleileiking@...>
14 messages
2012/07/23
[#46686] [ruby-trunk - Bug #6784][Open] Test failures related to numeric with x64 mingw — "h.shirosaki (Hiroshi Shirosaki)" <h.shirosaki@...>
5 messages
2012/07/23
[#46741] [ruby-trunk - Bug #6789][Open] parse.y compilation error due not updated id.h — "luislavena (Luis Lavena)" <luislavena@...>
8 messages
2012/07/24
[#46742] [ruby-trunk - Bug #6789][Assigned] parse.y compilation error due not updated id.h
— "luislavena (Luis Lavena)" <luislavena@...>
2012/07/24
[#46744] [ruby-trunk - Bug #6791][Open] ext/js on/generator/generator.c fails to compile on nightly build (AIX 6.1) — "pedz (Perry Smith)" <pedz@...>
4 messages
2012/07/25
[#46752] Re: [ruby-trunk - Bug #6791][Open] ext/js on/generator/generator.c fails to compile on nightly build (AIX 6.1)
— Yutaka Kanemoto <kinpoco@...>
2012/07/25
Hi Perry,
[#46772] Ruby 1.9.3 release? — Charles Oliver Nutter <headius@...>
JRuby will soon release 1.7.0pre2, the second preview of 1.7. Perhaps
6 messages
2012/07/25
[#46847] Re: Ruby 1.9.3 release?
— "NARUSE, Yui" <naruse@...>
2012/07/29
(2012/07/26 7:07), Charles Oliver Nutter wrote:
[#46864] Re: Ruby 1.9.3 release?
— Luis Lavena <luislavena@...>
2012/07/29
On Sat, Jul 28, 2012 at 10:59 PM, NARUSE, Yui <[email protected]> wrote:
[#46792] [ruby-trunk - Bug #6799][Open] Digest::*.hexdigest returns an ASCII-8BIT String — "Eregon (Benoit Daloze)" <redmine@...>
11 messages
2012/07/26
[#46832] [ruby-trunk - Bug #6807][Open] Can't compile ruby without ruby — "devcurmudgeon (Paul Sherwood)" <storitel@...>
13 messages
2012/07/28
[#46834] [ruby-trunk - Feature #6808][Open] Implicit index for enumerations — "trans (Thomas Sawyer)" <transfire@...>
15 messages
2012/07/28
[#46838] [ruby-trunk - Bug #6810][Open] `module A::B; end` is not equivalent to `module A; module B; end; end` with respect to constant lookup (scope) — "alexeymuranov (Alexey Muranov)" <redmine@...>
17 messages
2012/07/28
[#46854] [ruby-trunk - Feature #6811][Open] File, Dir and FileUtils should have bang-versions of singleton methods that fails silently — "prijutme4ty (Ilya Vorontsov)" <prijutme4ty@...>
7 messages
2012/07/29
[#46896] (Half-baked DRAFT) new `require' framework — SASADA Koichi <ko1@...>
Hi,
22 messages
2012/07/31
[#46897] Re: (Half-baked DRAFT) new `require' framework
— =?KOI8-R?B?4NLJyiDzz8vPzM/X?= <funny.falcon@...>
2012/07/31
2012/7/31 SASADA Koichi <[email protected]>
[#46899] Re: (Half-baked DRAFT) new `require' framework
— Alex Young <alex@...>
2012/07/31
On 31/07/12 13:29, SASADA Koichi wrote:
[#46902] Re: (Half-baked DRAFT) new `require' framework
— Trans <transfire@...>
2012/07/31
On Tue, Jul 31, 2012 at 12:07 PM, Alex Young <[email protected]> wrote:
[#46903] Re: (Half-baked DRAFT) new `require' framework
— Clifford Heath <clifford.heath@...>
2012/07/31
On 01/08/2012, at 5:59 AM, Trans wrote:
[#47040] Re: (Half-baked DRAFT) new `require' framework
— SASADA Koichi <ko1@...>
2012/08/07
(2012/07/31 21:29), SASADA Koichi wrote:
[#47054] Re: (Half-baked DRAFT) new `require' framework
— Rocky Bernstein <rockyb@...>
2012/08/07
If one is considering importing archive files like zip, tar, jar, or gem, I
[#47056] Re: (Half-baked DRAFT) new `require' framework
— Charles Oliver Nutter <headius@...>
2012/08/07
On Tue, Aug 7, 2012 at 8:48 AM, Rocky Bernstein <[email protected]> wrote:
[ruby-core:46573] Re: [ruby-trunk - Feature #6695] Configuration for Thread/Fiber creation
From:
SASADA Koichi <ko1@...>
Date:
2012-07-20 17:56:15 UTC
List:
ruby-core #46573
(2012/07/21 1:23), brixen (Brian Ford) wrote:
> I object to this API for setting stack size from Ruby for at least the two following reasons:
>
> 1. Stack size is an implementation detail and coupling Ruby code to details of a particular implementation is undesirable. Applications may be developed on one implementation and deployed on another. Or details affecting stack size may change between versions of a single implementation.
> 2. Stack size may depend on code not in the application or library (eg a library using Thread.new that calls application code or application code that different versions or implementations of a library).
>
> This setting should be a VM configuration option, not a Ruby method API.
Thank you for your comment.
This issue was started from Feature #3187 "Allow dynamic Fiber stack
size" by mperham (sorry, I had needed to show it).
mperham wants to specify fiber size per fiber because current fiber's
default size is too small to do complicated procedures and no way to
change this size.
Current default fiber's machine stack is small because of two reasons.
1. On the 32bit CPU, there are address space restriction and
Only a few thousand of fibers (and threads) can be made
if we allocate 1MB per fibers.
On the Fiber's usage, it is too small in some cases.
2. 1MB memory allocation hit performance in little.
(It is very weak reason for Rubyist)
To specify stack size, there are several way.
(1) VM configuration option (as Brian say)
(2) Global config by method (Thread.stack_size=) (as mperham proposed)
(3) Per-thread configuration (as I proposed)
I don't make any objection. It is acceptable.
However, we can't mix "small" and "large" size stack size in same
program. (2) is not perfect because it is not thread-safe.
BTW, the following API is thread-safe. Is it acceptable?
Thread.stack_size(size){
# only in this block, the stack size will be `size'
Thread.new(...){
...
}
}
On the #3187 thread, nagachika proposed more high level specifier
"factor" >https://bugs.ruby-lang.org/issues/3187#note-6>
> Fiber.stacksize = 2.0 # => twice the size of default stack size
I think using factor is also acceptable (the method name should be
changed, I guess). But we need to specify.
I also think more rough specifiers such as :small, :medium, :large are
also acceptable.
Fiber.new(size: :small){...}
(for example, at VM configuration with each sizes,
ruby -Xsmall_ss=4KB -Xmedium_ss=128KB -Xlarge_ss=1MB script
)
I don't care low-level or high-level specifier. But I think we need per
thread/fiber specifier.
----
From developer's perspective: We will increase default stack size for
Fiber (maybe same as Thread) after we introduce this feature. Most
case, larger stack size doesn't make problem (small stack size makes
problems such as mperham's situation).
If application or library needs huge number of Fibers, then this will
feature help.
----
> Thread#name doesn't need a new Thread.new API.
Yes. Thread#name and Thread#name= is enough if specify a name after
creation like:
Thread.new{
Thread.current.name = 'foo' # Thread.name may be also okay
...
}
But I want to specify a name at or before thread creation.
Is it only for me?
--
// SASADA Koichi at atdot dot net
