[#53893] [ruby-trunk - Bug #8204][Open] ObjectSpace.each_object(Bignum) can generate Bignums that are to small to be Bignums — "Hanmac (Hans Mackowiak)" <hanmac@...>
5 messages
2013/04/02
[#53914] [ruby-trunk - Feature #8206][Open] Should Ruby core implement String#blank? — "sam.saffron (Sam Saffron)" <sam.saffron@...>
10 messages
2013/04/03
[#53922] [ruby-trunk - Bug #8208][Open] Raise cached exceptions for nonblocking IO to avoid allocation/stack-copying costs — "headius (Charles Nutter)" <headius@...>
7 messages
2013/04/03
[#54349] [ruby-trunk - Bug #8208] Raise cached exceptions for nonblocking IO to avoid allocation/stack-copying costs
— "headius (Charles Nutter)" <headius@...>
2013/04/16
[#54374] Re: [ruby-trunk - Bug #8208] Raise cached exceptions for nonblocking IO to avoid allocation/stack-copying costs
— Eric Wong <normalperson@...>
2013/04/17
"headius (Charles Nutter)" <headius@headius.com> wrote:
[#53950] [ruby-trunk - Bug #8211][Open] Performance regression of method calls — "dunric (David Unric)" <dunric29a@...>
9 messages
2013/04/03
[#53974] [ruby-trunk - Feature #8215][Open] Support accessing Fiber-locals and backtraces for a Fiber — "halorgium (Tim Carey-Smith)" <ruby-lang-bugs@...>
14 messages
2013/04/03
[#54023] [ruby-trunk - Feature #8223][Open] Make Matrix more omnivorous. — "boris_stitnicky (Boris Stitnicky)" <boris@...>
9 messages
2013/04/05
[#54031] Question about r39944 — Aaron Patterson <tenderlove@...>
Hi,
9 messages
2013/04/05
[#54032] Re: Question about r39944
— "NARUSE, Yui" <naruse@...>
2013/04/05
Even if test directory should be on the load path on test-all, you should
[#54095] [ruby-trunk - Feature #8237][Open] Logical method chaining via inferred receiver — "wardrop (Tom Wardrop)" <tom@...>
34 messages
2013/04/08
[#54097] [ruby-trunk - Feature #8237] Logical method chaining via inferred receiver
— "rosenfeld (Rodrigo Rosenfeld Rosas)" <rr.rosas@...>
2013/04/08
[#54175] [ruby-trunk - Bug #8254][Open] Ruby segfaults on second SystemStackError from parser — "charliesome (Charlie Somerville)" <charlie@...>
3 messages
2013/04/11
[#54185] [CommonRuby - Feature #8257][Open] Exception#cause to carry originating exception along with new one — "headius (Charles Nutter)" <headius@...>
43 messages
2013/04/11
[#54629] Re: [CommonRuby - Feature #8257][Open] Exception#cause to carry originating exception along with new one
— SASADA Koichi <ko1@...>
2013/04/27
(2013/04/12 1:40), headius (Charles Nutter) wrote:
[#54637] Re: [CommonRuby - Feature #8257][Open] Exception#cause to carry originating exception along with new one
— Charles Oliver Nutter <headius@...>
2013/04/27
On Sat, Apr 27, 2013 at 5:19 PM, SASADA Koichi <ko1@atdot.net> wrote:
[#54196] Encouraging use of CommonRuby — Charles Oliver Nutter <headius@...>
I think we need to do more to encourage the use of the CommonRuby
20 messages
2013/04/11
[#54200] Re: Encouraging use of CommonRuby
— Marc-Andre Lafortune <ruby-core-mailing-list@...>
2013/04/11
Hi,
[#54214] Re: Encouraging use of CommonRuby
— Charles Oliver Nutter <headius@...>
2013/04/12
On Thu, Apr 11, 2013 at 3:50 PM, Marc-Andre Lafortune
[#54211] Re: Encouraging use of CommonRuby
— "NARUSE, Yui" <naruse@...>
2013/04/12
As far as I understand, what is CommonRuby and the process over CommonRuby
[#54215] Re: Encouraging use of CommonRuby
— Charles Oliver Nutter <headius@...>
2013/04/12
On Thu, Apr 11, 2013 at 11:25 PM, NARUSE, Yui <naruse@airemix.jp> wrote:
[#54220] Re: Encouraging use of CommonRuby
— "NARUSE, Yui" <naruse@...>
2013/04/12
(2013/04/12 16:40), Charles Oliver Nutter wrote:
[#54232] Re: Encouraging use of CommonRuby
— Charles Oliver Nutter <headius@...>
2013/04/12
On Fri, Apr 12, 2013 at 8:08 AM, NARUSE, Yui <naruse@airemix.jp> wrote:
[#54201] Has ObjectSpace changed recently? — Dave Thomas <dave@...>
I just noticed that in 2.0, I see this:
5 messages
2013/04/11
[#54207] [CommonRuby - Feature #8258][Open] Dir#escape_glob — "steveklabnik (Steve Klabnik)" <steve@...>
15 messages
2013/04/12
[#54266] [CommonRuby - Feature #8258] Dir#escape_glob
— "rkh (Konstantin Haase)" <me@...>
2013/04/14
[#54227] [CommonRuby - Feature #8258] Dir#escape_glob
— "rkh (Konstantin Haase)" <me@...>
2013/04/12
[#54218] [CommonRuby - Feature #8259][Open] Atomic attributes accessors — "funny_falcon (Yura Sokolov)" <funny.falcon@...>
43 messages
2013/04/12
[#63015] [CommonRuby - Feature #8259] Atomic attributes accessors
— headius@...
2014/06/09
Issue #8259 has been updated by Charles Nutter.
[#63016] Re: [CommonRuby - Feature #8259] Atomic attributes accessors
— Eric Wong <normalperson@...>
2014/06/09
I'm not sure if setting the attribute on the ivar is a good way to go.
[#54333] Requesting Commit Access — Aman Gupta <ruby@...1.net>
Hello ruby-core,
16 messages
2013/04/16
[#54368] Re: Requesting Commit Access
— Yukihiro Matsumoto <matz@...>
2013/04/16
Hi,
[#54415] [ruby-trunk - Bug #8286][Open] Can't decode non-MIME Base64 — "adacosta (Alan Da Costa)" <alandacosta@...>
10 messages
2013/04/17
[#54459] [CommonRuby - Feature #8291][Open] Allow retrieving the root Fiber of a Thread — "halorgium (Tim Carey-Smith)" <ruby-lang@...>
5 messages
2013/04/19
[#54473] [Backport 200 - Backport #8299][Open] Minor error in float parsing — "bobjalex (Bob Alexander)" <bobjalex@...>
27 messages
2013/04/19
[#54549] [ruby-trunk - Bug #8299] Minor error in float parsing
— "phasis68 (Heesob Park)" <phasis@...>
2013/04/24
[#54496] [ruby-trunk - Bug #8299] Minor error in float parsing
— "naruse (Yui NARUSE)" <naruse@...>
2013/04/22
[#54509] [ruby-trunk - Bug #8310][Open] resque-web crashes with segfault on Ruby 2.0.0-p0 only, Resque 1.24.1, Redis 2.6.12 — "vaharoni (Amit Aharoni)" <amit.sites@...>
5 messages
2013/04/22
[#54559] [ruby-trunk - Feature #8321][Open] Ripper: I would like coordinates for keywords — "ericp (Eric Promislow)" <eric.promislow@...>
7 messages
2013/04/24
[#54606] Plan to the first 2.0.0 patchlevel release. — Tomoyuki Chikanaga <nagachika00@...>
Hello, Rubyists.
8 messages
2013/04/26
[#54743] Re: Plan to the first 2.0.0 patchlevel release.
— Marc-Andre Lafortune <ruby-core-mailing-list@...>
2013/05/02
Hi,
[#54815] Re: Plan to the first 2.0.0 patchlevel release.
— Aman Gupta <ruby@...1.net>
2013/05/06
Could you please backport the following:
[#54621] [ruby-trunk - Feature #8339][Open] Introducing Geneartional Garbage Collection for CRuby/MRI — "ko1 (Koichi Sasada)" <redmine@...>
43 messages
2013/04/27
[#54643] [ruby-trunk - Feature #8339] Introducing Geneartional Garbage Collection for CRuby/MRI
— "authorNari (Narihiro Nakamura)" <authorNari@...>
2013/04/28
[#54649] Re: [ruby-trunk - Feature #8339] Introducing Geneartional Garbage Collection for CRuby/MRI
— SASADA Koichi <ko1@...>
2013/04/28
(2013/04/28 9:23), authorNari (Narihiro Nakamura) wrote:
[#54770] Re: [ruby-trunk - Feature #8339] Introducing Geneartional Garbage Collection for CRuby/MRI
— Narihiro Nakamura <authornari@...>
2013/05/04
2013/4/28 SASADA Koichi <ko1@atdot.net>:
[#54772] Re: [ruby-trunk - Feature #8339] Introducing Geneartional Garbage Collection for CRuby/MRI
— SASADA Koichi <ko1@...>
2013/05/04
(2013/05/04 12:08), Narihiro Nakamura wrote:
[#54813] Re: [ruby-trunk - Feature #8339] Introducing Geneartional Garbage Collection for CRuby/MRI
— Tanaka Akira <akr@...>
2013/05/06
2013/5/4 SASADA Koichi <ko1@atdot.net>:
[#54816] Re: [ruby-trunk - Feature #8339] Introducing Geneartional Garbage Collection for CRuby/MRI
— SASADA Koichi <ko1@...>
2013/05/06
(2013/05/06 11:50), Tanaka Akira wrote:
[#54822] Re: [ruby-trunk - Feature #8339] Introducing Geneartional Garbage Collection for CRuby/MRI
— Tanaka Akira <akr@...>
2013/05/06
2013/5/6 SASADA Koichi <ko1@atdot.net>:
[#54706] [ruby-trunk - Feature #8339] Introducing Geneartional Garbage Collection for CRuby/MRI
— "sam.saffron (Sam Saffron)" <sam.saffron@...>
2013/04/30
[#54657] Re: [ruby-trunk - Feature #8339][Open] Introducing Geneartional Garbage Collection for CRuby/MRI
— Magnus Holm <judofyr@...>
2013/04/28
On Sat, Apr 27, 2013 at 8:19 PM, ko1 (Koichi Sasada)
[#54662] Re: [ruby-trunk - Feature #8339][Open] Introducing Geneartional Garbage Collection for CRuby/MRI
— SASADA Koichi <ko1@...>
2013/04/28
(2013/04/28 21:40), Magnus Holm wrote:
[#54669] Re: [ruby-trunk - Feature #8339][Open] Introducing Geneartional Garbage Collection for CRuby/MRI
— SASADA Koichi <ko1@...>
2013/04/28
(2013/04/28 23:34), SASADA Koichi wrote:
[#54671] Re: [ruby-trunk - Feature #8339][Open] Introducing Geneartional Garbage Collection for CRuby/MRI
— Magnus Holm <judofyr@...>
2013/04/28
On Sun, Apr 28, 2013 at 6:07 PM, SASADA Koichi <ko1@atdot.net> wrote:
[#54672] Re: [ruby-trunk - Feature #8339][Open] Introducing Geneartional Garbage Collection for CRuby/MRI
— SASADA Koichi <ko1@...>
2013/04/28
(2013/04/29 1:19), Magnus Holm wrote:
[#54673] Re: [ruby-trunk - Feature #8339][Open] Introducing Geneartional Garbage Collection for CRuby/MRI
— Magnus Holm <judofyr@...>
2013/04/28
On Sun, Apr 28, 2013 at 6:29 PM, SASADA Koichi <ko1@atdot.net> wrote:
[#54676] Re: [ruby-trunk - Feature #8339][Open] Introducing Geneartional Garbage Collection for CRuby/MRI
— Eric Wong <normalperson@...>
2013/04/28
"ko1 (Koichi Sasada)" <redmine@ruby-lang.org> wrote:
[#54712] [ruby-trunk - Feature #8339] Introducing Geneartional Garbage Collection for CRuby/MRI
— "sam.saffron (Sam Saffron)" <sam.saffron@...>
2013/04/30
[#54665] [ruby-trunk - Bug #8344][Open] Status of Psych and Syck — "Eregon (Benoit Daloze)" <redmine@...>
18 messages
2013/04/28
[ruby-core:54349] [ruby-trunk - Bug #8208] Raise cached exceptions for nonblocking IO to avoid allocation/stack-copying costs
From:
"headius (Charles Nutter)" <headius@...>
Date:
2013-04-16 15:33:53 UTC
List:
ruby-core #54349
Issue #8208 has been updated by headius (Charles Nutter).
No comments?
Perf change for an implementation of this in JRuby:
Before:
$ jruby -rbenchmark -rsocket -e "s = TCPSocket.new('google.com', 80); 10.times { puts Benchmark.measure { 100_000.times { begin; s.read_nonblock(1000); rescue; end } } }"
3.020000 0.300000 3.320000 ( 1.967000)
0.610000 0.200000 0.810000 ( 0.725000)
0.620000 0.200000 0.820000 ( 0.737000)
0.540000 0.190000 0.730000 ( 0.719000)
0.580000 0.200000 0.780000 ( 0.756000)
0.570000 0.200000 0.770000 ( 0.748000)
0.560000 0.190000 0.750000 ( 0.743000)
0.560000 0.200000 0.760000 ( 0.741000)
0.570000 0.190000 0.760000 ( 0.747000)
0.570000 0.200000 0.770000 ( 0.747000)
After:
$ jruby -rbenchmark -rsocket -e "s = TCPSocket.new('google.com', 80); 10.times { puts Benchmark.measure { 100_000.times { begin; s.read_nonblock(1000); rescue; end } } }"
1.750000 0.230000 1.980000 ( 1.216000)
0.360000 0.180000 0.540000 ( 0.449000)
0.370000 0.190000 0.560000 ( 0.493000)
0.320000 0.180000 0.500000 ( 0.445000)
0.330000 0.170000 0.500000 ( 0.448000)
0.280000 0.170000 0.450000 ( 0.436000)
0.270000 0.160000 0.430000 ( 0.434000)
0.280000 0.170000 0.450000 ( 0.434000)
0.270000 0.160000 0.430000 ( 0.433000)
0.270000 0.170000 0.440000 ( 0.431000)
----------------------------------------
Bug #8208: Raise cached exceptions for nonblocking IO to avoid allocation/stack-copying costs
https://bugs.ruby-lang.org/issues/8208#change-38625
Author: headius (Charles Nutter)
Status: Open
Priority: Normal
Assignee:
Category:
Target version:
ruby -v: 2.0.0
Currently, all nonblocking IO APIs raise exceptions when the IO channel cannot perform the requested read or write without blocking. This adds a tremendous performance hit to nonblocking IO, since raising an exception requires both allocating the exception object and saving off the backtrace information from the call stack.
Two changes could reduce this cost:
1. Don't provide a full backtrace for these exceptions, or only provide it in a debug or verbose modes. 99% of the rescuers of these exceptions don't ever use the backtrace, so it's wasted overhead.
2. Always raise a cached exception. This completely avoids all allocation costs for nonblocking operations that fail.
JRuby does #1 in JRuby 1.7 series, but does not do #2. I will work on a patch for MRI to do both #1 and #2.
The benefit of this change would be that nonblocking IO has zero wasted overhead. The down side would be that the exceptions raised do not have a useful backtrace unless debug or verbose.
If we did this change, we would not have to introduce a separate API for nonblocking IO that doesn't raise exceptions.
Thoughts?
--
http://bugs.ruby-lang.org/