[#42344] [ruby-trunk - Feature #5964][Open] Make Symbols an Alternate Syntax for Strings — Tom Wardrop <tom@...>

23 messages 2012/02/03

[#42443] [ruby-trunk - Bug #5985][Open] miniruby skews "make benchmark" results — Eric Wong <normalperson@...>

21 messages 2012/02/08

[#42444] [ruby-trunk - Bug #5986][Open] Segmentation Fault — Luis Matta <levmatta@...>

16 messages 2012/02/08

[#42471] [ruby-trunk - Feature #5995][Open] calling io_advise_internal() in read_all() — Masaki Matsushita <glass.saga@...>

20 messages 2012/02/10

[#42560] [ruby-trunk - Bug #6011][Open] ruby-1.9.3-p0/lib/webrick/utils.rb:184: [BUG] Segmentation fault — Vit Ondruch <v.ondruch@...>

12 messages 2012/02/13

[#42579] [ruby-trunk - Bug #6012][Open] Proc#source_location also return the column — Roger Pack <rogerpack2005@...>

14 messages 2012/02/14

[#42685] [ruby-trunk - Bug #6036][Open] Test failures in Fedora Rawhide/17 — Bohuslav Kabrda <bkabrda@...>

14 messages 2012/02/16

[#42697] [ruby-trunk - Bug #6040][Open] Transcoding test failure: Big5 to UTF8 not defined (MinGW) — Luis Lavena <luislavena@...>

10 messages 2012/02/16

[#42813] [ruby-trunk - Feature #6065][Open] Allow Bignum marshalling/unmarshalling from C API — Martin Bosslet <Martin.Bosslet@...>

22 messages 2012/02/23

[#42815] [ruby-trunk - Bug #6066][Open] Fix "control may reach end of non-void function" warnings for clang — Eric Hodel <drbrain@...7.net>

15 messages 2012/02/23

[#42857] [ruby-trunk - Feature #6074][Open] Allow alias arguments to have a comma — Thomas Sawyer <transfire@...>

20 messages 2012/02/24

[#42891] [ruby-trunk - Feature #6083][Open] Hide a Bignum definition — Koichi Sasada <redmine@...>

23 messages 2012/02/25

[#42906] [ruby-trunk - Bug #6085][Open] Treatment of Wrong Number of Arguments — Marc-Andre Lafortune <ruby-core@...>

14 messages 2012/02/25

[#42949] [ruby-trunk - Bug #6089][Open] Test suite fails with OpenSSL 1.0.1 — Vit Ondruch <v.ondruch@...>

13 messages 2012/02/26

[ruby-core:42743] [ruby-trunk - Feature #5138] Add nonblocking IO that does not use exceptions for EOF and EWOULDBLOCK

From: Charles Nutter <headius@...>
Date: 2012-02-19 19:41:15 UTC
List: ruby-core #42743
Issue #5138 has been updated by Charles Nutter.


I never got to weigh in on this thread back in the day, and I'm working on IO stuff more recently, so a few thoughts...

Eric Wong wrote:
>  * Exceptions with the current *_nonblock methods are expensive.
>  
>    - extending with IO::Wait* increments the global VM state version

This is really poor form. I'm not sure why it hasn't been fixed yet. JRuby master solves this in a neat, compatible way: instead of raising singleton EAGAIN with WaitReable or WaitWritable mixed in, we raise EAGAINReadable or EAGAINWritable, subclasses that have already done the mixing. No other changes are needed.

>    - backtrace generation cost

In JRuby, EAGAIN does not generate a backtrace unless you pass a flag, since generally it's an "expected" exception and you don't want the backtrace anyway.

How does this translate to perf?

system ~/projects $ jruby eagain_bench.rb 
  1.553000   0.000000   1.553000 (  1.553000)
  0.673000   0.000000   0.673000 (  0.673000)
  0.667000   0.000000   0.667000 (  0.667000)
  0.669000   0.000000   0.669000 (  0.669000)
  0.696000   0.000000   0.696000 (  0.697000)

system ~/projects $ rvm 1.9.3 do ruby eagain_bench.rb 
  1.250000   0.150000   1.400000 (  1.396295)
  1.250000   0.150000   1.400000 (  1.408975)
  1.260000   0.150000   1.410000 (  1.405947)
  1.250000   0.150000   1.400000 (  1.405391)
  1.260000   0.150000   1.410000 (  1.407389)

system ~/projects $ rvm ruby-head do ruby eagain_bench.rb 
  1.310000   0.120000   1.430000 (  1.428627)
  1.300000   0.120000   1.420000 (  1.430652)
  1.310000   0.130000   1.440000 (  1.426011)
  1.310000   0.120000   1.430000 (  1.430642)
  1.310000   0.120000   1.430000 (  1.436441)

Rather well, I think.
----------------------------------------
Feature #5138: Add nonblocking IO that does not use exceptions for EOF and EWOULDBLOCK
https://bugs.ruby-lang.org/issues/5138

Author: Yehuda Katz
Status: Open
Priority: Normal
Assignee: Yukihiro Matsumoto
Category: core
Target version: 1.9.4


The current Ruby I/O classes have non-blocking methods (read_nonblock and write_nonblock). These methods will never block, and if they would block, they raise an exception instead (IO::WaitReadable or IO::WaitWritable). In addition, if the IO is at EOF, they raise an EOFError.

These exceptions are raised repeatedly in virtually every use of the non-blocking methods. This patch adds a pair of methods (try_read_nonblock and try_write_nonblock) that have the same semantics as the existing methods, but they return Symbols instead of raising exceptions for these routine cases:

* :read_would_block
* :write_would_block
* :eof

The patch contains updates for IO, StringIO, and OpenSSL. The updates are fully documented and tested.


-- 
http://bugs.ruby-lang.org/

In This Thread