[#41916] Proposal: Bitmap Marking GC — Narihiro Nakamura <authornari@...>

Hi.

18 messages 2012/01/05

[#41941] [ruby-trunk - Bug #5851][Open] make check fails when compiling with GCC 4.7 - *** longjmp causes uninitialized stack frame *** — Vit Ondruch <v.ondruch@...>

12 messages 2012/01/06

[#41979] [ruby-trunk - Bug #5865][Open] Exception#== should return false if the classes differ — Hiro Asari <asari.ruby@...>

10 messages 2012/01/08

[#42003] [ruby-trunk - Bug #5871][Open] regexp \W matches some word characters when inside a case-insensitive character class — Gareth Adams <gareth@...>

14 messages 2012/01/09

[#42016] [ruby-trunk - Feature #5873][Open] Adopt FFI over DL — Heesob Park <phasis@...>

15 messages 2012/01/10

[#42149] [ruby-trunk - Feature #5899][Open] chaining comparsions. — Ondrej Bilka <neleai@...>

12 messages 2012/01/16

[#42164] [ruby-trunk - Feature #5903][Open] Optimize st_table (take 2) — Yura Sokolov <funny.falcon@...>

18 messages 2012/01/17

[ruby-core:42137] Re: Comments wanted: about merging https://github.com/ruby/ruby/pull/72

From: Eric Wong <normalperson@...>
Date: 2012-01-15 19:20:40 UTC
List: ruby-core #42137
Urabe Shyouhei <shyouhei@ruby-lang.org> wrote:
> Hello,  I'm going to merge this pull request "Writing entire packet at once
> to avoid incomplete transmission".
> 
> https://github.com/ruby/ruby/pull/72 
> 
> If you have any opinions let me know (either in this list or at the web).

Looks good to me.

Each @socket.writeline call will correspond to one write(2)
and that's bad if you have a lot of small data that is ready
to be written.

Since TCP_NODELAY isn't enabled in net/http (vs net/http/persistent),
Nagle's algorithm will also delay the second packet as it tries to
bundle subsequent lines together.  This pull request does that
in userspace (more efficiently).

If TCP_NODELAY were enabled, each @socket.writeline will have 40 bytes
of TCP/IP header overhead.  Not the end of the world, but not good,
either.

In This Thread