[#58149] [ruby-trunk - Feature #9076][Open] New one-argument block syntax: &. — "asterite (Ary Borenszweig)" <ary@...>
23 messages
2013/11/04
[#58259] [ruby-trunk - Feature #9099][Open] Train emoji lambda operator — "charliesome (Charlie Somerville)" <charliesome@...>
9 messages
2013/11/10
[#58312] [ruby-trunk - Feature #9107][Open] Introduce YES and NO as aliases of true and false — "gsamokovarov (Genadi Samokovarov)" <gsamokovarov@...>
5 messages
2013/11/13
[#58350] [ruby-trunk - Feature #9113][Open] Ship Ruby for Linux with jemalloc out-of-the-box — "sam.saffron (Sam Saffron)" <sam.saffron@...>
59 messages
2013/11/15
[#60851] Re: [ruby-trunk - Feature #9113][Open] Ship Ruby for Linux with jemalloc out-of-the-box
— Eric Wong <normalperson@...>
2014/02/19
Btw, I also hope to experiment with a slab allocator since many internal
[#62721] [ruby-trunk - Feature #9113] Ship Ruby for Linux with jemalloc out-of-the-box
— nobu@...
2014/05/24
Issue #9113 has been updated by Nobuyoshi Nakada.
[#62735] [ruby-trunk - Feature #9113] Ship Ruby for Linux with jemalloc out-of-the-box
— normalperson@...
2014/05/25
Issue #9113 has been updated by Eric Wong.
[#58391] [ruby-trunk - Bug #9119][Assigned] TestTime#test_marshal_broken_offset broken under MinGW — "luislavena (Luis Lavena)" <luislavena@...>
10 messages
2013/11/17
[#58396] [ruby-trunk - Bug #9121][Open] [PATCH] Remove rbtree implementation of SortedSet due to performance regression — "xshay (Xavier Shay)" <contact@...>
15 messages
2013/11/18
[#58404] [ruby-trunk - Feature #9123][Open] Make Numeric#nonzero? behavior consistent with Numeric#zero? — "sferik (Erik Michaels-Ober)" <sferik@...>
40 messages
2013/11/18
[#58411] [ruby-trunk - Bug #9124][Open] TestSocket errors in test-all on Arch 64-bit — "jonforums (Jon Forums)" <redmine@...>
14 messages
2013/11/18
[#58515] [ruby-trunk - Bug #9124] TestSocket errors in test-all on Arch 64-bit
— "jonforums (Jon Forums)" <redmine@...>
2013/11/23
[#58841] [ruby-trunk - Bug #9124] TestSocket errors in test-all on Arch 64-bit
— "jonforums (Jon Forums)" <redmine@...>
2013/12/04
[#58842] Re: [ruby-trunk - Bug #9124] TestSocket errors in test-all on Arch 64-bit
— Eric Wong <normalperson@...>
2013/12/04
"jonforums (Jon Forums)" <[email protected]> wrote:
[#58452] [ruby-trunk - Bug #9133][Open] logger rotates log files more than expected — "no6v (Nobuhiro IMAI)" <nov@...>
8 messages
2013/11/21
[#58473] Object identity for string hash keys — Andrew Vit <andrew@...>
I'm not sure if this is a bug. I'm creating a hash like this:
5 messages
2013/11/21
[#58490] Re: [ruby-cvs:50910] drbrain:r43767 (trunk): * lib/rubygems: Update to RubyGems master 50a8210. Important changes — Tanaka Akira <akr@...>
2013/11/22 <[email protected]>:
4 messages
2013/11/22
[#58492] Re: [ruby-cvs:50910] drbrain:r43767 (trunk): * lib/rubygems: Update to RubyGems master 50a8210. Important changes
— Eric Wong <normalperson@...>
2013/11/22
Tanaka Akira <[email protected]> wrote:
[#58496] [ruby-trunk - Feature #9140][Open] Allow each_with_index to get start index — "rosenfeld (Rodrigo Rosenfeld Rosas)" <rr.rosas@...>
8 messages
2013/11/22
[#58545] [ruby-trunk - Feature #9145][Open] Queue#pop(true) return nil if empty instead of raising ThreadError — "jsc (Justin Collins)" <redmine@...>
9 messages
2013/11/24
[#58599] [ruby-trunk - Bug #9159][Open] [patch] use rb_fstring for internal strings — "tmm1 (Aman Gupta)" <[email protected]>
5 messages
2013/11/26
[#58653] [ruby-trunk - Bug #9170][Open] Math.sqrt returns different types when mathn is included; breaks various gems - this bug can be reproduced in Ruby 1.8 as well — "kranzky (Jason Hutchens)" <JasonHutchens@...>
7 messages
2013/11/28
[#58719] [ruby-trunk - Feature #5446] at_fork callback API — "tmm1 (Aman Gupta)" <[email protected]>
6 messages
2013/11/30
[ruby-core:58362] [ruby-trunk - Feature #9113] Ship Ruby for Linux with jemalloc out-of-the-box
From:
"sam.saffron (Sam Saffron)" <sam.saffron@...>
Date:
2013-11-15 07:16:59 UTC
List:
ruby-core #58362
Issue #9113 has been updated by sam.saffron (Sam Saffron). @martin this is a great "oldish" article by facebook about this http://www.facebook.com/notes/facebook-engineering/scalable-memory-allocation-using-jemalloc/480222803919 @nobu I guess my suggestion here is to include jemalloc source in the the repo, and compile on demand for linux (by default with an option to opt-out) that way everyone will pick this change up and it becomes "officially blessed" allocator. At Github @tmm1 has been using tcmalloc for years now, fragmentation is less good than jemalloc (tmm1 said perf is better, but I think its time to re-test cause I found jemalloc to perform better) Regardless, libc allocator is a problem and default compiles should not use it. ---------------------------------------- Feature #9113: Ship Ruby for Linux with jemalloc out-of-the-box https://bugs.ruby-lang.org/issues/9113#change-42959 Author: sam.saffron (Sam Saffron) Status: Third Party's Issue Priority: Normal Assignee: Category: build Target version: libc's malloc is a problem, it fragments badly meaning forks share less memory and is slow compared to tcmalloc or jemalloc. both jemalloc and tcmalloc are heavily battle tested and stable. 2 years ago redis picked up the jemalloc dependency see: http://oldblog.antirez.com/post/everything-about-redis-24.html To quote antirez: `` But an allocator is a serious thing. Since we introduced the specially encoded data types Redis started suffering from fragmentation. We tried different things to fix the problem, but basically the Linux default allocator in glibc sucks really, really hard. `` --- I recently benched Discourse with tcmalloc / jemalloc and default and noticed 2 very important thing: median request time reduce by up to 10% (under both) PSS (proportional share size) is reduced by 10% under jemalloc and 8% under tcmalloc. We can always use LD_PRELOAD to yank these in, but my concern is that standard distributions are using a far from optimal memory allocator. It would be awesome if the build, out-of-the-box, just checked if it was on Linux (eg: https://github.com/antirez/redis/blob/unstable/src/Makefile#L30-L34 ) and then used jemalloc instead. -- http://bugs.ruby-lang.org/