[#101981] [Ruby master Bug#17519] set_visibility fails when a prepended module and a refinement both exist — dbfeldman@...

Issue #17519 has been reported by fledman (David Feldman).

12 messages 2021/01/08

[#102003] [Ruby master Bug#17527] rb_io_wait_readable/writable with scheduler don't check errno — julien@...

Issue #17527 has been reported by ysbaddaden (Julien Portalier).

13 messages 2021/01/11

[#102065] [Ruby master Bug#17536] Segfault in `CFUNC :define_method` — v.ondruch@...

Issue #17536 has been reported by vo.x (Vit Ondruch).

13 messages 2021/01/13

[#102083] [Ruby master Bug#17540] A segfault due to Clang/LLVM optimization on 32-bit ARM Linux — xtkoba+ruby@...

Issue #17540 has been reported by xtkoba (Tee KOBAYASHI).

12 messages 2021/01/14

[#102102] [Ruby master Bug#17543] Ractor isolation broken by `self` in shareable proc — marcandre-ruby-core@...

Issue #17543 has been reported by marcandre (Marc-Andre Lafortune).

14 messages 2021/01/15

[#102118] [Ruby master Feature#17548] Need simple way to include symlink directories in Dir.glob — keithrbennett@...

Issue #17548 has been reported by keithrbennett (Keith Bennett).

8 messages 2021/01/17

[#102158] [Ruby master Bug#17560] Does `Module#ruby2_keywords` return `nil` or `self`? — nobu@...

Issue #17560 has been reported by nobu (Nobuyoshi Nakada).

9 messages 2021/01/19

[#102163] [Ruby master Bug#17561] The timeout option for Addrinfo.getaddrinfo is not reliable on Ruby 2.7.2 — sean@...

Issue #17561 has been reported by smcgivern (Sean McGivern).

8 messages 2021/01/19

[#102249] [Ruby master Bug#17583] Segfault on large stack(RUBY_THREAD_VM_STACK_SIZE) — yoshiokatsuneo@...

Issue #17583 has been reported by yoshiokatsuneo (Tsuneo Yoshioka).

12 messages 2021/01/26

[#102256] [Ruby master Bug#17585] DWAR5 support? — v.ondruch@...

Issue #17585 has been reported by vo.x (Vit Ondruch).

19 messages 2021/01/26

[#102301] [Ruby master Bug#17591] Test frameworks and REPLs do not show deprecation warnings by default — eregontp@...

Issue #17591 has been reported by Eregon (Benoit Daloze).

14 messages 2021/01/29

[#102305] [Ruby master Feature#17592] Ractor should allowing reading shareable class instance variables — marcandre-ruby-core@...

Issue #17592 has been reported by marcandre (Marc-Andre Lafortune).

25 messages 2021/01/29

[ruby-core:102339] [Ruby master Bug#17497] Ractor performance issue

From: keithrbennett@...
Date: 2021-01-31 20:57:03 UTC
List: ruby-core #102339
Issue #17497 has been updated by keithrbennett (Keith Bennett).


@ko1 - My apologies for not responding sooner. I guess I have not configured this forum correctly to receive notifications, I'll look into that.

I've tested my benchmark against Ruby head, and performance with multiple cores seem to have degraded. Perhaps I have made an error in my approaches, I don't know. I will paste my results below. My OS is "Ubuntu 20.04.2 LTS" (Kubuntu).

In case it's useful, I've made my script easier to use; it automatically tests and compares 1 ractor with (CPU_count) ractors. You can find it at https://github.com/keithrbennett/keithrbennett-ractor-test/blob/master/ractor-file-strings-test.rb. Information about configuring it, what it does, etc., is included in comments at the top of the script.
```

Small Data Set:

Ruby 3.0.0:

                        1 CPU         24 CPU's           Factor
----------------------------------------------------------------
User                  6.75500         52.34200          7.74863
System                0.00000          0.02800          0.00415
Total                 6.83900         52.40700          7.66296
Real                  6.83300          4.10500          0.60076


2021-01-31 Ruby head (ruby 3.1.0dev (2021-01-31T09:48:28Z master 22b8ddfd10) [x86_64-linux]):

                        1 CPU         24 CPU's           Factor
----------------------------------------------------------------
User                  6.18000         56.27400          9.10583
System                0.00400          0.02800          0.00453
Total                 6.26800         56.34200          8.98883
Real                  6.26100          4.26200          0.68072


================================================================

Larger Data Set:

Ruby 3.0.0:

                        1 CPU         24 CPU's           Factor
----------------------------------------------------------------
User                 51.01000        499.67200          9.79557
System                0.04000          0.25900          0.00508
Total                51.32300        500.12900          9.74473
Real                 51.31200         45.56600          0.88802


2021-01-31 Ruby head (ruby 3.1.0dev (2021-01-31T09:48:28Z master 22b8ddfd10) [x86_64-linux]):

                        1 CPU         24 CPU's           Factor
----------------------------------------------------------------
User                 47.08900        486.34400         10.32819
System                0.03200          0.20300          0.00431
Total                47.39500        486.74800         10.27003
Real                 47.38400         43.95100          0.92755

```

----------------------------------------
Bug #17497: Ractor performance issue
https://bugs.ruby-lang.org/issues/17497#change-90194

* Author: marcandre (Marc-Andre Lafortune)
* Status: Closed
* Priority: Normal
* Assignee: ko1 (Koichi Sasada)
* ruby -v: ruby 3.0.0p0 (2020-12-25 revision 95aff21468) [x86_64-darwin18]
* Backport: 2.5: UNKNOWN, 2.6: UNKNOWN, 2.7: UNKNOWN
----------------------------------------
There's a strange performance issue with Ractor (at least on MacOS, didn't run on other OS).

I ran a benchmark doing 3 different types of work:
* "fib": method calls (naive fibonacci calculation)
* "cpu": `(0...1000).inject(:+)`
* "sleep": call `sleep`

I get the kind of results I was excepting for the `fib` and for sleeping, but the results for the "cpu" workload show a problem.

It is so slow that my pure Ruby backport (using Threads) is 65x faster  on my Mac Pro (despite having 6 cores). Expected results would be 6x slower, so in that case Ractor is 400x slower than it should 仭

On my MacBook (2 cores) the results are not as bad, the `cpu` workload is 3x faster with my pure-Ruby backport (only) instead of ~2x slower, so the factor is 6x too slow.


```
$ gem install backports
Successfully installed backports-3.20.0
1 gem installed
$ ruby ractor_test.rb
<internal:ractor>:267: warning: Ractor is experimental, and the behavior may change in future versions of Ruby! Also there are many implementation issues.
fib: 110 ms  |  cpu: 22900 ms  |  sleep: 206 ms
$ B=t ruby ractor_test.rb
Using pure Ruby implementation
fib: 652 ms  |  cpu: 337 ms  |  sleep: 209 ms
```

Notice the `sleep` run takes similar time, which is good, and `fib` is ~6x faster on my 6-core CPU (and ~2x faster on my 2-core MacBook), again that's good as the pure ruby version uses Threads and thus runs with a single GVL.

The `cpu` version is the problem.

Script is here: https://gist.github.com/marcandre/bfed626e538a3d0fc7cad38dc026cf0e



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

Unsubscribe: <mailto:[email protected]?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>

In This Thread