[#78633] ruby/spec needs help from CRuby committers — Benoit Daloze <eregontp@...>
Currently, ruby/spec is maintained mostly by individuals and enjoys the
13 messages
2016/12/13
[#78963] Re: ruby/spec needs help from CRuby committers
— Urabe Shyouhei <shyouhei@...>
2017/01/04
I did ask attendees of last developer meeting to join this
[#78642] Re: ruby/spec needs help from CRuby committers
— Eric Wong <normalperson@...>
2016/12/14
Benoit Daloze <eregontp@gmail.com> wrote:
[ruby-core:78921] [Ruby trunk Bug#13085] io.c io_fwrite creates garbage
From:
normalperson@...
Date:
2016-12-31 01:01:49 UTC
List:
ruby-core #78921
Issue #13085 has been updated by Eric Wong.
File 0001-io.c-io_fwrite-temporarily-freeze-string-when-writin.patch added
Proposed patch to temporarily freeze string while copying
~~~
io.c (io_fwrite): temporarily freeze string when writing
This avoids garbage from IO#write for [Bug #13085].
Memory usage from benchmark/bm_io_copy_stream_write.rb
is reduced greatly:
target 0: a (ruby 2.5.0dev (2016-12-30 trunk 57236) [x86_64-linux])
target 1: b (ruby 2.5.0dev (2016-12-30) [x86_64-linux])
Memory usage (last size) (B)
name a b
io_copy_stream_write 82235392.000 6651904.000
Memory consuming ratio (size) with the result of `a' (greater is better)
name b
io_copy_stream_write 12.363
There is also a speedup in execution time:
Execution time (sec)
name a b
io_copy_stream_write 0.380 0.143
Speedup ratio: compare with the result of `a' (greater is better)
name b
io_copy_stream_write 2.651
Caveat, there is one potential race condition:
If another thread calls String#freeze on the string we are
currently writing; we will blindly unfreeze it during
fwrite_unfreeze from ensure. However, I do not expect this to
be a real-world case.
Ideally, Ruby should have a way of detecting threads which
are not visible to other threads.
~~~
----------------------------------------
Bug #13085: io.c io_fwrite creates garbage
https://bugs.ruby-lang.org/issues/13085#change-62335
* Author: Eric Wong
* Status: Open
* Priority: Normal
* Assignee:
* Target version:
* ruby -v:
* Backport: 2.2: UNKNOWN, 2.3: UNKNOWN, 2.4: UNKNOWN
----------------------------------------
Relying on rb_str_new_frozen for unconverted strings does not
save memory because copy-on-write is always triggered in
read-write I/O loops were subsequent IO#read calls will
clobber the given write buffer.
buf = ''.b
while input.read(16384, buf)
output.write(buf)
end
This generates a lot of garbage starting with Ruby 2.2 (r44471).
For my use case, even IO.copy_stream generates garbage, since
I wrap "write" to do Digest calculation in a single pass.
I tried using rb_str_replace and reusing the string as a hidden
(klass == 0) thread-local, but rb_str_replace attempts CoW
optimization by creating new frozen objects, too:
https://80x24.org/spew/20161229004417.12304-1-e@80x24.org/raw
So, I'm not sure what to do, temporal locking seems wrong for
writing strings (I guess it's for reading?). I get
test_threaded_flush failures with the following:
https://80x24.org/spew/20161229005701.9712-1-e@80x24.org/raw
IO#syswrite has the same problem with garbage. I can use
IO#write_nonblock on fast filesystems while holding GVL,
I guess...
---Files--------------------------------
0001-io.c-io_fwrite-temporarily-freeze-string-when-writin.patch (2.6 KB)
--
https://bugs.ruby-lang.org/
Unsubscribe: <mailto:ruby-core-request@ruby-lang.org?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>