Feature #4560
closed[PATCH] lib/net/protocol.rb: avoid exceptions in rbuf_fill
Description
Blindly hitting IO#read_nonblock() and raising is expensive due
to two factors:
- method cache being scanned/cleared when the IO::WaitReadable
extended class is GC-ed - backtrace generation
This reduces the likelyhood of an IO::WaitReadable exception,
but spurious wakeup can still occur due to bad TCP checksums.
This optimization only applies to non-OpenSSL sockets. I am
using IO#wait here instead of IO.select() since IO#wait is not
available on OpenSSL sockets.
Files
Updated by headius (Charles Nutter) over 13 years ago
=begin
This is an interesting one. JRuby recently changed how we generate backtraces to using the Java backtrace as the master. This means our backtraces are as expensive to generate as a full Java backtrace for the full stack (think generating a backtrace for all Ruby and C and intermediate calls in Ruby). As a result, any algorithms that generate backtraces as part of normal flow control took a big perf hit.
On JRuby master, I've made a change that does not generate backtraces for EAGAIN, to avoid the overhead of generating it for the expected case of read_nonblock having nothing available. But it's a bit of a band-aid. The overhead from even creating an exception can be weigh into a tight loop over read_nonblock when there's nothing available, and of course having the backtrace disabled could annoy someone if it leaked out (JRuby points them to a flag to turn the backtraces on). Not sure what's the best long-term solution.
Also, the 1.9 practice of mixing in WaitReadable is really dreadful. It's bad enough in JRuby that it has to construct a new singleton class for every raised exception, but the cache effects in 1.9 are really painful.
=end
Updated by normalperson (Eric Wong) over 13 years ago
=begin
redmine@ruby-lang.org wrote:
On JRuby master, I've made a change that does not generate backtraces
for EAGAIN, to avoid the overhead of generating it for the expected
case of read_nonblock having nothing available. But it's a bit of a
band-aid. The overhead from even creating an exception can be weigh
into a tight loop over read_nonblock when there's nothing available,
and of course having the backtrace disabled could annoy someone if it
leaked out (JRuby points them to a flag to turn the backtraces on).
Not sure what's the best long-term solution.
Perhaps something similar to the kgio[1] API with IO#tryread/trywrite
(that return :wait_readable/:wait_writable Symbols) can be moved into
the core Ruby IO class (and deprecate IO#read_nonblock/write_nonblock).
Also, the 1.9 practice of mixing in WaitReadable is really dreadful.
It's bad enough in JRuby that it has to construct a new singleton
class for every raised exception, but the cache effects in 1.9 are
really painful.
Yes, I've been trying to fix that in trunk, too: [ruby-core:35672]
[1] http://bogomips.org/kgio - *nix-only, but the API is fully RDoc-ed
--
Eric Wong
=end
Updated by mame (Yusuke Endoh) over 12 years ago
- Status changed from Open to Assigned
- Assignee set to akr (Akira Tanaka)
Updated by akr (Akira Tanaka) about 12 years ago
- Target version changed from 2.0.0 to 2.6
Updated by mame (Yusuke Endoh) almost 7 years ago
- Assignee changed from akr (Akira Tanaka) to normalperson (Eric Wong)
@normalperson (Eric Wong), I think that this issue has been fixed more elegantly by introducing read_nonblock(exception: false)
. Am I right?
Updated by jeremyevans0 (Jeremy Evans) about 3 years ago
- Status changed from Assigned to Closed