Antiarc has quit [Ping timeout: 258 seconds]
Antiarc has joined #jruby
bga57 has joined #jruby
havenwood has quit [Quit: ZNC 1.7.3 - https://znc.in]
havenwood has joined #jruby
havenwood has joined #jruby
havenwood has quit [Changing host]
bga57 has quit [Quit: Leaving.]
jeremyevans has quit [Quit: leaving]
jeremyevans has joined #jruby
shellac has joined #jruby
claudiuinberlin has joined #jruby
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
drbobbeaty has joined #jruby
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
claudiuinberlin has joined #jruby
drbobbeaty has joined #jruby
claudiuinberlin has quit [Ping timeout: 252 seconds]
shellac has quit [Quit: Computer has gone to sleep.]
shellac has joined #jruby
shellac has quit [Quit: Computer has gone to sleep.]
shellac has joined #jruby
shellac has quit [Ping timeout: 248 seconds]
rtyler has left #jruby ["Kicked by @appservice-irc:matrix.org : issued !quit command"]
rusk has quit [Remote host closed the connection]
subbu is now known as subbu|lunch
subbu|lunch is now known as subbu
<lopex> I gather the release is imminent
<lopex> ?
<enebo> lopex: trying to finish stuff but not imminent
enebo has left #jruby [#jruby]
enebo has joined #jruby
travis-ci has joined #jruby
travis-ci has left #jruby [#jruby]
<travis-ci> jruby/jruby (load_service_redux:3149d04 by Charles Oliver Nutter): The build has errored. https://travis-ci.org/jruby/jruby/builds/544424575 [237 min 40 sec]
<lopex> enebo: ah
<lopex> 6 planes and whole Poland is covered the whole time
<lopex> headius: ^^
<lopex> kares: you're too ^^
travis-ci has joined #jruby
<travis-ci> jruby/jruby (load_service_redux:3149d04 by Charles Oliver Nutter): The build has errored. https://travis-ci.org/jruby/jruby/builds/544424575 [270 min 13 sec]
travis-ci has left #jruby [#jruby]
<headius[m]> so close
<headius[m]> autoload specs are all green
<headius[m]> two minor things tagged in MRI autoload tests
<lopex> close to release ?
<lopex> still having and issue with puma and websockets, once tried to nail it down but failed
<lopex> s/and/an/
<headius[m]> yeah close to release
<headius[m]> websockets?
<lopex> something to do with hash/randoms
<lopex> yeah
<headius[m]> did we tell you about the problems we have had benchmarking puma? Seems to be issues with HTTP parsing or keepalive or something
<lopex> hmm
<headius[m]> we had a meeting at RailsConf and I think some work is being done but I haven't followed up
<headius[m]> what's the issue you're having?
<lopex> not exactly sure
<lopex> just narrowed it a bit in my mind
<lopex> it has to do with the cookies and/or ws ids
<lopex> but havent been able to repro a minimal case
<lopex> maybe hijacking doesnt work ?
<lopex> hijacking is a ws method for intercepting http calls so it can wokr on the same port
<lopex> more like a tcp level
<lopex> that's what puma on mri does for sure
<headius[m]> doesn't sound like my issue
<headius[m]> well maybe not
<headius[m]> I guess I don't know how puma's web socket support is at all really
<lopex> but hash/randoms might be
<lopex> yeah
<lopex> not blaming anyome atm :P
<lopex> same app
<lopex> works on mri
<lopex> and it's simple, just sinatra
<lopex> so rails shit goes out
<headius[m]> definitely could be some difference in the puma http stuff
<lopex> yeah
<headius[m]> there's a few "if JRuby" in puma, plus they use different exts for http parsing
<lopex> but I think we regressed at some point though
<lopex> or puma
<lopex> or it might be, mri changed and puma followed
<lopex> dunno
<lopex> headius[m]: ragel is not the most beloved nowadays ?
<lopex> headius[m]: maybe we should use http://kaitai.io/ for some stuff ?
<lopex> the gallery is not the most impressive though
<headius[m]> ragel is crap for Java parsers
<lopex> that's lost of commits
<headius[m]> it generates a funky state machine object and big switch because it can't do goto
<lopex> headius[m]: have you seen kaitai ?
<headius[m]> it's slower than the C/++ versions by a lot
<lopex> not sure about perf of it
<headius[m]> lopex: sounds like protobuf sort of thing?
<lopex> yeah
<headius[m]> my #1 target for a protobuf would be to make a zero-alloc IR interpreter from disk
<headius[m]> you just point at a memory-mapped serialized blob and start executing
travis-ci has joined #jruby
travis-ci has left #jruby [#jruby]
<travis-ci> jruby/jruby (load_service_redux:aea0e65 by Charles Oliver Nutter): The build failed. https://travis-ci.org/jruby/jruby/builds/544444268 [220 min 42 sec]
<headius[m]> I tried to do it but it's hard to represent polymorphic objects
<headius[m]> e.g. instructions, operands
<headius[m]> best I could do was to make them all fixed with with a type field, which is pretty wasteful
<lopex> yeah, just was about to ask about IR
<headius[m]> fixed width
<lopex> aka blitting in old terms
<lopex> is that right ?
<headius[m]> yeah
<headius[m]> I suppose that works
<lopex> kaitai has some interesting formats in their libs
<lopex> and an online ide to debug them
<lopex> but I'm not sure about their perf
<lopex> well, even as a tool it might come handy though
<lopex> for builds etc
<lopex> headius[m]: so, essentially you want the IR to collapse to just fread ?
<headius[m]> yeah
<lopex> hmm, and then keep ifnodes for fields that steer unions and the like ?
<lopex> just trying to grasp where the limit is
<lopex> the formats can be really bizarre
travis-ci has joined #jruby
travis-ci has left #jruby [#jruby]
<travis-ci> jruby/jruby (load_service_redux:586b07c by Charles Oliver Nutter): The build failed. https://travis-ci.org/jruby/jruby/builds/544449216 [222 min 26 sec]
<lopex> headius[m]: so the main diff is getConstantWithAutoload and fetchConstant wrt that pr ?
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<lopex> that IntList is a temp thing since you're always do toIntArray on it
<lopex> lol that pr is huge