pawnbox has quit [Ping timeout: 272 seconds]
andrewvc has joined #jruby
bruceadams has joined #jruby
deathy has joined #jruby
guilleiguaran__ has joined #jruby
ruurd has joined #jruby
ruurd is now known as Guest48944
dling` has joined #jruby
dling has quit [Ping timeout: 256 seconds]
<GitHub29> [jruby] segiddins opened issue #3631: Override of Gem.path_separator cause gem update --system to fail https://git.io/vgeDv
pawnbox has joined #jruby
pawnbox has quit [Ping timeout: 240 seconds]
nirvdrum has joined #jruby
pawnbox has joined #jruby
pawnbox has quit [Ping timeout: 250 seconds]
Puffball has quit [Read error: Connection reset by peer]
Puffball has joined #jruby
pawnbox has joined #jruby
tomjoro has joined #jruby
tomjoro has quit [Ping timeout: 272 seconds]
nirvdrum has quit [Ping timeout: 245 seconds]
dling has joined #jruby
Osho_ has joined #jruby
AlHafoudh_ has joined #jruby
dling` has quit [*.net *.split]
e_dub has quit [*.net *.split]
AlHafoudh_ is now known as AlHafoudh
skade has joined #jruby
skade has quit [Client Quit]
skade has joined #jruby
skade has quit [Client Quit]
e_dub has joined #jruby
halorgium has joined #jruby
thedarkone2 has quit [Quit: thedarkone2]
skade has joined #jruby
skade has quit [Read error: Connection reset by peer]
jeremyevans has quit [Read error: Connection reset by peer]
jeremyevans has joined #jruby
rsim has joined #jruby
rsim1 has joined #jruby
rsim has quit [Ping timeout: 240 seconds]
tomjoro has joined #jruby
pawnbox has quit [Remote host closed the connection]
pawnbox has joined #jruby
Guest48944 is now known as ruurd
blaxter has joined #jruby
jensnockert has joined #jruby
pawnbox has quit [Remote host closed the connection]
pawnbox has joined #jruby
vtunka has joined #jruby
ITXpander has joined #jruby
ITXpander has quit [Client Quit]
ITXpander has joined #jruby
lopex has joined #jruby
joast has joined #jruby
vifino has joined #jruby
<eregon> headius: Do I need any CLI option to run mandelbrot with unboxing?
drbobbeaty has joined #jruby
drbobbeaty has quit [Ping timeout: 272 seconds]
skade has joined #jruby
joast has quit [*.net *.split]
mj12` has joined #jruby
shellac has joined #jruby
joast has joined #jruby
<kares> bga57: you mean a debugger?
vtunka has quit [Quit: Leaving]
skade has quit [Ping timeout: 245 seconds]
vtunka has joined #jruby
skade has joined #jruby
<GitHub168> [jruby] headius pushed 5 new commits to master: https://git.io/vgvrK
<GitHub168> jruby/master d0a40ba Charles Oliver Nutter: Improve constant caching and factor logic out....
<GitHub168> jruby/master 628d386 Charles Oliver Nutter: Bind attribute accessors as direct instance variable accesses....
<GitHub168> jruby/master b4b7cd4 Charles Oliver Nutter: Bind Class#new directly to initialize....
headius2 has joined #jruby
<headius2> chrisseaton: how do I tell truffle not to print out "[truffle] opt fail" messages?
<headius2> I'm retesting PragTob's rubykon go thing with my recent indy improvements and wanted to compare with JRuby+Truffle but it coughs up a ton of "opt fail" warnings
<headius2> maybe I'm not running this right...it does terribly
<headius2> or perhaps you know, eregon
skade has quit [Quit: Computer has gone to sleep.]
joast has quit [Ping timeout: 246 seconds]
<headius2> ahh, arriving at airport, bbl
headius2 has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client]
robbyoconnor has quit [Ping timeout: 250 seconds]
<projectodd-ci> Project jruby-master-spec-ji build #2622: STILL FAILING in 33 sec: https://projectodd.ci.cloudbees.com/job/jruby-master-spec-ji/2622/
joast has joined #jruby
<projectodd-ci> Project jruby-master-test-slow_suites build #2574: STILL FAILING in 1 min 35 sec: https://projectodd.ci.cloudbees.com/job/jruby-master-test-slow_suites/2574/
drbobbeaty has joined #jruby
robbyoconnor has joined #jruby
robbyoconnor has quit [Client Quit]
<travis-ci> jruby/jruby (master:5f715f0 by Charles Oliver Nutter): The build was fixed. (https://travis-ci.org/jruby/jruby/builds/106187057)
<chrisseaton> headius / headius2: if it's printing 'opt fail' then there's something that needs to be fixed before it can be benchmarked (or you can benchmark it but it will be bad)
<chrisseaton> last time I patched benchmark/ips because it defeats optimistic optimisations right as it starts measuring
<chrisseaton> It runs with a constant number of iterations for warmup, then works out a new number of iterations to use, and starts measuring
<chrisseaton> So the first thing the compiled code does is 'ah I haven't seen this number of iterations before - it was always constant before so I compiled with that assumption - never deoptimise to sort this out'
vtunka has quit [Quit: Leaving]
<chrisseaton> never = better
pawnbox has quit [Remote host closed the connection]
vtunka has joined #jruby
<GitHub191> [jruby] chrisseaton pushed 1 new commit to master: https://git.io/vgvHp
<GitHub191> jruby/master 98070f3 Chris Seaton: [Truffle] More runtime.core to just core.
pawnbox has joined #jruby
joast has quit [Ping timeout: 246 seconds]
pawnbox has quit [Remote host closed the connection]
<projectodd-ci> Project jruby-master-spec-ji build #2623: STILL FAILING in 31 sec: https://projectodd.ci.cloudbees.com/job/jruby-master-spec-ji/2623/
lumrandir has joined #jruby
pawnbox has joined #jruby
<projectodd-ci> Project jruby-master-test-slow_suites build #2575: STILL FAILING in 1 min 39 sec: https://projectodd.ci.cloudbees.com/job/jruby-master-test-slow_suites/2575/
rsim1 has quit [Quit: Leaving.]
nirvdrum has joined #jruby
tomjoro has quit [Remote host closed the connection]
vtunka has quit [Quit: Leaving]
<yopp> hum
<yopp> is it possible to get cheap stats about how many objects are allocated right now?
<yopp> ideally by types, but even rough number will be fine
lumrandir has left #jruby [#jruby]
<yopp> chrisseaton, I bet you know :)
bbrowning has joined #jruby
vtunka has joined #jruby
<GitHub37> [jruby] eregon commented on commit ecf4c37: What's the purpose?... https://git.io/vgvhM
<GitHub40> [jruby] eregon commented on commit ecf4c37: (The less nodes using the context the better IMHO, and this is just going against it) https://git.io/vgvh9
<travis-ci> jruby/jruby (master:98070f3 by Chris Seaton): The build was broken. (https://travis-ci.org/jruby/jruby/builds/106201944)
vtunka has quit [Quit: Leaving]
<GitHub173> [jruby] eregon commented on commit a447e47: This should be @ExplodeLoop https://git.io/vgfvj
vtunka has joined #jruby
joast has joined #jruby
waka has joined #jruby
joast has quit [Ping timeout: 246 seconds]
joast has joined #jruby
joast is now known as 92AACA3RA
jensnockert has quit [Remote host closed the connection]
jensnockert has joined #jruby
vtunka has quit [Quit: Leaving]
vtunka has joined #jruby
enebo has joined #jruby
brightball has joined #jruby
brightball has quit [Quit: Leaving...]
skade has joined #jruby
jensnockert has quit [Remote host closed the connection]
skade has quit [Quit: Computer has gone to sleep.]
jensnockert has joined #jruby
skade has joined #jruby
jensnockert has quit [Remote host closed the connection]
ITXpander has quit [Quit: Leaving.]
jensnockert has joined #jruby
jensnockert has quit [Ping timeout: 264 seconds]
skade has quit [Quit: Computer has gone to sleep.]
<GitHub56> [jruby] chrisseaton commented on commit ecf4c37: It just seemed to me that it was something logical to tidy up - removing static state where possible. I think if we do some complex multi-tennant stuff in the future then we may want to set some options per-context.... https://git.io/vgfCB
rsim has joined #jruby
camlow325 has joined #jruby
<GitHub111> [jruby] chrisseaton commented on commit ecf4c37: Also, I think it's a limitation of JRuby when used with Nailgun/Drip that you can't change options. We'd like to be able to do that if we use a similar approach. https://git.io/vgflM
tomjoro has joined #jruby
<GitHub152> [jruby] chrisseaton pushed 3 new commits to master: https://git.io/vgf8E
<GitHub152> jruby/master 81fa791 Chris Seaton: [Truffle] Move POSIX stuff into a package.
<GitHub152> jruby/master 3f30d52 Chris Seaton: [Truffle] Move more native stuff.
<GitHub152> jruby/master 1fcd102 Chris Seaton: [Truffle] Move ext to the top level.
<projectodd-ci> Project jruby-master-spec-ji build #2624: STILL FAILING in 1 min 48 sec: https://projectodd.ci.cloudbees.com/job/jruby-master-spec-ji/2624/
vtunka has quit [Quit: Leaving]
vtunka has joined #jruby
<GitHub72> [jruby] chrisseaton pushed 1 new commit to truffle-head: https://git.io/vgfRa
<GitHub72> jruby/truffle-head 4578c7b Chris Seaton: Merge branch 'master' into truffle-head
<projectodd-ci> Project jruby-master-spec-compiler build #929: FAILURE in 30 sec: https://projectodd.ci.cloudbees.com/job/jruby-master-spec-compiler/929/
skade has joined #jruby
<projectodd-ci> Project jruby-master-test-slow_suites build #2576: STILL FAILING in 1 min 9 sec: https://projectodd.ci.cloudbees.com/job/jruby-master-test-slow_suites/2576/
thedarkone2 has joined #jruby
skade has quit [Read error: Connection reset by peer]
skade has joined #jruby
<GitHub45> [jruby] chrisseaton pushed 2 new commits to truffle-head: https://git.io/vgfz5
<GitHub45> jruby/truffle-head 71d774d Chris Seaton: [Truffle] Update Truffle version.
<GitHub45> jruby/truffle-head 6d2be03 Chris Seaton: Merge branch 'truffle-execjs-list-languages' into truffle-head
<GitHub166> [jruby] chrisseaton deleted truffle-execjs-list-languages at c465d06: https://git.io/vgfge
jensnockert has joined #jruby
jensnockert has quit [Ping timeout: 264 seconds]
vtunka has quit [Quit: Leaving]
slash_quit is now known as slash_join
hobodave has joined #jruby
jensnockert has joined #jruby
jensnockert has quit [Ping timeout: 264 seconds]
skade has quit [Ping timeout: 240 seconds]
<travis-ci> jruby/jruby (master:1fcd102 by Chris Seaton): The build was fixed. (https://travis-ci.org/jruby/jruby/builds/106243482)
<chrisseaton> yopp: I would try using JVisualVM, or Flight Recorder
<yopp> chrisseaton, and in the runtime?
<yopp> inside jruby?
<chrisseaton> not sure what you're asking?
<yopp> I need a simple metric for our app, to get an idea how many objects are left after each request
<chrisseaton> JVisualVM will do that
<yopp> I don't think that running jVVM on the production server is a great idea :|
<chrisseaton> well any instrumentation is likely to damage production performance
<chrisseaton> but you could attach just for ten minutes or whatever and then detach again
enebo has quit [Quit: enebo]
hobodave_ has joined #jruby
hobodave has quit [Ping timeout: 252 seconds]
<travis-ci> jruby/jruby (truffle-head:4578c7b by Chris Seaton): The build was fixed. (https://travis-ci.org/jruby/jruby/builds/106246922)
<travis-ci> jruby/jruby (truffle-head:6d2be03 by Chris Seaton): The build is still failing. (https://travis-ci.org/jruby/jruby/builds/106252411)
drbobbeaty has quit [Ping timeout: 250 seconds]
drbobbeaty has joined #jruby
shellac has quit [Quit: Ex-Chat]
mrmargolis has joined #jruby
drbobbeaty has quit [Ping timeout: 245 seconds]
<GitHub81> [jruby] nirvdrum commented on commit a447e47: Isn't it a concern that the String parts can be arbitrarily wrong? https://git.io/vgfdi
<travis-ci> jruby/jruby (truffle-head:ab12a8f by Chris Seaton): The build was fixed. (https://travis-ci.org/jruby/jruby/builds/106270671)
jensnockert has joined #jruby
bbrowning is now known as bbrowning_away
jensnockert has quit [Ping timeout: 264 seconds]
pietr0 has joined #jruby
cremes has joined #jruby
hobodave_ has quit [Quit: Computer has gone to sleep.]
tjohnson has joined #jruby
<chrisseaton> is headius on here while in flight?
<nirvdrum> Did he get a new one? He had tweeted that he missed his flight.
<chrisseaton> yeah he's tweeting like mad about optimisations
brightball has joined #jruby
camlow32_ has joined #jruby
brightball has quit [Client Quit]
camlow325 has quit [Ping timeout: 265 seconds]
brightball has joined #jruby
camlow32_ has quit [Remote host closed the connection]
<chrisseaton> or is enebo around instead?
brightball has quit [Quit: Leaving...]
<kares> obviously traveling does him some good optimizing ruby pieces :)
bbrowning_away is now known as bbrowning
headius2 has joined #jruby
headius2 has quit [Client Quit]
headius2 has joined #jruby
<headius2> greets
<headius2> chrisseaton: I figured the warnings were somewhat fatal but I did run with your shim loaded
<headius2> require 'shim.rb' at the top of full_playout.rb
<chrisseaton> I'm going to submit a PR to benchmark/ips soon
<headius2> When you tried this, did you get opt fail for stock ips?
subbu is now known as subbu|lunch
<chrisseaton> no, but it's not part of our integration tests - it should be - so it may have regressed
<chrisseaton> Is this a good benchmark for the exception backtrace removal thing? https://gist.github.com/chrisseaton/5e83d856aeba3b0b7f12
hobodave has joined #jruby
<chrisseaton> Need to implement it in Truffle as our backtraces are even more expensive than yours
jensnockert has joined #jruby
camlow325 has joined #jruby
jensnockert has quit [Ping timeout: 256 seconds]
camlow325 has quit [Remote host closed the connection]
<headius2> chrisseaton: sure, but you could just do foo rescue nil where foo is not defined
<headius2> you're including the overhead of attempting to parse the string as an integer in yours
<headius2> I guess mine includes overhead of searching for the method, but that would be less in JRuby proper
<chrisseaton> We'll cache that it doesn't exist so that'll be fine
<headius2> are you able to use the same strategy to eliminate backtraces?
<chrisseaton> I hope so
<headius2> yeah, we do in the class but not at the site
<headius2> so it's still a hash hit to repeatedly call an undefined method
<chrisseaton> I'll try a few difference ones
<chrisseaton> 37s for JRuby+Truffle vs 0.9s for JRuby
<chrisseaton> Yeah, we need this optimisation as well :)
<headius2> funny thing, our rewriting Java's backtrace and incorporating interpreter frames was a tiny fraction of the overhead
<headius2> I do it lazily anyway, but Java's stacktrace generate was still orders of magnitude slower than our rewriting the whole thing
<headius2> it doesn't seem right
<headius2> jeez, this is baffling
<chrisseaton> the problem is we deoptimise
<headius2> our Enumerable impls must be terrible
<headius2> you deopt to generate the stack trace?
<chrisseaton> yep
<headius2> holy ouch
<headius2> is that going to change?
<chrisseaton> I think we don't need to, and the API has a mechanism to not do it, but I don't think it's wired up at the moment
<headius2> I mean, you're going to be able to get rid of some with this trick but definitely not all
<headius2> not to mention Errno that are actually expected results...JRuby shims some of them to not have a trace
<headius2> EAGAIN for example
<headius2> we try to submit PRs to projects to remove exceptions as flow control but they keep popping up
92AACA3RA has quit [*.net *.split]
pawnbox has quit [Remote host closed the connection]
blaxter has quit [Ping timeout: 240 seconds]
skade has joined #jruby
pawnbox has joined #jruby
92AACA3RA has joined #jruby
subbu|lunch is now known as subbu
camlow325 has joined #jruby
camlow325 has quit [Remote host closed the connection]
camlow325 has joined #jruby
pawnbox has quit [Ping timeout: 246 seconds]
camlow325 has quit [Read error: Connection reset by peer]
camlow325 has joined #jruby
camlow325 has quit [Read error: Connection reset by peer]
camlow325 has joined #jruby
<headius2> well this is weird...if I put this pure-Ruby inject in jruby's Ruby kernel, it's slower than if I monkey-patch it after boot
<headius2> there's other defs in there so it's not some weird invalidation thing either
camlow325 has quit [Read error: Connection reset by peer]
camlow325 has joined #jruby
<headius2> like 10-15% slower...how strange
<headius2> subbu: do you remember if we treat the ruby kernel code differently in IR?
<headius2> like, do we not optimize it the same or something? I do see it JIT so I'm confused about this perf difference
<subbu> no, shouldn't be different.
<headius2> very strange
<subbu> if you are seeing it jit, then it should have had the same passes run on it.
<headius2> yeah
enebo has joined #jruby
<headius2> maybe it's just a fluke...this benchmark is pretty allocation-heavy, and that bumps up perf variability a lot
<headius2> chrisseaton: who's the best person to contact about crashes in Graal?
<headius2> bug tracker somewhere?
<chrisseaton> email Doug Simon probably
<headius2> most of the benchmarks I try to run crash reliably
<chrisseaton> I don't think there is any kind of public tracker at the moment I'm afraid
<headius2> ok
<headius2> someone should flip that switch on github
<chrisseaton> They will be very interested in normal Java code (not Truffle etc) that crashes
<headius2> actually I'll email the list
<chrisseaton> headius2: Truffle, R, Sulong etc are all on GH - Graal will move next
brightball has joined #jruby
yfeldblum has joined #jruby
brightball has quit [Client Quit]
<headius2> chrisseaton: that's excellent
rsim has quit [Quit: Leaving.]
<headius2> did you see the post to MLVM list about a C compiler to Java?
<headius2> now I have to ways to run C on JVM...we're going to need to investigate C exts again
<headius2> two
<chrisseaton> the Renjin one?
drbobbeaty has joined #jruby
robbyoconnor has joined #jruby
robbyoconnor has quit [Client Quit]
<GitHub60> [jruby] atambo closed issue #3621: SystemCallError: Unknown error - Unknown Error (0) - .gitignore https://git.io/vz9Vs
skade has quit [Quit: Computer has gone to sleep.]
skade has joined #jruby
<headius2> chrisseaton: yeah
skade has quit [Client Quit]
headius2 has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client]
headius2 has joined #jruby
hobodave has quit [Quit: Computer has gone to sleep.]
skade has joined #jruby
skade has quit [Client Quit]
hobodave has joined #jruby
jensnockert has joined #jruby
cremes has quit [Remote host closed the connection]
jensnockert has quit [Ping timeout: 245 seconds]
cremes has joined #jruby
brightball has joined #jruby
skade has joined #jruby
headius2 has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client]
headius2 has joined #jruby
kwando_ has quit [Remote host closed the connection]
brightball has quit [Read error: Connection timed out]
kwando has joined #jruby
hobodave has quit [Quit: Computer has gone to sleep.]
hobodave has joined #jruby
skade has quit [Quit: Computer has gone to sleep.]
bbrowning is now known as bbrowning_away
yfeldblum has quit [Ping timeout: 276 seconds]
headius2 has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client]
pawnbox has joined #jruby
pawnbox has quit [Ping timeout: 264 seconds]
yfeldblum has joined #jruby
<GitHub199> [jruby] nirvdrum pushed 1 new commit to truffle-pack-length-and-range: https://git.io/vgJjz
<GitHub199> jruby/truffle-pack-length-and-range e7c8caa Kevin Menard: [Truffle] Calculate code range as it's created in pack.
<GitHub142> [jruby] nirvdrum force-pushed truffle-pack-length-and-range from e7c8caa to fc38897: https://git.io/vgJjy
<GitHub142> jruby/truffle-pack-length-and-range 01ec0f5 Chris Seaton: [Truffle] Test length on pack as well bytes.
<GitHub142> jruby/truffle-pack-length-and-range 52d6684 Chris Seaton: [Truffle] Calculate string length as it's created in pack.
<GitHub142> jruby/truffle-pack-length-and-range fc38897 Kevin Menard: [Truffle] Calculate code range as it's created in pack.
<GitHub102> [jruby] nirvdrum force-pushed truffle-pack-length-and-range from fc38897 to 4211ade: https://git.io/vgJjy
<GitHub102> jruby/truffle-pack-length-and-range 4211ade Kevin Menard: [Truffle] Calculate code range as it's created in pack.
mrmargolis has quit [Remote host closed the connection]
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
pawnbox has joined #jruby
enebo has quit [Quit: enebo]
headius2 has joined #jruby
jensnockert has joined #jruby
hobodave has quit [Ping timeout: 250 seconds]
pawnbox has quit [Ping timeout: 250 seconds]
jensnockert has quit [Ping timeout: 264 seconds]
headius2 has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client]
gaustin has joined #jruby
gaustin has quit [Client Quit]
headius2 has joined #jruby
pawnbox has joined #jruby
pawnbox has quit [Ping timeout: 248 seconds]