ARM32 C1 and C2 jit's is comming to OpenJDK inside the AArch32 port

#openjdk #javaone 26 oct 2015

(14:28:25) xranby: Remember to talk & ask about the #OpenJDK #MobileProject & the #OpenJDK Port #AArch32 #ARM at #JavaOne #JavaOne2015 #Embedded #Mobile #Java!
(14:28:32) adinn: look under jdk9/jdk9/jdk or jdkN/jdkNu/jdk for N = 7/8
(14:30:36) xranby: both of these topics has not yet been granted infrastructure by the openjdk Governing Board
(14:31:56) xranby: Port: AArch32 http://mail.openjdk.java.net/pipermail/porters-dev/2015-October/000503.html
(14:32:14) xranby: Result: New Project: Mobile http://mail.openjdk.java.net/pipermail/announce/2015-October/000201.html
(14:32:32) xranby: however no infrastructure online at http://openjdk.java.net/
(14:34:58) adinn: xranby: the OpenJDK AArch32 Port project has only just been proposed so it's not surprising that it has no infrastructure as yet -- the GB has not actually convened since the proposal
(14:35:50) xranby: adinn: how about the Mobile Project? 
(14:36:03) adinn: xranby: I have no dog in that fight :-)
(14:36:31) adinn: xranby: but I don't think the GB has convened since that proposal was accepted either
(14:37:42) ***xranby is sitting on the sidelines watching..
(14:37:50) adinn: xranby: there's no reason to think anyone is dragging their feet here -- if that is what you were thinking?
(14:38:04) xranby: the Mobile Project claims to extend the work of Zero for use on iOS
(14:38:38) xranby: hence i am interested to see the modified zero code published if it is used for ios deployments
(14:38:39) adinn: xranby: yes I know -- which 3 letters explain why I am not i any way involved :-)
(14:39:02) adinn: ^i^in
(14:42:50) xranby: noone should honestly do iOS as long as there exist no free compiler to use for the os
(14:43:27) xranby: however.. people for some reason like to sign their libertys away using eula's
(14:44:37) adinn: xranby: I agree that having no freely available compiler does compromise the ability to maintain an open source release
(14:44:49) adinn: xranby: however, we have a Windows port which is based on use of a non-free compiler and it has proven very valuable to have
(14:45:23) adinn: xranby: without it many of our customers would not be able to maintain legacy products
(14:45:24) xranby: adinn: for microsoft using mingw usually does the trick
(14:46:11) adinn: xranby: well yes, perhaps, but current code is being developed using MS compiler as the one which has to work
(14:46:37) adinn: xranby: so maintenance on the assumption that we can get away with mingw is not exactly a trust-inspiring proposition
(14:47:45) xranby: for opengl bindings inside the jogl project mingw is enough, thats the only part i can put my word to
(14:48:06) xranby: we use it to compile all the windows 32 bit and 64 bit jni
(14:48:18) xranby: inside the jogamp project
(14:49:55) xranby: adinn: the aarch32 project looks awesome on paper.. repurposing the jit's from aarch64
(14:50:22) xranby: i have only seen the template interpreter in code
(14:50:28) adinn: xranby: I don't know how mush we will repurpose -- it's really more of a new port
(14:50:38) adinn: xranby: the architectures are not really the same
(14:50:56) xranby: oh. so it is not the same work as the demo done by linaro?
(14:51:12) adinn: xranby: yes, but Joseph Joyce merely implemented the template interpreter
(14:51:48) adinn: xranby: Ed's proposal is to debug the few minor nits in that code and then contribute a new C1 and C2
(14:52:25) xranby: everyone at javaone you have a massive scoop here 
(14:52:29) adinn: xranby: so it's not really got a lot in common with the AArch64 port
(14:52:59) xranby: adinn: thank you for this new gossip!
(14:53:02) adinn: xranby: hmm, well that's nice to know -- we weren't really thinking of it as a scoop :-)
(14:54:20) adinn: xranby: hmm, I'm not sure I added much to the story -- Ed's proposal already said this is what we were planning
(15:00:55) xranby: adinn: it is still a lett than 2 weeks old scoop!
(15:01:03) xranby: less than 2 weeks old
(15:01:51) xranby: the world will first know when the new project announcement is made
(15:02:20) xranby: that is when the ball is nicked
(15:02:24) xranby: and accepted
(15:13:06) xranby: adinn: thank you for the heads up that the GB has not yet been convened since the proposals
(15:13:23) xranby: this is the first time i have heard a plausible explenation
(15:14:06) adinn: xranby: well I'm not aware that it has -- Andrew Haley has not mentioned a meeting in his weekly status reports. he normally does that so I take it they have not met