Perl 6 - the future is here, just unevenly distributed

IRC log for #perl11, 2016-07-04

| Channels | #perl11 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:27 vytas joined #perl11
02:27 travis-ci RPerl build passed. Will Braswell says 'Learning RPerl, Exercises 1.1 & 1.3, Update Print Operators; Tests, Remove Deprecated Print Code'
02:27 travis-ci https://travis-ci.org/wbraswell/rperl/builds/142103717 https://github.com/wbraswell/rperl/compare/43c93e095388...e106e3d366c8
02:32 willthechill yay!
02:56 travis-ci RPerl build passed. Will Braswell says 'Tests, Update PMC Include Paths'
02:56 travis-ci https://travis-ci.org/wbraswell/rperl/builds/142106466 https://github.com/wbraswell/rperl/compare/e106e3d366c8...51abbec4e0dc
03:18 kentnl joined #perl11
04:33 travis-ci RPerl build passed. Will Braswell says 'Tests, Update PMC Include Paths, Part 2'
04:33 travis-ci https://travis-ci.org/wbraswell/rperl/builds/142114470 https://github.com/wbraswell/rperl/compare/51abbec4e0dc...6efc440912b5
04:36 willthechill yay
05:07 travis-ci MathPerl build passed. Will Braswell says 'Update PMC Include Paths & Subroutine Shims'
05:07 travis-ci https://travis-ci.org/wbraswell/mathperl/builds/142118048 https://github.com/wbraswell/mathperl/compare/2e5c5b97ed6e...d9c6dcf82763
05:07 travis-ci RPerl build passed. Will Braswell says 'Tests, Update PMC Include Paths, Part 3'
05:07 travis-ci https://travis-ci.org/wbraswell/rperl/builds/142117457 https://github.com/wbraswell/rperl/compare/6efc440912b5...1e0d1197f749
05:14 travis-ci PhysicsPerl build failed. Will Braswell says 'Update PMC Include Paths & Subroutine Shims'
05:14 travis-ci https://travis-ci.org/wbraswell/physicsperl/builds/142118116 https://github.com/wbraswell/physicsperl/compare/6d68702bd17b...c2711981de2b
05:20 willthechill yay, yay, darn!  :P
05:52 travis-ci PhysicsPerl build canceled. Will Braswell says 'Tests, Fix Missing Symbolic Link'
05:52 travis-ci https://travis-ci.org/wbraswell/physicsperl/builds/142123219 https://github.com/wbraswell/physicsperl/compare/c2711981de2b...625a0fb0ebe1
05:52 travis-ci MathPerl build passed. Will Braswell says 'Tests, Temporarily Revert Absolute Path Names For RPerl v1.75 Compatibility'
05:52 travis-ci https://travis-ci.org/wbraswell/mathperl/builds/142124051 https://github.com/wbraswell/mathperl/compare/fd2009b2eeb6...7796d78a7955
05:57 willthechill yay
05:58 travis-ci PhysicsPerl build passed. Will Braswell says 'Tests, Temporarily Revert Absolute Path Names For RPerl v1.75 Compatibility'
05:58 travis-ci https://travis-ci.org/wbraswell/physicsperl/builds/142124096 https://github.com/wbraswell/physicsperl/compare/625a0fb0ebe1...9769dcd939e4
05:58 willthechill double yay!  :D
05:58 willthechill all 3 passing now.
07:54 travis-ci perl11/cperl#1252 (smoke/gh173-sig2pp - c9679ef : Reini Urban): The build is still failing. https://travis-ci.org/perl11/cperl/builds/142135317
08:28 travis-ci perl11/cperl#1253 (master - 2fbf3da : Reini Urban): The build passed. https://travis-ci.org/perl11/cperl/builds/142135428
10:52 vytas joined #perl11
11:04 dalek joined #perl11
16:53 rurban_ joined #perl11
18:14 willthechill joined #perl11
18:25 travis-ci perl11/cperl#1254 (smoke/gh173-sig2pp - 35d7769 : Reini Urban): The build is still failing. https://travis-ci.org/perl11/cperl/builds/142269582
20:16 rurban joined #perl11
20:17 rurban Can now almost unexec perl programs via -u -> script.perldump. executable binaries: https://github.com/perl11/cperl/commits/feature/gh176-unexec much easier than with the old compiler.
20:18 rurban Just the initialization of perl.c needs to be tuned a bit. Bitrot since perl4
20:21 rurban https://github.com/perl11/cperl/issues/176
20:21 willthechill what does unexec mean?
20:21 rurban dump a script as executable
20:21 willthechill a.out format?
20:21 willthechill elf?
20:22 willthechill does it generate .c files and then call gcc?
20:22 rurban basically dumping the state of memory to an executable format, with some dynamic inits
20:22 rurban nope. elf, coff, darwin, cygwin, win32/64, hpux, aix, sunos/solaris, dos
20:22 willthechill okay so it reads binary from memory dumping to disk
20:22 willthechill it does not pass through any phase where it is C source code?
20:22 rurban cperl -u script.pl -> script.pl.perldump; ./script.pl.perldump is the binary
20:23 rurban no, directly to a binary. same as temacs => emacs
20:24 rurban it rearranges the binary layout a bit. the code is from emacs. it worked in perl4. I fixed it now for cperl
20:24 rurban only for darwin so far, but the others should be trivial
20:25 rurban but I fear B::C has superior memory advantages. lots of static and cow strings. in my new binaries everything is dynamic as with pure perl.
20:25 rurban no source code. a few milliseconds for everything
20:26 rurban parrot had such a backend also before they destroyed it
20:26 rurban basically a "native compiler backend"
20:27 willthechill well it is good to have a formal way to do the "Infamous Memory Dump Trick"!  :-)
20:28 willthechill rurban++
20:28 rurban I think so also
20:28 willthechill can you go into .o format instead of .exe?
20:28 willthechill so that I can combine your .o with my .o from rperl?
20:29 willthechill I can generate .o, .a, .so, and .exe
20:29 rurban nope. only directly to executables
20:29 willthechill okay I guess we will have to use B::CC to generate the .o files then
20:30 rurban I have to work a bit on dynaloaded modules also, I think. And the initialization order. And maybe with a shared libperl, esp. with ASLR.
20:30 rurban glibc will remove support for it soon, they said.
20:32 willthechill aslr?
20:38 rurban_ joined #perl11
20:56 rurban address space randomization of all dynaloaded symbols and funcs
20:56 rurban so all external pointers need to be remapped
20:56 rurban or just mmap'ed with MAP_FIXED
20:56 rurban or re-initialized
20:57 willthechill sounds good, will talk more about it soon, family time now for holiday celebration!  :-D
20:57 rurban have a good 4th july!
20:59 willthechill thanks, you too!
20:59 willthechill I will be back online in a few hours
20:59 willthechill more surprises to come today
20:59 willthechill ;-)
20:59 willthechill &
21:21 bulk88 rurban i ahve 4 GB installed, 3 GB available, the appveyor is timining out on e_perlcc.t
21:22 bulk88 rurban I didnt see the $(XCOPY) $(PERLDLL) ..\cpan\B-C\$(NULL) in the namke makefile, the change wasnt put int he other 2 ones obviously
23:39 rurban joined #perl11

| Channels | #perl11 index | Today | | Search | Google Search | Plain-Text | summary