Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2015-08-28

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

All times shown according to UTC.

Time Nick Message
01:48 ilbot3 joined #moarvm
01:48 Topic for #moarvm is now https://github.com/moarvm/moarvm | IRC logs at  http://irclog.perlgeek.de/moarvm/today
06:16 colomon_ joined #moarvm
07:28 zakharyas joined #moarvm
07:31 Ven joined #moarvm
07:34 FROGGS joined #moarvm
07:36 brrt joined #moarvm
07:37 dalek MoarVM/even-moar-jit: 4a3e2cf | brrt++ | src/jit/ (2 files):
07:37 dalek MoarVM/even-moar-jit: Copy is not a regular tile
07:37 dalek MoarVM/even-moar-jit:
07:37 dalek MoarVM/even-moar-jit: And even if it where, it should use a proper copy of the selected
07:37 dalek MoarVM/even-moar-jit: value.
07:37 dalek MoarVM/even-moar-jit: review: https://github.com/MoarVM/MoarVM/commit/4a3e2cf0ac
07:39 lizmat joined #moarvm
07:41 brrt that fixes one crash at least
07:42 brrt (we can now actually run JIT code, a bit)
07:42 JimmyZ brrt++
07:47 lizmat joined #moarvm
07:48 colomon joined #moarvm
07:51 lizmat joined #moarvm
08:34 Ven joined #moarvm
08:42 lizmat joined #moarvm
08:54 JimmyZ timotimo: do you mind if I do s/push/push_i/g of from_to in your branch?
08:55 colomon joined #moarvm
09:14 brrt joined #moarvm
09:47 timotimo please feel free :)
09:53 timotimo though ... where does the type of the list_i come from?
09:56 lizmat joined #moarvm
09:59 JimmyZ timotimo: looks like you branch is based on even_moar_jit branch
09:59 colomon joined #moarvm
10:13 lizmat joined #moarvm
10:14 timotimo yes
10:28 colomon joined #moarvm
10:29 zakharyas joined #moarvm
10:29 Peter_R joined #moarvm
10:29 JimmyZ joined #moarvm
10:29 psch joined #moarvm
10:30 dalek joined #moarvm
10:30 synbot6 joined #moarvm
10:30 camelia joined #moarvm
10:30 _longines joined #moarvm
10:32 synbot6 joined #moarvm
10:32 camelia joined #moarvm
10:32 _longines joined #moarvm
10:33 dalek joined #moarvm
10:44 JimmyZ timotimo: pushed
10:46 timotimo it looks like you moved some variable definitions away from the beginning of the block
10:46 timotimo er
10:46 timotimo no, i read the colors wrong
10:47 timotimo i wonder if we should keep the argument to the op that selects the type of list to initialize
10:49 JimmyZ not sure ,the type in nqp is repr NFA
10:50 JimmyZ well, from the code, i think is no
10:52 JimmyZ oh, it is yes
10:56 JimmyZ timotimo: after look at it again, I 'd say  no, we don't need it
10:57 JimmyZ it only for from_statelist
10:59 timotimo yeah
10:59 timotimo if you want to, you can patch that, too :)
11:21 lizmat joined #moarvm
11:22 dalek MoarVM/nfa_to_statelist: bd8bd41 | (Jimmy Zhuo)++ | / (6 files):
11:22 dalek MoarVM/nfa_to_statelist: remove a useless arg
11:22 dalek MoarVM/nfa_to_statelist: review: https://github.com/MoarVM/MoarVM/commit/bd8bd41b58
11:29 Ven joined #moarvm
11:31 FROGGS joined #moarvm
11:44 timotimo only the difficult part remains :|
11:44 timotimo figuring out where to put the "throw out statelist" and "recreate statelist" operations
11:49 lizmat joined #moarvm
11:58 timotimo src/jit/x64/tiles.dasc:128:57: warning: format specifies type 'int' but the argument has type 'MVMJitExprNode'
11:58 timotimo (aka 'long') [-Wformat]
11:58 timotimo MVM_oops(tc, "Unsupported load size: %d\n", args[0]);
11:58 timotimo should we change something here?
11:58 timotimo it potentially wants to be ->size?
11:59 timotimo hm, it seems like we expect those numbers to be 1, 2, 4 or 8
11:59 timotimo so it won't be a big problem
12:34 lizmat joined #moarvm
12:36 brrt joined #moarvm
12:36 Ven joined #moarvm
12:36 * brrt is going to do some union-busting next
12:37 timotimo o/
12:38 dalek MoarVM/nfa_to_statelist: ffad694 | timotimo++ | src/jit/core.expr:
12:38 dalek MoarVM/nfa_to_statelist: implement if_i
12:38 dalek MoarVM/nfa_to_statelist: review: https://github.com/MoarVM/MoarVM/commit/ffad6943f6
12:38 dalek MoarVM/nfa_to_statelist: 3d2f0f3 | timotimo++ | src/jit/x64/tile (3 files):
12:38 dalek MoarVM/nfa_to_statelist: have tiles for (n)equal (g/l)(than/equal)
12:38 dalek MoarVM/nfa_to_statelist: review: https://github.com/MoarVM/MoarVM/commit/3d2f0f3df0
12:38 timotimo ^- hoping this isn't wrong :)
12:44 Ven joined #moarvm
12:52 timotimo brrt: do you think it'll be a good idea to throw out all the c function calls from graph.c if they've been translated to the expr list?
12:52 brrt not yet
12:52 brrt eventually, yes
12:53 timotimo will it reduce moarvm's binary size?
12:54 brrt i am one dumb son of a .... feminine canine
12:55 brrt yes, removing it will reduce the memory size. i'm not sure if it will be a net benefit, since we'll offset it with the memory for templates
12:55 brrt but we're talking kilobytes, at best
12:56 brrt or worst
12:57 brrt i do hope it'll be a bit more hackable, less limited
12:58 timotimo OK
12:59 timotimo but this is code size, so we measure it against the instruction cache, which is small
12:59 brrt that is true
12:59 timotimo and templates are data which a small-ish bit of code works on
12:59 brrt yes
12:59 brrt i do hope that offsets the added complexity
12:59 timotimo i hope so, too
12:59 brrt actually, algorithmically the complexity is quite limited
12:59 brrt i've intentionally used only very simple data structures :-)
13:03 timotimo that does sound nice, yeah
13:11 brrt off, by, one
13:14 brrt i'm a dumbass :-)
13:19 dalek MoarVM/even-moar-jit: 7708b8d | brrt++ | src/jit/ (3 files):
13:19 dalek MoarVM/even-moar-jit: Malloc GPR buffer
13:19 dalek MoarVM/even-moar-jit:
13:19 dalek MoarVM/even-moar-jit: This caused an off-by-one because in POSIX ABI there are 9,
13:19 dalek MoarVM/even-moar-jit: not 8, volatile registers we use.
13:19 dalek MoarVM/even-moar-jit: review: https://github.com/MoarVM/MoarVM/commit/7708b8d6f1
13:28 timotimo do we still crash when compiling and running some of the generated code?
13:28 timotimo oh
13:29 timotimo i pushed the tiles and stuff to the wrong branch
13:32 dalek MoarVM/even-moar-jit: c7047ea | timotimo++ | src/jit/core.expr:
13:32 dalek MoarVM/even-moar-jit: implement if_i
13:32 dalek MoarVM/even-moar-jit: review: https://github.com/MoarVM/MoarVM/commit/c7047ea59f
13:32 dalek MoarVM/even-moar-jit: 966b091 | timotimo++ | src/jit/x64/tile (3 files):
13:32 dalek MoarVM/even-moar-jit: have tiles for (n)equal (g/l)(than/equal)
13:32 dalek MoarVM/even-moar-jit: review: https://github.com/MoarVM/MoarVM/commit/966b0914da
13:37 lizmat joined #moarvm
13:42 timotimo "Could not take register even after spilling"
13:44 Ven joined #moarvm
13:46 brrt working on it
13:46 brrt :-)
13:46 timotimo cool!
13:47 brrt yeah, that's not the same as 'it has been fixed', though :-)
13:49 timotimo right
13:59 dalek MoarVM/nfa_to_statelist: 134a35b | (Jimmy Zhuo)++ | src/6model/reprs/NFA.c:
13:59 dalek MoarVM/nfa_to_statelist: only pop 2
13:59 dalek MoarVM/nfa_to_statelist: review: https://github.com/MoarVM/MoarVM/commit/134a35ba09
14:07 timotimo good catch
14:11 timotimo JimmyZ: mind if i rebase nfa_to_statelist onto even-moar-jit or perhaps master?
14:13 JimmyZ timotimo: go ahead :)
14:17 dalek MoarVM/nfa_to_statelist: c9280de | timotimo++ | / (8 files):
14:17 dalek MoarVM/nfa_to_statelist: implement nfatostatelist
14:17 dalek MoarVM/nfa_to_statelist:
14:17 dalek MoarVM/nfa_to_statelist: isn't correct yet; something about temp roots is wrong
14:17 dalek MoarVM/nfa_to_statelist: review: https://github.com/MoarVM/MoarVM/commit/c9280def5d
14:17 dalek MoarVM/nfa_to_statelist: 36c420e | (Jimmy Zhuo)++ | src/6model/reprs/NFA.c:
14:17 dalek MoarVM/nfa_to_statelist: allocate intarray instead of array
14:17 dalek MoarVM/nfa_to_statelist: review: https://github.com/MoarVM/MoarVM/commit/36c420e874
14:17 dalek MoarVM/nfa_to_statelist: 4f585f7 | (Jimmy Zhuo)++ | / (6 files):
14:17 dalek MoarVM/nfa_to_statelist: remove a useless arg
14:17 dalek MoarVM/nfa_to_statelist: review: https://github.com/MoarVM/MoarVM/commit/4f585f771d
14:17 dalek MoarVM/nfa_to_statelist: 1d13ec4 | (Jimmy Zhuo)++ | src/6model/reprs/NFA.c:
14:17 timotimo i git push -f'd this to make things a bit simpler
14:19 lizmat joined #moarvm
14:22 brrt you can rebase if you want
14:23 timotimo too late, i already did :)
14:23 timotimo but only the nfa_to_statelist branch
14:23 timotimo which i don't expect you're working with at all, except when i asked you to help me with it :)
14:23 brrt yes, and by the time i was back, jnthn and JimmyZ had already fixed it :-)
14:24 timotimo right
14:24 JimmyZ what did I  fix...?
14:25 timotimo you fixed that i forgot to put the & in front of the things i pushed onto the temp roots
14:25 timotimo or was that jnthn? i think it was you
14:26 JimmyZ he said, and I pushed :P
14:27 timotimo tthat's good enough for me ;)
14:44 lizmat joined #moarvm
14:48 colomon joined #moarvm
15:14 Ven joined #moarvm
15:22 brrt damnit, somehow i'm trying to free an unassigned register
15:27 timotimo i'm trying to take a localref of an object register :|
15:31 brrt wohoooooow, no crashes so far, just an infinite loop :-)
15:32 brrt oh, i see
15:32 brrt hmm
15:33 brrt oh, damn, i see now
15:36 FROGGS m: for 1..3 { last; FIRST { say "hello $_" }; say "ohh $_" }
15:36 camelia rakudo-moar eb6de6: OUTPUT«hello 1␤»
15:36 FROGGS ww
15:59 Ven joined #moarvm
16:11 lizmat_ joined #moarvm
16:34 TimToady joined #moarvm
16:47 FROGGS joined #moarvm
17:20 lizmat joined #moarvm
17:31 FROGGS joined #moarvm
18:56 Ven joined #moarvm
20:16 FROGGS joined #moarvm
20:26 colomon joined #moarvm
21:37 dalek MoarVM: 09039bb | hoelzro++ | src/io/procops.c:
21:37 dalek MoarVM: Make sure we clean up the stdin handle when a process dies
21:37 dalek MoarVM:
21:37 dalek MoarVM: If I spawn a process with an open stdin handle, I don't explicitly
21:37 dalek MoarVM: close its stdin, and that process exits, that handle is never closed.
21:37 dalek MoarVM: This can result in a program exceeding its file descriptor limit
21:37 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/09039bb894
21:37 hoelzro http://i1.kym-cdn.com/photos/images/original/000/162/317/2vA1a.png
21:39 timotimo excellent, you found the mutex you need
21:41 japhb Woah, 09039bb might actually explain at least one of my crash modes
21:41 timotimo \o/
21:42 hoelzro yaaaay
21:42 hoelzro I'm helping!
21:45 japhb There was any doubt?   :-)
22:42 TEttinger joined #moarvm
22:42 rarara joined #moarvm
22:58 [Coke] brrt: Your name came up in the Q and A today.

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