Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2016-11-10

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

All times shown according to UTC.

Time Nick Message
00:01 pyrimidine joined #moarvm
00:21 pyrimidine joined #moarvm
01:10 pyrimidi_ joined #moarvm
02:18 pyrimidine joined #moarvm
02:38 pyrimidine joined #moarvm
02:48 ilbot3 joined #moarvm
02:48 Topic for #moarvm is now https://github.com/moarvm/moarvm | IRC logs at  http://irclog.perlgeek.de/moarvm/today
02:48 pyrimidine joined #moarvm
03:06 pyrimidine joined #moarvm
03:14 pyrimidine joined #moarvm
03:31 pyrimidine joined #moarvm
03:45 pyrimidine joined #moarvm
04:02 pyrimidine joined #moarvm
04:06 geekosaur joined #moarvm
05:44 pyrimidine joined #moarvm
05:54 pyrimidine joined #moarvm
06:04 pyrimidine joined #moarvm
06:14 pyrimidine joined #moarvm
06:24 pyrimidine joined #moarvm
06:25 brrt \o
06:30 brrt i am in fact no longer fully convinced that a second 'register assignment' loop in linear scan maear scan makes any sense at all
06:32 brrt because you need to be  able to find the ' current' live range that stands for a particular value, and that 'current' version lives  usuaully in the active set to begin with
06:34 pyrimidine joined #moarvm
06:39 brrt you *can* move it t o a later phase but it is kind of pointlyess
06:39 brrt pointless
06:40 pyrimidine joined #moarvm
06:50 pyrimidine joined #moarvm
06:59 domidumont joined #moarvm
07:04 brrt hmmm, okay, so let me think about that some more
07:05 brrt while running linear scan, I may have to update existing references to live ranges to other references because the live range themselves have been spilled
07:06 brrt if i wait with register assignment until after the first loop, I can simply read the registers per tile from the updates references
07:06 brrt however, since i already must do the update, updating the register number rather than the reference seems to me to be equivalent
07:09 brrt so what i need... is a register assignment primitive that updates the users of the live range to the new register
07:56 pyrimidine joined #moarvm
08:04 pyrimidine joined #moarvm
08:18 pyrimidine joined #moarvm
08:25 zakharyas joined #moarvm
08:25 pyrimidine joined #moarvm
08:34 pyrimidine joined #moarvm
08:54 pyrimidine joined #moarvm
09:09 pyrimidine joined #moarvm
09:34 pyrimidine joined #moarvm
09:41 pyrimidine joined #moarvm
09:50 pyrimidine joined #moarvm
09:53 Util joined #moarvm
10:02 vendethiel- joined #moarvm
10:05 pyrimidi_ joined #moarvm
10:14 pyrimidine joined #moarvm
10:19 brrt so, sucky news, but $employer has just suspended the hackathon time...
10:27 pyrimidi_ joined #moarvm
10:33 timotimo :o
10:50 pyrimidine joined #moarvm
11:00 brrt yes
11:01 brrt the good news bit is that I think I have a definitive answer to the second-loop problem; it is not necessary and it doesn't simplify anything
11:01 timotimo first-loop problems, man
11:02 brrt :-D
11:07 brrt note for future references: I'm going to store the index node-that-references the live range along in the use array of the live ranges, so that I have a complete mapping from live ranges to tile nodes, so that when I assign the actual register to the live range, I can find the tiles that have to be assigned the register to by simply looping over the 'uses' array
11:08 brrt which means that i no longer have to maintain a tile -> live range mapping because the mapping will be in the reverse order, which is the interesting order anyway
11:11 brrt and... if i do that for definitions as well as uses, this generalizes nicely because definitions are always in the '0' index, and I can still always 'reuse' the definitions/uses when splitting/spilling
11:11 brrt but first, lunch &
11:11 timotimo ah, nice
11:42 brrt` joined #moarvm
12:08 ggoebel joined #moarvm
12:35 pyrimidine joined #moarvm
12:44 pyrimidine joined #moarvm
12:54 pyrimidine joined #moarvm
13:03 pyrimidine joined #moarvm
13:13 pyrimidine joined #moarvm
13:15 japhb joined #moarvm
13:21 pyrimidine joined #moarvm
13:33 pyrimidine joined #moarvm
13:43 pyrimidine joined #moarvm
13:53 pyrimidine joined #moarvm
14:02 dalek joined #moarvm
14:45 pyrimidine joined #moarvm
14:52 pyrimidine joined #moarvm
15:03 nwc10 joined #moarvm
15:08 pyrimidine joined #moarvm
15:26 domidumont joined #moarvm
15:50 domidumont joined #moarvm
15:57 timotimo should i continue putting "but got REPR debug_name" stuff into places like the interpreter?
15:57 timotimo maybe smaller code would be better
15:57 timotimo so maybe there should be a function that gets called to build the error message from a few parts
15:57 timotimo i'm not convinced that's actually smaller
16:01 timotimo OTOH, we jit a bunch of things, so the size of the interpreter itself isn't as noticable for hot things
16:34 brrt i don't really know the answer to that, tbh
16:40 FROGGS joined #moarvm
17:00 pyrimidine joined #moarvm
17:06 pyrimidine joined #moarvm
17:58 japhb joined #moarvm
18:27 japhb joined #moarvm
19:43 domidumont joined #moarvm
19:57 pyrimidine joined #moarvm
20:26 pyrimidi_ joined #moarvm
20:35 zakharyas joined #moarvm
20:37 domidumont joined #moarvm
20:53 pyrimidine joined #moarvm
21:03 pyrimidine joined #moarvm
21:13 pyrimidine joined #moarvm
21:34 pyrimidine joined #moarvm
21:38 pyrimidi_ joined #moarvm
21:42 vendethiel joined #moarvm
21:49 pyrimidine joined #moarvm
22:02 pyrimidine joined #moarvm
22:29 pyrimidine joined #moarvm
23:08 pyrimidine joined #moarvm

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