Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2016-05-21

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

All times shown according to UTC.

Time Nick Message
01:56 tokuhirom joined #moarvm
03:51 tokuhirom joined #moarvm
05:24 tokuhirom joined #moarvm
05:53 dalek joined #moarvm
07:02 domidumont joined #moarvm
07:06 domidumont joined #moarvm
09:01 cognominal joined #moarvm
09:25 tokuhirom joined #moarvm
09:55 cognominal joined #moarvm
10:54 pyrimidine joined #moarvm
12:30 Ven joined #moarvm
12:45 domidumont joined #moarvm
14:17 Ven joined #moarvm
15:24 tokuhirom joined #moarvm
15:55 _longines joined #moarvm
17:05 domidumont joined #moarvm
18:06 TimToady joined #moarvm
19:25 tokuhirom joined #moarvm
20:00 dalek MoarVM/even-moar-jit: 40add84 | brrt++ | src/jit/ (4 files):
20:00 dalek MoarVM/even-moar-jit: Take register allocation out of code generation
20:00 dalek MoarVM/even-moar-jit:
20:00 dalek MoarVM/even-moar-jit: Trying to do online (i.e. without lookahead) register allocation
20:00 dalek MoarVM/even-moar-jit: was a poor idea (complex, ineffiicent, incorrect results). So it
20:00 dalek MoarVM/even-moar-jit: is moved to a separate step, and will be developed more further.
20:00 dalek MoarVM/even-moar-jit:
20:00 dalek MoarVM/even-moar-jit: As a result, allocating 'scratch' registers in tiles must be
20:00 dalek MoarVM/even-moar-jit: removed as well.
20:00 dalek MoarVM/even-moar-jit: review: https://github.com/MoarVM/MoarVM/commit/40add84363
20:00 dalek MoarVM/even-moar-jit: c9bf2fc | brrt++ | src/jit/ (2 files):
20:00 dalek MoarVM/even-moar-jit: Remove register locking logic
20:00 dalek MoarVM/even-moar-jit:
20:00 dalek MoarVM/even-moar-jit: Because we can't allocate scratch registers anyway, this is no
20:00 brrt joined #moarvm
20:07 brrt so, that is at least a bit of progress; my hypothesis is that by slowly moving bits into their right places, i can start working on a real, more complex and correct, allocator
20:11 brrt i.e. kill all the dumb decisions and simplify things first, then build it out
20:34 timotimo neat
20:54 jnthn brrt++
21:02 TimToady we prefer convergence on a correct solution even it gives the appearance of divergence occasionally :)
21:24 timotimo i just finished the paper on mementos, btw
21:27 tokuhirom joined #moarvm
21:28 timotimo it doesn't give implementation advice, though. like for how the memento map should look, or the allocation site data structure, really
21:28 timotimo they also have static frames be movable by the GC, which forced v8 to get a "zombie" state in which static frames survive an additional collection to keep heap integrity
21:30 TimToady joined #moarvm
21:45 jnthn Yes, most papers leave the reader with some amount of figuring out to do in terms of the details :-)
21:54 timotimo aye
23:23 tokuhirom joined #moarvm

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