Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2016-04-01

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

All times shown according to UTC.

Time Nick Message
03:00 japhb joined #moarvm
03:02 lizmat_ joined #moarvm
03:26 japhb joined #moarvm
03:32 geekosaur joined #moarvm
03:56 geekosaur joined #moarvm
06:28 leont joined #moarvm
06:44 geekosaur joined #moarvm
07:34 leont joined #moarvm
07:48 zakharyas joined #moarvm
08:27 lizmat joined #moarvm
11:05 domidumont joined #moarvm
11:50 lizmat joined #moarvm
11:52 leont joined #moarvm
11:55 dalek MoarVM: 45a071e | jnthn++ | src/strings/decode_stream.c:
11:55 dalek MoarVM: Fix typo in comment.
11:55 dalek MoarVM:
11:55 dalek MoarVM: The keys are like, right next to each other.
11:55 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/45a071ed85
12:39 dalek MoarVM: 5f76d03 | jnthn++ | src/strings/ (10 files):
12:39 dalek MoarVM: Make stream decodes return if stopper was reached.
12:39 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/5f76d030d7
12:39 dalek MoarVM: 7d59b00 | jnthn++ | src/strings/decode_stream.c:
12:39 dalek MoarVM: Don't do hopeless separator searches.
12:39 dalek MoarVM:
12:39 dalek MoarVM: If the decoder doesn't find the stopper, there's no point in going on
12:39 dalek MoarVM: a separator hunt with the newly decoded chars. Removes a bunch of the
12:39 dalek MoarVM: cost associated with reading really long lines.
12:39 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/7d59b005a4
13:08 zakharyas joined #moarvm
13:21 lizmat joined #moarvm
13:30 leont joined #moarvm
13:50 domidumont joined #moarvm
13:53 domidumont joined #moarvm
14:04 lizmat joined #moarvm
14:17 cognominal joined #moarvm
15:04 dalek MoarVM: b9afd77 | jnthn++ | src/strings/decode_stream. (2 files):
15:04 dalek MoarVM: Another big speed-up for reading long lines.
15:04 dalek MoarVM:
15:04 dalek MoarVM: Just skip over buffers that can't possibly contain the separator, so
15:04 dalek MoarVM: we save on expensively scanning them. With this and the previous
15:04 dalek MoarVM: commit, the heap profiler reaches its prompt within a second or so
15:04 dalek MoarVM: when reading in a 25MB heap profile.
15:04 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/b9afd772d5
15:05 jnthn OK, that makes moar-ha feel nicer :)
15:05 timotimo 25 megabytes? that's *nothing*! :P
15:06 moritz moar-ha? moar highly available? :-)
15:06 moritz (I guess "heap analyzer")
15:07 jnthn heap analyzer :)
15:07 jnthn timotimo: :P
15:45 domidumont joined #moarvm
16:39 geekosaur joined #moarvm
17:24 dalek MoarVM: 4afd7b6 | jnthn++ | src/strings/utf8_c8.c:
17:24 dalek MoarVM: Fix utf8-c8 decode crash caused by off-by-1.
17:24 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/4afd7b6242
17:27 timotimo huh! i thought i had tried that and it wasn't enough
17:27 timotimo though, what do i know :)
19:59 domidumont joined #moarvm
20:52 leont joined #moarvm
21:32 cognominal_ joined #moarvm
22:03 lizmat joined #moarvm
23:57 timotimo jnthn: do you have a good suggestion for how to deal with the safepoint thing?
23:58 timotimo i mean the leak that exists because we're doing "free at safepoint" over and over but never actually reach a safe point

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