Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2015-01-29

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

All times shown according to UTC.

Time Nick Message
00:34 japhb .oO( Converting unknown unknowns to known unknowns )
01:22 japhb joined #moarvm
01:30 lue 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
02:56 vendethiel joined #moarvm
03:39 vendethiel joined #moarvm
04:19 vendethiel joined #moarvm
04:44 vendethiel joined #moarvm
05:19 vendethiel joined #moarvm
06:33 dalek MoarVM: d235e77 | timotimo++ | src/core/nativecall.c:
06:33 dalek MoarVM: nativecall shall tell you what REPR you got in errors
06:33 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/d235e77c1e
06:35 timotimo not entirely sure how useful this will be for froggs, as it'll almost always give you P6Opaque ...
06:52 kjs_ joined #moarvm
07:14 arnsholt left #moarvm
07:14 arnsholt joined #moarvm
08:06 FROGGS joined #moarvm
08:07 zakharyas joined #moarvm
08:23 danaj joined #moarvm
09:02 vendethiel joined #moarvm
09:39 rurban_ joined #moarvm
09:42 kjs_ joined #moarvm
10:00 kjs__ joined #moarvm
10:13 kjs_ joined #moarvm
11:13 vendethiel joined #moarvm
11:25 kjs_ joined #moarvm
12:08 vendethiel joined #moarvm
13:04 dalek MoarVM: b24b0d4 | (Yun SangHo)++ | src/6model/reprs/P6bigint.c:
13:04 dalek MoarVM: Free the buffer after use.
13:04 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/b24b0d4bcc
13:04 dalek MoarVM: 9d25ef2 | (Jimmy Zhuo)++ | src/6model/reprs/P6bigint.c:
13:04 dalek MoarVM: Merge pull request #168 from foollbar/master
13:04 dalek MoarVM:
13:04 dalek MoarVM: Free the buffer after use.
13:04 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/9d25ef23c9
13:09 jnthn Aarrgh, no that's WRONG
13:09 FROGGS that's what I wanted to say
13:09 FROGGS JimmyZ: str is a GCable
13:09 jnthn +        MVM_free(str);
13:10 jnthn You can't just go and free that.
13:10 FROGGS or collectible or what we call it
13:10 jnthn If you're not up to reviewing patches properly, don't merge them.
13:10 FROGGS jnthn: I'm going to revert the PR and add notes
13:10 dalek MoarVM/revert-168-master: c5dbc71 | FROGGS++ | src/6model/reprs/P6bigint.c:
13:10 dalek MoarVM/revert-168-master: Revert "Free the buffer after use."
13:10 dalek MoarVM/revert-168-master: review: https://github.com/MoarVM/MoarVM/commit/c5dbc718bc
13:10 jnthn Well, the second half of it is probably OK
13:11 jnthn Yes, the second hunk is correct
13:11 jnthn First is really not
13:13 dalek MoarVM: c5dbc71 | FROGGS++ | src/6model/reprs/P6bigint.c:
13:13 dalek MoarVM: Revert "Free the buffer after use."
13:13 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/c5dbc718bc
13:13 dalek MoarVM: fbd7a2b | FROGGS++ | src/6model/reprs/P6bigint.c:
13:13 dalek MoarVM: Merge pull request #171 from MoarVM/revert-168-master
13:13 dalek MoarVM:
13:13 dalek MoarVM: Revert "Free the buffer after use."
13:13 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/fbd7a2b745
13:22 jnthn FROGGS++ # thanks
13:24 nwc10 OK, wrong code is wrong
13:24 nwc10 ==26402==ERROR: AddressSanitizer: attempting free on address which was not malloc()-ed: 0x62700226fec0 in thread T0
13:24 nwc10 now I'll test the right code...
13:25 nwc10 (that was during the setting compilation)
13:35 JimmyZ sorry, I don't know why my eye looked at it as  freeable :(
13:43 rurban joined #moarvm
14:00 FROGGS joined #moarvm
15:48 kjs_ joined #moarvm
16:05 kjs_ joined #moarvm
16:10 FROGGS_ joined #moarvm
18:25 kjs_ joined #moarvm
18:26 japhb Mistakes happen.  Even reviewers miss stuff.  If each person has a 20% chance of missing a bug, then even in the best of circumstances where the probabilities are independent, you still have a 4% chance that the coder and reviewer both miss it.
18:26 japhb And if (as with most bugs), the probabilities are *not* independent, it could be significantly higher.
18:27 japhb Still, if you're missing a lot of bugs in review, it's worth figuring out why.
18:28 japhb Also, Git was literally *designed* for multiple levels of review.  Most FOSS projects don't use that capability, but it's definitely there.
18:30 japhb I wonder how well GitHub's UI supports things like sign-off and editing PRs during review ...
19:02 jnthn Mistakes of this nature are also extremely costly when missed.
19:06 * timotimo is not mad at JimmyZ for missing it
19:08 jnthn timotimo: Yeah, but you're not the person who normally ends up tracking down the consequences of GC invariant breakages when they surface some months down the line in hard-to-reproduce situations.
19:08 timotimo that's true ;(
19:27 rurban joined #moarvm
19:53 muraiki joined #moarvm
19:55 rurban_ joined #moarvm
20:19 japhb jnthn: Do you have already a list of GC invariants?  Or for that matter, concurrency/locking invariants, invocation protocol, etc.?  (I haven't looked, so if "yes", then I can go looking on my own.)  I'm kinda wondering if there's any opportunity for machine-assisted invariant checking here.
20:22 timotimo anything that has an MVMObject header in it must not be MVM_Free'd
20:24 jnthn japhb: Somewhat in gc.markdown; it could no doubt be improved...
20:24 jnthn timotimo: MVMCollectable, more generally
20:24 timotimo er, yes
22:06 FROGGS_ joined #moarvm
23:39 kjs_ joined #moarvm

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