Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2015-08-30

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

All times shown according to UTC.

Time Nick Message
00:00 lizmat joined #moarvm
00:08 TEttinger joined #moarvm
01:15 vendethiel joined #moarvm
03:45 colomon success!  \o/
03:46 * colomon got his dop kit into his overstuffed suitcase
06:31 FROGGS[mobile] joined #moarvm
07:37 vendethiel joined #moarvm
07:56 virtualsue joined #moarvm
07:58 FROGGS joined #moarvm
08:04 vendethiel joined #moarvm
09:21 brrt joined #moarvm
09:35 dalek MoarVM/even-moar-jit: ac58687 | brrt++ | src/jit/ (7 files):
09:35 dalek MoarVM/even-moar-jit: Factor out the 'mem' abstraction
09:35 dalek MoarVM/even-moar-jit:
09:35 dalek MoarVM/even-moar-jit: The 'mem' value type could refer to any number of registers
09:35 dalek MoarVM/even-moar-jit: withoutt ensuring that they were live values. Adding support
09:35 dalek MoarVM/even-moar-jit: for such values would be complicated, and it's not necessary
09:35 dalek MoarVM/even-moar-jit: either (the tiler can take care of it).
09:35 dalek MoarVM/even-moar-jit: review: https://github.com/MoarVM/MoarVM/commit/ac586875e4
09:40 timotimo cool
09:42 timotimo colowhat is a "dop kit"?
09:44 timotimo "Trying to free too many registers" - aaw
09:45 dalek MoarVM/even-moar-jit: b900755 | brrt++ | build/Makefile.in:
09:45 dalek MoarVM/even-moar-jit: Remove JIT intermediates in make clean
09:45 dalek MoarVM/even-moar-jit: review: https://github.com/MoarVM/MoarVM/commit/b9007553fe
09:46 * brrt afk for now
14:52 FROGGS[mobile] joined #moarvm
19:37 colomon joined #moarvm
19:37 colomon o/
19:38 * colomon is back in the States
19:46 masak \o
19:50 TEttinger joined #moarvm
19:52 colomon what’s the current count?  :)
19:54 timotimo welcome back to the ground, colomon :)
19:55 colomon I can’t tell you how happy I was to get off that plane.  Big Sol, tiny cramped seat.
19:56 colomon at least I’ve got an aisle for my last leg home, and a few more hours to recoup before I have to board.
19:58 timotimo yeah, sol is really big
19:58 timotimo i'd go so far as to call it gigantic
20:09 colomon http://smoke.perl6.org/report — last column is (I think) the most account smoke test run in ages.
20:09 colomon woah, GLR count is 4?
20:10 timotimo glr count?
20:10 colomon 4 failing test files?
20:10 colomon at least, that’s what I just got locally
20:11 colomon well, techincally I am in Chicago and the server I ran it on is in Michigan, but that’s close enough to local for me.  ;)
20:14 colomon m: say <:13<01>/:13<07>>
20:14 camelia rakudo-moar cf3b24: OUTPUT«:13<01>/:13<07>␤»
20:14 colomon … errr ....
20:15 colomon maybe I’d better backlog
20:15 colomon oh, how’d I get in moarvm?  apologies, good folk
20:15 timotimo :)
20:15 timotimo we like visitors :)
20:23 vendethiel .oO( you're bound to stay now )
20:24 timotimo hey vendethiel
20:25 vendethiel hi hi
20:35 virtualsue joined #moarvm
21:16 rarara_ joined #moarvm
21:17 lizmat joined #moarvm
21:20 lizmat_ joined #moarvm
21:27 dalek MoarVM: 1172953 | paultcochrane++ | src/spesh/candidate.c:
21:27 dalek MoarVM: Remove potential double-free in MVM_spesh_candidate_setup
21:27 dalek MoarVM:
21:27 dalek MoarVM: The `before` and `after` variables were being potentially freed twice.  By
21:27 dalek MoarVM: only freeing them in the outer scope of their usage means that the resources
21:27 dalek MoarVM: are still freed before exiting the function and the code doesn't try to free
21:27 dalek MoarVM: already-freed variables.
21:27 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/11729536a8
21:27 dalek MoarVM: 511c467 | lizmat++ | src/spesh/candidate.c:
21:27 dalek MoarVM: Merge pull request #245 from paultcochrane/pr/remove-double-free
21:27 dalek MoarVM:
21:27 dalek MoarVM: Remove potential double-free in MVM_spesh_candidate_setup
21:28 dalek MoarVM: 613153e | paultcochrane++ | src/gc/orchestrate.c:
21:28 dalek MoarVM: Cast unsigned short to MVMunit64 to avoid integer overflow
21:28 dalek MoarVM:
21:28 dalek MoarVM: Coverity noticed that in the expression changed here, the variable
21:28 dalek MoarVM: `tc->instance->num_user_threads` was an unsigned short, which was then being
21:28 dalek MoarVM: implicitly promoted to int and then promoted again to unsigned long.
21:28 dalek MoarVM: However, if the result of the expression is sufficiently large, this won't
21:28 dalek MoarVM: calculate the correct value anymore since the value overflows.  By casting
21:28 dalek MoarVM: `num_user_threads` to MVMunit64 we ensure that the overflow is avoided.
21:28 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/613153e232
21:28 dalek MoarVM: 7e9f29e | lizmat++ | src/gc/orchestrate.c:
21:28 dalek MoarVM: Merge pull request #246 from paultcochrane/pr/avoid-integer-overflow
23:01 lizmat joined #moarvm

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