|
01:34
eater joined
02:59
ilbot3 joined
04:13
colomon joined
05:12
committable6 joined
06:04
releasable6 joined
06:35
brrt joined
|
|||
| brrt | good * #moarvm | 06:43 | |
|
06:56
domidumont joined
07:03
domidumont joined
07:05
geospeck joined
|
|||
| nwc10 | good *, brrt | 07:18 | |
| brrt | good * nwc10 | 07:20 | |
| have you also seen the blowup of CORE.setting recently? | |||
| nwc10 | blowup in what sense? It still works (builds) for me, but I didn't notice if it needs more time or more RAM | 07:23 | |
|
07:46
brrt joined
|
|||
| brrt | as in, ASAN crashes | 07:56 | |
| nwc10 | ASAN on "my" machine doesn't crash on it | 08:04 | |
| brrt | hmmm | 08:08 | |
| what architecture is that | |||
| nwc10 | x86_64 GNU/Linux | ||
| nwc10 goes to replace "emperor's new cup of coffee" with something a bit less vapourware | 08:16 | ||
|
08:18
committable6 joined,
benchable6 joined,
bisectable6 joined,
unicodable6 joined,
squashable6 joined
08:19
squashable6 joined,
bisectable6 joined
09:13
zakharyas joined
09:16
zakharyas joined
09:17
zakharyas joined
09:20
zakharyas joined
10:44
zakharyas joined
10:51
domidumont joined
11:11
AlexDaniel joined
11:13
quotable6 joined
11:22
AlexDaniel joined
13:58
zakharyas joined
13:59
robertle joined
14:38
AlexDaniel joined
|
|||
| lizmat | S01-perl-5-integration/basic.t is reliably segfaulting for me now | 14:58 | |
| brrt | yay | 15:07 | |
| have you tried to point jit-bisect —spesh at that | |||
| or just jit-bisect.pl (for the expr jit) | |||
| lizmat | MVM_JIT_EXPR_DISABLE=1 is the env var, right ? | 15:08 | |
| to switch off expr jit ? | 15:09 | ||
| brrt | yeah | ||
| but jit-bisect.pl will do that automatically for you | |||
| lizmat | it also segfaults with MVM_JIT_EXPR_DISABLE=1 , but not as often | 15:10 | |
| could it be that it doesn't make a difference between MVM_JIT_EXPR_DISABLE=1 and MVM_JIT_EXPR_DISABLE=0 ? | 15:11 | ||
| brrt | yes :-D | 15:18 | |
| oh, you mean, running jit-bisect.pl | |||
| yeah, that will clear out any such flags prior to running | |||
|
16:10
geospeck joined
16:13
brrt joined
17:00
zakharyas joined
17:06
AlexDaniel joined
17:29
brrt joined
17:44
AlexDaniel joined
17:57
geospeck joined
18:04
zakharyas joined
18:24
domidumont joined
18:41
AlexDaniel joined
|
|||
| AlexDaniel | samcv: created an issue for it anyway: github.com/MoarVM/MoarVM/issues/764 | 18:56 | |
|
19:00
geospeck joined
19:28
quotable6 joined
19:29
quotable6 joined
20:06
zakharyas joined
20:07
quotable6 joined
20:08
quotable6 joined,
bloatable6 joined,
nativecallable6 joined,
unicodable6 joined,
bisectable6 joined,
greppable6 joined,
coverable6 joined,
reportable6 joined,
benchable6 joined,
evalable6 joined,
releasable6 joined,
committable6 joined,
squashable6 joined,
statisfiable6 joined
20:12
geospeck joined
20:43
nine_ joined,
arnsholt joined
20:48
eater joined
21:13
geospeck joined
21:51
colomon_ joined
22:58
ilmari[m] joined
|
|||
| samcv | AlexDaniel: have the fix ready | 23:00 | |
| Geth | MoarVM: c749da122c | (Samantha McVey)++ | tools/ucd2c.pl ucd2c.pl: Explicitly set default values for several properties Also, set the default values for name, gencat_name and General_Category for codepoints not in UnicodeData.txt. |
23:02 | |
| MoarVM: 65fa583822 | (Samantha McVey)++ | tools/ucd2c.pl ucd2c.pl: Rename int and str enum variables to indicate their use This makes it much easier to tell which property is being added to, the switch for MVM_unicode_get_property_int or MVM_unicode_get_property_str. |
|||
| MoarVM: 34d01a811e | (Samantha McVey)++ | 2 files Fix the Unicode database from adding int properties to str function This was causing a compiler warning because of the type difference. We shouldn't be requesting, or at least should expect to get an empty string if we query an int property using the getuniprop_str op so there should be no difference in Perl6 functionality since it only requests using getuniprop_int. Fixes issue #764 |
|||