|
github.com/moarvm/moarvm | IRC logs at colabti.org/irclogger/irclogger_logs/moarvm Set by AlexDaniel on 12 June 2018. |
|||
|
00:15
patrickz joined
00:18
patrickb left
01:22
patrickz left
|
|||
| ugexe | MasterDuke: how do i fix github.com/MoarVM/MoarVM/pull/1083 ? for instance your comment is on operand 2, but the expression template goes like $1, $2, null, $3 when I expect $1, $2, null, $4 . I tried a few different ways and all of them seemed to work so I dont know what is actually correct | 02:11 | |
| s/operand 2/operand 3/ | 02:12 | ||
| MasterDuke | $1, $2, null, $3 is what i assumed should be used based on github.com/MoarVM/MoarVM/blob/mast...5293-L5297 | 02:17 | |
| ugexe | so inside graph.c i can just ignore operand 3? | 02:19 | |
| or ignore operand 4? | 02:21 | ||
| i know i still need to set a null inside MVMJitCallArg args | |||
| but i dont need to get that from anything, i can just do { MVM_JIT_LITERAL_PTR, { 0 } } | 02:23 | ||
| MasterDuke | i think i would just ignore ins->operands[3] | ||
| ugexe | so does operands[4] end up in $4 (and thus not used) or is put into $3? | 02:25 | |
| i feel like im overlooking something very obvious | 02:27 | ||
| MasterDuke | decodeconf should only have 3 operands, decoderepconf should have 4 | 02:30 | |
| as i understand it | |||
| they both call the same underlying moarvm function, but decodeconf calls it with a NULL as one of its arguments and decoderepconf calls it with all 4 of its arguments | 02:31 | ||
| i think grabbing 4 operands for decodeconf works because C doesn't bounds check and you're just grapping some random memory, which just happens to not cause any problems | 02:33 | ||
| ugexe | that does make sense, good explaination | 03:08 | |
| MasterDuke | oO(grapping? how did i make that typo?) | 03:12 | |
|
06:19
robertle left
06:57
nebuchadnezzar joined
07:12
domidumont joined
07:38
domidumont1 joined
07:41
domidumont left
08:03
zakharyas joined
08:04
robertle joined
08:45
domidumont1 left
10:14
domidumont joined
10:53
domidumont1 joined
10:56
domidumont left
11:48
lucasb left
12:15
zakharyas left
12:46
AlexDaniel joined
12:50
yoleaux joined
12:58
domidumont1 left
13:12
domidumont joined
13:27
zakharyas joined
13:29
AlexDaniel left
13:58
robertle left
14:01
robertle joined
15:03
lucasb joined
16:23
domidumont left
16:42
robertle left
17:31
robertle joined
18:09
mgoebel joined
18:47
zakharyas left
18:50
mgoebel left
18:57
domidumont joined
19:49
domidumont left
|
|||
| japhb | Why does 645a70c0d include the lego jit pieces for both backtrace and backtracestrings, but the expr jit template for only backtrace? | 22:04 | |
| ugexe | because there is already exprjit template for backtracestrings | 22:09 | |
| japhb | Ah. | 22:13 | |
|
23:12
lucasb left
|
|||