| ugexe | 🤔 | 01:47 | |
| Woodi | patrickb: s/helper program/helper function/ probably | 03:35 | |
| patrickb | Woodi: I actually meant helper program as in a separate executable file. But: I could just as well add a hidden command line option to the moar executable and make moar exec itself. But apart from saving myself to tune the build system, it's the same thing. | 07:17 | |
|
08:18
japhb left
08:40
sena_kun joined
|
|||
| Geth | MoarVM/trash_debugspam_for_azure: 1bb8215f40 | (Timo Paulssen)++ | 2 files sprinkle random telemetry output: CU load and NFA run |
08:53 | |
| MoarVM/trash_debugspam_for_azure: 9169e7a55a | (Timo Paulssen)++ | 3 files more telemeh info on gc, correctly align pointers in output |
09:20 | ||
|
09:24
sena_kun left
09:26
sena_kun joined,
sena_kun left
09:34
japhb joined
|
|||
| timo | i think turning off the jit may have made rakudo compile successfully | 09:36 | |
|
09:57
japhb left
09:59
japhb joined
|
|||
| ab5tract | timo: awesome! Is it clear what arch is used by these runners? I would assume x64 because Wintel, but aarch64 is supposedly cheaper and this is M$, so.. | 10:49 | |
| Also: my sincere apologies for (I think) partially prompting you to go down this rabbit hole | 10:50 | ||
| timo | there is no jit on aarch64 so turning off jit wouldn't have made a difference | 10:59 | |
| do you have time to look into the Test::Scheduler for that Throttle/vent-at=>0 thing? i think the PR is waiting for a signal from you | 11:03 | ||
| maybe someone wants to contribute a /Yc + /Yu patch for building moar on MSVC with precompiled headers | 11:17 | ||
| maybe that will make things faster | |||
| also check if /MT helps at all, but for that we need to put multiple input files in one invocation rather than the one-source-per-call we have right now which is obviously easier to make a makefile for, especially when working with nmake | |||
| also, maybe gmake is available in these runners and we can have a non-shit makefile | 11:18 | ||
| nope, the only "make" i see in the "included software" page is "CMake" | 11:19 | ||
| i bet powershell can be fun if you learn it a bit | |||
| ugexe | I pushed a small zef update to a branch and now my appveyor CI fails when compiling rakudo heh | 13:03 | |
| er moar rather | |||
| compiling src\main.obj | |||
| main.c | |||
| C:\projects\rakudo\nqp\MoarVM\src\math/num.h(23): error C2099: initializer is not a constant | |||
| timo | static const MVMnum64 MVM_NUM_NAN = NAN; | 13:04 | |
| ugexe | Did anything change related to that recently? | 13:06 | |
| The last successful ci run for zef was 2 months ago to give a time frame | 13:07 | ||
| timo | that file has last been touched at all 3 years ago | 13:09 | |
| ugexe | That makes it seem like NAAN has recently become defined so that it takes the first path in that conditional | 13:21 | |
| Since github.com/MoarVM/MoarVM/blob/570d.../num.h#L10 did not error | |||
| (Typing on a phone) | 13:22 | ||
| timo | you can try supplying /EP to the C compiler arguments, that will output what the preprocessor has done to standard output. it will be a huge output for every file you do it with, so ideally only do it for a single file. you can put that in the CL environment variable and the compiler will pick it up | 13:35 | |
| ab5tract | timo: I wasn’t sure if it was automatically disabled based on arch. Makes sense, but I’ve had my hands in the guts of that part of nqp-configure and straightforward it is not | 13:42 | |
| I’ve been banging my head on Comma and going in circles. I’ll take a break and look at that PR tonight. | 13:43 | ||
| (Re: jit, what I mean is I wouldn’t have been surprised to find that it was disabled based on os + arch, rather than on arch alone) | 13:44 | ||
| ugexe | i'll give that a try later | 14:29 | |
| timo | aw crap, since the cache runs at the very end, it will actually put rakudo + nqp + moar in the nqp and the moar cache instead of just the parts i want | 14:38 | |
| guess i have to move stuff sideways first ... | 14:39 | ||
|
14:41
[Coke]_ joined,
[Coke]_ is now known as a,
a is now known as Guest5605
14:44
Guest5605 left,
[Coke]_ joined
14:52
[Coke]_ left
15:42
[Coke]_ joined
15:45
[Coke] is now known as Guest1428
15:46
[Coke]_ is now known as [Coke]
15:53
lizmat_ joined
15:56
evalable6 joined,
bloatable6 joined
15:57
greppable6 joined
16:03
camelia left,
lizmat left,
bloatable6__ left,
greppable6__ left,
evalable6__ left,
jjatria left,
lizmat_ left,
lizmat joined
16:10
jjatria joined
17:01
camelia joined
17:34
Guest1428 left
17:42
[Coke]_ joined
19:15
[Coke] left
19:36
[Coke] joined
20:19
sena_kun joined
21:37
sena_kun left
22:00
sena_kun joined
22:31
sena_kun left
|
|||