|
01:45
japhb joined
08:46
sena_kun joined
09:06
sena_kun left
09:20
sena_kun joined
10:21
sena_kun left
|
|||
| patrickb | timo: I've added a big comment to the longjmp PR. I think we should keep the fixes for all the archs where MS tries to do stack unwinding even if we don't JIT yet. It doesn't hurt and we can then rely on the semantics of longjmp(). | 14:07 | |
| timo, ab5tract: If you don't object, I'd like to keep the longjmp PR the way it is now (i.e. don't extend the build system to detect the stack unwinder and only run the fix in affected scenarios). | 14:10 | ||
| I think trying to be smart about the longjmp situation might actually hurt. What if we do a build on a system where the fix isn't needed and then try to distribute the build to other Windows systems? | 14:11 | ||
| timo | OK, that sounds reasonable | 14:14 | |
| do you think it's too close to the next release to merge it? | |||
| lizmat | there's been no blin run yet, so I think it'd be ok | 14:16 | |
| timo | we don't blin for windows, do we? | 14:18 | |
| lizmat | fwiw, I don't think we do | ||
| so in that respect the question is moot anyway | |||
| timo | we don't have a big number of people who build moar&nqp&rakudo from main/main/main every day | 14:19 | |
| patrickb | lizmat: Has the pumpking situation for the next release been resolved? | 14:20 | |
| lizmat | ? you mean the release manager ? | 14:21 | |
| patrickb | Yes | ||
| lizmat | not yet... [Coke] said he'd consider it | ||
| timo | en.wikipedia.org/wiki/Court_shoe is wearing those a prerequisite? | 14:22 | |
| lizmat | I don't think so :-) | 14:23 | |
| timo | they are spelled "pumps" in german, too, but pronounced like ... pöms? doesn't really translate to text, IPA says /pœmps/ | 14:26 | |
| de.wikipedia.org/wiki/Pumps actually these look kinda different from the ones in the en.wikipedia.org article that it links to as "the english version of this" | 14:28 | ||
| patrickb | timo: Just curious, do you know why it's possible for you to push to a branch that lives under /patrickbkr/MoarVM ? | ||
| timo | yes, there's a setting you can set on a pull request | 14:30 | |
| i think it's called something like "allow maintainers to edit" or some such | |||
| patrickb | Ah! Cool. I didn't know about that.- | 14:31 | |
| timo | the comment is very good btw | ||
| patrickb | Thanks! | 14:32 | |
| [Coke] | Yes, it's too close to the release. | 15:25 | |
| Having my crappy blin run be the arbiter of "safe" is not entirely ... safe. | |||
| (also yes, I have not yet agreed :) | |||
| But am guessing I am the only blin alternative at this point. | 15:27 | ||
| pumps in english is more of a lady's shoe with heels, I think. | 15:28 | ||
| not "opera pumps". | |||
| lizmat | and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2024/11/18/2024-...butterfly/ | 17:07 | |
| timo | so, accessing a stale hash table's control structure currently oopses moarvm ... but does it really have to? | 19:06 | |
|
19:51
sena_kun joined
|
|||
| timo | do we want to investigate pthread thread cancellation? | 20:12 | |
|
22:00
sena_kun left
|
|||
| lizmat | nicholas originally implemented that to prevent random crashes: at least now we haz an exception? | 22:07 | |
| timo | well, an oops is more than just an exception | 22:08 | |
| it can't be caught, it exits moar ASAP | |||
| lizmat | true, but at least it points at a specific cause | 22:10 | |
| timo | sure | 22:17 | |