[00:02] *** reportable6 left [00:05] *** reportable6 joined [02:04] *** nativecallable6 left [02:04] *** statisfiable6 left [02:04] *** sourceable6 left [02:04] *** squashable6 left [02:04] *** bloatable6 left [02:04] *** bisectable6 left [02:04] *** evalable6 left [02:04] *** reportable6 left [02:04] *** greppable6 left [02:04] *** committable6 left [02:04] *** unicodable6 left [02:04] *** releasable6 left [02:04] *** notable6 left [02:04] *** quotable6 left [02:04] *** benchable6 left [02:04] *** linkable6 left [02:04] *** tellable6 left [02:04] *** shareable6 left [02:04] *** coverable6 left [02:04] *** coverable6 joined [02:04] *** sourceable6 joined [02:04] *** nativecallable6 joined [02:04] *** bisectable6 joined [02:05] *** quotable6 joined [02:05] *** unicodable6 joined [02:05] *** reportable6 joined [02:05] *** releasable6 joined [02:06] *** statisfiable6 joined [02:06] *** committable6 joined [02:06] *** squashable6 joined [02:06] *** linkable6 joined [02:06] *** notable6 joined [02:06] *** greppable6 joined [02:06] *** evalable6 joined [02:07] *** benchable6 joined [02:07] *** bloatable6 joined [02:07] *** shareable6 joined [02:07] *** tellable6 joined [05:32] *** Kaiepi left [05:33] *** Kaiepi joined [05:42] *** Kaiepi left [05:45] *** Kaiepi joined [06:00] *** Kaiepi left [06:00] *** Kaiepi joined [06:02] *** reportable6 left [06:05] *** reportable6 joined [06:45] *** frost joined [06:48] *** patrickb joined [08:18] *** patrickb left [08:45] Moarning. So it turns out I can turn one of the failing NQP tests after switching method dispatch to new-disp into a SEGV by deleting some of the test, and so probably the other odd failure mode is also a result of some kind of memory corruption... [08:46] "my machine" (several of them) are currently powered down because they in Oregon, and it's hot there. [08:47] so I don't have access to anything fast with ASAN or valgrind [08:50] I'm guessing "hot" means "very unusually hot"... [08:51] whole of the west coast of North America is, I believe. Canada is breaking records. [08:51] yeah, last few days have been hell [08:51] so powering down non-essential services seemed like a good plan [09:01] *** patrickb joined [09:13] *** patrickb left [09:21] uhh... [09:21] MVMObject *fateres = MVM_repr_alloc_init(tc, tc->instance->boot_types.BOOTIntArray); [09:21] MVM_ASSERT_NOT_FROMSPACE(tc, fateres); [09:21] The assert fails right after allocation?! [09:24] Ah, I guess it means something wrote into the previous nursery [09:25] So it's not all zeroed [09:40] *** jdv left [09:40] *** jdv joined [09:41] ooh fun. Once upon a time I had a hacky patch to allocate nurseries with mmap and change them to unaccessable when not in use. That would have exploded with this. [09:41] if your guess is right [09:42] hmm, I wonder *how* hacky it is, and whether it's viable to make it a conditional compile... [09:42] but not this morning. Or at least, this part of today. [09:44] Yeah, I did wonder about mprotect hacks [09:50] Can't rr it as it seems not to support AMD CPUs (or AMD CPUs don't support it...) [09:50] Aaa(rr)gh [09:50] At least one of the (powered off) machines could run rr [09:51] I can't remember if it was Intel, or if there is some new enough AMD CPU that rr can work with. [09:51] Either way, can't currently help [09:56] Ah, it seems there are patches in rr for zen and zen 2 architectures, but I think I've zen 3 [09:56] So mine is maybe too new [09:56] argh [10:04] Turns out the cheap way to get some an effect is to free the fromspace memory always after GC is done [10:05] And then let ASAN report the write to freed memory [10:11] Goodness, this finds all kinds [10:28] Bleh, though having now fixed those, it turns out they weren't the actual issue the test runs into, which instead seems to be about memory managed with the FSA [10:28] meeting and lunch now, guess I'll try to hunt it down later [10:48] Nicholas, jnthnwrthngtn: https://github.com/MoarVM/MoarVM/commit/9d468afcaf21b01a85126142c758674d9f5b579c [10:51] jnthnwrthngtn: rr runs just fine on zen3. I use https://github.com/rr-debugger/rr/wiki/Zen#kernel-module [10:52] Just load that kernel module and rr works like a charm [10:52] jnthnwrthngtn, https://github.com/rr-debugger/rr/wiki/Zen says it should work. :/ [11:03] Oh. When I ran rr it pointed me to a github issue about it, when I read through, and saw no mention of that :) [11:08] *** squashable6 left [11:09] *** squashable6 joined [11:13] *** patrickb joined [12:02] *** reportable6 left [12:04] *** reportable6 joined [12:12] *** frost left [12:25] *** dogbert17 joined [12:28] *** dogbert11 left [13:08] *** dogbert11 joined [13:11] *** dogbert12 joined [13:11] *** dogbert17 left [13:13] *** dogbert11 left [14:22] Current NQP HEAD compiles call and callmethod QAST nodes into dispatch instructions. [14:23] It builds. 3 failing tests, 2 a bug I need to hunt, 1 an NYI. Also an intermittent hang that I'm looking at now. [14:26] I also discovered that the QAST regex compiler will need updating since it directly emits findmeth/invoke. There are likely some other such places to update too [16:24] *** patrickb left [16:24] *** sortiz joined [17:02] *** sortiz left [18:02] *** reportable6 left [18:03] *** reportable6 joined [19:20] *** evalable6 left [19:20] *** linkable6 left [19:21] *** linkable6 joined [19:23] *** evalable6 joined [21:04] nine: that reminds me of the jvm XXXalot flags they use for testing [22:08] *** AlexDaniel left [22:09] *** AlexDaniel joined [22:39] *** cognominal left [22:41] *** cognominal joined