| 01:07:11
| <xaxxon> | koldbrutality, commit: a05f85a3db33860f7f9651d904fd4193ee757848 `v8::internal::Runtime_KeyedStoreIC_Miss(args_length=5, args_object=0x00007fff5fbfdfb8, isolate=0x0000000104800000) + 297 at ic.cc:2488 v8::internal::__RT_impl_Runtime_KeyedStoreIC_Miss(args=Arguments @ 0x00007fff5fbfde60, isolate=0x0000000104800000) + 775 at ic.cc:2502 v8::internal::KeyedStoreIC::Store(this=0x00007fff5fbfdd60, object=Handle<v8::internal::Object> @ 0x00007 |
| 01:07:11
| <xaxxon> | fff5fbfdcd8, key=Handle<v8::internal::Object> @ 0x00007fff5fbfdcd0, value=Handle<v8::internal::Object> @ 0x00007fff5fbfdcc8) + 2135 at ic.cc:2207 |
| 01:07:50
| <xaxxon> | vs v8::internal::Runtime_StoreIC_Miss(args_length=5, args_object=0x00007fff5b03cce8, isolate=0x00007fdf49801000) + 297 at ic.cc:2460 v8::internal::__RT_impl_Runtime_StoreIC_Miss(args=Arguments @ 0x00007fff5b03cb18, isolate=0x00007fdf49801000) + 945 at ic.cc:2475 v8::internal::StoreIC::Store(this=0x00007fff5b03ca10, object=Handle<v8::internal::Object> @ 0x00007fff5b03c858, name=Handle<v8::internal::Name> @ 0x00007fff5b03c850, value=Handle<v |
| 01:07:51
| <xaxxon> | 8::internal::Object> @ 0x00007fff5b03c848, store_mode=CERTAINLY_NOT_STORE_FROM_KEYED) + 1748 at ic.cc:1625 v8::internal::Object::SetProperty(it=0x00007fff5b03c6d8, value=Handle<v8::internal::Object> @ 0x00007fff5b03c610, language_mode=SLOPPY, store_mode=CERTAINLY_NOT_STORE_FROM_KEYED) + 82 at objects.cc:4713 |
| 01:07:53
| <xaxxon> | hrmm |
| 01:07:54
| <xaxxon> | that was dumb |
| 01:07:56
| <xaxxon> | han gon |
| 01:08:10
| <xaxxon> | but anyhow, that was the commit |
| 01:08:59
| <xaxxon> | koldbrutality, both stack traces: https://gist.github.com/xaxxon/5eabcc079a3103abd0f7b3239b2cf3b2 |
| 01:19:49
| <xaxxon> | does it matter if the object with the interceptor comes from a template the context was created from or if it was added after the context was created? |
| 01:48:45
| <xaxxon> | nope. it doesn't. |
| 02:21:04
| * plutoniix | joined |
| 02:58:31
| <koldbrutality> | it already creates an interceptor when you call it |
| 04:14:35
| <xaxxon> | ok, well I'm out of ideas :( Hopefully someone on v8-users will have some ideas |
| 04:31:09
| <xaxxon> | i can't seem to find anything on what it means to be a keyed store vs not |
| 05:04:31
| * plutoniix | quit (Quit: Leaving) |
| 05:15:33
| * thefourtheye | joined |
| 06:48:20
| * plutoniix | joined |
| 06:48:23
| * plutoniix | quit (Max SendQ exceeded) |
| 06:49:52
| * plutoniix | joined |
| 07:22:32
| <xaxxon> | plutoniix, thefourtheye anyone have any idea why when I register a named property handler, when I do a set, why it would call either the query or (if there is no query) a getter callback instead of a setter? I can't repro it with simplfied code, either. The only thing I see is the one that works does a keyed store and the one that doesn't isn't a keyed store. Thank you. |
| 07:34:02
| * Keverw | quit (Quit: Textual IRC Client: www.textualapp.com) |
| 07:55:05
| <xaxxon> | or I guess.. it means "holder isnt receiver" but I'm not 100% on holder. that means holder != this... but still not 100% on what it means |
| 08:02:13
| * thefourtheye | quit (Quit: Connection closed for inactivity) |
| 08:12:39
| <xaxxon> | reading that thread about it again |
| 08:27:32
| <xaxxon> | I guess I'm seeing why it doesn't work.. because I'm not asking it to write to the object I set the handler on |
| 08:27:44
| <xaxxon> | i'm not sure how/why it's a different object, but I guess it makes sense |
| 08:32:42
| <xaxxon> | does it somehow know if I put an object in an internal field? |
| 08:36:12
| * plutoniix | quit (Quit: Leaving) |
| 08:36:13
| <xaxxon> | so maybe a better question to ask... how do I map arbitrary property sets back to a std::map embedded in the internalfield of an object? |
| 08:56:30
| * plutoniix | joined |
| 09:07:40
| <xaxxon> | got my answer on mailing list. |
| 10:17:11
| <xaxxon> | but I still do'nt know exactly what it is I'm doing that causes the holder != this :( |
| 10:27:38
| * plutoniix | quit (Quit: Leaving) |
| 11:27:26
| * caitp- | changed nick to caitp |
| 12:32:34
| * xaxxon | quit (Quit: This computer has gone to sleep) |
| 13:01:36
| * bradleymeck | joined |
| 13:23:29
| * bradleymeck | quit (Quit: bradleymeck) |
| 13:26:43
| * bradleymeck | joined |
| 14:09:42
| * bradleymeck_ | joined |
| 14:10:05
| * bradleymeck | quit (Ping timeout: 265 seconds) |
| 14:10:05
| * bradleymeck_ | changed nick to bradleymeck |
| 14:19:54
| * mikolalysenko | quit (Ping timeout: 250 seconds) |
| 14:21:28
| * mikolalysenko | joined |
| 15:31:50
| * aklein | quit (Ping timeout: 250 seconds) |
| 15:33:25
| * aklein | joined |
| 15:52:52
| * seventh | joined |
| 16:04:11
| * RT|Chatzilla | quit (Quit: ChatZilla 0.9.86.1 [Firefox 2.0.0.22pre/2010030309]) |
| 16:07:34
| * bradleymeck | quit (Quit: bradleymeck) |
| 16:28:41
| * bradleymeck | joined |
| 16:45:33
| * Guest59 | joined |
| 16:47:54
| * Guest59 | quit (Read error: Connection reset by peer) |
| 16:48:31
| * Guest59 | joined |
| 16:55:43
| * plutoniix | joined |
| 18:41:09
| * bradleymeck | quit (Read error: Connection reset by peer) |
| 18:49:17
| * Guest59 | quit (Ping timeout: 260 seconds) |
| 19:17:58
| * bradleymeck | joined |
| 19:29:39
| * bradleymeck | quit (Read error: Connection reset by peer) |
| 19:36:48
| * seventh | quit (Quit: ...) |
| 19:53:44
| * Cube8 | joined |
| 20:04:05
| <caitp> | we throw a really weird error when calling Promise with an undefined new.target |
| 20:04:16
| <caitp> | "undefined is not a promise" |
| 20:05:06
| <caitp> | `throw %make_type_error(kNotAPromise, this);` doesn't seem to communicate what we want, imo |
| 20:05:18
| <caitp> | anyways |
| 20:14:41
| <gsathya> | oh damn, ff has a better error msg and links to the mdn with a desc even |
| 20:20:03
| * Cube8 | quit (Quit: Leaving) |
| 20:28:32
| * xaxxon | joined |
| 20:48:56
| <jwolfe> | when i get an assertion failure in d8, i only get a meaningful stack trace in debug builds. i thought there was a way to turn on debug symbols in release builds, but maybe i was mistaken. |
| 20:49:17
| <jwolfe> | passing debugsymbols=on to make doesn't work, and i don't think there's any way to make gn/ninja work. |
| 20:51:15
| <jwolfe> | "make gn/ninja work" meaning get debug symbols in release builds in gn/ninja. |
| 20:54:53
| <caitp> | I'm guessing it's related to the StackTrace stringification for your platform (I'm guessing stack_trace_posix.cc?) |
| 20:54:57
| <caitp> | jwolfe: ^ |
| 20:56:23
| <jwolfe> | caitp, are you saying that you can get a stack trace from release builds on your platform? |
| 20:57:10
| <caitp> | I'm using stack_trace_posix |
| 20:57:30
| <caitp> | hard to see what would be broken in the release case, but it should go through there afaik |
| 21:01:05
| <caitp> | looks like trace_symbols() is null, so it just takes the "output pointers" path instead |
| 21:01:14
| <caitp> | if I'm reading it correctly |
| 21:01:26
| <caitp> | so, what causes that to happen? no idea :> |
| 21:01:37
| <caitp> | `man backtrace_symbols` likely has an answer |
| 21:03:47
| <caitp> | http://stackoverflow.com/questions/6934659/how-to-make-backtrace-backtrace-symbols-print-the-function-names?rq=1 seems to have some suggestions that go beyond compiling with `-g` |
| 21:14:21
| <jwolfe> | caitp, so you're saying that no one can get a stack trace in release builds? you're suggesting i change the build options beyond what the existing make and gn infrastructures can do, and/or use a different runtime library in d8. those are all a lot more complicated than throwing a command line switch on a make invocation. |
| 21:14:54
| <caitp> | I'm sure there's some way, but I haven't tried. I'm just offering suggestions that could potentially work |
| 21:15:02
| <caitp> | given the tools that are being used |
| 21:23:52
| * zv | quit (Ping timeout: 250 seconds) |
| 21:40:25
| * zv | joined |
| 22:31:56
| * RT|Chatzilla | joined |
| 22:51:29
| <xaxxon> | can you get a stringification of a symbol? |
| 22:52:35
| <xaxxon> | also, is symbol still "experimental"? |
| 23:02:48
| * Guest59 | joined |
| 23:17:11
| <caitp> | xaxxon: v8::Symbol::Name() should get you the description |
| 23:17:38
| <caitp> | symbols are part of es6, they've been around for a while |
| 23:17:58
| <xaxxon> | caitp, yeah, on the docs it says "experimental use at your own risk" |
| 23:18:25
| <xaxxon> | and the name() method was hiding in plain site .. just all those static member functions made me not see it. thank you |
| 23:18:46
| <xaxxon> | http://bespin.cz/~ondras/html/classv8_1_1Symbol.html "This is an experimental feature. Use at your own risk." |
| 23:18:52
| <xaxxon> | oops |
| 23:18:58
| <caitp> | symbols aren't going anywhere |
| 23:19:03
| <caitp> | the "Name" method might, but not symbols |
| 23:19:12
| <xaxxon> | http://v8.paulfryzel.com/docs/master/classv8_1_1_symbol.html "This is an experimental feature. Use at your own risk." |
| 23:19:23
| <xaxxon> | docs generated today |
| 23:19:25
| <xaxxon> | cool |
| 23:21:11
| <caitp> | yeah, the api docs have probably not been touched in 2 or 3 years re: that class |
| 23:21:14
| <caitp> | if not longer |
| 23:22:29
| <caitp> | yeah, that line of docs hasn't been updated since 2013, i think you're safe using it :p |
| 23:22:46
| <xaxxon> | maybe I'll make a patch. |
| 23:22:57
| <xaxxon> | so I can be a 2-time contributor :) |
| 23:24:34
| * Guest59 | quit (Ping timeout: 252 seconds) |
| 23:53:33
| * Guest59 | joined |