00:19:02  * plutoniixjoined
01:47:29  * ncthom91joined
02:01:36  * bradleymeckquit (Quit: bradleymeck)
02:05:16  * bradleymeckjoined
02:19:03  * bradleymeckquit (Quit: bradleymeck)
02:21:14  * ncthom91quit (Quit: Textual IRC Client: www.textualapp.com)
09:14:51  <trungl-bot>Tree closed by [email protected]: closed (https://build.chromium.org/p/client.v8.ports/builders/V8%20Mips%20-%20builder/builds/10881 from 34ed247071e2ecfbbeff3ebefeb817616a79942b)
09:18:52  <trungl-bot>Tree opened by [email protected]: Tree is open (Automatic: (´・ω・`) )
10:25:12  * mylesborinsquit (Quit: farewell for now)
10:25:42  * mylesborinsjoined
10:56:31  <trungl-bot>Tree opened by [email protected]: Tree is open (Branch cut 6.1 happening)
12:00:30  * plutoniixquit (Quit: Leaving)
12:21:56  * Garbeejoined
12:30:29  * hashseedjoined
12:33:33  * hashseedquit (Quit: Page closed)
14:36:17  * plutoniixjoined
16:06:41  <aklein>caitp: do you know if all the typed array iterators really need their own instance types?
16:07:35  <caitp>I can't recall --- I can try to find the design doc for you where this was discussed
16:07:46  <caitp>it was about getting enough information to turbofan
16:09:43  <aklein>caitp: cool, if you can find the doc that'd be useful. we're just about out of InstanceTypes
16:09:57  <caitp>yeah...
16:10:09  <aklein>is this the one? https://docs.google.com/document/d/13z1fvRVpe_oEroplXEEX0a3WK94fhXorHjcOMsDmR-8/edit
16:10:35  <caitp>yep
16:11:30  * RT|Chatzillaquit (Quit: ChatZilla [Firefox])
16:12:04  <aklein>thanks
17:14:46  <caitp>aklein: note that we also have instance types for a lot of FixedArray subclasses, when we I think always know contextually which type they really are
18:07:20  * j0hnsm1thjoined
18:58:19  * xiinotulpjoined
19:01:53  * plutoniixquit (Ping timeout: 248 seconds)
19:20:06  <trungl-bot>Tree opened by [email protected]: Tree is open (Roller off because of branch cut)
20:12:52  <caitp>plus, we could give them same instance type and just a per-isolate map indicating their type, I think
20:14:15  <caitp>or rely on elements kind to know what fixed array type it is? probably an easy way to drop the top instance count by like 12
20:18:33  <caitp>another thing is to change FIRST_NONSTRING_TYPE to be something other than `128`, since most checks for string-ness don't test bit 7, but rather test if less than first non-string type
20:21:10  * kenansulaymanjoined
20:49:34  <aklein>caitp: yeah, there are lots of options. Yang, I think, went with a non-InstanceType thing for this case (which was, indeed, another FixedArray). there's also some chatter about making InstanceType a uint16_t, since apparently there's an extra byte somewhere in Map
20:51:30  * Garbeequit (Quit: Connection closed for inactivity)
20:52:38  <caitp>oh i thought you had some cleanup planned about it or something
20:58:33  <aklein>no, it was an email on the (internal, sigh) team list. could have easily gone to v8-dev
21:20:07  * Garbeejoined
22:12:40  * xiinotulpquit (Quit: Leaving)
22:27:13  * RT|Chatzillajoined
23:41:30  * Garbeequit (Quit: Connection closed for inactivity)