00:25:09  * plutoniixjoined
02:36:24  * arvquit (Ping timeout: 256 seconds)
02:36:24  * mmarchiniquit (Ping timeout: 256 seconds)
02:36:32  * mmarchinijoined
02:36:36  * arvjoined
05:20:19  * AtumT_joined
05:23:08  * AtumTquit (Ping timeout: 256 seconds)
05:33:49  * AtumTjoined
05:35:00  * AtumT_quit (Ping timeout: 260 seconds)
10:25:28  * mylesborinsquit (Quit: farewell for now)
10:25:58  * mylesborinsjoined
10:40:32  * plutoniixquit (Quit: Leaving)
15:33:23  <Wes->Is it possible to run two completely independent instances of v8 in the same process, on different threads?
15:53:52  <bradleymeck>Wes-: v8::Isolate , yes
15:54:15  <bradleymeck>if you mean v8::V8 , no
15:54:31  <bradleymeck>but i'd be curious about when you want separate v8::V8's anyway
15:56:31  <Wes->bradleymeck: Thanks. I don't have a use-case for v8::V8, I don't think. v8::Isolate is the same as a tab in Chrome from a security perspective?
16:00:33  * bradleymeckis not on chromium team / assumes Chromium is using separate process per tab due to https://chromium.googlesource.com/chromium/src/+/master/docs/security/side-channel-threat-model.md
16:12:37  * cloudshujoined
16:30:18  * Mariusz08_joined
16:31:59  * Mariusz08_part ("Leaving")
16:42:19  * Guest59_joined
16:44:50  * Guest59quit (Ping timeout: 256 seconds)
16:48:52  * RT|Chatzillaquit (Quit: ChatZilla [Firefox])
17:01:03  <trungl-bot>Tree closed by [email protected]: closed (https://ci.chromium.org/buildbot/client.v8/V8%20Linux%20-%20arm64%20-%20sim%20-%20MSAN/21562 from 9e27d4735f694a9e72a3d1c028a5c3044fc74d27)
17:13:12  <Wes->bradleymeck: Very interesting link, thanks. My thoughts are also impacted re. Spectre and friends.
17:13:35  <bradleymeck>Wes-: i'm not entirely convinced of that conclusion, but NDA wall
17:19:12  <Wes->bradleymeck: I'm not either - but I think it makes an excellent default position
17:20:03  <bradleymeck>Wes-: given that there are PoC of cross process exploits i'm not sure
17:22:42  <Wes->bradleymeck: AH. You're thinking in the /other/ direction. I'm still arguing locally about an arch change which would allow multiple hunks of untrusted code to run in multiple isolates in the same address space. My position -- opposed. My current thinking is that the only guaranteed way to stay save is to run them in fully virtualized VMs, but that has a serious cost.
17:23:24  <Wes->An intermediary position - separate address space - seems worth exploring, however.
17:35:19  <trungl-bot>Tree opened by [email protected]: open
17:37:41  <bradleymeck>Wes-: what kind of mem protection are you doing to create separate address spaces?
17:38:10  <bradleymeck>i feel like it would have to cause faults to enforce anything, but this is not my area of expertise
17:58:13  * seventhjoined
18:58:58  <trungl-bot>Tree closed by [email protected]: closed (https://ci.chromium.org/buildbot/client.v8/V8%20Win32/14255 from 796cdadd8b6874c6dd1ea1b652f218b3b76f5dcb)
19:01:00  <devsnek>on the latest lkgr
19:01:41  <devsnek>it calls Platform::GetForegroundTaskRunner when doing Isolate::New
19:01:54  <devsnek>but in most cases people key their task runners by isolate in the platform
19:02:09  <devsnek>so there's no task runner available yet because i haven't registered the isolate with the platform
19:04:30  <devsnek>can i just return nullptr or
19:04:34  <devsnek>is there some other magic
19:11:03  <trungl-bot>Tree opened by [email protected]: open
19:11:33  <devsnek>seems to work if i just return nullptr
20:00:29  * seventhquit (Remote host closed the connection)
22:49:00  * RT|Chatzillajoined
23:46:18  * plutoniixjoined