Comment #6 on issue 4335 by [email protected]: large google maps
performance regression in chrome canary (version 46)
https://code.google.com/p/v8/issues/detail?id=4335
I started to implement the thread-through approach, but then realized it
wouldn't work. We can't set the "requires slow elements" bit without
clearing ICs, because once it's set, any future attempt to set it for
another reason (that *would* require IC clearing) would return early.
So the only short-term fix I can see would be to stop setting that bit for
large indices, un-conflating its meaning to just be "there are elements
with non-standard attributes somewhere in this dictionary". That will
require some adaptations to the can/should convert to slow/fast elements
logic, which seems pretty weird in some cases anyway. I'll look into that
on Monday.
Or we could go for the medium-term solution right away, which is to track
interestingness of prototype chains in maps... but with the team spread
thin due to vacations, that might not make the next branch point.
--
You received this message because this project is configured to send all
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings
--
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
---
You received this message because you are subscribed to the Google Groups "v8-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
