Status: Accepted
Owner: ----
CC: [email protected], [email protected]
Labels: Type-Bug Priority-Medium
New issue 3766 by [email protected]: Performance regression between
3.29.88.17 -> 3.31.58
https://code.google.com/p/v8/issues/detail?id=3766
I was looking at the benchmark written by Matt Godbolt (@mattgodbolt) that
he posted at Mozilla bugtracker
https://bug1111000.bugzilla.mozilla.org/attachment.cgi?id=8537200
https://bugzilla.mozilla.org/show_bug.cgi?id=1111000#c12
And noticed that Chrome performance on it regressed.
In Chrome 39 (V8 3.29.88.17) I get 7548ms result
In Chrome 41 (V8 3.31.58) I get 8293ms result.
This is ~9% regression.
FF Nightly is at 79xx ms on my machine.
Note: this benchmark can be made around ~15-20% faster by commenting out
call this.pickBlitters(); at line 251. However M41 remains slightly slower
than M39:
M39: 59xx ms
M41: 62xx ms
(changing amount of frames for which benchmark is run makes this difference
more pronounced)
--
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.
