Comment #2 on issue 2087 by [email protected]: Share optimized code
across closure instances
http://code.google.com/p/v8/issues/detail?id=2087
Thanks for the quick feedback. I was clearly confused about deoptimize vs.
not optimize. When I use the --trace_opt option I got this:
10. 150
[marking DrawLine 0x20a34fc for recompilation]
[disabled optimization for: DrawLine / 20a34fd]
[marking MMulti 0x20a32a0 for recompilation]
[disabled optimization for: MMulti / 20a32a1]
[marking DrawQube 0x20a3520 for recompilation]
[disabled optimization for: DrawQube / 20a3521]
[marking CreateP 0x20a343c for recompilation]
[marking Init 0x20a330c for recompilation]
[disabled optimization for: CreateP / 20a343d]
[disabled optimization for: Init / 20a330d]
[marking Loop 0x20a32e8 for recompilation]
[optimizing: Loop / 20a32e9 - took 1.000 ms]
[marking RotateX 0x20a35e4 for recompilation]
[optimizing: RotateX / 20a35e5 - took 1.000 ms]
11. 648
[marking RotateX 0x20dd388 for recompilation]
I took the 'disabled optimization for' to mean deoptimize, which was a
mistake on my part.
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
