-
Notifications
You must be signed in to change notification settings - Fork 46
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[V8] About TurboFan #3
Comments
Actually TurboFan is getting closer to Crankshaft, but in smaller increments at the moment. We are planning to roll out TurboFan incrementally together with Ignition (the new bytecode interpreter), so we are focusing our efforts on that currently. The configuration shown on AreWeFastYet, which is Fullcodegen+TurboFan will probably never fully materialize in the wild. The long-term plan is to completely remove Fullcodegen and Crankshaft. Also note that TurboFan serves many purposes in Chrome/V8. For example TurboFan is also used as backend for the Ignition interpreter bytecode handlers, and we're also switching most of our code generation for EcmaScript builtins to TurboFan, which gave nice speedups in real world websites already (for example we could noticeably improve sites using frameworks like React or Ember.js, and at the same time we were able to improve Facebook load time, etc.). So the graph on AWFY only shows a very tiny fraction of the overall TurboFan impact. |
@bmeurer thanks for the insight! So if I understand the implications of what you said, Ignition will take the place of the low-latency JIT, right? |
It will take the place of the first tier, yes. |
Also, wasm. |
Acording to AWFY, V8 without Crankshaft it's not getting any closer in terms of performance. Are they planning to move all the optimizations or they act as complementary optimizing compilers?
The text was updated successfully, but these errors were encountered: