You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: posts/faster-jit-plan.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -46,7 +46,7 @@ At the core dev sprint, Brandt nerd-sniped me to rewrite the entire JIT frontend
46
46

47
47
(Image credits to Meta's Free-Threading Benchmarking Runner). Anything below 1.00x on the graph is a slowdown.
48
48
49
-
The details of the impplementation are quite interesting to me, so you might want to give the [PR](https://github.com/python/cpython/pull/140310) a read. The key idea is to maintain two dispatch tables in a mechanism I call "dual dispatch". One table is the standard interpreter, the other is the tracing interpreter. we (ab)use computed gotos/tail calling to dispatch from one table to the other.
49
+
The details of the implementation are quite interesting to me, so you might want to give the [PR](https://github.com/python/cpython/pull/140310) a read. The key idea is to maintain two dispatch tables in a mechanism I call "dual dispatch". One table is the standard interpreter, the other is the tracing interpreter. we (ab)use computed gotos/tail calling to dispatch from one table to the other.
0 commit comments