←back to thread

The ITTAGE indirect branch predictor

(blog.nelhage.com)
49 points Bogdanp | 1 comments | 04 Jul 25 23:57 UTC | HN request time: 0.228s | source
Show context
nynx ◴[05 Jul 25 00:40 UTC] No.44469206[source]▶
I must be missing something here. How would this help predict interpreter dispatch? Those won’t be a function of previous branch history or pc, which may very well be independent of the next opcode. They’d be a function of state in memory or registers.
replies(4): >>44469293 #>>44469360 #>>44469495 #>>44469524 #
achierius ◴[05 Jul 25 01:00 UTC] No.44469293[source]▶
"very well may be" but oftentimes isn't. Branch history does in practice do a very good job of predicting what target you're going to take for an indirect branch.
replies(1): >>44469354 #
1. nynx ◴[05 Jul 25 01:15 UTC] No.44469354[source]▶
Sure. I can easily see that often being the case for arbitrary code but not interpreter dispatch loops.