x64: Break more data dependencies in float-related instructions#7818
x64: Break more data dependencies in float-related instructions#7818fitzgen merged 3 commits intobytecodealliance:mainfrom
Conversation
This commit takes a stab at bytecodealliance#7816 without diving a whole lot into it. I noticed that the loop started with `vcvtss2sd` which is along the same lines as previous false dependencies found earlier in PRs such as bytecodealliance#7098. I had forgotten these instructions at the time and meant to go back and touch them up and bytecodealliance#7731 has provided sufficient motivation to do so! Locally this takes that test case from 1.6s to 0.4s for me.
github-actions
bot
added
cranelift
Subscribe to Label ActionDetailsThis issue or pull request has been labeled: "cranelift", "cranelift:area:x64", "winch"Thus the following users have been cc'd because of the following labels:
To subscribe or unsubscribe from this label, edit the |
fitzgen
left a comment
fitzgen
left a comment
There was a problem hiding this comment.
Nice!! Out of curiosity, how did you end up root causing that perf bug to this false dependency?
|
Ah it was mostly from previous experience. I knew there were a set of instructions in the back of my mind which we still did the "fake the output register as the input" for AVX (e.g. the instructions modified here) and when I ran |
2 participants
This commit takes a stab at #7816 without diving a whole lot into it. I noticed that the loop started with
vcvtss2sdwhich is along the same lines as previous false dependencies found earlier in PRs such as #7098. I had forgotten these instructions at the time and meant to go back and touch them up and #7731 has provided sufficient motivation to do so!Locally this takes that test case from 1.6s to 0.4s for me.