We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
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
As discussed on Zulip, build times with cg_clif appear to be consistently faster than llvm with ld, but slower with lld.
cg_clif
ld
lld
https://github.com/theduke/rustc-cranelift-bench/blob/main/timings.csv
The text was updated successfully, but these errors were encountered:
I can reproduce this. While trying to repro I even found a rustc ICE: rust-lang/rust#83138
Sorry, something went wrong.
Excepts from -Ztime-passes:
-Ztime-passes
cg_llvm:
time: 0.056; rss: 67MB -> 126MB ( +59MB) configure_and_expand time: 0.208; rss: 142MB -> 143MB ( +0MB) link time: 0.360; rss: 51MB -> 141MB ( +90MB) total
cg_clif:
time: 0.201; rss: 51MB -> 276MB ( +224MB) configure_and_expand time: 0.149; rss: 154MB -> 150MB ( -4MB) link time: 0.455; rss: 45MB -> 101MB ( +56MB) total
I think this is #927.
With #927 implemented in b1d14ca, I get:
time: 0.104; rss: 50MB -> 130MB ( +81MB) configure_and_expand time: 0.155; rss: 135MB -> 132MB ( -4MB) link time: 0.379; rss: 44MB -> 94MB ( +51MB) total
According to hyperfine cg_clif is now 8% slower than cg_llvm instead of the ~25% I believe it was before.
I think the difference is small enough now.
No branches or pull requests
As discussed on Zulip, build times with
cg_clif
appear to be consistently faster than llvm withld
, but slower withlld
.https://github.com/theduke/rustc-cranelift-bench/blob/main/timings.csv
The text was updated successfully, but these errors were encountered: