Open Drvi opened 3 months ago
Not sure if hardcoding
-O0
would be preferable... I'm open to suggestions!
It seems reasonable to me that PackageCompiler should just hardcode -O0
during the first pass over the snoop file, independently from letting the user pass custom julia-args. I don't see any reason to run the snoop file with anything other than -O0
.
I guess some precompilation workloads are not particularly light computationally and could take longer without optimizations... But yeah maybe -O0
should be the default :+1:
All modified and coverable lines are covered by tests :white_check_mark:
Project coverage is 84.62%. Comparing base (
8cd96a1
) to head (80c4211
).:exclamation: Current head 80c4211 differs from pull request most recent head feacfe0. Consider uploading reports for the commit feacfe0 to get more accurate results
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
I guess some precompilation workloads are not particularly light computationally and could take longer without optimizations... But yeah maybe
-O0
should be the default 👍
ah true, good point. 👍
Are the tests on Nightly unrelated?
I checked a couple of nighly failures and there seems to be a bounds error:
BoundsError(a=Array{Base.Partr.taskheap, 1}(dims=(0,), mem=Memory{Base.Partr.taskheap}(0, 0x7f3fe5b3e0b0)[]), i=(0,))
This is a good idea. My hope is that propagating and adding all sorts of these configurations will be easier with, https://github.com/JuliaLang/PackageCompiler.jl/pull/929
Can we split the difference and do -O1
?
So they can run their potentially expensive precompile execution scripts with
-O0
.Not sure if hardcoding
-O0
would be preferable... I'm open to suggestions!