-
-
Notifications
You must be signed in to change notification settings - Fork 5.5k
New issue
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
minor followups on recent CodeInstance
refactors
#53581
Conversation
f8d7b0f
to
fb28d67
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The metadata argument is mandatory to be a CodeInstance, as it contains some of the required context for uncompressing and populating the CodeInfo correctly and efficiently
base/compiler/typeinfer.jl
Outdated
function transform_result_for_cache(interp::AbstractInterpreter, | ||
linfo::MethodInstance, valid_worlds::WorldRange, result::InferenceResult, | ||
can_discard_trees::Bool=may_discard_trees(interp)) | ||
function transform_result_for_cache(interp::AbstractInterpreter, result::InferenceResult) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is fine to change, but it had seemed inconvenient to cause breakage for external clients which are supposed to overload this method signature.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Well, that does happen frequently. The overload signature of transform_result_for_cache
was changed in #53219, and this PR serves as a follow-up to that change. So I believe making modifications at this PR wouldn't be an issue. We need to add some updates to external abstract interpreter packages anyway.
Is it not feasible to check the identity of |
fb28d67
to
5c8e5d3
Compare
5c8e5d3
to
772fd06
Compare
Having looked over #52415, I've gotten why |
@nanosoldier |
Your benchmark job has completed - no performance regressions were detected. A full report can be found here. |
772fd06
to
fd81aa4
Compare
I'd like to move this forward. @vtjnash can you give another review on this? |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I am on vacation this month, so proceed as you see fit. I will try to remove my "request changes" if any
I dismissed it from my side. Have a great vacation! |
simplifies the signature oftransform_result_for_cache
makejl_uncompress_ir
takeMethodInstance
instead ofCodeInstance
and simplifies the inlining algorithm
codeinst::CodeInstace
objects