Add edge kind for access to non-explicit partitioned bindings #57009
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Our binding partion invalidation code scans the original source for any GlobalRefs that need to be invalidated. However, this is not the only source of access to binding partitions. Various intrinsics (in particular the `*global` ones) also operate on bindings. Since these are not manifest in the source, we instead use the existing `edges` mechanism to give them forward edges. This PR only includes the basic info type, and validation in the replacement case. There's a bit more work to do there, but I'm waiting on #56499 for that part, precompilation in particular.
Keno
force-pushed
the
kf/implicitedges
branch
from
42080a8 to
6448ad8
Compare
aviatesk
added a commit
to JuliaDebug/Cthulhu.jl
that referenced
this pull request
Jan 28, 2025
aviatesk
added a commit
that referenced
this pull request
Jan 28, 2025
`bpart`s there are all known to be `Core.BindingPartition` and never be `nothing`.
aviatesk
added a commit
that referenced
this pull request
Jan 28, 2025
`bpart`s there are all known to be `Core.BindingPartition` and never be `nothing`.
Keno
added a commit
that referenced
this pull request
Jan 31, 2025
This PR adds limited backedge support for Bindings. There are two classes of bindings that get backedges: 1. Cross-module `GlobalRef` bindings (new in this PR) 2. Any globals accesses through intrinsics (i.e. those with forward edges from #57009) This is a time/space trade-off for invalidation. As a result of the first category, invalidating a binding now only needs to scan all the methods defined in the same module as the binding. At the same time, it is anticipated that most binding references are to bindings in the same module, keeping the list of bindings that need explicit (back)edges small.
Keno
added a commit
that referenced
this pull request
Feb 1, 2025
This PR adds limited backedge support for Bindings. There are two classes of bindings that get backedges: 1. Cross-module `GlobalRef` bindings (new in this PR) 2. Any globals accesses through intrinsics (i.e. those with forward edges from #57009) This is a time/space trade-off for invalidation. As a result of the first category, invalidating a binding now only needs to scan all the methods defined in the same module as the binding. At the same time, it is anticipated that most binding references are to bindings in the same module, keeping the list of bindings that need explicit (back)edges small.
Keno
added a commit
that referenced
this pull request
Feb 2, 2025
This PR adds limited backedge support for Bindings. There are two classes of bindings that get backedges: 1. Cross-module `GlobalRef` bindings (new in this PR) 2. Any globals accesses through intrinsics (i.e. those with forward edges from #57009) This is a time/space trade-off for invalidation. As a result of the first category, invalidating a binding now only needs to scan all the methods defined in the same module as the binding. At the same time, it is anticipated that most binding references are to bindings in the same module, keeping the list of bindings that need explicit (back)edges small.
Keno
added a commit
that referenced
this pull request
Feb 2, 2025
This PR adds limited backedge support for Bindings. There are two classes of bindings that get backedges: 1. Cross-module `GlobalRef` bindings (new in this PR) 2. Any globals accesses through intrinsics (i.e. those with forward edges from #57009) This is a time/space trade-off for invalidation. As a result of the first category, invalidating a binding now only needs to scan all the methods defined in the same module as the binding. At the same time, it is anticipated that most binding references are to bindings in the same module, keeping the list of bindings that need explicit (back)edges small.
serenity4
pushed a commit
to serenity4/julia
that referenced
this pull request
May 1, 2025
…ang#57009) Our binding partion invalidation code scans the original source for any GlobalRefs that need to be invalidated. However, this is not the only source of access to binding partitions. Various intrinsics (in particular the `*global` ones) also operate on bindings. Since these are not manifest in the source, we instead use the existing `edges` mechanism to give them forward edges. This PR only includes the basic info type, and validation in the replacement case. There's a bit more work to do there, but I'm waiting on JuliaLang#56499 for that part, precompilation in particular.
serenity4
pushed a commit
to serenity4/julia
that referenced
this pull request
May 1, 2025
This PR adds limited backedge support for Bindings. There are two classes of bindings that get backedges: 1. Cross-module `GlobalRef` bindings (new in this PR) 2. Any globals accesses through intrinsics (i.e. those with forward edges from JuliaLang#57009) This is a time/space trade-off for invalidation. As a result of the first category, invalidating a binding now only needs to scan all the methods defined in the same module as the binding. At the same time, it is anticipated that most binding references are to bindings in the same module, keeping the list of bindings that need explicit (back)edges small.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Our binding partion invalidation code scans the original source for any GlobalRefs that need to be invalidated. However, this is not the only source of access to binding partitions. Various intrinsics (in particular the
*globalones) also operate on bindings. Since these are not manifest in the source, we instead use the existingedgesmechanism to give them forward edges.This PR only includes the basic info type, and validation in the replacement case. There's a bit more work to do there, but I'm waiting on #56499 for that part, precompilation in particular.