lattice: Allow external lattice to have a say in how to widen #47307
Conversation
This file contains 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
Keno
force-pushed
the
kf/widenlatticeelem
branch
from
0890df3 to
dd1a215
Compare
vtjnash
reviewed
Oct 24, 2022
Comment on lines
+471
to
473
| typea = widenconditional(typea) | ||
| typeb = widenconditional(typeb) | ||
| return tmerge(widenlattice(lattice), typea, typeb) |
There was a problem hiding this comment.
no lattice? (overall, SGTM, though I only partially follow the intended need here)
There was a problem hiding this comment.
You mean why spell this as widenconditional rather than widenlattice? The lack of widenconditional here was just an oversight in the original PR. It's there in the non-IPO version of this function. It was exposed here, because I stopped doing an unconditional widenconst at the Consts layer. I do think this could be spelled as widenlattice also, but I wanted to minimize the impact for the time being, since I don't actually have an example of a lattice layer inserted at a place where it would make a difference.
aviatesk
reviewed
Oct 25, 2022
Keno
force-pushed
the
kf/widenlatticeelem
branch
from
dd1a215 to
409134d
Compare
aviatesk
approved these changes
Oct 27, 2022
Keno
force-pushed
the
kf/widenlatticeelem
branch
from
409134d to
e58e076
Compare
Currently, when e.g. a PartialStruct is not of interest to the lattice code, it just calls `widenconst` on it to pass it to the next lattice layer. This might be insufficient if an intermediate lattice layer has some other representation that is wider that the `PartialStruct`, but narrower than the corresponding `widenconst`. This adds `widenconst(::AbstractLatice, ::Any)` to allow the lattices to insert custom widening code. By default, it ends up calling `widenconst`, so there's no functional change in base, but custom lattices can make use of the extra hook. I'm not entirely sure that this is what we want the final interface to look like (I think it probably does too many type checks), but it works reasonably well and I think is good enough to experiment with.
Keno
force-pushed
the
kf/widenlatticeelem
branch
from
e58e076 to
cce900c
Compare
|
I think this is ready to go! |
aviatesk
added a commit
that referenced
this pull request
Dec 16, 2022
When merging `PartialStruct`, we are currently merging their fields a bit aggressively (#47307) in order to accelerate convergence. However, when `PartialStruct` wraps external lattice elements, this can be too aggressive since it does not use `tmerge(𝕃, fields...)` recursively and thus the external lattice elements are not merged as expected. This commit adds an additional lattice hook, `tmerge_field`, inside `tmerge(::PartialsLattice)` so that external lattice implementation can provide its own field-merge strategies.
aviatesk
added a commit
that referenced
this pull request
Dec 19, 2022
When merging `PartialStruct`, we are currently merging their fields a bit aggressively (#47307) in order to accelerate convergence. However, when `PartialStruct` wraps external lattice elements, this can be too aggressive since it does not use `tmerge(𝕃, fields...)` recursively and thus the external lattice elements are not merged as expected. This commit adds an additional lattice hook, `tmerge_field`, inside `tmerge(::PartialsLattice)` so that external lattice implementation can provide its own field-merge strategies.
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.
Currently, when e.g. a PartialStruct is not of interest to the lattice code, it just calls
widenconston it to pass it to the next lattice layer. This might be insufficient if an intermediate lattice layer has some other representation that is wider that thePartialStruct, but narrower than the correspondingwidenconst. This addswidenconst(::AbstractLatice, ::Any)to allow the lattices to insert custom widening code. By default, it ends up callingwidenconst, so there's no functional change in base, but custom lattices can make use of the extra hook.I'm not entirely sure that this is what we want the final interface to look like (I think it probably does too many type checks), but it works reasonably well and I think is good enough to experiment with.