Use DeBruijn levels in SExpr0. #11987
Merged
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
nickchapman-da
force-pushed
the
nick-db-levels-for-sexpr0
branch
from
76a7b7f to
46c5063
Compare
nickchapman-da
changed the title
nickchapman-da
requested review from
cocreature and
remyhaemmerle-da
as code owners
…des both index and level, and performs a runtime check. changelog_begin changelog_end add runtime check in freeVars: determination that a variable is-free using levels instead of indexes remove DB-indexes and runtime check; simplify freeVars computation in closure-conversion
nickchapman-da
force-pushed
the
nick-db-levels-for-sexpr0
branch
from
d051ead to
1783596
Compare
remyhaemmerle-da
approved these changes
Dec 9, 2021
Comment on lines
97
to
+107
| def decrement1(x: SExpr): SExpr = SEApp(SEBuiltin(SBSubInt64), List(x, SEValue(SInt64(1)))) | ||
| val decrement = SEAbs(1, decrement1(SEVar(1))) | ||
| val decrement = SEAbs(1, decrement1(mkVar(0))) | ||
|
|
||
| def subtract2(x: SExpr, y: SExpr): SExpr = SEApp(SEBuiltin(SBSubInt64), List(x, y)) | ||
| val subtract = SEAbs(2, subtract2(SEVar(2), SEVar(1))) | ||
| val subtract = SEAbs(2, subtract2(mkVar(0), mkVar(1))) | ||
|
|
||
| def twice2(f: SExpr, x: SExpr): SExpr = SEApp(f, List(SEApp(f, List(x)))) | ||
| val twice = SEAbs(2, twice2(SEVar(2), SEVar(1))) | ||
| val twice = SEAbs(2, twice2(mkVar(3), mkVar(4))) | ||
|
|
||
| def thrice2(f: SExpr, x: SExpr): SExpr = SEApp(f, List(SEApp(f, List(SEApp(f, List(x)))))) | ||
| val thrice = SEAbs(2, thrice2(SEVar(2), SEVar(1))) | ||
| val thrice = SEAbs(2, thrice2(mkVar(0), mkVar(1))) |
There was a problem hiding this comment.
Not very important, but I do not really like the use of global index in this class.
How are they calculated ? ...
I assume the level is somehow global, and those expression should be use in a very specific place.
There was a problem hiding this comment.
I agree.
Actually, I think there is very little value in this class. I should delete it.
It's left over form my very first explorations into speedy.
Comment on lines
+28
to
+30
| mapping.get(v) match { | ||
| case Some(loc) => loc | ||
| case None => | ||
| throw sys.error(s"lookup($abs),in:$mapping") | ||
| case None => sys.error(s"lookup($v),in:$mapping") |
There was a problem hiding this comment.
why not using getOrElse ?
realvictorprm
pushed a commit
that referenced
this pull request
Dec 10, 2021
…des both index and level, and performs a runtime check. (#11987) changelog_begin changelog_end add runtime check in freeVars: determination that a variable is-free using levels instead of indexes remove DB-indexes and runtime check; simplify freeVars computation in closure-conversion
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.
Move to using DB-levels (not DB-indexes) in
SExpr0.ClosureConversioncode to computefreeVarsis much simpler.Compiler.scala(During the dev for this change, I ran with both index and level numbers, and ensure behaviour is unchanged.)
This PR is waiting for the fix in: #12047
This PR is part of work to cleanup the speedy compiler: #11669