feat[codegen]: deallocate variables after last use #4219
Conversation
deallocate variables after last use, improving memory usage.
charles-cooper
changed the title
|
marked this for inclusion in 0.4.1, but depending on when our next audit can be, might push to 0.4.2 |
use_count inside of Expr is dangerous, since Expr() can be called multiple times for a given node. separate the analysis.
| var = self.vars[varname] | ||
| for s in self._scopes: | ||
| if s not in var.blockscopes: | ||
| # defer deallocation until we hit the end of its scope |
There was a problem hiding this comment.
is this accurate? we defer the dealoc until we hit its scope again (not necessarily its end)
There was a problem hiding this comment.
or if "it" refers to the scope then still we might have to wait until multiple scopes finish
There was a problem hiding this comment.
"it" refers to the scope of the variable
There was a problem hiding this comment.
but we sweep after statements, right?
a: uint256 = 0
for i: uint256 in range(1, 3):
a += i # <- can't dealoc, defer
b: uint256 = 0 # a already deallocated, swept after the for
b += 1
return b # actual scope end but a is already deallocatedThere was a problem hiding this comment.
yea, this is the idea here
| @@ -40,6 +40,7 @@ class VariableRecord: | |||
| defined_at: Any = None | |||
| is_internal: bool = False | |||
| alloca: Optional[Alloca] = None | |||
| system: bool = False | |||
There was a problem hiding this comment.
can't we rather modify the analysis than create a new type of variable given it's used only at one place?
There was a problem hiding this comment.
yea, it's a bit of a kludge, until we refactor sqrt to be either pure IR or a library function
|
Given that we log write as both write+read, the dealoc won't happen until the last write (although the memory variable might be long dead at that point). So, in this context, the write operation is considered as "use". |
i think a write is a use |
deallocate variables after last use, improving memory usage.
What I did
How I did it
How to verify it
Commit message
Commit message for the final, squashed PR. (Optional, but reviewers will appreciate it! Please see our commit message style guide for what we would ideally like to see in a commit message.)
Description for the changelog
Cute Animal Picture