gh-112075: Use PyMem_* for allocating dict keys objects #114543
Merged
+23
−43
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Outdated suggestions cannot be applied.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.
Use
PyMem_*
for allocating dict keys objectsCurrently dict keys are allocated using
PyObject_Malloc
. This isn't right for free-threaded builds because we don't want non-object allocations to overlap with the object heap. Doing so breaks the guarantees that we can apply an incref to a value which may have been freed and back it out if we lost the reference to the logic. This just switches the allocations to usePyMem_*
instead.This also eliminates some duplication by moving the decref into
dictkeys_decref
when the ref count drops to zero. This is the only place we need to do this, but the logic to move to the free list or free is currently duplicated. This will become a little more complex with QSBR so we may as well isolate it all to one spot.dict
objects thread-safe in--disable-gil
builds #112075