Fix underestimation of array length in constrainAload (0.48) #212
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Outdated suggestions cannot be applied.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.
To this point, VP has constrained the length of an array in
constrainAload
to be between 0 andTR::getMaxSigned<TR::Int32>() / elementSize
elements long. This was likely a due to a historical limitation. Since arrays can be longer thanTR::getMaxSigned<TR::Int32>()
bytes, if such an array were copied, it would fail theArrayCopyBNDCHK
, causing erroneous removal of trees that would prevent the copy from being performed. I discussed my findings during my investigation of this issue in a series of comments in openj9 #19247.This PR increases the high bound of the constraint on the length of an array in
constrainAload
by usingJ9::ObjectModel::maxArraySizeInElements()
instead, which produces a better upper bound based on the size of the heap. We already use this method in other parts of VP, like constrainArraylength.This PR ports omr #7461 to the 0.48 release.
Fixes: openj9 #19247, openj9 #19403, openj9 #15500 (most likely)