Repository navigation
Miscellaneous bug patches - #832
Open
TysonRayJones wants to merge 27 commits into
Open
TysonRayJones wants to merge 27 commits into
TysonRayJones wants to merge 27 commits into
Conversation
We require a validation failure on one node to reach all other nodes; this requires an OR across nodes. The code previously erroneously performed an AND, requiring the failure occur on all nodes in order to be propagated.
Pretty much pointless since there's no new deployment that switches on only at the absolute maximum Qureg size before overflow, but eh, an aesthetic bug
which erroneously passed (numRows,numRows) instead of (numRows,numCols), to the C API. This meant passing a rectangular (non-square) set of amplitudes either caused a seg-fault, or discarding some of the passed amplitudes.
destroyQureg previously erroneously used the non-NUMA dealloc
which previously did not validate matrix diagonals were real, and ergo returned true for non-Hermitian matrices!
which previously numerically differed from the doc
Avoid imprecise floating-point page counts, support multiword node masks, and skip binding empty page ranges
Previous erroneous logic sometimes overestimated distributed capacity by 1-2 qubits, permitting validation to succeed when insufficient memory was available to allocate distributed structures, like Qureg and SuperOp
which previously referenced a qindex overflow, rather than a size_t Commit also includes yet another ridiculous edge-case protection
which failed to update the full array on each node when the number of local probabilities exceeded max-int (about 32 qubits worth)
Member
Author
|
@lucjaulmes Does this patch of the NUMA allocation look right? |
Member
Author
|
Note to self: I changed the CPTP Kraus-map validation logic, tightening it, which requires loosening our unit test thresholds. This could prove a painful breaking change for users, so doc this carefully! |
This branch has not been deployed
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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.
as discovered by GPT-5.6 Sol, and stylised by moi