Fix VF2PostLayout with uncoupled qubits in strict_direction=True#14998
Merged
mtreinish merged 1 commit intoQiskit:mainfrom Sep 8, 2025
Merged
Fix VF2PostLayout with uncoupled qubits in strict_direction=True#14998mtreinish merged 1 commit intoQiskit:mainfrom
VF2PostLayout with uncoupled qubits in strict_direction=True#14998mtreinish merged 1 commit intoQiskit:mainfrom
Conversation
jakelishman
added
Changelog: Fixed
Collaborator
|
One or more of the following people are relevant to this code:
|
When `VF2PostLayout` is set in strict mode, we have to include even active but uncoupled qubits in the interaction graph, because of the semantic constraints. We need to limit the number of these uncoupled qubits we'll consider, though, because there's a factorial overhead to handling them. A previous commit[^1] added this skip to `strict_direction=False`, but didn't correctly handle the continuation case where we're within the limit. This commit correctly clears out the uncoupled-qubit metadata that was causing problems in later layout reconstruction. [^1]: b4094dd: Fix calls to `VF2PostLayout` after optimization loop at O3
jakelishman
force-pushed
the
fix-vf2post-uncoupled
branch
from
632525f to
90b2665
Compare
jakelishman
added
the
stable backport potential
Pull Request Test Coverage Report for Build 17548526532Details
💛 - Coveralls |
mtreinish
approved these changes
Sep 8, 2025
Member
mtreinish
left a comment
mtreinish
left a comment
There was a problem hiding this comment.
LGTM, thanks for catching this and the quick fix.
Comment on lines
+163
to
+168
| if len(free_nodes) > 1: | ||
| self.property_set["VF2PostLayout_stop_reason"] = ( | ||
| VF2PostLayoutStopReason.NO_BETTER_SOLUTION_FOUND | ||
| ) | ||
| return | ||
| free_nodes.clear() |
Member
There was a problem hiding this comment.
It took me a sec to remember that I had an if not strict_direction for the free node removal in the vf2_utils code that generated the free_nodes. With that this does the correct thing as described in the comment. But I was initially worried that the case of a single free node would be missing from the interaction graph.
Member
Author
There was a problem hiding this comment.
That logic in build_interaction_graph is why it took me ~30 minutes to actually produce this patch too haha.
I wasn't too worried about changing the code much because it all disappears in #14863 and becomes Rust.
mergify Bot
pushed a commit
that referenced
this pull request
Sep 8, 2025
…14998) When `VF2PostLayout` is set in strict mode, we have to include even active but uncoupled qubits in the interaction graph, because of the semantic constraints. We need to limit the number of these uncoupled qubits we'll consider, though, because there's a factorial overhead to handling them. A previous commit[^1] added this skip to `strict_direction=False`, but didn't correctly handle the continuation case where we're within the limit. This commit correctly clears out the uncoupled-qubit metadata that was causing problems in later layout reconstruction. [^1]: b4094dd: Fix calls to `VF2PostLayout` after optimization loop at O3 (cherry picked from commit 2d73e5d)
github-merge-queue Bot
pushed a commit
that referenced
this pull request
Sep 8, 2025
…14998) (#15001) When `VF2PostLayout` is set in strict mode, we have to include even active but uncoupled qubits in the interaction graph, because of the semantic constraints. We need to limit the number of these uncoupled qubits we'll consider, though, because there's a factorial overhead to handling them. A previous commit[^1] added this skip to `strict_direction=False`, but didn't correctly handle the continuation case where we're within the limit. This commit correctly clears out the uncoupled-qubit metadata that was causing problems in later layout reconstruction. [^1]: b4094dd: Fix calls to `VF2PostLayout` after optimization loop at O3 (cherry picked from commit 2d73e5d) Co-authored-by: Jake Lishman <jake.lishman@ibm.com>
aaryav-3
pushed a commit
to aaryav-3/qiskit
that referenced
this pull request
Sep 28, 2025
…iskit#14998) When `VF2PostLayout` is set in strict mode, we have to include even active but uncoupled qubits in the interaction graph, because of the semantic constraints. We need to limit the number of these uncoupled qubits we'll consider, though, because there's a factorial overhead to handling them. A previous commit[^1] added this skip to `strict_direction=False`, but didn't correctly handle the continuation case where we're within the limit. This commit correctly clears out the uncoupled-qubit metadata that was causing problems in later layout reconstruction. [^1]: b4094dd: Fix calls to `VF2PostLayout` after optimization loop at O3
aaryav-3
pushed a commit
to aaryav-3/qiskit
that referenced
this pull request
Oct 21, 2025
…iskit#14998) When `VF2PostLayout` is set in strict mode, we have to include even active but uncoupled qubits in the interaction graph, because of the semantic constraints. We need to limit the number of these uncoupled qubits we'll consider, though, because there's a factorial overhead to handling them. A previous commit[^1] added this skip to `strict_direction=False`, but didn't correctly handle the continuation case where we're within the limit. This commit correctly clears out the uncoupled-qubit metadata that was causing problems in later layout reconstruction. [^1]: b4094dd: Fix calls to `VF2PostLayout` after optimization loop at O3
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
5 participants
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.
When
VF2PostLayoutis set in strict mode, we have to include even active but uncoupled qubits in the interaction graph, because of the semantic constraints. We need to limit the number of these uncoupled qubits we'll consider, though, because there's a factorial overhead to handling them.A previous commit1 added this skip to
strict_direction=False, but didn't correctly handle the continuation case where we're within the limit. This commitups the limit to 3 uncoupled qubits (a potential 6x overhead) andcorrectly clears out the uncoupled-qubit metadata that was causing problems in later layout reconstruction.Summary
Details and comments
No release note because the bug isn't released.
Fix #14997
Footnotes
b4094dd: Fix calls to
VF2PostLayoutafter optimization loop at O3 ↩