Skip to content

SMT2: fix extractbit with a non-constant index - #9295

Open
tautschnig wants to merge 1 commit into
diffblue:developfrom
tautschnig:kiro/rr-37-extractbit-index-cast
Open

tautschnig wants to merge 1 commit into
diffblue:developfrom
tautschnig:kiro/rr-37-extractbit-index-cast

Conversation

@tautschnig

@tautschnig tautschnig commented Oct 28, 2026 •

Copy link
Copy Markdown
Collaborator

For a non-constant index, extractbit(src, index) is encoded as bit 0
of src shifted right by index. The index was converted to the type of
the source with a typecast_exprt, which converts the value: for a
fixedbv source it scales the index by 2^fraction_bits, for a c_bool
source it maps any non-zero index to 1, and for a source narrower
than the index it truncates the index, so that an out-of-range index
such as 256 for an 8-bit source selects bit 0.

Instead, zero-extend both the source and the index to the wider of the
two widths. Then an out-of-range index shifts out all bits and the
result is false. (The SAT back end leaves the result unconstrained for
out-of-range indices.)

C programs do not produce such expressions, but other front ends do.

Fixes #9367

Found by Jeroen Robben's specification-conformance pipeline (https://reports.robben.io/methodology/).

  • Each commit message has a non-empty body, explaining why the change was made.
  • Methods or procedures I have added are documented, following the guidelines provided in CODING_STANDARD.md.
  • n/a The feature or user visible behaviour I have added or modified has been documented in the User Guide in doc/cprover-manual/
  • Regression or unit tests are included, or existing tests cover the modified code (in this case I have detailed which ones those are in the commit message).
  • n/a My commit message includes data points confirming performance improvements (if claimed).
  • My PR is restricted to a single feature or bugfix.
  • n/a White-space or formatting changes outside the feature-related changed lines are in commits of their own.

Copilot AI balanced review requested due to automatic review settings October 28, 2026 11:06

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota.

An organization owner can choose the repository organization as the billing source for reviews requested by members with a paid Copilot license. Learn about Copilot code review billing. Review the Copilot code review billing settings.

For a non-constant index, extractbit(src, index) is encoded as bit 0
of src shifted right by index. The index was converted to the type of
the source with a typecast_exprt, which converts the value: for a
fixedbv source it scales the index by 2^fraction_bits, for a c_bool
source it maps any non-zero index to 1, and for a source narrower
than the index it truncates the index, so that an out-of-range index
such as 256 for an 8-bit source selects bit 0.

Instead, zero-extend both the source and the index to the wider of the
two widths. Then an out-of-range index shifts out all bits and the
result is false. (The SAT back end leaves the result unconstrained for
out-of-range indices.)

C programs do not produce such expressions, but other front ends do.

Reported by Jeroen Robben (https://reports.robben.io/methodology/).

Fixes diffblue#9367

Co-authored-by: Kiro <kiro-agent@users.noreply.github.com>
@tautschnig
tautschnig force-pushed the kiro/rr-37-extractbit-index-cast branch from 8863651 to f733327 Compare October 28, 2026 11:39

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

SMT2 backend: extractbit with a non-constant index casts the index to the source type

3 participants