In the IR, some memory accesses are "must" accesses (the entire memory location is always read or written), and some are "may" accesses (some, all, or none of the bits in the location are written). We previously had to special case specific "may" accesses in a few places. This change regularizes our handling of "may" accesses.
The `MemoryAccessKind` enumeration now describes only the extent of the access (the set of locations potentially accessed), but does not distinguish "must" from "may". The new predicates `Operand.hasMayMemoryAccess()` and `Instruction.hasResultMayMemoryAccess()` hold when the access is a "may" access.
Unaliased SSA now correctly ignores variables that are ever accessed via a "may" access.
Aliased SSA now distinguishes `MemoryLocation`s for "may" and "must" accesses. I've refactored `getOverlap()` into the core `getExtentOverlap()`, which considers only the extent, but not the "may" vs. "must", and `getOverlap()`, which tweaks the result of `getExtentOverlap()` based on "may" vs. "must" and read-only locations.
When determining the overlap between a `Phi` operand and its definition, we now use the result of the defining `Chi` instruction, if one exists. This gives exact definitions for `Phi` operands for virtual variables.
The `loopVariant` predicate in `ComparisonWithWiderType.ql` is intended
to identify loop counters, but it was too much of a stretch to apply it
to any subexpression of the small side of the comparison.
This change fixes two false positives on arvidn/libtorrent and many
others seen in the wild (on Linux, CoreCLR, ffmpeg, ...).
This fixes a failing qltest and makes the exclusion similar to what's in
`PointerOverflow.ql`. It's possible we should exclude based on both `+`
and `<`, but we can revisit that if false positives show up.
Before this change, evaluating `cpp/constant-comparison` followed by
`cpp/signed-overflow-check` would result in re-evaluation of almost all
the cached stages they share: CFG, basic blocks, SSA, and range
analysis. The same effect could be seen on `cpp/bad-strncpy-size`, which
also uses the GVN library.
We were privately importing `semmle.code.<lang>.ir.internal.Overlap`, but `PrintSSA.qll` was depending on it being public. This is made a little more complicated by the presence of cross-langage pyrameterized modules.
