Partial data flow had a semantic merge conflict with this branch. The
problem is that partial data flow doesn't (and shouldn't) cause the
initial pruning steps to run, but the length-2 access paths depend on
the `consCand` information that comes from that initial pruning. The
solution is to restore the old `AccessPath` class, now called
`PartialAccessPath` for use only by partial data flow.
With this change, partial data flow will in some cases allow more field
flow than non-partial data flow.
This commit removes fields from the responsibilities of `FlowVar.qll`.
The treatment of fields in that file was slow and imprecise.
It then adds another copy of the shared global data flow library, used
only to find local field flow, and it exposes that local field flow
through `localFlow` and `localFlowStep`.
This has a performance cost. It adds two cached stages to any query that
uses `localFlow`: the stage from `DataFlowImplCommon`, which is shared
with all queries that use global data flow, and a new stage just for
`localFlowStep`.
Lack of support for the GCC vector extensions was causing a bunch of sanity failures in the syntax zoo. This PR adds minimal IR generation support for these types.
Added `VectorAggregateLiteral`, and factored most of `ArrayAggregateLiteral` out into the common base class `ArrayOrVectorAggregateLiteral`. I'd be happy to merge these all into `ArrayAggregateLiteral` if we don't care about the distinction.
Made a few tweaks to `TranslatedArrayExpr` to compute the element type by looking at the result type of the `ArrayExpr`, not the type of the base operand. Note that this means that for `T a[10]; a[i] = foo;`, the result of the `PointerAdd` for `a[i]` will now be `glvalue<T>`, not `T*`. This is actually more faithful to the source language, and has no semantic difference on the IR.
Added some missing `getInstructionElementSize()` overrides.
Added the new `BuiltIn` opcode, renamed the existing `BuiltInInstruction` to `BuiltInOperationInstruction`, and made any `BuiltInOperation` that we don't specifically handle translate to `BuiltIn`. `BuiltInOperationInstruction` now has a way to get the specific `BuiltInOperation`.
Added `getCanonicalQLClass()` overrides for `GNUVectorType` and `BuiltInOperation`.
Added a simple IR test for vector types.
