implicitly-generated AST nodes. We previously built instantiated nodes
for each of these AST nodes, then passed them on to Sema, which was
not prepared to see already-type-checked nodes (see PR5755). In some
places, we had ugly workarounds to try to avoid re-type-checking
(e.g., in VarDecl initializer instantiation).
Now, we skip implicitly-generated nodes when performing instantiation,
preferring instead to build just the AST nodes that directly reflect
what was written in the source code. This has several advantages:
- We don't need to instantiate anything that doesn't have a direct
correlation to the source code, so we can have better location
information.
- Semantic analysis sees the same thing at template instantiation
time that it would see for a non-template.
- At least one ugly hack (VarDecl initializers) goes away.
Fixes PR5755.
llvm-svn: 91218
in a potentially potentially evaluated context, queue those
diagnostics and only emit them if the context ends up being
potentially evaluated. This completes the fix for PR5761.
llvm-svn: 91213
__builtin_offsetof, passing through an ellipsis) when we're in an
unevaluated context. This is the first part of the fix to PR5761,
which deals with the simple case of an unevaluated context.
llvm-svn: 91210
just issues an error for the moment. The front end won't yet generate these
intrinsics for ARM, so this is behind the scenes until complete.
llvm-svn: 91200
This implements a new flag -fcatch-undefined-behavior. The flag turns
on additional runtime checks for:
T a[I];
a[i] abort when i < 0 or i >= I.
Future stuff includes shifts by >= bitwidth amounts.
llvm-svn: 91198
While scanning through the uses of an alloca, keep track of the current offset
relative to the start of the alloca, and check memory references to see if
the offset & size correspond to a component within the alloca. This has the
nice benefit of unifying much of the code from isSafeUseOfAllocation,
isSafeElementUse, and isSafeUseOfBitCastedAllocation. The code to rewrite
the uses of a promoted alloca, after it is determined to be safe, is
reorganized in the same way.
Also, when rewriting GEP instructions, mark them as "in-bounds" since all the
indices are known to be safe.
llvm-svn: 91184
branches only to a landing pad. Without this check, the compiler would go into
an infinite loop because the branch to a landing pad is an "abnormal" edge which
wasn't being taken into account.
This is the meat of that fix:
if (!PrevBB.canFallThrough() && !MBB->BranchesToLandingPad(MBB)) {
The other stuff is simplification of the "branches to a landing pad" code.
llvm-svn: 91161
Evan Cheng