This extends https://github.com/llvm/llvm-project/pull/138577 to more UBSan checks, by changing SanitizerDebugLocation (formerly SanitizerScope) to add annotations if enabled for the specified ordinals. Annotations will use the ordinal name if there is exactly one ordinal specified in the SanitizerDebugLocation; otherwise, it will use the handler name. Updates the tests from https://github.com/llvm/llvm-project/pull/141814. --------- Co-authored-by: Vitaly Buka <vitalybuka@google.com>
The InstrProf headers are very expensive. Avoid including them in all of CodeGen/ by moving the CodeGenPGO member behind a unqiue_ptr. This reduces clang build time by 0.8%.
It was enabled for "aarch64", which covers arm64e but not arm64_32. Co-authored-by: Ahmed Bougacha <ahmed@bougacha.org>
The `Checked` parameter of `CodeGenFunction::EmitCheck` is of type `ArrayRef<std::pair<llvm::Value *, SanitizerMask>>`, which is overly generalized: SanitizerMask can denote that zero or more sanitizers are enabled, but `EmitCheck` requires that exactly one sanitizer is specified in the SanitizerMask (e.g., `SanitizeTrap.has(Checked[i].second)` enforces that). This patch replaces SanitizerMask with SanitizerOrdinal in the `Checked` parameter of `EmitCheck` and code that transitively relies on it. This should not affect the behavior of UBSan, but it has the advantages that: - the code is clearer: it avoids ambiguity in EmitCheck about what to do if multiple bits are set - specifying the wrong number of sanitizers in `Checked[i].second` will be detected as a compile-time error, rather than a runtime assertion failure Suggested by Vitaly in https://github.com/llvm/llvm-project/pull/122392 as an alternative to adding an explicit runtime assertion that the SanitizerMask contains exactly one sanitizer.
Identified with misc-include-cleaner.
Follow up to #109133.
As part of the LLVM effort to eliminate debug-info intrinsics, we're moving to a world where only iterators should be used to insert instructions. This isn't a problem in clang when instructions get generated before any debug-info is inserted, however we're planning on deprecating and removing the instruction-pointer insertion routines. Scatter some calls to getIterator in a few places, remove a deref-then-addrof on another iterator, and add an overload for the createLoadInstBefore utility. Some callers passes a null insertion point, which we need to handle explicitly now.
When accessing data in the buffer, we know we won't overrun the buffer, so we know it is inbounds. In addition, we know that the addition to increase the index is also NUW because the buffer's end has to be unsigned-greater-than 0, which becomes untrue if the bounds ever has an unsigned wrap.
Reverts the above commit, as it updates a common header function and did not update all callsites: https://lab.llvm.org/buildbot/#/builders/29/builds/382 This reverts commit 6481dc57612671ebe77fe9c34214fba94e1b3b27.