Base revision:
282c83c32384cb2f37030c28650fef4150a8b67c
Revert “Add BITINT_MAXWIDTH support”
[AArch64] Bail out for float operands in SetCC optimization.
[hwasan] work around lifetime issue with setjmp.
Revert "[DAG] Extend SearchForAndLoads with any_extend
Revert “[ASan] Not linking asan_static library for DSO.”
Revert "DebugInfo: Don't put types in type units if they
Test fixes for prior patch
Revert “[Clang] Propagate guaranteed alignment for malloc and others”
[AArch64][SVE] Remove false register dependency for unary FP
7b03725097872fbd3369a7213c1d98b372aa2d78.patch
[msan] Guard FP_XSTATE_MAGIC1 usage with SANITIZER_GLIBC
f457863ae345d2635026501f5383e0e625869639.patch
[ARM] Patterns for vector conversion between half and float
[CodeView] Avoid integer overflow while parsing long version
[CodeView] Match any backend version in the new test
[tsan] Make __fxstat code path glibc only
[sanitizer] Check if directory exists before trying to create
[ELF] Check COMMON symbols for PROVIDE and don't redefine
[ELF][X86] Don't create IBT .plt if there is no PLT entry
[AArch64] Cleanup and extend cast costs. NFC
[AArch64] Add some tests for the cost of extending an
[AArch64] Add some fp16 conversion cost tests. NFC
[AArch64] Update costs for some fp16 converts
0e96d95d13d9f7b2a96bcaa569ce0a0181a6c7f3.patch
Pass through more LIBCXX_* variables to libfuzzer's custom
[CompilerRT] Fix build of compiler-rt with musl
[AArch64] Move fp16 cast tests.
[AArch64] Improve costs for some conversions to fp16.
Driver: Make macOS the default target OS for -arch arm64
[lldb] Fix interpreting absolute Windows paths with forward
1db59dc8e28819b1960dae8e7fe6d79ad4b03340.patch
[AArch64] Ensure fixed point fptoi_sat has correct saturation
[asan] Always skip first object from dl_iterate_phdr
[BOLT] Compact legacy profiles
Support the min of module flags when linking, use for AArch64
Rename-disable-noundef-analysis-flag-to-no-enable-noundef-analysis.patch
Revert-two-changes-that-break-Android-builds.v7.patch
Add-stubs-and-headers-for-nl_types-APIs.patch
Ensure-that-we-use-our-toolchain-s-lipo-and-not-the-.patch
Undef-FS-macro-for-Android-x86_64.patch
AArch64-Use-Tbl.patch
[AArch64] Add extra fptoint_sat tests for larger than legal
[AArch64] Use simd mov to materialize big fp constants
[AArch64] Some tests to show reconstructing truncates. NFC
[AArch64] Turn truncating buildvectors into truncates
[AArch64] Add tests with free shuffles for indexed fma
[AArch64] Add additional tests for sinking free shuffles for
[AArch64] Add support for FMA intrinsics to
[AArch64] Add fcmp fast math tests
[AArch64] Emit vector FP cmp when LE is used with fast-math
Loop-Vectorizer-shouldMaximizeVectorBandwidth.patch