Thank you Andrea for reviewing v2 and the great feedback.
Changes since v2:
* Improved potentially flaky if condition in cse_main()
* Made search for tampered register sources more efficient.
Dylan Rees (2):
aarch64: Improve simd highpart builtin folding
middle-end: Eliminate redundant scalar duplication at different vector
widths
gcc/config/aarch64/aarch64-builtins.cc | 16 +-
gcc/cse.cc | 230 +++++++++++++++++-
.../aarch64/simd/fold_to_highpart_7.c | 28 +++
gcc/testsuite/gcc.target/i386/pr81501-9a.c | 4 +-
gcc/testsuite/gcc.target/i386/pr81501-9b.c | 2 +-
5 files changed, 272 insertions(+), 8 deletions(-)
create mode 100644 gcc/testsuite/gcc.target/aarch64/simd/fold_to_highpart_7.c
--
2.43.0
