[Bug target/113247] RISC-V: Performance bug in SHA256 after enabling RVV vectorization

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113247

--- Comment #11 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Pan Li <panli@gcc.gnu.org>:

https://gcc.gnu.org/g:7be87b7d2e330afd14a7cc028f64d88f80e12f40

commit r14-7245-g7be87b7d2e330afd14a7cc028f64d88f80e12f40
Author: Juzhe-Zhong <juzhe.zhong@rivai.ai>
Date:   Mon Jan 15 20:00:14 2024 +0800

    RISC-V: Fix regression (GCC-14 compare with GCC-13.2) of SHA256 from
coremark-pro

    This patch fixes -70% performance drop from GCC-13.2 to GCC-14 with
-march=rv64gcv in real hardware.

    The root cause is incorrect cost model cause inefficient vectorization
which makes us performance drop significantly.

    So this patch does:

    1. Adjust vector to scalar cost by introducing v to scalar reg move.
    2. Adjust vec_construct cost since we does spend NUNITS instructions to
construct the vector.

    Tested on both RV32/RV64 no regression, Rebase to the trunk and commit it
as it is approved by Robin.

            PR target/113247

    gcc/ChangeLog:

            * config/riscv/riscv-protos.h (struct regmove_vector_cost): Add
vector to scalar regmove.
            * config/riscv/riscv-vector-costs.cc (adjust_stmt_cost): Ditto.
            * config/riscv/riscv.cc (riscv_builtin_vectorization_cost): Adjust
vec_construct cost.

    gcc/testsuite/ChangeLog:

            * gcc.target/riscv/rvv/autovec/vls/reduc-19.c: Adapt test.
            * gcc.target/riscv/rvv/autovec/vls/reduc-20.c: Ditto.
            * gcc.target/riscv/rvv/autovec/vls/reduc-21.c: Ditto.
            * gcc.dg/vect/costmodel/riscv/rvv/pr113247-1.c: New test.
            * gcc.dg/vect/costmodel/riscv/rvv/pr113247-2.c: New test.
            * gcc.dg/vect/costmodel/riscv/rvv/pr113247-3.c: New test.
            * gcc.dg/vect/costmodel/riscv/rvv/pr113247-4.c: New test.