This is not complete, but shows the direction I'd like to go.
Version 2 extracts more bits from my sve2 branch. There's
still more to pull back, especially for crypto_helper.c, where
there are also tail clearing bugs to fix.
Version 3 rebases on master, which has some of the arm neon
decodetree conversion applied.
r~
Richard Henderson (16):
target/arm: Create gen_gvec_[us]sra
target/arm: Create gen_gvec_{u,s}{rshr,rsra}
target/arm: Create gen_gvec_{sri,sli}
target/arm: Remove unnecessary range check for VSHL
target/arm: Tidy handle_vec_simd_shri
target/arm: Create gen_gvec_{ceq,clt,cle,cgt,cge}0
target/arm: Create gen_gvec_{mla,mls}
target/arm: Swap argument order for VSHL during decode
target/arm: Create gen_gvec_{cmtst,ushl,sshl}
target/arm: Create gen_gvec_{uqadd,sqadd,uqsub,sqsub}
target/arm: Remove fp_status from helper_{recpe,rsqrte}_u32
target/arm: Create gen_gvec_{qrdmla,qrdmls}
target/arm: Pass pointer to qc to qrdmla/qrdmls
target/arm: Clear tail in gvec_fmul_idx_*, gvec_fmla_idx_*
target/arm: Vectorize SABD/UABD
target/arm: Vectorize SABA/UABA
target/arm/helper.h | 71 +-
target/arm/translate.h | 84 +-
target/arm/neon-dp.decode | 9 +-
target/arm/neon_helper.c | 10 -
target/arm/translate-a64.c | 210 +---
target/arm/translate-neon.inc.c | 59 +-
target/arm/translate.c | 1908 ++++++++++++++++++++-----------
target/arm/vec_helper.c | 233 +++-
target/arm/vfp_helper.c | 4 +-
9 files changed, 1667 insertions(+), 921 deletions(-)
--
2.20.1
