[Bug target/62275] ARM should use vcvta instructions when possible for float -> int rounding

[ktkachov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62275

--- Comment #3 from ktkachov at gcc dot gnu.org ---
Author: ktkachov
Date: Tue Sep  2 16:00:01 2014
New Revision: 214826

URL: https://gcc.gnu.org/viewcvs?rev=214826&root=gcc&view=rev
Log:
[2/2] Vectorise lroundf, lfloorf, lceilf using the new ARMv8-A vcvt*
instructions.

    PR target/62275
    * config/arm/neon.md
    (neon_vcvt<NEON_VCVT:nvrint_variant><su_optab><VCVTF:mode>
    <v_cmp_result>): New pattern.
    * config/arm/iterators.md (NEON_VCVT): New int iterator.
    * config/arm/arm_neon_builtins.def (vcvtav2sf, vcvtav4sf, vcvtauv2sf,
    vcvtauv4sf, vcvtpv2sf, vcvtpv4sf, vcvtpuv2sf, vcvtpuv4sf, vcvtmv2sf,
    vcvtmv4sf, vcvtmuv2sf, vcvtmuv4sf): New builtin definitions.
    * config/arm/arm.c (arm_builtin_vectorized_function): Handle
    BUILT_IN_LROUNDF, BUILT_IN_LFLOORF, BUILT_IN_LCEILF.

    PR target/62275
    * gcc.target/arm/vect-lceilf_1.c: New test.
    * gcc.target/arm/vect-lfloorf_1.c: Likewise.
    * gcc.target/arm/vect-lroundf_1.c: Likewise.

Added:
    trunk/gcc/testsuite/gcc.target/arm/vect-lceilf_1.c
    trunk/gcc/testsuite/gcc.target/arm/vect-lfloorf_1.c
    trunk/gcc/testsuite/gcc.target/arm/vect-lroundf_1.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/config/arm/arm.c
    trunk/gcc/config/arm/arm_neon_builtins.def
    trunk/gcc/config/arm/iterators.md
    trunk/gcc/config/arm/neon.md
    trunk/gcc/testsuite/ChangeLog