[Bug target/43364] Suboptimal code for the use of ARM NEON intrinsic "vset_lane_f32"

[siarhei.siamashka at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43364



Siarhei Siamashka <siarhei.siamashka at gmail dot com> changed:



           What    |Removed                     |Added

----------------------------------------------------------------------------

             Status|NEW                         |RESOLVED

         Resolution|                            |FIXED

      Known to fail|                            |



--- Comment #5 from Siarhei Siamashka <siarhei.siamashka at gmail dot com> 
2012-12-10 02:12:05 UTC ---

This seems to have improved a lot. Thanks for your hard work.



    .cpu cortex-a8

    .eabi_attribute 27, 3

    .eabi_attribute 28, 1

    .fpu neon

    .eabi_attribute 20, 1

    .eabi_attribute 21, 1

    .eabi_attribute 23, 3

    .eabi_attribute 24, 1

    .eabi_attribute 25, 1

    .eabi_attribute 26, 1

    .eabi_attribute 30, 2

    .eabi_attribute 34, 1

    .eabi_attribute 18, 4

    .file    "test.c"

    .text

    .align    2

    .global    neon_add

    .type    neon_add, %function

neon_add:

    @ args = 0, pretend = 0, frame = 0

    @ frame_needed = 0, uses_anonymous_args = 0

    @ link register save eliminated.

    vmov.f32    d16, #0.0  @ v2sf

    vmov    d17, d16  @ v2sf

    vld1.32    {d16[0]}, [r1]

    vld1.32    {d17[0]}, [r2]

    vadd.f32    d16, d16, d17

    vst1.32    {d16[0]}, [r0]

    bx    lr

    .size    neon_add, .-neon_add

    .ident    "GCC: (GNU) 4.8.0 20121209 (experimental)"