[Bug target/56313] aarch64 backend not using fmls instruction

[ramana at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=56313

Ramana Radhakrishnan ramana at gcc dot gnu.org changed:

   What|Removed |Added

 CC||ramana at gcc dot gnu.org

--- Comment #3 from Ramana Radhakrishnan ramana at gcc dot gnu.org ---
4.9 and trunk both generate. 


foo:
adrpx0, .LANCHOR0
add x0, x0, :lo12:.LANCHOR0
add x2, x0, 8
add x1, x0, 16
ldr d0, [x0]
ldr d2, [x2]
ldr d1, [x1]
fmlsv0.2s, v2.2s, v1.2s
str d0, [x0]
ret

[Bug target/56313] aarch64 backend not using fmls instruction

[rearnsha at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56313

Richard Earnshaw rearnsha at gcc dot gnu.org changed:

   What|Removed |Added

 Status|UNCONFIRMED |RESOLVED
 Resolution|--- |FIXED
   Target Milestone|--- |4.9.0

--- Comment #2 from Richard Earnshaw rearnsha at gcc dot gnu.org ---
Fixed on trunk.  Probably with:

2013-09-16  James Greenhalgh  james.greenha...@arm.com

* config/aarch64/aarch64-simd-builtins.def (fma): New.
* config/aarch64/aarch64-simd.md
(aarch64_mla_eltmode): New.
(aarch64_mla_elt_vswap_width_namemode): Likewise.
(aarch64_mls_eltmode): Likewise.
(aarch64_mls_elt_vswap_width_namemode): Likewise.
(aarch64_fma4_eltmode): Likewise.
(aarch64_fma4_elt_vswap_width_namemode): Likewise.
(aarch64_fma4_elt_to_128v2df): Likewise.
(aarch64_fma4_elt_to_64df): Likewise.
(fnmamode4): Likewise.
(aarch64_fnma4_eltmode): Likewise.
(aarch64_fnma4_elt_vswap_width_namemode): Likewise.
(aarch64_fnma4_elt_to_128v2df): Likewise.
(aarch64_fnma4_elt_to_64df): Likewise.
* config/aarch64/iterators.md (VDQSF): New.
* config/aarch64/arm_neon.h
(vfmassdq_laneq_f32, 64): Convert to C implementation.
(vmlsaq_laneq_fsu16, 32, 64): Likewise.

[Bug target/56313] aarch64 backend not using fmls instruction

[josh.m.conner at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56313



--- Comment #1 from Joshua Conner josh.m.conner at gmail dot com 2013-02-14 
01:39:55 UTC ---

In case it helps, the pattern for aarch64_vmlsmode is written as:



  (set (op0)

(minus (op1)

  (mult (op2)

(op3



Restructuring this to:



  (set (op0)

(fma (neg (op1))

  (op2)

  (op3)))



Allows the combiner to take advantage of the pattern.