Vector sincosf for x86_64 and tests.

Here is implementation of vectorized sincosf containing SSE, AVX,
AVX2 and AVX512 versions according to Vector ABI
<https://groups.google.com/forum/#!topic/x86-64-abi/LmppCfN1rZ4>.

    * NEWS: Mention addition of x86_64 vector sincosf.
    * math/test-float-vlen16.h: Added wrapper for sincosf tests.
    * math/test-float-vlen4.h: Likewise.
    * math/test-float-vlen8.h: Likewise.
    * sysdeps/unix/sysv/linux/x86_64/libmvec.abilist: New symbols added.
    * sysdeps/x86/fpu/bits/math-vector.h: Added sincosf SIMD declaration.
    * sysdeps/x86_64/fpu/Makefile (libmvec-support): Added new files.
    * sysdeps/x86_64/fpu/Versions: New versions added.
    * sysdeps/x86_64/fpu/libm-test-ulps: Regenerated.
    * sysdeps/x86_64/fpu/multiarch/Makefile (libmvec-sysdep_routines):
    Added build of SSE, AVX2 and AVX512 IFUNC versions.
    * sysdeps/x86_64/fpu/multiarch/svml_s_sincosf16_core.S
    * sysdeps/x86_64/fpu/multiarch/svml_s_sincosf16_core_avx512.S
    * sysdeps/x86_64/fpu/multiarch/svml_s_sincosf4_core.S
    * sysdeps/x86_64/fpu/multiarch/svml_s_sincosf4_core_sse4.S
    * sysdeps/x86_64/fpu/multiarch/svml_s_sincosf8_core.S
    * sysdeps/x86_64/fpu/multiarch/svml_s_sincosf8_core_avx2.S
    * sysdeps/x86_64/fpu/svml_s_sincosf16_core.S
    * sysdeps/x86_64/fpu/svml_s_sincosf4_core.S
    * sysdeps/x86_64/fpu/svml_s_sincosf8_core.S
    * sysdeps/x86_64/fpu/svml_s_sincosf8_core_avx.S
    * sysdeps/x86_64/fpu/svml_s_sincosf_data.S: New file.
    * sysdeps/x86_64/fpu/svml_s_sincosf_data.h: New file.
    * sysdeps/x86_64/fpu/svml_s_wrapper_impl.h: Added 3 argument wrappers.
    * sysdeps/x86_64/fpu/test-float-vlen16.c: : Vector sincosf tests.
    * sysdeps/x86_64/fpu/test-float-vlen16-wrappers.c: Likewise.
    * sysdeps/x86_64/fpu/test-float-vlen4-wrappers.c: Likewise.
    * sysdeps/x86_64/fpu/test-float-vlen4.c: Likewise.
    * sysdeps/x86_64/fpu/test-float-vlen8-avx2-wrappers.c: Likewise.
    * sysdeps/x86_64/fpu/test-float-vlen8-avx2.c: Likewise.
    * sysdeps/x86_64/fpu/test-float-vlen8-wrappers.c: Likewise.
    * sysdeps/x86_64/fpu/test-float-vlen8.c: Likewise.

1 files changed, 154 insertions, 39 deletions

diff --git a/sysdeps/x86_64/fpu/svml_s_wrapper_impl.h b/sysdeps/x86_64/fpu/svml_s_wrapper_impl.h
index f88e30f054..66bb081c9d 100644
--- a/sysdeps/x86_64/fpu/svml_s_wrapper_impl.h
+++ b/sysdeps/x86_64/fpu/svml_s_wrapper_impl.h
@@ -76,6 +76,67 @@
         ret
 .endm
 
+/* 3 argument SSE2 ISA version as wrapper to scalar.  */
+.macro WRAPPER_IMPL_SSE2_fFF callee
+        pushq   %rbp
+        cfi_adjust_cfa_offset (8)
+        cfi_rel_offset (%rbp, 0)
+        pushq   %rbx
+        cfi_adjust_cfa_offset (8)
+        cfi_rel_offset (%rbx, 0)
+        movq    %rdi, %rbp
+        movq    %rsi, %rbx
+        subq    $40, %rsp
+        cfi_adjust_cfa_offset(40)
+        leaq    24(%rsp), %rsi
+        leaq    28(%rsp), %rdi
+        movaps  %xmm0, (%rsp)
+        call    \callee@PLT
+        leaq    24(%rsp), %rsi
+        leaq    28(%rsp), %rdi
+        movss   28(%rsp), %xmm0
+        movss   %xmm0, 0(%rbp)
+        movaps  (%rsp), %xmm1
+        movss   24(%rsp), %xmm0
+        movss   %xmm0, (%rbx)
+        movaps  %xmm1, %xmm0
+        shufps  $85, %xmm1, %xmm0
+        call    \callee@PLT
+        movss   28(%rsp), %xmm0
+        leaq    24(%rsp), %rsi
+        movss   %xmm0, 4(%rbp)
+        leaq    28(%rsp), %rdi
+        movaps  (%rsp), %xmm1
+        movss   24(%rsp), %xmm0
+        movss   %xmm0, 4(%rbx)
+        movaps  %xmm1, %xmm0
+        unpckhps        %xmm1, %xmm0
+        call    \callee@PLT
+        movaps  (%rsp), %xmm1
+        leaq    24(%rsp), %rsi
+        leaq    28(%rsp), %rdi
+        movss   28(%rsp), %xmm0
+        shufps  $255, %xmm1, %xmm1
+        movss   %xmm0, 8(%rbp)
+        movss   24(%rsp), %xmm0
+        movss   %xmm0, 8(%rbx)
+        movaps  %xmm1, %xmm0
+        call    \callee@PLT
+        movss   28(%rsp), %xmm0
+        movss   %xmm0, 12(%rbp)
+        movss   24(%rsp), %xmm0
+        movss   %xmm0, 12(%rbx)
+        addq    $40, %rsp
+        cfi_adjust_cfa_offset(-40)
+        popq    %rbx
+        cfi_adjust_cfa_offset (-8)
+        cfi_restore (%rbx)
+        popq    %rbp
+        cfi_adjust_cfa_offset (-8)
+        cfi_restore (%rbp)
+        ret
+.endm
+
 /* AVX/AVX2 ISA version as wrapper to SSE ISA version.  */
 .macro WRAPPER_IMPL_AVX callee
         pushq     	%rbp
@@ -130,6 +191,52 @@
         ret
 .endm
 
+/* 3 argument AVX/AVX2 ISA version as wrapper to SSE ISA version.  */
+.macro WRAPPER_IMPL_AVX_fFF callee
+        pushq     %rbp
+        cfi_adjust_cfa_offset (8)
+        cfi_rel_offset (%rbp, 0)
+        movq      %rsp, %rbp
+        cfi_def_cfa_register (%rbp)
+        andq      $-32, %rsp
+        pushq     %r13
+        cfi_adjust_cfa_offset (8)
+        cfi_rel_offset (%r13, 0)
+        pushq     %r14
+        cfi_adjust_cfa_offset (8)
+        cfi_rel_offset (%r14, 0)
+        subq      $48, %rsp
+        movq      %rsi, %r14
+        vmovaps   %ymm0, (%rsp)
+        movq      %rdi, %r13
+        vmovaps   16(%rsp), %xmm1
+        vmovaps   %xmm1, 32(%rsp)
+        vzeroupper
+        vmovaps   (%rsp), %xmm0
+        call      HIDDEN_JUMPTARGET(\callee)
+        vmovaps   32(%rsp), %xmm0
+        lea       (%rsp), %rdi
+        lea       16(%rsp), %rsi
+        call      HIDDEN_JUMPTARGET(\callee)
+        vmovaps   (%rsp), %xmm0
+        vmovaps   16(%rsp), %xmm1
+        vmovaps   %xmm0, 16(%r13)
+        vmovaps   %xmm1, 16(%r14)
+        addq      $48, %rsp
+        popq      %r14
+        cfi_adjust_cfa_offset (-8)
+        cfi_restore (%r14)
+        popq      %r13
+        cfi_adjust_cfa_offset (-8)
+        cfi_restore (%r13)
+        movq      %rbp, %rsp
+        cfi_def_cfa_register (%rsp)
+        popq      %rbp
+        cfi_adjust_cfa_offset (-8)
+        cfi_restore (%rbp)
+        ret
+.endm
+
 /* AVX512 ISA version as wrapper to AVX2 ISA version.  */
 .macro WRAPPER_IMPL_AVX512 callee
         pushq	%rbp
@@ -147,20 +254,9 @@
         .byte	0x29
         .byte	0x04
         .byte	0x24
-/* Below is encoding for vmovaps (%rsp), %ymm0.  */
-        .byte	0xc5
-        .byte	0xfc
-        .byte	0x28
-        .byte	0x04
-        .byte	0x24
+        vmovaps (%rsp), %ymm0
         call	HIDDEN_JUMPTARGET(\callee)
-/* Below is encoding for vmovaps 32(%rsp), %ymm0.  */
-        .byte	0xc5
-        .byte	0xfc
-        .byte	0x28
-        .byte	0x44
-        .byte	0x24
-        .byte	0x20
+        vmovaps 32(%rsp), %ymm0
         call	HIDDEN_JUMPTARGET(\callee)
         movq	%rbp, %rsp
         cfi_def_cfa_register (%rsp)
@@ -195,38 +291,57 @@
         .byte	0x29
         .byte	0x4c
         .byte	0x24
-/* Below is encoding for vmovaps (%rsp), %ymm0.  */
-        .byte	0xc5
-        .byte	0xfc
-        .byte	0x28
+        vmovaps (%rsp), %ymm0
+        vmovaps 64(%rsp), %ymm1
+        call      HIDDEN_JUMPTARGET(\callee)
+        vmovaps 32(%rsp), %ymm0
+        vmovaps 96(%rsp), %ymm1
+        call      HIDDEN_JUMPTARGET(\callee)
+        movq      %rbp, %rsp
+        cfi_def_cfa_register (%rsp)
+        popq      %rbp
+        cfi_adjust_cfa_offset (-8)
+        cfi_restore (%rbp)
+        ret
+.endm
+
+/* 3 argument AVX512 ISA version as wrapper to AVX2 ISA version.  */
+.macro WRAPPER_IMPL_AVX512_fFF callee
+        pushq     %rbp
+        cfi_adjust_cfa_offset (8)
+        cfi_rel_offset (%rbp, 0)
+        movq	%rsp, %rbp
+        cfi_def_cfa_register (%rbp)
+        andq      $-64, %rsp
+        pushq     %r12
+        pushq     %r13
+        subq      $176, %rsp
+        movq      %rsi, %r13
+/* Below is encoding for vmovaps %zmm0, (%rsp).  */
+        .byte	0x62
+        .byte	0xf1
+        .byte	0x7c
+        .byte	0x48
+        .byte	0x29
         .byte	0x04
         .byte	0x24
-/* Below is encoding for vmovaps 64(%rsp), %ymm1.  */
-        .byte	0xc5
-        .byte	0xfc
-        .byte	0x28
-        .byte	0x4c
-        .byte	0x24
-        .byte	0x40
+        movq      %rdi, %r12
+        vmovaps   (%rsp), %ymm0
         call      HIDDEN_JUMPTARGET(\callee)
-/* Below is encoding for vmovaps 32(%rsp), %ymm0.  */
-        .byte	0xc5
-        .byte	0xfc
-        .byte	0x28
-        .byte	0x44
-        .byte	0x24
-        .byte	0x20
-/* Below is encoding for vmovaps 96(%rsp), %ymm1.  */
-        .byte	0xc5
-        .byte	0xfc
-        .byte	0x28
-        .byte	0x4c
-        .byte	0x24
-        .byte	0x60
+        vmovaps   32(%rsp), %ymm0
+        lea       64(%rsp), %rdi
+        lea       96(%rsp), %rsi
         call      HIDDEN_JUMPTARGET(\callee)
+        vmovaps   64(%rsp), %ymm0
+        vmovaps   96(%rsp), %ymm1
+        vmovaps   %ymm0, 32(%r12)
+        vmovaps   %ymm1, 32(%r13)
+        addq      $176, %rsp
+        popq      %r13
+        popq      %r12
         movq      %rbp, %rsp
         cfi_def_cfa_register (%rsp)
-        popq      %rbp
+        popq	%rbp
         cfi_adjust_cfa_offset (-8)
         cfi_restore (%rbp)
         ret