Hi,
The attached patch removes calls to builtins in vmul_n* (a, b) with __a * __b.
For float variants, it gates multiplication on __FAST_MATH__.
Since we are not removing all calls to builtins, I am not sure if we
should remove
entry for vmul_n from arm_neon_builtins.def ?
Testing the patch showed fallout for armv8_2-fp16-neon-2.c, because the patch
generates better code.
Code-gen diff:
--- armv8_2-fp16-neon-2.s 2021-05-26 11:34:30.870304900 +0530
+++ armv8_2-fp16-neon-2-after.s 2021-05-26 11:19:13.990304900 +0530
@@ -84,21 +84,9 @@
test_vmul_n_16x4:
- @ args = 0, pretend = 0, frame = 8
+ @ args = 0, pretend = 0, frame = 0
@ frame_needed = 0, uses_anonymous_args = 0
@ link register save eliminated.
- sub sp, sp, #8
- vldr d7, .L8
- add r3, sp, #6
- vst1.16 {d1[0]}, [r3]
- vld1.16 {d7[0]}, [r3]
- vmul.f16 d0, d0, d7[0]
- add sp, sp, #8
- @ sp needed
+ vmov.f16 r3, s2 @ __fp16
+ vdup.16 d16, r3
+ vmul.f16 d0, d16, d0
bx lr
-.L9:
- .align 3
-.L8:
- .short 0
- .short 0
- .short 0
- .short 0
.size test_vmul_n_16x4, .-test_vmul_n_16x4
@@ -113,21 +101,9 @@
test_vmul_n_16x8:
- @ args = 0, pretend = 0, frame = 8
+ @ args = 0, pretend = 0, frame = 0
@ frame_needed = 0, uses_anonymous_args = 0
@ link register save eliminated.
- sub sp, sp, #8
- vldr d7, .L12
- add r3, sp, #6
- vst1.16 {d2[0]}, [r3]
- vld1.16 {d7[0]}, [r3]
- vmul.f16 q0, q0, d7[0]
- add sp, sp, #8
- @ sp needed
+ vmov.f16 r3, s4 @ __fp16
+ vdup.16 q8, r3
+ vmul.f16 q0, q8, q0
bx lr
-.L13:
- .align 3
-.L12:
- .short 0
- .short 0
- .short 0
- .short 0
.size test_vmul_n_16x8, .-test_vmul_n_16x8
Adjusted the test, to fix the failing tests.
OK to commit if testing passes ?
Thanks,
Prathamesh
2021-26-05 Prathamesh Kulkarni <prathamesh.kulka...@linaro.org>
PR target/66791
* config/arm/arm_neon.h (vmul_n_s16): Replace call to builtin with __a
* __b.
(vmul_n_s32): Likewise.
(vmul_n_u16): Likewise.
(vmul_n_u32): Likewise.
(vmulq_n_s16): Likewise.
(vmulq_n_s32): Likewise.
(vmulq_n_u16): Likewise.
(vmulq_n_u32): Likewise.
(vmul_n_f32): Gate __a * __b conditionally on __FAST_MATH__.
(vmulq_n_f32): Likewise.
(vmul_n_f16): Likewise.
(vmulq_n_f16): Likewise.
testsuite/
* gcc.target/arm/armv8_2-fp16-neon-2.c: Adjust.
diff --git a/gcc/config/arm/arm_neon.h b/gcc/config/arm/arm_neon.h
index dcd533fd003..8ac00774e6c 100644
--- a/gcc/config/arm/arm_neon.h
+++ b/gcc/config/arm/arm_neon.h
@@ -8331,70 +8331,78 @@ __extension__ extern __inline int16x4_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
vmul_n_s16 (int16x4_t __a, int16_t __b)
{
- return (int16x4_t)__builtin_neon_vmul_nv4hi (__a, (__builtin_neon_hi) __b);
+ return __a * __b;
}
__extension__ extern __inline int32x2_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
vmul_n_s32 (int32x2_t __a, int32_t __b)
{
- return (int32x2_t)__builtin_neon_vmul_nv2si (__a, (__builtin_neon_si) __b);
+ return __a * __b;
}
__extension__ extern __inline float32x2_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
vmul_n_f32 (float32x2_t __a, float32_t __b)
{
+#ifdef __FAST_MATH__
+ return __a * __b;
+#else
return (float32x2_t)__builtin_neon_vmul_nv2sf (__a, (__builtin_neon_sf) __b);
+#endif
}
__extension__ extern __inline uint16x4_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
vmul_n_u16 (uint16x4_t __a, uint16_t __b)
{
- return (uint16x4_t)__builtin_neon_vmul_nv4hi ((int16x4_t) __a,
(__builtin_neon_hi) __b);
+ return __a * __b;
}
__extension__ extern __inline uint32x2_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
vmul_n_u32 (uint32x2_t __a, uint32_t __b)
{
- return (uint32x2_t)__builtin_neon_vmul_nv2si ((int32x2_t) __a,
(__builtin_neon_si) __b);
+ return __a * __b;
}
__extension__ extern __inline int16x8_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
vmulq_n_s16 (int16x8_t __a, int16_t __b)
{
- return (int16x8_t)__builtin_neon_vmul_nv8hi (__a, (__builtin_neon_hi) __b);
+ return __a * __b;
}
__extension__ extern __inline int32x4_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
vmulq_n_s32 (int32x4_t __a, int32_t __b)
{
- return (int32x4_t)__builtin_neon_vmul_nv4si (__a, (__builtin_neon_si) __b);
+ return __a * __b;
}
__extension__ extern __inline float32x4_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
vmulq_n_f32 (float32x4_t __a, float32_t __b)
{
+#ifdef __FAST_MATH__
+ return __a * __b;
+#else
return (float32x4_t)__builtin_neon_vmul_nv4sf (__a, (__builtin_neon_sf) __b);
+#endif
}
__extension__ extern __inline uint16x8_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
vmulq_n_u16 (uint16x8_t __a, uint16_t __b)
{
- return (uint16x8_t)__builtin_neon_vmul_nv8hi ((int16x8_t) __a,
(__builtin_neon_hi) __b);
+ return __a * __b;
}
__extension__ extern __inline uint32x4_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
vmulq_n_u32 (uint32x4_t __a, uint32_t __b)
{
- return (uint32x4_t)__builtin_neon_vmul_nv4si ((int32x4_t) __a,
(__builtin_neon_si) __b);
+ return __a * __b;
}
__extension__ extern __inline int32x4_t
@@ -17661,7 +17669,11 @@ __extension__ extern __inline float16x4_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
vmul_n_f16 (float16x4_t __a, float16_t __b)
{
+#ifdef __FAST_MATH__
+ return __a * __b;
+#else
return __builtin_neon_vmul_nv4hf (__a, __b);
+#endif
}
__extension__ extern __inline float16x8_t
@@ -17686,7 +17698,11 @@ __extension__ extern __inline float16x8_t
__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
vmulq_n_f16 (float16x8_t __a, float16_t __b)
{
+#ifdef __FAST_MATH__
+ return __a * __b;
+#else
return __builtin_neon_vmul_nv8hf (__a, __b);
+#endif
}
__extension__ extern __inline float16x4_t
diff --git a/gcc/testsuite/gcc.target/arm/armv8_2-fp16-neon-2.c
b/gcc/testsuite/gcc.target/arm/armv8_2-fp16-neon-2.c
index 50f689352ca..2d26bc0ac26 100644
--- a/gcc/testsuite/gcc.target/arm/armv8_2-fp16-neon-2.c
+++ b/gcc/testsuite/gcc.target/arm/armv8_2-fp16-neon-2.c
@@ -327,13 +327,13 @@ BINOP_TEST (vminnm)
BINOP_TEST (vmul)
/* { dg-final { scan-assembler-times {vmul\.f16\td[0-9]+, d[0-9]+, d[0-9]+} 3
} }
- { dg-final { scan-assembler-times {vmul\.f16\tq[0-9]+, q[0-9]+, q[0-9]+} 1
} } */
+ { dg-final { scan-assembler-times {vmul\.f16\tq[0-9]+, q[0-9]+, q[0-9]+} 2
} } */
BINOP_LANE_TEST (vmul, 2)
/* { dg-final { scan-assembler-times {vmul\.f16\td[0-9]+, d[0-9]+,
d[0-9]+\[2\]} 1 } }
{ dg-final { scan-assembler-times {vmul\.f16\tq[0-9]+, q[0-9]+,
d[0-9]+\[2\]} 1 } } */
BINOP_N_TEST (vmul)
-/* { dg-final { scan-assembler-times {vmul\.f16\td[0-9]+, d[0-9]+,
d[0-9]+\[0\]} 1 } }
- { dg-final { scan-assembler-times {vmul\.f16\tq[0-9]+, q[0-9]+,
d[0-9]+\[0\]} 1 } }*/
+/* { dg-final { scan-assembler-times {vmul\.f16\td[0-9]+, d[0-9]+, d[0-9]+} 3
} }
+ { dg-final { scan-assembler-times {vmul\.f16\tq[0-9]+, q[0-9]+, q[0-9]+} 2
} }*/
float16x4_t
test_vpadd_16x4 (float16x4_t a, float16x4_t b)
@@ -387,7 +387,7 @@ test_vdup_n_f16 (float16_t a)
{
return vdup_n_f16 (a);
}
-/* { dg-final { scan-assembler-times {vdup\.16\td[0-9]+, r[0-9]+} 2 } } */
+/* { dg-final { scan-assembler-times {vdup\.16\td[0-9]+, r[0-9]+} 3 } } */
float16x8_t
test_vmovq_n_f16 (float16_t a)
