|
|
|
@@ -18,7 +18,9 @@ ConvDwFp16Center: |
|
|
|
// https://github.com/ARM-software/abi-aa/blob/master/aapcs64/aapcs64.rst#simd-and-floating-point-registers |
|
|
|
// x19 ~ x29 should be also preserved |
|
|
|
// whereas our coding style do not permit such amount of parameters |
|
|
|
sub sp, sp, #48 |
|
|
|
sub sp, sp, #176 |
|
|
|
st1 {v8.4s, v9.4s, v10.4s, v11.4s}, [sp], #64 |
|
|
|
st1 {v12.4s, v13.4s, v14.4s, v15.4s}, [sp], #64 |
|
|
|
stp x19, x20, [sp], #16 |
|
|
|
stp x21, x22, [sp], #16 |
|
|
|
stp x23, x24, [sp], #16 |
|
|
|
@@ -286,7 +288,9 @@ ConvDwFp16Center: |
|
|
|
subs x4, x4, #1 |
|
|
|
bne LoopH |
|
|
|
|
|
|
|
sub sp, sp, #48 |
|
|
|
sub sp, sp, #176 |
|
|
|
ld1 {v8.4s, v9.4s, v10.4s, v11.4s}, [sp], #64 |
|
|
|
ld1 {v12.4s, v13.4s, v14.4s, v15.4s}, [sp], #64 |
|
|
|
ldp x19, x20, [sp], #16 |
|
|
|
ldp x21, x22, [sp], #16 |
|
|
|
ldp x23, x24, [sp], #16 |
|
|
|
|