aaron/ispc
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Update __load() builtins to take const pointers.

Browse Source
This commit is contained in:
Matt Pharr
2012-07-06 08:47:47 -07:00
parent 0d3993fa25
commit e5fe0eabdc
5 changed files with 23 additions and 23 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
examples/intrinsics/knc.h
View File

@@ -465,8 +465,8 @@ static FORCEINLINE void __insert_element(__vec16_i1 *vec, int index,
}
*/
template <int ALIGN> static FORCEINLINE __vec16_i1 __load(__vec16_i1 *p) {
uint16_t *ptr = (uint16_t *)p;
template <int ALIGN> static FORCEINLINE __vec16_i1 __load(const __vec16_i1 *p) {
const uint16_t *ptr = (const uint16_t *)p;
__vec16_i1 r;
r.m = *ptr;
return r;
@@ -729,14 +729,14 @@ static FORCEINLINE __vec16_i32 __shuffle2_i32(__vec16_i32 v0, __vec16_i32 v1, __
}
*/
template <int ALIGN> static FORCEINLINE __vec16_i32 __load(__vec16_i32 *p) {
template <int ALIGN> static FORCEINLINE __vec16_i32 __load(const __vec16_i32 *p) {
__vec16_i32 v;
v = _mm512_extloadunpackhi_epi32(v, p, _MM_UPCONV_EPI32_NONE, _MM_HINT_NONE);
v = _mm512_extloadunpacklo_epi32(v, p, _MM_UPCONV_EPI32_NONE, _MM_HINT_NONE);
return v;
}
template <> static FORCEINLINE __vec16_i32 __load<64>(__vec16_i32 *p) {
template <> static FORCEINLINE __vec16_i32 __load<64>(const __vec16_i32 *p) {
return _mm512_load_epi32(p);
}
@@ -827,7 +827,7 @@ SHUFFLES(__vec16_i64, i64, int64_t)
LOAD_STORE(__vec16_i64, int64_t)
template <int ALIGN> static FORCEINLINE __vec16_i64 __load(__vec16_i64 *p) {
template <int ALIGN> static FORCEINLINE __vec16_i64 __load(const __vec16_i64 *p) {
__m512i v1;
__m512i v2;
v2 = _mm512_extloadunpackhi_epi32(v1, p, _MM_UPCONV_EPI32_NONE, _MM_HINT_NONE);
@@ -851,7 +851,7 @@ template <int ALIGN> static FORCEINLINE __vec16_i64 __load(__vec16_i64 *p) {
return ret;
}
template <> static FORCEINLINE __vec16_i64 __load<64>(__vec16_i64 *p) {
template <> static FORCEINLINE __vec16_i64 __load<64>(const __vec16_i64 *p) {
__m512i v2 = _mm512_load_epi32(p);
__m512i v1 = _mm512_load_epi32(((uint8_t*)p)+64);
__vec16_i64 ret;
@@ -1015,14 +1015,14 @@ static FORCEINLINE __vec16_f __shuffle2_float(__vec16_f v0, __vec16_f v1, __vec1
}
*/
template <int ALIGN> static FORCEINLINE __vec16_f __load(__vec16_f *p) {
template <int ALIGN> static FORCEINLINE __vec16_f __load(const __vec16_f *p) {
__vec16_f v;
v = _mm512_extloadunpackhi_ps(v, p, _MM_UPCONV_PS_NONE, _MM_HINT_NONE);
v = _mm512_extloadunpacklo_ps(v, p, _MM_UPCONV_PS_NONE, _MM_HINT_NONE);
return v;
}
template <> static FORCEINLINE __vec16_f __load<64>(__vec16_f *p) {
template <> static FORCEINLINE __vec16_f __load<64>(const __vec16_f *p) {
return _mm512_load_ps(p);
}
@@ -1184,7 +1184,7 @@ static FORCEINLINE __vec16_f __shuffle2_float(__vec16_d v0, __vec16_d v1, __vec1
}
*/
template <int ALIGN> static FORCEINLINE __vec16_d __load(__vec16_d *p) {
template <int ALIGN> static FORCEINLINE __vec16_d __load(const __vec16_d *p) {
__vec16_d ret;
ret.v1 = _mm512_extloadunpackhi_pd(ret.v1, p, _MM_UPCONV_PD_NONE, _MM_HINT_NONE);
ret.v1 = _mm512_extloadunpacklo_pd(ret.v1, p, _MM_UPCONV_PD_NONE, _MM_HINT_NONE);
@@ -1193,7 +1193,7 @@ template <int ALIGN> static FORCEINLINE __vec16_d __load(__vec16_d *p) {
return ret;
}
template <> static FORCEINLINE __vec16_d __load<64>(__vec16_d *p) {
template <> static FORCEINLINE __vec16_d __load<64>(const __vec16_d *p) {
__vec16_d ret;
ret.v1 = _mm512_load_pd(p);
ret.v2 = _mm512_load_pd(((uint8_t*)p)+64);