Skip to content

S
simde
Commit 484fcce2 authored by jinboson's avatar jinboson Committed by GitHub
Browse files
Options

x86 avx: use INT64_C when the destination is i64 (#1238)

parent 5e225b1c
Hide whitespace changes
Inline Side-by-side
Showing with 14 additions and 14 deletions
x86/avx.h
View file @ 484fcce2
...@@ -2686,7 +2686,7 @@ simde_mm256_cmp_pd ...@@ -2686,7 +2686,7 @@ simde_mm256_cmp_pd
#else #else
SIMDE_VECTORIZE SIMDE_VECTORIZE
for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) { for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) {
r_.i64[i] = (a_.f64[i] == b_.f64[i]) ? ~INT32_C(0) : INT32_C(0); r_.i64[i] = (a_.f64[i] == b_.f64[i]) ? ~INT64_C(0) : INT64_C(0);
} }
#endif #endif
break; break;
...@@ -2698,7 +2698,7 @@ simde_mm256_cmp_pd ...@@ -2698,7 +2698,7 @@ simde_mm256_cmp_pd
#else #else
SIMDE_VECTORIZE SIMDE_VECTORIZE
for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) { for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) {
r_.i64[i] = (a_.f64[i] < b_.f64[i]) ? ~INT32_C(0) : INT32_C(0); r_.i64[i] = (a_.f64[i] < b_.f64[i]) ? ~INT64_C(0) : INT64_C(0);
} }
#endif #endif
break; break;
...@@ -2710,7 +2710,7 @@ simde_mm256_cmp_pd ...@@ -2710,7 +2710,7 @@ simde_mm256_cmp_pd
#else #else
SIMDE_VECTORIZE SIMDE_VECTORIZE
for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) { for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) {
r_.i64[i] = (a_.f64[i] <= b_.f64[i]) ? ~INT32_C(0) : INT32_C(0); r_.i64[i] = (a_.f64[i] <= b_.f64[i]) ? ~INT64_C(0) : INT64_C(0);
} }
#endif #endif
break; break;
...@@ -2722,7 +2722,7 @@ simde_mm256_cmp_pd ...@@ -2722,7 +2722,7 @@ simde_mm256_cmp_pd
#else #else
SIMDE_VECTORIZE SIMDE_VECTORIZE
for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) { for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) {
r_.i64[i] = ((a_.f64[i] != a_.f64[i]) || (b_.f64[i] != b_.f64[i])) ? ~INT32_C(0) : INT32_C(0); r_.i64[i] = ((a_.f64[i] != a_.f64[i]) || (b_.f64[i] != b_.f64[i])) ? ~INT64_C(0) : INT64_C(0);
} }
#endif #endif
break; break;
...@@ -2734,7 +2734,7 @@ simde_mm256_cmp_pd ...@@ -2734,7 +2734,7 @@ simde_mm256_cmp_pd
#else #else
SIMDE_VECTORIZE SIMDE_VECTORIZE
for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) { for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) {
r_.i64[i] = (a_.f64[i] != b_.f64[i]) ? ~INT32_C(0) : INT32_C(0); r_.i64[i] = (a_.f64[i] != b_.f64[i]) ? ~INT64_C(0) : INT64_C(0);
} }
#endif #endif
break; break;
...@@ -2746,7 +2746,7 @@ simde_mm256_cmp_pd ...@@ -2746,7 +2746,7 @@ simde_mm256_cmp_pd
#else #else
SIMDE_VECTORIZE SIMDE_VECTORIZE
for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) { for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) {
r_.i64[i] = ((a_.f64[i] == a_.f64[i]) & (b_.f64[i] == b_.f64[i]) & (a_.f64[i] != b_.f64[i])) ? ~INT32_C(0) : INT32_C(0); r_.i64[i] = ((a_.f64[i] == a_.f64[i]) & (b_.f64[i] == b_.f64[i]) & (a_.f64[i] != b_.f64[i])) ? ~INT64_C(0) : INT64_C(0);
} }
#endif #endif
break; break;
...@@ -2758,7 +2758,7 @@ simde_mm256_cmp_pd ...@@ -2758,7 +2758,7 @@ simde_mm256_cmp_pd
#else #else
SIMDE_VECTORIZE SIMDE_VECTORIZE
for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) { for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) {
r_.i64[i] = !(a_.f64[i] < b_.f64[i]) ? ~INT32_C(0) : INT32_C(0); r_.i64[i] = !(a_.f64[i] < b_.f64[i]) ? ~INT64_C(0) : INT64_C(0);
} }
#endif #endif
break; break;
...@@ -2770,7 +2770,7 @@ simde_mm256_cmp_pd ...@@ -2770,7 +2770,7 @@ simde_mm256_cmp_pd
#else #else
SIMDE_VECTORIZE SIMDE_VECTORIZE
for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) { for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) {
r_.i64[i] = !(a_.f64[i] <= b_.f64[i]) ? ~INT32_C(0) : INT32_C(0); r_.i64[i] = !(a_.f64[i] <= b_.f64[i]) ? ~INT64_C(0) : INT64_C(0);
} }
#endif #endif
break; break;
...@@ -2782,7 +2782,7 @@ simde_mm256_cmp_pd ...@@ -2782,7 +2782,7 @@ simde_mm256_cmp_pd
#else #else
SIMDE_VECTORIZE SIMDE_VECTORIZE
for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) { for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) {
r_.i64[i] = ((a_.f64[i] == a_.f64[i]) & (b_.f64[i] == b_.f64[i])) ? ~INT32_C(0) : INT32_C(0); r_.i64[i] = ((a_.f64[i] == a_.f64[i]) & (b_.f64[i] == b_.f64[i])) ? ~INT64_C(0) : INT64_C(0);
} }
#endif #endif
break; break;
...@@ -2794,7 +2794,7 @@ simde_mm256_cmp_pd ...@@ -2794,7 +2794,7 @@ simde_mm256_cmp_pd
#else #else
SIMDE_VECTORIZE SIMDE_VECTORIZE
for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) { for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) {
r_.i64[i] = ((a_.f64[i] != a_.f64[i]) | (b_.f64[i] != b_.f64[i]) | (a_.f64[i] == b_.f64[i])) ? ~INT32_C(0) : INT32_C(0); r_.i64[i] = ((a_.f64[i] != a_.f64[i]) | (b_.f64[i] != b_.f64[i]) | (a_.f64[i] == b_.f64[i])) ? ~INT64_C(0) : INT64_C(0);
} }
#endif #endif
break; break;
...@@ -2806,7 +2806,7 @@ simde_mm256_cmp_pd ...@@ -2806,7 +2806,7 @@ simde_mm256_cmp_pd
#else #else
SIMDE_VECTORIZE SIMDE_VECTORIZE
for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) { for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) {
r_.i64[i] = !(a_.f64[i] >= b_.f64[i]) ? ~INT32_C(0) : INT32_C(0); r_.i64[i] = !(a_.f64[i] >= b_.f64[i]) ? ~INT64_C(0) : INT64_C(0);
} }
#endif #endif
break; break;
...@@ -2818,7 +2818,7 @@ simde_mm256_cmp_pd ...@@ -2818,7 +2818,7 @@ simde_mm256_cmp_pd
#else #else
SIMDE_VECTORIZE SIMDE_VECTORIZE
for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) { for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) {
r_.i64[i] = !(a_.f64[i] > b_.f64[i]) ? ~INT32_C(0) : INT32_C(0); r_.i64[i] = !(a_.f64[i] > b_.f64[i]) ? ~INT64_C(0) : INT64_C(0);
} }
#endif #endif
break; break;
...@@ -2835,7 +2835,7 @@ simde_mm256_cmp_pd ...@@ -2835,7 +2835,7 @@ simde_mm256_cmp_pd
#else #else
SIMDE_VECTORIZE SIMDE_VECTORIZE
for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) { for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) {
r_.i64[i] = (a_.f64[i] >= b_.f64[i]) ? ~INT32_C(0) : INT32_C(0); r_.i64[i] = (a_.f64[i] >= b_.f64[i]) ? ~INT64_C(0) : INT64_C(0);
} }
#endif #endif
break; break;
...@@ -2847,7 +2847,7 @@ simde_mm256_cmp_pd ...@@ -2847,7 +2847,7 @@ simde_mm256_cmp_pd
#else #else
SIMDE_VECTORIZE SIMDE_VECTORIZE
for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) { for (size_t i = 0 ; i < (sizeof(r_.f64) / sizeof(r_.f64[0])) ; i++) {
r_.i64[i] = (a_.f64[i] > b_.f64[i]) ? ~INT32_C(0) : INT32_C(0); r_.i64[i] = (a_.f64[i] > b_.f64[i]) ? ~INT64_C(0) : INT64_C(0);
} }
#endif #endif
break; break;
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment