| author | |
| committer | |
| log | 74619cc04586866013e40228f001e3fa79932d8e |
| tree | bc3d7e0a184b1fe53a0bc8f8eb53c7dd468cdbc8 |
| parent | a8b36fbe34e4acfea1fcb348fbed321b05611fd3 |
| signature | Commit is signed but in an unrecognized format. |
3 files changed, 5 insertions(+), 5 deletions(-)
lib/include/__clang_cuda_intrinsics.h+2-2| ... | @@ -45,7 +45,7 @@ | ... | @@ -45,7 +45,7 @@ |
| 45 | _Static_assert(sizeof(__val) == sizeof(__Bits)); \ | 45 | _Static_assert(sizeof(__val) == sizeof(__Bits)); \ |
| 46 | _Static_assert(sizeof(__Bits) == 2 * sizeof(int)); \ | 46 | _Static_assert(sizeof(__Bits) == 2 * sizeof(int)); \ |
| 47 | __Bits __tmp; \ | 47 | __Bits __tmp; \ |
| 48 | memcpy(&__val, &__tmp, sizeof(__val)); \ | 48 | memcpy(&__tmp, &__val, sizeof(__val)); \ |
| 49 | __tmp.__a = ::__FnName(__tmp.__a, __offset, __width); \ | 49 | __tmp.__a = ::__FnName(__tmp.__a, __offset, __width); \ |
| 50 | __tmp.__b = ::__FnName(__tmp.__b, __offset, __width); \ | 50 | __tmp.__b = ::__FnName(__tmp.__b, __offset, __width); \ |
| 51 | long long __ret; \ | 51 | long long __ret; \ |
| ... | @@ -129,7 +129,7 @@ __MAKE_SHUFFLES(__shfl_xor, __nvvm_shfl_bfly_i32, __nvvm_shfl_bfly_f32, 0x1f, | ... | @@ -129,7 +129,7 @@ __MAKE_SHUFFLES(__shfl_xor, __nvvm_shfl_bfly_i32, __nvvm_shfl_bfly_f32, 0x1f, |
| 129 | _Static_assert(sizeof(__val) == sizeof(__Bits)); \ | 129 | _Static_assert(sizeof(__val) == sizeof(__Bits)); \ |
| 130 | _Static_assert(sizeof(__Bits) == 2 * sizeof(int)); \ | 130 | _Static_assert(sizeof(__Bits) == 2 * sizeof(int)); \ |
| 131 | __Bits __tmp; \ | 131 | __Bits __tmp; \ |
| 132 | memcpy(&__val, &__tmp, sizeof(__val)); \ | 132 | memcpy(&__tmp, &__val, sizeof(__val)); \ |
| 133 | __tmp.__a = ::__FnName(__mask, __tmp.__a, __offset, __width); \ | 133 | __tmp.__a = ::__FnName(__mask, __tmp.__a, __offset, __width); \ |
| 134 | __tmp.__b = ::__FnName(__mask, __tmp.__b, __offset, __width); \ | 134 | __tmp.__b = ::__FnName(__mask, __tmp.__b, __offset, __width); \ |
| 135 | long long __ret; \ | 135 | long long __ret; \ |
lib/include/__clang_cuda_runtime_wrapper.h+1-1| ... | @@ -48,7 +48,7 @@ | ... | @@ -48,7 +48,7 @@ |
| 48 | #include "cuda.h" | 48 | #include "cuda.h" |
| 49 | #if !defined(CUDA_VERSION) | 49 | #if !defined(CUDA_VERSION) |
| 50 | #error "cuda.h did not define CUDA_VERSION" | 50 | #error "cuda.h did not define CUDA_VERSION" |
| 51 | #elif CUDA_VERSION < 7000 || CUDA_VERSION > 10010 | 51 | #elif CUDA_VERSION < 7000 |
| 52 | #error "Unsupported CUDA version!" | 52 | #error "Unsupported CUDA version!" |
| 53 | #endif | 53 | #endif |
| 54 | 54 |
lib/include/xmmintrin.h+2-2| ... | @@ -2181,7 +2181,7 @@ void _mm_sfence(void); | ... | @@ -2181,7 +2181,7 @@ void _mm_sfence(void); |
| 2181 | /// 3: Bits [63:48] are copied to the destination. | 2181 | /// 3: Bits [63:48] are copied to the destination. |
| 2182 | /// \returns A 16-bit integer containing the extracted 16 bits of packed data. | 2182 | /// \returns A 16-bit integer containing the extracted 16 bits of packed data. |
| 2183 | #define _mm_extract_pi16(a, n) \ | 2183 | #define _mm_extract_pi16(a, n) \ |
| 2184 | (int)__builtin_ia32_vec_ext_v4hi((__m64)a, (int)n) | 2184 | (int)__builtin_ia32_vec_ext_v4hi((__v4hi)a, (int)n) |
| 2185 | 2185 | ||
| 2186 | /// Copies data from the 64-bit vector of [4 x i16] to the destination, | 2186 | /// Copies data from the 64-bit vector of [4 x i16] to the destination, |
| 2187 | /// and inserts the lower 16-bits of an integer operand at the 16-bit offset | 2187 | /// and inserts the lower 16-bits of an integer operand at the 16-bit offset |
| ... | @@ -2212,7 +2212,7 @@ void _mm_sfence(void); | ... | @@ -2212,7 +2212,7 @@ void _mm_sfence(void); |
| 2212 | /// \returns A 64-bit integer vector containing the copied packed data from the | 2212 | /// \returns A 64-bit integer vector containing the copied packed data from the |
| 2213 | /// operands. | 2213 | /// operands. |
| 2214 | #define _mm_insert_pi16(a, d, n) \ | 2214 | #define _mm_insert_pi16(a, d, n) \ |
| 2215 | (__m64)__builtin_ia32_vec_set_v4hi((__m64)a, (int)d, (int)n) | 2215 | (__m64)__builtin_ia32_vec_set_v4hi((__v4hi)a, (int)d, (int)n) |
| 2216 | 2216 | ||
| 2217 | /// Compares each of the corresponding packed 16-bit integer values of | 2217 | /// Compares each of the corresponding packed 16-bit integer values of |
| 2218 | /// the 64-bit integer vectors, and writes the greater value to the | 2218 | /// the 64-bit integer vectors, and writes the greater value to the |