refactor pad_reflect_1d to make the UT case pass (#17204)
Co-authored-by: Zhang Jianyu <zhang.jianyu@outlook.com>
This commit is contained in:
co-authored by
Zhang Jianyu
parent
4abef75f2c
commit
efaaccdd69
@@ -617,4 +617,30 @@ static __dpct_inline__ float get_alibi_slope(const float max_bias,
|
||||
return dpct::pow(base, exph);
|
||||
}
|
||||
|
||||
static const sycl::uint3 init_fastdiv_values(uint32_t d) {
|
||||
GGML_ASSERT(d != 0);
|
||||
|
||||
uint32_t L = 0;
|
||||
while (L < 32 && (uint32_t{ 1 } << L) < d) {
|
||||
L++;
|
||||
}
|
||||
|
||||
uint32_t mp = (uint32_t) ((uint64_t{ 1 } << 32) * ((uint64_t{ 1 } << L) - d) / d + 1);
|
||||
return sycl::uint3(mp, L, d);
|
||||
}
|
||||
|
||||
|
||||
static __dpct_inline__ uint32_t fastdiv(uint32_t n, const sycl::uint3 fastdiv_values) {
|
||||
const uint32_t hi = sycl::mul_hi<unsigned>(n, fastdiv_values.x());
|
||||
return (hi + n) >> fastdiv_values.y();
|
||||
}
|
||||
|
||||
|
||||
static __dpct_inline__ sycl::uint2 fast_div_modulo(uint32_t n, const sycl::uint3 fastdiv_values) {
|
||||
const uint32_t div_val = fastdiv(n, fastdiv_values);
|
||||
const uint32_t mod_val = n - div_val * fastdiv_values.z();
|
||||
return sycl::uint2(div_val, mod_val);
|
||||
}
|
||||
|
||||
|
||||
#endif // GGML_SYCL_COMMON_HPP
|
||||
|
||||
Reference in New Issue
Block a user