[torch.compile] Add torch inductor pass for fusing silu_and_mul with subsequent scaled_fp8_quant operations (#10867)
Signed-off-by: Sage Moore <sage@neuralmagic.com>
This commit is contained in:
@@ -7,3 +7,22 @@ inline constexpr uint32_t next_pow_2(uint32_t const num) {
|
||||
if (num <= 1) return num;
|
||||
return 1 << (CHAR_BIT * sizeof(num) - __builtin_clz(num - 1));
|
||||
}
|
||||
|
||||
template <typename A, typename B>
|
||||
static inline constexpr auto div_ceil(A a, B b) {
|
||||
return (a + b - 1) / b;
|
||||
}
|
||||
|
||||
// Round a down to the next multiple of b. The caller is responsible for making
|
||||
// sure that b is non-zero
|
||||
template <typename T>
|
||||
inline constexpr T round_to_previous_multiple_of(T a, T b) {
|
||||
return a % b == 0 ? a : (a / b) * b;
|
||||
}
|
||||
|
||||
// Round a up to the next multiple of b. The caller is responsible for making
|
||||
// sure that b is non-zero
|
||||
template <typename T>
|
||||
inline constexpr T round_to_next_multiple_of(T a, T b) {
|
||||
return a % b == 0 ? a : ((a / b) + 1) * b;
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user