Skip to content

device-libs: Replace MATH_SQRT f32 macro with function #880

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
arsenm wants to merge 1 commit into
base: amd-staging
Choose a base branch
from
Open

device-libs: Replace MATH_SQRT f32 macro with function #880

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
28 changes: 1 addition & 27 deletions amd/device-libs/ocml/src/privF.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,30 +21,4 @@

#define MATH_FAST_SQRT(X) BUILTIN_AMDGPU_SQRT_F32(X)

#define MATH_SQRT(X) ({ \
float _sqrt_x = X; \
bool _sqrt_b = _sqrt_x < 0x1.0p-96f; \
_sqrt_x *= _sqrt_b ? 0x1.0p+32f : 1.0f; \
float _sqrt_s; \
if (!DAZ_OPT()) { \
_sqrt_s = BUILTIN_AMDGPU_SQRT_F32(_sqrt_x); \
float _sqrt_sp = AS_FLOAT(AS_INT(_sqrt_s) - 1); \
float _sqrt_ss = AS_FLOAT(AS_INT(_sqrt_s) + 1); \
float _sqrt_vp = BUILTIN_FMA_F32(-_sqrt_sp, _sqrt_s, _sqrt_x); \
float _sqrt_vs = BUILTIN_FMA_F32(-_sqrt_ss, _sqrt_s, _sqrt_x); \
_sqrt_s = _sqrt_vp <= 0.0f ? _sqrt_sp : _sqrt_s; \
_sqrt_s = _sqrt_vs > 0.0f ? _sqrt_ss : _sqrt_s; \
} else { \
float _sqrt_r = BUILTIN_AMDGPU_RSQRT_F32(_sqrt_x); \
_sqrt_s = _sqrt_x * _sqrt_r; \
float _sqrt_h = 0.5f * _sqrt_r; \
float _sqrt_e = BUILTIN_FMA_F32(-_sqrt_h, _sqrt_s, 0.5f); \
_sqrt_h = BUILTIN_FMA_F32(_sqrt_h, _sqrt_e, _sqrt_h); \
_sqrt_s = BUILTIN_FMA_F32(_sqrt_s, _sqrt_e, _sqrt_s); \
float _sqrt_d = BUILTIN_FMA_F32(-_sqrt_s, _sqrt_s, _sqrt_x); \
_sqrt_s = BUILTIN_FMA_F32(_sqrt_d, _sqrt_h, _sqrt_s); \
} \
_sqrt_s *= _sqrt_b ? 0x1.0p-16f : 1.0f; \
_sqrt_s = BUILTIN_CLASS_F32(_sqrt_x, CLASS_PZER|CLASS_NZER|CLASS_PINF) ? _sqrt_x : _sqrt_s; \
_sqrt_s; \
})
#define MATH_SQRT(X) __ocml_sqrt_f32(X)