Skip to content

device-libs: Use elementwise builtins for ldexp #1374

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: Use elementwise builtins for ldexp #1374

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
24 changes: 20 additions & 4 deletions amd/device-libs/opencl/src/math/wrapbs.cl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -146,10 +146,6 @@ SWRAPST(fmin,float,float)
SWRAPST(fmin,double,double)
PWRAPST(fmin,half,half)

SWRAPST(ldexp,float,int)
SWRAPST(ldexp,double,int)
PWRAPST(ldexp,half,int)

SWRAPST(max,float,float)
SWRAPST(max,double,double)
PWRAPST(max,half,half)
Expand All @@ -166,3 +162,23 @@ SWRAPT(rootn,float,int)
SWRAPT(rootn,double,int)
PWRAPT(rootn,half,int)

#define WRAP_ELEMENTWISE_TYPE(F, T, N, B) \
ATTR T##N F(T##N x, int##N y) { return B(x, y); }

#define WRAP_ELEMENTWISE_SCALAR(F, T, B) \
ATTR T F(T x, int y) { return B(x, y); }

#define WRAP_ELEMENTWISE_VECSIZES(F, T, B) \
WRAP_ELEMENTWISE_TYPE(F, T, 16, B) \
WRAP_ELEMENTWISE_TYPE(F, T, 8, B) \
WRAP_ELEMENTWISE_TYPE(F, T, 4, B) \
WRAP_ELEMENTWISE_TYPE(F, T, 3, B) \
WRAP_ELEMENTWISE_TYPE(F, T, 2, B) \
WRAP_ELEMENTWISE_SCALAR(F, T, B)

#define WRAP_ELEMENTWISE(F, B) \
WRAP_ELEMENTWISE_VECSIZES(F, half, B) \
WRAP_ELEMENTWISE_VECSIZES(F, float, B) \
WRAP_ELEMENTWISE_VECSIZES(F, double, B)

WRAP_ELEMENTWISE(ldexp, __builtin_elementwise_ldexp)
Loading