Skip to content
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

Add some more fastmath functions #2047

Merged
merged 1 commit into from
Aug 22, 2023
Merged

Add some more fastmath functions #2047

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
4 changes: 4 additions & 0 deletions src/device/intrinsics/math.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -131,6 +131,7 @@ end

@device_override Base.exp2(x::Float64) = ccall("extern __nv_exp2", llvmcall, Cdouble, (Cdouble,), x)
@device_override Base.exp2(x::Float32) = ccall("extern __nv_exp2f", llvmcall, Cfloat, (Cfloat,), x)
@device_override FastMath.exp2_fast(x::Union{Float32, Float64}) = exp2(x)
# TODO: enable once PTX > 7.0 is supported
# @device_override Base.exp2(x::Float16) = @asmcall("ex2.approx.f16 \$0, \$1", "=h,h", Float16, Tuple{Float16}, x)

Expand Down Expand Up @@ -221,6 +222,7 @@ end

@device_override Base.sqrt(x::Float64) = ccall("extern __nv_sqrt", llvmcall, Cdouble, (Cdouble,), x)
@device_override Base.sqrt(x::Float32) = ccall("extern __nv_sqrtf", llvmcall, Cfloat, (Cfloat,), x)
@device_override FastMath.sqrt_fast(x::Union{Float32, Float64}) = sqrt(x)

@device_function rsqrt(x::Float64) = ccall("extern __nv_rsqrt", llvmcall, Cdouble, (Cdouble,), x)
@device_function rsqrt(x::Float32) = ccall("extern __nv_rsqrtf", llvmcall, Cfloat, (Cfloat,), x)
Expand Down Expand Up @@ -306,6 +308,8 @@ end

@device_override FastMath.div_fast(x::Float32, y::Float32) = ccall("extern __nv_fast_fdividef", llvmcall, Cfloat, (Cfloat, Cfloat), x, y)

@device_override Base.inv(x::Float32) = ccall("extern __nv_frcp_rn", llvmcall, Cfloat, (Cfloat,), x)
@device_override FastMath.inv_fast(x::Union{Float32, Float64}) = @fastmath one(x) / x

## distributions

Expand Down