Skip to content

[flang][cuda] Add bind name for __mul24 and __umul24 #153307

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.

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

Merged
merged 1 commit into from
Aug 13, 2025
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
14 changes: 10 additions & 4 deletions flang/module/cudadevice.f90
Original file line number Diff line number Diff line change
Expand Up @@ -372,6 +372,12 @@ attributes(device) subroutine sincospi(x, y, z) bind(c,name='__nv_sincospi')
procedure :: sincospi
end interface

interface
attributes(device) real(4) function __cosf(x) bind(c, name='__nv_cosf')
real(4), value :: x
end function
end interface

interface
attributes(device) real(4) function cospif(x) bind(c,name='__nv_cospif')
real(4), value :: x
Expand Down Expand Up @@ -431,14 +437,14 @@ attributes(device) real(8) function sinpi(x) bind(c,name='__nv_sinpi')
end interface

interface __float2half_rn
attributes(device) real(2) function __float2half_rn(r) bind(c)
attributes(device) real(2) function __float2half_rn(r) bind(c, name='__nv_float2half_rn')
!dir$ ignore_tkr (d) r
real, value :: r
end function
end interface

interface __half2float
attributes(device) real function __half2float(i) bind(c)
attributes(device) real function __half2float(i) bind(c, name='__nv_half2float')
!dir$ ignore_tkr (d) i
real(2), value :: i
end function
Expand Down Expand Up @@ -676,14 +682,14 @@ attributes(device) real(8) function sinpi(x) bind(c,name='__nv_sinpi')
end interface

interface __mul24
attributes(device) integer function __mul24(i,j) bind(c)
attributes(device) integer function __mul24(i,j) bind(c, name='__nv_mul24')
!dir$ ignore_tkr (d) i, (d) j
integer, value :: i,j
end function
end interface

interface __umul24
attributes(device) integer function __umul24(i,j) bind(c)
attributes(device) integer function __umul24(i,j) bind(c, name='__nv_umul24')
!dir$ ignore_tkr (d) i, (d) j
integer, value :: i,j
end function
Expand Down
7 changes: 7 additions & 0 deletions flang/test/Lower/CUDA/cuda-device-proc.cuf
Original file line number Diff line number Diff line change
Expand Up @@ -58,6 +58,9 @@ attributes(global) subroutine devsub()
res = __ffs(al)
res = __brev(ai)
resl = __brev(al)

ai = __mul24(ai, ai)
ai = __umul24(ai, ai)
end

! CHECK-LABEL: func.func @_QPdevsub() attributes {cuf.proc_attr = #cuf.cuda_proc<global>}
Expand Down Expand Up @@ -105,6 +108,10 @@ end
! CHECK: %{{.*}} = fir.call @__nv_brev(%{{.*}}) proc_attrs<bind_c> fastmath<contract> : (i32) -> i32
! CHECK: %{{.*}} = fir.call @__nv_brevll(%{{.*}}) proc_attrs<bind_c> fastmath<contract> : (i64) -> i64

! CHECK: %{{.*}} = fir.call @__nv_mul24(%{{.*}}, %{{.*}}) proc_attrs<bind_c> fastmath<contract> : (i32, i32) -> i32
! CHECK: %{{.*}} = fir.call @__nv_umul24(%{{.*}}, %{{.*}}) proc_attrs<bind_c> fastmath<contract> : (i32, i32) -> i32


subroutine host1()
integer, device :: a(32)
integer, device :: ret
Expand Down