diff --git a/flang/module/cudadevice.f90 b/flang/module/cudadevice.f90
index 03860e200cdd8..4f3625a8b07af 100644
--- a/flang/module/cudadevice.f90
+++ b/flang/module/cudadevice.f90
@@ -639,29 +639,29 @@ attributes(device) real(8) function sinpi(x) bind(c,name='__nv_sinpi')
     end function
   end interface
 
-  interface __ll2double_rn
-    attributes(device) double precision function __ll2double_rn(i) bind(c)
+  interface __ll2double_rd
+    attributes(device) double precision function __ll2double_rd(i) bind(c, name='__nv_ll2double_rd')
       !dir$ ignore_tkr (d) i
       integer(8), value :: i
     end function
   end interface
 
-  interface __ll2double_rz
-    attributes(device) double precision function __ll2double_rz(i) bind(c)
+  interface __ll2double_rn
+    attributes(device) double precision function __ll2double_rn(i) bind(c, name='__nv_ll2double_rn')
       !dir$ ignore_tkr (d) i
       integer(8), value :: i
     end function
   end interface
 
   interface __ll2double_ru
-    attributes(device) double precision function __ll2double_ru(i) bind(c)
+    attributes(device) double precision function __ll2double_ru(i) bind(c, name='__nv_ll2double_ru')
       !dir$ ignore_tkr (d) i
       integer(8), value :: i
     end function
   end interface
 
-  interface __ll2double_rd
-    attributes(device) double precision function __ll2double_rd(i) bind(c)
+  interface __ll2double_rz
+    attributes(device) double precision function __ll2double_rz(i) bind(c, name='__nv_ll2double_rz')
       !dir$ ignore_tkr (d) i
       integer(8), value :: i
     end function
diff --git a/flang/test/Lower/CUDA/cuda-device-proc.cuf b/flang/test/Lower/CUDA/cuda-device-proc.cuf
index b146f81030657..ab90dec790046 100644
--- a/flang/test/Lower/CUDA/cuda-device-proc.cuf
+++ b/flang/test/Lower/CUDA/cuda-device-proc.cuf
@@ -90,6 +90,10 @@ attributes(global) subroutine devsub()
   ad = __ull2double_rz(al)
   r2 = __float2half_rn(af)
   af = __half2float(r2)
+  ad = __ll2double_rd(al)
+  ad = __ll2double_rn(al)
+  ad = __ll2double_ru(al)
+  ad = __ll2double_rz(al)
 end
 
 ! CHECK-LABEL: func.func @_QPdevsub() attributes {cuf.proc_attr = #cuf.cuda_proc<global>}
@@ -166,6 +170,10 @@ end
 ! CHECK: %{{.*}} = fir.call @__nv_ull2double_rz(%{{.*}}) proc_attrs<bind_c> fastmath<contract> : (i64) -> f64
 ! CHECK: %{{.*}} = fir.call @__nv_float2half_rn(%{{.*}}) proc_attrs<bind_c> fastmath<contract> : (f32) -> f16
 ! CHECK: %{{.*}} = fir.call @__nv_half2float(%{{.*}}) proc_attrs<bind_c> fastmath<contract> : (f16) -> f32
+! CHECK: %{{.*}} = fir.call @__nv_ll2double_rd(%{{.*}}) proc_attrs<bind_c> fastmath<contract> : (i64) -> f64
+! CHECK: %{{.*}} = fir.call @__nv_ll2double_rn(%{{.*}}) proc_attrs<bind_c> fastmath<contract> : (i64) -> f64
+! CHECK: %{{.*}} = fir.call @__nv_ll2double_ru(%{{.*}}) proc_attrs<bind_c> fastmath<contract> : (i64) -> f64
+! CHECK: %{{.*}} = fir.call @__nv_ll2double_rz(%{{.*}}) proc_attrs<bind_c> fastmath<contract> : (i64) -> f64
 
 subroutine host1()
   integer, device :: a(32)