[CUDA] Added missing functions.

[tra]

Summary

Initial commit missed sincos(float), llabs() and few atomics that we
used to pull in from device_functions.hpp, which we no longer include.

[tra]

Added missing __threadfence_system().

[jlebar]

For my information, how are we verifying that we've caught everything?

[tra]

In D43602#1015370, @jlebar wrote:

For my information, how are we verifying that we've caught everything?

for v in 8.0 9.0 9.1 ;  do 
  /usr/local/cuda-$v/bin/nvcc -c -x cu /dev/null -o /tmp/null.o -arch=sm_60 -keep-dir=nvcc-$v -keep -v
  dump-func-sig nvcc-$v/empty.cpp1.ii -- -x cuda-cpp-output -nocudainc -nocudalib --cuda-host-only -ferror-limit=0 -std=c++11  > nvcc-$v/decls; 
  dump-func-sig /dev/null -- -x cuda --cuda-path=/usr/local/cuda-$v --cuda-host-only -ferror-limit=0 -std=c++11 --cuda-gpu-arch=sm_60 \
      |grep -v typename |grep -v curand | grep -v _Complex | grep -v fetch_builtin > clang-$v; 
done

dump-func-sig is a tool I hacked together which uses clang tooling to parse the files as much as it can and then prints out reconstructed function signature and it's location.

Ten do diff of clang-N and nvcc-N/decls with a lot of regex filtering for argument names, etc. No diff tool does good enough job, so I've missed few functions.

lib/Headers/__clang_cuda_device_functions.h

__DEVICE__ float __ull2float_rn(unsigned long long __a) {
   return __nv_ull2float_rn(__a);
 }
 __DEVICE__ float __ull2float_ru(unsigned long long __a) {
   return __nv_ull2float_ru(__a);
 }
 __DEVICE__ float __ull2float_rz(unsigned long long __a) {
   return __nv_ull2float_rz(__a);
 }
+__DEVICE__ unsigned long long __ullAtomicAdd(unsigned long long *__p,
+                                             unsigned long long __v) {
+  return __nvvm_atom_add_gen_ll((long long *)__p, __v);
+}
 __DEVICE__ unsigned long long __ullAtomicAdd_block(unsigned long long *__p,
                                                    unsigned long long __v) {
   return __nvvm_atom_cta_add_gen_ll((long long *)__p, __v);
 }
 __DEVICE__ unsigned long long __ullAtomicAdd_system(unsigned long long *__p,
                                                     unsigned long long __v) {
   return __nvvm_atom_sys_add_gen_ll((long long *)__p, __v);
 }
 __DEVICE__ unsigned long long __ullAtomicAnd(unsigned long long *__p,
                                              unsigned long long __v) {
   return __nvvm_atom_and_gen_ll((long long *)__p, __v);
 }
 __DEVICE__ unsigned long long __ullAtomicAnd_block(unsigned long long *__p,
                                                    unsigned long long __v) {
   return __nvvm_atom_cta_and_gen_ll((long long *)__p, __v);
 }
 __DEVICE__ unsigned long long __ullAtomicAnd_system(unsigned long long *__p,
                                                     unsigned long long __v) {
   return __nvvm_atom_sys_and_gen_ll((long long *)__p, __v);
 }
+__DEVICE__ unsigned long long __ullAtomicCAS(unsigned long long *__p,
+                                             unsigned long long __cmp,
+                                             unsigned long long __v) {
+  return __nvvm_atom_cas_gen_ll((long long *)__p, __cmp, __v);
+}
 __DEVICE__ unsigned long long __ullAtomicCAS_block(unsigned long long *__p,
                                                    unsigned long long __cmp,
                                                    unsigned long long __v) {
   return __nvvm_atom_cta_cas_gen_ll((long long *)__p, __cmp, __v);
 }
 __DEVICE__ unsigned long long __ullAtomicCAS_system(unsigned long long *__p,
                                                     unsigned long long __cmp,
                                                     unsigned long long __v) {
   return __nvvm_atom_sys_cas_gen_ll((long long *)__p, __cmp, __v);
 }
+__DEVICE__ unsigned long long __ullAtomicExch(unsigned long long *__p,
+                                              unsigned long long __v) {
+  return __nvvm_atom_xchg_gen_ll((long long *)__p, __v);
+}
 __DEVICE__ unsigned long long __ullAtomicExch_block(unsigned long long *__p,
                                                     unsigned long long __v) {
   return __nvvm_atom_cta_xchg_gen_ll((long long *)__p, __v);
 }
 __DEVICE__ unsigned long long __ullAtomicExch_system(unsigned long long *__p,
                                                      unsigned long long __v) {
   return __nvvm_atom_sys_xchg_gen_ll((long long *)__p, __v);
 }

 __DEVICE__ int ilogb(double __a) { return __nv_ilogb(__a); }
 __DEVICE__ int ilogbf(float __a) { return __nv_ilogbf(__a); }
 __DEVICE__ double j0(double __a) { return __nv_j0(__a); }
 __DEVICE__ float j0f(float __a) { return __nv_j0f(__a); }
 __DEVICE__ double j1(double __a) { return __nv_j1(__a); }
 __DEVICE__ float j1f(float __a) { return __nv_j1f(__a); }
 __DEVICE__ double jn(int __n, double __a) { return __nv_jn(__n, __a); }
 __DEVICE__ float jnf(int __n, float __a) { return __nv_jnf(__n, __a); }
+#if defined(__LP64__)
+__DEVICE__ long labs(long __a) { return llabs(__a); };
+#else
+__DEVICE__ long labs(long __a) { return __nv_abs(__a); };
+#endif
 __DEVICE__ double ldexp(double __a, int __b) { return __nv_ldexp(__a, __b); }
 __DEVICE__ float ldexpf(float __a, int __b) { return __nv_ldexpf(__a, __b); }
 __DEVICE__ double lgamma(double __a) { return __nv_lgamma(__a); }
 __DEVICE__ float lgammaf(float __a) { return __nv_lgammaf(__a); }
+__DEVICE__ long long llabs(long long __a) { return __nv_llabs(__a); }
 __DEVICE__ long long llmax(long long __a, long long __b) {
   return __nv_llmax(__a, __b);
 }
 __DEVICE__ long long llmin(long long __a, long long __b) {
   return __nv_llmin(__a, __b);
 }
 __DEVICE__ long long llrint(double __a) { return __nv_llrint(__a); }
 __DEVICE__ long long llrintf(float __a) { return __nv_llrintf(__a); }

 __DEVICE__ float scalblnf(float __a, long __b) {
   if (__b > INT_MAX)
     return __a > 0 ? HUGE_VALF : -HUGE_VALF;
   if (__b < INT_MIN)
     return __a > 0 ? 0.f : -0.f;
   return scalbnf(__a, (int)__b);
 }
 __DEVICE__ double sin(double __a) { return __nv_sin(__a); }
+__DEVICE__ void sincos(double __a, double *__sptr, double *__cptr) {
+  return __nv_sincos(__a, __sptr, __cptr);
+}
 __DEVICE__ void sincosf(float __a, float *__sptr, float *__cptr) {
   return __FAST_OR_SLOW(__nv_fast_sincosf, __nv_sincosf)(__a, __sptr, __cptr);
 }
 __DEVICE__ void sincospi(double __a, double *__sptr, double *__cptr) {
   return __nv_sincospi(__a, __sptr, __cptr);
 }
 __DEVICE__ void sincospif(float __a, float *__sptr, float *__cptr) {
   return __nv_sincospif(__a, __sptr, __cptr);