10 Commits

Author	SHA1	Message	Date
Yaxun Sam Liu	11c91c2fdd	Fix ambiguity of __clz and __clzll	2018-08-31 18:07:57 -04:00
Aaron Enye Shi	1f871736e5	Add func attributes to match ocml and ockl	2018-07-10 18:27:32 +00:00
Aaron Enye Shi	d1323f4f9b	Remove activelanemask asm using ockl and llvm instrinsics ... Replace implementation of __any and __all functions using OCKL functions and replaced __ballot implementation to use llvm intrinsic llvm.amdgcn.icmp.i32 instead of calls to __activelanemask_v4_b64_b1 which is not convergent.	2018-07-10 17:56:57 +00:00
Yaxun (Sam) Liu	5ec7973a6f	Add workaround __local_to_generic	2018-07-09 17:56:33 +00:00
Aaron Enye Shi	47d78e372e	Implement min/max functions in HIP header ... Remove using hc::precise_math min and max. Instead we can use ocml directly for device and std:: for host.	2018-07-05 20:15:41 +00:00
Aaron Enye Shi	930a16bccd	Implement Memory Fence Functions in header ... Enabled __llvm_fence_* functions for seq_cst.	2018-07-04 23:35:24 +00:00
Aaron Enye Shi	07de5cb334	Workaround cast warning of smaller integer type for __to_local ... For now, guard the __to_local function for device compile only since a local pointer should be same size as unsigned int on GPU compile. Also change to void* instead of char*.	2018-07-04 20:52:55 +00:00
Aaron Enye Shi	8ac864c2e3	Replace __hip_hc_ir_ inline asm with __ockl_* functions	2018-06-20 20:40:14 +00:00
Aaron Enye Shi	e02fc7e680	Implement device_functions.cpp into HIP headers ... Move all Integer Intrinsics, device_functions.cpp definitions and HIP specific device functions into HIP headers. Implement the device functions using llvm_intrinsics and device-libs functions instead of calling hc::__* functions. Remove device_functions.cpp since everything is now defined in header.	2018-06-20 17:39:23 +00:00
Aaron Enye Shi	cfe37484c9	Split __llvm and device lib funcs into new headers	2018-06-20 17:39:06 +00:00