- Add API table versioning defines to hip_api_trace.hpp
- Add rocprofiler-register implementation in hip_api_trace.cpp
- created ToolInit implementation
- moved some local functions into anonymous namespace
- renamed some function to be overloads
- made most of the initialization code templated (reduced code duplication)
- enforce ABI
- Updated hipamd/src/CMakeLists.txt
- find rocprofiler-register package (enabled by default but not required)
- Address review comments
- simplify size calculation for dispatch table
- remove setting .size field in GetDispatchTableImpl
- Fix windows build
- Do not enforce ABI on Windows
Change-Id: I08766e8a083528a1236996274bf4522e0e8a1b9f
The following builtins from the CUDA spec are implemented:
- __all_sync, __any_sync, __ballot_sync and __activemask
- __match_any_sync and __match_all_sync
- __shfl_sync, __shfl_up_sync, __shfl_down_sync, and __shfl_xor_sync
The following builtins are NOT implemented, pending support in the compiler:
- __reduce_add_sync, __reduce_min_sync, __reduce_max_sync
- __reduce_and_sync, __reduce_or_sync, __reduce_xor_sync
Change-Id: I07dedbbfe5449f4b5c9b040bed59f5603ccec8c3
It cannot be moved to amd_device_functions.h because that causes circular
dependences when trying to use the macro in other files. So we create a new
header and move all assert/abort macros to that common header.
As a side-effect, also fix the macro to correctly expand the entire condition
argument, and also consume the trailing semicolon.
Change-Id: I43688c8e61183503a3a1a039b91321a3779152af
If the compiler decides not to inline these functions, we might break ODR (one definition rule) due to this file being included in multiple files and being linked together
Change-Id: Iacbfdabb53f5b4e5db8c690b23f3730ec9af16c0
This reverts commit 353dbe6e3b.
Reason for revert: This is considered a breaking change and requires
multiple apps to change their behavior. This will be reintroduced in later releases.
Change-Id: I3481627115af1872785585a155cc6a0ecfbe1372
This reverts commit 629e279f72.
Reason for revert: This is considered a breaking change and requires
multiple apps to change their behavior. This will be reintroduced in later releases.
Change-Id: I0354ce4e0f5e6c402499a7a8c2aaf43bf5b1bfc7
__HIP_CLANG_ONLY__ is not recognized in HIPRTC, due to which some
math functions like amd_mixed_dot were not included in hiprtc builtins.
Change-Id: I1fe41e1ddc8911f6a5b5b1405dd4730d0170a4f7
Defining int64_t, uint64_t, int32_t, uint32_t in HIPRTC
seem to result in conflicts with some apps as they use
their own definitions for these types. NVRTC also doesn't
define these. Hence remove them to match the behavior.
Change-Id: I77ef70e846950698cb00375f5d0501b907f01fe3
__smid() needs to use both HW_ID and XCC_ID for gfx940, gfx941, and
gfx942. Previously, we only did this for gfx940 and thus XCC_ID
was incorrectly not passed back on the other two architectures.
Change-Id: I9fb13b6cef3280e15463443a180174629d03f8b2
Compiler seem to be stricter in compiler staging builds related to
safe buffer programming when compared to other component staging builds.
This seem to result in additional errors when -Werror is enabled
in MIGraphX tests.
Removes all the clang pragmas to ignore several type of warnings in all
the headers and adds a single pragma which ignores all warnings using
#pragma clang diagnostic ignored "-Weverything" in hiprtc builtins.
Change-Id: I95f302bb285b2451b19dd5dfdb7df29164b0f750
Add __host__ and __device__ to bunch of operator/function matching CUDA
Fix some bugs seen in __hisinf
Change-Id: I9e67e3e3eb2083b463158f3e250e5221c89b2896
Remove duplicated operators of hipComplexFloat and
hipComplexDouble.
If users need complex number multiplication and division,
they should call
hipCmulf() and hipCdivf() for hipComplexFloat,
hipCmul() and hipCdiv() for hipComplexComplex
SWDEV-428198 - Add missing operators
Add missing operators of vectors in host
Change-Id: Ie58d1642d579e7119997db49a9fd6a6641b666fd
Remove duplicated operators of hipComplexFloat and
hipComplexDouble.
If users need complex number multiplication and division,
they should call
hipCmulf() and hipCdivf() for hipComplexFloat,
hipCmul() and hipCdiv() for hipComplexComplex
Change-Id: Ie58d1642d578e7118997db40a9fd6a6641b666fd
HIPRTC implementation of few type traits are defined in std
namespace which causes conflicts with the std C++ headers when
apps try to include those std headers. Instead, use the
specific implementation of these headers from __hip_internal
namespace to avoid this.
Change-Id: Ia358df159af8783d1eba9a19cd458e192bf7e9e6
- hipArray will be an internal struct from rocm6.0
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: Icf97fe96b87be8532098cd7f9ceaad099f99c9b9
- alias hipGetDeviceProperties to hipGetDevicePropertiesR0600
- alias hipDeviceProp_t to hipDeviceProp_tR0600
- remove gcnArch from new device property struct
- add new requested struct members
Change-Id: If3f5dbef3d608487d9f6f419285f4bf577ea9bf0
This reverts commit 051e1005f4.
Reason for revert: This causes build failure in hipBLASLt & hipSPARSELt
Change-Id: I59f6aeb8b82261b77eb978e53407bc56be1d0cf1
Support hipExternalMemoryGetMappedMipmappedArray().
Add ImageExternalBuffer to differiate ImageBuffer.
Currently we only support tiling_optimal mode as
vulkan driver doesn't provide tiling information.
Change-Id: I7e3524cdde53e4df9f728894bcebf4bd3f58d4d9
Jaydeep Patel
Jonathan R. Madsen
sdashmiz
Sameer Sahasrabuddhe
jujiang
Rahul Garg
Anusha GodavarthySurya
Ioannis Assiouras
Jatin Chaudhary
Satyanvesh Dittakavi
jiabaxie
Joseph Greathouse
taosang2
Maneesh Gupta
Rakesh Roy