libc++ defines fma as template function for auto promotion of mixed-type
arguments. libc++ does not handle _Float16 as _Float16 is not a supported
type by C++ standard. As such, it is unlikely we can commit our fix for
_Float16 to libc++ trunk.
Therefore we handle _Float16 with a template specialization of
__numeric_type in HIP headers.
Change-Id: If01960a657ebf1a7a67463cdcf66fab7458dff3c
Fixes github issue: #1754
- When ResourceDesc::resType is hipResourceTypeLinear ignore address mode and filter mode.
- When textureDesc::normalizedCoords is set to zero, AddressModeWrap and AddressModeMirror won't be supported and will be switched to AddressModeClamp.
there is a build error when building HIP with latest HCC from GitHub after PR#1935 merged into HIP master branch. this PR changed blockDimX to blockDim and two lines missed this change where added in the current PR.
By default hipcc passes -mllvm options to let HIP-Clang inline all device functions.
--hipcc-func-supp enables function support and disables inline all.
--hipcc-no-func-supp disable function support and enables inline all.
This is a temporary solution to match HCC behavior for performance.
This option is mainly for debugging purpose.
Change-Id: I0c44ac1812bb3cea5c3e5b6e14ebaa45919236f6
* Fix cooperative launch APIs to set hipGetLastError
Previously, the cooperative launch APIs did not properly log their
errors in the global hipGetLastError variable before returning back
to the user. As such, the APIs would leave hipSuccess in the
last error, which would break some use cases.
This fixes that problem by making a trampoline function that does
the HIP_INIT_API and ihipLogStatus.
* Add missing flag to the log of multi-GPU launch
nvcc treats .c program as C program and .cpp program as C++ program.
Currently hipcc treats .c and .cpp programs as HIP programs.
It is desirable to let hipcc behave like nvcc.
Currently it is not feasible to let hipcc treat .cpp programs as C++ program
since there are too many HIP applications use .cpp as extension for HIP programs.
However we should be able let hipcc treat .c program C program since there
are few applications use .c as extensioin for HIP programs.
HIP-Clang cuda_wrapper headers require clang include path before standard C++ include path.
However libc++ include path requires to be before clang include path.
To workaround this, we pass -isystem with the parent directory of clang include
path instead of the clang include path itself.
This PR takes ensures that the maxThreadsPerBlock returned by hipFuncGetAttributes is both a multiple of the warp size and that the register usage of the maximum block does not exceed the number of available registers.
Fixes#1662
GCC emits a warning about using static functions like
hipCUDAErrorTohipError inside this function, because it has an
inline directive, but it's not static. Adding static to this function
to silence warnings (and prevent potential problems in the future).
NVCC warned if you tried to use hipOccupancyMaxActiveBlocksPerMultiprocessor
because when passing in a device function pointer, "const void* func" was
insufficient to describe it accurately. Adding a C++ templated class type
definition for this function.
Fixes SWDEV-207362,
The output file name should not contribute to picking up the right flags for the compiler. This fix solves issues when the output has conflicting extensions which confuses hipcc to treat them as the source files and add the required flags for them.
PS: Output file refers to the file followed by -o
Example: hipcc test.o -o test.hip will add the flags for .hip compilation ignoring the fact that it is an output file
Query ROCr to see if we have the proper lower-level support for
cooperative groups -- GWS support through the firmware, driver,
thunk, and ROCr. ROCr does these checks for us, and presents a
query that allows us to see if GWS entries are available for use.
If so, then we have all the lower-level technologies needed, and
we should enable cooperative groups support for HIP.
Fixes SWDEV-226025,
Right now -x c++ can come before libhip_hcc.so which forces the compiler to treat libhip_hcc.so as a text file and generates a lot of gibberish unicode. This PR changes the order of flags ensuring that -x c++ and similar flags come after libhip_hcc.so
Hopefully, this will not have any negative side effect.
Rahul Garg
ansurya
Yaxun (Sam) Liu
eshcherb
Sarbojit2019
Maneesh Gupta
satyanveshd
Siu Chi Chan
Jatin Chaudhary
Aryan Salmanpour
Nick Curtis
Benjamin Sherman
Evgeny Mankov
Paul Fultz II
Joseph Greathouse
Jeff Daily
nelsonc-amd
Nagy-Egri Máté Ferenc
jglaser
zhaozhangjian
Sameer Sahasrabuddhe
Reshabh Sharma