[Issue]: Build errors with 6.3.x release

[abagusetty]

Problem Description

Hi,

I was trying to build the tests from 6.3.x branch and faced this error for Nvidia backend

[ 77%] Building CXX object catch_tests/unit/context/CMakeFiles/Context.dir/hipDrvMemcpy.cc.o
cd /home/abagusetty/rocm/6.3.2/hip-tests/build/catch_tests/unit/context && /soft/compilers/rocm/6.3.2/hipcc-install/bin/hipcc -DKERNELS_PATH=\"/home/abagusetty/rocm/6.3.2/hip-tests/catch/kernels/\" -I/home/abagusetty/rocm/6.3.2/hip-tests/catch/external/Catch2 -I/home/abagusetty/rocm/6.3.2/hip-tests/catch/./include -I/home/abagusetty/rocm/6.3.2/hip-tests/catch/./kernels -I/soft/compilers/rocm/6.3.2/hipcc-install/include -I/home/abagusetty/rocm/6.3.2/hip-tests/catch/external/picojson --std=c++17 --extended-lambda -MD -MT catch_tests/unit/context/CMakeFiles/Context.dir/hipDrvMemcpy.cc.o -MF CMakeFiles/Context.dir/hipDrvMemcpy.cc.o.d -o CMakeFiles/Context.dir/hipDrvMemcpy.cc.o -c /home/abagusetty/rocm/6.3.2/hip-tests/catch/unit/context/hipDrvMemcpy.cc
/home/abagusetty/rocm/6.3.2/hip-tests/catch/unit/context/hipDrvMemcpy.cc(58): error: argument of type "hipDeviceptr_t" (aka "unsigned long long") is incompatible with parameter of type "void *"
    { hipError_t localError = hipFree(Ad); if ((localError != hipSuccess) && (localError != hipErrorPeerAccessAlreadyEnabled)) { 
                                      ^

/home/abagusetty/rocm/6.3.2/hip-tests/catch/unit/context/hipDrvMemcpy.cc(59): error: argument of type "hipDeviceptr_t" (aka "unsigned long long") is incompatible with parameter of type "void *"
    { hipError_t localError = hipFree(Bd); if ((localError != hipSuccess) && (localError != hipErrorPeerAccessAlreadyEnabled)) { 
                                      ^

Operating System

SLES 15-SP5

CPU

AMD EPYC 7543 32-Core Processor

GPU

Nvidia A100

ROCm Version

ROCm 6.3.2

ROCm Component

HIP

Steps to Reproduce

No response

(Optional for Linux users) Output of /opt/rocm/bin/rocminfo --support

No response

Additional Information

No response

[darren-amd]

Hi @abagusetty,

I gave this a try on CUDA 12.6 and have been unable to reproduce the issue. Could you please provide the steps on how you are building as well as the full output log? Thanks!