Introduce `HostHalf` wrapper class to eliminate explicit
`cl_half_from_float`
and `cl_half_to_float` conversions throughout the test code. The wrapper
provides semantic value constructors/operators and automatic
conversions,
simplifying half-precision arithmetic operations.
Key improvements:
- `HostHalf` class with operator overloading for arithmetic and
comparisons
- Type traits `is_host_atomic_fp_v` and `is_host_fp_v` for generic FP
handling
- Unified floating-point atomic operations (add/sub/min/max/exchange)
- Removed 300+ lines of half-specific conditional branches
- Consistent calculation for all FP types
Related to #2142, according to the work plan extended `CBasicTestStore`
with support for `atomic_half`.
Optimization remark: in tests related to `CBasicTestStore` kernel source
code is mostly composed with arguments following similar pattern:
`__kernel void test_atomic_kernel(uint threadCount, uint numDestItems,
__global int *finalDest, __global int *oldValues, volatile __local
atomic_int *destMemory)`
`oldValues` buffer is initialized with a host pointer, after kernel
execution it is read back to the host pointer but it is unused in
neither of the kernels I verified.
* Add memory_scope_all_devices testing
This duplicats memory_scope_all_svm_devices testing, but it seems pretty quick
so I don't think it hurts.
Fixes#990
* Address clang-format failures
* Address a further clang-format failure
* Change setup code in `KernelCode()` to use `_explicit` builtin
variants that are common to both OpenCL-2.X and OpenCL-3.0.
* Only test optional supported builtin variants (`_explicit` signature
memory_order/scope) for OpenCL-3.0.
* Disable program scope global variable and generic address space tests
for a OpenCL-3.0 driver which does not optionally support these
features.
