Build native Windows packages for both RDNA4 targets

Add explicit gfx1200/gfx1201 builds, runtime selection and per-target
code objects. Guard staging, packaging and installation against mixed
architectures; retain gfx1201 compatibility and strict K/M arithmetic.

Adapt realdody's target configuration and MSVC binary16/BOOL portability
fixes from ae530b0. Do not import its forced gfx12 layout on RDNA3.
Add exhaustive binary16 and C++/PowerShell target contract regressions.

Validation: 20 CTest gates on RX 9070 XT; four software-only gfx1200 gates;
MSVC portability compile checks; exact 4K K async and M control outputs;
isolated package install/restore and mixed-target rejection tests.
No physical gfx1200 validation is claimed.

Co-authored-by: realdody <[email protected]>

tools/windows/build_native_texture.py

@@ -1,4 +1,4 @@
-"""Build a private gfx1201 texture override from the user's local DLSS DLL.
+"""Build a private RDNA4 texture override from the user's local DLSS DLL.
 
 The resulting PTX/code object contains NVIDIA code and must remain local.
 No Wine, shell script, GPU, CUDA SDK or system installation is required.
@@ -15,6 +15,14 @@ import subprocess
 import sys
 
 
+def require_code_object_target(path, target):
+    with path.open('rb') as file:
+        header = file.read(64)
+    if (len(header) != 64 or header[:6] != b'\x7fELF\x02\x01' or
+            header[18:20] != b'\xe0\x00' or header[48] != {'gfx1200': 0x48, 'gfx1201': 0x4e}[target]):
+        raise RuntimeError(f'Code object does not target {target}: {path}')
+
+
 def main():
     parser = argparse.ArgumentParser(description=__doc__)
     parser.add_argument('--dlss-dll', type=pathlib.Path, required=True)
@@ -22,6 +30,7 @@ def main():
     parser.add_argument('--zluda-root', type=pathlib.Path, required=True)
     parser.add_argument('--kernel', required=True)
     parser.add_argument('--output-directory', type=pathlib.Path, required=True)
+    parser.add_argument('--gpu-arch', choices=['gfx1200', 'gfx1201'], default='gfx1201')
     parser.add_argument('--shader-mode', choices=['cu', 'wgp'], default='cu')
     args = parser.parse_args()
     if not re.fullmatch(r'hiluma_engine_(?:output_depth(?:inv|reg)_mv(?:hi|lo)_(?:hdr|ldr)(?:_max)?|'
@@ -30,6 +39,8 @@ def main():
     repo = pathlib.Path(__file__).resolve().parents[2]
     output = args.output_directory.resolve()
     output.mkdir(parents=True, exist_ok=True)
+    for existing in output.glob('*.hsaco'):
+        require_code_object_target(existing, args.gpu_arch)
     work = output / ('work-' + args.kernel)
     work.mkdir(exist_ok=True)
     hip = args.hip_root.resolve()
@@ -70,7 +81,7 @@ def main():
         raise RuntimeError('PTX recipe made no changes; refuse a no-op override')
     raw, ir, extra = work / 'raw.bc', work / 'raw.ll', work / 'extra.bc'
     run('compile-helper', [tools['clang++'], '-x', 'hip', '-std=c++20', '-nogpuinc', '-nogpulib', '-O3',
-                          '-mno-wavefrontsize64', '--offload-device-only', '--offload-arch=gfx1201',
+                          '-mno-wavefrontsize64', '--offload-device-only', '--offload-arch=' + args.gpu_arch,
                           '-fgpu-rdc', '-emit-llvm', '-c', '-Xclang', '-fdenormal-fp-math=dynamic',
                           '-DD4R_ACCURACY', '-DD4R_KERNEL_NAME=' + args.kernel,
                           '-o', raw, repo / 'kernels/tex/sust_only.hip'])
@@ -79,12 +90,12 @@ def main():
     for line in ir.read_text().splitlines():
         if re.search(r'@llvm.used|wchar_size|llvm.module.flags|__hip_cuid', line):
             continue
-        line = line.replace('optnone', '').replace('"target-cpu"="gfx1201"', '')
+        line = line.replace('optnone', '').replace(f'"target-cpu"="{args.gpu_arch}"', '')
         lines.append(re.sub(r'"target-features"="[^"]+"', '', line))
     ir.write_text('\n'.join(lines) + '\n')
     run('assemble', [tools['llvm-as'], ir, '-o', extra])
     env['D4R_ZLUDA_EXTRA_BC'] = str(extra)
-    run('emit', [emitter, edited, work / 'emitted', 'gfx1201'])
+    run('emit', [emitter, edited, work / 'emitted', args.gpu_arch])
     optimized = (work / 'emitted/opt.ll').read_text()
     definition = re.search(r'^define amdgpu_kernel .*@' + re.escape(args.kernel) + r'\(.*#(\d+)\s*\{', optimized, re.M)
     if not definition:
@@ -93,9 +104,10 @@ def main():
     expected_mode = '-cumode' if args.shader_mode == 'wgp' else '+cumode'
     if not attributes or expected_mode not in attributes[1]:
         raise RuntimeError(f'Emitter did not honor {args.shader_mode} mode; build the corresponding ZLUDA patch first')
+    require_code_object_target(work / 'emitted/module.hsaco', args.gpu_arch)
     shutil.copyfile(work / 'emitted/module.hsaco', output / (args.kernel + '.hsaco'))
     run('manifest', [sys.executable, repo / 'kernels/tools/kernel_manifest.py', output, args.dlss_dll])
-    metadata = dict(architecture='gfx1201', accuracy=True, nativeFP8=False, shaderMode=args.shader_mode,
+    metadata = dict(architecture=args.gpu_arch, accuracy=True, nativeFP8=False, shaderMode=args.shader_mode,
                     lastBuiltKernel=args.kernel, lastRecipeRewrites=rewrite_counts,
                     privateNvidiaDerivedCode=True, validation='not yet validated; do not install before comparison',
                     dlssSha256=hashlib.sha256(args.dlss_dll.read_bytes()).hexdigest(),
@@ -103,7 +115,7 @@ def main():
                     compiler=subprocess.check_output([str(tools['clang++']), '--version'], env=env).decode(errors='replace'),
                     objects={p.name: hashlib.sha256(p.read_bytes()).hexdigest() for p in output.glob('*.hsaco')})
     (output / 'build-info.json').write_text(json.dumps(metadata, indent=2) + '\n')
-    print(f'Built PRIVATE gfx1201 texture override: {output}; validation required before use')
+    print(f'Built PRIVATE {args.gpu_arch} texture override: {output}; validation required before use')
 
 
 if __name__ == '__main__':