xdfnx-devandrealdody
98de8e3a3e
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]>
31 lines
2.0 KiB
PowerShell
31 lines
2.0 KiB
PowerShell
[CmdletBinding()]
|
|
param([Parameter(Mandatory=$true)][string]$DlssDll, [string]$PackageRoot, [string]$OutputDirectory, [ValidateSet('gfx1200','gfx1201')][string]$GpuArch)
|
|
$ErrorActionPreference = 'Stop'
|
|
. (Join-Path $PSScriptRoot 'gpu-target.ps1')
|
|
$repo = [IO.Path]::GetFullPath((Join-Path $PSScriptRoot '../..'))
|
|
if (!$PackageRoot) { $PackageRoot = Join-Path $repo 'dist/windows-rdna4-therock' }
|
|
$GpuArch=Get-D4RGpuTarget $PackageRoot $GpuArch
|
|
if (!$OutputDirectory) { $OutputDirectory = Join-Path $repo "build/native-m-$GpuArch" }
|
|
if (![IO.Path]::IsPathRooted($DlssDll) -or !(Test-Path -LiteralPath $DlssDll)) {
|
|
throw 'Supply the absolute path to your local nvngx_dlss.dll.'
|
|
}
|
|
New-Item -ItemType Directory -Force $OutputDirectory | Out-Null
|
|
foreach ($object in Get-ChildItem -LiteralPath $OutputDirectory -Filter '*.hsaco') {
|
|
Assert-D4RCodeObjectTarget $object.FullName $GpuArch
|
|
}
|
|
foreach ($layer in @('enc1','enc2','enc3_tube','dec2','dec1')) {
|
|
$name = "rrlite_${layer}_4x4"
|
|
$source = Join-Path $PackageRoot "experimental/m/${name}_${GpuArch}.hsaco"
|
|
if (!(Test-Path -LiteralPath $source)) { throw "Native M module missing: $source" }
|
|
Assert-D4RCodeObjectTarget $source $GpuArch
|
|
Copy-Item -LiteralPath $source -Destination (Join-Path $OutputDirectory "$name.hsaco") -Force
|
|
}
|
|
& python (Join-Path $repo 'kernels/tools/kernel_manifest.py') $OutputDirectory $DlssDll
|
|
if ($LASTEXITCODE) { throw "Native manifest generation failed ($LASTEXITCODE)" }
|
|
@{architecture=$GpuArch; family='M'; numerics='SWIN_EXACT; SWIN_EXACT_PV; FP16 widening; native FP8 disabled';
|
|
dlssSha256=(Get-FileHash -LiteralPath $DlssDll -Algorithm SHA256).Hash;
|
|
validation='staged only; GPU execution and DLSS capture/reference validation required';
|
|
modules=@(Get-ChildItem -LiteralPath $OutputDirectory -Filter '*.hsaco' | Get-FileHash -Algorithm SHA256)} |
|
|
ConvertTo-Json -Depth 5 | Set-Content -LiteralPath (Join-Path $OutputDirectory 'build-info.json') -Encoding UTF8
|
|
Write-Host "Experimental native M directory: $OutputDirectory"
|