feat: add d4r inference engines and Windows parity tooling
Add native K/M/RR inference engines, HIP kernels, replay and validation harnesses, and Windows DLSS setup and runtime support. Include GPU profiling tools, build and packaging scripts, platform integration patches, and implementation documentation.
This commit is contained in:
ayo committed
1 parent
25493b3198
commit
7c5d3c9157
430 files changed
+76191
No files matched your search
@@ -0,0 +1,158 @@
|
||||
[CmdletBinding()]
|
||||
param(
|
||||
[string]$PackageRoot, [string]$DiagnosticRoot, [string]$OptiScalerRoot,
|
||||
[string]$ZludaRoot, [string]$HipRoot, [string]$KernelKRoot, [string]$KernelMRoot,
|
||||
[string]$GpuArch,
|
||||
[string]$ArchivePath
|
||||
)
|
||||
$ErrorActionPreference = 'Stop'
|
||||
. (Join-Path $PSScriptRoot 'gpu-target.ps1')
|
||||
$repo = [IO.Path]::GetFullPath((Join-Path $PSScriptRoot '../..'))
|
||||
$defaults = @{PackageRoot='dist/windows-rdna4-game'; DiagnosticRoot='dist/windows-rdna4-command-list';
|
||||
OptiScalerRoot='dist/optiscaler-windows-d4r'; ZludaRoot='dist/zluda-windows-final';
|
||||
HipRoot='.tools/therock-10.2.0a20260929/_rocm_sdk_core'}
|
||||
foreach ($name in $defaults.Keys) {
|
||||
if (!(Get-Variable -Name $name -ValueOnly)) { Set-Variable -Name $name -Value (Join-Path $repo $defaults[$name]) }
|
||||
}
|
||||
$GpuArch=Get-D4RGpuTarget $DiagnosticRoot $GpuArch
|
||||
if ($GpuArch -ne 'gfx1201' -and !$PSBoundParameters.ContainsKey('PackageRoot')) { $PackageRoot += "-$GpuArch" }
|
||||
if (!$KernelKRoot) { $KernelKRoot=Join-Path $repo "build/native-k-$GpuArch" }
|
||||
if (!$KernelMRoot) { $KernelMRoot=Join-Path $repo "build/native-m-$GpuArch" }
|
||||
foreach ($directory in @($KernelKRoot,$KernelMRoot)) {
|
||||
$info=Join-Path $directory 'build-info.json'
|
||||
if ((Test-Path -LiteralPath $info) -and (Get-Content -LiteralPath $info -Raw | ConvertFrom-Json).architecture -ne $GpuArch) {
|
||||
throw "Native kernel directory does not target ${GpuArch}: $directory"
|
||||
}
|
||||
}
|
||||
$PackageRoot = [IO.Path]::GetFullPath($PackageRoot)
|
||||
if (Test-Path -LiteralPath (Join-Path $PackageRoot 'gpu-target.json')) {
|
||||
[void](Get-D4RGpuTarget $PackageRoot $GpuArch)
|
||||
}
|
||||
$sourceCommit=(& git -C $repo rev-parse HEAD).Trim()
|
||||
$sourceDirty=[bool](& git -C $repo status --porcelain)
|
||||
if ($ArchivePath -and $sourceDirty) { throw 'Commit source changes before creating the distributable archive.' }
|
||||
New-Item -ItemType Directory -Force $PackageRoot | Out-Null
|
||||
$stagedFiles = [Collections.Generic.List[string]]::new()
|
||||
function Stage([string]$source, [string]$relative) {
|
||||
if (!(Test-Path -LiteralPath $source -PathType Leaf)) { throw "Required file missing: $source" }
|
||||
if ($source.EndsWith('.hsaco')) { Assert-D4RCodeObjectTarget $source $GpuArch }
|
||||
$destination = Join-Path $PackageRoot $relative
|
||||
New-Item -ItemType Directory -Force (Split-Path $destination) | Out-Null
|
||||
Copy-Item -LiteralPath $source -Destination $destination -Force
|
||||
$stagedFiles.Add($relative)
|
||||
}
|
||||
Stage (Join-Path $OptiScalerRoot 'OptiScaler.dll') 'OptiScaler.dll'
|
||||
Stage (Join-Path $DiagnosticRoot 'bin/d4r_nvngx.dll') 'd4r/_nvngx.dll'
|
||||
Stage (Join-Path $DiagnosticRoot "bin/pixel_convert_${GpuArch}.hsaco") "d4r/pixel_convert_${GpuArch}.hsaco"
|
||||
Stage (Join-Path $DiagnosticRoot 'nvapi-compat/nvapi64.dll') 'd4r/nvapi/nvapi64.dll'
|
||||
Stage (Join-Path $DiagnosticRoot 'nvapi-compat/nvapi64.dll') 'd4r/zluda/nvapi64.dll'
|
||||
Stage (Join-Path $DiagnosticRoot 'bin/d4r_debug_launcher.exe') 'd4r/d4r_debug_launcher.exe'
|
||||
Stage (Join-Path $DiagnosticRoot 'bin/d4r_gpu_inventory.exe') 'd4r/d4r_gpu_inventory.exe'
|
||||
foreach ($file in @('nvcuda.dll','libc++.dll','libunwind.dll')) { Stage (Join-Path $ZludaRoot $file) "d4r/zluda/$file" }
|
||||
Stage (Join-Path $ZludaRoot 'nvapi64.dll') 'd4r/nvapi/backend/nvapi64.dll'
|
||||
foreach ($file in @('amdhip64_7.dll','amd_comgr.dll','hiprtc0717.dll','hiprtc-builtins0717.dll','rocm_kpack.dll','AMD.ROCM.Comgr.MANIFEST','.hipVersion')) {
|
||||
Stage (Join-Path $HipRoot "bin/$file") "d4r/hip/bin/$file"
|
||||
}
|
||||
$entries = @{}
|
||||
foreach ($directory in @($KernelKRoot, $KernelMRoot)) {
|
||||
foreach ($line in Get-Content -LiteralPath (Join-Path $directory 'd4r-kernels.txt')) {
|
||||
if (!$line.Trim() -or $line.StartsWith('#')) { continue }
|
||||
if ($line -notmatch '^([A-Za-z0-9_]+) ([0-9a-f]{16})$') { throw "Invalid native manifest entry: $line" }
|
||||
if ($entries.ContainsKey($Matches[1])) { throw "Duplicate native kernel: $($Matches[1])" }
|
||||
$entries[$Matches[1]] = $line
|
||||
Stage (Join-Path $directory ($Matches[1] + '.hsaco')) ('d4r/native/' + $Matches[1] + '.hsaco')
|
||||
}
|
||||
}
|
||||
if ($entries.Count -ne 16) { throw "Expected 11 K and 5 M native objects; found $($entries.Count)" }
|
||||
@("# Audited DLSS 310.9.1: K + M $GpuArch native kernels", '# Generated by package-windows-game.ps1') + @($entries.Values | Sort-Object) |
|
||||
Set-Content -LiteralPath (Join-Path $PackageRoot 'd4r/native/d4r-kernels.txt') -Encoding ASCII
|
||||
$stagedFiles.Add('d4r/native/d4r-kernels.txt')
|
||||
Stage (Join-Path $repo 'scripts/windows/windows-game.ps1') 'windows-game.ps1'
|
||||
Stage (Join-Path $repo 'scripts/windows/game-summary.ps1') 'game-summary.ps1'
|
||||
Stage (Join-Path $repo 'scripts/windows/gpu-target.ps1') 'gpu-target.ps1'
|
||||
Stage (Join-Path $repo 'scripts/windows/gpu-preflight.ps1') 'gpu-preflight.ps1'
|
||||
Stage (Join-Path $repo 'tools/windows/gpu-targets.json') 'gpu-targets.json'
|
||||
if (Test-Path -LiteralPath (Join-Path $DiagnosticRoot 'gpu-target.json')) {
|
||||
Stage (Join-Path $DiagnosticRoot 'gpu-target.json') 'gpu-target.json'
|
||||
} else {
|
||||
# Builds predating target metadata were gfx1201 only. The resolver rejects
|
||||
# any attempt to label such a build gfx1200.
|
||||
@{schema=1; architecture=$GpuArch; wavefrontSize=32; validation='Legacy gfx1201 build; run the GPU diagnostics.'} |
|
||||
ConvertTo-Json | Set-Content -LiteralPath (Join-Path $PackageRoot 'gpu-target.json') -Encoding UTF8
|
||||
$stagedFiles.Add('gpu-target.json')
|
||||
}
|
||||
Stage (Join-Path $repo 'scripts/windows/capture-game-window.ps1') 'capture-game-window.ps1'
|
||||
Stage (Join-Path $repo 'scripts/windows/send-game-key.ps1') 'send-game-key.ps1'
|
||||
Stage (Join-Path $repo 'docs/windows-game.md') 'README.md'
|
||||
Stage (Join-Path $repo 'docs/windows-performance.md') 'windows-performance.md'
|
||||
Stage (Join-Path $repo 'docs/windows-gpu-support.md') 'windows-gpu-support.md'
|
||||
Stage (Join-Path $repo 'docs/windows-rdna4-port.md') 'windows-rdna4-port.md'
|
||||
Stage (Join-Path $repo 'packaging/windows/d4r.ini') 'd4r/d4r.ini'
|
||||
Stage (Join-Path $repo 'tools/windows/dll-pins.txt') 'd4r/dll-pins.txt'
|
||||
Stage (Join-Path $repo 'LICENSE') 'licenses/d4r.txt'
|
||||
Stage (Join-Path $OptiScalerRoot 'GPL-3.0.txt') 'licenses/OptiScaler.txt'
|
||||
Stage (Join-Path $repo 'external/ZLUDA/LICENSE-MIT') 'licenses/ZLUDA-MIT.txt'
|
||||
Stage (Join-Path $repo 'external/ZLUDA/LICENSE-APACHE') 'licenses/ZLUDA-APACHE.txt'
|
||||
Stage (Join-Path $repo 'external/ZLUDA/ext/llvm-project/LICENSE.TXT') 'licenses/LLVM.txt'
|
||||
Stage (Join-Path $repo 'external/MinHook/LICENSE.txt') 'licenses/MinHook.txt'
|
||||
Stage (Join-Path $HipRoot 'share/doc/amd_comgr/LICENSE.txt') 'licenses/amd_comgr.txt'
|
||||
Stage (Join-Path $repo 'external/clr/LICENSE.md') 'licenses/HIP-CLR.txt'
|
||||
foreach ($patch in Get-ChildItem -LiteralPath (Join-Path $repo 'patches/optiscaler') -Filter '*.patch') { Stage $patch.FullName ('source-patches/optiscaler/' + $patch.Name) }
|
||||
foreach ($patch in Get-ChildItem -LiteralPath (Join-Path $repo 'patches/zluda') -Filter '*.patch') { Stage $patch.FullName ('source-patches/zluda/' + $patch.Name) }
|
||||
foreach ($patch in Get-ChildItem -LiteralPath (Join-Path $repo 'patches/zluda-llvm') -Filter '*.patch') { Stage $patch.FullName ('source-patches/zluda-llvm/' + $patch.Name) }
|
||||
$sourceArchive=Join-Path $repo 'build/d4r-windows-rdna4-source.zip'
|
||||
New-Item -ItemType Directory -Force (Split-Path $sourceArchive) | Out-Null
|
||||
& git -C $repo archive --format=zip --output $sourceArchive $sourceCommit
|
||||
if ($LASTEXITCODE) { throw 'Cannot archive the committed d4r source.' }
|
||||
Stage $sourceArchive 'source/d4r-windows-rdna4-source.zip'
|
||||
$mBuildInfoPath=Join-Path $KernelMRoot 'build-info.json'
|
||||
$nativeFp8M=$false; $tokenLanesM=$false
|
||||
if (Test-Path -LiteralPath $mBuildInfoPath) {
|
||||
$mBuildInfo=Get-Content -LiteralPath $mBuildInfoPath -Raw | ConvertFrom-Json
|
||||
if ($mBuildInfo.architecture -ne $GpuArch -or $mBuildInfo.family -ne 'M') { throw "M build metadata does not match ${GpuArch}: $mBuildInfoPath" }
|
||||
$nativeFp8M=[bool]$mBuildInfo.nativeFp8
|
||||
$tokenLanesM=[bool]$mBuildInfo.tokenLanes
|
||||
}
|
||||
if ($nativeFp8M) {
|
||||
if (!$tokenLanesM -or $GpuArch -notmatch '^gfx12') { throw 'Native FP8 M packages require the gfx12 token-lane build.' }
|
||||
$marker=Join-Path $PackageRoot 'd4r/native/d4r-native-fp8.txt'
|
||||
New-Item -ItemType Directory -Force (Split-Path $marker) | Out-Null
|
||||
@('D4R_NATIVE_FP8 1',"architecture=$GpuArch",'family=M','tokenLanes=true') |
|
||||
Set-Content -LiteralPath $marker -Encoding ASCII
|
||||
$stagedFiles.Add('d4r/native/d4r-native-fp8.txt')
|
||||
}
|
||||
$metadata = [ordered]@{
|
||||
architecture=$GpuArch; hardwareValidation=$(if ($GpuArch -eq 'gfx1201') { 'RX9070XT tested; other SKUs unverified' } else { 'compile-only; hardware unverified' }); nativeKernels=16; dlssSha256='3975567b8943c53acce397f2b72380092f84f162d00b0d2c7d08a1025c563983';
|
||||
nativeFp8M=$nativeFp8M; tokenLanesM=$tokenLanesM;
|
||||
d4rCommit=$sourceCommit;
|
||||
d4rWorkingTreeDirty=$sourceDirty;
|
||||
optiScaler=(Get-Content -LiteralPath (Join-Path $OptiScalerRoot 'build-info.json') -Raw | ConvertFrom-Json);
|
||||
zludaBuild=(Get-Content -LiteralPath (Join-Path $ZludaRoot 'build-info.json') -Raw | ConvertFrom-Json);
|
||||
hipRuntime='TheRock 10.2.0a20260929 / HIP 7.17.26386';
|
||||
files=@($stagedFiles | ForEach-Object { Get-Item -LiteralPath (Join-Path $PackageRoot $_) } | ForEach-Object {
|
||||
@{path=$_.FullName.Substring($PackageRoot.Length+1); sha256=(Get-FileHash -LiteralPath $_.FullName -Algorithm SHA256).Hash; version=$_.VersionInfo.FileVersion}
|
||||
})
|
||||
}
|
||||
$metadata | ConvertTo-Json -Depth 10 | Set-Content -LiteralPath (Join-Path $PackageRoot 'package.json') -Encoding UTF8
|
||||
if (Get-ChildItem -LiteralPath $PackageRoot -Recurse -File | Where-Object { $_.Name -eq 'nvngx_dlss.dll' -or ($_.Name -eq '_nvngx.dll' -and $_.DirectoryName -ne (Join-Path $PackageRoot 'd4r')) }) {
|
||||
throw 'Proprietary NVIDIA binaries must not be in the package.'
|
||||
}
|
||||
Write-Host "Windows game development package: $PackageRoot (NVIDIA DLLs excluded)"
|
||||
if ($ArchivePath) {
|
||||
$ArchivePath=[IO.Path]::GetFullPath($ArchivePath)
|
||||
if ($ArchivePath.StartsWith($PackageRoot+'\',[StringComparison]::OrdinalIgnoreCase)) { throw 'Place the ZIP outside the package directory.' }
|
||||
Add-Type -AssemblyName System.IO.Compression,System.IO.Compression.FileSystem
|
||||
New-Item -ItemType Directory -Force (Split-Path $ArchivePath) | Out-Null
|
||||
$stream=[IO.File]::Open($ArchivePath,[IO.FileMode]::Create,[IO.FileAccess]::ReadWrite,[IO.FileShare]::None)
|
||||
$archive=$null
|
||||
try {
|
||||
$archive=[IO.Compression.ZipArchive]::new($stream,[IO.Compression.ZipArchiveMode]::Create)
|
||||
# Never archive caches, results, private texture outputs or a local DLL
|
||||
# merely because it happens to be somewhere under PackageRoot.
|
||||
foreach ($relative in @($stagedFiles)+@('package.json')) {
|
||||
[void][IO.Compression.ZipFileExtensions]::CreateEntryFromFile($archive,
|
||||
(Join-Path $PackageRoot $relative),$relative.Replace('\','/'),[IO.Compression.CompressionLevel]::Optimal)
|
||||
}
|
||||
} finally { if ($archive) { $archive.Dispose() }; $stream.Dispose() }
|
||||
Write-Host "Public package archive: $ArchivePath"
|
||||
}
|
||||
Reference in new issue
Block a user