wouter/d4r
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
af450a92b959d554cff0f495d529fb6a935d25bc
d4r/tools/wine_nvcuda_bridge.spec
T
ayo 162bbd77c8 shim: adaptive marker polling, opt-in GPU-side input wait/completion and RGBA8 output 
- [Interop] MarkerPollUs = auto is the new default: the input-marker wait learns when
  each feature's inputs become ready, sleeps in <=200 us chunks, yields near the
  prediction and backs off when late (d4r_input_wait.h); a number keeps fixed pacing
  (0 yields). d4r_config.py no longer exports 200 when the key is unset

- D4R_SHIM_GPU_INPUT_WAIT=1 (experimental, split VRAM, age 0): NGX's null stream
  waits on a coherent Vulkan word imported into HIP (d4rStreamWaitValue32Eq) instead
  of CPU polling. The original readback marker stays the authority; a CPU watchdog
  releases discarded lists and timeouts. Denoiser, verify, capture and dilation
  frames keep the CPU wait

- D4R_SHIM_GPU_COMPLETION=1 (needs GPU_INPUT_WAIT): the bridge writes a completion
  word right after the native output kernel (d4rNgxWriteValue32), and the game's queue
  polls it with a bounded compute wait (kernels/support/vulkan_wait.comp), replacing
  the split list's CPU timeline signal. frame_retired writes the same word on every
  failure path

- D4R_SHIM_NATIVE_COMPLETION=1: a null-stream callback in the bridge signals the
  split timeline (d4rNgxTimeline*); D4rFrameCompletion::can_signal only releases the
  oldest outstanding frame

- RGBA8 output: tex_common.h adds an 'R4' redirect tag, so K's native output kernel
  stores packed RGBA8 directly (D4R_SHIM_OUTPUT_RGBA8=1, checked against the kernel
  manifest by d4rOutputKernelRgba8). Host packing handles RGBA8 too.
  patches/vkd3d-proton/0004 (VKD3D_D4R_LINEAR_RGBA8=WxH) makes such outputs linear and
  exportable so the result can be written in place

- D4R_SHIM_FUSE_RGBA8_INPUT=1: one Vulkan pass converts and copies RGBA8 color input
  (vulkan_read.comp mode 5 matches the blit's toward-zero halves, plus a row stride)

- profiling: thread CPU time from the Unix bridge (Wine's GetThreadTimes ticks at
  10 ms); game-queue D3D12 timestamps (vk_input/vk_dlss_wait/vk_output/vk_total);
  gpu_eval split into input_gpu_wait and dlss_gpu; D4R_PROFILE_GPU_EVENTS,
  _TIMESTAMPS, _LOG and _CPU_TIME=0 switch off individual parts for bisection

- bridge: D4R_NGX_STREAM_PRIORITY=high|normal runs NGX on a priority stream
  (off by default; no gain measured)

- release_feature: a feature already marked retiring by a failed drain can still
  be released normally

- harness: RGBA8 with motion and jitter scenes (re-quantized each frame),
  D4R_HARNESS_RGBA8_RAMP, D4R_HARNESS_THRASH_MB to evict L2/Infinity Cache before
  each evaluation

- kernels/native: in_hoist.py and split_output_paths.py (opt-in scheduling
  experiments); scripts/build_k_ldr_output.sh builds K's LDR output with an
  optional D4R_LDR_OUTPUT_MAX_VGPR cap

- tests: test_input_wait.py and test_latency_optimizations.py; MarkerPollUs and
  frame-completion tests updated
2026-10-13 22:30:41 -04:00

128 lines
7.2 KiB
RPMSpec
Raw Blame History

@ stdcall cuInit(long) cuInit
@ stdcall cuDeviceGetCount(ptr) cuDeviceGetCount
@ stdcall cuDeviceGet(ptr long) cuDeviceGet
@ stdcall cuCtxPushCurrent_v2(ptr) cuCtxPushCurrent_v2
@ stdcall cuCtxPopCurrent_v2(ptr) cuCtxPopCurrent_v2
@ stdcall cuCtxGetDevice(ptr) cuCtxGetDevice
@ stdcall cuCtxCreate_v2(ptr long long) cuCtxCreate_v2
@ stdcall cuCtxDestroy_v2(ptr) cuCtxDestroy_v2
@ stdcall cuCtxSetCurrent(ptr) cuCtxSetCurrent
@ stdcall cuCtxSynchronize() cuCtxSynchronize
@ stdcall cuDeviceGetLuid(ptr ptr long) cuDeviceGetLuid
@ stdcall cuDeviceGetAttribute(ptr long long) cuDeviceGetAttribute
@ stdcall cuDeviceGetUuid(ptr long) cuDeviceGetUuid
@ stdcall cuGetErrorString(long ptr) cuGetErrorString
@ stdcall cuGetProcAddress(ptr ptr long long) cuGetProcAddress
@ stdcall cuGetProcAddress_v2(ptr ptr long long ptr) cuGetProcAddress_v2
@ stdcall cuArrayCreate(ptr ptr) cuArrayCreate
@ stdcall cuArray3DCreate(ptr ptr) cuArray3DCreate
@ stdcall cuArrayDestroy(ptr) cuArrayDestroy
@ stdcall cuArrayGetDescriptor(ptr ptr) cuArrayGetDescriptor
@ stdcall cuDestroyExternalMemory(ptr) cuDestroyExternalMemory
@ stdcall cuLaunchKernel(ptr long long long long long long long ptr ptr ptr) cuLaunchKernel
@ stdcall cuMemAlloc(ptr long long) cuMemAlloc
@ stdcall cuMemAllocHost(ptr long long) cuMemAllocHost
@ stdcall cuMemFree(long long) cuMemFree
@ stdcall cuMemFreeHost(ptr) cuMemFreeHost
@ stdcall cuMemcpy2D(ptr) cuMemcpy2D
@ stdcall cuMemcpyDtoH(ptr long long) cuMemcpyDtoH
@ stdcall cuMemcpyHtoDAsync(long long ptr long long ptr) cuMemcpyHtoDAsync
@ stdcall cuMipmappedArrayDestroy(ptr) cuMipmappedArrayDestroy
@ stdcall cuModuleGetFunction(ptr ptr ptr) cuModuleGetFunction
@ stdcall cuModuleLoad(ptr ptr) cuModuleLoad
@ stdcall cuModuleLoadData(ptr ptr) cuModuleLoadData
@ stdcall cuModuleLoadDataEx(ptr ptr long ptr ptr) cuModuleLoadDataEx
@ stdcall cuModuleUnload(ptr) cuModuleUnload
@ stdcall cuSurfObjectCreate(ptr ptr) cuSurfObjectCreate
@ stdcall cuSurfObjectDestroy(long long) cuSurfObjectDestroy
@ stdcall cuSurfObjectGetResourceDesc(ptr long long) cuSurfObjectGetResourceDesc
@ stdcall cuTexObjectCreate(ptr ptr ptr ptr) cuTexObjectCreate
@ stdcall cuTexObjectDestroy(long long) cuTexObjectDestroy
@ stdcall cuTexObjectGetResourceDesc(ptr long long) cuTexObjectGetResourceDesc
@ stdcall cuEventCreate(ptr ptr) cuEventCreate
@ stdcall cuEventDestroy(ptr) cuEventDestroy
@ stdcall cuEventDestroy_v2(ptr) cuEventDestroy_v2
@ stdcall cuEventRecord(ptr ptr) cuEventRecord
@ stdcall cuEventRecordWithFlags(ptr ptr ptr) cuEventRecordWithFlags
@ stdcall cuEventSynchronize(ptr) cuEventSynchronize
@ stdcall cuEventQuery(ptr) cuEventQuery
@ stdcall cuEventElapsedTime(ptr ptr ptr) cuEventElapsedTime
@ stdcall cuStreamCreate(ptr ptr) cuStreamCreate
@ stdcall cuStreamCreateWithPriority(ptr ptr ptr) cuStreamCreateWithPriority
@ stdcall cuStreamDestroy(ptr) cuStreamDestroy
@ stdcall cuStreamDestroy_v2(ptr) cuStreamDestroy_v2
@ stdcall cuStreamSynchronize(ptr) cuStreamSynchronize
@ stdcall cuStreamQuery(ptr) cuStreamQuery
@ stdcall cuStreamWaitEvent(ptr ptr ptr) cuStreamWaitEvent
@ stdcall cuStreamGetFlags(ptr ptr) cuStreamGetFlags
@ stdcall cuStreamGetPriority(ptr ptr) cuStreamGetPriority
@ stdcall cuCtxGetCurrent(ptr) cuCtxGetCurrent
@ stdcall cuCtxGetStreamPriorityRange(ptr ptr) cuCtxGetStreamPriorityRange
@ stdcall cuCtxGetLimit(ptr ptr) cuCtxGetLimit
@ stdcall cuCtxSetLimit(ptr ptr) cuCtxSetLimit
@ stdcall cuDriverGetVersion(ptr) cuDriverGetVersion
@ stdcall cuDeviceGetName(ptr ptr ptr) cuDeviceGetName
@ stdcall cuDeviceTotalMem(ptr ptr) cuDeviceTotalMem
@ stdcall cuDeviceTotalMem_v2(ptr ptr) cuDeviceTotalMem_v2
@ stdcall cuDeviceComputeCapability(ptr ptr ptr) cuDeviceComputeCapability
@ stdcall cuMemGetInfo(ptr ptr) cuMemGetInfo
@ stdcall cuMemGetInfo_v2(ptr ptr) cuMemGetInfo_v2
@ stdcall cuMemcpyHtoD(ptr ptr ptr) cuMemcpyHtoD
@ stdcall cuMemcpyHtoD_v2(ptr ptr ptr) cuMemcpyHtoD_v2
@ stdcall cuMemcpyDtoD(ptr ptr ptr) cuMemcpyDtoD
@ stdcall cuMemcpyDtoDAsync(ptr ptr ptr ptr) cuMemcpyDtoDAsync
@ stdcall cuMemcpyDtoHAsync(ptr ptr ptr ptr) cuMemcpyDtoHAsync
@ stdcall cuMemsetD8(ptr ptr ptr) cuMemsetD8
@ stdcall cuMemsetD16(ptr ptr ptr) cuMemsetD16
@ stdcall cuMemsetD32(ptr ptr ptr) cuMemsetD32
@ stdcall cuMemsetD8Async(ptr ptr ptr ptr) cuMemsetD8Async
@ stdcall cuMemsetD16Async(ptr ptr ptr ptr) cuMemsetD16Async
@ stdcall cuMemsetD32Async(ptr ptr ptr ptr) cuMemsetD32Async
@ stdcall cuMemsetD2D8(ptr ptr ptr ptr ptr) cuMemsetD2D8
@ stdcall cuMemsetD2D32(ptr ptr ptr ptr ptr) cuMemsetD2D32
@ stdcall cuFuncSetAttribute(ptr ptr ptr) cuFuncSetAttribute
@ stdcall cuFuncGetAttribute(ptr ptr ptr) cuFuncGetAttribute
@ stdcall cuModuleGetGlobal(ptr ptr ptr ptr) cuModuleGetGlobal
@ stdcall cuOccupancyMaxActiveBlocksPerMultiprocessor(ptr ptr ptr ptr) cuOccupancyMaxActiveBlocksPerMultiprocessor
@ stdcall cuOccupancyMaxPotentialBlockSize(ptr ptr ptr ptr ptr ptr) cuOccupancyMaxPotentialBlockSize
@ stdcall cuMemAllocPitch(ptr ptr ptr ptr ptr) cuMemAllocPitch
@ stdcall cuMemHostAlloc(ptr ptr ptr) cuMemHostAlloc
@ stdcall cuPointerGetAttribute(ptr ptr ptr) cuPointerGetAttribute
@ stdcall d4rImportVulkanMemory(ptr int64 int64 ptr ptr) d4rImportVulkanMemory
@ stdcall d4rReleaseVulkanMemory(ptr) d4rReleaseVulkanMemory
@ stdcall d4rEngineNetCreate(ptr int64 ptr long long long int64 int64 ptr) d4rEngineNetCreate
@ stdcall d4rEngineNetLaunch(ptr long) d4rEngineNetLaunch
@ stdcall d4rEngineNetDone(ptr ptr) d4rEngineNetDone
@ stdcall d4rEngineNetDestroy(ptr) d4rEngineNetDestroy
@ stdcall d4rEngineNetInputWaitReady(ptr int64 ptr) d4rEngineNetInputWaitReady
@ stdcall d4rEngineNetLaunchAt(ptr long) d4rEngineNetLaunchAt
@ stdcall d4rEngineNetReleaseInput(ptr long) d4rEngineNetReleaseInput
@ stdcall d4rEngineNetSetTimeline(ptr ptr int64) d4rEngineNetSetTimeline
@ stdcall d4rEngineNetSignalTimeline(ptr int64) d4rEngineNetSignalTimeline
@ stdcall d4rMemcpy2DAsync(ptr ptr) d4rMemcpy2DAsync
@ stdcall d4rCtxSynchronize() d4rCtxSynchronize
@ stdcall d4rSetShutdownHook(ptr) d4rSetShutdownHook
@ stdcall d4rEventSynchronize(ptr) d4rEventSynchronize
@ stdcall d4rStreamWaitValue32(int64 long) d4rStreamWaitValue32
@ stdcall d4rStreamWaitValue32Eq(int64 long) d4rStreamWaitValue32Eq
@ stdcall d4rNgxStreamPriority(ptr) d4rNgxStreamPriority
@ stdcall d4rNgxWriteValue32(int64 long) d4rNgxWriteValue32
@ stdcall d4rNgxArmCompletion(int64 long long) d4rNgxArmCompletion
@ stdcall d4rNgxCompletionFired() d4rNgxCompletionFired
@ stdcall d4rThreadCpuTimeMs() d4rThreadCpuTimeMs
@ stdcall d4rWriteValue32(int64 long) d4rWriteValue32
@ stdcall d4rSetArrayRedirect(ptr int64 long) d4rSetArrayRedirect
@ stdcall d4rRegisterLinearTexture(int64 int64 int64 long long) d4rRegisterLinearTexture
@ stdcall d4rSetEnv(str str long) d4rSetEnv
@ stdcall d4rLoadError() d4rLoadError
@ stdcall d4rOutputKernelNative() d4rOutputKernelNative
@ stdcall d4rOutputKernelRgba8() d4rOutputKernelRgba8
@ stdcall d4rNgxTimelineCreate(ptr int64 ptr) d4rNgxTimelineCreate
@ stdcall d4rNgxTimelinePrepare(ptr) d4rNgxTimelinePrepare
@ stdcall d4rNgxTimelineCancel(ptr) d4rNgxTimelineCancel
@ stdcall d4rNgxTimelineQueue(ptr int64) d4rNgxTimelineQueue
@ stdcall d4rNgxTimelineSignal(ptr int64) d4rNgxTimelineSignal
@ stdcall d4rNgxTimelineDestroy(ptr) d4rNgxTimelineDestroy
@ stdcall d4rNgxTimelineWait(ptr) d4rNgxTimelineWait
@ stdcall d4rDilateMotionVectors(ptr) d4rDilateMotionVectors
Reference in new issue View Git Blame Copy Permalink