xdfnx-devandrealdody 3fa963fd1b Enable experimental native Windows RDNA3 and RDNA4 targets ...

Share explicit target metadata across build/runtime/package guards, preserve native gfx11 WMMA with a Windows LDS lane exchange, select HIP devices by D3D12 LUID, and reject wrong-target/vkd3d installations before changing game files. Review all issue #10 logs and document compile-only coverage plus RX9070XT regressions.

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

2026-10-02 17:10:42 +03:00

..

benchmark_native_replay.py

Guard paired GPU replay timings and normalize Windows diagnostic environments

2026-10-02 15:03:32 +03:00

binary16_probe.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

binary16.h

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

build_native_texture.py

Enable experimental native Windows RDNA3 and RDNA4 targets

2026-10-02 17:10:42 +03:00

cuda_api.h

Remove game log backpressure and measure Windows K/M kernels

2026-10-01 18:37:14 +03:00

cuda_driver_probe.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

cuda_image_api.h

Batch Windows DLSS GPU inputs and add deferred kernel diagnostics

2026-10-02 12:06:15 +03:00

cuda_image_probe.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

cuda_mma_probe.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

d3d12_boundary_timing.h

Measure Windows interop GPU boundaries without added completion waits

2026-10-02 10:08:47 +03:00

d3d12_command_hooks.cpp

Queue native Windows DLSS asynchronously and drain feature consumers

2026-10-02 10:00:26 +03:00

d3d12_command_hooks.generated.h

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

d3d12_command_hooks.h

Resolve texture states in queue order across DLSS command lists

2026-10-01 16:41:52 +03:00

d3d12_command_probe.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

d3d12_dlss_probe.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

d3d12_external.h

Queue native Windows DLSS asynchronously and drain feature consumers

2026-10-02 10:00:26 +03:00

d3d12_hip_interop_probe.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

d3d12_newer_commands.h

Track Windows D3D12 indirect state and newer command interfaces

2026-10-01 14:15:01 +03:00

d3d12_pixel_probe.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

d3d12_retirement.h

Queue native Windows DLSS asynchronously and drain feature consumers

2026-10-02 10:00:26 +03:00

d4r_nvngx_windows.cpp

Enable experimental native Windows RDNA3 and RDNA4 targets

2026-10-02 17:10:42 +03:00

debug_launcher.cpp

Measure Windows submission costs and validate faster private K output stores

2026-10-01 19:54:41 +03:00

diagnostic.h

Guard paired GPU replay timings and normalize Windows diagnostic environments

2026-10-02 15:03:32 +03:00

fp8_pack_probe.hip

Validate Windows native M FP8 and reduce operand packing cost

2026-10-01 21:16:28 +03:00

fp8_pack_validate.py

Validate Windows native M FP8 and reduce operand packing cost

2026-10-01 21:16:28 +03:00

frame_compare.py

Convert native Windows D3D12 formats on gfx1201 with GPU validation

2026-10-01 14:48:42 +03:00

generate_d3d12_hooks.py

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

gfx12_wmma_probe.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

gpu_inventory.cpp

Enable experimental native Windows RDNA3 and RDNA4 targets

2026-10-02 17:10:42 +03:00

gpu_target_probe.cpp

Enable experimental native Windows RDNA3 and RDNA4 targets

2026-10-02 17:10:42 +03:00

gpu_target.h

Enable experimental native Windows RDNA3 and RDNA4 targets

2026-10-02 17:10:42 +03:00

gpu_targets.generated.h.in

Enable experimental native Windows RDNA3 and RDNA4 targets

2026-10-02 17:10:42 +03:00

gpu_targets.py

Enable experimental native Windows RDNA3 and RDNA4 targets

2026-10-02 17:10:42 +03:00

gpu_telemetry.c

Record controlled K timings and add read-only AMD GPU telemetry

2026-10-02 12:39:14 +03:00

gpu-target.json.in

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

gpu-targets.json

Enable experimental native Windows RDNA3 and RDNA4 targets

2026-10-02 17:10:42 +03:00

hip_api.h

Enable experimental native Windows RDNA3 and RDNA4 targets

2026-10-02 17:10:42 +03:00

hip_gfx1201_probe.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

hip_stream_lifecycle_probe.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

k_layer_validate.py

Enable experimental native Windows RDNA3 and RDNA4 targets

2026-10-02 17:10:42 +03:00

k_module_probe.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

k_ptx_validate.py

Execute native Windows K DLSS and validate real transformer weights

2026-09-30 18:23:01 +03:00

k_replay_validate.py

Enable experimental native Windows RDNA3 and RDNA4 targets

2026-10-02 17:10:42 +03:00

m_capture_compare.py

Correct M attention precision and validate every temporal transformer block

2026-09-30 21:30:05 +03:00

m_native_replay_compare.py

Validate Windows native M FP8 and reduce operand packing cost

2026-10-01 21:16:28 +03:00

m_ptx_validate.py

Correct M attention precision and validate every temporal transformer block

2026-09-30 21:30:05 +03:00

m_replay_validate.py

Enable experimental native Windows RDNA3 and RDNA4 targets

2026-10-02 17:10:42 +03:00

m_temporal_validate.py

Correct M attention precision and validate every temporal transformer block

2026-09-30 21:30:05 +03:00

native_replay_probe.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

ngx_cuda_evaluate.h

Build patched Windows ZLUDA and execute native NGX DLSS diagnostics

2026-09-30 16:35:30 +03:00

ngx_cuda_init_probe.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

ngx_parameters.h

Run native Windows D3D12 DLSS K and M through shared VRAM

2026-09-30 20:22:03 +03:00

ngx_public_abi.h

Forward Windows NGX project identity and validate frontend native network use

2026-10-01 15:45:55 +03:00

ngx_windows_runtime.h

Enable experimental native Windows RDNA3 and RDNA4 targets

2026-10-02 17:10:42 +03:00

nvapi_trace.cpp

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

pixel_conversion.h

Build native Windows packages for both RDNA4 targets

2026-10-02 16:18:18 +03:00

pixel_convert_gfx1201.hip

Validate native Windows DLSS output on GPU and record actual M gameplay

2026-10-01 17:54:24 +03:00

pixel_format.h

Convert native Windows D3D12 formats on gfx1201 with GPU validation

2026-10-01 14:48:42 +03:00

probe_kernel.hip

Execute native Windows K DLSS and validate real transformer weights

2026-09-30 18:23:01 +03:00

profile_report.py

Guard paired GPU replay timings and normalize Windows diagnostic environments

2026-10-02 15:03:32 +03:00

swin_debug_compare.py

Correct M attention precision and validate every temporal transformer block

2026-09-30 21:30:05 +03:00

test_ptx_reference.py

Validate native Windows M against real PTX and correct patch-merge bounds

2026-09-30 19:01:18 +03:00

wmma_f16_compare.hip

Reproduce gfx12 FP16 accumulator differences and preserve K baseline

2026-10-02 13:03:19 +03:00

wmma_f16_compare.py

Reproduce gfx12 FP16 accumulator differences and preserve K baseline

2026-10-02 13:03:19 +03:00

wmma_gfx1201.hip

Enable experimental native Windows RDNA3 and RDNA4 targets

2026-10-02 17:10:42 +03:00