ayo
4334a0c3c4
RDNA4 (experimental; untested on RDNA4 hardware) - kernels/common/wmma_layout.h picks the WMMA operand/accumulator layout per target; K (pwin_*) and M (swin_*) index through it, so the same sources build for gfx11 and gfx12. D4R_WMMA_LAYOUT=12 on gfx11 is a test shim that runs the gfx12 indexing on gfx11 WMMA. - M can use native e4m3 FP8 WMMA on gfx12 (D4R_NATIVE_FP8=1). Texture tails (enc0_tail, dec0_head) get gfx12 f16 and FP8 paths that match ZLUDA's gfx12 lowering. - ZLUDA patch 0007: gfx12 WMMA helpers, optional native FP8 MMA (D4R_ZLUDA_WMMA_FP8_NATIVE), a layout override for testing, and an architecture argument for d4r_emit. Texture kernels are now compiled offline per target through d4r_emit (D4R_ZLUDA_EMIT) instead of on the local GPU. - New d4r.ini option [Kernels] NativeFp8 (default on), wired through the shim and d4r_config.py. On gfx12 the bridge serves kernels/<arch>-fp8 when it is on. - The bridge picks the KFD GPU with the most SIMDs (so an iGPU is not chosen over a discrete GPU); D4R_GPU_ARCH overrides it. - kernels/build.sh and package_release.sh build gfx1100-gfx1103 and gfx1200/gfx1201 (plus -fp8 variants), and D4R_BUNDLE_TEX takes per-target folders. Performance (RDNA3; harness byte-identical unless noted) - ZLUDA patch 0006: m16n8k8 f16 MMAs on WMMA (E; not bit-exact, 52-55 dB per layer, D4R_ZLUDA_WMMA_K8=0 disables it), per-key weight-image slots for native prep kernels (d4r_prep_key_at, d4r_prep_key_slots), and wave64 builds of MMA-free kernels (D4R_ZLUDA_WAVE64). - M: packed FP8 encoding, conflict-free patch-merge reads, scalar weight loads in enc1, and weight prep once per weight set for the six enc3 tube blocks. - K: CU mode for the six non-wide pwin layers. - Texture kernels: every DLSS flag variant is built natively (24 K output, 8 M post, 4 M downsample, 4 M enc0), not only Townfall's; post and K output run as wave64. - M direct output: rrlite_downsample_kernel writes the shim's output buffer; the shim's default direct-output presets are now 11,13. Tools - dump_runner takes the prep grid from d4r_prep_blocks and refuses texture/surface dumps. - New pwin_check.py, swin_check.py, model_enc3.py and tail_check.hip. Docs: README GPU table and RDNA4 notes, building (patches 0002-0007, d4r_emit), native-kernels (RDNA4 section), performance (new numbers), packaging README. Version 0.1.2.
34 lines
2.3 KiB
Bash
Executable File
34 lines
2.3 KiB
Bash
Executable File
#!/usr/bin/env bash
|
|
# build_tex.sh KERNEL SRC OUT_DIR: one texture kernel. NVIDIA's PTX for KERNEL, edited by make_ptx.py so
|
|
# part of it calls functions from SRC.hip, is compiled by ZLUDA with SRC's bitcode linked in
|
|
# (D4R_ZLUDA_EXTRA_BC) and the resulting code object is saved as OUT_DIR/KERNEL.hsaco.
|
|
# Called by kernels/build.sh, which sets D4R_ROCM_DIR, D4R_GPU_ARCH, D4R_DLSS_PTX_DIR, D4R_ZLUDA_EMIT,
|
|
# D4R_ZLUDA_WAVE64 (1 = wave64 code object) and D4R_TEX_FP8 (1 = the gfx12 native-FP8 variant).
|
|
# ZLUDA compiles offline for D4R_GPU_ARCH with its d4r_emit tool, so any target builds on any machine.
|
|
set -euo pipefail
|
|
K=$1 SRC=$2 OUT=$3
|
|
D="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
|
|
R="${D4R_ROCM_DIR:-/opt/rocm}"
|
|
ARCH="${D4R_GPU_ARCH:-gfx1101}"
|
|
FP8="${D4R_TEX_FP8:-0}"
|
|
LLVM="$R/lib/llvm/bin"
|
|
B="$(mktemp -d)"
|
|
trap 'rm -rf "$B"' EXIT
|
|
# D4R_TEX_CFLAGS: extra flags for SRC.hip (tests, e.g. -DD4R_WMMA_LAYOUT=12 for the gfx12-layout shim on gfx11)
|
|
read -r -a FLAGS <<< "${D4R_TEX_CFLAGS:-}"
|
|
[[ "$FP8" == 1 ]] && FLAGS+=(-DD4R_TEX_FP8)
|
|
# SRC.hip -> bitcode without target attributes, so ZLUDA's own target settings apply after linking
|
|
"$LLVM/clang++" -x hip -std=c++20 -nogpuinc -nogpulib -O3 -mno-wavefrontsize64 --offload-device-only \
|
|
--offload-arch="$ARCH" -fgpu-rdc -emit-llvm -c -Xclang -fdenormal-fp-math=dynamic -DD4R_KERNEL_NAME="$K" \
|
|
"${FLAGS[@]}" -o "$B/raw.bc" "$D/$SRC.hip"
|
|
"$LLVM/llvm-dis" "$B/raw.bc" -o "$B/raw.ll"
|
|
sed -E -e '/@llvm.used/d' -e '/wchar_size/d' -e '/llvm.module.flags/d' -e '/__hip_cuid/d' -e 's/optnone//g' \
|
|
-e "s/\"target-cpu\"=\"$ARCH\"//g" -e 's/"target-features"="[^"]+"//g' "$B/raw.ll" | "$LLVM/llvm-as" -o "$B/extra.bc" -
|
|
python3 "$D/make_ptx.py" "$K" "$B/$K.ptx" >/dev/null
|
|
# the code-generation switches of the runtime (d4r.ini [Kernels]), which ZLUDA's module cache keys on
|
|
env LD_LIBRARY_PATH="$R/lib" D4R_ZLUDA_EXTRA_BC="$B/extra.bc" D4R_ZLUDA_WMMA=1 D4R_ZLUDA_WMMA_FP8=1 \
|
|
D4R_ZLUDA_WMMA_FP8_NATIVE="$FP8" D4R_ZLUDA_IGNORE_DENORMAL=1 D4R_ZLUDA_IMPLICIT_MAX_BLOCK=256 \
|
|
D4R_ZLUDA_WAVE64="${D4R_ZLUDA_WAVE64:-0}" "$D4R_ZLUDA_EMIT" "$B/$K.ptx" "$B/dump" "$ARCH" >/dev/null
|
|
cp "$B/dump/module.hsaco" "$OUT/$K.hsaco"
|
|
printf '%-52s %s\n' "$K" "$(grep -E '^\s+\.(vgpr_count|private_segment_fixed_size|wavefront_size):' "$B/dump/asm.s" | tr -s ' ' | paste -sd' ')"
|