bisect-issue1.sh

# wget https://sharkpublic.blob.core.windows.net/sharkpublic/chi/llama/atten/fp8_attn.mlir
iree-compile \
  /sharedfile/attn/128/fp8_attn.mlir \
  --iree-hip-target=gfx942 \
  -o=/sharedfile/attn/128/fp8_attn.vmfb \
  --iree-hal-target-device=hip \
  --iree-dispatch-creation-enable-aggressive-fusion=true \
  --iree-global-opt-propagate-transposes=true \
  --iree-opt-aggressively-propagate-transposes=true \
  --iree-opt-data-tiling=false \
  --iree-preprocessing-pass-pipeline='builtin.module(util.func(iree-preprocessing-generalize-linalg-matmul-experimental))' \
  --iree-hal-indirect-command-buffers=true \
  --iree-stream-resource-memory-model=discrete \
  --iree-hal-memoization=true \
  --iree-opt-strip-assertions

# All the input.bin  can be found on  ssh chi@SharkMi300X machine corrensponding patch
ROCR_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \
iree-benchmark-module \
--hip_use_streams=true \
--module=/sharedfile/attn/128/fp8_attn.vmfb \
--parameters=model=/sharedfile/attn/fp8_attn.irpa \
--device=hip://4 \
--function=prefill_bs4 \
--input=4x128xi64=@/sharedfile/128/prefill/prefill_token_ids_4x128xi64.bin \
--input=4xi64=@/sharedfile/128/prefill/prefill_seq_lens_4xi64.bin \
--input=4x4xi64=@/sharedfile/128/prefill/prefill_seq_block_ids_4x4xi64.bin \
--input=261x2097152xf8E4M3FNUZ=@/sharedfile/128/prefill/prefill_cache_state_261x2097152xf8E4M3FNUZ.bin \
--benchmark_repetitions=3