Add enable_fusion_modeling for conv2d and conv3d

[jerryzh168]

Summary:
This is to have more accurate benchmarking for fusion

Test Plan:

RECIPE_NAME="tensorwise"
SHAPE_GEN_NAME="custom"
SCRIPT_PATH="benchmarks/float8/float8_inference_roofline.py"
M=1
K=160
N=320
D=3
H=194
W=130
kernel_size=3

OUTPUT_FILE="~/local/tmp/test_${M}_${K}_${N}_${D}_${H}_${W}_${kernel_size}.csv"
python $SCRIPT_PATH $OUTPUT_FILE \
    --recipe_name $RECIPE_NAME \
    --shape_gen_name $SHAPE_GEN_NAME \
    --M $M --K $K --N $N  \
    --D $D --H $H --W $W  \
    --kernel_size $kernel_size \
    --enable_fusion_modeling \
    --op_name conv3d

Reviewers:

Subscribers:

Tasks:

Tags:

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/3343

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit 4adbcaa with merge base ff0e461 ():
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[vkuzo]

is there a reason to have an epilogue here? It's nice to have a prologue to measure the fusion of preceding activation, but unless the conv kernel does epilogue fusion I'm not sure we need an epilogue here, it would just measure the same thing for both bf16 and lowp.

[jerryzh168]

oh I just saw conv was typically surrounded by activations/norms, do we only support prologue fusions typically?

[jerryzh168]

just checked, we'll have epilogue fusion for conv op. and can explore prologue fusion for the quant kernels later. so the current test is OK I think