add mlonmcu_benchmarks_spike_vlen128_gcc14_{spike_rv64,llvm19_k230_ss…

…h}.sh script (TODO: combine)

Integration/MLonMCU/mlonmcu_benchmarks_spike_vlen128_gcc14_llvm19_k230_ssh.sh

@@ -0,0 +1,22 @@
+#!/bin/bash
+# TODO: docs and license
+
+set -e
+
+source common.sh
+
+source $VENV_DIR/bin/activate
+
+TIMESTAMP=$(date +%Y%m%dT%H%M%S)
+
+export MLONMCU_HOME=$WORKSPACE_DIR
+
+PARALLEL=1
+TARGET=canmv_k230_ssh
+
+MODELS=(aww vww resnet toycar)
+
+for MODEL in "${MODELS[@]}"
+do
+    python gen_muriscnn_benchmarks.py $MODEL -b tflmi -t $TARGET --vlen 128 --post -p --parallel $PARALLEL -f muriscvnn -f vext -f auto_vectorize --toolchain gcc --toolchain llvm --label ${TIMESTAMP}-${TARGET}-${MODEL}-tflm-gccllvm --opt 3 --opt s --unroll 0 --unroll 1 --custom-unroll 0 --custom-unroll 1 --portable --baseline 0 --skip-default --runs-per-stage
+done

Integration/MLonMCU/mlonmcu_benchmarks_spike_vlen128_gcc14_llvm19_spike_rv64.sh

@@ -0,0 +1,21 @@
+#!/bin/bash
+# TODO: docs and license
+
+set -e
+
+source common.sh
+
+source $VENV_DIR/bin/activate
+
+TIMESTAMP=$(date +%Y%m%dT%H%M%S)
+
+export MLONMCU_HOME=$WORKSPACE_DIR
+
+PARALLEL=$(($(nproc) / 2))
+
+MODELS=(aww vww resnet toycar)
+
+for MODEL in "${MODELS[@]}"
+do
+    python gen_muriscnn_benchmarks.py $MODEL -b tflmi -t $TARGET --vlen 128 --post -p --parallel $PARALLEL -f muriscvnn -f vext -f auto_vectorize --toolchain gcc --toolchain llvm --label ${TIMESTAMP}-${TARGET}-${MODEL}-tflm-gccllvm --opt 3 --opt s --unroll 0 --unroll 1 --custom-unroll 0 --custom-unroll 1 --portable --baseline 0 --skip-default
+done