wasm-micro-runtime/tests/benchmarks/polybench/run_interp.sh
Wenyong Huang 76be848ec3
Implement the segue optimization for LLVM AOT/JIT (#2230)
Segue is an optimization technology which uses x86 segment register to store
the WebAssembly linear memory base address, so as to remove most of the cost
of SFI (Software-based Fault Isolation) base addition and free up a general
purpose register, by this way it may:
- Improve the performance of JIT/AOT
- Reduce the footprint of JIT/AOT, the JIT/AOT code generated is smaller
- Reduce the compilation time of JIT/AOT

This PR uses the x86-64 GS segment register to apply the optimization, currently
it supports linux and linux-sgx platforms on x86-64 target. By default it is disabled,
developer can use the option below to enable it for wamrc and iwasm(with LLVM
JIT enabled):
```bash
wamrc --enable-segue=[<flags>] -o output_file wasm_file
iwasm --enable-segue=[<flags>] wasm_file [args...]
```
`flags` can be:
    i32.load, i64.load, f32.load, f64.load, v128.load,
    i32.store, i64.store, f32.store, f64.store, v128.store
Use comma to separate them, e.g. `--enable-segue=i32.load,i64.store`,
and `--enable-segue` means all flags are added.

Acknowledgement:
Many thanks to Intel Labs, UC San Diego and UT Austin teams for introducing this
technology and the great support and guidance!

Signed-off-by: Wenyong Huang <wenyong.huang@intel.com>
Co-authored-by: Vahldiek-oberwagner, Anjo Lucas <anjo.lucas.vahldiek-oberwagner@intel.com>
2023-05-26 10:13:33 +08:00

56 lines
1.5 KiB
Bash
Executable File

#!/bin/bash
# Copyright (C) 2019 Intel Corporation. All rights reserved.
# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
CUR_DIR=$PWD
OUT_DIR=$CUR_DIR/out
REPORT=$CUR_DIR/report.txt
TIME=/usr/bin/time
PLATFORM=$(uname -s | tr A-Z a-z)
IWASM_CMD=$CUR_DIR/../../../product-mini/platforms/${PLATFORM}/build/iwasm
BENCH_NAME_MAX_LEN=20
POLYBENCH_CASES="2mm 3mm adi atax bicg cholesky correlation covariance \
deriche doitgen durbin fdtd-2d floyd-warshall gemm gemver \
gesummv gramschmidt heat-3d jacobi-1d jacobi-2d ludcmp lu \
mvt nussinov seidel-2d symm syr2k syrk trisolv trmm"
rm -f $REPORT
touch $REPORT
function print_bench_name()
{
name=$1
echo -en "$name" >> $REPORT
name_len=${#name}
if [ $name_len -lt $BENCH_NAME_MAX_LEN ]
then
spaces=$(( $BENCH_NAME_MAX_LEN - $name_len ))
for i in $(eval echo "{1..$spaces}"); do echo -n " " >> $REPORT; done
fi
}
echo "Start to run cases, the result is written to report.txt"
#run benchmarks
cd $OUT_DIR
echo -en "\t\t\t\t\t native\tiwasm-interp\n" >> $REPORT
for t in $POLYBENCH_CASES
do
print_bench_name $t
echo "run $t with native .."
echo -en "\t" >> $REPORT
$TIME -f "real-%e-time" ./${t}_native 2>&1 | grep "real-.*-time" | awk -F '-' '{ORS=""; print $2}' >> $REPORT
echo "run $t with iwasm interp .."
echo -en "\t" >> $REPORT
$TIME -f "real-%e-time" $IWASM_CMD ${t}.wasm 2>&1 | grep "real-.*-time" | awk -F '-' '{ORS=""; print $2}' >> $REPORT
echo -en "\n" >> $REPORT
done