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>

tests/benchmarks/libsodium/build.sh

@@ -16,6 +16,8 @@ libsodium_CASES="aead_aes256gcm2 aead_aes256gcm aead_chacha20poly13052 aead_chac
                  sodium_utils3 sodium_utils sodium_version stream2 stream3 stream4 stream verify1 \
                  xchacha20"
 
+PLATFORM=$(uname -s | tr A-Z a-z)
+
 readonly WAMRC_CMD=$PWD/../../../wamr-compiler/build/wamrc
 readonly OUT_DIR=$PWD/libsodium/zig-out/bin
 
@@ -34,9 +36,16 @@ zig build -Drelease-fast -Denable_benchmarks=true -Dtarget=wasm32-wasi
 for case in ${libsodium_CASES}
 do
     ${WAMRC_CMD} -o ${OUT_DIR}/${case}.aot ${OUT_DIR}/${case}.wasm
-
     if [ "$?" != 0 ]; then
         echo -e "Error while compiling ${case}.wasm to ${case}.aot"
         exit
     fi
+
+    if [[ ${PLATFORM} == "linux" ]]; then
+        ${WAMRC_CMD} --enable-segue -o ${OUT_DIR}/${case}_segue.aot ${OUT_DIR}/${case}.wasm
+        if [ "$?" != 0 ]; then
+            echo -e "Error while compiling ${case}.wasm to ${case}_segue.aot"
+            exit
+        fi
+    fi
 done

tests/benchmarks/libsodium/run_aot.sh

@@ -13,12 +13,14 @@ libsodium_CASES="aead_aes256gcm2 aead_aes256gcm aead_chacha20poly13052 aead_chac
                  scalarmult6 scalarmult7 scalarmult8 scalarmult_ed25519 scalarmult_ristretto255 \
                  scalarmult secretbox2 secretbox7 secretbox8 secretbox_easy2 secretbox_easy \
                  secretbox secretstream shorthash sign siphashx24 sodium_core sodium_utils2 \
-                 sodium_utils3 sodium_utils sodium_version stream2 stream3 stream4 stream verify1 \
-                 xchacha20"
+                 sodium_utils stream2 stream3 stream4 stream verify1 xchacha20"
+
+PLATFORM=$(uname -s | tr A-Z a-z)
 
 readonly OUT_DIR=$PWD/libsodium/zig-out/bin
 readonly REPORT=$PWD/report.txt
-readonly IWASM_CMD=$PWD/../../../product-mini/platforms/linux/build/iwasm
+readonly IWASM_CMD=$PWD/../../../product-mini/platforms/${PLATFORM}/build/iwasm
+readonly TIME=/usr/bin/time
 
 BENCH_NAME_MAX_LEN=20
 
@@ -40,7 +42,11 @@ function print_bench_name()
 # run benchmarks
 cd $OUT_DIR
 
-echo -en "\t\t\t\t\t\tnative\tiwasm-aot\n" >> $REPORT
+if [[ ${PLATFORM} == "linux" ]]; then
+    echo -en "\t\t\t\t\t\tnative\tiwasm-aot\tiwasm-aot-segue\n" >> $REPORT
+else
+    echo -en "\t\t\t\t\t\tnative\tiwasm-aot\n" >> $REPORT
+fi
 
 for t in $libsodium_CASES
 do
@@ -48,11 +54,38 @@ do
 
     echo "run $t with native..."
     echo -en "\t" >> $REPORT
-    ./${t} | awk -F '-' 'BEGIN{FIELDWIDTHS="10"}{ORS=""; print $1 / 1000000.0}' >> $REPORT
+    if [[ $t != "sodium_utils2" ]]; then
+        ./${t} | awk '{printf "%-10.2f", $0/1000000.0}' >> $REPORT
+    else
+        # sodium_utils2 doesn't print the result,
+        # use time command to get result instead
+        $TIME -f "real-%e-time" ./${t} 2>&1 | grep "real-.*-time" |
+            awk -F '-' '{printf "%-10.2f", $2}' >> $REPORT
+    fi
 
     echo "run $t with iwasm aot..."
     echo -en "\t  \t" >> $REPORT
-    $IWASM_CMD ${t}.aot | awk -F '-' 'BEGIN{FIELDWIDTHS="10"}{ORS=""; print $1 / 1000000.0}' >> $REPORT
+    if [[ $t != "sodium_utils2" ]]; then
+        $IWASM_CMD ${t}.aot | awk '{printf "%-10.2f", $0/1000000.0}' >> $REPORT
+    else
+        # sodium_utils2 doesn't print the result,
+        # use time command to get result instead
+        $TIME -f "real-%e-time" $IWASM_CMD ${t}.aot 2>&1 | grep "real-.*-time" |
+            awk -F '-' '{printf "%-10.2f", $2}' >> $REPORT
+    fi
+
+    if [[ ${PLATFORM} == "linux" ]]; then
+        echo "run $t with iwasm aot segue..."
+        echo -en "\t  \t" >> $REPORT
+        if [[ $t != "sodium_utils2" ]]; then
+            $IWASM_CMD ${t}_segue.aot | awk '{printf "%.2f", $0/1000000.0}' >> $REPORT
+        else
+            # sodium_utils2 doesn't print the result,
+            # use time command to get result instead
+            $TIME -f "real-%e-time" $IWASM_CMD ${t}_segue.aot 2>&1 | grep "real-.*-time" |
+                awk -F '-' '{printf "%.2f", $2}' >> $REPORT
+        fi
+    fi
 
     echo -en "\n" >> $REPORT
 done