diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index 32ea391..c20b6d5 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -14,21 +14,30 @@ env:
 
 jobs:
   test:
-    name: Rust ${{matrix.rust}}
+    name: Test
     runs-on: ubuntu-latest
     strategy:
       fail-fast: false
       matrix:
+        target: [x86_64-unknown-linux-gnu, aarch64-unknown-linux-gnu]
         rust: [nightly, beta, stable, 1.64]
     timeout-minutes: 45
     steps:
       - uses: actions/checkout@v3
       - uses: dtolnay/rust-toolchain@master
         with:
-          toolchain: ${{matrix.rust}}
+          toolchain: ${{ matrix.rust }}
+          target: ${{ matrix.target }}
+      - uses: taiki-e/setup-cross-toolchain-action@v1
+        with:
+          target: ${{ matrix.target }}
+
       - name: Enable type layout randomization
         run: echo RUSTFLAGS=${RUSTFLAGS}\ -Zrandomize-layout >> $GITHUB_ENV
         if: matrix.rust == 'nightly'
+
+      - uses: Swatinem/rust-cache@v2
+
       - run: cargo build
       - run: cargo test
       - run: cargo test --release
diff --git a/src/x86.rs b/src/x86.rs
index b34ea48..d255450 100644
--- a/src/x86.rs
+++ b/src/x86.rs
@@ -45,9 +45,8 @@ pub(super) unsafe fn encode<const UPPER: bool>(input: &[u8], output: *mut u8) {
 
         // Store result into the output buffer.
         _mm_storeu_si128(output.add(i).cast(), hex_lo);
-        i += CHUNK_SIZE;
-        _mm_storeu_si128(output.add(i).cast(), hex_hi);
-        i += CHUNK_SIZE;
+        _mm_storeu_si128(output.add(i + CHUNK_SIZE).cast(), hex_hi);
+        i += CHUNK_SIZE * 2;
     }
 
     if !input_remainder.is_empty() {