[PATCH 17/20] target/arm: Implement AESEMC

Signed-off-by: Richard Henderson <[email protected]>
---
 target/arm/tcg/helper-defs.h   |  1 +
 target/arm/tcg/crypto_helper.c | 31 +++++++++++++++++++++++++++++++
 target/arm/tcg/translate-sve.c |  2 ++
 target/arm/tcg/sve.decode      |  3 +++
 4 files changed, 37 insertions(+)

diff --git a/target/arm/tcg/helper-defs.h b/target/arm/tcg/helper-defs.h
index 3719487754..5ab89dc886 100644
--- a/target/arm/tcg/helper-defs.h
+++ b/target/arm/tcg/helper-defs.h
@@ -461,6 +461,7 @@ DEF_HELPER_FLAGS_4(crypto_aese, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
 DEF_HELPER_FLAGS_4(crypto_aesd, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
 DEF_HELPER_FLAGS_3(crypto_aesmc, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
 DEF_HELPER_FLAGS_3(crypto_aesimc, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(crypto_aesemc, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
 
 DEF_HELPER_FLAGS_4(crypto_aese_idx, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
 DEF_HELPER_FLAGS_4(crypto_aesd_idx, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
diff --git a/target/arm/tcg/crypto_helper.c b/target/arm/tcg/crypto_helper.c
index 05f193b3d1..ae677cdf20 100644
--- a/target/arm/tcg/crypto_helper.c
+++ b/target/arm/tcg/crypto_helper.c
@@ -193,6 +193,37 @@ void HELPER(crypto_aesimc)(void *vd, void *vm, uint32_t desc)
     clear_tail(vd, opr_sz, simd_maxsz(desc));
 }
 
+void HELPER(crypto_aesemc)(void *vd, void *vn, void *vm, uint32_t desc)
+{
+    intptr_t opr_sz = simd_oprsz(desc);
+    intptr_t idx = simd_data(desc);
+    void *vm_idx = vm + idx * 16;
+    intptr_t s = opr_sz - 16;
+
+    do {
+        intptr_t base = ROUND_DOWN(s, 4 * 16);
+        AESState rk = *(AESState *)(vm_idx + base);
+        do {
+            AESState *ad = (AESState *)(vd + s);
+            AESState *st = (AESState *)(vn + s);
+            AESState t;
+
+            /* Like above: Arm AK comes first; api AK comes last */
+            if (HOST_BIG_ENDIAN) {
+                t.d[0] = st->d[1] ^ rk.d[1];
+                t.d[1] = st->d[0] ^ rk.d[0];
+                aesenc_SB_SR_MC_AK(&t, &t, &aes_zero, false);
+                ad->d[0] = t.d[1];
+                ad->d[1] = t.d[0];
+            } else {
+                t.v = st->v ^ rk.v;
+                aesenc_SB_SR_MC_AK(ad, &t, &aes_zero, false);
+            }
+            s -= 16;
+        } while (s >= base);
+    } while (s > 0);
+}
+
 /*
  * SHA-1 logical functions
  */
diff --git a/target/arm/tcg/translate-sve.c b/target/arm/tcg/translate-sve.c
index 340ed7d975..fb809b905f 100644
--- a/target/arm/tcg/translate-sve.c
+++ b/target/arm/tcg/translate-sve.c
@@ -8336,6 +8336,8 @@ TRANS_FEAT_STREAMING_IF(AESE_idx, aa64_sve_aes2, aa64_ssve_aes,
                         do_aes2_idx, a, gen_helper_crypto_aese_idx)
 TRANS_FEAT_STREAMING_IF(AESD_idx, aa64_sve_aes2, aa64_ssve_aes,
                         do_aes2_idx, a, gen_helper_crypto_aesd_idx)
+TRANS_FEAT_STREAMING_IF(AESEMC, aa64_sve_aes2, aa64_ssve_aes,
+                        do_aes2_idx, a, gen_helper_crypto_aesemc)
 
 TRANS_FEAT(FCVTNT_sh_m, aa64_sme_or_sve2, gen_gvec_fpst_arg_zpz,
            gen_helper_sve2_fcvtnt_sh, a, 0, FPST_A64)
diff --git a/target/arm/tcg/sve.decode b/target/arm/tcg/sve.decode
index 1ab1b4042b..5832e0190c 100644
--- a/target/arm/tcg/sve.decode
+++ b/target/arm/tcg/sve.decode
@@ -1988,6 +1988,9 @@ AESE_idx        01000101 00 1 .. .10111010 ..... ...00   @aes2_4
 AESD_idx        01000101 00 1 .. .10111011 ..... ....0   @aes2_2
 AESD_idx        01000101 00 1 .. .10111011 ..... ...00   @aes2_4
 
+AESEMC          01000101 00 1 .. .11111010 ..... ....0   @aes2_2
+AESEMC          01000101 00 1 .. .11111010 ..... ...00   @aes2_4
+
 # SVE2 crypto constructive binary operations
 SM4EKEY         01000101 00 1 ..... 11110 0 ..... .....  @rd_rn_rm_e0
 RAX1            01000101 00 1 ..... 11110 1 ..... .....  @rd_rn_rm_e0
-- 
2.43.0