[PATCH v2 08/38] target/arm: Demultiplex AESE and AESMC

[Richard Henderson]

Split these helpers so that we are not passing 'decrypt'
within the simd descriptor.

Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
 target/arm/helper.h             |  2 ++
 target/arm/tcg/sve.decode       |  4 ++--
 target/arm/tcg/crypto_helper.c  | 37 +++++++++++++++++++++++----------
 target/arm/tcg/translate-a64.c  | 13 ++++--------
 target/arm/tcg/translate-neon.c |  4 ++--
 target/arm/tcg/translate-sve.c  |  8 ++++---
 6 files changed, 41 insertions(+), 27 deletions(-)

diff --git a/target/arm/helper.h b/target/arm/helper.h
index 3335c2b10b..95e32a697a 100644
--- a/target/arm/helper.h
+++ b/target/arm/helper.h
@@ -552,7 +552,9 @@ DEF_HELPER_FLAGS_2(neon_qzip16, TCG_CALL_NO_RWG, void, ptr, 
ptr)
 DEF_HELPER_FLAGS_2(neon_qzip32, TCG_CALL_NO_RWG, void, ptr, ptr)
 
 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_sha1su0, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
 DEF_HELPER_FLAGS_4(crypto_sha1c, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
diff --git a/target/arm/tcg/sve.decode b/target/arm/tcg/sve.decode
index 14b3a69c36..04b6fcc0cf 100644
--- a/target/arm/tcg/sve.decode
+++ b/target/arm/tcg/sve.decode
@@ -1629,8 +1629,8 @@ STNT1_zprz      1110010 .. 10 ..... 001 ... ..... ..... \
 ### SVE2 Crypto Extensions
 
 # SVE2 crypto unary operations
-# AESMC and AESIMC
-AESMC           01000101 00 10000011100 decrypt:1 00000 rd:5
+AESMC           01000101 00 10000011100 0 00000 rd:5
+AESIMC          01000101 00 10000011100 1 00000 rd:5
 
 # SVE2 crypto destructive binary operations
 AESE            01000101 00 10001 0 11100 0 ..... .....  @rdn_rm_e0
diff --git a/target/arm/tcg/crypto_helper.c b/target/arm/tcg/crypto_helper.c
index 06254939d2..75882d9ea3 100644
--- a/target/arm/tcg/crypto_helper.c
+++ b/target/arm/tcg/crypto_helper.c
@@ -45,11 +45,9 @@ static void clear_tail_16(void *vd, uint32_t desc)
     clear_tail(vd, opr_sz, max_sz);
 }
 
-static void do_crypto_aese(uint64_t *rd, uint64_t *rn,
-                           uint64_t *rm, bool decrypt)
+static void do_crypto_aese(uint64_t *rd, uint64_t *rn, uint64_t *rm,
+                           const uint8_t *sbox, const uint8_t *shift)
 {
-    static uint8_t const * const sbox[2] = { AES_sbox, AES_isbox };
-    static uint8_t const * const shift[2] = { AES_shifts, AES_ishifts };
     union CRYPTO_STATE rk = { .l = { rm[0], rm[1] } };
     union CRYPTO_STATE st = { .l = { rn[0], rn[1] } };
     int i;
@@ -60,7 +58,7 @@ static void do_crypto_aese(uint64_t *rd, uint64_t *rn,
 
     /* combine ShiftRows operation and sbox substitution */
     for (i = 0; i < 16; i++) {
-        CR_ST_BYTE(st, i) = sbox[decrypt][CR_ST_BYTE(rk, shift[decrypt][i])];
+        CR_ST_BYTE(st, i) = sbox[CR_ST_BYTE(rk, shift[i])];
     }
 
     rd[0] = st.l[0];
@@ -70,18 +68,26 @@ static void do_crypto_aese(uint64_t *rd, uint64_t *rn,
 void HELPER(crypto_aese)(void *vd, void *vn, void *vm, uint32_t desc)
 {
     intptr_t i, opr_sz = simd_oprsz(desc);
-    bool decrypt = simd_data(desc);
 
     for (i = 0; i < opr_sz; i += 16) {
-        do_crypto_aese(vd + i, vn + i, vm + i, decrypt);
+        do_crypto_aese(vd + i, vn + i, vm + i, AES_sbox, AES_shifts);
     }
     clear_tail(vd, opr_sz, simd_maxsz(desc));
 }
 
-static void do_crypto_aesmc(uint64_t *rd, uint64_t *rm, bool decrypt)
+void HELPER(crypto_aesd)(void *vd, void *vn, void *vm, uint32_t desc)
+{
+    intptr_t i, opr_sz = simd_oprsz(desc);
+
+    for (i = 0; i < opr_sz; i += 16) {
+        do_crypto_aese(vd + i, vn + i, vm + i, AES_isbox, AES_ishifts);
+    }
+    clear_tail(vd, opr_sz, simd_maxsz(desc));
+}
+
+static void do_crypto_aesmc(uint64_t *rd, uint64_t *rm, const uint32_t *mc)
 {
     union CRYPTO_STATE st = { .l = { rm[0], rm[1] } };
-    const uint32_t *mc = decrypt ? AES_imc_rot : AES_mc_rot;
     int i;
 
     for (i = 0; i < 16; i += 4) {
@@ -99,10 +105,19 @@ static void do_crypto_aesmc(uint64_t *rd, uint64_t *rm, 
bool decrypt)
 void HELPER(crypto_aesmc)(void *vd, void *vm, uint32_t desc)
 {
     intptr_t i, opr_sz = simd_oprsz(desc);
-    bool decrypt = simd_data(desc);
 
     for (i = 0; i < opr_sz; i += 16) {
-        do_crypto_aesmc(vd + i, vm + i, decrypt);
+        do_crypto_aesmc(vd + i, vm + i, AES_mc_rot);
+    }
+    clear_tail(vd, opr_sz, simd_maxsz(desc));
+}
+
+void HELPER(crypto_aesimc)(void *vd, void *vm, uint32_t desc)
+{
+    intptr_t i, opr_sz = simd_oprsz(desc);
+
+    for (i = 0; i < opr_sz; i += 16) {
+        do_crypto_aesmc(vd + i, vm + i, AES_imc_rot);
     }
     clear_tail(vd, opr_sz, simd_maxsz(desc));
 }
diff --git a/target/arm/tcg/translate-a64.c b/target/arm/tcg/translate-a64.c
index aa93f37e21..8b7337ad01 100644
--- a/target/arm/tcg/translate-a64.c
+++ b/target/arm/tcg/translate-a64.c
@@ -13566,7 +13566,6 @@ static void disas_crypto_aes(DisasContext *s, uint32_t 
insn)
     int opcode = extract32(insn, 12, 5);
     int rn = extract32(insn, 5, 5);
     int rd = extract32(insn, 0, 5);
-    int decrypt;
     gen_helper_gvec_2 *genfn2 = NULL;
     gen_helper_gvec_3 *genfn3 = NULL;
 
@@ -13577,20 +13576,16 @@ static void disas_crypto_aes(DisasContext *s, 
uint32_t insn)
 
     switch (opcode) {
     case 0x4: /* AESE */
-        decrypt = 0;
         genfn3 = gen_helper_crypto_aese;
         break;
     case 0x6: /* AESMC */
-        decrypt = 0;
         genfn2 = gen_helper_crypto_aesmc;
         break;
     case 0x5: /* AESD */
-        decrypt = 1;
-        genfn3 = gen_helper_crypto_aese;
+        genfn3 = gen_helper_crypto_aesd;
         break;
     case 0x7: /* AESIMC */
-        decrypt = 1;
-        genfn2 = gen_helper_crypto_aesmc;
+        genfn2 = gen_helper_crypto_aesimc;
         break;
     default:
         unallocated_encoding(s);
@@ -13601,9 +13596,9 @@ static void disas_crypto_aes(DisasContext *s, uint32_t 
insn)
         return;
     }
     if (genfn2) {
-        gen_gvec_op2_ool(s, true, rd, rn, decrypt, genfn2);
+        gen_gvec_op2_ool(s, true, rd, rn, 0, genfn2);
     } else {
-        gen_gvec_op3_ool(s, true, rd, rd, rn, decrypt, genfn3);
+        gen_gvec_op3_ool(s, true, rd, rd, rn, 0, genfn3);
     }
 }
 
diff --git a/target/arm/tcg/translate-neon.c b/target/arm/tcg/translate-neon.c
index 03913de047..8de4ceb203 100644
--- a/target/arm/tcg/translate-neon.c
+++ b/target/arm/tcg/translate-neon.c
@@ -3451,9 +3451,9 @@ static bool trans_VMVN(DisasContext *s, arg_2misc *a)
     }
 
 WRAP_2M_3_OOL_FN(gen_AESE, gen_helper_crypto_aese, 0)
-WRAP_2M_3_OOL_FN(gen_AESD, gen_helper_crypto_aese, 1)
+WRAP_2M_3_OOL_FN(gen_AESD, gen_helper_crypto_aesd, 0)
 WRAP_2M_2_OOL_FN(gen_AESMC, gen_helper_crypto_aesmc, 0)
-WRAP_2M_2_OOL_FN(gen_AESIMC, gen_helper_crypto_aesmc, 1)
+WRAP_2M_2_OOL_FN(gen_AESIMC, gen_helper_crypto_aesimc, 0)
 WRAP_2M_2_OOL_FN(gen_SHA1H, gen_helper_crypto_sha1h, 0)
 WRAP_2M_2_OOL_FN(gen_SHA1SU1, gen_helper_crypto_sha1su1, 0)
 WRAP_2M_2_OOL_FN(gen_SHA256SU0, gen_helper_crypto_sha256su0, 0)
diff --git a/target/arm/tcg/translate-sve.c b/target/arm/tcg/translate-sve.c
index ff050626e6..b98f469cb1 100644
--- a/target/arm/tcg/translate-sve.c
+++ b/target/arm/tcg/translate-sve.c
@@ -7151,12 +7151,14 @@ TRANS_FEAT(USDOT_zzzz, aa64_sve_i8mm, 
gen_gvec_ool_arg_zzzz,
            a->esz == 2 ? gen_helper_gvec_usdot_b : NULL, a, 0)
 
 TRANS_FEAT_NONSTREAMING(AESMC, aa64_sve2_aes, gen_gvec_ool_zz,
-                        gen_helper_crypto_aesmc, a->rd, a->rd, a->decrypt)
+                        gen_helper_crypto_aesmc, a->rd, a->rd, 0)
+TRANS_FEAT_NONSTREAMING(AESIMC, aa64_sve2_aes, gen_gvec_ool_zz,
+                        gen_helper_crypto_aesimc, a->rd, a->rd, 0)
 
 TRANS_FEAT_NONSTREAMING(AESE, aa64_sve2_aes, gen_gvec_ool_arg_zzz,
-                        gen_helper_crypto_aese, a, false)
+                        gen_helper_crypto_aese, a, 0)
 TRANS_FEAT_NONSTREAMING(AESD, aa64_sve2_aes, gen_gvec_ool_arg_zzz,
-                        gen_helper_crypto_aese, a, true)
+                        gen_helper_crypto_aesd, a, 0)
 
 TRANS_FEAT_NONSTREAMING(SM4E, aa64_sve2_sm4, gen_gvec_ool_arg_zzz,
                         gen_helper_crypto_sm4e, a, 0)
-- 
2.34.1