qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
* [PATCH RFC 0/3] target/arm: Implement SVE2 Crypto Extensions
@ 2020-04-23 22:37 Stephen Long
  2020-04-23 22:37 ` [PATCH RFC 1/3] target/arm: Implement SVE2 AESMC, AESIMC Stephen Long
                   ` (2 more replies)
  0 siblings, 3 replies; 5+ messages in thread
From: Stephen Long @ 2020-04-23 22:37 UTC (permalink / raw)
  To: qemu-devel; +Cc: qemu-arm, richard.henderson, apazos

Used the helper functions in crypto_helper.c to implement the helper
functions for the crypto insns.

Stephen Long (3):
  target/arm: Implement SVE2 AESMC, AESIMC
  target/arm: Implement SVE2 AESE, AESD, SM4E
  target/arm: Implement SVE2 SM4EKEY, RAX1

 target/arm/cpu.h           |  5 +++
 target/arm/helper-sve.h    | 10 ++++++
 target/arm/sve.decode      | 20 ++++++++++++
 target/arm/sve_helper.c    | 59 +++++++++++++++++++++++++++++++++++
 target/arm/translate-sve.c | 64 ++++++++++++++++++++++++++++++++++++++
 5 files changed, 158 insertions(+)

-- 
2.17.1



^ permalink raw reply	[flat|nested] 5+ messages in thread

* [PATCH RFC 1/3] target/arm: Implement SVE2 AESMC, AESIMC
  2020-04-23 22:37 [PATCH RFC 0/3] target/arm: Implement SVE2 Crypto Extensions Stephen Long
@ 2020-04-23 22:37 ` Stephen Long
  2020-04-24 21:41   ` Richard Henderson
  2020-04-23 22:37 ` [PATCH RFC 2/3] target/arm: Implement SVE2 AESE, AESD, SM4E Stephen Long
  2020-04-23 22:37 ` [PATCH RFC 3/3] target/arm: Implement SVE2 SM4EKEY, RAX1 Stephen Long
  2 siblings, 1 reply; 5+ messages in thread
From: Stephen Long @ 2020-04-23 22:37 UTC (permalink / raw)
  To: qemu-devel; +Cc: qemu-arm, richard.henderson, apazos

Signed-off-by: Stephen Long <steplong@quicinc.com>
---
 target/arm/helper-sve.h    |  3 +++
 target/arm/sve.decode      | 10 ++++++++++
 target/arm/sve_helper.c    | 13 +++++++++++++
 target/arm/translate-sve.c | 18 ++++++++++++++++++
 4 files changed, 44 insertions(+)

diff --git a/target/arm/helper-sve.h b/target/arm/helper-sve.h
index f6ae814021..6e8421991c 100644
--- a/target/arm/helper-sve.h
+++ b/target/arm/helper-sve.h
@@ -2687,3 +2687,6 @@ DEF_HELPER_FLAGS_5(sve2_sqrdcmlah_zzzz_s, TCG_CALL_NO_RWG,
                    void, ptr, ptr, ptr, ptr, i32)
 DEF_HELPER_FLAGS_5(sve2_sqrdcmlah_zzzz_d, TCG_CALL_NO_RWG,
                    void, ptr, ptr, ptr, ptr, i32)
+
+DEF_HELPER_FLAGS_3(sve2_aesmc, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
+DEF_HELPER_FLAGS_3(sve2_aesimc, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
diff --git a/target/arm/sve.decode b/target/arm/sve.decode
index 3a2a4a7f1c..a83420e690 100644
--- a/target/arm/sve.decode
+++ b/target/arm/sve.decode
@@ -92,6 +92,10 @@
 # Named instruction formats.  These are generally used to
 # reduce the amount of duplication between instruction patterns.
 
+# One operand with unused vector element size
+@rdn_e0         ........ .. ........... . ..... rd:5 \
+                &rr_esz rn=%reg_movprfx esz=0
+
 # Two operand with unused vector element size
 @pd_pn_e0       ........ ........ ....... rn:4 . rd:4           &rr_esz esz=0
 
@@ -1387,3 +1391,9 @@ UMLSLT_zzzw     01000100 .. 0 ..... 010 111 ..... .....  @rda_rn_rm
 
 CMLA_zzzz       01000100 esz:2 0 rm:5 0010 rot:2 rn:5 rd:5  ra=%reg_movprfx
 SQRDCMLAH_zzzz  01000100 esz:2 0 rm:5 0011 rot:2 rn:5 rd:5  ra=%reg_movprfx
+
+#### SVE2 Crypto Extensions
+
+## SVE2 crypto unary operations
+AESMC           01000101 00 10000011100 0 00000 .....   @rdn_e0
+AESIMC          01000101 00 10000011100 1 00000 .....   @rdn_e0
diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c
index 55e2c32f03..f25bb5338d 100644
--- a/target/arm/sve_helper.c
+++ b/target/arm/sve_helper.c
@@ -7428,3 +7428,16 @@ void HELPER(sve2_xar_s)(void *vd, void *vn, void *vm, uint32_t desc)
         d[i] = ror32(n[i] ^ m[i], shr);
     }
 }
+
+#define DO_CRYPTO_AESMC(NAME, DECRYPT)                          \
+void HELPER(NAME)(void *vd, void *vn, uint32_t desc)            \
+{                                                               \
+    intptr_t i, opr_sz = simd_oprsz(desc);                      \
+    void *d = vd, *n = vn;                                      \
+    for (i = 0; i < opr_sz; i += 16) {                          \
+        HELPER(crypto_aesmc)(vd + i, vn + i, DECRYPT);          \
+    }
+}
+
+DO_CRYPTO_AESMC(sve2_aesmc, 0);
+DO_CRYPTO_AESMC(sve2_aesimc, 1);
diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c
index 20eb588cb3..03463308ca 100644
--- a/target/arm/translate-sve.c
+++ b/target/arm/translate-sve.c
@@ -7882,3 +7882,21 @@ static bool trans_SQRDCMLAH_zzzz(DisasContext *s, arg_CMLA_zzzz *a)
     };
     return do_sve2_zzzz_fn(s, a->rd, a->rn, a->rm, a->ra, fns[a->esz], a->rot);
 }
+
+#define DO_SVE2_AES_CRYPTO(NAME, name)                                  \
+static bool trans_##NAME(DisasContext *s, arg_rr_esz *a)                \
+{                                                                       \
+    if (!dc_isar_feature(aa64_sve2_aes, s)) {                           \
+        return false;                                                   \
+    }                                                                   \
+    if (sve_access_check(s)) {                                          \
+        unsigned vsz = vec_full_reg_size(s);                            \
+        tcg_gen_gvec_2_ool(vec_full_reg_offset(s, a->rd),               \
+                           vec_full_reg_offset(s, a->rn),               \
+                           vsz, vsz, 0, gen_helper_sve2_##name);        \
+    }                                                                   \
+    return true;                                                        \
+}
+
+DO_SVE2_AES_CRYPTO(AESMC, aesmc)
+DO_SVE2_AES_CRYPTO(AESIMC, aesimc)
-- 
2.17.1



^ permalink raw reply related	[flat|nested] 5+ messages in thread

* [PATCH RFC 2/3] target/arm: Implement SVE2 AESE, AESD, SM4E
  2020-04-23 22:37 [PATCH RFC 0/3] target/arm: Implement SVE2 Crypto Extensions Stephen Long
  2020-04-23 22:37 ` [PATCH RFC 1/3] target/arm: Implement SVE2 AESMC, AESIMC Stephen Long
@ 2020-04-23 22:37 ` Stephen Long
  2020-04-23 22:37 ` [PATCH RFC 3/3] target/arm: Implement SVE2 SM4EKEY, RAX1 Stephen Long
  2 siblings, 0 replies; 5+ messages in thread
From: Stephen Long @ 2020-04-23 22:37 UTC (permalink / raw)
  To: qemu-devel; +Cc: qemu-arm, richard.henderson, apazos

Signed-off-by: Stephen Long <steplong@quicinc.com>
---
 target/arm/cpu.h           |  5 +++++
 target/arm/helper-sve.h    |  4 ++++
 target/arm/sve.decode      |  6 ++++++
 target/arm/sve_helper.c    | 25 +++++++++++++++++++++++++
 target/arm/translate-sve.c | 16 ++++++++++++++++
 5 files changed, 56 insertions(+)

diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index b7c7946771..4dda0cf6c1 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -3870,6 +3870,11 @@ static inline bool isar_feature_aa64_sve2_bitperm(const ARMISARegisters *id)
     return FIELD_EX64(id->id_aa64zfr0, ID_AA64ZFR0, BITPERM) != 0;
 }
 
+static inline bool isar_feature_aa64_sve2_sm4(const ARMISARegisters *id)
+{
+    return FIELD_EX64(id->id_aa64zfr0, ID_AA64ZFR0, SM4) != 0;
+}
+
 /*
  * Feature tests for "does this exist in either 32-bit or 64-bit?"
  */
diff --git a/target/arm/helper-sve.h b/target/arm/helper-sve.h
index 6e8421991c..3da9590da5 100644
--- a/target/arm/helper-sve.h
+++ b/target/arm/helper-sve.h
@@ -2690,3 +2690,7 @@ DEF_HELPER_FLAGS_5(sve2_sqrdcmlah_zzzz_d, TCG_CALL_NO_RWG,
 
 DEF_HELPER_FLAGS_3(sve2_aesmc, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
 DEF_HELPER_FLAGS_3(sve2_aesimc, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
+
+DEF_HELPER_FLAGS_3(sve2_aese, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
+DEF_HELPER_FLAGS_3(sve2_aesd, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
+DEF_HELPER_FLAGS_3(sve2_sm4e, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
diff --git a/target/arm/sve.decode b/target/arm/sve.decode
index a83420e690..4bbf219cb6 100644
--- a/target/arm/sve.decode
+++ b/target/arm/sve.decode
@@ -98,6 +98,7 @@
 
 # Two operand with unused vector element size
 @pd_pn_e0       ........ ........ ....... rn:4 . rd:4           &rr_esz esz=0
+@pd5_pn5_e0     ........ ........ ...... rn:5 rd:5              &rr_esz esz=0
 
 # Two operand
 @pd_pn          ........ esz:2 .. .... ....... rn:4 . rd:4      &rr_esz
@@ -1397,3 +1398,8 @@ SQRDCMLAH_zzzz  01000100 esz:2 0 rm:5 0011 rot:2 rn:5 rd:5  ra=%reg_movprfx
 ## SVE2 crypto unary operations
 AESMC           01000101 00 10000011100 0 00000 .....   @rdn_e0
 AESIMC          01000101 00 10000011100 1 00000 .....   @rdn_e0
+
+## SVE2 crpyto destructive binary operations
+AESE            01000101 00 10001 0 11100 0 ..... .....  @pd5_pn5_e0
+AESD            01000101 00 10001 0 11100 1 ..... .....  @pd5_pn5_e0
+SM4E            01000101 00 10001 1 11100 0 ..... .....  @pd5_pn5_e0
diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c
index f25bb5338d..0581f33fc7 100644
--- a/target/arm/sve_helper.c
+++ b/target/arm/sve_helper.c
@@ -7441,3 +7441,28 @@ void HELPER(NAME)(void *vd, void *vn, uint32_t desc)            \
 
 DO_CRYPTO_AESMC(sve2_aesmc, 0);
 DO_CRYPTO_AESMC(sve2_aesimc, 1);
+
+#define DO_CRYPTO_AESE(NAME, DECRYPT)                           \
+void HELPER(NAME)(void *vd, void *vn, uint32_t desc)            \
+{                                                               \
+    intptr_t i, opr_sz = simd_oprsz(desc);                      \
+    void *d = vd, *n = vn;                                      \
+    for (i = 0; i < opr_sz; i += 16) {                          \
+        HELPER(crypto_aese)(vd + i, vn + i, DECRYPT);           \
+    }
+}
+
+DO_CRYPTO_AESE(sve2_aese, 0);
+DO_CRYPTO_AESE(sve2_aesd, 1);
+
+#undef DO_CRYPTO_AESE
+#undef DO_CRYPTO_AESMC
+
+void HELPER(sve2_sm4e)(void *vd, void *vn, uint32_t desc)
+{
+    intptr_t i, opr_sz = simd_oprsz(desc);
+    void *d = vd, *n = vn;
+    for (i = 0; i < opr_sz; i += 16) {
+        HELPER(crypto_sm4e)(vd + i, vn + i);
+    }
+}
diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c
index 03463308ca..d991dcdb1c 100644
--- a/target/arm/translate-sve.c
+++ b/target/arm/translate-sve.c
@@ -7900,3 +7900,19 @@ static bool trans_##NAME(DisasContext *s, arg_rr_esz *a)                \
 
 DO_SVE2_AES_CRYPTO(AESMC, aesmc)
 DO_SVE2_AES_CRYPTO(AESIMC, aesimc)
+DO_SVE2_AES_CRYPTO(AESE, aese)
+DO_SVE2_AES_CRYPTO(AESD, aesd)
+
+static bool trans_SM4E(DisasContext *s, arg_rr_esz *a)
+{
+    if (!dc_isar_feature(aa64_sve2_sm4, s)) {
+        return false;
+    }
+    if (sve_access_check(s)) {
+        unsigned vsz = vec_full_reg_size(s);
+        tcg_gen_gvec_2_ool(vec_full_reg_offset(s, a->rd),
+                           vec_full_reg_offset(s, a->rn),
+                           vsz, vsz, 0, gen_helper_sve2_sm4e);
+    }
+    return true;
+}
-- 
2.17.1



^ permalink raw reply related	[flat|nested] 5+ messages in thread

* [PATCH RFC 3/3] target/arm: Implement SVE2 SM4EKEY, RAX1
  2020-04-23 22:37 [PATCH RFC 0/3] target/arm: Implement SVE2 Crypto Extensions Stephen Long
  2020-04-23 22:37 ` [PATCH RFC 1/3] target/arm: Implement SVE2 AESMC, AESIMC Stephen Long
  2020-04-23 22:37 ` [PATCH RFC 2/3] target/arm: Implement SVE2 AESE, AESD, SM4E Stephen Long
@ 2020-04-23 22:37 ` Stephen Long
  2 siblings, 0 replies; 5+ messages in thread
From: Stephen Long @ 2020-04-23 22:37 UTC (permalink / raw)
  To: qemu-devel; +Cc: qemu-arm, richard.henderson, apazos

Signed-off-by: Stephen Long <steplong@quicinc.com>
---
 target/arm/helper-sve.h    |  3 +++
 target/arm/sve.decode      |  4 ++++
 target/arm/sve_helper.c    | 21 +++++++++++++++++++++
 target/arm/translate-sve.c | 30 ++++++++++++++++++++++++++++++
 4 files changed, 58 insertions(+)

diff --git a/target/arm/helper-sve.h b/target/arm/helper-sve.h
index 3da9590da5..07681728ba 100644
--- a/target/arm/helper-sve.h
+++ b/target/arm/helper-sve.h
@@ -2694,3 +2694,6 @@ DEF_HELPER_FLAGS_3(sve2_aesimc, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
 DEF_HELPER_FLAGS_3(sve2_aese, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
 DEF_HELPER_FLAGS_3(sve2_aesd, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
 DEF_HELPER_FLAGS_3(sve2_sm4e, TCG_CALL_NO_RWG, void, ptr, ptr, i32)
+
+DEF_HELPER_FLAGS_4(sve2_sm4ekey, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
+DEF_HELPER_FLAGS_4(sve2_rax1, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32)
diff --git a/target/arm/sve.decode b/target/arm/sve.decode
index 4bbf219cb6..1e98c6aa2f 100644
--- a/target/arm/sve.decode
+++ b/target/arm/sve.decode
@@ -1403,3 +1403,7 @@ AESIMC          01000101 00 10000011100 1 00000 .....   @rdn_e0
 AESE            01000101 00 10001 0 11100 0 ..... .....  @pd5_pn5_e0
 AESD            01000101 00 10001 0 11100 1 ..... .....  @pd5_pn5_e0
 SM4E            01000101 00 10001 1 11100 0 ..... .....  @pd5_pn5_e0
+
+## 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
diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c
index 0581f33fc7..e8299b33ee 100644
--- a/target/arm/sve_helper.c
+++ b/target/arm/sve_helper.c
@@ -7466,3 +7466,24 @@ void HELPER(sve2_sm4e)(void *vd, void *vn, uint32_t desc)
         HELPER(crypto_sm4e)(vd + i, vn + i);
     }
 }
+
+void HELPER(sve2_sm4ekey)(void *vd, void *vn, void *vm, uint32_t desc)
+{
+    intptr_t i, opr_sz = simd_oprsz(desc);
+    void *d = vd, *n = vn;
+    for (i = 0; i < opr_sz; i += 16) {
+        HELPER(crypto_sm4ekey)(vd + i, vn + i, vm + i);
+    }
+}
+
+void HELPER(sve2_rax1)(void *vd, void *vn, void *vm, uint32_t desc)
+{
+    intptr_t i, opr_sz = simd_oprsz(desc) / 8;
+    uint64_t *d = vd, *n = vn, *m = vm;
+
+    for (i = 0; i < opr_sz; ++i) {
+        uint64_t nn = n[i];
+        uint64_t mm = m[i];
+        d[i] = nn ^ rol64(mm, 1);
+    }
+}
diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c
index d991dcdb1c..671f09efa7 100644
--- a/target/arm/translate-sve.c
+++ b/target/arm/translate-sve.c
@@ -7916,3 +7916,33 @@ static bool trans_SM4E(DisasContext *s, arg_rr_esz *a)
     }
     return true;
 }
+
+static bool trans_SM4EKEY(DisasContext *s, arg_rrr_esz *a)
+{
+    if (!dc_isar_feature(aa64_sve2_sm4, s)) {
+        return false;
+    }
+    if (sve_access_check(s)) {
+        unsigned vsz = vec_full_reg_size(s);
+        tcg_gen_gvec_3_ool(vec_full_reg_offset(s, a->rd),
+                           vec_full_reg_offset(s, a->rn),
+                           vec_full_reg_offset(s, a->rm),
+                           vsz, vsz, 0, gen_helper_sve2_sm4ekey);
+    }
+    return true;
+}
+
+static bool trans_RAX1(DisasContext *s, arg_rrr_esz *a)
+{
+    if (!dc_isar_feature(aa64_sve2_sm4, s)) {
+        return false;
+    }
+    if (sve_access_check(s)) {
+        unsigned vsz = vec_full_reg_size(s);
+        tcg_gen_gvec_3_ool(vec_full_reg_offset(s, a->rd),
+                           vec_full_reg_offset(s, a->rn),
+                           vec_full_reg_offset(s, a->rm),
+                           vsz, vsz, 0, gen_helper_sve2_rax1);
+    }
+    return true;
+}
-- 
2.17.1



^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH RFC 1/3] target/arm: Implement SVE2 AESMC, AESIMC
  2020-04-23 22:37 ` [PATCH RFC 1/3] target/arm: Implement SVE2 AESMC, AESIMC Stephen Long
@ 2020-04-24 21:41   ` Richard Henderson
  0 siblings, 0 replies; 5+ messages in thread
From: Richard Henderson @ 2020-04-24 21:41 UTC (permalink / raw)
  To: Stephen Long, qemu-devel; +Cc: qemu-arm, apazos

On 4/23/20 3:37 PM, Stephen Long wrote:
> +#define DO_CRYPTO_AESMC(NAME, DECRYPT)                          \
> +void HELPER(NAME)(void *vd, void *vn, uint32_t desc)            \
> +{                                                               \
> +    intptr_t i, opr_sz = simd_oprsz(desc);                      \
> +    void *d = vd, *n = vn;                                      \
> +    for (i = 0; i < opr_sz; i += 16) {                          \
> +        HELPER(crypto_aesmc)(vd + i, vn + i, DECRYPT);          \
> +    }
> +}

Better, IMO, is to add a "desc" parameter to crypto_aesmc and move the loop
there.  Then all variants of this operation use the exact same helper.  The
separate decrypt parameter would become simd_data().

Same for the other two patches in this series.


r~


^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2020-04-24 21:42 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2020-04-23 22:37 [PATCH RFC 0/3] target/arm: Implement SVE2 Crypto Extensions Stephen Long
2020-04-23 22:37 ` [PATCH RFC 1/3] target/arm: Implement SVE2 AESMC, AESIMC Stephen Long
2020-04-24 21:41   ` Richard Henderson
2020-04-23 22:37 ` [PATCH RFC 2/3] target/arm: Implement SVE2 AESE, AESD, SM4E Stephen Long
2020-04-23 22:37 ` [PATCH RFC 3/3] target/arm: Implement SVE2 SM4EKEY, RAX1 Stephen Long

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).