From: Max Chou <max.chou@sifive.com>
To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org
Cc: dbarboza@ventanamicro.com,
Lawrence Hunter <lawrence.hunter@codethink.co.uk>,
Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>,
Max Chou <max.chou@sifive.com>,
Palmer Dabbelt <palmer@dabbelt.com>,
Alistair Francis <alistair.francis@wdc.com>,
Bin Meng <bin.meng@windriver.com>,
Weiwei Li <liweiwei@iscas.ac.cn>,
Liu Zhiwei <zhiwei_liu@linux.alibaba.com>,
Nazar Kazakov <nazar.kazakov@codethink.co.uk>,
William Salmon <will.salmon@codethink.co.uk>
Subject: [PATCH v4 12/17] target/riscv: Add Zvksh ISA extension support
Date: Fri, 23 Jun 2023 00:16:28 +0800 [thread overview]
Message-ID: <20230622161646.32005-13-max.chou@sifive.com> (raw)
In-Reply-To: <20230622161646.32005-1-max.chou@sifive.com>
From: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
This commit adds support for the Zvksh vector-crypto extension, which
consists of the following instructions:
* vsm3me.vv
* vsm3c.vi
Translation functions are defined in
`target/riscv/insn_trans/trans_rvvk.c.inc` and helpers are defined in
`target/riscv/vcrypto_helper.c`.
Co-authored-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
[max.chou@sifive.com: Replaced vstart checking by TCG op]
Signed-off-by: Kiran Ostrolenk <kiran.ostrolenk@codethink.co.uk>
Signed-off-by: Lawrence Hunter <lawrence.hunter@codethink.co.uk>
Signed-off-by: Max Chou <max.chou@sifive.com>
---
target/riscv/cpu.c | 5 +-
target/riscv/cpu_cfg.h | 1 +
target/riscv/helper.h | 3 +
target/riscv/insn32.decode | 4 +
target/riscv/insn_trans/trans_rvvk.c.inc | 31 ++++++
target/riscv/vcrypto_helper.c | 134 +++++++++++++++++++++++
6 files changed, 176 insertions(+), 2 deletions(-)
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index 6bba8ba8c9..c9a9ff80cd 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -121,6 +121,7 @@ static const struct isa_ext_data isa_edata_arr[] = {
ISA_EXT_DATA_ENTRY(zvkned, PRIV_VERSION_1_12_0, ext_zvkned),
ISA_EXT_DATA_ENTRY(zvknha, PRIV_VERSION_1_12_0, ext_zvknha),
ISA_EXT_DATA_ENTRY(zvknhb, PRIV_VERSION_1_12_0, ext_zvknhb),
+ ISA_EXT_DATA_ENTRY(zvksh, PRIV_VERSION_1_12_0, ext_zvksh),
ISA_EXT_DATA_ENTRY(zhinx, PRIV_VERSION_1_12_0, ext_zhinx),
ISA_EXT_DATA_ENTRY(zhinxmin, PRIV_VERSION_1_12_0, ext_zhinxmin),
ISA_EXT_DATA_ENTRY(smaia, PRIV_VERSION_1_12_0, ext_smaia),
@@ -1197,8 +1198,8 @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp)
* In principle Zve*x would also suffice here, were they supported
* in qemu
*/
- if ((cpu->cfg.ext_zvbb || cpu->cfg.ext_zvkned || cpu->cfg.ext_zvknha) &&
- !cpu->cfg.ext_zve32f) {
+ if ((cpu->cfg.ext_zvbb || cpu->cfg.ext_zvkned || cpu->cfg.ext_zvknha ||
+ cpu->cfg.ext_zvksh) && !cpu->cfg.ext_zve32f) {
error_setg(errp,
"Vector crypto extensions require V or Zve* extensions");
return;
diff --git a/target/riscv/cpu_cfg.h b/target/riscv/cpu_cfg.h
index 41cce87ffc..f859d9e2f5 100644
--- a/target/riscv/cpu_cfg.h
+++ b/target/riscv/cpu_cfg.h
@@ -88,6 +88,7 @@ struct RISCVCPUConfig {
bool ext_zvkned;
bool ext_zvknha;
bool ext_zvknhb;
+ bool ext_zvksh;
bool ext_zmmul;
bool ext_zvfh;
bool ext_zvfhmin;
diff --git a/target/riscv/helper.h b/target/riscv/helper.h
index 19f5a8a28d..9220af18e6 100644
--- a/target/riscv/helper.h
+++ b/target/riscv/helper.h
@@ -1238,3 +1238,6 @@ DEF_HELPER_5(vaeskf2_vi, void, ptr, ptr, i32, env, i32)
DEF_HELPER_5(vsha2ms_vv, void, ptr, ptr, ptr, env, i32)
DEF_HELPER_5(vsha2ch_vv, void, ptr, ptr, ptr, env, i32)
DEF_HELPER_5(vsha2cl_vv, void, ptr, ptr, ptr, env, i32)
+
+DEF_HELPER_5(vsm3me_vv, void, ptr, ptr, ptr, env, i32)
+DEF_HELPER_5(vsm3c_vi, void, ptr, ptr, i32, env, i32)
diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode
index d2cfb2729c..5ca83e8462 100644
--- a/target/riscv/insn32.decode
+++ b/target/riscv/insn32.decode
@@ -953,3 +953,7 @@ vaeskf2_vi 101010 1 ..... ..... 010 ..... 1110111 @r_vm_1
vsha2ms_vv 101101 1 ..... ..... 010 ..... 1110111 @r_vm_1
vsha2ch_vv 101110 1 ..... ..... 010 ..... 1110111 @r_vm_1
vsha2cl_vv 101111 1 ..... ..... 010 ..... 1110111 @r_vm_1
+
+# *** Zvksh vector crypto extension ***
+vsm3me_vv 100000 1 ..... ..... 010 ..... 1110111 @r_vm_1
+vsm3c_vi 101011 1 ..... ..... 010 ..... 1110111 @r_vm_1
diff --git a/target/riscv/insn_trans/trans_rvvk.c.inc b/target/riscv/insn_trans/trans_rvvk.c.inc
index 528a0d3b32..af1fb74c38 100644
--- a/target/riscv/insn_trans/trans_rvvk.c.inc
+++ b/target/riscv/insn_trans/trans_rvvk.c.inc
@@ -479,3 +479,34 @@ static bool vsha_check(DisasContext *s, arg_rmrr *a)
GEN_VV_UNMASKED_TRANS(vsha2ms_vv, vsha_check, ZVKNH_EGS)
GEN_VV_UNMASKED_TRANS(vsha2cl_vv, vsha_check, ZVKNH_EGS)
GEN_VV_UNMASKED_TRANS(vsha2ch_vv, vsha_check, ZVKNH_EGS)
+
+/*
+ * Zvksh
+ */
+
+#define ZVKSH_EGS 8
+
+static inline bool vsm3_check(DisasContext *s, arg_rmrr *a)
+{
+ int egw_bytes = ZVKSH_EGS << s->sew;
+ int mult = 1 << MAX(s->lmul, 0);
+ return s->cfg_ptr->ext_zvksh == true &&
+ require_rvv(s) &&
+ vext_check_isa_ill(s) &&
+ !is_overlapped(a->rd, mult, a->rs2, mult) &&
+ MAXSZ(s) >= egw_bytes &&
+ s->sew == MO_32;
+}
+
+static inline bool vsm3me_check(DisasContext *s, arg_rmrr *a)
+{
+ return vsm3_check(s, a) && vext_check_sss(s, a->rd, a->rs1, a->rs2, a->vm);
+}
+
+static inline bool vsm3c_check(DisasContext *s, arg_rmrr *a)
+{
+ return vsm3_check(s, a) && vext_check_ss(s, a->rd, a->rs2, a->vm);
+}
+
+GEN_VV_UNMASKED_TRANS(vsm3me_vv, vsm3me_check, ZVKSH_EGS)
+GEN_VI_UNMASKED_TRANS(vsm3c_vi, vsm3c_check, ZVKSH_EGS)
diff --git a/target/riscv/vcrypto_helper.c b/target/riscv/vcrypto_helper.c
index ca09062c6c..06c8f4adc7 100644
--- a/target/riscv/vcrypto_helper.c
+++ b/target/riscv/vcrypto_helper.c
@@ -717,3 +717,137 @@ void HELPER(vsha2cl_vv)(void *vd, void *vs1, void *vs2, CPURISCVState *env,
vext_set_elems_1s(vd, vta, env->vl * esz, total_elems * esz);
env->vstart = 0;
}
+
+static inline uint32_t p1(uint32_t x)
+{
+ return x ^ rol32(x, 15) ^ rol32(x, 23);
+}
+
+static inline uint32_t zvksh_w(uint32_t m16, uint32_t m9, uint32_t m3,
+ uint32_t m13, uint32_t m6)
+{
+ return p1(m16 ^ m9 ^ rol32(m3, 15)) ^ rol32(m13, 7) ^ m6;
+}
+
+void HELPER(vsm3me_vv)(void *vd_vptr, void *vs1_vptr, void *vs2_vptr,
+ CPURISCVState *env, uint32_t desc)
+{
+ uint32_t esz = memop_size(FIELD_EX64(env->vtype, VTYPE, VSEW));
+ uint32_t total_elems = vext_get_total_elems(env, desc, esz);
+ uint32_t vta = vext_vta(desc);
+ uint32_t *vd = vd_vptr;
+ uint32_t *vs1 = vs1_vptr;
+ uint32_t *vs2 = vs2_vptr;
+
+ for (int i = env->vstart / 8; i < env->vl / 8; i++) {
+ uint32_t w[24];
+ for (int j = 0; j < 8; j++) {
+ w[j] = bswap32(vs1[H4((i * 8) + j)]);
+ w[j + 8] = bswap32(vs2[H4((i * 8) + j)]);
+ }
+ for (int j = 0; j < 8; j++) {
+ w[j + 16] =
+ zvksh_w(w[j], w[j + 7], w[j + 13], w[j + 3], w[j + 10]);
+ }
+ for (int j = 0; j < 8; j++) {
+ vd[(i * 8) + j] = bswap32(w[H4(j + 16)]);
+ }
+ }
+ vext_set_elems_1s(vd_vptr, vta, env->vl * esz, total_elems * esz);
+ env->vstart = 0;
+}
+
+static inline uint32_t ff1(uint32_t x, uint32_t y, uint32_t z)
+{
+ return x ^ y ^ z;
+}
+
+static inline uint32_t ff2(uint32_t x, uint32_t y, uint32_t z)
+{
+ return (x & y) | (x & z) | (y & z);
+}
+
+static inline uint32_t ff_j(uint32_t x, uint32_t y, uint32_t z, uint32_t j)
+{
+ return (j <= 15) ? ff1(x, y, z) : ff2(x, y, z);
+}
+
+static inline uint32_t gg1(uint32_t x, uint32_t y, uint32_t z)
+{
+ return x ^ y ^ z;
+}
+
+static inline uint32_t gg2(uint32_t x, uint32_t y, uint32_t z)
+{
+ return (x & y) | (~x & z);
+}
+
+static inline uint32_t gg_j(uint32_t x, uint32_t y, uint32_t z, uint32_t j)
+{
+ return (j <= 15) ? gg1(x, y, z) : gg2(x, y, z);
+}
+
+static inline uint32_t t_j(uint32_t j)
+{
+ return (j <= 15) ? 0x79cc4519 : 0x7a879d8a;
+}
+
+static inline uint32_t p_0(uint32_t x)
+{
+ return x ^ rol32(x, 9) ^ rol32(x, 17);
+}
+
+static void sm3c(uint32_t *vd, uint32_t *vs1, uint32_t *vs2, uint32_t uimm)
+{
+ uint32_t x0, x1;
+ uint32_t j;
+ uint32_t ss1, ss2, tt1, tt2;
+ x0 = vs2[0] ^ vs2[4];
+ x1 = vs2[1] ^ vs2[5];
+ j = 2 * uimm;
+ ss1 = rol32(rol32(vs1[0], 12) + vs1[4] + rol32(t_j(j), j % 32), 7);
+ ss2 = ss1 ^ rol32(vs1[0], 12);
+ tt1 = ff_j(vs1[0], vs1[1], vs1[2], j) + vs1[3] + ss2 + x0;
+ tt2 = gg_j(vs1[4], vs1[5], vs1[6], j) + vs1[7] + ss1 + vs2[0];
+ vs1[3] = vs1[2];
+ vd[3] = rol32(vs1[1], 9);
+ vs1[1] = vs1[0];
+ vd[1] = tt1;
+ vs1[7] = vs1[6];
+ vd[7] = rol32(vs1[5], 19);
+ vs1[5] = vs1[4];
+ vd[5] = p_0(tt2);
+ j = 2 * uimm + 1;
+ ss1 = rol32(rol32(vd[1], 12) + vd[5] + rol32(t_j(j), j % 32), 7);
+ ss2 = ss1 ^ rol32(vd[1], 12);
+ tt1 = ff_j(vd[1], vs1[1], vd[3], j) + vs1[3] + ss2 + x1;
+ tt2 = gg_j(vd[5], vs1[5], vd[7], j) + vs1[7] + ss1 + vs2[1];
+ vd[2] = rol32(vs1[1], 9);
+ vd[0] = tt1;
+ vd[6] = rol32(vs1[5], 19);
+ vd[4] = p_0(tt2);
+}
+
+void HELPER(vsm3c_vi)(void *vd_vptr, void *vs2_vptr, uint32_t uimm,
+ CPURISCVState *env, uint32_t desc)
+{
+ uint32_t esz = memop_size(FIELD_EX64(env->vtype, VTYPE, VSEW));
+ uint32_t total_elems = vext_get_total_elems(env, desc, esz);
+ uint32_t vta = vext_vta(desc);
+ uint32_t *vd = vd_vptr;
+ uint32_t *vs2 = vs2_vptr;
+ uint32_t v1[8], v2[8], v3[8];
+
+ for (int i = env->vstart / 8; i < env->vl / 8; i++) {
+ for (int k = 0; k < 8; k++) {
+ v2[k] = bswap32(vd[H4(i * 8 + k)]);
+ v3[k] = bswap32(vs2[H4(i * 8 + k)]);
+ }
+ sm3c(v1, v2, v3, uimm);
+ for (int k = 0; k < 8; k++) {
+ vd[i * 8 + k] = bswap32(v1[H4(k)]);
+ }
+ }
+ vext_set_elems_1s(vd_vptr, vta, env->vl * esz, total_elems * esz);
+ env->vstart = 0;
+}
--
2.31.1
next prev parent reply other threads:[~2023-06-22 16:20 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-06-22 16:16 [PATCH v4 00/17] Add RISC-V vector cryptographic instruction set support Max Chou
2023-06-22 16:16 ` [PATCH v4 01/17] target/riscv: Refactor some of the generic vector functionality Max Chou
2023-06-22 16:16 ` [PATCH v4 02/17] target/riscv: Refactor vector-vector translation macro Max Chou
2023-06-22 16:16 ` [PATCH v4 03/17] target/riscv: Remove redundant "cpu_vl == 0" checks Max Chou
2023-06-22 16:16 ` [PATCH v4 04/17] target/riscv: Add Zvbc ISA extension support Max Chou
2023-06-22 16:16 ` [PATCH v4 05/17] target/riscv: Move vector translation checks Max Chou
2023-06-22 16:16 ` [PATCH v4 06/17] target/riscv: Refactor translation of vector-widening instruction Max Chou
2023-06-22 16:16 ` [PATCH v4 07/17] target/riscv: Refactor some of the generic vector functionality Max Chou
2023-06-22 16:16 ` [PATCH v4 08/17] tcg: Fix temporary variable in tcg_gen_gvec_andcs Max Chou
2023-06-22 17:30 ` Daniel Henrique Barboza
2023-06-23 6:51 ` Richard Henderson
2023-06-22 16:16 ` [PATCH v4 09/17] target/riscv: Add Zvbb ISA extension support Max Chou
2023-06-22 17:48 ` Daniel Henrique Barboza
2023-06-22 16:16 ` [PATCH v4 10/17] target/riscv: Add Zvkned " Max Chou
2023-06-22 18:03 ` Daniel Henrique Barboza
2023-06-23 7:33 ` Richard Henderson
2023-06-26 8:02 ` Max Chou
2023-06-22 16:16 ` [PATCH v4 11/17] target/riscv: Add Zvknh " Max Chou
2023-06-22 18:06 ` Daniel Henrique Barboza
2023-06-22 16:16 ` Max Chou [this message]
2023-06-22 18:09 ` [PATCH v4 12/17] target/riscv: Add Zvksh " Daniel Henrique Barboza
2023-06-22 16:16 ` [PATCH v4 13/17] target/riscv: Add Zvkg " Max Chou
2023-06-22 18:10 ` Daniel Henrique Barboza
2023-06-22 16:16 ` [PATCH v4 14/17] crypto: Create sm4_subword Max Chou
2023-06-22 16:16 ` [PATCH v4 15/17] crypto: Add SM4 constant parameter CK Max Chou
2023-06-22 16:16 ` [PATCH v4 16/17] target/riscv: Add Zvksed ISA extension support Max Chou
2023-06-22 16:16 ` [PATCH v4 17/17] target/riscv: Expose Zvk* and Zvb[b, c] cpu properties Max Chou via
2023-06-22 17:41 ` [PATCH v4 17/17] target/riscv: Expose Zvk* and Zvb[b,c] " Daniel Henrique Barboza
2023-06-26 8:08 ` Max Chou
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230622161646.32005-13-max.chou@sifive.com \
--to=max.chou@sifive.com \
--cc=alistair.francis@wdc.com \
--cc=bin.meng@windriver.com \
--cc=dbarboza@ventanamicro.com \
--cc=kiran.ostrolenk@codethink.co.uk \
--cc=lawrence.hunter@codethink.co.uk \
--cc=liweiwei@iscas.ac.cn \
--cc=nazar.kazakov@codethink.co.uk \
--cc=palmer@dabbelt.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-riscv@nongnu.org \
--cc=will.salmon@codethink.co.uk \
--cc=zhiwei_liu@linux.alibaba.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).