1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462

```
#[cfg(test)]
use stdarch_test::assert_instr;
extern "unadjusted" {
#[link_name = "llvm.riscv.sm4ed"]
fn _sm4ed(rs1: i32, rs2: i32, bs: i32) -> i32;
#[link_name = "llvm.riscv.sm4ks"]
fn _sm4ks(rs1: i32, rs2: i32, bs: i32) -> i32;
#[link_name = "llvm.riscv.sm3p0"]
fn _sm3p0(rs1: i32) -> i32;
#[link_name = "llvm.riscv.sm3p1"]
fn _sm3p1(rs1: i32) -> i32;
#[link_name = "llvm.riscv.sha256sig0"]
fn _sha256sig0(rs1: i32) -> i32;
#[link_name = "llvm.riscv.sha256sig1"]
fn _sha256sig1(rs1: i32) -> i32;
#[link_name = "llvm.riscv.sha256sum0"]
fn _sha256sum0(rs1: i32) -> i32;
#[link_name = "llvm.riscv.sha256sum1"]
fn _sha256sum1(rs1: i32) -> i32;
}
#[cfg(target_arch = "riscv32")]
extern "unadjusted" {
#[link_name = "llvm.riscv.xperm8.i32"]
fn _xperm8_32(rs1: i32, rs2: i32) -> i32;
#[link_name = "llvm.riscv.xperm4.i32"]
fn _xperm4_32(rs1: i32, rs2: i32) -> i32;
}
#[cfg(target_arch = "riscv64")]
extern "unadjusted" {
#[link_name = "llvm.riscv.xperm8.i64"]
fn _xperm8_64(rs1: i64, rs2: i64) -> i64;
#[link_name = "llvm.riscv.xperm4.i64"]
fn _xperm4_64(rs1: i64, rs2: i64) -> i64;
}
/// Byte-wise lookup of indicies into a vector in registers.
///
/// The xperm8 instruction operates on bytes. The rs1 register contains a vector of XLEN/8
/// 8-bit elements. The rs2 register contains a vector of XLEN/8 8-bit indexes. The result is
/// each element in rs2 replaced by the indexed element in rs1, or zero if the index into rs2
/// is out of bounds.
///
/// Source: RISC-V Cryptography Extensions Volume I: Scalar & Entropy Source Instructions
///
/// Version: v1.0.1
///
/// Section: 3.47
///
/// # Safety
///
/// This function is safe to use if the `zbkx` target feature is present.
#[unstable(feature = "riscv_ext_intrinsics", issue = "114544")]
#[target_feature(enable = "zbkx")]
#[cfg_attr(test, assert_instr(xperm8))]
#[inline]
pub unsafe fn xperm8(rs1: usize, rs2: usize) -> usize {
#[cfg(target_arch = "riscv32")]
{
_xperm8_32(rs1 as i32, rs2 as i32) as usize
}
#[cfg(target_arch = "riscv64")]
{
_xperm8_64(rs1 as i64, rs2 as i64) as usize
}
}
/// Nibble-wise lookup of indicies into a vector.
///
/// The xperm4 instruction operates on nibbles. The rs1 register contains a vector of XLEN/4
/// 4-bit elements. The rs2 register contains a vector of XLEN/4 4-bit indexes. The result is
/// each element in rs2 replaced by the indexed element in rs1, or zero if the index into rs2
/// is out of bounds.
///
/// Source: RISC-V Cryptography Extensions Volume I: Scalar & Entropy Source Instructions
///
/// Version: v1.0.1
///
/// Section: 3.48
///
/// # Safety
///
/// This function is safe to use if the `zbkx` target feature is present.
#[unstable(feature = "riscv_ext_intrinsics", issue = "114544")]
#[target_feature(enable = "zbkx")]
#[cfg_attr(test, assert_instr(xperm4))]
#[inline]
pub unsafe fn xperm4(rs1: usize, rs2: usize) -> usize {
#[cfg(target_arch = "riscv32")]
{
_xperm4_32(rs1 as i32, rs2 as i32) as usize
}
#[cfg(target_arch = "riscv64")]
{
_xperm4_64(rs1 as i64, rs2 as i64) as usize
}
}
/// Implements the Sigma0 transformation function as used in the SHA2-256 hash function \[49\]
/// (Section 4.1.2).
///
/// This instruction is supported for both RV32 and RV64 base architectures. For RV32, the
/// entire XLEN source register is operated on. For RV64, the low 32 bits of the source
/// register are operated on, and the result sign extended to XLEN bits. Though named for
/// SHA2-256, the instruction works for both the SHA2-224 and SHA2-256 parameterisations as
/// described in \[49\]. This instruction must always be implemented such that its execution
/// latency does not depend on the data being operated on.
///
/// Source: RISC-V Cryptography Extensions Volume I: Scalar & Entropy Source Instructions
///
/// Version: v1.0.1
///
/// Section: 3.27
///
/// # Safety
///
/// This function is safe to use if the `zknh` target feature is present.
#[unstable(feature = "riscv_ext_intrinsics", issue = "114544")]
#[target_feature(enable = "zknh")]
#[cfg_attr(test, assert_instr(sha256sig0))]
#[inline]
pub unsafe fn sha256sig0(rs1: u32) -> u32 {
_sha256sig0(rs1 as i32) as u32
}
/// Implements the Sigma1 transformation function as used in the SHA2-256 hash function \[49\]
/// (Section 4.1.2).
///
/// This instruction is supported for both RV32 and RV64 base architectures. For RV32, the
/// entire XLEN source register is operated on. For RV64, the low 32 bits of the source
/// register are operated on, and the result sign extended to XLEN bits. Though named for
/// SHA2-256, the instruction works for both the SHA2-224 and SHA2-256 parameterisations as
/// described in \[49\]. This instruction must always be implemented such that its execution
/// latency does not depend on the data being operated on.
///
/// Source: RISC-V Cryptography Extensions Volume I: Scalar & Entropy Source Instructions
///
/// Version: v1.0.1
///
/// Section: 3.28
///
/// # Safety
///
/// This function is safe to use if the `zknh` target feature is present.
#[unstable(feature = "riscv_ext_intrinsics", issue = "114544")]
#[target_feature(enable = "zknh")]
#[cfg_attr(test, assert_instr(sha256sig1))]
#[inline]
pub unsafe fn sha256sig1(rs1: u32) -> u32 {
_sha256sig1(rs1 as i32) as u32
}
/// Implements the Sum0 transformation function as used in the SHA2-256 hash function \[49\]
/// (Section 4.1.2).
///
/// This instruction is supported for both RV32 and RV64 base architectures. For RV32, the
/// entire XLEN source register is operated on. For RV64, the low 32 bits of the source
/// register are operated on, and the result sign extended to XLEN bits. Though named for
/// SHA2-256, the instruction works for both the SHA2-224 and SHA2-256 parameterisations as
/// described in \[49\]. This instruction must always be implemented such that its execution
/// latency does not depend on the data being operated on.
///
/// Source: RISC-V Cryptography Extensions Volume I: Scalar & Entropy Source Instructions
///
/// Version: v1.0.1
///
/// Section: 3.29
///
/// # Safety
///
/// This function is safe to use if the `zknh` target feature is present.
#[unstable(feature = "riscv_ext_intrinsics", issue = "114544")]
#[target_feature(enable = "zknh")]
#[cfg_attr(test, assert_instr(sha256sum0))]
#[inline]
pub unsafe fn sha256sum0(rs1: u32) -> u32 {
_sha256sum0(rs1 as i32) as u32
}
/// Implements the Sum1 transformation function as used in the SHA2-256 hash function \[49\]
/// (Section 4.1.2).
///
/// This instruction is supported for both RV32 and RV64 base architectures. For RV32, the
/// entire XLEN source register is operated on. For RV64, the low 32 bits of the source
/// register are operated on, and the result sign extended to XLEN bits. Though named for
/// SHA2-256, the instruction works for both the SHA2-224 and SHA2-256 parameterisations as
/// described in \[49\]. This instruction must always be implemented such that its execution
/// latency does not depend on the data being operated on.
///
/// Source: RISC-V Cryptography Extensions Volume I: Scalar & Entropy Source Instructions
///
/// Version: v1.0.1
///
/// Section: 3.30
///
/// # Safety
///
/// This function is safe to use if the `zknh` target feature is present.
#[unstable(feature = "riscv_ext_intrinsics", issue = "114544")]
#[target_feature(enable = "zknh")]
#[cfg_attr(test, assert_instr(sha256sum1))]
#[inline]
pub unsafe fn sha256sum1(rs1: u32) -> u32 {
_sha256sum1(rs1 as i32) as u32
}
/// Accelerates the block encrypt/decrypt operation of the SM4 block cipher \[5, 31\].
///
/// Implements a T-tables in hardware style approach to accelerating the SM4 round function. A
/// byte is extracted from rs2 based on bs, to which the SBox and linear layer transforms are
/// applied, before the result is XOR’d with rs1 and written back to rd. This instruction
/// exists on RV32 and RV64 base architectures. On RV64, the 32-bit result is sign extended to
/// XLEN bits. This instruction must always be implemented such that its execution latency does
/// not depend on the data being operated on.
///
/// Source: RISC-V Cryptography Extensions Volume I: Scalar & Entropy Source Instructions
///
/// Version: v1.0.1
///
/// Section: 3.43
///
/// # Note
///
/// The `BS` parameter is expected to be a constant value and only the bottom 2 bits of `bs` are
/// used.
///
/// # Safety
///
/// This function is safe to use if the `zksed` target feature is present.
///
/// # Details
///
/// Accelerates the round function `F` in the SM4 block cipher algorithm
///
/// This instruction is included in extension `Zksed`. It's defined as:
///
/// ```text
/// SM4ED(x, a, BS) = x ⊕ T(ai)
/// ... where
/// ai = a.bytes[BS]
/// T(ai) = L(τ(ai))
/// bi = τ(ai) = SM4-S-Box(ai)
/// ci = L(bi) = bi ⊕ (bi ≪ 2) ⊕ (bi ≪ 10) ⊕ (bi ≪ 18) ⊕ (bi ≪ 24)
/// SM4ED = (ci ≪ (BS * 8)) ⊕ x
/// ```
///
/// where `⊕` represents 32-bit xor, and `≪ k` represents rotate left by `k` bits.
/// As is defined above, `T` is a combined transformation of non linear S-Box transform `τ`
/// and linear layer transform `L`.
///
/// In the SM4 algorithm, the round function `F` is defined as:
///
/// ```text
/// F(x0, x1, x2, x3, rk) = x0 ⊕ T(x1 ⊕ x2 ⊕ x3 ⊕ rk)
/// ... where
/// T(A) = L(τ(A))
/// B = τ(A) = (SM4-S-Box(a0), SM4-S-Box(a1), SM4-S-Box(a2), SM4-S-Box(a3))
/// C = L(B) = B ⊕ (B ≪ 2) ⊕ (B ≪ 10) ⊕ (B ≪ 18) ⊕ (B ≪ 24)
/// ```
///
/// It can be implemented by `sm4ed` instruction like:
///
/// ```no_run
/// # #[cfg(any(target_arch = "riscv32", target_arch = "riscv64"))]
/// # fn round_function(x0: u32, x1: u32, x2: u32, x3: u32, rk: u32) -> u32 {
/// # #[cfg(target_arch = "riscv32")] use core::arch::riscv32::sm4ed;
/// # #[cfg(target_arch = "riscv64")] use core::arch::riscv64::sm4ed;
/// let a = x1 ^ x2 ^ x3 ^ rk;
/// let c0 = sm4ed(x0, a, 0);
/// let c1 = sm4ed(c0, a, 1); // c1 represents c[0..=1], etc.
/// let c2 = sm4ed(c1, a, 2);
/// let c3 = sm4ed(c2, a, 3);
/// return c3; // c3 represents c[0..=3]
/// # }
/// ```
#[unstable(feature = "riscv_ext_intrinsics", issue = "114544")]
#[target_feature(enable = "zksed")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(sm4ed, BS = 0))]
#[inline]
pub unsafe fn sm4ed<const BS: u8>(rs1: u32, rs2: u32) -> u32 {
static_assert!(BS < 4);
_sm4ed(rs1 as i32, rs2 as i32, BS as i32) as u32
}
/// Accelerates the Key Schedule operation of the SM4 block cipher \[5, 31\] with `bs=0`.
///
/// Implements a T-tables in hardware style approach to accelerating the SM4 Key Schedule. A
/// byte is extracted from rs2 based on bs, to which the SBox and linear layer transforms are
/// applied, before the result is XOR’d with rs1 and written back to rd. This instruction
/// exists on RV32 and RV64 base architectures. On RV64, the 32-bit result is sign extended to
/// XLEN bits. This instruction must always be implemented such that its execution latency does
/// not depend on the data being operated on.
///
/// Source: RISC-V Cryptography Extensions Volume I: Scalar & Entropy Source Instructions
///
/// Version: v1.0.1
///
/// Section: 3.44
///
/// # Note
///
/// The `BS` parameter is expected to be a constant value and only the bottom 2 bits of `bs` are
/// used.
///
/// # Safety
///
/// This function is safe to use if the `zksed` target feature is present.
///
/// # Details
///
/// Accelerates the round function `F` in the SM4 block cipher algorithm
///
/// This instruction is included in extension `Zksed`. It's defined as:
///
/// ```text
/// SM4ED(x, a, BS) = x ⊕ T(ai)
/// ... where
/// ai = a.bytes[BS]
/// T(ai) = L(τ(ai))
/// bi = τ(ai) = SM4-S-Box(ai)
/// ci = L(bi) = bi ⊕ (bi ≪ 2) ⊕ (bi ≪ 10) ⊕ (bi ≪ 18) ⊕ (bi ≪ 24)
/// SM4ED = (ci ≪ (BS * 8)) ⊕ x
/// ```
///
/// where `⊕` represents 32-bit xor, and `≪ k` represents rotate left by `k` bits.
/// As is defined above, `T` is a combined transformation of non linear S-Box transform `τ`
/// and linear layer transform `L`.
///
/// In the SM4 algorithm, the round function `F` is defined as:
///
/// ```text
/// F(x0, x1, x2, x3, rk) = x0 ⊕ T(x1 ⊕ x2 ⊕ x3 ⊕ rk)
/// ... where
/// T(A) = L(τ(A))
/// B = τ(A) = (SM4-S-Box(a0), SM4-S-Box(a1), SM4-S-Box(a2), SM4-S-Box(a3))
/// C = L(B) = B ⊕ (B ≪ 2) ⊕ (B ≪ 10) ⊕ (B ≪ 18) ⊕ (B ≪ 24)
/// ```
///
/// It can be implemented by `sm4ed` instruction like:
///
/// ```no_run
/// # #[cfg(any(target_arch = "riscv32", target_arch = "riscv64"))]
/// # fn round_function(x0: u32, x1: u32, x2: u32, x3: u32, rk: u32) -> u32 {
/// # #[cfg(target_arch = "riscv32")] use core::arch::riscv32::sm4ed;
/// # #[cfg(target_arch = "riscv64")] use core::arch::riscv64::sm4ed;
/// let a = x1 ^ x2 ^ x3 ^ rk;
/// let c0 = sm4ed(x0, a, 0);
/// let c1 = sm4ed(c0, a, 1); // c1 represents c[0..=1], etc.
/// let c2 = sm4ed(c1, a, 2);
/// let c3 = sm4ed(c2, a, 3);
/// return c3; // c3 represents c[0..=3]
/// # }
/// ```
#[unstable(feature = "riscv_ext_intrinsics", issue = "114544")]
#[target_feature(enable = "zksed")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(sm4ks, BS = 0))]
#[inline]
pub unsafe fn sm4ks<const BS: u8>(rs1: u32, rs2: u32) -> u32 {
static_assert!(BS < 4);
_sm4ks(rs1 as i32, rs2 as i32, BS as i32) as u32
}
/// Implements the P0 transformation function as used in the SM3 hash function [4, 30].
///
/// This instruction is supported for the RV32 and RV64 base architectures. It implements the
/// P0 transform of the SM3 hash function [4, 30]. This instruction must always be implemented
/// such that its execution latency does not depend on the data being operated on.
///
/// Source: RISC-V Cryptography Extensions Volume I: Scalar & Entropy Source Instructions
///
/// Version: v1.0.1
///
/// Section: 3.41
///
/// # Safety
///
/// This function is safe to use if the `zksh` target feature is present.
///
/// # Details
///
/// `P0` transformation function as is used in the SM3 hash algorithm
///
/// This function is included in `Zksh` extension. It's defined as:
///
/// ```text
/// P0(X) = X ⊕ (X ≪ 9) ⊕ (X ≪ 17)
/// ```
///
/// where `⊕` represents 32-bit xor, and `≪ k` represents rotate left by `k` bits.
///
/// In the SM3 algorithm, the `P0` transformation is used as `E ← P0(TT2)` when the
/// compression function `CF` uses the intermediate value `TT2` to calculate
/// the variable `E` in one iteration for subsequent processes.
#[unstable(feature = "riscv_ext_intrinsics", issue = "114544")]
#[target_feature(enable = "zksh")]
#[cfg_attr(test, assert_instr(sm3p0))]
#[inline]
pub unsafe fn sm3p0(rs1: u32) -> u32 {
_sm3p0(rs1 as i32) as u32
}
/// Implements the P1 transformation function as used in the SM3 hash function [4, 30].
///
/// This instruction is supported for the RV32 and RV64 base architectures. It implements the
/// P1 transform of the SM3 hash function [4, 30]. This instruction must always be implemented
/// such that its execution latency does not depend on the data being operated on.
///
/// Source: RISC-V Cryptography Extensions Volume I: Scalar & Entropy Source Instructions
///
/// Version: v1.0.1
///
/// Section: 3.42
///
/// # Safety
///
/// This function is safe to use if the `zksh` target feature is present.
///
/// # Details
///
/// `P1` transformation function as is used in the SM3 hash algorithm
///
/// This function is included in `Zksh` extension. It's defined as:
///
/// ```text
/// P1(X) = X ⊕ (X ≪ 15) ⊕ (X ≪ 23)
/// ```
///
/// where `⊕` represents 32-bit xor, and `≪ k` represents rotate left by `k` bits.
///
/// In the SM3 algorithm, the `P1` transformation is used to expand message,
/// where expanded word `Wj` can be generated from the previous words.
/// The whole process can be described as the following pseudocode:
///
/// ```text
/// FOR j=16 TO 67
/// Wj ← P1(Wj−16 ⊕ Wj−9 ⊕ (Wj−3 ≪ 15)) ⊕ (Wj−13 ≪ 7) ⊕ Wj−6
/// ENDFOR
/// ```
#[unstable(feature = "riscv_ext_intrinsics", issue = "114544")]
#[target_feature(enable = "zksh")]
#[cfg_attr(test, assert_instr(sm3p1))]
#[inline]
pub unsafe fn sm3p1(rs1: u32) -> u32 {
_sm3p1(rs1 as i32) as u32
}
```