Add a new file cpacf_sha256.c which implements sha256. Add support for the sha256 subfuction for CPACF kimd and klmd.
Signed-off-by: Harald Freudenberger <[email protected]> --- target/s390x/gen-features.c | 2 + target/s390x/tcg/cpacf.h | 4 + target/s390x/tcg/cpacf_sha256.c | 229 +++++++++++++++++++++++++++++++ target/s390x/tcg/crypto_helper.c | 8 ++ target/s390x/tcg/meson.build | 1 + 5 files changed, 244 insertions(+) create mode 100644 target/s390x/tcg/cpacf_sha256.c diff --git a/target/s390x/gen-features.c b/target/s390x/gen-features.c index 8218e6470e..5cf5b92c37 100644 --- a/target/s390x/gen-features.c +++ b/target/s390x/gen-features.c @@ -916,7 +916,9 @@ static uint16_t qemu_V7_1[] = { */ static uint16_t qemu_MAX[] = { S390_FEAT_MSA_EXT_5, + S390_FEAT_KIMD_SHA_256, S390_FEAT_KIMD_SHA_512, + S390_FEAT_KLMD_SHA_256, S390_FEAT_KLMD_SHA_512, S390_FEAT_PRNO_TRNG, }; diff --git a/target/s390x/tcg/cpacf.h b/target/s390x/tcg/cpacf.h index 99b8a11232..79f05e1e14 100644 --- a/target/s390x/tcg/cpacf.h +++ b/target/s390x/tcg/cpacf.h @@ -8,6 +8,10 @@ #ifndef S390X_CPACF_H #define S390X_CPACF_H +/* from crypto_sha256.c */ +int cpacf_sha256(CPUS390XState *env, uintptr_t ra, uint64_t param_addr, + uint64_t *message_reg, uint64_t *len_reg, uint32_t type); + /* from crypto_sha512.c */ int cpacf_sha512(CPUS390XState *env, uintptr_t ra, uint64_t param_addr, uint64_t *message_reg, uint64_t *len_reg, uint32_t type); diff --git a/target/s390x/tcg/cpacf_sha256.c b/target/s390x/tcg/cpacf_sha256.c new file mode 100644 index 0000000000..5190aef6fa --- /dev/null +++ b/target/s390x/tcg/cpacf_sha256.c @@ -0,0 +1,229 @@ +/* + * s390 cpacf sha256 + * + * Authors: + * Harald Freudenberger <[email protected]> + * + * The sha256 implementation here is more or less a copy-and-paste + * from Jason A. Donenfeld's implementation of sha 512 with adaptions + * for sha 256. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#include "qemu/osdep.h" +#include "s390x-internal.h" +#include "tcg_s390x.h" +#include "accel/tcg/cpu-ldst.h" +#include "cpacf.h" + +static uint32_t R(uint32_t x, int c) +{ + return (x >> c) | (x << (32 - c)); +} +static uint32_t Ch(uint32_t x, uint32_t y, uint32_t z) +{ + return (x & y) ^ (~x & z); +} +static uint32_t Maj(uint32_t x, uint32_t y, uint32_t z) +{ + return (x & y) ^ (x & z) ^ (y & z); +} +static uint32_t Sigma0(uint32_t x) +{ + return R(x, 2) ^ R(x, 13) ^ R(x, 22); +} +static uint32_t Sigma1(uint32_t x) +{ + return R(x, 6) ^ R(x, 11) ^ R(x, 25); +} +static uint32_t sigma0(uint32_t x) +{ + return R(x, 7) ^ R(x, 18) ^ (x >> 3); +} +static uint32_t sigma1(uint32_t x) +{ + return R(x, 17) ^ R(x, 19) ^ (x >> 10); +} + +static const uint32_t K[64] = { + 0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5, 0x3956c25b, 0x59f111f1, + 0x923f82a4, 0xab1c5ed5, 0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3, + 0x72be5d74, 0x80deb1fe, 0x9bdc06a7, 0xc19bf174, 0xe49b69c1, 0xefbe4786, + 0x0fc19dc6, 0x240ca1cc, 0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da, + 0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7, 0xc6e00bf3, 0xd5a79147, + 0x06ca6351, 0x14292967, 0x27b70a85, 0x2e1b2138, 0x4d2c6dfc, 0x53380d13, + 0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85, 0xa2bfe8a1, 0xa81a664b, + 0xc24b8b70, 0xc76c51a3, 0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070, + 0x19a4c116, 0x1e376c08, 0x2748774c, 0x34b0bcb5, 0x391c0cb3, 0x4ed8aa4a, + 0x5b9cca4f, 0x682e6ff3, 0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208, + 0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2, +}; + +/* a is icv/ocv, w is a single message block. w will get reused internally. */ +static void sha256_bda(uint32_t a[8], uint32_t w[16]) +{ + uint32_t t, z[8], b[8]; + int i, j; + + memcpy(z, a, sizeof(z)); + for (i = 0; i < 64; i++) { + memcpy(b, a, sizeof(b)); + + t = a[7] + Sigma1(a[4]) + Ch(a[4], a[5], a[6]) + K[i] + w[i % 16]; + b[7] = t + Sigma0(a[0]) + Maj(a[0], a[1], a[2]); + b[3] += t; + for (j = 0; j < 8; ++j) { + a[(j + 1) % 8] = b[j]; + } + if (i % 16 == 15) { + for (j = 0; j < 16; ++j) { + w[j] += w[(j + 9) % 16] + sigma0(w[(j + 1) % 16]) + + sigma1(w[(j + 14) % 16]); + } + } + } + + for (i = 0; i < 8; i++) { + a[i] += z[i]; + } +} + +/* a is icv/ocv, w is a single message block that needs be32 conversion. */ +static void sha256_bda_be32(uint32_t a[8], uint32_t w[16]) +{ + uint32_t t[16]; + int i; + + for (i = 0; i < 16; i++) { + t[i] = be32_to_cpu(w[i]); + } + sha256_bda(a, t); +} + +static void sha256_read_icv(CPUS390XState *env, uint64_t addr, + uint32_t a[8], uintptr_t ra) +{ + int i; + + for (i = 0; i < 8; i++, addr += 4) { + addr = wrap_address(env, addr); + a[i] = cpu_ldl_be_data_ra(env, addr, ra); + } +} + +static void sha256_write_ocv(CPUS390XState *env, uint64_t addr, + uint32_t a[8], uintptr_t ra) +{ + int i; + + for (i = 0; i < 8; i++, addr += 4) { + addr = wrap_address(env, addr); + cpu_stl_be_data_ra(env, addr, a[i], ra); + } +} + +static void sha256_read_block(CPUS390XState *env, uint64_t addr, + uint32_t a[16], uintptr_t ra) +{ + int i; + + for (i = 0; i < 16; i++, addr += 4) { + addr = wrap_address(env, addr); + a[i] = cpu_ldl_be_data_ra(env, addr, ra); + } +} + +static void sha256_read_mbl(CPUS390XState *env, uint64_t addr, + uint8_t a[8], uintptr_t ra) +{ + int i; + + for (i = 0; i < 8; i++, addr += 1) { + addr = wrap_address(env, addr); + a[i] = cpu_ldub_data_ra(env, addr, ra); + } +} + +int cpacf_sha256(CPUS390XState *env, uintptr_t ra, uint64_t param_addr, + uint64_t *message_reg, uint64_t *len_reg, uint32_t type) +{ + enum { MAX_BLOCKS_PER_RUN = 128 }; /* 128 * 64 = 8K */ + uint64_t len = *len_reg, processed = 0; + int i, message_reg_len = 64; + uint32_t a[8]; + + g_assert(type == S390_FEAT_TYPE_KIMD || type == S390_FEAT_TYPE_KLMD); + + if (!(env->psw.mask & PSW_MASK_64)) { + len = (uint32_t)len; + message_reg_len = (env->psw.mask & PSW_MASK_32) ? 32 : 24; + } + + /* KIMD: length has to be properly aligned. */ + if (type == S390_FEAT_TYPE_KIMD && !QEMU_IS_ALIGNED(len, 64)) { + tcg_s390_program_interrupt(env, PGM_SPECIFICATION, ra); + } + + sha256_read_icv(env, param_addr, a, ra); + + /* Process full blocks first. */ + for (; len >= 64; len -= 64, processed += 64) { + uint32_t w[16]; + + if (processed >= MAX_BLOCKS_PER_RUN * 64) { + break; + } + + sha256_read_block(env, *message_reg + processed, w, ra); + sha256_bda(a, w); + } + + /* KLMD: Process partial/empty block last. */ + if (type == S390_FEAT_TYPE_KLMD && len < 64) { + uint8_t x[64]; + + /* Read the remainder of the message byte-per-byte. */ + for (i = 0; i < len; i++) { + uint64_t addr = wrap_address(env, *message_reg + processed + i); + + x[i] = cpu_ldub_data_ra(env, addr, ra); + } + /* Pad the remainder with zero and set the top bit. */ + memset(x + len, 0, 64 - len); + x[len] = 0x80; + + /* + * Place the MBL either into this block (if there is space left), + * or use an additional one. + */ + if (len < 56) { + sha256_read_mbl(env, param_addr + 32, x + 56, ra); + } + sha256_bda_be32(a, (uint32_t *)x); + + if (len >= 56) { + memset(x, 0, 56); + sha256_read_mbl(env, param_addr + 32, x + 56, ra); + sha256_bda_be32(a, (uint32_t *)x); + } + + processed += len; + len = 0; + } + + /* + * Modify memory after we read all inputs and modify registers only after + * writing memory succeeded. + * + * TODO: if writing fails halfway through (e.g., when crossing page + * boundaries), we're in trouble. We'd need something like access_prepare(). + */ + sha256_write_ocv(env, param_addr, a, ra); + *message_reg = deposit64(*message_reg, 0, message_reg_len, + *message_reg + processed); + *len_reg -= processed; + + return !len ? 0 : 3; +} diff --git a/target/s390x/tcg/crypto_helper.c b/target/s390x/tcg/crypto_helper.c index 2d1c0290b5..b69dbb61a6 100644 --- a/target/s390x/tcg/crypto_helper.c +++ b/target/s390x/tcg/crypto_helper.c @@ -51,6 +51,10 @@ static int cpacf_kimd(CPUS390XState *env, const uintptr_t ra, int rc = 0; switch (fc) { + case 0x02: /* CPACF_KIMD_SHA_256 */ + rc = cpacf_sha256(env, ra, env->regs[1], &env->regs[r2], + &env->regs[r2 + 1], S390_FEAT_TYPE_KIMD); + break; case 0x03: /* CPACF_KIMD_SHA_512 */ rc = cpacf_sha512(env, ra, env->regs[1], &env->regs[r2], &env->regs[r2 + 1], S390_FEAT_TYPE_KIMD); @@ -68,6 +72,10 @@ static int cpacf_klmd(CPUS390XState *env, const uintptr_t ra, int rc = 0; switch (fc) { + case 0x02: /* CPACF_KLMD_SHA_256 */ + rc = cpacf_sha256(env, ra, env->regs[1], &env->regs[r2], + &env->regs[r2 + 1], S390_FEAT_TYPE_KLMD); + break; case 0x03: /* CPACF_KLMD_SHA_512 */ rc = cpacf_sha512(env, ra, env->regs[1], &env->regs[r2], &env->regs[r2 + 1], S390_FEAT_TYPE_KLMD); diff --git a/target/s390x/tcg/meson.build b/target/s390x/tcg/meson.build index 6dbc7ead65..4115f4c704 100644 --- a/target/s390x/tcg/meson.build +++ b/target/s390x/tcg/meson.build @@ -1,5 +1,6 @@ s390x_ss.add(when: 'CONFIG_TCG', if_true: files( 'cc_helper.c', + 'cpacf_sha256.c', 'cpacf_sha512.c', 'crypto_helper.c', 'excp_helper.c', -- 2.43.0
