On 6/15/21 4:36 AM, Paul Menzel wrote:
Dear Stefan,
Am 14.06.21 um 19:35 schrieb Stefan Berger:
Signed-off-by: Stefan Berger stefanb@linux.ibm.com
Makefile | 2 +- src/sha.h | 11 +++ src/sha1.c | 8 +- src/sha1.h | 8 -- src/sha256.c | 211 +++++++++++++++++++++++++++++++++++++++++++ src/sha512.c | 244 ++++++++++++++++++++++++++++++++++++++++++++++++++ src/tcgbios.c | 2 +- src/x86.h | 7 ++ 8 files changed, 479 insertions(+), 14 deletions(-) create mode 100644 src/sha.h delete mode 100644 src/sha1.h create mode 100644 src/sha256.c create mode 100644 src/sha512.c
Thank you for the patch. As the diffstatt is quite big, I am wondering how the correctness can be verified? Did you run some tests? Can I reproduce this in QEMU somehow?
I had tested the sha implementations with the first 3 of these vectors here and some other test data measured by trusted grub (on SLOF):
https://www.di-mgt.com.au/sha_testvectors.html
I ran tests inside QEMU with Fedora 34 using the tss2-1.6.0 package for example:
cp /sys/kernel/security/tpm/binary_bios_measurements ./
tsseventextend -if ./binary_bios_measurements -v -sim | less
Did you copy the code from somewhere?
No, it's not copied. SLOF has had the sha256 implementation for quite a while (I contributed it) and sha512 and sha384 are 64 bit derivatives more or less of the sha256 implementation.
Stefan
Kind regards,
Paul
diff --git a/Makefile b/Makefile index 3d8943e..c108f87 100644 --- a/Makefile +++ b/Makefile @@ -46,7 +46,7 @@ SRC32FLAT=$(SRCBOTH) post.c e820map.c malloc.c romfile.c x86.c \ fw/mtrr.c fw/xen.c fw/acpi.c fw/mptable.c fw/pirtable.c \ fw/smbios.c fw/romfile_loader.c fw/dsdt_parser.c hw/virtio-ring.c \ hw/virtio-pci.c hw/virtio-mmio.c hw/virtio-blk.c hw/virtio-scsi.c \ - hw/tpm_drivers.c hw/nvme.c + hw/tpm_drivers.c hw/nvme.c sha256.c sha512.c SRC32SEG=string.c output.c pcibios.c apm.c stacks.c hw/pci.c hw/serialio.c DIRS=src src/hw src/fw vgasrc diff --git a/src/sha.h b/src/sha.h new file mode 100644 index 0000000..edc9437 --- /dev/null +++ b/src/sha.h @@ -0,0 +1,11 @@ +#ifndef __SHA_H +#define __SHA_H
+#include "types.h" // u32
+void sha1(const u8 *data, u32 length, u8 *hash); +void sha256(const u8 *data, u32 length, u8 *hash); +void sha384(const u8 *data, u32 length, u8 *hash); +void sha512(const u8 *data, u32 length, u8 *hash);
+#endif // sha.h diff --git a/src/sha1.c b/src/sha1.c index 2ecb3cb..e6d80c8 100644 --- a/src/sha1.c +++ b/src/sha1.c @@ -13,7 +13,7 @@ #include "config.h" #include "byteorder.h" // cpu_to_*, __swab64 -#include "sha1.h" // sha1 +#include "sha.h" // sha1 #include "string.h" // memcpy #include "x86.h" // rol @@ -126,11 +126,11 @@ sha1_do(sha1_ctx *ctx, const u8 *data32, u32 length) } -u32 +void sha1(const u8 *data, u32 length, u8 *hash) { if (!CONFIG_TCGBIOS) - return 0; + return; sha1_ctx ctx = { .h[0] = 0x67452301, @@ -143,5 +143,5 @@ sha1(const u8 *data, u32 length, u8 *hash) sha1_do(&ctx, data, length); memcpy(hash, &ctx.h[0], 20); - return 0; + return; } diff --git a/src/sha1.h b/src/sha1.h deleted file mode 100644 index 07aabf3..0000000 --- a/src/sha1.h +++ /dev/null @@ -1,8 +0,0 @@ -#ifndef __SHA1_H -#define __SHA1_H
-#include "types.h" // u32
-u32 sha1(const u8 *data, u32 length, u8 *hash);
-#endif // sha1.h diff --git a/src/sha256.c b/src/sha256.c new file mode 100644 index 0000000..72c3df8 --- /dev/null +++ b/src/sha256.c @@ -0,0 +1,211 @@ +/*****************************************************************************
- Copyright (c) 2015-2020 IBM Corporation
- All rights reserved.
- This program and the accompanying materials
- are made available under the terms of the BSD License
- which accompanies this distribution, and is available at
- Contributors:
- * IBM Corporation - initial implementation
*****************************************************************************/
+/*
- * See: NIST standard for SHA-256 in FIPS PUB 180-4
- */
+#include "config.h" +#include "byteorder.h" +#include "sha.h" +#include "string.h" +#include "x86.h"
+typedef struct _sha256_ctx { + u32 h[8]; +} sha256_ctx;
+static inline u32 Ch(u32 x, u32 y, u32 z) +{ + return (x & y) | ((x ^ 0xffffffff) & z); +}
+static inline u32 Maj(u32 x, u32 y, u32 z) +{ + return (x & y) | (x & z) | (y & z); +}
+static inline u32 sum0(u32 x) +{ + return ror(x, 2) ^ ror(x, 13) ^ ror(x, 22); +}
+static inline u32 sum1(u32 x) +{ + return ror(x, 6) ^ ror(x, 11) ^ ror(x, 25); +}
+static inline u32 sigma0(u32 x) +{ + return ror(x, 7) ^ ror(x, 18) ^ (x >> 3); +}
+static inline u32 sigma1(u32 x) +{ + return ror(x, 17) ^ ror(x, 19) ^ (x >> 10); +}
+static void sha256_block(u32 *w, sha256_ctx *ctx) +{ + u32 t; + u32 a, b, c, d, e, f, g, h; + u32 T1, T2;
+ /* + * FIPS 180-4 4.2.2: SHA256 Constants + */ + static const u32 sha_ko[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 + };
+ /* + * FIPS 180-4 6.2.2: step 1 + * + * 0 <= i <= 15: + * W(t) = M(t) + * 16 <= i <= 63: + * W(t) = sigma1(W(t-2)) + W(t-7) + sigma0(W(t-15)) + W(t-16) + */
+ /* w(0)..w(15) are in big endian format */ + for (t = 0; t <= 15; t++) + w[t] = be32_to_cpu(w[t]);
+ for (t = 16; t <= 63; t++) + w[t] = sigma1(w[t-2]) + w[t-7] + sigma0(w[t-15]) + w[t-16];
+ /* + * step 2: a = H0, b = H1, c = H2, d = H3, e = H4, f = H5, g = H6, h = H7 + */ + a = ctx->h[0]; + b = ctx->h[1]; + c = ctx->h[2]; + d = ctx->h[3]; + e = ctx->h[4]; + f = ctx->h[5]; + g = ctx->h[6]; + h = ctx->h[7];
+ /* + * step 3: For i = 0 to 63: + * T1 = h + sum1(e) + Ch(e,f,g) + K(t) + W(t); + * T2 = sum0(a) + Maj(a,b,c) + * h = g; g = f; f = e; e = d + T1; d = c; c = b; b = a; a + T1 + T2 + */ + for (t = 0; t <= 63; t++) { + T1 = h + sum1(e) + Ch(e, f, g) + sha_ko[t] + w[t]; + T2 = sum0(a) + Maj(a, b, c); + h = g; + g = f; + f = e; + e = d + T1; + d = c; + c = b; + b = a; + a = T1 + T2; + }
+ /* + * step 4: + * H0 = a + H0, H1 = b + H1, H2 = c + H2, H3 = d + H3, H4 = e
- H4
+ */ + ctx->h[0] += a; + ctx->h[1] += b; + ctx->h[2] += c; + ctx->h[3] += d; + ctx->h[4] += e; + ctx->h[5] += f; + ctx->h[6] += g; + ctx->h[7] += h; +}
+static void sha256_do(sha256_ctx *ctx, const u8 *data32, u32 length) +{ + u32 offset; + u16 num; + u32 bits = 0; + u32 w[64]; + u64 tmp;
+ /* treat data in 64-byte chunks */ + for (offset = 0; length - offset >= 64; offset += 64) { + memcpy(w, data32 + offset, 64); + sha256_block((u32 *)w, ctx); + bits += (64 * 8); + }
+ /* last block with less than 64 bytes */ + num = length - offset; + bits += (num << 3);
+ memcpy(w, data32 + offset, num); + /* + * FIPS 180-4 5.1: Padding the Message + */ + ((u8 *)w)[num] = 0x80; + if (64 - (num + 1) > 0) + memset( &((u8 *)w)[num + 1], 0, 64 - (num + 1));
+ if (num >= 56) { + /* cannot append number of bits here */ + sha256_block((u32 *)w, ctx); + memset(w, 0, 60); + }
+ /* write number of bits to end of block */ + tmp = cpu_to_be64(bits); + memcpy(&w[14], &tmp, 8);
+ sha256_block(w, ctx);
+ /* need to switch result's endianness */ + for (num = 0; num < 8; num++) + ctx->h[num] = cpu_to_be32(ctx->h[num]); +}
+void sha256(const u8 *data, u32 length, u8 *hash) +{ + sha256_ctx ctx = { + .h = { + /* + * FIPS 180-4: 6.2.1 + * -> 5.3.3: initial hash value + */ + 0x6a09e667, + 0xbb67ae85, + 0x3c6ef372, + 0xa54ff53a, + 0x510e527f, + 0x9b05688c, + 0x1f83d9ab, + 0x5be0cd19 + } + };
+ sha256_do(&ctx, data, length); + memcpy(hash, ctx.h, sizeof(ctx.h)); +} diff --git a/src/sha512.c b/src/sha512.c new file mode 100644 index 0000000..fddd9fa --- /dev/null +++ b/src/sha512.c @@ -0,0 +1,244 @@ +/*****************************************************************************
- Copyright (c) 2021 IBM Corporation
- All rights reserved.
- This program and the accompanying materials
- are made available under the terms of the BSD License
- which accompanies this distribution, and is available at
- Contributors:
- * IBM Corporation - initial implementation
*****************************************************************************/
+/*
- * See: NIST standard for SHA-512 and SHA-384 in FIPS PUB 180-4 &
RFC 6234
- */
+#include "config.h" +#include "byteorder.h" +#include "sha.h" +#include "string.h"
+typedef struct _sha512_ctx { + u64 h[8]; +} sha512_ctx;
+static inline u64 ror64(u64 x, u8 n) +{ + return (x >> n) | (x << (64 - n)); +}
+static inline u64 Ch64(u64 x, u64 y, u64 z) +{ + return (x & y) ^ ((x ^ 0xffffffffffffffffULL) & z); +}
+static inline u64 Maj64(u64 x, u64 y, u64 z) +{ + return (x & y) ^ (x & z) ^ (y & z); +}
+static inline u64 sum0_64(u64 x) +{ + return ror64(x, 28) ^ ror64(x, 34) ^ ror64(x, 39); +}
+static inline u64 sum1_64(u64 x) +{ + return ror64(x, 14) ^ ror64(x, 18) ^ ror64(x, 41); +}
+static inline u64 sigma0_64(u64 x) +{ + return ror64(x, 1) ^ ror64(x, 8) ^ (x >> 7); +}
+static inline u64 sigma1_64(u64 x) +{ + return ror64(x, 19) ^ ror64(x, 61) ^ (x >> 6); +}
+static void sha512_block(u64 *w, sha512_ctx *ctx) +{ + u32 t; + u64 a, b, c, d, e, f, g, h; + u64 T1, T2;
+ /* + * FIPS 180-4 4.2.2: SHA512 Constants + */ + static const u64 sha_ko[80] = { + 0x428a2f98d728ae22, 0x7137449123ef65cd, 0xb5c0fbcfec4d3b2f, 0xe9b5dba58189dbbc, + 0x3956c25bf348b538, 0x59f111f1b605d019, 0x923f82a4af194f9b, 0xab1c5ed5da6d8118, + 0xd807aa98a3030242, 0x12835b0145706fbe, 0x243185be4ee4b28c, 0x550c7dc3d5ffb4e2, + 0x72be5d74f27b896f, 0x80deb1fe3b1696b1, 0x9bdc06a725c71235, 0xc19bf174cf692694, + 0xe49b69c19ef14ad2, 0xefbe4786384f25e3, 0x0fc19dc68b8cd5b5, 0x240ca1cc77ac9c65, + 0x2de92c6f592b0275, 0x4a7484aa6ea6e483, 0x5cb0a9dcbd41fbd4, 0x76f988da831153b5, + 0x983e5152ee66dfab, 0xa831c66d2db43210, 0xb00327c898fb213f, 0xbf597fc7beef0ee4, + 0xc6e00bf33da88fc2, 0xd5a79147930aa725, 0x06ca6351e003826f, 0x142929670a0e6e70, + 0x27b70a8546d22ffc, 0x2e1b21385c26c926, 0x4d2c6dfc5ac42aed, 0x53380d139d95b3df, + 0x650a73548baf63de, 0x766a0abb3c77b2a8, 0x81c2c92e47edaee6, 0x92722c851482353b, + 0xa2bfe8a14cf10364, 0xa81a664bbc423001, 0xc24b8b70d0f89791, 0xc76c51a30654be30, + 0xd192e819d6ef5218, 0xd69906245565a910, 0xf40e35855771202a, 0x106aa07032bbd1b8, + 0x19a4c116b8d2d0c8, 0x1e376c085141ab53, 0x2748774cdf8eeb99, 0x34b0bcb5e19b48a8, + 0x391c0cb3c5c95a63, 0x4ed8aa4ae3418acb, 0x5b9cca4f7763e373, 0x682e6ff3d6b2b8a3, + 0x748f82ee5defb2fc, 0x78a5636f43172f60, 0x84c87814a1f0ab72, 0x8cc702081a6439ec, + 0x90befffa23631e28, 0xa4506cebde82bde9, 0xbef9a3f7b2c67915, 0xc67178f2e372532b, + 0xca273eceea26619c, 0xd186b8c721c0c207, 0xeada7dd6cde0eb1e, 0xf57d4f7fee6ed178, + 0x06f067aa72176fba, 0x0a637dc5a2c898a6, 0x113f9804bef90dae, 0x1b710b35131c471b, + 0x28db77f523047d84, 0x32caab7b40c72493, 0x3c9ebe0a15c9bebc, 0x431d67c49c100d4c, + 0x4cc5d4becb3e42b6, 0x597f299cfc657e2a, 0x5fcb6fab3ad6faec, 0x6c44198c4a475817 + };
+ /* + * FIPS 180-4 6.4.2: step 1 + * + * 0 <= i <= 15: + * W(t) = M(t) + * 16 <= i <= 79: + * W(t) = sigma1(W(t-2)) + W(t-7) + sigma0(W(t-15)) + W(t-16) + */
+ /* w(0)..w(15) are in big endian format */ + for (t = 0; t <= 15; t++) + w[t] = be64_to_cpu(w[t]);
+ for (t = 16; t <= 79; t++) + w[t] = sigma1_64(w[t-2]) + w[t-7] + sigma0_64(w[t-15]) + w[t-16];
+ /* + * step 2: a = H0, b = H1, c = H2, d = H3, e = H4, f = H5, g = H6, h = H7 + */ + a = ctx->h[0]; + b = ctx->h[1]; + c = ctx->h[2]; + d = ctx->h[3]; + e = ctx->h[4]; + f = ctx->h[5]; + g = ctx->h[6]; + h = ctx->h[7];
+ /* + * step 3: For i = 0 to 79: + * T1 = h + sum1(e) + Ch(e,f,g) + K(t) + W(t); + * T2 = sum0(a) + Maj(a,b,c) + * h = g; g = f; f = e; e = d + T1; d = c; c = b; b = a; a + T1 + T2 + */ + for (t = 0; t <= 79; t++) { + T1 = h + sum1_64(e) + Ch64(e, f, g) + sha_ko[t] + w[t]; + T2 = sum0_64(a) + Maj64(a, b, c); + h = g; + g = f; + f = e; + e = d + T1; + d = c; + c = b; + b = a; + a = T1 + T2; + }
+ /* + * step 4: + * H0 = a + H0, H1 = b + H1, H2 = c + H2, H3 = d + H3, H4 = e
- H4
+ */ + ctx->h[0] += a; + ctx->h[1] += b; + ctx->h[2] += c; + ctx->h[3] += d; + ctx->h[4] += e; + ctx->h[5] += f; + ctx->h[6] += g; + ctx->h[7] += h; +}
+static void sha512_do(sha512_ctx *ctx, const u8 *data32, u32 length) +{ + u32 offset; + u16 num; + u64 bits = 0; + u64 w[80]; + u64 tmp;
+ /* treat data in 128-byte/1024 bit chunks */ + for (offset = 0; length - offset >= 128; offset += 128) { + memcpy(w, data32 + offset, 128); + sha512_block(w, ctx); + bits += (128 * 8); + }
+ /* last block with less than 128 bytes */ + num = length - offset; + bits += (num << 3);
+ memcpy(w, data32 + offset, num); + /* + * FIPS 180-4 5.1: Padding the Message + */ + ((u8 *)w)[num] = 0x80; + if (128 - (num + 1) > 0) + memset( &((u8 *)w)[num + 1], 0, 128 - (num + 1));
+ if (num >= 112) { + /* cannot append number of bits here; + * need space for 128 bits (16 bytes) + */ + sha512_block((u64 *)w, ctx); + memset(w, 0, 128); + }
+ /* write number of bits to end of the block; we write 64 bits */ + tmp = cpu_to_be64(bits); + memcpy(&w[15], &tmp, 8);
+ sha512_block(w, ctx);
+ /* need to switch result's endianness */ + for (num = 0; num < 8; num++) + ctx->h[num] = cpu_to_be64(ctx->h[num]); +}
+void sha384(const u8 *data, u32 length, u8 *hash) +{ + sha512_ctx ctx = { + .h = { + /* + * FIPS 180-4: 6.2.1 + * -> 5.3.4: initial hash value + */ + 0xcbbb9d5dc1059ed8, + 0x629a292a367cd507, + 0x9159015a3070dd17, + 0x152fecd8f70e5939, + 0x67332667ffc00b31, + 0x8eb44a8768581511, + 0xdb0c2e0d64f98fa7, + 0x47b5481dbefa4fa4 + } + };
+ sha512_do(&ctx, data, length); + memcpy(hash, ctx.h, 384/8); +}
+void sha512(const u8 *data, u32 length, u8 *hash) +{ + sha512_ctx ctx = { + .h = { + /* + * FIPS 180-4: 6.2.1 + * -> 5.3.5: initial hash value + */ + 0x6a09e667f3bcc908, + 0xbb67ae8584caa73b, + 0x3c6ef372fe94f82b, + 0xa54ff53a5f1d36f1, + 0x510e527fade682d1, + 0x9b05688c2b3e6c1f, + 0x1f83d9abfb41bd6b, + 0x5be0cd19137e2179 + } + };
+ sha512_do(&ctx, data, length); + memcpy(hash, ctx.h, sizeof(ctx.h)); +} diff --git a/src/tcgbios.c b/src/tcgbios.c index 4150aa0..a9d62c8 100644 --- a/src/tcgbios.c +++ b/src/tcgbios.c @@ -16,7 +16,7 @@ #include "fw/paravirt.h" // runningOnXen #include "hw/tpm_drivers.h" // tpm_drivers[] #include "output.h" // dprintf -#include "sha1.h" // sha1 +#include "sha.h" // sha1, sha256, ... #include "std/acpi.h" // RSDP_SIGNATURE, rsdt_descriptor #include "std/smbios.h" // struct smbios_entry_point #include "std/tcg.h" // TCG_PC_LOGOVERFLOW diff --git a/src/x86.h b/src/x86.h index c7bb60d..43a9e6d 100644 --- a/src/x86.h +++ b/src/x86.h @@ -140,6 +140,13 @@ static inline u32 rol(u32 val, u16 rol) { return res; } +static inline u32 ror(u32 val, u16 ror) { + u32 res; + asm volatile("rorl %%cl, %%eax" + : "=a" (res) : "a" (val), "c" (ror)); + return res; +}
static inline void outb(u8 value, u16 port) { __asm__ __volatile__("outb %b0, %w1" : : "a"(value), "Nd"(port)); }