Merge remote-tracking branch 'upstream/dev' into dev

This commit is contained in:
SChernykh 2019-03-04 17:43:22 +01:00
commit 123e7cf0b8
11 changed files with 96 additions and 3244 deletions

View file

@ -23,7 +23,7 @@ if (WITH_ASM AND NOT XMRIG_ARM AND CMAKE_SIZEOF_VOID_P EQUAL 8)
if (WIN32 AND CMAKE_C_COMPILER_ID MATCHES GNU)
set(XMRIG_ASM_FILES
"src/crypto/asm/win64/cn_main_loop.S"
"src/crypto/asm/win64/CryptonightR_template.S"
"src/crypto/asm/CryptonightR_template.S"
)
else()
set(XMRIG_ASM_FILES

View file

@ -202,8 +202,8 @@ template<> inline constexpr Variant cn_base_variant<VARIANT_4>() { return VA
template<Variant variant> inline constexpr bool cn_is_cryptonight_r() { return false; }
template<> inline constexpr bool cn_is_cryptonight_r<VARIANT_WOW>() { return true; }
template<> inline constexpr bool cn_is_cryptonight_r<VARIANT_4>() { return true; }
template<> inline constexpr bool cn_is_cryptonight_r<VARIANT_WOW>() { return true; }
template<> inline constexpr bool cn_is_cryptonight_r<VARIANT_4>() { return true; }
} /* namespace xmrig */

View file

@ -58,27 +58,58 @@ const static uint8_t test_input[380] = {
0xCF, 0x50, 0x29, 0x6A, 0x07, 0x0B, 0x93, 0x8F, 0x8F, 0xA8, 0x10, 0x04
};
const static char* test_input_WOW = R"===(9d47bf4c41b7e8e727e681715acb47fa1677cdba9ca7bcb05ad8cc8abd5daa66 5468697320697320612074657374205468697320697320612074657374205468697320697320612074657374 1806260
0d4a495cb844a3ca8ba4edb8e6bcf829ef1c06d9cdea2b62ca46c2a21b8b0a79 4c6f72656d20697073756d20646f6c6f722073697420616d65742c20636f6e73656374657475722061646970697363696e67 1806261
a1d6d848b5c5915fccd2f64cf216c6b1a02cf7c77bc80d8d4e51b419e88ff0dd 656c69742c2073656420646f20656975736d6f642074656d706f7220696e6369646964756e74207574206c61626f7265 1806262
af3a8544a0221a148c2ac90484b19861e3afca33fe17021efb8ad6496b567915 657420646f6c6f7265206d61676e6120616c697175612e20557420656e696d206164206d696e696d2076656e69616d2c 1806263
313399e0963ae8a99dab8af66d343e097dae0c0feb08dbc43ccdafef5515f413 71756973206e6f737472756420657865726369746174696f6e20756c6c616d636f206c61626f726973206e697369 1806264
6021c6ef90bff9ae94a7506d623d3a7a86c1756d655f50dd558f716d64622a34 757420616c697175697020657820656120636f6d6d6f646f20636f6e7365717561742e20447569732061757465 1806265
2b13000535f3db5f9b9b84a65c4351f386cd2cdedebb8c3ad2eab086e6a3fee5 697275726520646f6c6f7220696e20726570726568656e646572697420696e20766f6c7570746174652076656c6974 1806266
fc0e1dad8e895749dc90eb690bc1ba059a1cd772afaaf65a106bf9e5e6b80503 657373652063696c6c756d20646f6c6f726520657520667567696174206e756c6c612070617269617475722e 1806267
b60b0afe144deff7d903ed2d5545e77ebe66a3c51fee7016eeb8fee9eb630c0f 4578636570746575722073696e74206f6363616563617420637570696461746174206e6f6e2070726f6964656e742c 1806268
64774b27e7d5fec862fc4c0c13ac6bf09123b6f05bb0e4b75c97f379a2b3a679 73756e7420696e2063756c706120717569206f666669636961206465736572756e74206d6f6c6c697420616e696d20696420657374206c61626f72756d2e 1806269)===";
const static char* test_input_R = R"===(f759588ad57e758467295443a9bd71490abff8e9dad1b95b6bf2f5d0d78387bc 5468697320697320612074657374205468697320697320612074657374205468697320697320612074657374 1806260
5bb833deca2bdd7252a9ccd7b4ce0b6a4854515794b56c207262f7a5b9bdb566 4c6f72656d20697073756d20646f6c6f722073697420616d65742c20636f6e73656374657475722061646970697363696e67 1806261
1ee6728da60fbd8d7d55b2b1ade487a3cf52a2c3ac6f520db12c27d8921f6cab 656c69742c2073656420646f20656975736d6f642074656d706f7220696e6369646964756e74207574206c61626f7265 1806262
6969fe2ddfb758438d48049f302fc2108a4fcc93e37669170e6db4b0b9b4c4cb 657420646f6c6f7265206d61676e6120616c697175612e20557420656e696d206164206d696e696d2076656e69616d2c 1806263
7f3048b4e90d0cbe7a57c0394f37338a01fae3adfdc0e5126d863a895eb04e02 71756973206e6f737472756420657865726369746174696f6e20756c6c616d636f206c61626f726973206e697369 1806264
1d290443a4b542af04a82f6b2494a6ee7f20f2754c58e0849032483a56e8e2ef 757420616c697175697020657820656120636f6d6d6f646f20636f6e7365717561742e20447569732061757465 1806265
c43cc6567436a86afbd6aa9eaa7c276e9806830334b614b2bee23cc76634f6fd 697275726520646f6c6f7220696e20726570726568656e646572697420696e20766f6c7570746174652076656c6974 1806266
87be2479c0c4e8edfdfaa5603e93f4265b3f8224c1c5946feb424819d18990a4 657373652063696c6c756d20646f6c6f726520657520667567696174206e756c6c612070617269617475722e 1806267
dd9d6a6d8e47465cceac0877ef889b93e7eba979557e3935d7f86dce11b070f3 4578636570746575722073696e74206f6363616563617420637570696461746174206e6f6e2070726f6964656e742c 1806268
75c6f2ae49a20521de97285b431e717125847fb8935ed84a61e7f8d36a2c3d8e 73756e7420696e2063756c706120717569206f666669636961206465736572756e74206d6f6c6c697420616e696d20696420657374206c61626f72756d2e 1806269)===";
struct cn_r_test_input_data
{
uint64_t height;
size_t size;
uint8_t data[64];
};
const static cn_r_test_input_data cn_r_test_input[] = {
{ 1806260, 44, { 0x54, 0x68, 0x69, 0x73, 0x20, 0x69, 0x73, 0x20, 0x61, 0x20, 0x74, 0x65, 0x73, 0x74, 0x20, 0x54, 0x68, 0x69, 0x73, 0x20, 0x69, 0x73, 0x20, 0x61, 0x20, 0x74, 0x65, 0x73, 0x74, 0x20, 0x54, 0x68, 0x69, 0x73, 0x20, 0x69, 0x73, 0x20, 0x61, 0x20, 0x74, 0x65, 0x73, 0x74 } },
{ 1806261, 50, { 0x4c, 0x6f, 0x72, 0x65, 0x6d, 0x20, 0x69, 0x70, 0x73, 0x75, 0x6d, 0x20, 0x64, 0x6f, 0x6c, 0x6f, 0x72, 0x20, 0x73, 0x69, 0x74, 0x20, 0x61, 0x6d, 0x65, 0x74, 0x2c, 0x20, 0x63, 0x6f, 0x6e, 0x73, 0x65, 0x63, 0x74, 0x65, 0x74, 0x75, 0x72, 0x20, 0x61, 0x64, 0x69, 0x70, 0x69, 0x73, 0x63, 0x69, 0x6e, 0x67 } },
{ 1806262, 48, { 0x65, 0x6c, 0x69, 0x74, 0x2c, 0x20, 0x73, 0x65, 0x64, 0x20, 0x64, 0x6f, 0x20, 0x65, 0x69, 0x75, 0x73, 0x6d, 0x6f, 0x64, 0x20, 0x74, 0x65, 0x6d, 0x70, 0x6f, 0x72, 0x20, 0x69, 0x6e, 0x63, 0x69, 0x64, 0x69, 0x64, 0x75, 0x6e, 0x74, 0x20, 0x75, 0x74, 0x20, 0x6c, 0x61, 0x62, 0x6f, 0x72, 0x65 } },
{ 1806263, 48, { 0x65, 0x74, 0x20, 0x64, 0x6f, 0x6c, 0x6f, 0x72, 0x65, 0x20, 0x6d, 0x61, 0x67, 0x6e, 0x61, 0x20, 0x61, 0x6c, 0x69, 0x71, 0x75, 0x61, 0x2e, 0x20, 0x55, 0x74, 0x20, 0x65, 0x6e, 0x69, 0x6d, 0x20, 0x61, 0x64, 0x20, 0x6d, 0x69, 0x6e, 0x69, 0x6d, 0x20, 0x76, 0x65, 0x6e, 0x69, 0x61, 0x6d, 0x2c } },
{ 1806264, 46, { 0x71, 0x75, 0x69, 0x73, 0x20, 0x6e, 0x6f, 0x73, 0x74, 0x72, 0x75, 0x64, 0x20, 0x65, 0x78, 0x65, 0x72, 0x63, 0x69, 0x74, 0x61, 0x74, 0x69, 0x6f, 0x6e, 0x20, 0x75, 0x6c, 0x6c, 0x61, 0x6d, 0x63, 0x6f, 0x20, 0x6c, 0x61, 0x62, 0x6f, 0x72, 0x69, 0x73, 0x20, 0x6e, 0x69, 0x73, 0x69 } },
{ 1806265, 45, { 0x75, 0x74, 0x20, 0x61, 0x6c, 0x69, 0x71, 0x75, 0x69, 0x70, 0x20, 0x65, 0x78, 0x20, 0x65, 0x61, 0x20, 0x63, 0x6f, 0x6d, 0x6d, 0x6f, 0x64, 0x6f, 0x20, 0x63, 0x6f, 0x6e, 0x73, 0x65, 0x71, 0x75, 0x61, 0x74, 0x2e, 0x20, 0x44, 0x75, 0x69, 0x73, 0x20, 0x61, 0x75, 0x74, 0x65 } },
{ 1806266, 47, { 0x69, 0x72, 0x75, 0x72, 0x65, 0x20, 0x64, 0x6f, 0x6c, 0x6f, 0x72, 0x20, 0x69, 0x6e, 0x20, 0x72, 0x65, 0x70, 0x72, 0x65, 0x68, 0x65, 0x6e, 0x64, 0x65, 0x72, 0x69, 0x74, 0x20, 0x69, 0x6e, 0x20, 0x76, 0x6f, 0x6c, 0x75, 0x70, 0x74, 0x61, 0x74, 0x65, 0x20, 0x76, 0x65, 0x6c, 0x69, 0x74 } },
{ 1806267, 44, { 0x65, 0x73, 0x73, 0x65, 0x20, 0x63, 0x69, 0x6c, 0x6c, 0x75, 0x6d, 0x20, 0x64, 0x6f, 0x6c, 0x6f, 0x72, 0x65, 0x20, 0x65, 0x75, 0x20, 0x66, 0x75, 0x67, 0x69, 0x61, 0x74, 0x20, 0x6e, 0x75, 0x6c, 0x6c, 0x61, 0x20, 0x70, 0x61, 0x72, 0x69, 0x61, 0x74, 0x75, 0x72, 0x2e } },
{ 1806268, 47, { 0x45, 0x78, 0x63, 0x65, 0x70, 0x74, 0x65, 0x75, 0x72, 0x20, 0x73, 0x69, 0x6e, 0x74, 0x20, 0x6f, 0x63, 0x63, 0x61, 0x65, 0x63, 0x61, 0x74, 0x20, 0x63, 0x75, 0x70, 0x69, 0x64, 0x61, 0x74, 0x61, 0x74, 0x20, 0x6e, 0x6f, 0x6e, 0x20, 0x70, 0x72, 0x6f, 0x69, 0x64, 0x65, 0x6e, 0x74, 0x2c } },
{ 1806269, 62, { 0x73, 0x75, 0x6e, 0x74, 0x20, 0x69, 0x6e, 0x20, 0x63, 0x75, 0x6c, 0x70, 0x61, 0x20, 0x71, 0x75, 0x69, 0x20, 0x6f, 0x66, 0x66, 0x69, 0x63, 0x69, 0x61, 0x20, 0x64, 0x65, 0x73, 0x65, 0x72, 0x75, 0x6e, 0x74, 0x20, 0x6d, 0x6f, 0x6c, 0x6c, 0x69, 0x74, 0x20, 0x61, 0x6e, 0x69, 0x6d, 0x20, 0x69, 0x64, 0x20, 0x65, 0x73, 0x74, 0x20, 0x6c, 0x61, 0x62, 0x6f, 0x72, 0x75, 0x6d, 0x2e } },
};
// "cn/wow"
const static uint8_t test_output_wow[] = {
0x9d, 0x47, 0xbf, 0x4c, 0x41, 0xb7, 0xe8, 0xe7, 0x27, 0xe6, 0x81, 0x71, 0x5a, 0xcb, 0x47, 0xfa, 0x16, 0x77, 0xcd, 0xba, 0x9c, 0xa7, 0xbc, 0xb0, 0x5a, 0xd8, 0xcc, 0x8a, 0xbd, 0x5d, 0xaa, 0x66,
0x0d, 0x4a, 0x49, 0x5c, 0xb8, 0x44, 0xa3, 0xca, 0x8b, 0xa4, 0xed, 0xb8, 0xe6, 0xbc, 0xf8, 0x29, 0xef, 0x1c, 0x06, 0xd9, 0xcd, 0xea, 0x2b, 0x62, 0xca, 0x46, 0xc2, 0xa2, 0x1b, 0x8b, 0x0a, 0x79,
0xa1, 0xd6, 0xd8, 0x48, 0xb5, 0xc5, 0x91, 0x5f, 0xcc, 0xd2, 0xf6, 0x4c, 0xf2, 0x16, 0xc6, 0xb1, 0xa0, 0x2c, 0xf7, 0xc7, 0x7b, 0xc8, 0x0d, 0x8d, 0x4e, 0x51, 0xb4, 0x19, 0xe8, 0x8f, 0xf0, 0xdd,
0xaf, 0x3a, 0x85, 0x44, 0xa0, 0x22, 0x1a, 0x14, 0x8c, 0x2a, 0xc9, 0x04, 0x84, 0xb1, 0x98, 0x61, 0xe3, 0xaf, 0xca, 0x33, 0xfe, 0x17, 0x02, 0x1e, 0xfb, 0x8a, 0xd6, 0x49, 0x6b, 0x56, 0x79, 0x15,
0x31, 0x33, 0x99, 0xe0, 0x96, 0x3a, 0xe8, 0xa9, 0x9d, 0xab, 0x8a, 0xf6, 0x6d, 0x34, 0x3e, 0x09, 0x7d, 0xae, 0x0c, 0x0f, 0xeb, 0x08, 0xdb, 0xc4, 0x3c, 0xcd, 0xaf, 0xef, 0x55, 0x15, 0xf4, 0x13,
0x60, 0x21, 0xc6, 0xef, 0x90, 0xbf, 0xf9, 0xae, 0x94, 0xa7, 0x50, 0x6d, 0x62, 0x3d, 0x3a, 0x7a, 0x86, 0xc1, 0x75, 0x6d, 0x65, 0x5f, 0x50, 0xdd, 0x55, 0x8f, 0x71, 0x6d, 0x64, 0x62, 0x2a, 0x34,
0x2b, 0x13, 0x00, 0x05, 0x35, 0xf3, 0xdb, 0x5f, 0x9b, 0x9b, 0x84, 0xa6, 0x5c, 0x43, 0x51, 0xf3, 0x86, 0xcd, 0x2c, 0xde, 0xde, 0xbb, 0x8c, 0x3a, 0xd2, 0xea, 0xb0, 0x86, 0xe6, 0xa3, 0xfe, 0xe5,
0xfc, 0x0e, 0x1d, 0xad, 0x8e, 0x89, 0x57, 0x49, 0xdc, 0x90, 0xeb, 0x69, 0x0b, 0xc1, 0xba, 0x05, 0x9a, 0x1c, 0xd7, 0x72, 0xaf, 0xaa, 0xf6, 0x5a, 0x10, 0x6b, 0xf9, 0xe5, 0xe6, 0xb8, 0x05, 0x03,
0xb6, 0x0b, 0x0a, 0xfe, 0x14, 0x4d, 0xef, 0xf7, 0xd9, 0x03, 0xed, 0x2d, 0x55, 0x45, 0xe7, 0x7e, 0xbe, 0x66, 0xa3, 0xc5, 0x1f, 0xee, 0x70, 0x16, 0xee, 0xb8, 0xfe, 0xe9, 0xeb, 0x63, 0x0c, 0x0f,
0x64, 0x77, 0x4b, 0x27, 0xe7, 0xd5, 0xfe, 0xc8, 0x62, 0xfc, 0x4c, 0x0c, 0x13, 0xac, 0x6b, 0xf0, 0x91, 0x23, 0xb6, 0xf0, 0x5b, 0xb0, 0xe4, 0xb7, 0x5c, 0x97, 0xf3, 0x79, 0xa2, 0xb3, 0xa6, 0x79,
};
// "cn/r"
const static uint8_t test_output_r[] = {
0xf7, 0x59, 0x58, 0x8a, 0xd5, 0x7e, 0x75, 0x84, 0x67, 0x29, 0x54, 0x43, 0xa9, 0xbd, 0x71, 0x49, 0x0a, 0xbf, 0xf8, 0xe9, 0xda, 0xd1, 0xb9, 0x5b, 0x6b, 0xf2, 0xf5, 0xd0, 0xd7, 0x83, 0x87, 0xbc,
0x5b, 0xb8, 0x33, 0xde, 0xca, 0x2b, 0xdd, 0x72, 0x52, 0xa9, 0xcc, 0xd7, 0xb4, 0xce, 0x0b, 0x6a, 0x48, 0x54, 0x51, 0x57, 0x94, 0xb5, 0x6c, 0x20, 0x72, 0x62, 0xf7, 0xa5, 0xb9, 0xbd, 0xb5, 0x66,
0x1e, 0xe6, 0x72, 0x8d, 0xa6, 0x0f, 0xbd, 0x8d, 0x7d, 0x55, 0xb2, 0xb1, 0xad, 0xe4, 0x87, 0xa3, 0xcf, 0x52, 0xa2, 0xc3, 0xac, 0x6f, 0x52, 0x0d, 0xb1, 0x2c, 0x27, 0xd8, 0x92, 0x1f, 0x6c, 0xab,
0x69, 0x69, 0xfe, 0x2d, 0xdf, 0xb7, 0x58, 0x43, 0x8d, 0x48, 0x04, 0x9f, 0x30, 0x2f, 0xc2, 0x10, 0x8a, 0x4f, 0xcc, 0x93, 0xe3, 0x76, 0x69, 0x17, 0x0e, 0x6d, 0xb4, 0xb0, 0xb9, 0xb4, 0xc4, 0xcb,
0x7f, 0x30, 0x48, 0xb4, 0xe9, 0x0d, 0x0c, 0xbe, 0x7a, 0x57, 0xc0, 0x39, 0x4f, 0x37, 0x33, 0x8a, 0x01, 0xfa, 0xe3, 0xad, 0xfd, 0xc0, 0xe5, 0x12, 0x6d, 0x86, 0x3a, 0x89, 0x5e, 0xb0, 0x4e, 0x02,
0x1d, 0x29, 0x04, 0x43, 0xa4, 0xb5, 0x42, 0xaf, 0x04, 0xa8, 0x2f, 0x6b, 0x24, 0x94, 0xa6, 0xee, 0x7f, 0x20, 0xf2, 0x75, 0x4c, 0x58, 0xe0, 0x84, 0x90, 0x32, 0x48, 0x3a, 0x56, 0xe8, 0xe2, 0xef,
0xc4, 0x3c, 0xc6, 0x56, 0x74, 0x36, 0xa8, 0x6a, 0xfb, 0xd6, 0xaa, 0x9e, 0xaa, 0x7c, 0x27, 0x6e, 0x98, 0x06, 0x83, 0x03, 0x34, 0xb6, 0x14, 0xb2, 0xbe, 0xe2, 0x3c, 0xc7, 0x66, 0x34, 0xf6, 0xfd,
0x87, 0xbe, 0x24, 0x79, 0xc0, 0xc4, 0xe8, 0xed, 0xfd, 0xfa, 0xa5, 0x60, 0x3e, 0x93, 0xf4, 0x26, 0x5b, 0x3f, 0x82, 0x24, 0xc1, 0xc5, 0x94, 0x6f, 0xeb, 0x42, 0x48, 0x19, 0xd1, 0x89, 0x90, 0xa4,
0xdd, 0x9d, 0x6a, 0x6d, 0x8e, 0x47, 0x46, 0x5c, 0xce, 0xac, 0x08, 0x77, 0xef, 0x88, 0x9b, 0x93, 0xe7, 0xeb, 0xa9, 0x79, 0x55, 0x7e, 0x39, 0x35, 0xd7, 0xf8, 0x6d, 0xce, 0x11, 0xb0, 0x70, 0xf3,
0x75, 0xc6, 0xf2, 0xae, 0x49, 0xa2, 0x05, 0x21, 0xde, 0x97, 0x28, 0x5b, 0x43, 0x1e, 0x71, 0x71, 0x25, 0x84, 0x7f, 0xb8, 0x93, 0x5e, 0xd8, 0x4a, 0x61, 0xe7, 0xf8, 0xd3, 0x6a, 0x2c, 0x3d, 0x8e,
};
// "cn/0"
const static uint8_t test_output_v0[160] = {
@ -295,8 +326,6 @@ const static uint8_t test_output_pico_trtl[160] = {
};
#endif
unsigned char hf_hex2bin(char c, bool &err);
char hf_bin2hex(unsigned char c);
#ifndef XMRIG_NO_CN_GPU
// "cn/gpu"

View file

@ -1,279 +0,0 @@
PUBLIC FN_PREFIX(CryptonightR_soft_aes_template_part1)
PUBLIC FN_PREFIX(CryptonightR_soft_aes_template_mainloop)
PUBLIC FN_PREFIX(CryptonightR_soft_aes_template_part2)
PUBLIC FN_PREFIX(CryptonightR_soft_aes_template_part3)
PUBLIC FN_PREFIX(CryptonightR_soft_aes_template_end)
ALIGN(64)
FN_PREFIX(CryptonightR_soft_aes_template_part1):
mov QWORD PTR [rsp+8], rcx
push rbx
push rbp
push rsi
push rdi
push r12
push r13
push r14
push r15
sub rsp, 232
mov eax, [rcx+96]
mov ebx, [rcx+100]
mov esi, [rcx+104]
mov edx, [rcx+108]
mov [rsp+144], eax
mov [rsp+148], ebx
mov [rsp+152], esi
mov [rsp+156], edx
mov rax, QWORD PTR [rcx+48]
mov r10, rcx
xor rax, QWORD PTR [rcx+16]
mov r8, QWORD PTR [rcx+32]
xor r8, QWORD PTR [rcx]
mov r9, QWORD PTR [rcx+40]
xor r9, QWORD PTR [rcx+8]
movd xmm4, rax
mov rdx, QWORD PTR [rcx+56]
xor rdx, QWORD PTR [rcx+24]
mov r11, QWORD PTR [rcx+224]
mov rcx, QWORD PTR [rcx+88]
xor rcx, QWORD PTR [r10+72]
mov rax, QWORD PTR [r10+80]
movd xmm0, rdx
xor rax, QWORD PTR [r10+64]
movaps XMMWORD PTR [rsp+16], xmm6
movaps XMMWORD PTR [rsp+32], xmm7
movaps XMMWORD PTR [rsp+48], xmm8
movaps XMMWORD PTR [rsp+64], xmm9
movaps XMMWORD PTR [rsp+80], xmm10
movaps XMMWORD PTR [rsp+96], xmm11
movaps XMMWORD PTR [rsp+112], xmm12
movaps XMMWORD PTR [rsp+128], xmm13
movd xmm5, rax
mov rax, r8
punpcklqdq xmm4, xmm0
and eax, 2097136
movd xmm10, QWORD PTR [r10+96]
movd xmm0, rcx
mov rcx, QWORD PTR [r10+104]
xorps xmm9, xmm9
mov QWORD PTR [rsp+328], rax
movd xmm12, r11
mov QWORD PTR [rsp+320], r9
punpcklqdq xmm5, xmm0
movd xmm13, rcx
mov r12d, 524288
ALIGN(64)
FN_PREFIX(CryptonightR_soft_aes_template_mainloop):
movd xmm11, r12d
mov r12, QWORD PTR [r10+272]
lea r13, QWORD PTR [rax+r11]
mov esi, DWORD PTR [r13]
movd xmm0, r9
mov r10d, DWORD PTR [r13+4]
movd xmm7, r8
mov ebp, DWORD PTR [r13+12]
mov r14d, DWORD PTR [r13+8]
mov rdx, QWORD PTR [rsp+328]
movzx ecx, sil
shr esi, 8
punpcklqdq xmm7, xmm0
mov r15d, DWORD PTR [r12+rcx*4]
movzx ecx, r10b
shr r10d, 8
mov edi, DWORD PTR [r12+rcx*4]
movzx ecx, r14b
shr r14d, 8
mov ebx, DWORD PTR [r12+rcx*4]
movzx ecx, bpl
shr ebp, 8
mov r9d, DWORD PTR [r12+rcx*4]
movzx ecx, r10b
shr r10d, 8
xor r15d, DWORD PTR [r12+rcx*4+1024]
movzx ecx, r14b
shr r14d, 8
mov eax, r14d
shr eax, 8
xor edi, DWORD PTR [r12+rcx*4+1024]
add eax, 256
movzx ecx, bpl
shr ebp, 8
xor ebx, DWORD PTR [r12+rcx*4+1024]
movzx ecx, sil
shr esi, 8
xor r9d, DWORD PTR [r12+rcx*4+1024]
add r12, 2048
movzx ecx, r10b
shr r10d, 8
add r10d, 256
mov r11d, DWORD PTR [r12+rax*4]
xor r11d, DWORD PTR [r12+rcx*4]
xor r11d, r9d
movzx ecx, sil
mov r10d, DWORD PTR [r12+r10*4]
shr esi, 8
add esi, 256
xor r10d, DWORD PTR [r12+rcx*4]
movzx ecx, bpl
xor r10d, ebx
shr ebp, 8
movd xmm1, r11d
add ebp, 256
movd r11, xmm12
mov r9d, DWORD PTR [r12+rcx*4]
xor r9d, DWORD PTR [r12+rsi*4]
mov eax, DWORD PTR [r12+rbp*4]
xor r9d, edi
movzx ecx, r14b
movd xmm0, r10d
movd xmm2, r9d
xor eax, DWORD PTR [r12+rcx*4]
mov rcx, rdx
xor eax, r15d
punpckldq xmm2, xmm1
xor rcx, 16
movd xmm6, eax
mov rax, rdx
punpckldq xmm6, xmm0
xor rax, 32
punpckldq xmm6, xmm2
xor rdx, 48
movdqu xmm2, XMMWORD PTR [rcx+r11]
pxor xmm6, xmm2
pxor xmm6, xmm7
paddq xmm2, xmm4
movdqu xmm1, XMMWORD PTR [rax+r11]
movdqu xmm0, XMMWORD PTR [rdx+r11]
pxor xmm6, xmm1
pxor xmm6, xmm0
paddq xmm0, xmm5
movdqu XMMWORD PTR [rcx+r11], xmm0
movdqu XMMWORD PTR [rax+r11], xmm2
movd rcx, xmm13
paddq xmm1, xmm7
movdqu XMMWORD PTR [rdx+r11], xmm1
movd rdi, xmm6
mov r10, rdi
and r10d, 2097136
movdqa xmm0, xmm6
pxor xmm0, xmm4
movdqu XMMWORD PTR [r13], xmm0
mov ebx, [rsp+144]
mov ebp, [rsp+152]
add ebx, [rsp+148]
add ebp, [rsp+156]
shl rbp, 32
or rbx, rbp
xor rbx, QWORD PTR [r10+r11]
lea r14, QWORD PTR [r10+r11]
mov rbp, QWORD PTR [r14+8]
mov [rsp+160], rbx
mov [rsp+168], rdi
mov [rsp+176], rbp
mov [rsp+184], r10
mov r10, rsp
mov ebx, [rsp+144]
mov esi, [rsp+148]
mov edi, [rsp+152]
mov ebp, [rsp+156]
movd esp, xmm7
movaps xmm0, xmm7
psrldq xmm0, 8
movd r15d, xmm0
movd eax, xmm4
movd edx, xmm5
movaps xmm0, xmm5
psrldq xmm0, 8
movd r9d, xmm0
FN_PREFIX(CryptonightR_soft_aes_template_part2):
mov rsp, r10
mov [rsp+144], ebx
mov [rsp+148], esi
mov [rsp+152], edi
mov [rsp+156], ebp
mov edi, edi
shl rbp, 32
or rbp, rdi
xor r8, rbp
mov ebx, ebx
shl rsi, 32
or rsi, rbx
xor QWORD PTR [rsp+320], rsi
mov rbx, [rsp+160]
mov rdi, [rsp+168]
mov rbp, [rsp+176]
mov r10, [rsp+184]
mov r9, r10
xor r9, 16
mov rcx, r10
xor rcx, 32
xor r10, 48
mov rax, rbx
mul rdi
movdqu xmm2, XMMWORD PTR [r9+r11]
movdqu xmm1, XMMWORD PTR [rcx+r11]
pxor xmm6, xmm2
pxor xmm6, xmm1
paddq xmm1, xmm7
add r8, rdx
movdqu xmm0, XMMWORD PTR [r10+r11]
pxor xmm6, xmm0
paddq xmm0, xmm5
paddq xmm2, xmm4
movdqu XMMWORD PTR [r9+r11], xmm0
movdqa xmm5, xmm4
mov r9, QWORD PTR [rsp+320]
movdqa xmm4, xmm6
add r9, rax
movdqu XMMWORD PTR [rcx+r11], xmm2
movdqu XMMWORD PTR [r10+r11], xmm1
mov r10, QWORD PTR [rsp+304]
movd r12d, xmm11
mov QWORD PTR [r14], r8
xor r8, rbx
mov rax, r8
mov QWORD PTR [r14+8], r9
and eax, 2097136
xor r9, rbp
mov QWORD PTR [rsp+320], r9
mov QWORD PTR [rsp+328], rax
sub r12d, 1
jne FN_PREFIX(CryptonightR_soft_aes_template_mainloop)
FN_PREFIX(CryptonightR_soft_aes_template_part3):
movaps xmm6, XMMWORD PTR [rsp+16]
movaps xmm7, XMMWORD PTR [rsp+32]
movaps xmm8, XMMWORD PTR [rsp+48]
movaps xmm9, XMMWORD PTR [rsp+64]
movaps xmm10, XMMWORD PTR [rsp+80]
movaps xmm11, XMMWORD PTR [rsp+96]
movaps xmm12, XMMWORD PTR [rsp+112]
movaps xmm13, XMMWORD PTR [rsp+128]
add rsp, 232
pop r15
pop r14
pop r13
pop r12
pop rdi
pop rsi
pop rbp
pop rbx
ret
FN_PREFIX(CryptonightR_soft_aes_template_end):

File diff suppressed because it is too large Load diff

View file

@ -1,531 +0,0 @@
PUBLIC FN_PREFIX(CryptonightR_template_part1)
PUBLIC FN_PREFIX(CryptonightR_template_mainloop)
PUBLIC FN_PREFIX(CryptonightR_template_part2)
PUBLIC FN_PREFIX(CryptonightR_template_part3)
PUBLIC FN_PREFIX(CryptonightR_template_end)
PUBLIC FN_PREFIX(CryptonightR_template_double_part1)
PUBLIC FN_PREFIX(CryptonightR_template_double_mainloop)
PUBLIC FN_PREFIX(CryptonightR_template_double_part2)
PUBLIC FN_PREFIX(CryptonightR_template_double_part3)
PUBLIC FN_PREFIX(CryptonightR_template_double_part4)
PUBLIC FN_PREFIX(CryptonightR_template_double_end)
ALIGN(64)
FN_PREFIX(CryptonightR_template_part1):
mov QWORD PTR [rsp+16], rbx
mov QWORD PTR [rsp+24], rbp
mov QWORD PTR [rsp+32], rsi
push r10
push r11
push r12
push r13
push r14
push r15
push rdi
sub rsp, 64
mov r12, rcx
mov r8, QWORD PTR [r12+32]
mov rdx, r12
xor r8, QWORD PTR [r12]
mov r15, QWORD PTR [r12+40]
mov r9, r8
xor r15, QWORD PTR [r12+8]
mov r11, QWORD PTR [r12+224]
mov r12, QWORD PTR [r12+56]
xor r12, QWORD PTR [rdx+24]
mov rax, QWORD PTR [rdx+48]
xor rax, QWORD PTR [rdx+16]
movaps XMMWORD PTR [rsp+48], xmm6
movd xmm0, r12
movaps XMMWORD PTR [rsp+32], xmm7
movaps XMMWORD PTR [rsp+16], xmm8
movaps XMMWORD PTR [rsp], xmm9
mov r12, QWORD PTR [rdx+88]
xor r12, QWORD PTR [rdx+72]
movd xmm6, rax
mov rax, QWORD PTR [rdx+80]
xor rax, QWORD PTR [rdx+64]
punpcklqdq xmm6, xmm0
and r9d, 2097136
movd xmm0, r12
movd xmm7, rax
punpcklqdq xmm7, xmm0
mov r10d, r9d
movd xmm9, rsp
mov rsp, r8
mov r8d, 524288
mov ebx, [rdx+96]
mov esi, [rdx+100]
mov edi, [rdx+104]
mov ebp, [rdx+108]
ALIGN(64)
FN_PREFIX(CryptonightR_template_mainloop):
movdqa xmm5, XMMWORD PTR [r9+r11]
movd xmm0, r15
movd xmm4, rsp
punpcklqdq xmm4, xmm0
lea rdx, QWORD PTR [r9+r11]
aesenc xmm5, xmm4
mov r13d, r9d
mov eax, r9d
xor r9d, 48
xor r13d, 16
xor eax, 32
movdqu xmm0, XMMWORD PTR [r9+r11]
movaps xmm3, xmm0
movdqu xmm2, XMMWORD PTR [r13+r11]
movdqu xmm1, XMMWORD PTR [rax+r11]
pxor xmm0, xmm2
pxor xmm5, xmm1
pxor xmm5, xmm0
movd r12, xmm5
movd r10d, xmm5
and r10d, 2097136
paddq xmm3, xmm7
paddq xmm2, xmm6
paddq xmm1, xmm4
movdqu XMMWORD PTR [r13+r11], xmm3
movdqu XMMWORD PTR [rax+r11], xmm2
movdqu XMMWORD PTR [r9+r11], xmm1
movdqa xmm0, xmm5
pxor xmm0, xmm6
movdqu XMMWORD PTR [rdx], xmm0
lea r13d, [ebx+esi]
lea edx, [edi+ebp]
shl rdx, 32
or r13, rdx
movd eax, xmm6
movd edx, xmm7
pextrd r9d, xmm7, 2
xor r13, QWORD PTR [r10+r11]
mov r14, QWORD PTR [r10+r11+8]
FN_PREFIX(CryptonightR_template_part2):
lea rcx, [r10+r11]
mov eax, edi
mov edx, ebp
shl rdx, 32
or rax, rdx
xor rsp, rax
mov eax, ebx
mov edx, esi
shl rdx, 32
or rax, rdx
xor r15, rax
mov rax, r13
mul r12
add r15, rax
add rsp, rdx
mov r9d, r10d
mov r12d, r10d
xor r9d, 16
xor r12d, 32
xor r10d, 48
movdqa xmm1, XMMWORD PTR [r12+r11]
movaps xmm3, xmm1
movdqa xmm2, XMMWORD PTR [r9+r11]
movdqa xmm0, XMMWORD PTR [r10+r11]
pxor xmm1, xmm2
pxor xmm5, xmm0
pxor xmm5, xmm1
paddq xmm3, xmm4
paddq xmm2, xmm6
paddq xmm0, xmm7
movdqu XMMWORD PTR [r9+r11], xmm0
movdqu XMMWORD PTR [r12+r11], xmm2
movdqu XMMWORD PTR [r10+r11], xmm3
movdqa xmm7, xmm6
mov QWORD PTR [rcx], rsp
xor rsp, r13
mov r9d, esp
mov QWORD PTR [rcx+8], r15
and r9d, 2097136
xor r15, r14
movdqa xmm6, xmm5
dec r8d
jnz FN_PREFIX(CryptonightR_template_mainloop)
FN_PREFIX(CryptonightR_template_part3):
movd rsp, xmm9
mov rbx, QWORD PTR [rsp+136]
mov rbp, QWORD PTR [rsp+144]
mov rsi, QWORD PTR [rsp+152]
movaps xmm6, XMMWORD PTR [rsp+48]
movaps xmm7, XMMWORD PTR [rsp+32]
movaps xmm8, XMMWORD PTR [rsp+16]
movaps xmm9, XMMWORD PTR [rsp]
add rsp, 64
pop rdi
pop r15
pop r14
pop r13
pop r12
pop r11
pop r10
ret 0
FN_PREFIX(CryptonightR_template_end):
ALIGN(64)
FN_PREFIX(CryptonightR_template_double_part1):
mov QWORD PTR [rsp+24], rbx
push rbp
push rsi
push rdi
push r12
push r13
push r14
push r15
sub rsp, 320
mov r14, QWORD PTR [rcx+32]
mov r8, rcx
xor r14, QWORD PTR [rcx]
mov r12, QWORD PTR [rcx+40]
mov ebx, r14d
mov rsi, QWORD PTR [rcx+224]
and ebx, 2097136
xor r12, QWORD PTR [rcx+8]
mov rcx, QWORD PTR [rcx+56]
xor rcx, QWORD PTR [r8+24]
mov rax, QWORD PTR [r8+48]
xor rax, QWORD PTR [r8+16]
mov r15, QWORD PTR [rdx+32]
xor r15, QWORD PTR [rdx]
movd xmm0, rcx
mov rcx, QWORD PTR [r8+88]
xor rcx, QWORD PTR [r8+72]
mov r13, QWORD PTR [rdx+40]
mov rdi, QWORD PTR [rdx+224]
xor r13, QWORD PTR [rdx+8]
movaps XMMWORD PTR [rsp+160], xmm6
movaps XMMWORD PTR [rsp+176], xmm7
movaps XMMWORD PTR [rsp+192], xmm8
movaps XMMWORD PTR [rsp+208], xmm9
movaps XMMWORD PTR [rsp+224], xmm10
movaps XMMWORD PTR [rsp+240], xmm11
movaps XMMWORD PTR [rsp+256], xmm12
movaps XMMWORD PTR [rsp+272], xmm13
movaps XMMWORD PTR [rsp+288], xmm14
movaps XMMWORD PTR [rsp+304], xmm15
movd xmm7, rax
mov rax, QWORD PTR [r8+80]
xor rax, QWORD PTR [r8+64]
movaps xmm1, XMMWORD PTR [rdx+96]
movaps xmm2, XMMWORD PTR [r8+96]
movaps XMMWORD PTR [rsp], xmm1
movaps XMMWORD PTR [rsp+16], xmm2
mov r8d, r15d
punpcklqdq xmm7, xmm0
movd xmm0, rcx
mov rcx, QWORD PTR [rdx+56]
xor rcx, QWORD PTR [rdx+24]
movd xmm9, rax
mov QWORD PTR [rsp+128], rsi
mov rax, QWORD PTR [rdx+48]
xor rax, QWORD PTR [rdx+16]
punpcklqdq xmm9, xmm0
movd xmm0, rcx
mov rcx, QWORD PTR [rdx+88]
xor rcx, QWORD PTR [rdx+72]
movd xmm8, rax
mov QWORD PTR [rsp+136], rdi
mov rax, QWORD PTR [rdx+80]
xor rax, QWORD PTR [rdx+64]
punpcklqdq xmm8, xmm0
and r8d, 2097136
movd xmm0, rcx
mov r11d, 524288
movd xmm10, rax
punpcklqdq xmm10, xmm0
movd xmm14, QWORD PTR [rsp+128]
movd xmm15, QWORD PTR [rsp+136]
ALIGN(64)
FN_PREFIX(CryptonightR_template_double_mainloop):
movdqu xmm6, XMMWORD PTR [rbx+rsi]
movd xmm0, r12
mov ecx, ebx
movd xmm3, r14
punpcklqdq xmm3, xmm0
xor ebx, 16
aesenc xmm6, xmm3
movd xmm4, r15
movdqu xmm0, XMMWORD PTR [rbx+rsi]
pxor xmm6, xmm0
xor ebx, 48
paddq xmm0, xmm7
movdqu xmm1, XMMWORD PTR [rbx+rsi]
pxor xmm6, xmm1
movdqu XMMWORD PTR [rbx+rsi], xmm0
paddq xmm1, xmm3
xor ebx, 16
mov eax, ebx
xor rax, 32
movdqu xmm0, XMMWORD PTR [rbx+rsi]
pxor xmm6, xmm0
movd rdx, xmm6
movdqu XMMWORD PTR [rbx+rsi], xmm1
paddq xmm0, xmm9
movdqu XMMWORD PTR [rax+rsi], xmm0
movdqa xmm0, xmm6
pxor xmm0, xmm7
movdqu XMMWORD PTR [rcx+rsi], xmm0
mov esi, edx
movdqu xmm5, XMMWORD PTR [r8+rdi]
and esi, 2097136
mov ecx, r8d
movd xmm0, r13
punpcklqdq xmm4, xmm0
xor r8d, 16
aesenc xmm5, xmm4
movdqu xmm0, XMMWORD PTR [r8+rdi]
pxor xmm5, xmm0
xor r8d, 48
paddq xmm0, xmm8
movdqu xmm1, XMMWORD PTR [r8+rdi]
pxor xmm5, xmm1
movdqu XMMWORD PTR [r8+rdi], xmm0
paddq xmm1, xmm4
xor r8d, 16
mov eax, r8d
xor rax, 32
movdqu xmm0, XMMWORD PTR [r8+rdi]
pxor xmm5, xmm0
movdqu XMMWORD PTR [r8+rdi], xmm1
paddq xmm0, xmm10
movdqu XMMWORD PTR [rax+rdi], xmm0
movdqa xmm0, xmm5
pxor xmm0, xmm8
movdqu XMMWORD PTR [rcx+rdi], xmm0
movd rdi, xmm5
movd rcx, xmm14
mov ebp, edi
mov r8, QWORD PTR [rcx+rsi]
mov r10, QWORD PTR [rcx+rsi+8]
lea r9, QWORD PTR [rcx+rsi]
xor esi, 16
movd xmm0, rsp
movd xmm1, rsi
movd xmm2, rdi
movd xmm11, rbp
movd xmm12, r15
movd xmm13, rdx
mov [rsp+104], rcx
mov [rsp+112], r9
mov ebx, DWORD PTR [rsp+16]
mov esi, DWORD PTR [rsp+20]
mov edi, DWORD PTR [rsp+24]
mov ebp, DWORD PTR [rsp+28]
lea eax, [ebx+esi]
lea edx, [edi+ebp]
shl rdx, 32
or rax, rdx
xor r8, rax
movd esp, xmm3
pextrd r15d, xmm3, 2
movd eax, xmm7
movd edx, xmm9
pextrd r9d, xmm9, 2
FN_PREFIX(CryptonightR_template_double_part2):
mov eax, edi
mov edx, ebp
shl rdx, 32
or rax, rdx
xor r14, rax
mov eax, ebx
mov edx, esi
shl rdx, 32
or rax, rdx
xor r12, rax
movd rsp, xmm0
mov DWORD PTR [rsp+16], ebx
mov DWORD PTR [rsp+20], esi
mov DWORD PTR [rsp+24], edi
mov DWORD PTR [rsp+28], ebp
movd rsi, xmm1
movd rdi, xmm2
movd rbp, xmm11
movd r15, xmm12
movd rdx, xmm13
mov rcx, [rsp+104]
mov r9, [rsp+112]
mov rbx, r8
mov rax, r8
mul rdx
and ebp, 2097136
mov r8, rax
movdqu xmm1, XMMWORD PTR [rcx+rsi]
pxor xmm6, xmm1
xor esi, 48
paddq xmm1, xmm7
movdqu xmm2, XMMWORD PTR [rsi+rcx]
pxor xmm6, xmm2
paddq xmm2, xmm3
movdqu XMMWORD PTR [rsi+rcx], xmm1
xor esi, 16
mov eax, esi
mov rsi, rcx
movdqu xmm0, XMMWORD PTR [rax+rcx]
pxor xmm6, xmm0
movdqu XMMWORD PTR [rax+rcx], xmm2
paddq xmm0, xmm9
add r12, r8
xor rax, 32
add r14, rdx
movdqa xmm9, xmm7
movdqa xmm7, xmm6
movdqu XMMWORD PTR [rax+rcx], xmm0
mov QWORD PTR [r9+8], r12
xor r12, r10
mov QWORD PTR [r9], r14
movd rcx, xmm15
xor r14, rbx
mov r10d, ebp
mov ebx, r14d
xor ebp, 16
and ebx, 2097136
mov r8, QWORD PTR [r10+rcx]
mov r9, QWORD PTR [r10+rcx+8]
movd xmm0, rsp
movd xmm1, rbx
movd xmm2, rsi
movd xmm11, rdi
movd xmm12, rbp
movd xmm13, r15
mov [rsp+104], rcx
mov [rsp+112], r9
mov ebx, DWORD PTR [rsp]
mov esi, DWORD PTR [rsp+4]
mov edi, DWORD PTR [rsp+8]
mov ebp, DWORD PTR [rsp+12]
lea eax, [ebx+esi]
lea edx, [edi+ebp]
shl rdx, 32
or rax, rdx
xor r8, rax
movd xmm3, r8
movd esp, xmm4
pextrd r15d, xmm4, 2
movd eax, xmm8
movd edx, xmm10
pextrd r9d, xmm10, 2
FN_PREFIX(CryptonightR_template_double_part3):
movd r15, xmm13
mov eax, edi
mov edx, ebp
shl rdx, 32
or rax, rdx
xor r15, rax
mov eax, ebx
mov edx, esi
shl rdx, 32
or rax, rdx
xor r13, rax
movd rsp, xmm0
mov DWORD PTR [rsp], ebx
mov DWORD PTR [rsp+4], esi
mov DWORD PTR [rsp+8], edi
mov DWORD PTR [rsp+12], ebp
movd rbx, xmm1
movd rsi, xmm2
movd rdi, xmm11
movd rbp, xmm12
mov rcx, [rsp+104]
mov r9, [rsp+112]
mov rax, r8
mul rdi
mov rdi, rcx
mov r8, rax
movdqu xmm1, XMMWORD PTR [rbp+rcx]
pxor xmm5, xmm1
xor ebp, 48
paddq xmm1, xmm8
add r13, r8
movdqu xmm2, XMMWORD PTR [rbp+rcx]
pxor xmm5, xmm2
add r15, rdx
movdqu XMMWORD PTR [rbp+rcx], xmm1
paddq xmm2, xmm4
xor ebp, 16
mov eax, ebp
xor rax, 32
movdqu xmm0, XMMWORD PTR [rbp+rcx]
pxor xmm5, xmm0
movdqu XMMWORD PTR [rbp+rcx], xmm2
paddq xmm0, xmm10
movdqu XMMWORD PTR [rax+rcx], xmm0
movd rax, xmm3
movdqa xmm10, xmm8
mov QWORD PTR [r10+rcx], r15
movdqa xmm8, xmm5
xor r15, rax
mov QWORD PTR [r10+rcx+8], r13
mov r8d, r15d
xor r13, r9
and r8d, 2097136
dec r11d
jnz FN_PREFIX(CryptonightR_template_double_mainloop)
FN_PREFIX(CryptonightR_template_double_part4):
mov rbx, QWORD PTR [rsp+400]
movaps xmm6, XMMWORD PTR [rsp+160]
movaps xmm7, XMMWORD PTR [rsp+176]
movaps xmm8, XMMWORD PTR [rsp+192]
movaps xmm9, XMMWORD PTR [rsp+208]
movaps xmm10, XMMWORD PTR [rsp+224]
movaps xmm11, XMMWORD PTR [rsp+240]
movaps xmm12, XMMWORD PTR [rsp+256]
movaps xmm13, XMMWORD PTR [rsp+272]
movaps xmm14, XMMWORD PTR [rsp+288]
movaps xmm15, XMMWORD PTR [rsp+304]
add rsp, 320
pop r15
pop r14
pop r13
pop r12
pop rdi
pop rsi
pop rbp
ret 0
FN_PREFIX(CryptonightR_template_double_end):

View file

@ -1,266 +0,0 @@
PUBLIC FN_PREFIX(CryptonightWOW_soft_aes_template_part1)
PUBLIC FN_PREFIX(CryptonightWOW_soft_aes_template_mainloop)
PUBLIC FN_PREFIX(CryptonightWOW_soft_aes_template_part2)
PUBLIC FN_PREFIX(CryptonightWOW_soft_aes_template_part3)
PUBLIC FN_PREFIX(CryptonightWOW_soft_aes_template_end)
ALIGN(64)
FN_PREFIX(CryptonightWOW_soft_aes_template_part1):
mov QWORD PTR [rsp+8], rcx
push rbx
push rbp
push rsi
push rdi
push r12
push r13
push r14
push r15
sub rsp, 232
mov eax, [rcx+96]
mov ebx, [rcx+100]
mov esi, [rcx+104]
mov edx, [rcx+108]
mov [rsp+144], eax
mov [rsp+148], ebx
mov [rsp+152], esi
mov [rsp+156], edx
mov rax, QWORD PTR [rcx+48]
mov r10, rcx
xor rax, QWORD PTR [rcx+16]
mov r8, QWORD PTR [rcx+32]
xor r8, QWORD PTR [rcx]
mov r9, QWORD PTR [rcx+40]
xor r9, QWORD PTR [rcx+8]
movd xmm4, rax
mov rdx, QWORD PTR [rcx+56]
xor rdx, QWORD PTR [rcx+24]
mov r11, QWORD PTR [rcx+224]
mov rcx, QWORD PTR [rcx+88]
xor rcx, QWORD PTR [r10+72]
mov rax, QWORD PTR [r10+80]
movd xmm0, rdx
xor rax, QWORD PTR [r10+64]
movaps XMMWORD PTR [rsp+16], xmm6
movaps XMMWORD PTR [rsp+32], xmm7
movaps XMMWORD PTR [rsp+48], xmm8
movaps XMMWORD PTR [rsp+64], xmm9
movaps XMMWORD PTR [rsp+80], xmm10
movaps XMMWORD PTR [rsp+96], xmm11
movaps XMMWORD PTR [rsp+112], xmm12
movaps XMMWORD PTR [rsp+128], xmm13
movd xmm5, rax
mov rax, r8
punpcklqdq xmm4, xmm0
and eax, 2097136
movd xmm10, QWORD PTR [r10+96]
movd xmm0, rcx
mov rcx, QWORD PTR [r10+104]
xorps xmm9, xmm9
mov QWORD PTR [rsp+328], rax
movd xmm12, r11
mov QWORD PTR [rsp+320], r9
punpcklqdq xmm5, xmm0
movd xmm13, rcx
mov r12d, 524288
ALIGN(64)
FN_PREFIX(CryptonightWOW_soft_aes_template_mainloop):
movd xmm11, r12d
mov r12, QWORD PTR [r10+272]
lea r13, QWORD PTR [rax+r11]
mov esi, DWORD PTR [r13]
movd xmm0, r9
mov r10d, DWORD PTR [r13+4]
movd xmm7, r8
mov ebp, DWORD PTR [r13+12]
mov r14d, DWORD PTR [r13+8]
mov rdx, QWORD PTR [rsp+328]
movzx ecx, sil
shr esi, 8
punpcklqdq xmm7, xmm0
mov r15d, DWORD PTR [r12+rcx*4]
movzx ecx, r10b
shr r10d, 8
mov edi, DWORD PTR [r12+rcx*4]
movzx ecx, r14b
shr r14d, 8
mov ebx, DWORD PTR [r12+rcx*4]
movzx ecx, bpl
shr ebp, 8
mov r9d, DWORD PTR [r12+rcx*4]
movzx ecx, r10b
shr r10d, 8
xor r15d, DWORD PTR [r12+rcx*4+1024]
movzx ecx, r14b
shr r14d, 8
mov eax, r14d
shr eax, 8
xor edi, DWORD PTR [r12+rcx*4+1024]
add eax, 256
movzx ecx, bpl
shr ebp, 8
xor ebx, DWORD PTR [r12+rcx*4+1024]
movzx ecx, sil
shr esi, 8
xor r9d, DWORD PTR [r12+rcx*4+1024]
add r12, 2048
movzx ecx, r10b
shr r10d, 8
add r10d, 256
mov r11d, DWORD PTR [r12+rax*4]
xor r11d, DWORD PTR [r12+rcx*4]
xor r11d, r9d
movzx ecx, sil
mov r10d, DWORD PTR [r12+r10*4]
shr esi, 8
add esi, 256
xor r10d, DWORD PTR [r12+rcx*4]
movzx ecx, bpl
xor r10d, ebx
shr ebp, 8
movd xmm1, r11d
add ebp, 256
movd r11, xmm12
mov r9d, DWORD PTR [r12+rcx*4]
xor r9d, DWORD PTR [r12+rsi*4]
mov eax, DWORD PTR [r12+rbp*4]
xor r9d, edi
movzx ecx, r14b
movd xmm0, r10d
movd xmm2, r9d
xor eax, DWORD PTR [r12+rcx*4]
mov rcx, rdx
xor eax, r15d
punpckldq xmm2, xmm1
xor rcx, 16
movd xmm6, eax
mov rax, rdx
punpckldq xmm6, xmm0
xor rax, 32
punpckldq xmm6, xmm2
xor rdx, 48
movdqu xmm2, XMMWORD PTR [rcx+r11]
pxor xmm6, xmm7
paddq xmm2, xmm4
movdqu xmm1, XMMWORD PTR [rax+r11]
movdqu xmm0, XMMWORD PTR [rdx+r11]
paddq xmm0, xmm5
movdqu XMMWORD PTR [rcx+r11], xmm0
movdqu XMMWORD PTR [rax+r11], xmm2
movd rcx, xmm13
paddq xmm1, xmm7
movdqu XMMWORD PTR [rdx+r11], xmm1
movd rdi, xmm6
mov r10, rdi
and r10d, 2097136
movdqa xmm0, xmm6
pxor xmm0, xmm4
movdqu XMMWORD PTR [r13], xmm0
mov ebx, [rsp+144]
mov ebp, [rsp+152]
add ebx, [rsp+148]
add ebp, [rsp+156]
shl rbp, 32
or rbx, rbp
xor rbx, QWORD PTR [r10+r11]
lea r14, QWORD PTR [r10+r11]
mov rbp, QWORD PTR [r14+8]
mov [rsp+160], rbx
mov [rsp+168], rdi
mov [rsp+176], rbp
mov [rsp+184], r10
mov r10, rsp
mov ebx, [rsp+144]
mov esi, [rsp+148]
mov edi, [rsp+152]
mov ebp, [rsp+156]
movd esp, xmm7
movaps xmm0, xmm7
psrldq xmm0, 8
movd r15d, xmm0
movd eax, xmm4
movd edx, xmm5
FN_PREFIX(CryptonightWOW_soft_aes_template_part2):
mov rsp, r10
mov [rsp+144], ebx
mov [rsp+148], esi
mov [rsp+152], edi
mov [rsp+156], ebp
mov rbx, [rsp+160]
mov rdi, [rsp+168]
mov rbp, [rsp+176]
mov r10, [rsp+184]
mov r9, r10
xor r9, 16
mov rcx, r10
xor rcx, 32
xor r10, 48
mov rax, rbx
mul rdi
movdqu xmm2, XMMWORD PTR [r9+r11]
movdqu xmm1, XMMWORD PTR [rcx+r11]
paddq xmm1, xmm7
movd xmm0, rax
movd xmm3, rdx
xor rax, QWORD PTR [r11+rcx+8]
xor rdx, QWORD PTR [rcx+r11]
punpcklqdq xmm3, xmm0
add r8, rdx
movdqu xmm0, XMMWORD PTR [r10+r11]
pxor xmm2, xmm3
paddq xmm0, xmm5
paddq xmm2, xmm4
movdqu XMMWORD PTR [r9+r11], xmm0
movdqa xmm5, xmm4
mov r9, QWORD PTR [rsp+320]
movdqa xmm4, xmm6
add r9, rax
movdqu XMMWORD PTR [rcx+r11], xmm2
movdqu XMMWORD PTR [r10+r11], xmm1
mov r10, QWORD PTR [rsp+304]
movd r12d, xmm11
mov QWORD PTR [r14], r8
xor r8, rbx
mov rax, r8
mov QWORD PTR [r14+8], r9
and eax, 2097136
xor r9, rbp
mov QWORD PTR [rsp+320], r9
mov QWORD PTR [rsp+328], rax
sub r12d, 1
jne FN_PREFIX(CryptonightWOW_soft_aes_template_mainloop)
FN_PREFIX(CryptonightWOW_soft_aes_template_part3):
movaps xmm6, XMMWORD PTR [rsp+16]
movaps xmm7, XMMWORD PTR [rsp+32]
movaps xmm8, XMMWORD PTR [rsp+48]
movaps xmm9, XMMWORD PTR [rsp+64]
movaps xmm10, XMMWORD PTR [rsp+80]
movaps xmm11, XMMWORD PTR [rsp+96]
movaps xmm12, XMMWORD PTR [rsp+112]
movaps xmm13, XMMWORD PTR [rsp+128]
add rsp, 232
pop r15
pop r14
pop r13
pop r12
pop rdi
pop rsi
pop rbp
pop rbx
ret
FN_PREFIX(CryptonightWOW_soft_aes_template_end):

View file

@ -1,486 +0,0 @@
PUBLIC FN_PREFIX(CryptonightWOW_template_part1)
PUBLIC FN_PREFIX(CryptonightWOW_template_mainloop)
PUBLIC FN_PREFIX(CryptonightWOW_template_part2)
PUBLIC FN_PREFIX(CryptonightWOW_template_part3)
PUBLIC FN_PREFIX(CryptonightWOW_template_end)
PUBLIC FN_PREFIX(CryptonightWOW_template_double_part1)
PUBLIC FN_PREFIX(CryptonightWOW_template_double_mainloop)
PUBLIC FN_PREFIX(CryptonightWOW_template_double_part2)
PUBLIC FN_PREFIX(CryptonightWOW_template_double_part3)
PUBLIC FN_PREFIX(CryptonightWOW_template_double_part4)
PUBLIC FN_PREFIX(CryptonightWOW_template_double_end)
ALIGN(64)
FN_PREFIX(CryptonightWOW_template_part1):
mov QWORD PTR [rsp+16], rbx
mov QWORD PTR [rsp+24], rbp
mov QWORD PTR [rsp+32], rsi
push r10
push r11
push r12
push r13
push r14
push r15
push rdi
sub rsp, 64
mov r12, rcx
mov r8, QWORD PTR [r12+32]
mov rdx, r12
xor r8, QWORD PTR [r12]
mov r15, QWORD PTR [r12+40]
mov r9, r8
xor r15, QWORD PTR [r12+8]
mov r11, QWORD PTR [r12+224]
mov r12, QWORD PTR [r12+56]
xor r12, QWORD PTR [rdx+24]
mov rax, QWORD PTR [rdx+48]
xor rax, QWORD PTR [rdx+16]
movaps XMMWORD PTR [rsp+48], xmm6
movd xmm0, r12
movaps XMMWORD PTR [rsp+32], xmm7
movaps XMMWORD PTR [rsp+16], xmm8
movaps XMMWORD PTR [rsp], xmm9
mov r12, QWORD PTR [rdx+88]
xor r12, QWORD PTR [rdx+72]
movd xmm6, rax
mov rax, QWORD PTR [rdx+80]
xor rax, QWORD PTR [rdx+64]
punpcklqdq xmm6, xmm0
and r9d, 2097136
movd xmm0, r12
movd xmm7, rax
punpcklqdq xmm7, xmm0
mov r10d, r9d
movd xmm9, rsp
mov rsp, r8
mov r8d, 524288
mov ebx, [rdx+96]
mov esi, [rdx+100]
mov edi, [rdx+104]
mov ebp, [rdx+108]
ALIGN(64)
FN_PREFIX(CryptonightWOW_template_mainloop):
movdqa xmm5, XMMWORD PTR [r9+r11]
movd xmm0, r15
movd xmm4, rsp
punpcklqdq xmm4, xmm0
lea rdx, QWORD PTR [r9+r11]
aesenc xmm5, xmm4
movd r10d, xmm5
and r10d, 2097136
mov r12d, r9d
mov eax, r9d
xor r9d, 48
xor r12d, 16
xor eax, 32
movdqu xmm0, XMMWORD PTR [r9+r11]
movdqu xmm2, XMMWORD PTR [r12+r11]
movdqu xmm1, XMMWORD PTR [rax+r11]
paddq xmm0, xmm7
paddq xmm2, xmm6
paddq xmm1, xmm4
movdqu XMMWORD PTR [r12+r11], xmm0
movd r12, xmm5
movdqu XMMWORD PTR [rax+r11], xmm2
movdqu XMMWORD PTR [r9+r11], xmm1
movdqa xmm0, xmm5
pxor xmm0, xmm6
movdqu XMMWORD PTR [rdx], xmm0
lea r13d, [ebx+esi]
lea edx, [edi+ebp]
shl rdx, 32
or r13, rdx
xor r13, QWORD PTR [r10+r11]
mov r14, QWORD PTR [r10+r11+8]
movd eax, xmm6
movd edx, xmm7
pextrd r9d, xmm7, 2
FN_PREFIX(CryptonightWOW_template_part2):
mov rax, r13
mul r12
movd xmm0, rax
movd xmm3, rdx
punpcklqdq xmm3, xmm0
mov r9d, r10d
mov r12d, r10d
xor r9d, 16
xor r12d, 32
xor r10d, 48
movdqa xmm1, XMMWORD PTR [r12+r11]
xor rdx, QWORD PTR [r12+r11]
xor rax, QWORD PTR [r11+r12+8]
movdqa xmm2, XMMWORD PTR [r9+r11]
pxor xmm3, xmm2
paddq xmm7, XMMWORD PTR [r10+r11]
paddq xmm1, xmm4
paddq xmm3, xmm6
movdqu XMMWORD PTR [r9+r11], xmm7
movdqu XMMWORD PTR [r12+r11], xmm3
movdqu XMMWORD PTR [r10+r11], xmm1
movdqa xmm7, xmm6
add r15, rax
add rsp, rdx
xor r10, 48
mov QWORD PTR [r10+r11], rsp
xor rsp, r13
mov r9d, esp
mov QWORD PTR [r10+r11+8], r15
and r9d, 2097136
xor r15, r14
movdqa xmm6, xmm5
dec r8d
jnz FN_PREFIX(CryptonightWOW_template_mainloop)
FN_PREFIX(CryptonightWOW_template_part3):
movd rsp, xmm9
mov rbx, QWORD PTR [rsp+136]
mov rbp, QWORD PTR [rsp+144]
mov rsi, QWORD PTR [rsp+152]
movaps xmm6, XMMWORD PTR [rsp+48]
movaps xmm7, XMMWORD PTR [rsp+32]
movaps xmm8, XMMWORD PTR [rsp+16]
movaps xmm9, XMMWORD PTR [rsp]
add rsp, 64
pop rdi
pop r15
pop r14
pop r13
pop r12
pop r11
pop r10
ret 0
FN_PREFIX(CryptonightWOW_template_end):
ALIGN(64)
FN_PREFIX(CryptonightWOW_template_double_part1):
mov QWORD PTR [rsp+24], rbx
push rbp
push rsi
push rdi
push r12
push r13
push r14
push r15
sub rsp, 320
mov r14, QWORD PTR [rcx+32]
mov r8, rcx
xor r14, QWORD PTR [rcx]
mov r12, QWORD PTR [rcx+40]
mov ebx, r14d
mov rsi, QWORD PTR [rcx+224]
and ebx, 2097136
xor r12, QWORD PTR [rcx+8]
mov rcx, QWORD PTR [rcx+56]
xor rcx, QWORD PTR [r8+24]
mov rax, QWORD PTR [r8+48]
xor rax, QWORD PTR [r8+16]
mov r15, QWORD PTR [rdx+32]
xor r15, QWORD PTR [rdx]
movd xmm0, rcx
mov rcx, QWORD PTR [r8+88]
xor rcx, QWORD PTR [r8+72]
mov r13, QWORD PTR [rdx+40]
mov rdi, QWORD PTR [rdx+224]
xor r13, QWORD PTR [rdx+8]
movaps XMMWORD PTR [rsp+160], xmm6
movaps XMMWORD PTR [rsp+176], xmm7
movaps XMMWORD PTR [rsp+192], xmm8
movaps XMMWORD PTR [rsp+208], xmm9
movaps XMMWORD PTR [rsp+224], xmm10
movaps XMMWORD PTR [rsp+240], xmm11
movaps XMMWORD PTR [rsp+256], xmm12
movaps XMMWORD PTR [rsp+272], xmm13
movaps XMMWORD PTR [rsp+288], xmm14
movaps XMMWORD PTR [rsp+304], xmm15
movd xmm7, rax
mov rax, QWORD PTR [r8+80]
xor rax, QWORD PTR [r8+64]
movaps xmm1, XMMWORD PTR [rdx+96]
movaps xmm2, XMMWORD PTR [r8+96]
movaps XMMWORD PTR [rsp], xmm1
movaps XMMWORD PTR [rsp+16], xmm2
mov r8d, r15d
punpcklqdq xmm7, xmm0
movd xmm0, rcx
mov rcx, QWORD PTR [rdx+56]
xor rcx, QWORD PTR [rdx+24]
movd xmm9, rax
mov QWORD PTR [rsp+128], rsi
mov rax, QWORD PTR [rdx+48]
xor rax, QWORD PTR [rdx+16]
punpcklqdq xmm9, xmm0
movd xmm0, rcx
mov rcx, QWORD PTR [rdx+88]
xor rcx, QWORD PTR [rdx+72]
movd xmm8, rax
mov QWORD PTR [rsp+136], rdi
mov rax, QWORD PTR [rdx+80]
xor rax, QWORD PTR [rdx+64]
punpcklqdq xmm8, xmm0
and r8d, 2097136
movd xmm0, rcx
mov r11d, 524288
movd xmm10, rax
punpcklqdq xmm10, xmm0
movd xmm14, QWORD PTR [rsp+128]
movd xmm15, QWORD PTR [rsp+136]
ALIGN(64)
FN_PREFIX(CryptonightWOW_template_double_mainloop):
movdqu xmm6, XMMWORD PTR [rbx+rsi]
movd xmm0, r12
mov ecx, ebx
movd xmm3, r14
punpcklqdq xmm3, xmm0
xor ebx, 16
aesenc xmm6, xmm3
movd rdx, xmm6
movd xmm4, r15
movdqu xmm0, XMMWORD PTR [rbx+rsi]
xor ebx, 48
paddq xmm0, xmm7
movdqu xmm1, XMMWORD PTR [rbx+rsi]
movdqu XMMWORD PTR [rbx+rsi], xmm0
paddq xmm1, xmm3
xor ebx, 16
mov eax, ebx
xor rax, 32
movdqu xmm0, XMMWORD PTR [rbx+rsi]
movdqu XMMWORD PTR [rbx+rsi], xmm1
paddq xmm0, xmm9
movdqu XMMWORD PTR [rax+rsi], xmm0
movdqa xmm0, xmm6
pxor xmm0, xmm7
movdqu XMMWORD PTR [rcx+rsi], xmm0
mov esi, edx
movdqu xmm5, XMMWORD PTR [r8+rdi]
and esi, 2097136
mov ecx, r8d
movd xmm0, r13
punpcklqdq xmm4, xmm0
xor r8d, 16
aesenc xmm5, xmm4
movdqu xmm0, XMMWORD PTR [r8+rdi]
xor r8d, 48
paddq xmm0, xmm8
movdqu xmm1, XMMWORD PTR [r8+rdi]
movdqu XMMWORD PTR [r8+rdi], xmm0
paddq xmm1, xmm4
xor r8d, 16
mov eax, r8d
xor rax, 32
movdqu xmm0, XMMWORD PTR [r8+rdi]
movdqu XMMWORD PTR [r8+rdi], xmm1
paddq xmm0, xmm10
movdqu XMMWORD PTR [rax+rdi], xmm0
movdqa xmm0, xmm5
pxor xmm0, xmm8
movdqu XMMWORD PTR [rcx+rdi], xmm0
movd rdi, xmm5
movd rcx, xmm14
mov ebp, edi
mov r8, QWORD PTR [rcx+rsi]
mov r10, QWORD PTR [rcx+rsi+8]
lea r9, QWORD PTR [rcx+rsi]
xor esi, 16
movd xmm0, rsp
movd xmm1, rsi
movd xmm2, rdi
movd xmm11, rbp
movd xmm12, r15
movd xmm13, rdx
mov [rsp+104], rcx
mov [rsp+112], r9
mov ebx, DWORD PTR [rsp+16]
mov esi, DWORD PTR [rsp+20]
mov edi, DWORD PTR [rsp+24]
mov ebp, DWORD PTR [rsp+28]
lea eax, [ebx+esi]
lea edx, [edi+ebp]
shl rdx, 32
or rax, rdx
xor r8, rax
movd esp, xmm3
pextrd r15d, xmm3, 2
movd eax, xmm7
movd edx, xmm9
pextrd r9d, xmm9, 2
FN_PREFIX(CryptonightWOW_template_double_part2):
movd rsp, xmm0
mov DWORD PTR [rsp+16], ebx
mov DWORD PTR [rsp+20], esi
mov DWORD PTR [rsp+24], edi
mov DWORD PTR [rsp+28], ebp
movd rsi, xmm1
movd rdi, xmm2
movd rbp, xmm11
movd r15, xmm12
movd rdx, xmm13
mov rcx, [rsp+104]
mov r9, [rsp+112]
mov rbx, r8
mov rax, r8
mul rdx
and ebp, 2097136
mov r8, rax
movd xmm1, rdx
movd xmm0, r8
punpcklqdq xmm1, xmm0
pxor xmm1, XMMWORD PTR [rcx+rsi]
xor esi, 48
paddq xmm1, xmm7
movdqu xmm2, XMMWORD PTR [rsi+rcx]
xor rdx, QWORD PTR [rsi+rcx]
paddq xmm2, xmm3
xor r8, QWORD PTR [rsi+rcx+8]
movdqu XMMWORD PTR [rsi+rcx], xmm1
xor esi, 16
mov eax, esi
mov rsi, rcx
movdqu xmm0, XMMWORD PTR [rax+rcx]
movdqu XMMWORD PTR [rax+rcx], xmm2
paddq xmm0, xmm9
add r12, r8
xor rax, 32
add r14, rdx
movdqa xmm9, xmm7
movdqa xmm7, xmm6
movdqu XMMWORD PTR [rax+rcx], xmm0
mov QWORD PTR [r9+8], r12
xor r12, r10
mov QWORD PTR [r9], r14
movd rcx, xmm15
xor r14, rbx
mov r10d, ebp
mov ebx, r14d
xor ebp, 16
and ebx, 2097136
mov r8, QWORD PTR [r10+rcx]
mov r9, QWORD PTR [r10+rcx+8]
movd xmm0, rsp
movd xmm1, rbx
movd xmm2, rsi
movd xmm11, rdi
movd xmm12, rbp
movd xmm13, r15
mov [rsp+104], rcx
mov [rsp+112], r9
mov ebx, DWORD PTR [rsp]
mov esi, DWORD PTR [rsp+4]
mov edi, DWORD PTR [rsp+8]
mov ebp, DWORD PTR [rsp+12]
lea eax, [ebx+esi]
lea edx, [edi+ebp]
shl rdx, 32
or rax, rdx
xor r8, rax
movd xmm3, r8
movd esp, xmm4
pextrd r15d, xmm4, 2
movd eax, xmm8
movd edx, xmm10
pextrd r9d, xmm10, 2
FN_PREFIX(CryptonightWOW_template_double_part3):
movd rsp, xmm0
mov DWORD PTR [rsp], ebx
mov DWORD PTR [rsp+4], esi
mov DWORD PTR [rsp+8], edi
mov DWORD PTR [rsp+12], ebp
movd rbx, xmm1
movd rsi, xmm2
movd rdi, xmm11
movd rbp, xmm12
movd r15, xmm13
mov rcx, [rsp+104]
mov r9, [rsp+112]
mov rax, r8
mul rdi
movd xmm1, rdx
movd xmm0, rax
punpcklqdq xmm1, xmm0
mov rdi, rcx
mov r8, rax
pxor xmm1, XMMWORD PTR [rbp+rcx]
xor ebp, 48
paddq xmm1, xmm8
xor r8, QWORD PTR [rbp+rcx+8]
xor rdx, QWORD PTR [rbp+rcx]
add r13, r8
movdqu xmm2, XMMWORD PTR [rbp+rcx]
add r15, rdx
movdqu XMMWORD PTR [rbp+rcx], xmm1
paddq xmm2, xmm4
xor ebp, 16
mov eax, ebp
xor rax, 32
movdqu xmm0, XMMWORD PTR [rbp+rcx]
movdqu XMMWORD PTR [rbp+rcx], xmm2
paddq xmm0, xmm10
movdqu XMMWORD PTR [rax+rcx], xmm0
movd rax, xmm3
movdqa xmm10, xmm8
mov QWORD PTR [r10+rcx], r15
movdqa xmm8, xmm5
xor r15, rax
mov QWORD PTR [r10+rcx+8], r13
mov r8d, r15d
xor r13, r9
and r8d, 2097136
dec r11d
jnz FN_PREFIX(CryptonightWOW_template_double_mainloop)
FN_PREFIX(CryptonightWOW_template_double_part4):
mov rbx, QWORD PTR [rsp+400]
movaps xmm6, XMMWORD PTR [rsp+160]
movaps xmm7, XMMWORD PTR [rsp+176]
movaps xmm8, XMMWORD PTR [rsp+192]
movaps xmm9, XMMWORD PTR [rsp+208]
movaps xmm10, XMMWORD PTR [rsp+224]
movaps xmm11, XMMWORD PTR [rsp+240]
movaps xmm12, XMMWORD PTR [rsp+256]
movaps xmm13, XMMWORD PTR [rsp+272]
movaps xmm14, XMMWORD PTR [rsp+288]
movaps xmm15, XMMWORD PTR [rsp+304]
add rsp, 320
pop r15
pop r14
pop r13
pop r12
pop rdi
pop rsi
pop rbp
ret 0
FN_PREFIX(CryptonightWOW_template_double_end):

View file

@ -5,7 +5,8 @@
* Copyright 2014-2016 Wolf9466 <https://github.com/OhGodAPet>
* Copyright 2016 Jay D Dee <jayddee246@gmail.com>
* Copyright 2017-2018 XMR-Stak <https://github.com/fireice-uk>, <https://github.com/psychocrypt>
* Copyright 2016-2018 XMRig <https://github.com/xmrig>, <support@xmrig.com>
* Copyright 2018-2019 SChernykh <https://github.com/SChernykh>
* Copyright 2016-2019 XMRig <https://github.com/xmrig>, <support@xmrig.com>
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
@ -21,8 +22,8 @@
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
#ifndef __IWORKER_H__
#define __IWORKER_H__
#ifndef XMRIG_IWORKER_H
#define XMRIG_IWORKER_H
#include <stdint.h>
@ -31,7 +32,7 @@
class IWorker
{
public:
virtual ~IWorker() {}
virtual ~IWorker() = default;
virtual bool selfTest() = 0;
virtual size_t id() const = 0;
@ -41,4 +42,4 @@ public:
};
#endif // __IWORKER_H__
#endif // XMRIG_IWORKER_H

View file

@ -25,7 +25,6 @@
#include <thread>
#include <sstream>
#include "crypto/CryptoNight_test.h"
@ -56,23 +55,16 @@ bool MultiWorker<N>::selfTest()
using namespace xmrig;
if (m_thread->algorithm() == CRYPTONIGHT) {
if (!verify2(VARIANT_WOW, test_input_WOW)) {
LOG_WARN("CryptonightR (Wownero) self-test failed");
return false;
}
if (!verify2(VARIANT_4, test_input_R)) {
LOG_WARN("CryptonightR self-test failed");
return false;
}
const bool rc = verify(VARIANT_0, test_output_v0) &&
verify(VARIANT_1, test_output_v1) &&
verify(VARIANT_2, test_output_v2) &&
verify(VARIANT_XTL, test_output_xtl) &&
verify(VARIANT_MSR, test_output_msr) &&
verify(VARIANT_XAO, test_output_xao) &&
verify(VARIANT_RTO, test_output_rto) &&
verify(VARIANT_HALF, test_output_half);
const bool rc = verify(VARIANT_0, test_output_v0) &&
verify(VARIANT_1, test_output_v1) &&
verify(VARIANT_2, test_output_v2) &&
verify(VARIANT_XTL, test_output_xtl) &&
verify(VARIANT_MSR, test_output_msr) &&
verify(VARIANT_XAO, test_output_xao) &&
verify(VARIANT_RTO, test_output_rto) &&
verify(VARIANT_HALF, test_output_half) &&
verify2(VARIANT_WOW, test_output_wow) &&
verify2(VARIANT_4, test_output_r);
# ifndef XMRIG_NO_CN_GPU
if (!rc || N > 1) {
@ -179,61 +171,48 @@ bool MultiWorker<N>::verify(xmrig::Variant variant, const uint8_t *referenceValu
template<size_t N>
bool MultiWorker<N>::verify2(xmrig::Variant variant, const char *test_data)
bool MultiWorker<N>::verify2(xmrig::Variant variant, const uint8_t *referenceValue)
{
xmrig::CpuThread::cn_hash_fun func = m_thread->fn(variant);
if (!func) {
return false;
}
std::stringstream s(test_data);
std::string expected_hex;
std::string input_hex;
uint64_t height;
while (!s.eof())
{
uint8_t referenceValue[N * 32];
uint8_t input[N * 256];
s >> expected_hex;
s >> input_hex;
s >> height;
if ((expected_hex.length() != 64) || (input_hex.length() > 512))
{
return false;
for (size_t i = 0; i < (sizeof(cn_r_test_input) / sizeof(cn_r_test_input[0])); ++i) {
const size_t size = cn_r_test_input[i].size;
for (size_t k = 0; k < N; ++k) {
memcpy(m_state.blob + (k * size), cn_r_test_input[i].data, size);
}
bool err = false;
func(m_state.blob, size, m_hash, m_ctx, cn_r_test_input[i].height);
for (int i = 0; i < 32; ++i)
{
referenceValue[i] = (hf_hex2bin(expected_hex[i * 2], err) << 4) + hf_hex2bin(expected_hex[i * 2 + 1], err);
for (size_t k = 0; k < N; ++k) {
if (memcmp(m_hash + k * 32, referenceValue + i * 32, sizeof m_hash / N) != 0) {
return false;
}
}
}
const size_t input_len = input_hex.length() / 2;
for (size_t i = 0; i < input_len; ++i)
{
input[i] = (hf_hex2bin(input_hex[i * 2], err) << 4) + hf_hex2bin(input_hex[i * 2 + 1], err);
}
return true;
}
if (err)
{
return false;
}
for (size_t i = 1; i < N; ++i)
{
memcpy(input + i * input_len, input, input_len);
memcpy(referenceValue + i * 32, referenceValue, 32);
}
template<>
bool MultiWorker<1>::verify2(xmrig::Variant variant, const uint8_t *referenceValue)
{
xmrig::CpuThread::cn_hash_fun func = m_thread->fn(variant);
if (!func) {
return false;
}
func(input, input_len, m_hash, m_ctx, height);
if (memcmp(m_hash, referenceValue, sizeof m_hash) != 0)
{
for (size_t i = 0; i < (sizeof(cn_r_test_input) / sizeof(cn_r_test_input[0])); ++i) {
func(cn_r_test_input[i].data, cn_r_test_input[i].size, m_hash, m_ctx, cn_r_test_input[i].height);
if (memcmp(m_hash, referenceValue + i * 32, sizeof m_hash) != 0) {
return false;
}
}
return true;
}

View file

@ -50,7 +50,7 @@ protected:
private:
bool resume(const xmrig::Job &job);
bool verify(xmrig::Variant variant, const uint8_t *referenceValue);
bool verify2(xmrig::Variant variant, const char *test_data);
bool verify2(xmrig::Variant variant, const uint8_t *referenceValue);
void consumeJob();
void save(const xmrig::Job &job);