diff --git a/.clang-format b/.clang-format new file mode 100644 index 0000000..065dcdf --- /dev/null +++ b/.clang-format @@ -0,0 +1,225 @@ +--- +Language: Cpp +# BasedOnStyle: Mozilla +AccessModifierOffset: -2 +AlignAfterOpenBracket: Align +AlignArrayOfStructures: None +AlignConsecutiveAssignments: + Enabled: false + AcrossEmptyLines: false + AcrossComments: false + AlignCompound: false + PadOperators: true +AlignConsecutiveBitFields: + Enabled: false + AcrossEmptyLines: false + AcrossComments: false + AlignCompound: false + PadOperators: false +AlignConsecutiveDeclarations: + Enabled: false + AcrossEmptyLines: false + AcrossComments: false + AlignCompound: false + PadOperators: false +AlignConsecutiveMacros: + Enabled: false + AcrossEmptyLines: false + AcrossComments: false + AlignCompound: false + PadOperators: false +AlignEscapedNewlines: Right +AlignOperands: Align +AlignTrailingComments: + Kind: Always + OverEmptyLines: 0 +AllowAllArgumentsOnNextLine: true +AllowAllParametersOfDeclarationOnNextLine: false +AllowShortBlocksOnASingleLine: Never +AllowShortCaseLabelsOnASingleLine: false +AllowShortEnumsOnASingleLine: true +AllowShortFunctionsOnASingleLine: Inline +AllowShortIfStatementsOnASingleLine: Never +AllowShortLambdasOnASingleLine: All +AllowShortLoopsOnASingleLine: false +AlwaysBreakAfterDefinitionReturnType: TopLevel +AlwaysBreakAfterReturnType: TopLevel +AlwaysBreakBeforeMultilineStrings: false +AlwaysBreakTemplateDeclarations: Yes +AttributeMacros: + - __capability +BinPackArguments: false +BinPackParameters: false +BitFieldColonSpacing: Both +BraceWrapping: + AfterCaseLabel: false + AfterClass: true + AfterControlStatement: Never + AfterEnum: true + AfterExternBlock: true + AfterFunction: true + AfterNamespace: false + AfterObjCDeclaration: false + AfterStruct: true + AfterUnion: true + BeforeCatch: false + BeforeElse: false + BeforeLambdaBody: false + BeforeWhile: false + IndentBraces: false + SplitEmptyFunction: true + SplitEmptyRecord: false + SplitEmptyNamespace: true +BreakAfterAttributes: Never +BreakAfterJavaFieldAnnotations: false +BreakArrays: true +BreakBeforeBinaryOperators: None +BreakBeforeConceptDeclarations: Always +BreakBeforeBraces: Mozilla +BreakBeforeInlineASMColon: OnlyMultiline +BreakBeforeTernaryOperators: true +BreakConstructorInitializers: BeforeComma +BreakInheritanceList: BeforeComma +BreakStringLiterals: true +ColumnLimit: 80 +CommentPragmas: '^ IWYU pragma:' +CompactNamespaces: false +ConstructorInitializerIndentWidth: 2 +ContinuationIndentWidth: 2 +Cpp11BracedListStyle: false +DerivePointerAlignment: false +DisableFormat: false +EmptyLineAfterAccessModifier: Never +EmptyLineBeforeAccessModifier: LogicalBlock +ExperimentalAutoDetectBinPacking: false +FixNamespaceComments: false +ForEachMacros: + - foreach + - Q_FOREACH + - BOOST_FOREACH +IfMacros: + - KJ_IF_MAYBE +IncludeBlocks: Preserve +IncludeCategories: + - Regex: '^"(llvm|llvm-c|clang|clang-c)/' + Priority: 2 + SortPriority: 0 + CaseSensitive: false + - Regex: '^(<|"(gtest|gmock|isl|json)/)' + Priority: 3 + SortPriority: 0 + CaseSensitive: false + - Regex: '.*' + Priority: 1 + SortPriority: 0 + CaseSensitive: false +IncludeIsMainRegex: '(Test)?$' +IncludeIsMainSourceRegex: '' +IndentAccessModifiers: false +IndentCaseBlocks: false +IndentCaseLabels: true +IndentExternBlock: AfterExternBlock +IndentGotoLabels: true +IndentPPDirectives: None +IndentRequiresClause: true +IndentWidth: 2 +IndentWrappedFunctionNames: false +InsertBraces: false +InsertNewlineAtEOF: false +InsertTrailingCommas: None +IntegerLiteralSeparator: + Binary: 0 + BinaryMinDigits: 0 + Decimal: 0 + DecimalMinDigits: 0 + Hex: 0 + HexMinDigits: 0 +JavaScriptQuotes: Leave +JavaScriptWrapImports: true +KeepEmptyLinesAtTheStartOfBlocks: true +LambdaBodyIndentation: Signature +LineEnding: DeriveLF +MacroBlockBegin: '' +MacroBlockEnd: '' +MaxEmptyLinesToKeep: 1 +NamespaceIndentation: None +ObjCBinPackProtocolList: Auto +ObjCBlockIndentWidth: 2 +ObjCBreakBeforeNestedBlockParam: true +ObjCSpaceAfterProperty: true +ObjCSpaceBeforeProtocolList: false +PackConstructorInitializers: BinPack +PenaltyBreakAssignment: 2 +PenaltyBreakBeforeFirstCallParameter: 19 +PenaltyBreakComment: 300 +PenaltyBreakFirstLessLess: 120 +PenaltyBreakOpenParenthesis: 0 +PenaltyBreakString: 1000 +PenaltyBreakTemplateDeclaration: 10 +PenaltyExcessCharacter: 1000000 +PenaltyIndentedWhitespace: 0 +PenaltyReturnTypeOnItsOwnLine: 200 +PointerAlignment: Left +PPIndentWidth: -1 +QualifierAlignment: Leave +ReferenceAlignment: Pointer +ReflowComments: true +RemoveBracesLLVM: false +RemoveSemicolon: false +RequiresClausePosition: OwnLine +RequiresExpressionIndentation: OuterScope +SeparateDefinitionBlocks: Leave +ShortNamespaceLines: 1 +SortIncludes: CaseSensitive +SortJavaStaticImport: Before +SortUsingDeclarations: LexicographicNumeric +SpaceAfterCStyleCast: false +SpaceAfterLogicalNot: false +SpaceAfterTemplateKeyword: false +SpaceAroundPointerQualifiers: Default +SpaceBeforeAssignmentOperators: true +SpaceBeforeCaseColon: false +SpaceBeforeCpp11BracedList: false +SpaceBeforeCtorInitializerColon: true +SpaceBeforeInheritanceColon: true +SpaceBeforeParens: ControlStatements +SpaceBeforeParensOptions: + AfterControlStatements: true + AfterForeachMacros: true + AfterFunctionDefinitionName: false + AfterFunctionDeclarationName: false + AfterIfMacros: true + AfterOverloadedOperator: false + AfterRequiresInClause: false + AfterRequiresInExpression: false + BeforeNonEmptyParentheses: false +SpaceBeforeRangeBasedForLoopColon: true +SpaceBeforeSquareBrackets: false +SpaceInEmptyBlock: false +SpaceInEmptyParentheses: false +SpacesBeforeTrailingComments: 1 +SpacesInAngles: Never +SpacesInConditionalStatement: false +SpacesInContainerLiterals: true +SpacesInCStyleCastParentheses: false +SpacesInLineCommentPrefix: + Minimum: 1 + Maximum: -1 +SpacesInParentheses: false +SpacesInSquareBrackets: false +Standard: Latest +StatementAttributeLikeMacros: + - Q_EMIT +StatementMacros: + - Q_UNUSED + - QT_REQUIRE_VERSION +TabWidth: 8 +UseTab: Never +WhitespaceSensitiveMacros: + - BOOST_PP_STRINGIZE + - CF_SWIFT_NAME + - NS_SWIFT_NAME + - PP_STRINGIZE + - STRINGIZE +... + diff --git a/Makefile b/Makefile index 13d6604..1494fbb 100644 --- a/Makefile +++ b/Makefile @@ -1,5 +1,11 @@ all: dut_aes32 dut_aesbitsliced dut_donna dut_donnabad dut_simple +SRC_DIR = src +SOURCES = $(wildcard $(SRC_DIR)/*.h) +EXAMPLE_DIR = examples +EXAMPLE_HEADERS = $(wildcard $(EXAMPLE_DIR)/*/*.h) +EXAMPLE_SOURCES = $(wildcard $(EXAMPLE_DIR)/*/*.c) + OBJS_AES32 = examples/aes32/rijndael-alg-fst.o OBJS_DONNA = examples/donna/curve25519-donna.o OBJS_DONNABAD = examples/donnabad/curve25519-donnabad.o @@ -43,5 +49,9 @@ dut_simple: examples/simple/example.c clean: rm -f $(OBJS_AES32) $(OBJS_AESBITSLICED) $(OBJS_DONNA) $(OBJS_DONNABAD) dudect_* *.exe a.out +format: $(SOURCES) $(EXAMPLE_HEADERS) $(EXAMPLE_SOURCES) + clang-format -i $^ + + test: all python test.py diff --git a/examples/aes32/dut_aes32.c b/examples/aes32/dut_aes32.c index 4e3e678..310fa5e 100644 --- a/examples/aes32/dut_aes32.c +++ b/examples/aes32/dut_aes32.c @@ -1,16 +1,18 @@ -#include -#include -#include #include "rijndael-alg-fst.h" +#include +#include +#include #define DUDECT_IMPLEMENTATION #include "dudect.h" -static uint32_t rk[44] = {0}; +static uint32_t rk[44] = { 0 }; -uint8_t do_one_computation(uint8_t *data) { - uint8_t in[16] = {0}; - uint8_t out[16] = {0}; +uint8_t +do_one_computation(uint8_t* data) +{ + uint8_t in[16] = { 0 }; + uint8_t out[16] = { 0 }; uint8_t ret = 0; memcpy(in, data, 16); @@ -18,11 +20,14 @@ uint8_t do_one_computation(uint8_t *data) { rijndaelEncrypt(rk, 10, in, out); ret ^= out[0]; - /* return some computation output to try to tame a clever optimizing compiler */ + /* return some computation output to try to tame a clever optimizing compiler + */ return ret; } -void prepare_inputs(dudect_config_t *c, uint8_t *input_data, uint8_t *classes) { +void +prepare_inputs(dudect_config_t* c, uint8_t* input_data, uint8_t* classes) +{ randombytes(input_data, c->number_measurements * c->chunk_size); for (size_t i = 0; i < c->number_measurements; i++) { classes[i] = randombit(); @@ -34,18 +39,19 @@ void prepare_inputs(dudect_config_t *c, uint8_t *input_data, uint8_t *classes) { } } -int main(int argc, char **argv) +int +main(int argc, char** argv) { (void)argc; (void)argv; dudect_config_t config = { - .chunk_size = 16, - .number_measurements = 1e6, + .chunk_size = 16, + .number_measurements = 1e6, }; dudect_ctx_t ctx; - uint8_t cipherKey[16] = {0}; + uint8_t cipherKey[16] = { 0 }; rijndaelKeySetupEnc(rk, cipherKey, 128); dudect_init(&ctx, &config); @@ -68,4 +74,3 @@ int main(int argc, char **argv) dudect_free(&ctx); return (int)state; } - diff --git a/examples/aes32/rijndael-alg-fst.c b/examples/aes32/rijndael-alg-fst.c index 0f8ebf1..ee70aa1 100644 --- a/examples/aes32/rijndael-alg-fst.c +++ b/examples/aes32/rijndael-alg-fst.c @@ -45,681 +45,484 @@ Td4[x] = Si[x].[01, 01, 01, 01]; */ static const u32 Te0[256] = { - 0xc66363a5U, 0xf87c7c84U, 0xee777799U, 0xf67b7b8dU, - 0xfff2f20dU, 0xd66b6bbdU, 0xde6f6fb1U, 0x91c5c554U, - 0x60303050U, 0x02010103U, 0xce6767a9U, 0x562b2b7dU, - 0xe7fefe19U, 0xb5d7d762U, 0x4dababe6U, 0xec76769aU, - 0x8fcaca45U, 0x1f82829dU, 0x89c9c940U, 0xfa7d7d87U, - 0xeffafa15U, 0xb25959ebU, 0x8e4747c9U, 0xfbf0f00bU, - 0x41adadecU, 0xb3d4d467U, 0x5fa2a2fdU, 0x45afafeaU, - 0x239c9cbfU, 0x53a4a4f7U, 0xe4727296U, 0x9bc0c05bU, - 0x75b7b7c2U, 0xe1fdfd1cU, 0x3d9393aeU, 0x4c26266aU, - 0x6c36365aU, 0x7e3f3f41U, 0xf5f7f702U, 0x83cccc4fU, - 0x6834345cU, 0x51a5a5f4U, 0xd1e5e534U, 0xf9f1f108U, - 0xe2717193U, 0xabd8d873U, 0x62313153U, 0x2a15153fU, - 0x0804040cU, 0x95c7c752U, 0x46232365U, 0x9dc3c35eU, - 0x30181828U, 0x379696a1U, 0x0a05050fU, 0x2f9a9ab5U, - 0x0e070709U, 0x24121236U, 0x1b80809bU, 0xdfe2e23dU, - 0xcdebeb26U, 0x4e272769U, 0x7fb2b2cdU, 0xea75759fU, - 0x1209091bU, 0x1d83839eU, 0x582c2c74U, 0x341a1a2eU, - 0x361b1b2dU, 0xdc6e6eb2U, 0xb45a5aeeU, 0x5ba0a0fbU, - 0xa45252f6U, 0x763b3b4dU, 0xb7d6d661U, 0x7db3b3ceU, - 0x5229297bU, 0xdde3e33eU, 0x5e2f2f71U, 0x13848497U, - 0xa65353f5U, 0xb9d1d168U, 0x00000000U, 0xc1eded2cU, - 0x40202060U, 0xe3fcfc1fU, 0x79b1b1c8U, 0xb65b5bedU, - 0xd46a6abeU, 0x8dcbcb46U, 0x67bebed9U, 0x7239394bU, - 0x944a4adeU, 0x984c4cd4U, 0xb05858e8U, 0x85cfcf4aU, - 0xbbd0d06bU, 0xc5efef2aU, 0x4faaaae5U, 0xedfbfb16U, - 0x864343c5U, 0x9a4d4dd7U, 0x66333355U, 0x11858594U, - 0x8a4545cfU, 0xe9f9f910U, 0x04020206U, 0xfe7f7f81U, - 0xa05050f0U, 0x783c3c44U, 0x259f9fbaU, 0x4ba8a8e3U, - 0xa25151f3U, 0x5da3a3feU, 0x804040c0U, 0x058f8f8aU, - 0x3f9292adU, 0x219d9dbcU, 0x70383848U, 0xf1f5f504U, - 0x63bcbcdfU, 0x77b6b6c1U, 0xafdada75U, 0x42212163U, - 0x20101030U, 0xe5ffff1aU, 0xfdf3f30eU, 0xbfd2d26dU, - 0x81cdcd4cU, 0x180c0c14U, 0x26131335U, 0xc3ecec2fU, - 0xbe5f5fe1U, 0x359797a2U, 0x884444ccU, 0x2e171739U, - 0x93c4c457U, 0x55a7a7f2U, 0xfc7e7e82U, 0x7a3d3d47U, - 0xc86464acU, 0xba5d5de7U, 0x3219192bU, 0xe6737395U, - 0xc06060a0U, 0x19818198U, 0x9e4f4fd1U, 0xa3dcdc7fU, - 0x44222266U, 0x542a2a7eU, 0x3b9090abU, 0x0b888883U, - 0x8c4646caU, 0xc7eeee29U, 0x6bb8b8d3U, 0x2814143cU, - 0xa7dede79U, 0xbc5e5ee2U, 0x160b0b1dU, 0xaddbdb76U, - 0xdbe0e03bU, 0x64323256U, 0x743a3a4eU, 0x140a0a1eU, - 0x924949dbU, 0x0c06060aU, 0x4824246cU, 0xb85c5ce4U, - 0x9fc2c25dU, 0xbdd3d36eU, 0x43acacefU, 0xc46262a6U, - 0x399191a8U, 0x319595a4U, 0xd3e4e437U, 0xf279798bU, - 0xd5e7e732U, 0x8bc8c843U, 0x6e373759U, 0xda6d6db7U, - 0x018d8d8cU, 0xb1d5d564U, 0x9c4e4ed2U, 0x49a9a9e0U, - 0xd86c6cb4U, 0xac5656faU, 0xf3f4f407U, 0xcfeaea25U, - 0xca6565afU, 0xf47a7a8eU, 0x47aeaee9U, 0x10080818U, - 0x6fbabad5U, 0xf0787888U, 0x4a25256fU, 0x5c2e2e72U, - 0x381c1c24U, 0x57a6a6f1U, 0x73b4b4c7U, 0x97c6c651U, - 0xcbe8e823U, 0xa1dddd7cU, 0xe874749cU, 0x3e1f1f21U, - 0x964b4bddU, 0x61bdbddcU, 0x0d8b8b86U, 0x0f8a8a85U, - 0xe0707090U, 0x7c3e3e42U, 0x71b5b5c4U, 0xcc6666aaU, - 0x904848d8U, 0x06030305U, 0xf7f6f601U, 0x1c0e0e12U, - 0xc26161a3U, 0x6a35355fU, 0xae5757f9U, 0x69b9b9d0U, - 0x17868691U, 0x99c1c158U, 0x3a1d1d27U, 0x279e9eb9U, - 0xd9e1e138U, 0xebf8f813U, 0x2b9898b3U, 0x22111133U, - 0xd26969bbU, 0xa9d9d970U, 0x078e8e89U, 0x339494a7U, - 0x2d9b9bb6U, 0x3c1e1e22U, 0x15878792U, 0xc9e9e920U, - 0x87cece49U, 0xaa5555ffU, 0x50282878U, 0xa5dfdf7aU, - 0x038c8c8fU, 0x59a1a1f8U, 0x09898980U, 0x1a0d0d17U, - 0x65bfbfdaU, 0xd7e6e631U, 0x844242c6U, 0xd06868b8U, - 0x824141c3U, 0x299999b0U, 0x5a2d2d77U, 0x1e0f0f11U, - 0x7bb0b0cbU, 0xa85454fcU, 0x6dbbbbd6U, 0x2c16163aU, + 0xc66363a5U, 0xf87c7c84U, 0xee777799U, 0xf67b7b8dU, 0xfff2f20dU, 0xd66b6bbdU, + 0xde6f6fb1U, 0x91c5c554U, 0x60303050U, 0x02010103U, 0xce6767a9U, 0x562b2b7dU, + 0xe7fefe19U, 0xb5d7d762U, 0x4dababe6U, 0xec76769aU, 0x8fcaca45U, 0x1f82829dU, + 0x89c9c940U, 0xfa7d7d87U, 0xeffafa15U, 0xb25959ebU, 0x8e4747c9U, 0xfbf0f00bU, + 0x41adadecU, 0xb3d4d467U, 0x5fa2a2fdU, 0x45afafeaU, 0x239c9cbfU, 0x53a4a4f7U, + 0xe4727296U, 0x9bc0c05bU, 0x75b7b7c2U, 0xe1fdfd1cU, 0x3d9393aeU, 0x4c26266aU, + 0x6c36365aU, 0x7e3f3f41U, 0xf5f7f702U, 0x83cccc4fU, 0x6834345cU, 0x51a5a5f4U, + 0xd1e5e534U, 0xf9f1f108U, 0xe2717193U, 0xabd8d873U, 0x62313153U, 0x2a15153fU, + 0x0804040cU, 0x95c7c752U, 0x46232365U, 0x9dc3c35eU, 0x30181828U, 0x379696a1U, + 0x0a05050fU, 0x2f9a9ab5U, 0x0e070709U, 0x24121236U, 0x1b80809bU, 0xdfe2e23dU, + 0xcdebeb26U, 0x4e272769U, 0x7fb2b2cdU, 0xea75759fU, 0x1209091bU, 0x1d83839eU, + 0x582c2c74U, 0x341a1a2eU, 0x361b1b2dU, 0xdc6e6eb2U, 0xb45a5aeeU, 0x5ba0a0fbU, + 0xa45252f6U, 0x763b3b4dU, 0xb7d6d661U, 0x7db3b3ceU, 0x5229297bU, 0xdde3e33eU, + 0x5e2f2f71U, 0x13848497U, 0xa65353f5U, 0xb9d1d168U, 0x00000000U, 0xc1eded2cU, + 0x40202060U, 0xe3fcfc1fU, 0x79b1b1c8U, 0xb65b5bedU, 0xd46a6abeU, 0x8dcbcb46U, + 0x67bebed9U, 0x7239394bU, 0x944a4adeU, 0x984c4cd4U, 0xb05858e8U, 0x85cfcf4aU, + 0xbbd0d06bU, 0xc5efef2aU, 0x4faaaae5U, 0xedfbfb16U, 0x864343c5U, 0x9a4d4dd7U, + 0x66333355U, 0x11858594U, 0x8a4545cfU, 0xe9f9f910U, 0x04020206U, 0xfe7f7f81U, + 0xa05050f0U, 0x783c3c44U, 0x259f9fbaU, 0x4ba8a8e3U, 0xa25151f3U, 0x5da3a3feU, + 0x804040c0U, 0x058f8f8aU, 0x3f9292adU, 0x219d9dbcU, 0x70383848U, 0xf1f5f504U, + 0x63bcbcdfU, 0x77b6b6c1U, 0xafdada75U, 0x42212163U, 0x20101030U, 0xe5ffff1aU, + 0xfdf3f30eU, 0xbfd2d26dU, 0x81cdcd4cU, 0x180c0c14U, 0x26131335U, 0xc3ecec2fU, + 0xbe5f5fe1U, 0x359797a2U, 0x884444ccU, 0x2e171739U, 0x93c4c457U, 0x55a7a7f2U, + 0xfc7e7e82U, 0x7a3d3d47U, 0xc86464acU, 0xba5d5de7U, 0x3219192bU, 0xe6737395U, + 0xc06060a0U, 0x19818198U, 0x9e4f4fd1U, 0xa3dcdc7fU, 0x44222266U, 0x542a2a7eU, + 0x3b9090abU, 0x0b888883U, 0x8c4646caU, 0xc7eeee29U, 0x6bb8b8d3U, 0x2814143cU, + 0xa7dede79U, 0xbc5e5ee2U, 0x160b0b1dU, 0xaddbdb76U, 0xdbe0e03bU, 0x64323256U, + 0x743a3a4eU, 0x140a0a1eU, 0x924949dbU, 0x0c06060aU, 0x4824246cU, 0xb85c5ce4U, + 0x9fc2c25dU, 0xbdd3d36eU, 0x43acacefU, 0xc46262a6U, 0x399191a8U, 0x319595a4U, + 0xd3e4e437U, 0xf279798bU, 0xd5e7e732U, 0x8bc8c843U, 0x6e373759U, 0xda6d6db7U, + 0x018d8d8cU, 0xb1d5d564U, 0x9c4e4ed2U, 0x49a9a9e0U, 0xd86c6cb4U, 0xac5656faU, + 0xf3f4f407U, 0xcfeaea25U, 0xca6565afU, 0xf47a7a8eU, 0x47aeaee9U, 0x10080818U, + 0x6fbabad5U, 0xf0787888U, 0x4a25256fU, 0x5c2e2e72U, 0x381c1c24U, 0x57a6a6f1U, + 0x73b4b4c7U, 0x97c6c651U, 0xcbe8e823U, 0xa1dddd7cU, 0xe874749cU, 0x3e1f1f21U, + 0x964b4bddU, 0x61bdbddcU, 0x0d8b8b86U, 0x0f8a8a85U, 0xe0707090U, 0x7c3e3e42U, + 0x71b5b5c4U, 0xcc6666aaU, 0x904848d8U, 0x06030305U, 0xf7f6f601U, 0x1c0e0e12U, + 0xc26161a3U, 0x6a35355fU, 0xae5757f9U, 0x69b9b9d0U, 0x17868691U, 0x99c1c158U, + 0x3a1d1d27U, 0x279e9eb9U, 0xd9e1e138U, 0xebf8f813U, 0x2b9898b3U, 0x22111133U, + 0xd26969bbU, 0xa9d9d970U, 0x078e8e89U, 0x339494a7U, 0x2d9b9bb6U, 0x3c1e1e22U, + 0x15878792U, 0xc9e9e920U, 0x87cece49U, 0xaa5555ffU, 0x50282878U, 0xa5dfdf7aU, + 0x038c8c8fU, 0x59a1a1f8U, 0x09898980U, 0x1a0d0d17U, 0x65bfbfdaU, 0xd7e6e631U, + 0x844242c6U, 0xd06868b8U, 0x824141c3U, 0x299999b0U, 0x5a2d2d77U, 0x1e0f0f11U, + 0x7bb0b0cbU, 0xa85454fcU, 0x6dbbbbd6U, 0x2c16163aU, }; static const u32 Te1[256] = { - 0xa5c66363U, 0x84f87c7cU, 0x99ee7777U, 0x8df67b7bU, - 0x0dfff2f2U, 0xbdd66b6bU, 0xb1de6f6fU, 0x5491c5c5U, - 0x50603030U, 0x03020101U, 0xa9ce6767U, 0x7d562b2bU, - 0x19e7fefeU, 0x62b5d7d7U, 0xe64dababU, 0x9aec7676U, - 0x458fcacaU, 0x9d1f8282U, 0x4089c9c9U, 0x87fa7d7dU, - 0x15effafaU, 0xebb25959U, 0xc98e4747U, 0x0bfbf0f0U, - 0xec41adadU, 0x67b3d4d4U, 0xfd5fa2a2U, 0xea45afafU, - 0xbf239c9cU, 0xf753a4a4U, 0x96e47272U, 0x5b9bc0c0U, - 0xc275b7b7U, 0x1ce1fdfdU, 0xae3d9393U, 0x6a4c2626U, - 0x5a6c3636U, 0x417e3f3fU, 0x02f5f7f7U, 0x4f83ccccU, - 0x5c683434U, 0xf451a5a5U, 0x34d1e5e5U, 0x08f9f1f1U, - 0x93e27171U, 0x73abd8d8U, 0x53623131U, 0x3f2a1515U, - 0x0c080404U, 0x5295c7c7U, 0x65462323U, 0x5e9dc3c3U, - 0x28301818U, 0xa1379696U, 0x0f0a0505U, 0xb52f9a9aU, - 0x090e0707U, 0x36241212U, 0x9b1b8080U, 0x3ddfe2e2U, - 0x26cdebebU, 0x694e2727U, 0xcd7fb2b2U, 0x9fea7575U, - 0x1b120909U, 0x9e1d8383U, 0x74582c2cU, 0x2e341a1aU, - 0x2d361b1bU, 0xb2dc6e6eU, 0xeeb45a5aU, 0xfb5ba0a0U, - 0xf6a45252U, 0x4d763b3bU, 0x61b7d6d6U, 0xce7db3b3U, - 0x7b522929U, 0x3edde3e3U, 0x715e2f2fU, 0x97138484U, - 0xf5a65353U, 0x68b9d1d1U, 0x00000000U, 0x2cc1ededU, - 0x60402020U, 0x1fe3fcfcU, 0xc879b1b1U, 0xedb65b5bU, - 0xbed46a6aU, 0x468dcbcbU, 0xd967bebeU, 0x4b723939U, - 0xde944a4aU, 0xd4984c4cU, 0xe8b05858U, 0x4a85cfcfU, - 0x6bbbd0d0U, 0x2ac5efefU, 0xe54faaaaU, 0x16edfbfbU, - 0xc5864343U, 0xd79a4d4dU, 0x55663333U, 0x94118585U, - 0xcf8a4545U, 0x10e9f9f9U, 0x06040202U, 0x81fe7f7fU, - 0xf0a05050U, 0x44783c3cU, 0xba259f9fU, 0xe34ba8a8U, - 0xf3a25151U, 0xfe5da3a3U, 0xc0804040U, 0x8a058f8fU, - 0xad3f9292U, 0xbc219d9dU, 0x48703838U, 0x04f1f5f5U, - 0xdf63bcbcU, 0xc177b6b6U, 0x75afdadaU, 0x63422121U, - 0x30201010U, 0x1ae5ffffU, 0x0efdf3f3U, 0x6dbfd2d2U, - 0x4c81cdcdU, 0x14180c0cU, 0x35261313U, 0x2fc3ececU, - 0xe1be5f5fU, 0xa2359797U, 0xcc884444U, 0x392e1717U, - 0x5793c4c4U, 0xf255a7a7U, 0x82fc7e7eU, 0x477a3d3dU, - 0xacc86464U, 0xe7ba5d5dU, 0x2b321919U, 0x95e67373U, - 0xa0c06060U, 0x98198181U, 0xd19e4f4fU, 0x7fa3dcdcU, - 0x66442222U, 0x7e542a2aU, 0xab3b9090U, 0x830b8888U, - 0xca8c4646U, 0x29c7eeeeU, 0xd36bb8b8U, 0x3c281414U, - 0x79a7dedeU, 0xe2bc5e5eU, 0x1d160b0bU, 0x76addbdbU, - 0x3bdbe0e0U, 0x56643232U, 0x4e743a3aU, 0x1e140a0aU, - 0xdb924949U, 0x0a0c0606U, 0x6c482424U, 0xe4b85c5cU, - 0x5d9fc2c2U, 0x6ebdd3d3U, 0xef43acacU, 0xa6c46262U, - 0xa8399191U, 0xa4319595U, 0x37d3e4e4U, 0x8bf27979U, - 0x32d5e7e7U, 0x438bc8c8U, 0x596e3737U, 0xb7da6d6dU, - 0x8c018d8dU, 0x64b1d5d5U, 0xd29c4e4eU, 0xe049a9a9U, - 0xb4d86c6cU, 0xfaac5656U, 0x07f3f4f4U, 0x25cfeaeaU, - 0xafca6565U, 0x8ef47a7aU, 0xe947aeaeU, 0x18100808U, - 0xd56fbabaU, 0x88f07878U, 0x6f4a2525U, 0x725c2e2eU, - 0x24381c1cU, 0xf157a6a6U, 0xc773b4b4U, 0x5197c6c6U, - 0x23cbe8e8U, 0x7ca1ddddU, 0x9ce87474U, 0x213e1f1fU, - 0xdd964b4bU, 0xdc61bdbdU, 0x860d8b8bU, 0x850f8a8aU, - 0x90e07070U, 0x427c3e3eU, 0xc471b5b5U, 0xaacc6666U, - 0xd8904848U, 0x05060303U, 0x01f7f6f6U, 0x121c0e0eU, - 0xa3c26161U, 0x5f6a3535U, 0xf9ae5757U, 0xd069b9b9U, - 0x91178686U, 0x5899c1c1U, 0x273a1d1dU, 0xb9279e9eU, - 0x38d9e1e1U, 0x13ebf8f8U, 0xb32b9898U, 0x33221111U, - 0xbbd26969U, 0x70a9d9d9U, 0x89078e8eU, 0xa7339494U, - 0xb62d9b9bU, 0x223c1e1eU, 0x92158787U, 0x20c9e9e9U, - 0x4987ceceU, 0xffaa5555U, 0x78502828U, 0x7aa5dfdfU, - 0x8f038c8cU, 0xf859a1a1U, 0x80098989U, 0x171a0d0dU, - 0xda65bfbfU, 0x31d7e6e6U, 0xc6844242U, 0xb8d06868U, - 0xc3824141U, 0xb0299999U, 0x775a2d2dU, 0x111e0f0fU, - 0xcb7bb0b0U, 0xfca85454U, 0xd66dbbbbU, 0x3a2c1616U, + 0xa5c66363U, 0x84f87c7cU, 0x99ee7777U, 0x8df67b7bU, 0x0dfff2f2U, 0xbdd66b6bU, + 0xb1de6f6fU, 0x5491c5c5U, 0x50603030U, 0x03020101U, 0xa9ce6767U, 0x7d562b2bU, + 0x19e7fefeU, 0x62b5d7d7U, 0xe64dababU, 0x9aec7676U, 0x458fcacaU, 0x9d1f8282U, + 0x4089c9c9U, 0x87fa7d7dU, 0x15effafaU, 0xebb25959U, 0xc98e4747U, 0x0bfbf0f0U, + 0xec41adadU, 0x67b3d4d4U, 0xfd5fa2a2U, 0xea45afafU, 0xbf239c9cU, 0xf753a4a4U, + 0x96e47272U, 0x5b9bc0c0U, 0xc275b7b7U, 0x1ce1fdfdU, 0xae3d9393U, 0x6a4c2626U, + 0x5a6c3636U, 0x417e3f3fU, 0x02f5f7f7U, 0x4f83ccccU, 0x5c683434U, 0xf451a5a5U, + 0x34d1e5e5U, 0x08f9f1f1U, 0x93e27171U, 0x73abd8d8U, 0x53623131U, 0x3f2a1515U, + 0x0c080404U, 0x5295c7c7U, 0x65462323U, 0x5e9dc3c3U, 0x28301818U, 0xa1379696U, + 0x0f0a0505U, 0xb52f9a9aU, 0x090e0707U, 0x36241212U, 0x9b1b8080U, 0x3ddfe2e2U, + 0x26cdebebU, 0x694e2727U, 0xcd7fb2b2U, 0x9fea7575U, 0x1b120909U, 0x9e1d8383U, + 0x74582c2cU, 0x2e341a1aU, 0x2d361b1bU, 0xb2dc6e6eU, 0xeeb45a5aU, 0xfb5ba0a0U, + 0xf6a45252U, 0x4d763b3bU, 0x61b7d6d6U, 0xce7db3b3U, 0x7b522929U, 0x3edde3e3U, + 0x715e2f2fU, 0x97138484U, 0xf5a65353U, 0x68b9d1d1U, 0x00000000U, 0x2cc1ededU, + 0x60402020U, 0x1fe3fcfcU, 0xc879b1b1U, 0xedb65b5bU, 0xbed46a6aU, 0x468dcbcbU, + 0xd967bebeU, 0x4b723939U, 0xde944a4aU, 0xd4984c4cU, 0xe8b05858U, 0x4a85cfcfU, + 0x6bbbd0d0U, 0x2ac5efefU, 0xe54faaaaU, 0x16edfbfbU, 0xc5864343U, 0xd79a4d4dU, + 0x55663333U, 0x94118585U, 0xcf8a4545U, 0x10e9f9f9U, 0x06040202U, 0x81fe7f7fU, + 0xf0a05050U, 0x44783c3cU, 0xba259f9fU, 0xe34ba8a8U, 0xf3a25151U, 0xfe5da3a3U, + 0xc0804040U, 0x8a058f8fU, 0xad3f9292U, 0xbc219d9dU, 0x48703838U, 0x04f1f5f5U, + 0xdf63bcbcU, 0xc177b6b6U, 0x75afdadaU, 0x63422121U, 0x30201010U, 0x1ae5ffffU, + 0x0efdf3f3U, 0x6dbfd2d2U, 0x4c81cdcdU, 0x14180c0cU, 0x35261313U, 0x2fc3ececU, + 0xe1be5f5fU, 0xa2359797U, 0xcc884444U, 0x392e1717U, 0x5793c4c4U, 0xf255a7a7U, + 0x82fc7e7eU, 0x477a3d3dU, 0xacc86464U, 0xe7ba5d5dU, 0x2b321919U, 0x95e67373U, + 0xa0c06060U, 0x98198181U, 0xd19e4f4fU, 0x7fa3dcdcU, 0x66442222U, 0x7e542a2aU, + 0xab3b9090U, 0x830b8888U, 0xca8c4646U, 0x29c7eeeeU, 0xd36bb8b8U, 0x3c281414U, + 0x79a7dedeU, 0xe2bc5e5eU, 0x1d160b0bU, 0x76addbdbU, 0x3bdbe0e0U, 0x56643232U, + 0x4e743a3aU, 0x1e140a0aU, 0xdb924949U, 0x0a0c0606U, 0x6c482424U, 0xe4b85c5cU, + 0x5d9fc2c2U, 0x6ebdd3d3U, 0xef43acacU, 0xa6c46262U, 0xa8399191U, 0xa4319595U, + 0x37d3e4e4U, 0x8bf27979U, 0x32d5e7e7U, 0x438bc8c8U, 0x596e3737U, 0xb7da6d6dU, + 0x8c018d8dU, 0x64b1d5d5U, 0xd29c4e4eU, 0xe049a9a9U, 0xb4d86c6cU, 0xfaac5656U, + 0x07f3f4f4U, 0x25cfeaeaU, 0xafca6565U, 0x8ef47a7aU, 0xe947aeaeU, 0x18100808U, + 0xd56fbabaU, 0x88f07878U, 0x6f4a2525U, 0x725c2e2eU, 0x24381c1cU, 0xf157a6a6U, + 0xc773b4b4U, 0x5197c6c6U, 0x23cbe8e8U, 0x7ca1ddddU, 0x9ce87474U, 0x213e1f1fU, + 0xdd964b4bU, 0xdc61bdbdU, 0x860d8b8bU, 0x850f8a8aU, 0x90e07070U, 0x427c3e3eU, + 0xc471b5b5U, 0xaacc6666U, 0xd8904848U, 0x05060303U, 0x01f7f6f6U, 0x121c0e0eU, + 0xa3c26161U, 0x5f6a3535U, 0xf9ae5757U, 0xd069b9b9U, 0x91178686U, 0x5899c1c1U, + 0x273a1d1dU, 0xb9279e9eU, 0x38d9e1e1U, 0x13ebf8f8U, 0xb32b9898U, 0x33221111U, + 0xbbd26969U, 0x70a9d9d9U, 0x89078e8eU, 0xa7339494U, 0xb62d9b9bU, 0x223c1e1eU, + 0x92158787U, 0x20c9e9e9U, 0x4987ceceU, 0xffaa5555U, 0x78502828U, 0x7aa5dfdfU, + 0x8f038c8cU, 0xf859a1a1U, 0x80098989U, 0x171a0d0dU, 0xda65bfbfU, 0x31d7e6e6U, + 0xc6844242U, 0xb8d06868U, 0xc3824141U, 0xb0299999U, 0x775a2d2dU, 0x111e0f0fU, + 0xcb7bb0b0U, 0xfca85454U, 0xd66dbbbbU, 0x3a2c1616U, }; static const u32 Te2[256] = { - 0x63a5c663U, 0x7c84f87cU, 0x7799ee77U, 0x7b8df67bU, - 0xf20dfff2U, 0x6bbdd66bU, 0x6fb1de6fU, 0xc55491c5U, - 0x30506030U, 0x01030201U, 0x67a9ce67U, 0x2b7d562bU, - 0xfe19e7feU, 0xd762b5d7U, 0xabe64dabU, 0x769aec76U, - 0xca458fcaU, 0x829d1f82U, 0xc94089c9U, 0x7d87fa7dU, - 0xfa15effaU, 0x59ebb259U, 0x47c98e47U, 0xf00bfbf0U, - 0xadec41adU, 0xd467b3d4U, 0xa2fd5fa2U, 0xafea45afU, - 0x9cbf239cU, 0xa4f753a4U, 0x7296e472U, 0xc05b9bc0U, - 0xb7c275b7U, 0xfd1ce1fdU, 0x93ae3d93U, 0x266a4c26U, - 0x365a6c36U, 0x3f417e3fU, 0xf702f5f7U, 0xcc4f83ccU, - 0x345c6834U, 0xa5f451a5U, 0xe534d1e5U, 0xf108f9f1U, - 0x7193e271U, 0xd873abd8U, 0x31536231U, 0x153f2a15U, - 0x040c0804U, 0xc75295c7U, 0x23654623U, 0xc35e9dc3U, - 0x18283018U, 0x96a13796U, 0x050f0a05U, 0x9ab52f9aU, - 0x07090e07U, 0x12362412U, 0x809b1b80U, 0xe23ddfe2U, - 0xeb26cdebU, 0x27694e27U, 0xb2cd7fb2U, 0x759fea75U, - 0x091b1209U, 0x839e1d83U, 0x2c74582cU, 0x1a2e341aU, - 0x1b2d361bU, 0x6eb2dc6eU, 0x5aeeb45aU, 0xa0fb5ba0U, - 0x52f6a452U, 0x3b4d763bU, 0xd661b7d6U, 0xb3ce7db3U, - 0x297b5229U, 0xe33edde3U, 0x2f715e2fU, 0x84971384U, - 0x53f5a653U, 0xd168b9d1U, 0x00000000U, 0xed2cc1edU, - 0x20604020U, 0xfc1fe3fcU, 0xb1c879b1U, 0x5bedb65bU, - 0x6abed46aU, 0xcb468dcbU, 0xbed967beU, 0x394b7239U, - 0x4ade944aU, 0x4cd4984cU, 0x58e8b058U, 0xcf4a85cfU, - 0xd06bbbd0U, 0xef2ac5efU, 0xaae54faaU, 0xfb16edfbU, - 0x43c58643U, 0x4dd79a4dU, 0x33556633U, 0x85941185U, - 0x45cf8a45U, 0xf910e9f9U, 0x02060402U, 0x7f81fe7fU, - 0x50f0a050U, 0x3c44783cU, 0x9fba259fU, 0xa8e34ba8U, - 0x51f3a251U, 0xa3fe5da3U, 0x40c08040U, 0x8f8a058fU, - 0x92ad3f92U, 0x9dbc219dU, 0x38487038U, 0xf504f1f5U, - 0xbcdf63bcU, 0xb6c177b6U, 0xda75afdaU, 0x21634221U, - 0x10302010U, 0xff1ae5ffU, 0xf30efdf3U, 0xd26dbfd2U, - 0xcd4c81cdU, 0x0c14180cU, 0x13352613U, 0xec2fc3ecU, - 0x5fe1be5fU, 0x97a23597U, 0x44cc8844U, 0x17392e17U, - 0xc45793c4U, 0xa7f255a7U, 0x7e82fc7eU, 0x3d477a3dU, - 0x64acc864U, 0x5de7ba5dU, 0x192b3219U, 0x7395e673U, - 0x60a0c060U, 0x81981981U, 0x4fd19e4fU, 0xdc7fa3dcU, - 0x22664422U, 0x2a7e542aU, 0x90ab3b90U, 0x88830b88U, - 0x46ca8c46U, 0xee29c7eeU, 0xb8d36bb8U, 0x143c2814U, - 0xde79a7deU, 0x5ee2bc5eU, 0x0b1d160bU, 0xdb76addbU, - 0xe03bdbe0U, 0x32566432U, 0x3a4e743aU, 0x0a1e140aU, - 0x49db9249U, 0x060a0c06U, 0x246c4824U, 0x5ce4b85cU, - 0xc25d9fc2U, 0xd36ebdd3U, 0xacef43acU, 0x62a6c462U, - 0x91a83991U, 0x95a43195U, 0xe437d3e4U, 0x798bf279U, - 0xe732d5e7U, 0xc8438bc8U, 0x37596e37U, 0x6db7da6dU, - 0x8d8c018dU, 0xd564b1d5U, 0x4ed29c4eU, 0xa9e049a9U, - 0x6cb4d86cU, 0x56faac56U, 0xf407f3f4U, 0xea25cfeaU, - 0x65afca65U, 0x7a8ef47aU, 0xaee947aeU, 0x08181008U, - 0xbad56fbaU, 0x7888f078U, 0x256f4a25U, 0x2e725c2eU, - 0x1c24381cU, 0xa6f157a6U, 0xb4c773b4U, 0xc65197c6U, - 0xe823cbe8U, 0xdd7ca1ddU, 0x749ce874U, 0x1f213e1fU, - 0x4bdd964bU, 0xbddc61bdU, 0x8b860d8bU, 0x8a850f8aU, - 0x7090e070U, 0x3e427c3eU, 0xb5c471b5U, 0x66aacc66U, - 0x48d89048U, 0x03050603U, 0xf601f7f6U, 0x0e121c0eU, - 0x61a3c261U, 0x355f6a35U, 0x57f9ae57U, 0xb9d069b9U, - 0x86911786U, 0xc15899c1U, 0x1d273a1dU, 0x9eb9279eU, - 0xe138d9e1U, 0xf813ebf8U, 0x98b32b98U, 0x11332211U, - 0x69bbd269U, 0xd970a9d9U, 0x8e89078eU, 0x94a73394U, - 0x9bb62d9bU, 0x1e223c1eU, 0x87921587U, 0xe920c9e9U, - 0xce4987ceU, 0x55ffaa55U, 0x28785028U, 0xdf7aa5dfU, - 0x8c8f038cU, 0xa1f859a1U, 0x89800989U, 0x0d171a0dU, - 0xbfda65bfU, 0xe631d7e6U, 0x42c68442U, 0x68b8d068U, - 0x41c38241U, 0x99b02999U, 0x2d775a2dU, 0x0f111e0fU, - 0xb0cb7bb0U, 0x54fca854U, 0xbbd66dbbU, 0x163a2c16U, + 0x63a5c663U, 0x7c84f87cU, 0x7799ee77U, 0x7b8df67bU, 0xf20dfff2U, 0x6bbdd66bU, + 0x6fb1de6fU, 0xc55491c5U, 0x30506030U, 0x01030201U, 0x67a9ce67U, 0x2b7d562bU, + 0xfe19e7feU, 0xd762b5d7U, 0xabe64dabU, 0x769aec76U, 0xca458fcaU, 0x829d1f82U, + 0xc94089c9U, 0x7d87fa7dU, 0xfa15effaU, 0x59ebb259U, 0x47c98e47U, 0xf00bfbf0U, + 0xadec41adU, 0xd467b3d4U, 0xa2fd5fa2U, 0xafea45afU, 0x9cbf239cU, 0xa4f753a4U, + 0x7296e472U, 0xc05b9bc0U, 0xb7c275b7U, 0xfd1ce1fdU, 0x93ae3d93U, 0x266a4c26U, + 0x365a6c36U, 0x3f417e3fU, 0xf702f5f7U, 0xcc4f83ccU, 0x345c6834U, 0xa5f451a5U, + 0xe534d1e5U, 0xf108f9f1U, 0x7193e271U, 0xd873abd8U, 0x31536231U, 0x153f2a15U, + 0x040c0804U, 0xc75295c7U, 0x23654623U, 0xc35e9dc3U, 0x18283018U, 0x96a13796U, + 0x050f0a05U, 0x9ab52f9aU, 0x07090e07U, 0x12362412U, 0x809b1b80U, 0xe23ddfe2U, + 0xeb26cdebU, 0x27694e27U, 0xb2cd7fb2U, 0x759fea75U, 0x091b1209U, 0x839e1d83U, + 0x2c74582cU, 0x1a2e341aU, 0x1b2d361bU, 0x6eb2dc6eU, 0x5aeeb45aU, 0xa0fb5ba0U, + 0x52f6a452U, 0x3b4d763bU, 0xd661b7d6U, 0xb3ce7db3U, 0x297b5229U, 0xe33edde3U, + 0x2f715e2fU, 0x84971384U, 0x53f5a653U, 0xd168b9d1U, 0x00000000U, 0xed2cc1edU, + 0x20604020U, 0xfc1fe3fcU, 0xb1c879b1U, 0x5bedb65bU, 0x6abed46aU, 0xcb468dcbU, + 0xbed967beU, 0x394b7239U, 0x4ade944aU, 0x4cd4984cU, 0x58e8b058U, 0xcf4a85cfU, + 0xd06bbbd0U, 0xef2ac5efU, 0xaae54faaU, 0xfb16edfbU, 0x43c58643U, 0x4dd79a4dU, + 0x33556633U, 0x85941185U, 0x45cf8a45U, 0xf910e9f9U, 0x02060402U, 0x7f81fe7fU, + 0x50f0a050U, 0x3c44783cU, 0x9fba259fU, 0xa8e34ba8U, 0x51f3a251U, 0xa3fe5da3U, + 0x40c08040U, 0x8f8a058fU, 0x92ad3f92U, 0x9dbc219dU, 0x38487038U, 0xf504f1f5U, + 0xbcdf63bcU, 0xb6c177b6U, 0xda75afdaU, 0x21634221U, 0x10302010U, 0xff1ae5ffU, + 0xf30efdf3U, 0xd26dbfd2U, 0xcd4c81cdU, 0x0c14180cU, 0x13352613U, 0xec2fc3ecU, + 0x5fe1be5fU, 0x97a23597U, 0x44cc8844U, 0x17392e17U, 0xc45793c4U, 0xa7f255a7U, + 0x7e82fc7eU, 0x3d477a3dU, 0x64acc864U, 0x5de7ba5dU, 0x192b3219U, 0x7395e673U, + 0x60a0c060U, 0x81981981U, 0x4fd19e4fU, 0xdc7fa3dcU, 0x22664422U, 0x2a7e542aU, + 0x90ab3b90U, 0x88830b88U, 0x46ca8c46U, 0xee29c7eeU, 0xb8d36bb8U, 0x143c2814U, + 0xde79a7deU, 0x5ee2bc5eU, 0x0b1d160bU, 0xdb76addbU, 0xe03bdbe0U, 0x32566432U, + 0x3a4e743aU, 0x0a1e140aU, 0x49db9249U, 0x060a0c06U, 0x246c4824U, 0x5ce4b85cU, + 0xc25d9fc2U, 0xd36ebdd3U, 0xacef43acU, 0x62a6c462U, 0x91a83991U, 0x95a43195U, + 0xe437d3e4U, 0x798bf279U, 0xe732d5e7U, 0xc8438bc8U, 0x37596e37U, 0x6db7da6dU, + 0x8d8c018dU, 0xd564b1d5U, 0x4ed29c4eU, 0xa9e049a9U, 0x6cb4d86cU, 0x56faac56U, + 0xf407f3f4U, 0xea25cfeaU, 0x65afca65U, 0x7a8ef47aU, 0xaee947aeU, 0x08181008U, + 0xbad56fbaU, 0x7888f078U, 0x256f4a25U, 0x2e725c2eU, 0x1c24381cU, 0xa6f157a6U, + 0xb4c773b4U, 0xc65197c6U, 0xe823cbe8U, 0xdd7ca1ddU, 0x749ce874U, 0x1f213e1fU, + 0x4bdd964bU, 0xbddc61bdU, 0x8b860d8bU, 0x8a850f8aU, 0x7090e070U, 0x3e427c3eU, + 0xb5c471b5U, 0x66aacc66U, 0x48d89048U, 0x03050603U, 0xf601f7f6U, 0x0e121c0eU, + 0x61a3c261U, 0x355f6a35U, 0x57f9ae57U, 0xb9d069b9U, 0x86911786U, 0xc15899c1U, + 0x1d273a1dU, 0x9eb9279eU, 0xe138d9e1U, 0xf813ebf8U, 0x98b32b98U, 0x11332211U, + 0x69bbd269U, 0xd970a9d9U, 0x8e89078eU, 0x94a73394U, 0x9bb62d9bU, 0x1e223c1eU, + 0x87921587U, 0xe920c9e9U, 0xce4987ceU, 0x55ffaa55U, 0x28785028U, 0xdf7aa5dfU, + 0x8c8f038cU, 0xa1f859a1U, 0x89800989U, 0x0d171a0dU, 0xbfda65bfU, 0xe631d7e6U, + 0x42c68442U, 0x68b8d068U, 0x41c38241U, 0x99b02999U, 0x2d775a2dU, 0x0f111e0fU, + 0xb0cb7bb0U, 0x54fca854U, 0xbbd66dbbU, 0x163a2c16U, }; static const u32 Te3[256] = { - 0x6363a5c6U, 0x7c7c84f8U, 0x777799eeU, 0x7b7b8df6U, - 0xf2f20dffU, 0x6b6bbdd6U, 0x6f6fb1deU, 0xc5c55491U, - 0x30305060U, 0x01010302U, 0x6767a9ceU, 0x2b2b7d56U, - 0xfefe19e7U, 0xd7d762b5U, 0xababe64dU, 0x76769aecU, - 0xcaca458fU, 0x82829d1fU, 0xc9c94089U, 0x7d7d87faU, - 0xfafa15efU, 0x5959ebb2U, 0x4747c98eU, 0xf0f00bfbU, - 0xadadec41U, 0xd4d467b3U, 0xa2a2fd5fU, 0xafafea45U, - 0x9c9cbf23U, 0xa4a4f753U, 0x727296e4U, 0xc0c05b9bU, - 0xb7b7c275U, 0xfdfd1ce1U, 0x9393ae3dU, 0x26266a4cU, - 0x36365a6cU, 0x3f3f417eU, 0xf7f702f5U, 0xcccc4f83U, - 0x34345c68U, 0xa5a5f451U, 0xe5e534d1U, 0xf1f108f9U, - 0x717193e2U, 0xd8d873abU, 0x31315362U, 0x15153f2aU, - 0x04040c08U, 0xc7c75295U, 0x23236546U, 0xc3c35e9dU, - 0x18182830U, 0x9696a137U, 0x05050f0aU, 0x9a9ab52fU, - 0x0707090eU, 0x12123624U, 0x80809b1bU, 0xe2e23ddfU, - 0xebeb26cdU, 0x2727694eU, 0xb2b2cd7fU, 0x75759feaU, - 0x09091b12U, 0x83839e1dU, 0x2c2c7458U, 0x1a1a2e34U, - 0x1b1b2d36U, 0x6e6eb2dcU, 0x5a5aeeb4U, 0xa0a0fb5bU, - 0x5252f6a4U, 0x3b3b4d76U, 0xd6d661b7U, 0xb3b3ce7dU, - 0x29297b52U, 0xe3e33eddU, 0x2f2f715eU, 0x84849713U, - 0x5353f5a6U, 0xd1d168b9U, 0x00000000U, 0xeded2cc1U, - 0x20206040U, 0xfcfc1fe3U, 0xb1b1c879U, 0x5b5bedb6U, - 0x6a6abed4U, 0xcbcb468dU, 0xbebed967U, 0x39394b72U, - 0x4a4ade94U, 0x4c4cd498U, 0x5858e8b0U, 0xcfcf4a85U, - 0xd0d06bbbU, 0xefef2ac5U, 0xaaaae54fU, 0xfbfb16edU, - 0x4343c586U, 0x4d4dd79aU, 0x33335566U, 0x85859411U, - 0x4545cf8aU, 0xf9f910e9U, 0x02020604U, 0x7f7f81feU, - 0x5050f0a0U, 0x3c3c4478U, 0x9f9fba25U, 0xa8a8e34bU, - 0x5151f3a2U, 0xa3a3fe5dU, 0x4040c080U, 0x8f8f8a05U, - 0x9292ad3fU, 0x9d9dbc21U, 0x38384870U, 0xf5f504f1U, - 0xbcbcdf63U, 0xb6b6c177U, 0xdada75afU, 0x21216342U, - 0x10103020U, 0xffff1ae5U, 0xf3f30efdU, 0xd2d26dbfU, - 0xcdcd4c81U, 0x0c0c1418U, 0x13133526U, 0xecec2fc3U, - 0x5f5fe1beU, 0x9797a235U, 0x4444cc88U, 0x1717392eU, - 0xc4c45793U, 0xa7a7f255U, 0x7e7e82fcU, 0x3d3d477aU, - 0x6464acc8U, 0x5d5de7baU, 0x19192b32U, 0x737395e6U, - 0x6060a0c0U, 0x81819819U, 0x4f4fd19eU, 0xdcdc7fa3U, - 0x22226644U, 0x2a2a7e54U, 0x9090ab3bU, 0x8888830bU, - 0x4646ca8cU, 0xeeee29c7U, 0xb8b8d36bU, 0x14143c28U, - 0xdede79a7U, 0x5e5ee2bcU, 0x0b0b1d16U, 0xdbdb76adU, - 0xe0e03bdbU, 0x32325664U, 0x3a3a4e74U, 0x0a0a1e14U, - 0x4949db92U, 0x06060a0cU, 0x24246c48U, 0x5c5ce4b8U, - 0xc2c25d9fU, 0xd3d36ebdU, 0xacacef43U, 0x6262a6c4U, - 0x9191a839U, 0x9595a431U, 0xe4e437d3U, 0x79798bf2U, - 0xe7e732d5U, 0xc8c8438bU, 0x3737596eU, 0x6d6db7daU, - 0x8d8d8c01U, 0xd5d564b1U, 0x4e4ed29cU, 0xa9a9e049U, - 0x6c6cb4d8U, 0x5656faacU, 0xf4f407f3U, 0xeaea25cfU, - 0x6565afcaU, 0x7a7a8ef4U, 0xaeaee947U, 0x08081810U, - 0xbabad56fU, 0x787888f0U, 0x25256f4aU, 0x2e2e725cU, - 0x1c1c2438U, 0xa6a6f157U, 0xb4b4c773U, 0xc6c65197U, - 0xe8e823cbU, 0xdddd7ca1U, 0x74749ce8U, 0x1f1f213eU, - 0x4b4bdd96U, 0xbdbddc61U, 0x8b8b860dU, 0x8a8a850fU, - 0x707090e0U, 0x3e3e427cU, 0xb5b5c471U, 0x6666aaccU, - 0x4848d890U, 0x03030506U, 0xf6f601f7U, 0x0e0e121cU, - 0x6161a3c2U, 0x35355f6aU, 0x5757f9aeU, 0xb9b9d069U, - 0x86869117U, 0xc1c15899U, 0x1d1d273aU, 0x9e9eb927U, - 0xe1e138d9U, 0xf8f813ebU, 0x9898b32bU, 0x11113322U, - 0x6969bbd2U, 0xd9d970a9U, 0x8e8e8907U, 0x9494a733U, - 0x9b9bb62dU, 0x1e1e223cU, 0x87879215U, 0xe9e920c9U, - 0xcece4987U, 0x5555ffaaU, 0x28287850U, 0xdfdf7aa5U, - 0x8c8c8f03U, 0xa1a1f859U, 0x89898009U, 0x0d0d171aU, - 0xbfbfda65U, 0xe6e631d7U, 0x4242c684U, 0x6868b8d0U, - 0x4141c382U, 0x9999b029U, 0x2d2d775aU, 0x0f0f111eU, - 0xb0b0cb7bU, 0x5454fca8U, 0xbbbbd66dU, 0x16163a2cU, + 0x6363a5c6U, 0x7c7c84f8U, 0x777799eeU, 0x7b7b8df6U, 0xf2f20dffU, 0x6b6bbdd6U, + 0x6f6fb1deU, 0xc5c55491U, 0x30305060U, 0x01010302U, 0x6767a9ceU, 0x2b2b7d56U, + 0xfefe19e7U, 0xd7d762b5U, 0xababe64dU, 0x76769aecU, 0xcaca458fU, 0x82829d1fU, + 0xc9c94089U, 0x7d7d87faU, 0xfafa15efU, 0x5959ebb2U, 0x4747c98eU, 0xf0f00bfbU, + 0xadadec41U, 0xd4d467b3U, 0xa2a2fd5fU, 0xafafea45U, 0x9c9cbf23U, 0xa4a4f753U, + 0x727296e4U, 0xc0c05b9bU, 0xb7b7c275U, 0xfdfd1ce1U, 0x9393ae3dU, 0x26266a4cU, + 0x36365a6cU, 0x3f3f417eU, 0xf7f702f5U, 0xcccc4f83U, 0x34345c68U, 0xa5a5f451U, + 0xe5e534d1U, 0xf1f108f9U, 0x717193e2U, 0xd8d873abU, 0x31315362U, 0x15153f2aU, + 0x04040c08U, 0xc7c75295U, 0x23236546U, 0xc3c35e9dU, 0x18182830U, 0x9696a137U, + 0x05050f0aU, 0x9a9ab52fU, 0x0707090eU, 0x12123624U, 0x80809b1bU, 0xe2e23ddfU, + 0xebeb26cdU, 0x2727694eU, 0xb2b2cd7fU, 0x75759feaU, 0x09091b12U, 0x83839e1dU, + 0x2c2c7458U, 0x1a1a2e34U, 0x1b1b2d36U, 0x6e6eb2dcU, 0x5a5aeeb4U, 0xa0a0fb5bU, + 0x5252f6a4U, 0x3b3b4d76U, 0xd6d661b7U, 0xb3b3ce7dU, 0x29297b52U, 0xe3e33eddU, + 0x2f2f715eU, 0x84849713U, 0x5353f5a6U, 0xd1d168b9U, 0x00000000U, 0xeded2cc1U, + 0x20206040U, 0xfcfc1fe3U, 0xb1b1c879U, 0x5b5bedb6U, 0x6a6abed4U, 0xcbcb468dU, + 0xbebed967U, 0x39394b72U, 0x4a4ade94U, 0x4c4cd498U, 0x5858e8b0U, 0xcfcf4a85U, + 0xd0d06bbbU, 0xefef2ac5U, 0xaaaae54fU, 0xfbfb16edU, 0x4343c586U, 0x4d4dd79aU, + 0x33335566U, 0x85859411U, 0x4545cf8aU, 0xf9f910e9U, 0x02020604U, 0x7f7f81feU, + 0x5050f0a0U, 0x3c3c4478U, 0x9f9fba25U, 0xa8a8e34bU, 0x5151f3a2U, 0xa3a3fe5dU, + 0x4040c080U, 0x8f8f8a05U, 0x9292ad3fU, 0x9d9dbc21U, 0x38384870U, 0xf5f504f1U, + 0xbcbcdf63U, 0xb6b6c177U, 0xdada75afU, 0x21216342U, 0x10103020U, 0xffff1ae5U, + 0xf3f30efdU, 0xd2d26dbfU, 0xcdcd4c81U, 0x0c0c1418U, 0x13133526U, 0xecec2fc3U, + 0x5f5fe1beU, 0x9797a235U, 0x4444cc88U, 0x1717392eU, 0xc4c45793U, 0xa7a7f255U, + 0x7e7e82fcU, 0x3d3d477aU, 0x6464acc8U, 0x5d5de7baU, 0x19192b32U, 0x737395e6U, + 0x6060a0c0U, 0x81819819U, 0x4f4fd19eU, 0xdcdc7fa3U, 0x22226644U, 0x2a2a7e54U, + 0x9090ab3bU, 0x8888830bU, 0x4646ca8cU, 0xeeee29c7U, 0xb8b8d36bU, 0x14143c28U, + 0xdede79a7U, 0x5e5ee2bcU, 0x0b0b1d16U, 0xdbdb76adU, 0xe0e03bdbU, 0x32325664U, + 0x3a3a4e74U, 0x0a0a1e14U, 0x4949db92U, 0x06060a0cU, 0x24246c48U, 0x5c5ce4b8U, + 0xc2c25d9fU, 0xd3d36ebdU, 0xacacef43U, 0x6262a6c4U, 0x9191a839U, 0x9595a431U, + 0xe4e437d3U, 0x79798bf2U, 0xe7e732d5U, 0xc8c8438bU, 0x3737596eU, 0x6d6db7daU, + 0x8d8d8c01U, 0xd5d564b1U, 0x4e4ed29cU, 0xa9a9e049U, 0x6c6cb4d8U, 0x5656faacU, + 0xf4f407f3U, 0xeaea25cfU, 0x6565afcaU, 0x7a7a8ef4U, 0xaeaee947U, 0x08081810U, + 0xbabad56fU, 0x787888f0U, 0x25256f4aU, 0x2e2e725cU, 0x1c1c2438U, 0xa6a6f157U, + 0xb4b4c773U, 0xc6c65197U, 0xe8e823cbU, 0xdddd7ca1U, 0x74749ce8U, 0x1f1f213eU, + 0x4b4bdd96U, 0xbdbddc61U, 0x8b8b860dU, 0x8a8a850fU, 0x707090e0U, 0x3e3e427cU, + 0xb5b5c471U, 0x6666aaccU, 0x4848d890U, 0x03030506U, 0xf6f601f7U, 0x0e0e121cU, + 0x6161a3c2U, 0x35355f6aU, 0x5757f9aeU, 0xb9b9d069U, 0x86869117U, 0xc1c15899U, + 0x1d1d273aU, 0x9e9eb927U, 0xe1e138d9U, 0xf8f813ebU, 0x9898b32bU, 0x11113322U, + 0x6969bbd2U, 0xd9d970a9U, 0x8e8e8907U, 0x9494a733U, 0x9b9bb62dU, 0x1e1e223cU, + 0x87879215U, 0xe9e920c9U, 0xcece4987U, 0x5555ffaaU, 0x28287850U, 0xdfdf7aa5U, + 0x8c8c8f03U, 0xa1a1f859U, 0x89898009U, 0x0d0d171aU, 0xbfbfda65U, 0xe6e631d7U, + 0x4242c684U, 0x6868b8d0U, 0x4141c382U, 0x9999b029U, 0x2d2d775aU, 0x0f0f111eU, + 0xb0b0cb7bU, 0x5454fca8U, 0xbbbbd66dU, 0x16163a2cU, }; static const u32 Te4[256] = { - 0x63636363U, 0x7c7c7c7cU, 0x77777777U, 0x7b7b7b7bU, - 0xf2f2f2f2U, 0x6b6b6b6bU, 0x6f6f6f6fU, 0xc5c5c5c5U, - 0x30303030U, 0x01010101U, 0x67676767U, 0x2b2b2b2bU, - 0xfefefefeU, 0xd7d7d7d7U, 0xababababU, 0x76767676U, - 0xcacacacaU, 0x82828282U, 0xc9c9c9c9U, 0x7d7d7d7dU, - 0xfafafafaU, 0x59595959U, 0x47474747U, 0xf0f0f0f0U, - 0xadadadadU, 0xd4d4d4d4U, 0xa2a2a2a2U, 0xafafafafU, - 0x9c9c9c9cU, 0xa4a4a4a4U, 0x72727272U, 0xc0c0c0c0U, - 0xb7b7b7b7U, 0xfdfdfdfdU, 0x93939393U, 0x26262626U, - 0x36363636U, 0x3f3f3f3fU, 0xf7f7f7f7U, 0xccccccccU, - 0x34343434U, 0xa5a5a5a5U, 0xe5e5e5e5U, 0xf1f1f1f1U, - 0x71717171U, 0xd8d8d8d8U, 0x31313131U, 0x15151515U, - 0x04040404U, 0xc7c7c7c7U, 0x23232323U, 0xc3c3c3c3U, - 0x18181818U, 0x96969696U, 0x05050505U, 0x9a9a9a9aU, - 0x07070707U, 0x12121212U, 0x80808080U, 0xe2e2e2e2U, - 0xebebebebU, 0x27272727U, 0xb2b2b2b2U, 0x75757575U, - 0x09090909U, 0x83838383U, 0x2c2c2c2cU, 0x1a1a1a1aU, - 0x1b1b1b1bU, 0x6e6e6e6eU, 0x5a5a5a5aU, 0xa0a0a0a0U, - 0x52525252U, 0x3b3b3b3bU, 0xd6d6d6d6U, 0xb3b3b3b3U, - 0x29292929U, 0xe3e3e3e3U, 0x2f2f2f2fU, 0x84848484U, - 0x53535353U, 0xd1d1d1d1U, 0x00000000U, 0xededededU, - 0x20202020U, 0xfcfcfcfcU, 0xb1b1b1b1U, 0x5b5b5b5bU, - 0x6a6a6a6aU, 0xcbcbcbcbU, 0xbebebebeU, 0x39393939U, - 0x4a4a4a4aU, 0x4c4c4c4cU, 0x58585858U, 0xcfcfcfcfU, - 0xd0d0d0d0U, 0xefefefefU, 0xaaaaaaaaU, 0xfbfbfbfbU, - 0x43434343U, 0x4d4d4d4dU, 0x33333333U, 0x85858585U, - 0x45454545U, 0xf9f9f9f9U, 0x02020202U, 0x7f7f7f7fU, - 0x50505050U, 0x3c3c3c3cU, 0x9f9f9f9fU, 0xa8a8a8a8U, - 0x51515151U, 0xa3a3a3a3U, 0x40404040U, 0x8f8f8f8fU, - 0x92929292U, 0x9d9d9d9dU, 0x38383838U, 0xf5f5f5f5U, - 0xbcbcbcbcU, 0xb6b6b6b6U, 0xdadadadaU, 0x21212121U, - 0x10101010U, 0xffffffffU, 0xf3f3f3f3U, 0xd2d2d2d2U, - 0xcdcdcdcdU, 0x0c0c0c0cU, 0x13131313U, 0xececececU, - 0x5f5f5f5fU, 0x97979797U, 0x44444444U, 0x17171717U, - 0xc4c4c4c4U, 0xa7a7a7a7U, 0x7e7e7e7eU, 0x3d3d3d3dU, - 0x64646464U, 0x5d5d5d5dU, 0x19191919U, 0x73737373U, - 0x60606060U, 0x81818181U, 0x4f4f4f4fU, 0xdcdcdcdcU, - 0x22222222U, 0x2a2a2a2aU, 0x90909090U, 0x88888888U, - 0x46464646U, 0xeeeeeeeeU, 0xb8b8b8b8U, 0x14141414U, - 0xdedededeU, 0x5e5e5e5eU, 0x0b0b0b0bU, 0xdbdbdbdbU, - 0xe0e0e0e0U, 0x32323232U, 0x3a3a3a3aU, 0x0a0a0a0aU, - 0x49494949U, 0x06060606U, 0x24242424U, 0x5c5c5c5cU, - 0xc2c2c2c2U, 0xd3d3d3d3U, 0xacacacacU, 0x62626262U, - 0x91919191U, 0x95959595U, 0xe4e4e4e4U, 0x79797979U, - 0xe7e7e7e7U, 0xc8c8c8c8U, 0x37373737U, 0x6d6d6d6dU, - 0x8d8d8d8dU, 0xd5d5d5d5U, 0x4e4e4e4eU, 0xa9a9a9a9U, - 0x6c6c6c6cU, 0x56565656U, 0xf4f4f4f4U, 0xeaeaeaeaU, - 0x65656565U, 0x7a7a7a7aU, 0xaeaeaeaeU, 0x08080808U, - 0xbabababaU, 0x78787878U, 0x25252525U, 0x2e2e2e2eU, - 0x1c1c1c1cU, 0xa6a6a6a6U, 0xb4b4b4b4U, 0xc6c6c6c6U, - 0xe8e8e8e8U, 0xddddddddU, 0x74747474U, 0x1f1f1f1fU, - 0x4b4b4b4bU, 0xbdbdbdbdU, 0x8b8b8b8bU, 0x8a8a8a8aU, - 0x70707070U, 0x3e3e3e3eU, 0xb5b5b5b5U, 0x66666666U, - 0x48484848U, 0x03030303U, 0xf6f6f6f6U, 0x0e0e0e0eU, - 0x61616161U, 0x35353535U, 0x57575757U, 0xb9b9b9b9U, - 0x86868686U, 0xc1c1c1c1U, 0x1d1d1d1dU, 0x9e9e9e9eU, - 0xe1e1e1e1U, 0xf8f8f8f8U, 0x98989898U, 0x11111111U, - 0x69696969U, 0xd9d9d9d9U, 0x8e8e8e8eU, 0x94949494U, - 0x9b9b9b9bU, 0x1e1e1e1eU, 0x87878787U, 0xe9e9e9e9U, - 0xcecececeU, 0x55555555U, 0x28282828U, 0xdfdfdfdfU, - 0x8c8c8c8cU, 0xa1a1a1a1U, 0x89898989U, 0x0d0d0d0dU, - 0xbfbfbfbfU, 0xe6e6e6e6U, 0x42424242U, 0x68686868U, - 0x41414141U, 0x99999999U, 0x2d2d2d2dU, 0x0f0f0f0fU, - 0xb0b0b0b0U, 0x54545454U, 0xbbbbbbbbU, 0x16161616U, + 0x63636363U, 0x7c7c7c7cU, 0x77777777U, 0x7b7b7b7bU, 0xf2f2f2f2U, 0x6b6b6b6bU, + 0x6f6f6f6fU, 0xc5c5c5c5U, 0x30303030U, 0x01010101U, 0x67676767U, 0x2b2b2b2bU, + 0xfefefefeU, 0xd7d7d7d7U, 0xababababU, 0x76767676U, 0xcacacacaU, 0x82828282U, + 0xc9c9c9c9U, 0x7d7d7d7dU, 0xfafafafaU, 0x59595959U, 0x47474747U, 0xf0f0f0f0U, + 0xadadadadU, 0xd4d4d4d4U, 0xa2a2a2a2U, 0xafafafafU, 0x9c9c9c9cU, 0xa4a4a4a4U, + 0x72727272U, 0xc0c0c0c0U, 0xb7b7b7b7U, 0xfdfdfdfdU, 0x93939393U, 0x26262626U, + 0x36363636U, 0x3f3f3f3fU, 0xf7f7f7f7U, 0xccccccccU, 0x34343434U, 0xa5a5a5a5U, + 0xe5e5e5e5U, 0xf1f1f1f1U, 0x71717171U, 0xd8d8d8d8U, 0x31313131U, 0x15151515U, + 0x04040404U, 0xc7c7c7c7U, 0x23232323U, 0xc3c3c3c3U, 0x18181818U, 0x96969696U, + 0x05050505U, 0x9a9a9a9aU, 0x07070707U, 0x12121212U, 0x80808080U, 0xe2e2e2e2U, + 0xebebebebU, 0x27272727U, 0xb2b2b2b2U, 0x75757575U, 0x09090909U, 0x83838383U, + 0x2c2c2c2cU, 0x1a1a1a1aU, 0x1b1b1b1bU, 0x6e6e6e6eU, 0x5a5a5a5aU, 0xa0a0a0a0U, + 0x52525252U, 0x3b3b3b3bU, 0xd6d6d6d6U, 0xb3b3b3b3U, 0x29292929U, 0xe3e3e3e3U, + 0x2f2f2f2fU, 0x84848484U, 0x53535353U, 0xd1d1d1d1U, 0x00000000U, 0xededededU, + 0x20202020U, 0xfcfcfcfcU, 0xb1b1b1b1U, 0x5b5b5b5bU, 0x6a6a6a6aU, 0xcbcbcbcbU, + 0xbebebebeU, 0x39393939U, 0x4a4a4a4aU, 0x4c4c4c4cU, 0x58585858U, 0xcfcfcfcfU, + 0xd0d0d0d0U, 0xefefefefU, 0xaaaaaaaaU, 0xfbfbfbfbU, 0x43434343U, 0x4d4d4d4dU, + 0x33333333U, 0x85858585U, 0x45454545U, 0xf9f9f9f9U, 0x02020202U, 0x7f7f7f7fU, + 0x50505050U, 0x3c3c3c3cU, 0x9f9f9f9fU, 0xa8a8a8a8U, 0x51515151U, 0xa3a3a3a3U, + 0x40404040U, 0x8f8f8f8fU, 0x92929292U, 0x9d9d9d9dU, 0x38383838U, 0xf5f5f5f5U, + 0xbcbcbcbcU, 0xb6b6b6b6U, 0xdadadadaU, 0x21212121U, 0x10101010U, 0xffffffffU, + 0xf3f3f3f3U, 0xd2d2d2d2U, 0xcdcdcdcdU, 0x0c0c0c0cU, 0x13131313U, 0xececececU, + 0x5f5f5f5fU, 0x97979797U, 0x44444444U, 0x17171717U, 0xc4c4c4c4U, 0xa7a7a7a7U, + 0x7e7e7e7eU, 0x3d3d3d3dU, 0x64646464U, 0x5d5d5d5dU, 0x19191919U, 0x73737373U, + 0x60606060U, 0x81818181U, 0x4f4f4f4fU, 0xdcdcdcdcU, 0x22222222U, 0x2a2a2a2aU, + 0x90909090U, 0x88888888U, 0x46464646U, 0xeeeeeeeeU, 0xb8b8b8b8U, 0x14141414U, + 0xdedededeU, 0x5e5e5e5eU, 0x0b0b0b0bU, 0xdbdbdbdbU, 0xe0e0e0e0U, 0x32323232U, + 0x3a3a3a3aU, 0x0a0a0a0aU, 0x49494949U, 0x06060606U, 0x24242424U, 0x5c5c5c5cU, + 0xc2c2c2c2U, 0xd3d3d3d3U, 0xacacacacU, 0x62626262U, 0x91919191U, 0x95959595U, + 0xe4e4e4e4U, 0x79797979U, 0xe7e7e7e7U, 0xc8c8c8c8U, 0x37373737U, 0x6d6d6d6dU, + 0x8d8d8d8dU, 0xd5d5d5d5U, 0x4e4e4e4eU, 0xa9a9a9a9U, 0x6c6c6c6cU, 0x56565656U, + 0xf4f4f4f4U, 0xeaeaeaeaU, 0x65656565U, 0x7a7a7a7aU, 0xaeaeaeaeU, 0x08080808U, + 0xbabababaU, 0x78787878U, 0x25252525U, 0x2e2e2e2eU, 0x1c1c1c1cU, 0xa6a6a6a6U, + 0xb4b4b4b4U, 0xc6c6c6c6U, 0xe8e8e8e8U, 0xddddddddU, 0x74747474U, 0x1f1f1f1fU, + 0x4b4b4b4bU, 0xbdbdbdbdU, 0x8b8b8b8bU, 0x8a8a8a8aU, 0x70707070U, 0x3e3e3e3eU, + 0xb5b5b5b5U, 0x66666666U, 0x48484848U, 0x03030303U, 0xf6f6f6f6U, 0x0e0e0e0eU, + 0x61616161U, 0x35353535U, 0x57575757U, 0xb9b9b9b9U, 0x86868686U, 0xc1c1c1c1U, + 0x1d1d1d1dU, 0x9e9e9e9eU, 0xe1e1e1e1U, 0xf8f8f8f8U, 0x98989898U, 0x11111111U, + 0x69696969U, 0xd9d9d9d9U, 0x8e8e8e8eU, 0x94949494U, 0x9b9b9b9bU, 0x1e1e1e1eU, + 0x87878787U, 0xe9e9e9e9U, 0xcecececeU, 0x55555555U, 0x28282828U, 0xdfdfdfdfU, + 0x8c8c8c8cU, 0xa1a1a1a1U, 0x89898989U, 0x0d0d0d0dU, 0xbfbfbfbfU, 0xe6e6e6e6U, + 0x42424242U, 0x68686868U, 0x41414141U, 0x99999999U, 0x2d2d2d2dU, 0x0f0f0f0fU, + 0xb0b0b0b0U, 0x54545454U, 0xbbbbbbbbU, 0x16161616U, }; static const u32 Td0[256] = { - 0x51f4a750U, 0x7e416553U, 0x1a17a4c3U, 0x3a275e96U, - 0x3bab6bcbU, 0x1f9d45f1U, 0xacfa58abU, 0x4be30393U, - 0x2030fa55U, 0xad766df6U, 0x88cc7691U, 0xf5024c25U, - 0x4fe5d7fcU, 0xc52acbd7U, 0x26354480U, 0xb562a38fU, - 0xdeb15a49U, 0x25ba1b67U, 0x45ea0e98U, 0x5dfec0e1U, - 0xc32f7502U, 0x814cf012U, 0x8d4697a3U, 0x6bd3f9c6U, - 0x038f5fe7U, 0x15929c95U, 0xbf6d7aebU, 0x955259daU, - 0xd4be832dU, 0x587421d3U, 0x49e06929U, 0x8ec9c844U, - 0x75c2896aU, 0xf48e7978U, 0x99583e6bU, 0x27b971ddU, - 0xbee14fb6U, 0xf088ad17U, 0xc920ac66U, 0x7dce3ab4U, - 0x63df4a18U, 0xe51a3182U, 0x97513360U, 0x62537f45U, - 0xb16477e0U, 0xbb6bae84U, 0xfe81a01cU, 0xf9082b94U, - 0x70486858U, 0x8f45fd19U, 0x94de6c87U, 0x527bf8b7U, - 0xab73d323U, 0x724b02e2U, 0xe31f8f57U, 0x6655ab2aU, - 0xb2eb2807U, 0x2fb5c203U, 0x86c57b9aU, 0xd33708a5U, - 0x302887f2U, 0x23bfa5b2U, 0x02036abaU, 0xed16825cU, - 0x8acf1c2bU, 0xa779b492U, 0xf307f2f0U, 0x4e69e2a1U, - 0x65daf4cdU, 0x0605bed5U, 0xd134621fU, 0xc4a6fe8aU, - 0x342e539dU, 0xa2f355a0U, 0x058ae132U, 0xa4f6eb75U, - 0x0b83ec39U, 0x4060efaaU, 0x5e719f06U, 0xbd6e1051U, - 0x3e218af9U, 0x96dd063dU, 0xdd3e05aeU, 0x4de6bd46U, - 0x91548db5U, 0x71c45d05U, 0x0406d46fU, 0x605015ffU, - 0x1998fb24U, 0xd6bde997U, 0x894043ccU, 0x67d99e77U, - 0xb0e842bdU, 0x07898b88U, 0xe7195b38U, 0x79c8eedbU, - 0xa17c0a47U, 0x7c420fe9U, 0xf8841ec9U, 0x00000000U, - 0x09808683U, 0x322bed48U, 0x1e1170acU, 0x6c5a724eU, - 0xfd0efffbU, 0x0f853856U, 0x3daed51eU, 0x362d3927U, - 0x0a0fd964U, 0x685ca621U, 0x9b5b54d1U, 0x24362e3aU, - 0x0c0a67b1U, 0x9357e70fU, 0xb4ee96d2U, 0x1b9b919eU, - 0x80c0c54fU, 0x61dc20a2U, 0x5a774b69U, 0x1c121a16U, - 0xe293ba0aU, 0xc0a02ae5U, 0x3c22e043U, 0x121b171dU, - 0x0e090d0bU, 0xf28bc7adU, 0x2db6a8b9U, 0x141ea9c8U, - 0x57f11985U, 0xaf75074cU, 0xee99ddbbU, 0xa37f60fdU, - 0xf701269fU, 0x5c72f5bcU, 0x44663bc5U, 0x5bfb7e34U, - 0x8b432976U, 0xcb23c6dcU, 0xb6edfc68U, 0xb8e4f163U, - 0xd731dccaU, 0x42638510U, 0x13972240U, 0x84c61120U, - 0x854a247dU, 0xd2bb3df8U, 0xaef93211U, 0xc729a16dU, - 0x1d9e2f4bU, 0xdcb230f3U, 0x0d8652ecU, 0x77c1e3d0U, - 0x2bb3166cU, 0xa970b999U, 0x119448faU, 0x47e96422U, - 0xa8fc8cc4U, 0xa0f03f1aU, 0x567d2cd8U, 0x223390efU, - 0x87494ec7U, 0xd938d1c1U, 0x8ccaa2feU, 0x98d40b36U, - 0xa6f581cfU, 0xa57ade28U, 0xdab78e26U, 0x3fadbfa4U, - 0x2c3a9de4U, 0x5078920dU, 0x6a5fcc9bU, 0x547e4662U, - 0xf68d13c2U, 0x90d8b8e8U, 0x2e39f75eU, 0x82c3aff5U, - 0x9f5d80beU, 0x69d0937cU, 0x6fd52da9U, 0xcf2512b3U, - 0xc8ac993bU, 0x10187da7U, 0xe89c636eU, 0xdb3bbb7bU, - 0xcd267809U, 0x6e5918f4U, 0xec9ab701U, 0x834f9aa8U, - 0xe6956e65U, 0xaaffe67eU, 0x21bccf08U, 0xef15e8e6U, - 0xbae79bd9U, 0x4a6f36ceU, 0xea9f09d4U, 0x29b07cd6U, - 0x31a4b2afU, 0x2a3f2331U, 0xc6a59430U, 0x35a266c0U, - 0x744ebc37U, 0xfc82caa6U, 0xe090d0b0U, 0x33a7d815U, - 0xf104984aU, 0x41ecdaf7U, 0x7fcd500eU, 0x1791f62fU, - 0x764dd68dU, 0x43efb04dU, 0xccaa4d54U, 0xe49604dfU, - 0x9ed1b5e3U, 0x4c6a881bU, 0xc12c1fb8U, 0x4665517fU, - 0x9d5eea04U, 0x018c355dU, 0xfa877473U, 0xfb0b412eU, - 0xb3671d5aU, 0x92dbd252U, 0xe9105633U, 0x6dd64713U, - 0x9ad7618cU, 0x37a10c7aU, 0x59f8148eU, 0xeb133c89U, - 0xcea927eeU, 0xb761c935U, 0xe11ce5edU, 0x7a47b13cU, - 0x9cd2df59U, 0x55f2733fU, 0x1814ce79U, 0x73c737bfU, - 0x53f7cdeaU, 0x5ffdaa5bU, 0xdf3d6f14U, 0x7844db86U, - 0xcaaff381U, 0xb968c43eU, 0x3824342cU, 0xc2a3405fU, - 0x161dc372U, 0xbce2250cU, 0x283c498bU, 0xff0d9541U, - 0x39a80171U, 0x080cb3deU, 0xd8b4e49cU, 0x6456c190U, - 0x7bcb8461U, 0xd532b670U, 0x486c5c74U, 0xd0b85742U, + 0x51f4a750U, 0x7e416553U, 0x1a17a4c3U, 0x3a275e96U, 0x3bab6bcbU, 0x1f9d45f1U, + 0xacfa58abU, 0x4be30393U, 0x2030fa55U, 0xad766df6U, 0x88cc7691U, 0xf5024c25U, + 0x4fe5d7fcU, 0xc52acbd7U, 0x26354480U, 0xb562a38fU, 0xdeb15a49U, 0x25ba1b67U, + 0x45ea0e98U, 0x5dfec0e1U, 0xc32f7502U, 0x814cf012U, 0x8d4697a3U, 0x6bd3f9c6U, + 0x038f5fe7U, 0x15929c95U, 0xbf6d7aebU, 0x955259daU, 0xd4be832dU, 0x587421d3U, + 0x49e06929U, 0x8ec9c844U, 0x75c2896aU, 0xf48e7978U, 0x99583e6bU, 0x27b971ddU, + 0xbee14fb6U, 0xf088ad17U, 0xc920ac66U, 0x7dce3ab4U, 0x63df4a18U, 0xe51a3182U, + 0x97513360U, 0x62537f45U, 0xb16477e0U, 0xbb6bae84U, 0xfe81a01cU, 0xf9082b94U, + 0x70486858U, 0x8f45fd19U, 0x94de6c87U, 0x527bf8b7U, 0xab73d323U, 0x724b02e2U, + 0xe31f8f57U, 0x6655ab2aU, 0xb2eb2807U, 0x2fb5c203U, 0x86c57b9aU, 0xd33708a5U, + 0x302887f2U, 0x23bfa5b2U, 0x02036abaU, 0xed16825cU, 0x8acf1c2bU, 0xa779b492U, + 0xf307f2f0U, 0x4e69e2a1U, 0x65daf4cdU, 0x0605bed5U, 0xd134621fU, 0xc4a6fe8aU, + 0x342e539dU, 0xa2f355a0U, 0x058ae132U, 0xa4f6eb75U, 0x0b83ec39U, 0x4060efaaU, + 0x5e719f06U, 0xbd6e1051U, 0x3e218af9U, 0x96dd063dU, 0xdd3e05aeU, 0x4de6bd46U, + 0x91548db5U, 0x71c45d05U, 0x0406d46fU, 0x605015ffU, 0x1998fb24U, 0xd6bde997U, + 0x894043ccU, 0x67d99e77U, 0xb0e842bdU, 0x07898b88U, 0xe7195b38U, 0x79c8eedbU, + 0xa17c0a47U, 0x7c420fe9U, 0xf8841ec9U, 0x00000000U, 0x09808683U, 0x322bed48U, + 0x1e1170acU, 0x6c5a724eU, 0xfd0efffbU, 0x0f853856U, 0x3daed51eU, 0x362d3927U, + 0x0a0fd964U, 0x685ca621U, 0x9b5b54d1U, 0x24362e3aU, 0x0c0a67b1U, 0x9357e70fU, + 0xb4ee96d2U, 0x1b9b919eU, 0x80c0c54fU, 0x61dc20a2U, 0x5a774b69U, 0x1c121a16U, + 0xe293ba0aU, 0xc0a02ae5U, 0x3c22e043U, 0x121b171dU, 0x0e090d0bU, 0xf28bc7adU, + 0x2db6a8b9U, 0x141ea9c8U, 0x57f11985U, 0xaf75074cU, 0xee99ddbbU, 0xa37f60fdU, + 0xf701269fU, 0x5c72f5bcU, 0x44663bc5U, 0x5bfb7e34U, 0x8b432976U, 0xcb23c6dcU, + 0xb6edfc68U, 0xb8e4f163U, 0xd731dccaU, 0x42638510U, 0x13972240U, 0x84c61120U, + 0x854a247dU, 0xd2bb3df8U, 0xaef93211U, 0xc729a16dU, 0x1d9e2f4bU, 0xdcb230f3U, + 0x0d8652ecU, 0x77c1e3d0U, 0x2bb3166cU, 0xa970b999U, 0x119448faU, 0x47e96422U, + 0xa8fc8cc4U, 0xa0f03f1aU, 0x567d2cd8U, 0x223390efU, 0x87494ec7U, 0xd938d1c1U, + 0x8ccaa2feU, 0x98d40b36U, 0xa6f581cfU, 0xa57ade28U, 0xdab78e26U, 0x3fadbfa4U, + 0x2c3a9de4U, 0x5078920dU, 0x6a5fcc9bU, 0x547e4662U, 0xf68d13c2U, 0x90d8b8e8U, + 0x2e39f75eU, 0x82c3aff5U, 0x9f5d80beU, 0x69d0937cU, 0x6fd52da9U, 0xcf2512b3U, + 0xc8ac993bU, 0x10187da7U, 0xe89c636eU, 0xdb3bbb7bU, 0xcd267809U, 0x6e5918f4U, + 0xec9ab701U, 0x834f9aa8U, 0xe6956e65U, 0xaaffe67eU, 0x21bccf08U, 0xef15e8e6U, + 0xbae79bd9U, 0x4a6f36ceU, 0xea9f09d4U, 0x29b07cd6U, 0x31a4b2afU, 0x2a3f2331U, + 0xc6a59430U, 0x35a266c0U, 0x744ebc37U, 0xfc82caa6U, 0xe090d0b0U, 0x33a7d815U, + 0xf104984aU, 0x41ecdaf7U, 0x7fcd500eU, 0x1791f62fU, 0x764dd68dU, 0x43efb04dU, + 0xccaa4d54U, 0xe49604dfU, 0x9ed1b5e3U, 0x4c6a881bU, 0xc12c1fb8U, 0x4665517fU, + 0x9d5eea04U, 0x018c355dU, 0xfa877473U, 0xfb0b412eU, 0xb3671d5aU, 0x92dbd252U, + 0xe9105633U, 0x6dd64713U, 0x9ad7618cU, 0x37a10c7aU, 0x59f8148eU, 0xeb133c89U, + 0xcea927eeU, 0xb761c935U, 0xe11ce5edU, 0x7a47b13cU, 0x9cd2df59U, 0x55f2733fU, + 0x1814ce79U, 0x73c737bfU, 0x53f7cdeaU, 0x5ffdaa5bU, 0xdf3d6f14U, 0x7844db86U, + 0xcaaff381U, 0xb968c43eU, 0x3824342cU, 0xc2a3405fU, 0x161dc372U, 0xbce2250cU, + 0x283c498bU, 0xff0d9541U, 0x39a80171U, 0x080cb3deU, 0xd8b4e49cU, 0x6456c190U, + 0x7bcb8461U, 0xd532b670U, 0x486c5c74U, 0xd0b85742U, }; static const u32 Td1[256] = { - 0x5051f4a7U, 0x537e4165U, 0xc31a17a4U, 0x963a275eU, - 0xcb3bab6bU, 0xf11f9d45U, 0xabacfa58U, 0x934be303U, - 0x552030faU, 0xf6ad766dU, 0x9188cc76U, 0x25f5024cU, - 0xfc4fe5d7U, 0xd7c52acbU, 0x80263544U, 0x8fb562a3U, - 0x49deb15aU, 0x6725ba1bU, 0x9845ea0eU, 0xe15dfec0U, - 0x02c32f75U, 0x12814cf0U, 0xa38d4697U, 0xc66bd3f9U, - 0xe7038f5fU, 0x9515929cU, 0xebbf6d7aU, 0xda955259U, - 0x2dd4be83U, 0xd3587421U, 0x2949e069U, 0x448ec9c8U, - 0x6a75c289U, 0x78f48e79U, 0x6b99583eU, 0xdd27b971U, - 0xb6bee14fU, 0x17f088adU, 0x66c920acU, 0xb47dce3aU, - 0x1863df4aU, 0x82e51a31U, 0x60975133U, 0x4562537fU, - 0xe0b16477U, 0x84bb6baeU, 0x1cfe81a0U, 0x94f9082bU, - 0x58704868U, 0x198f45fdU, 0x8794de6cU, 0xb7527bf8U, - 0x23ab73d3U, 0xe2724b02U, 0x57e31f8fU, 0x2a6655abU, - 0x07b2eb28U, 0x032fb5c2U, 0x9a86c57bU, 0xa5d33708U, - 0xf2302887U, 0xb223bfa5U, 0xba02036aU, 0x5ced1682U, - 0x2b8acf1cU, 0x92a779b4U, 0xf0f307f2U, 0xa14e69e2U, - 0xcd65daf4U, 0xd50605beU, 0x1fd13462U, 0x8ac4a6feU, - 0x9d342e53U, 0xa0a2f355U, 0x32058ae1U, 0x75a4f6ebU, - 0x390b83ecU, 0xaa4060efU, 0x065e719fU, 0x51bd6e10U, - 0xf93e218aU, 0x3d96dd06U, 0xaedd3e05U, 0x464de6bdU, - 0xb591548dU, 0x0571c45dU, 0x6f0406d4U, 0xff605015U, - 0x241998fbU, 0x97d6bde9U, 0xcc894043U, 0x7767d99eU, - 0xbdb0e842U, 0x8807898bU, 0x38e7195bU, 0xdb79c8eeU, - 0x47a17c0aU, 0xe97c420fU, 0xc9f8841eU, 0x00000000U, - 0x83098086U, 0x48322bedU, 0xac1e1170U, 0x4e6c5a72U, - 0xfbfd0effU, 0x560f8538U, 0x1e3daed5U, 0x27362d39U, - 0x640a0fd9U, 0x21685ca6U, 0xd19b5b54U, 0x3a24362eU, - 0xb10c0a67U, 0x0f9357e7U, 0xd2b4ee96U, 0x9e1b9b91U, - 0x4f80c0c5U, 0xa261dc20U, 0x695a774bU, 0x161c121aU, - 0x0ae293baU, 0xe5c0a02aU, 0x433c22e0U, 0x1d121b17U, - 0x0b0e090dU, 0xadf28bc7U, 0xb92db6a8U, 0xc8141ea9U, - 0x8557f119U, 0x4caf7507U, 0xbbee99ddU, 0xfda37f60U, - 0x9ff70126U, 0xbc5c72f5U, 0xc544663bU, 0x345bfb7eU, - 0x768b4329U, 0xdccb23c6U, 0x68b6edfcU, 0x63b8e4f1U, - 0xcad731dcU, 0x10426385U, 0x40139722U, 0x2084c611U, - 0x7d854a24U, 0xf8d2bb3dU, 0x11aef932U, 0x6dc729a1U, - 0x4b1d9e2fU, 0xf3dcb230U, 0xec0d8652U, 0xd077c1e3U, - 0x6c2bb316U, 0x99a970b9U, 0xfa119448U, 0x2247e964U, - 0xc4a8fc8cU, 0x1aa0f03fU, 0xd8567d2cU, 0xef223390U, - 0xc787494eU, 0xc1d938d1U, 0xfe8ccaa2U, 0x3698d40bU, - 0xcfa6f581U, 0x28a57adeU, 0x26dab78eU, 0xa43fadbfU, - 0xe42c3a9dU, 0x0d507892U, 0x9b6a5fccU, 0x62547e46U, - 0xc2f68d13U, 0xe890d8b8U, 0x5e2e39f7U, 0xf582c3afU, - 0xbe9f5d80U, 0x7c69d093U, 0xa96fd52dU, 0xb3cf2512U, - 0x3bc8ac99U, 0xa710187dU, 0x6ee89c63U, 0x7bdb3bbbU, - 0x09cd2678U, 0xf46e5918U, 0x01ec9ab7U, 0xa8834f9aU, - 0x65e6956eU, 0x7eaaffe6U, 0x0821bccfU, 0xe6ef15e8U, - 0xd9bae79bU, 0xce4a6f36U, 0xd4ea9f09U, 0xd629b07cU, - 0xaf31a4b2U, 0x312a3f23U, 0x30c6a594U, 0xc035a266U, - 0x37744ebcU, 0xa6fc82caU, 0xb0e090d0U, 0x1533a7d8U, - 0x4af10498U, 0xf741ecdaU, 0x0e7fcd50U, 0x2f1791f6U, - 0x8d764dd6U, 0x4d43efb0U, 0x54ccaa4dU, 0xdfe49604U, - 0xe39ed1b5U, 0x1b4c6a88U, 0xb8c12c1fU, 0x7f466551U, - 0x049d5eeaU, 0x5d018c35U, 0x73fa8774U, 0x2efb0b41U, - 0x5ab3671dU, 0x5292dbd2U, 0x33e91056U, 0x136dd647U, - 0x8c9ad761U, 0x7a37a10cU, 0x8e59f814U, 0x89eb133cU, - 0xeecea927U, 0x35b761c9U, 0xede11ce5U, 0x3c7a47b1U, - 0x599cd2dfU, 0x3f55f273U, 0x791814ceU, 0xbf73c737U, - 0xea53f7cdU, 0x5b5ffdaaU, 0x14df3d6fU, 0x867844dbU, - 0x81caaff3U, 0x3eb968c4U, 0x2c382434U, 0x5fc2a340U, - 0x72161dc3U, 0x0cbce225U, 0x8b283c49U, 0x41ff0d95U, - 0x7139a801U, 0xde080cb3U, 0x9cd8b4e4U, 0x906456c1U, - 0x617bcb84U, 0x70d532b6U, 0x74486c5cU, 0x42d0b857U, + 0x5051f4a7U, 0x537e4165U, 0xc31a17a4U, 0x963a275eU, 0xcb3bab6bU, 0xf11f9d45U, + 0xabacfa58U, 0x934be303U, 0x552030faU, 0xf6ad766dU, 0x9188cc76U, 0x25f5024cU, + 0xfc4fe5d7U, 0xd7c52acbU, 0x80263544U, 0x8fb562a3U, 0x49deb15aU, 0x6725ba1bU, + 0x9845ea0eU, 0xe15dfec0U, 0x02c32f75U, 0x12814cf0U, 0xa38d4697U, 0xc66bd3f9U, + 0xe7038f5fU, 0x9515929cU, 0xebbf6d7aU, 0xda955259U, 0x2dd4be83U, 0xd3587421U, + 0x2949e069U, 0x448ec9c8U, 0x6a75c289U, 0x78f48e79U, 0x6b99583eU, 0xdd27b971U, + 0xb6bee14fU, 0x17f088adU, 0x66c920acU, 0xb47dce3aU, 0x1863df4aU, 0x82e51a31U, + 0x60975133U, 0x4562537fU, 0xe0b16477U, 0x84bb6baeU, 0x1cfe81a0U, 0x94f9082bU, + 0x58704868U, 0x198f45fdU, 0x8794de6cU, 0xb7527bf8U, 0x23ab73d3U, 0xe2724b02U, + 0x57e31f8fU, 0x2a6655abU, 0x07b2eb28U, 0x032fb5c2U, 0x9a86c57bU, 0xa5d33708U, + 0xf2302887U, 0xb223bfa5U, 0xba02036aU, 0x5ced1682U, 0x2b8acf1cU, 0x92a779b4U, + 0xf0f307f2U, 0xa14e69e2U, 0xcd65daf4U, 0xd50605beU, 0x1fd13462U, 0x8ac4a6feU, + 0x9d342e53U, 0xa0a2f355U, 0x32058ae1U, 0x75a4f6ebU, 0x390b83ecU, 0xaa4060efU, + 0x065e719fU, 0x51bd6e10U, 0xf93e218aU, 0x3d96dd06U, 0xaedd3e05U, 0x464de6bdU, + 0xb591548dU, 0x0571c45dU, 0x6f0406d4U, 0xff605015U, 0x241998fbU, 0x97d6bde9U, + 0xcc894043U, 0x7767d99eU, 0xbdb0e842U, 0x8807898bU, 0x38e7195bU, 0xdb79c8eeU, + 0x47a17c0aU, 0xe97c420fU, 0xc9f8841eU, 0x00000000U, 0x83098086U, 0x48322bedU, + 0xac1e1170U, 0x4e6c5a72U, 0xfbfd0effU, 0x560f8538U, 0x1e3daed5U, 0x27362d39U, + 0x640a0fd9U, 0x21685ca6U, 0xd19b5b54U, 0x3a24362eU, 0xb10c0a67U, 0x0f9357e7U, + 0xd2b4ee96U, 0x9e1b9b91U, 0x4f80c0c5U, 0xa261dc20U, 0x695a774bU, 0x161c121aU, + 0x0ae293baU, 0xe5c0a02aU, 0x433c22e0U, 0x1d121b17U, 0x0b0e090dU, 0xadf28bc7U, + 0xb92db6a8U, 0xc8141ea9U, 0x8557f119U, 0x4caf7507U, 0xbbee99ddU, 0xfda37f60U, + 0x9ff70126U, 0xbc5c72f5U, 0xc544663bU, 0x345bfb7eU, 0x768b4329U, 0xdccb23c6U, + 0x68b6edfcU, 0x63b8e4f1U, 0xcad731dcU, 0x10426385U, 0x40139722U, 0x2084c611U, + 0x7d854a24U, 0xf8d2bb3dU, 0x11aef932U, 0x6dc729a1U, 0x4b1d9e2fU, 0xf3dcb230U, + 0xec0d8652U, 0xd077c1e3U, 0x6c2bb316U, 0x99a970b9U, 0xfa119448U, 0x2247e964U, + 0xc4a8fc8cU, 0x1aa0f03fU, 0xd8567d2cU, 0xef223390U, 0xc787494eU, 0xc1d938d1U, + 0xfe8ccaa2U, 0x3698d40bU, 0xcfa6f581U, 0x28a57adeU, 0x26dab78eU, 0xa43fadbfU, + 0xe42c3a9dU, 0x0d507892U, 0x9b6a5fccU, 0x62547e46U, 0xc2f68d13U, 0xe890d8b8U, + 0x5e2e39f7U, 0xf582c3afU, 0xbe9f5d80U, 0x7c69d093U, 0xa96fd52dU, 0xb3cf2512U, + 0x3bc8ac99U, 0xa710187dU, 0x6ee89c63U, 0x7bdb3bbbU, 0x09cd2678U, 0xf46e5918U, + 0x01ec9ab7U, 0xa8834f9aU, 0x65e6956eU, 0x7eaaffe6U, 0x0821bccfU, 0xe6ef15e8U, + 0xd9bae79bU, 0xce4a6f36U, 0xd4ea9f09U, 0xd629b07cU, 0xaf31a4b2U, 0x312a3f23U, + 0x30c6a594U, 0xc035a266U, 0x37744ebcU, 0xa6fc82caU, 0xb0e090d0U, 0x1533a7d8U, + 0x4af10498U, 0xf741ecdaU, 0x0e7fcd50U, 0x2f1791f6U, 0x8d764dd6U, 0x4d43efb0U, + 0x54ccaa4dU, 0xdfe49604U, 0xe39ed1b5U, 0x1b4c6a88U, 0xb8c12c1fU, 0x7f466551U, + 0x049d5eeaU, 0x5d018c35U, 0x73fa8774U, 0x2efb0b41U, 0x5ab3671dU, 0x5292dbd2U, + 0x33e91056U, 0x136dd647U, 0x8c9ad761U, 0x7a37a10cU, 0x8e59f814U, 0x89eb133cU, + 0xeecea927U, 0x35b761c9U, 0xede11ce5U, 0x3c7a47b1U, 0x599cd2dfU, 0x3f55f273U, + 0x791814ceU, 0xbf73c737U, 0xea53f7cdU, 0x5b5ffdaaU, 0x14df3d6fU, 0x867844dbU, + 0x81caaff3U, 0x3eb968c4U, 0x2c382434U, 0x5fc2a340U, 0x72161dc3U, 0x0cbce225U, + 0x8b283c49U, 0x41ff0d95U, 0x7139a801U, 0xde080cb3U, 0x9cd8b4e4U, 0x906456c1U, + 0x617bcb84U, 0x70d532b6U, 0x74486c5cU, 0x42d0b857U, }; static const u32 Td2[256] = { - 0xa75051f4U, 0x65537e41U, 0xa4c31a17U, 0x5e963a27U, - 0x6bcb3babU, 0x45f11f9dU, 0x58abacfaU, 0x03934be3U, - 0xfa552030U, 0x6df6ad76U, 0x769188ccU, 0x4c25f502U, - 0xd7fc4fe5U, 0xcbd7c52aU, 0x44802635U, 0xa38fb562U, - 0x5a49deb1U, 0x1b6725baU, 0x0e9845eaU, 0xc0e15dfeU, - 0x7502c32fU, 0xf012814cU, 0x97a38d46U, 0xf9c66bd3U, - 0x5fe7038fU, 0x9c951592U, 0x7aebbf6dU, 0x59da9552U, - 0x832dd4beU, 0x21d35874U, 0x692949e0U, 0xc8448ec9U, - 0x896a75c2U, 0x7978f48eU, 0x3e6b9958U, 0x71dd27b9U, - 0x4fb6bee1U, 0xad17f088U, 0xac66c920U, 0x3ab47dceU, - 0x4a1863dfU, 0x3182e51aU, 0x33609751U, 0x7f456253U, - 0x77e0b164U, 0xae84bb6bU, 0xa01cfe81U, 0x2b94f908U, - 0x68587048U, 0xfd198f45U, 0x6c8794deU, 0xf8b7527bU, - 0xd323ab73U, 0x02e2724bU, 0x8f57e31fU, 0xab2a6655U, - 0x2807b2ebU, 0xc2032fb5U, 0x7b9a86c5U, 0x08a5d337U, - 0x87f23028U, 0xa5b223bfU, 0x6aba0203U, 0x825ced16U, - 0x1c2b8acfU, 0xb492a779U, 0xf2f0f307U, 0xe2a14e69U, - 0xf4cd65daU, 0xbed50605U, 0x621fd134U, 0xfe8ac4a6U, - 0x539d342eU, 0x55a0a2f3U, 0xe132058aU, 0xeb75a4f6U, - 0xec390b83U, 0xefaa4060U, 0x9f065e71U, 0x1051bd6eU, + 0xa75051f4U, 0x65537e41U, 0xa4c31a17U, 0x5e963a27U, 0x6bcb3babU, 0x45f11f9dU, + 0x58abacfaU, 0x03934be3U, 0xfa552030U, 0x6df6ad76U, 0x769188ccU, 0x4c25f502U, + 0xd7fc4fe5U, 0xcbd7c52aU, 0x44802635U, 0xa38fb562U, 0x5a49deb1U, 0x1b6725baU, + 0x0e9845eaU, 0xc0e15dfeU, 0x7502c32fU, 0xf012814cU, 0x97a38d46U, 0xf9c66bd3U, + 0x5fe7038fU, 0x9c951592U, 0x7aebbf6dU, 0x59da9552U, 0x832dd4beU, 0x21d35874U, + 0x692949e0U, 0xc8448ec9U, 0x896a75c2U, 0x7978f48eU, 0x3e6b9958U, 0x71dd27b9U, + 0x4fb6bee1U, 0xad17f088U, 0xac66c920U, 0x3ab47dceU, 0x4a1863dfU, 0x3182e51aU, + 0x33609751U, 0x7f456253U, 0x77e0b164U, 0xae84bb6bU, 0xa01cfe81U, 0x2b94f908U, + 0x68587048U, 0xfd198f45U, 0x6c8794deU, 0xf8b7527bU, 0xd323ab73U, 0x02e2724bU, + 0x8f57e31fU, 0xab2a6655U, 0x2807b2ebU, 0xc2032fb5U, 0x7b9a86c5U, 0x08a5d337U, + 0x87f23028U, 0xa5b223bfU, 0x6aba0203U, 0x825ced16U, 0x1c2b8acfU, 0xb492a779U, + 0xf2f0f307U, 0xe2a14e69U, 0xf4cd65daU, 0xbed50605U, 0x621fd134U, 0xfe8ac4a6U, + 0x539d342eU, 0x55a0a2f3U, 0xe132058aU, 0xeb75a4f6U, 0xec390b83U, 0xefaa4060U, + 0x9f065e71U, 0x1051bd6eU, - 0x8af93e21U, 0x063d96ddU, 0x05aedd3eU, 0xbd464de6U, - 0x8db59154U, 0x5d0571c4U, 0xd46f0406U, 0x15ff6050U, - 0xfb241998U, 0xe997d6bdU, 0x43cc8940U, 0x9e7767d9U, - 0x42bdb0e8U, 0x8b880789U, 0x5b38e719U, 0xeedb79c8U, - 0x0a47a17cU, 0x0fe97c42U, 0x1ec9f884U, 0x00000000U, - 0x86830980U, 0xed48322bU, 0x70ac1e11U, 0x724e6c5aU, - 0xfffbfd0eU, 0x38560f85U, 0xd51e3daeU, 0x3927362dU, - 0xd9640a0fU, 0xa621685cU, 0x54d19b5bU, 0x2e3a2436U, - 0x67b10c0aU, 0xe70f9357U, 0x96d2b4eeU, 0x919e1b9bU, - 0xc54f80c0U, 0x20a261dcU, 0x4b695a77U, 0x1a161c12U, - 0xba0ae293U, 0x2ae5c0a0U, 0xe0433c22U, 0x171d121bU, - 0x0d0b0e09U, 0xc7adf28bU, 0xa8b92db6U, 0xa9c8141eU, - 0x198557f1U, 0x074caf75U, 0xddbbee99U, 0x60fda37fU, - 0x269ff701U, 0xf5bc5c72U, 0x3bc54466U, 0x7e345bfbU, - 0x29768b43U, 0xc6dccb23U, 0xfc68b6edU, 0xf163b8e4U, - 0xdccad731U, 0x85104263U, 0x22401397U, 0x112084c6U, - 0x247d854aU, 0x3df8d2bbU, 0x3211aef9U, 0xa16dc729U, - 0x2f4b1d9eU, 0x30f3dcb2U, 0x52ec0d86U, 0xe3d077c1U, - 0x166c2bb3U, 0xb999a970U, 0x48fa1194U, 0x642247e9U, - 0x8cc4a8fcU, 0x3f1aa0f0U, 0x2cd8567dU, 0x90ef2233U, - 0x4ec78749U, 0xd1c1d938U, 0xa2fe8ccaU, 0x0b3698d4U, - 0x81cfa6f5U, 0xde28a57aU, 0x8e26dab7U, 0xbfa43fadU, - 0x9de42c3aU, 0x920d5078U, 0xcc9b6a5fU, 0x4662547eU, - 0x13c2f68dU, 0xb8e890d8U, 0xf75e2e39U, 0xaff582c3U, - 0x80be9f5dU, 0x937c69d0U, 0x2da96fd5U, 0x12b3cf25U, - 0x993bc8acU, 0x7da71018U, 0x636ee89cU, 0xbb7bdb3bU, - 0x7809cd26U, 0x18f46e59U, 0xb701ec9aU, 0x9aa8834fU, - 0x6e65e695U, 0xe67eaaffU, 0xcf0821bcU, 0xe8e6ef15U, - 0x9bd9bae7U, 0x36ce4a6fU, 0x09d4ea9fU, 0x7cd629b0U, - 0xb2af31a4U, 0x23312a3fU, 0x9430c6a5U, 0x66c035a2U, - 0xbc37744eU, 0xcaa6fc82U, 0xd0b0e090U, 0xd81533a7U, - 0x984af104U, 0xdaf741ecU, 0x500e7fcdU, 0xf62f1791U, - 0xd68d764dU, 0xb04d43efU, 0x4d54ccaaU, 0x04dfe496U, - 0xb5e39ed1U, 0x881b4c6aU, 0x1fb8c12cU, 0x517f4665U, - 0xea049d5eU, 0x355d018cU, 0x7473fa87U, 0x412efb0bU, - 0x1d5ab367U, 0xd25292dbU, 0x5633e910U, 0x47136dd6U, - 0x618c9ad7U, 0x0c7a37a1U, 0x148e59f8U, 0x3c89eb13U, - 0x27eecea9U, 0xc935b761U, 0xe5ede11cU, 0xb13c7a47U, - 0xdf599cd2U, 0x733f55f2U, 0xce791814U, 0x37bf73c7U, - 0xcdea53f7U, 0xaa5b5ffdU, 0x6f14df3dU, 0xdb867844U, - 0xf381caafU, 0xc43eb968U, 0x342c3824U, 0x405fc2a3U, - 0xc372161dU, 0x250cbce2U, 0x498b283cU, 0x9541ff0dU, - 0x017139a8U, 0xb3de080cU, 0xe49cd8b4U, 0xc1906456U, - 0x84617bcbU, 0xb670d532U, 0x5c74486cU, 0x5742d0b8U, + 0x8af93e21U, 0x063d96ddU, 0x05aedd3eU, 0xbd464de6U, 0x8db59154U, 0x5d0571c4U, + 0xd46f0406U, 0x15ff6050U, 0xfb241998U, 0xe997d6bdU, 0x43cc8940U, 0x9e7767d9U, + 0x42bdb0e8U, 0x8b880789U, 0x5b38e719U, 0xeedb79c8U, 0x0a47a17cU, 0x0fe97c42U, + 0x1ec9f884U, 0x00000000U, 0x86830980U, 0xed48322bU, 0x70ac1e11U, 0x724e6c5aU, + 0xfffbfd0eU, 0x38560f85U, 0xd51e3daeU, 0x3927362dU, 0xd9640a0fU, 0xa621685cU, + 0x54d19b5bU, 0x2e3a2436U, 0x67b10c0aU, 0xe70f9357U, 0x96d2b4eeU, 0x919e1b9bU, + 0xc54f80c0U, 0x20a261dcU, 0x4b695a77U, 0x1a161c12U, 0xba0ae293U, 0x2ae5c0a0U, + 0xe0433c22U, 0x171d121bU, 0x0d0b0e09U, 0xc7adf28bU, 0xa8b92db6U, 0xa9c8141eU, + 0x198557f1U, 0x074caf75U, 0xddbbee99U, 0x60fda37fU, 0x269ff701U, 0xf5bc5c72U, + 0x3bc54466U, 0x7e345bfbU, 0x29768b43U, 0xc6dccb23U, 0xfc68b6edU, 0xf163b8e4U, + 0xdccad731U, 0x85104263U, 0x22401397U, 0x112084c6U, 0x247d854aU, 0x3df8d2bbU, + 0x3211aef9U, 0xa16dc729U, 0x2f4b1d9eU, 0x30f3dcb2U, 0x52ec0d86U, 0xe3d077c1U, + 0x166c2bb3U, 0xb999a970U, 0x48fa1194U, 0x642247e9U, 0x8cc4a8fcU, 0x3f1aa0f0U, + 0x2cd8567dU, 0x90ef2233U, 0x4ec78749U, 0xd1c1d938U, 0xa2fe8ccaU, 0x0b3698d4U, + 0x81cfa6f5U, 0xde28a57aU, 0x8e26dab7U, 0xbfa43fadU, 0x9de42c3aU, 0x920d5078U, + 0xcc9b6a5fU, 0x4662547eU, 0x13c2f68dU, 0xb8e890d8U, 0xf75e2e39U, 0xaff582c3U, + 0x80be9f5dU, 0x937c69d0U, 0x2da96fd5U, 0x12b3cf25U, 0x993bc8acU, 0x7da71018U, + 0x636ee89cU, 0xbb7bdb3bU, 0x7809cd26U, 0x18f46e59U, 0xb701ec9aU, 0x9aa8834fU, + 0x6e65e695U, 0xe67eaaffU, 0xcf0821bcU, 0xe8e6ef15U, 0x9bd9bae7U, 0x36ce4a6fU, + 0x09d4ea9fU, 0x7cd629b0U, 0xb2af31a4U, 0x23312a3fU, 0x9430c6a5U, 0x66c035a2U, + 0xbc37744eU, 0xcaa6fc82U, 0xd0b0e090U, 0xd81533a7U, 0x984af104U, 0xdaf741ecU, + 0x500e7fcdU, 0xf62f1791U, 0xd68d764dU, 0xb04d43efU, 0x4d54ccaaU, 0x04dfe496U, + 0xb5e39ed1U, 0x881b4c6aU, 0x1fb8c12cU, 0x517f4665U, 0xea049d5eU, 0x355d018cU, + 0x7473fa87U, 0x412efb0bU, 0x1d5ab367U, 0xd25292dbU, 0x5633e910U, 0x47136dd6U, + 0x618c9ad7U, 0x0c7a37a1U, 0x148e59f8U, 0x3c89eb13U, 0x27eecea9U, 0xc935b761U, + 0xe5ede11cU, 0xb13c7a47U, 0xdf599cd2U, 0x733f55f2U, 0xce791814U, 0x37bf73c7U, + 0xcdea53f7U, 0xaa5b5ffdU, 0x6f14df3dU, 0xdb867844U, 0xf381caafU, 0xc43eb968U, + 0x342c3824U, 0x405fc2a3U, 0xc372161dU, 0x250cbce2U, 0x498b283cU, 0x9541ff0dU, + 0x017139a8U, 0xb3de080cU, 0xe49cd8b4U, 0xc1906456U, 0x84617bcbU, 0xb670d532U, + 0x5c74486cU, 0x5742d0b8U, }; static const u32 Td3[256] = { - 0xf4a75051U, 0x4165537eU, 0x17a4c31aU, 0x275e963aU, - 0xab6bcb3bU, 0x9d45f11fU, 0xfa58abacU, 0xe303934bU, - 0x30fa5520U, 0x766df6adU, 0xcc769188U, 0x024c25f5U, - 0xe5d7fc4fU, 0x2acbd7c5U, 0x35448026U, 0x62a38fb5U, - 0xb15a49deU, 0xba1b6725U, 0xea0e9845U, 0xfec0e15dU, - 0x2f7502c3U, 0x4cf01281U, 0x4697a38dU, 0xd3f9c66bU, - 0x8f5fe703U, 0x929c9515U, 0x6d7aebbfU, 0x5259da95U, - 0xbe832dd4U, 0x7421d358U, 0xe0692949U, 0xc9c8448eU, - 0xc2896a75U, 0x8e7978f4U, 0x583e6b99U, 0xb971dd27U, - 0xe14fb6beU, 0x88ad17f0U, 0x20ac66c9U, 0xce3ab47dU, - 0xdf4a1863U, 0x1a3182e5U, 0x51336097U, 0x537f4562U, - 0x6477e0b1U, 0x6bae84bbU, 0x81a01cfeU, 0x082b94f9U, - 0x48685870U, 0x45fd198fU, 0xde6c8794U, 0x7bf8b752U, - 0x73d323abU, 0x4b02e272U, 0x1f8f57e3U, 0x55ab2a66U, - 0xeb2807b2U, 0xb5c2032fU, 0xc57b9a86U, 0x3708a5d3U, - 0x2887f230U, 0xbfa5b223U, 0x036aba02U, 0x16825cedU, - 0xcf1c2b8aU, 0x79b492a7U, 0x07f2f0f3U, 0x69e2a14eU, - 0xdaf4cd65U, 0x05bed506U, 0x34621fd1U, 0xa6fe8ac4U, - 0x2e539d34U, 0xf355a0a2U, 0x8ae13205U, 0xf6eb75a4U, - 0x83ec390bU, 0x60efaa40U, 0x719f065eU, 0x6e1051bdU, - 0x218af93eU, 0xdd063d96U, 0x3e05aeddU, 0xe6bd464dU, - 0x548db591U, 0xc45d0571U, 0x06d46f04U, 0x5015ff60U, - 0x98fb2419U, 0xbde997d6U, 0x4043cc89U, 0xd99e7767U, - 0xe842bdb0U, 0x898b8807U, 0x195b38e7U, 0xc8eedb79U, - 0x7c0a47a1U, 0x420fe97cU, 0x841ec9f8U, 0x00000000U, - 0x80868309U, 0x2bed4832U, 0x1170ac1eU, 0x5a724e6cU, - 0x0efffbfdU, 0x8538560fU, 0xaed51e3dU, 0x2d392736U, - 0x0fd9640aU, 0x5ca62168U, 0x5b54d19bU, 0x362e3a24U, - 0x0a67b10cU, 0x57e70f93U, 0xee96d2b4U, 0x9b919e1bU, - 0xc0c54f80U, 0xdc20a261U, 0x774b695aU, 0x121a161cU, - 0x93ba0ae2U, 0xa02ae5c0U, 0x22e0433cU, 0x1b171d12U, - 0x090d0b0eU, 0x8bc7adf2U, 0xb6a8b92dU, 0x1ea9c814U, - 0xf1198557U, 0x75074cafU, 0x99ddbbeeU, 0x7f60fda3U, - 0x01269ff7U, 0x72f5bc5cU, 0x663bc544U, 0xfb7e345bU, - 0x4329768bU, 0x23c6dccbU, 0xedfc68b6U, 0xe4f163b8U, - 0x31dccad7U, 0x63851042U, 0x97224013U, 0xc6112084U, - 0x4a247d85U, 0xbb3df8d2U, 0xf93211aeU, 0x29a16dc7U, - 0x9e2f4b1dU, 0xb230f3dcU, 0x8652ec0dU, 0xc1e3d077U, - 0xb3166c2bU, 0x70b999a9U, 0x9448fa11U, 0xe9642247U, - 0xfc8cc4a8U, 0xf03f1aa0U, 0x7d2cd856U, 0x3390ef22U, - 0x494ec787U, 0x38d1c1d9U, 0xcaa2fe8cU, 0xd40b3698U, - 0xf581cfa6U, 0x7ade28a5U, 0xb78e26daU, 0xadbfa43fU, - 0x3a9de42cU, 0x78920d50U, 0x5fcc9b6aU, 0x7e466254U, - 0x8d13c2f6U, 0xd8b8e890U, 0x39f75e2eU, 0xc3aff582U, - 0x5d80be9fU, 0xd0937c69U, 0xd52da96fU, 0x2512b3cfU, - 0xac993bc8U, 0x187da710U, 0x9c636ee8U, 0x3bbb7bdbU, - 0x267809cdU, 0x5918f46eU, 0x9ab701ecU, 0x4f9aa883U, - 0x956e65e6U, 0xffe67eaaU, 0xbccf0821U, 0x15e8e6efU, - 0xe79bd9baU, 0x6f36ce4aU, 0x9f09d4eaU, 0xb07cd629U, - 0xa4b2af31U, 0x3f23312aU, 0xa59430c6U, 0xa266c035U, - 0x4ebc3774U, 0x82caa6fcU, 0x90d0b0e0U, 0xa7d81533U, - 0x04984af1U, 0xecdaf741U, 0xcd500e7fU, 0x91f62f17U, - 0x4dd68d76U, 0xefb04d43U, 0xaa4d54ccU, 0x9604dfe4U, - 0xd1b5e39eU, 0x6a881b4cU, 0x2c1fb8c1U, 0x65517f46U, - 0x5eea049dU, 0x8c355d01U, 0x877473faU, 0x0b412efbU, - 0x671d5ab3U, 0xdbd25292U, 0x105633e9U, 0xd647136dU, - 0xd7618c9aU, 0xa10c7a37U, 0xf8148e59U, 0x133c89ebU, - 0xa927eeceU, 0x61c935b7U, 0x1ce5ede1U, 0x47b13c7aU, - 0xd2df599cU, 0xf2733f55U, 0x14ce7918U, 0xc737bf73U, - 0xf7cdea53U, 0xfdaa5b5fU, 0x3d6f14dfU, 0x44db8678U, - 0xaff381caU, 0x68c43eb9U, 0x24342c38U, 0xa3405fc2U, - 0x1dc37216U, 0xe2250cbcU, 0x3c498b28U, 0x0d9541ffU, - 0xa8017139U, 0x0cb3de08U, 0xb4e49cd8U, 0x56c19064U, - 0xcb84617bU, 0x32b670d5U, 0x6c5c7448U, 0xb85742d0U, + 0xf4a75051U, 0x4165537eU, 0x17a4c31aU, 0x275e963aU, 0xab6bcb3bU, 0x9d45f11fU, + 0xfa58abacU, 0xe303934bU, 0x30fa5520U, 0x766df6adU, 0xcc769188U, 0x024c25f5U, + 0xe5d7fc4fU, 0x2acbd7c5U, 0x35448026U, 0x62a38fb5U, 0xb15a49deU, 0xba1b6725U, + 0xea0e9845U, 0xfec0e15dU, 0x2f7502c3U, 0x4cf01281U, 0x4697a38dU, 0xd3f9c66bU, + 0x8f5fe703U, 0x929c9515U, 0x6d7aebbfU, 0x5259da95U, 0xbe832dd4U, 0x7421d358U, + 0xe0692949U, 0xc9c8448eU, 0xc2896a75U, 0x8e7978f4U, 0x583e6b99U, 0xb971dd27U, + 0xe14fb6beU, 0x88ad17f0U, 0x20ac66c9U, 0xce3ab47dU, 0xdf4a1863U, 0x1a3182e5U, + 0x51336097U, 0x537f4562U, 0x6477e0b1U, 0x6bae84bbU, 0x81a01cfeU, 0x082b94f9U, + 0x48685870U, 0x45fd198fU, 0xde6c8794U, 0x7bf8b752U, 0x73d323abU, 0x4b02e272U, + 0x1f8f57e3U, 0x55ab2a66U, 0xeb2807b2U, 0xb5c2032fU, 0xc57b9a86U, 0x3708a5d3U, + 0x2887f230U, 0xbfa5b223U, 0x036aba02U, 0x16825cedU, 0xcf1c2b8aU, 0x79b492a7U, + 0x07f2f0f3U, 0x69e2a14eU, 0xdaf4cd65U, 0x05bed506U, 0x34621fd1U, 0xa6fe8ac4U, + 0x2e539d34U, 0xf355a0a2U, 0x8ae13205U, 0xf6eb75a4U, 0x83ec390bU, 0x60efaa40U, + 0x719f065eU, 0x6e1051bdU, 0x218af93eU, 0xdd063d96U, 0x3e05aeddU, 0xe6bd464dU, + 0x548db591U, 0xc45d0571U, 0x06d46f04U, 0x5015ff60U, 0x98fb2419U, 0xbde997d6U, + 0x4043cc89U, 0xd99e7767U, 0xe842bdb0U, 0x898b8807U, 0x195b38e7U, 0xc8eedb79U, + 0x7c0a47a1U, 0x420fe97cU, 0x841ec9f8U, 0x00000000U, 0x80868309U, 0x2bed4832U, + 0x1170ac1eU, 0x5a724e6cU, 0x0efffbfdU, 0x8538560fU, 0xaed51e3dU, 0x2d392736U, + 0x0fd9640aU, 0x5ca62168U, 0x5b54d19bU, 0x362e3a24U, 0x0a67b10cU, 0x57e70f93U, + 0xee96d2b4U, 0x9b919e1bU, 0xc0c54f80U, 0xdc20a261U, 0x774b695aU, 0x121a161cU, + 0x93ba0ae2U, 0xa02ae5c0U, 0x22e0433cU, 0x1b171d12U, 0x090d0b0eU, 0x8bc7adf2U, + 0xb6a8b92dU, 0x1ea9c814U, 0xf1198557U, 0x75074cafU, 0x99ddbbeeU, 0x7f60fda3U, + 0x01269ff7U, 0x72f5bc5cU, 0x663bc544U, 0xfb7e345bU, 0x4329768bU, 0x23c6dccbU, + 0xedfc68b6U, 0xe4f163b8U, 0x31dccad7U, 0x63851042U, 0x97224013U, 0xc6112084U, + 0x4a247d85U, 0xbb3df8d2U, 0xf93211aeU, 0x29a16dc7U, 0x9e2f4b1dU, 0xb230f3dcU, + 0x8652ec0dU, 0xc1e3d077U, 0xb3166c2bU, 0x70b999a9U, 0x9448fa11U, 0xe9642247U, + 0xfc8cc4a8U, 0xf03f1aa0U, 0x7d2cd856U, 0x3390ef22U, 0x494ec787U, 0x38d1c1d9U, + 0xcaa2fe8cU, 0xd40b3698U, 0xf581cfa6U, 0x7ade28a5U, 0xb78e26daU, 0xadbfa43fU, + 0x3a9de42cU, 0x78920d50U, 0x5fcc9b6aU, 0x7e466254U, 0x8d13c2f6U, 0xd8b8e890U, + 0x39f75e2eU, 0xc3aff582U, 0x5d80be9fU, 0xd0937c69U, 0xd52da96fU, 0x2512b3cfU, + 0xac993bc8U, 0x187da710U, 0x9c636ee8U, 0x3bbb7bdbU, 0x267809cdU, 0x5918f46eU, + 0x9ab701ecU, 0x4f9aa883U, 0x956e65e6U, 0xffe67eaaU, 0xbccf0821U, 0x15e8e6efU, + 0xe79bd9baU, 0x6f36ce4aU, 0x9f09d4eaU, 0xb07cd629U, 0xa4b2af31U, 0x3f23312aU, + 0xa59430c6U, 0xa266c035U, 0x4ebc3774U, 0x82caa6fcU, 0x90d0b0e0U, 0xa7d81533U, + 0x04984af1U, 0xecdaf741U, 0xcd500e7fU, 0x91f62f17U, 0x4dd68d76U, 0xefb04d43U, + 0xaa4d54ccU, 0x9604dfe4U, 0xd1b5e39eU, 0x6a881b4cU, 0x2c1fb8c1U, 0x65517f46U, + 0x5eea049dU, 0x8c355d01U, 0x877473faU, 0x0b412efbU, 0x671d5ab3U, 0xdbd25292U, + 0x105633e9U, 0xd647136dU, 0xd7618c9aU, 0xa10c7a37U, 0xf8148e59U, 0x133c89ebU, + 0xa927eeceU, 0x61c935b7U, 0x1ce5ede1U, 0x47b13c7aU, 0xd2df599cU, 0xf2733f55U, + 0x14ce7918U, 0xc737bf73U, 0xf7cdea53U, 0xfdaa5b5fU, 0x3d6f14dfU, 0x44db8678U, + 0xaff381caU, 0x68c43eb9U, 0x24342c38U, 0xa3405fc2U, 0x1dc37216U, 0xe2250cbcU, + 0x3c498b28U, 0x0d9541ffU, 0xa8017139U, 0x0cb3de08U, 0xb4e49cd8U, 0x56c19064U, + 0xcb84617bU, 0x32b670d5U, 0x6c5c7448U, 0xb85742d0U, }; static const u32 Td4[256] = { - 0x52525252U, 0x09090909U, 0x6a6a6a6aU, 0xd5d5d5d5U, - 0x30303030U, 0x36363636U, 0xa5a5a5a5U, 0x38383838U, - 0xbfbfbfbfU, 0x40404040U, 0xa3a3a3a3U, 0x9e9e9e9eU, - 0x81818181U, 0xf3f3f3f3U, 0xd7d7d7d7U, 0xfbfbfbfbU, - 0x7c7c7c7cU, 0xe3e3e3e3U, 0x39393939U, 0x82828282U, - 0x9b9b9b9bU, 0x2f2f2f2fU, 0xffffffffU, 0x87878787U, - 0x34343434U, 0x8e8e8e8eU, 0x43434343U, 0x44444444U, - 0xc4c4c4c4U, 0xdedededeU, 0xe9e9e9e9U, 0xcbcbcbcbU, - 0x54545454U, 0x7b7b7b7bU, 0x94949494U, 0x32323232U, - 0xa6a6a6a6U, 0xc2c2c2c2U, 0x23232323U, 0x3d3d3d3dU, - 0xeeeeeeeeU, 0x4c4c4c4cU, 0x95959595U, 0x0b0b0b0bU, - 0x42424242U, 0xfafafafaU, 0xc3c3c3c3U, 0x4e4e4e4eU, - 0x08080808U, 0x2e2e2e2eU, 0xa1a1a1a1U, 0x66666666U, - 0x28282828U, 0xd9d9d9d9U, 0x24242424U, 0xb2b2b2b2U, - 0x76767676U, 0x5b5b5b5bU, 0xa2a2a2a2U, 0x49494949U, - 0x6d6d6d6dU, 0x8b8b8b8bU, 0xd1d1d1d1U, 0x25252525U, - 0x72727272U, 0xf8f8f8f8U, 0xf6f6f6f6U, 0x64646464U, - 0x86868686U, 0x68686868U, 0x98989898U, 0x16161616U, - 0xd4d4d4d4U, 0xa4a4a4a4U, 0x5c5c5c5cU, 0xccccccccU, - 0x5d5d5d5dU, 0x65656565U, 0xb6b6b6b6U, 0x92929292U, - 0x6c6c6c6cU, 0x70707070U, 0x48484848U, 0x50505050U, - 0xfdfdfdfdU, 0xededededU, 0xb9b9b9b9U, 0xdadadadaU, - 0x5e5e5e5eU, 0x15151515U, 0x46464646U, 0x57575757U, - 0xa7a7a7a7U, 0x8d8d8d8dU, 0x9d9d9d9dU, 0x84848484U, - 0x90909090U, 0xd8d8d8d8U, 0xababababU, 0x00000000U, - 0x8c8c8c8cU, 0xbcbcbcbcU, 0xd3d3d3d3U, 0x0a0a0a0aU, - 0xf7f7f7f7U, 0xe4e4e4e4U, 0x58585858U, 0x05050505U, - 0xb8b8b8b8U, 0xb3b3b3b3U, 0x45454545U, 0x06060606U, - 0xd0d0d0d0U, 0x2c2c2c2cU, 0x1e1e1e1eU, 0x8f8f8f8fU, - 0xcacacacaU, 0x3f3f3f3fU, 0x0f0f0f0fU, 0x02020202U, - 0xc1c1c1c1U, 0xafafafafU, 0xbdbdbdbdU, 0x03030303U, - 0x01010101U, 0x13131313U, 0x8a8a8a8aU, 0x6b6b6b6bU, - 0x3a3a3a3aU, 0x91919191U, 0x11111111U, 0x41414141U, - 0x4f4f4f4fU, 0x67676767U, 0xdcdcdcdcU, 0xeaeaeaeaU, - 0x97979797U, 0xf2f2f2f2U, 0xcfcfcfcfU, 0xcecececeU, - 0xf0f0f0f0U, 0xb4b4b4b4U, 0xe6e6e6e6U, 0x73737373U, - 0x96969696U, 0xacacacacU, 0x74747474U, 0x22222222U, - 0xe7e7e7e7U, 0xadadadadU, 0x35353535U, 0x85858585U, - 0xe2e2e2e2U, 0xf9f9f9f9U, 0x37373737U, 0xe8e8e8e8U, - 0x1c1c1c1cU, 0x75757575U, 0xdfdfdfdfU, 0x6e6e6e6eU, - 0x47474747U, 0xf1f1f1f1U, 0x1a1a1a1aU, 0x71717171U, - 0x1d1d1d1dU, 0x29292929U, 0xc5c5c5c5U, 0x89898989U, - 0x6f6f6f6fU, 0xb7b7b7b7U, 0x62626262U, 0x0e0e0e0eU, - 0xaaaaaaaaU, 0x18181818U, 0xbebebebeU, 0x1b1b1b1bU, - 0xfcfcfcfcU, 0x56565656U, 0x3e3e3e3eU, 0x4b4b4b4bU, - 0xc6c6c6c6U, 0xd2d2d2d2U, 0x79797979U, 0x20202020U, - 0x9a9a9a9aU, 0xdbdbdbdbU, 0xc0c0c0c0U, 0xfefefefeU, - 0x78787878U, 0xcdcdcdcdU, 0x5a5a5a5aU, 0xf4f4f4f4U, - 0x1f1f1f1fU, 0xddddddddU, 0xa8a8a8a8U, 0x33333333U, - 0x88888888U, 0x07070707U, 0xc7c7c7c7U, 0x31313131U, - 0xb1b1b1b1U, 0x12121212U, 0x10101010U, 0x59595959U, - 0x27272727U, 0x80808080U, 0xececececU, 0x5f5f5f5fU, - 0x60606060U, 0x51515151U, 0x7f7f7f7fU, 0xa9a9a9a9U, - 0x19191919U, 0xb5b5b5b5U, 0x4a4a4a4aU, 0x0d0d0d0dU, - 0x2d2d2d2dU, 0xe5e5e5e5U, 0x7a7a7a7aU, 0x9f9f9f9fU, - 0x93939393U, 0xc9c9c9c9U, 0x9c9c9c9cU, 0xefefefefU, - 0xa0a0a0a0U, 0xe0e0e0e0U, 0x3b3b3b3bU, 0x4d4d4d4dU, - 0xaeaeaeaeU, 0x2a2a2a2aU, 0xf5f5f5f5U, 0xb0b0b0b0U, - 0xc8c8c8c8U, 0xebebebebU, 0xbbbbbbbbU, 0x3c3c3c3cU, - 0x83838383U, 0x53535353U, 0x99999999U, 0x61616161U, - 0x17171717U, 0x2b2b2b2bU, 0x04040404U, 0x7e7e7e7eU, - 0xbabababaU, 0x77777777U, 0xd6d6d6d6U, 0x26262626U, - 0xe1e1e1e1U, 0x69696969U, 0x14141414U, 0x63636363U, - 0x55555555U, 0x21212121U, 0x0c0c0c0cU, 0x7d7d7d7dU, + 0x52525252U, 0x09090909U, 0x6a6a6a6aU, 0xd5d5d5d5U, 0x30303030U, 0x36363636U, + 0xa5a5a5a5U, 0x38383838U, 0xbfbfbfbfU, 0x40404040U, 0xa3a3a3a3U, 0x9e9e9e9eU, + 0x81818181U, 0xf3f3f3f3U, 0xd7d7d7d7U, 0xfbfbfbfbU, 0x7c7c7c7cU, 0xe3e3e3e3U, + 0x39393939U, 0x82828282U, 0x9b9b9b9bU, 0x2f2f2f2fU, 0xffffffffU, 0x87878787U, + 0x34343434U, 0x8e8e8e8eU, 0x43434343U, 0x44444444U, 0xc4c4c4c4U, 0xdedededeU, + 0xe9e9e9e9U, 0xcbcbcbcbU, 0x54545454U, 0x7b7b7b7bU, 0x94949494U, 0x32323232U, + 0xa6a6a6a6U, 0xc2c2c2c2U, 0x23232323U, 0x3d3d3d3dU, 0xeeeeeeeeU, 0x4c4c4c4cU, + 0x95959595U, 0x0b0b0b0bU, 0x42424242U, 0xfafafafaU, 0xc3c3c3c3U, 0x4e4e4e4eU, + 0x08080808U, 0x2e2e2e2eU, 0xa1a1a1a1U, 0x66666666U, 0x28282828U, 0xd9d9d9d9U, + 0x24242424U, 0xb2b2b2b2U, 0x76767676U, 0x5b5b5b5bU, 0xa2a2a2a2U, 0x49494949U, + 0x6d6d6d6dU, 0x8b8b8b8bU, 0xd1d1d1d1U, 0x25252525U, 0x72727272U, 0xf8f8f8f8U, + 0xf6f6f6f6U, 0x64646464U, 0x86868686U, 0x68686868U, 0x98989898U, 0x16161616U, + 0xd4d4d4d4U, 0xa4a4a4a4U, 0x5c5c5c5cU, 0xccccccccU, 0x5d5d5d5dU, 0x65656565U, + 0xb6b6b6b6U, 0x92929292U, 0x6c6c6c6cU, 0x70707070U, 0x48484848U, 0x50505050U, + 0xfdfdfdfdU, 0xededededU, 0xb9b9b9b9U, 0xdadadadaU, 0x5e5e5e5eU, 0x15151515U, + 0x46464646U, 0x57575757U, 0xa7a7a7a7U, 0x8d8d8d8dU, 0x9d9d9d9dU, 0x84848484U, + 0x90909090U, 0xd8d8d8d8U, 0xababababU, 0x00000000U, 0x8c8c8c8cU, 0xbcbcbcbcU, + 0xd3d3d3d3U, 0x0a0a0a0aU, 0xf7f7f7f7U, 0xe4e4e4e4U, 0x58585858U, 0x05050505U, + 0xb8b8b8b8U, 0xb3b3b3b3U, 0x45454545U, 0x06060606U, 0xd0d0d0d0U, 0x2c2c2c2cU, + 0x1e1e1e1eU, 0x8f8f8f8fU, 0xcacacacaU, 0x3f3f3f3fU, 0x0f0f0f0fU, 0x02020202U, + 0xc1c1c1c1U, 0xafafafafU, 0xbdbdbdbdU, 0x03030303U, 0x01010101U, 0x13131313U, + 0x8a8a8a8aU, 0x6b6b6b6bU, 0x3a3a3a3aU, 0x91919191U, 0x11111111U, 0x41414141U, + 0x4f4f4f4fU, 0x67676767U, 0xdcdcdcdcU, 0xeaeaeaeaU, 0x97979797U, 0xf2f2f2f2U, + 0xcfcfcfcfU, 0xcecececeU, 0xf0f0f0f0U, 0xb4b4b4b4U, 0xe6e6e6e6U, 0x73737373U, + 0x96969696U, 0xacacacacU, 0x74747474U, 0x22222222U, 0xe7e7e7e7U, 0xadadadadU, + 0x35353535U, 0x85858585U, 0xe2e2e2e2U, 0xf9f9f9f9U, 0x37373737U, 0xe8e8e8e8U, + 0x1c1c1c1cU, 0x75757575U, 0xdfdfdfdfU, 0x6e6e6e6eU, 0x47474747U, 0xf1f1f1f1U, + 0x1a1a1a1aU, 0x71717171U, 0x1d1d1d1dU, 0x29292929U, 0xc5c5c5c5U, 0x89898989U, + 0x6f6f6f6fU, 0xb7b7b7b7U, 0x62626262U, 0x0e0e0e0eU, 0xaaaaaaaaU, 0x18181818U, + 0xbebebebeU, 0x1b1b1b1bU, 0xfcfcfcfcU, 0x56565656U, 0x3e3e3e3eU, 0x4b4b4b4bU, + 0xc6c6c6c6U, 0xd2d2d2d2U, 0x79797979U, 0x20202020U, 0x9a9a9a9aU, 0xdbdbdbdbU, + 0xc0c0c0c0U, 0xfefefefeU, 0x78787878U, 0xcdcdcdcdU, 0x5a5a5a5aU, 0xf4f4f4f4U, + 0x1f1f1f1fU, 0xddddddddU, 0xa8a8a8a8U, 0x33333333U, 0x88888888U, 0x07070707U, + 0xc7c7c7c7U, 0x31313131U, 0xb1b1b1b1U, 0x12121212U, 0x10101010U, 0x59595959U, + 0x27272727U, 0x80808080U, 0xececececU, 0x5f5f5f5fU, 0x60606060U, 0x51515151U, + 0x7f7f7f7fU, 0xa9a9a9a9U, 0x19191919U, 0xb5b5b5b5U, 0x4a4a4a4aU, 0x0d0d0d0dU, + 0x2d2d2d2dU, 0xe5e5e5e5U, 0x7a7a7a7aU, 0x9f9f9f9fU, 0x93939393U, 0xc9c9c9c9U, + 0x9c9c9c9cU, 0xefefefefU, 0xa0a0a0a0U, 0xe0e0e0e0U, 0x3b3b3b3bU, 0x4d4d4d4dU, + 0xaeaeaeaeU, 0x2a2a2a2aU, 0xf5f5f5f5U, 0xb0b0b0b0U, 0xc8c8c8c8U, 0xebebebebU, + 0xbbbbbbbbU, 0x3c3c3c3cU, 0x83838383U, 0x53535353U, 0x99999999U, 0x61616161U, + 0x17171717U, 0x2b2b2b2bU, 0x04040404U, 0x7e7e7e7eU, 0xbabababaU, 0x77777777U, + 0xd6d6d6d6U, 0x26262626U, 0xe1e1e1e1U, 0x69696969U, 0x14141414U, 0x63636363U, + 0x55555555U, 0x21212121U, 0x0c0c0c0cU, 0x7d7d7d7dU, }; static const u32 rcon[] = { - 0x01000000, 0x02000000, 0x04000000, 0x08000000, - 0x10000000, 0x20000000, 0x40000000, 0x80000000, - 0x1B000000, 0x36000000, /* for 128-bit blocks, Rijndael never uses more than 10 rcon values */ + 0x01000000, 0x02000000, 0x04000000, 0x08000000, 0x10000000, 0x20000000, + 0x40000000, 0x80000000, 0x1B000000, 0x36000000, /* for 128-bit blocks, + Rijndael never uses more + than 10 rcon values */ }; #define SWAP(x) (_lrotl(x, 8) & 0x00ff00ff | _lrotr(x, 8) & 0xff00ff00) #ifdef _MSC_VER -#define GETU32(p) SWAP(*((u32 *)(p))) -#define PUTU32(ct, st) { *((u32 *)(ct)) = SWAP((st)); } +#define GETU32(p) SWAP(*((u32*)(p))) +#define PUTU32(ct, st) \ + { \ + *((u32*)(ct)) = SWAP((st)); \ + } #else -#define GETU32(pt) (((u32)(pt)[0] << 24) ^ ((u32)(pt)[1] << 16) ^ ((u32)(pt)[2] << 8) ^ ((u32)(pt)[3])) -#define PUTU32(ct, st) { (ct)[0] = (u8)((st) >> 24); (ct)[1] = (u8)((st) >> 16); (ct)[2] = (u8)((st) >> 8); (ct)[3] = (u8)(st); } +#define GETU32(pt) \ + (((u32)(pt)[0] << 24) ^ ((u32)(pt)[1] << 16) ^ ((u32)(pt)[2] << 8) ^ \ + ((u32)(pt)[3])) +#define PUTU32(ct, st) \ + { \ + (ct)[0] = (u8)((st) >> 24); \ + (ct)[1] = (u8)((st) >> 16); \ + (ct)[2] = (u8)((st) >> 8); \ + (ct)[3] = (u8)(st); \ + } #endif /** @@ -727,85 +530,80 @@ static const u32 rcon[] = { * * @return the number of rounds for the given cipher key size. */ -int rijndaelKeySetupEnc(u32 rk[/*4*(Nr + 1)*/], const u8 cipherKey[], int keyBits) { - int i = 0; - u32 temp; +int +rijndaelKeySetupEnc(u32 rk[/*4*(Nr + 1)*/], const u8 cipherKey[], int keyBits) +{ + int i = 0; + u32 temp; - rk[0] = GETU32(cipherKey ); - rk[1] = GETU32(cipherKey + 4); - rk[2] = GETU32(cipherKey + 8); - rk[3] = GETU32(cipherKey + 12); - if (keyBits == 128) { - for (;;) { - temp = rk[3]; - rk[4] = rk[0] ^ - (Te4[(temp >> 16) & 0xff] & 0xff000000) ^ - (Te4[(temp >> 8) & 0xff] & 0x00ff0000) ^ - (Te4[(temp ) & 0xff] & 0x0000ff00) ^ - (Te4[(temp >> 24) ] & 0x000000ff) ^ - rcon[i]; - rk[5] = rk[1] ^ rk[4]; - rk[6] = rk[2] ^ rk[5]; - rk[7] = rk[3] ^ rk[6]; - if (++i == 10) { - return 10; - } - rk += 4; - } - } - rk[4] = GETU32(cipherKey + 16); - rk[5] = GETU32(cipherKey + 20); - if (keyBits == 192) { - for (;;) { - temp = rk[ 5]; - rk[ 6] = rk[ 0] ^ - (Te4[(temp >> 16) & 0xff] & 0xff000000) ^ - (Te4[(temp >> 8) & 0xff] & 0x00ff0000) ^ - (Te4[(temp ) & 0xff] & 0x0000ff00) ^ - (Te4[(temp >> 24) ] & 0x000000ff) ^ - rcon[i]; - rk[ 7] = rk[ 1] ^ rk[ 6]; - rk[ 8] = rk[ 2] ^ rk[ 7]; - rk[ 9] = rk[ 3] ^ rk[ 8]; - if (++i == 8) { - return 12; - } - rk[10] = rk[ 4] ^ rk[ 9]; - rk[11] = rk[ 5] ^ rk[10]; - rk += 6; - } - } - rk[6] = GETU32(cipherKey + 24); - rk[7] = GETU32(cipherKey + 28); - if (keyBits == 256) { - for (;;) { - temp = rk[ 7]; - rk[ 8] = rk[ 0] ^ - (Te4[(temp >> 16) & 0xff] & 0xff000000) ^ - (Te4[(temp >> 8) & 0xff] & 0x00ff0000) ^ - (Te4[(temp ) & 0xff] & 0x0000ff00) ^ - (Te4[(temp >> 24) ] & 0x000000ff) ^ - rcon[i]; - rk[ 9] = rk[ 1] ^ rk[ 8]; - rk[10] = rk[ 2] ^ rk[ 9]; - rk[11] = rk[ 3] ^ rk[10]; - if (++i == 7) { - return 14; - } - temp = rk[11]; - rk[12] = rk[ 4] ^ - (Te4[(temp >> 24) ] & 0xff000000) ^ - (Te4[(temp >> 16) & 0xff] & 0x00ff0000) ^ - (Te4[(temp >> 8) & 0xff] & 0x0000ff00) ^ - (Te4[(temp ) & 0xff] & 0x000000ff); - rk[13] = rk[ 5] ^ rk[12]; - rk[14] = rk[ 6] ^ rk[13]; - rk[15] = rk[ 7] ^ rk[14]; + rk[0] = GETU32(cipherKey); + rk[1] = GETU32(cipherKey + 4); + rk[2] = GETU32(cipherKey + 8); + rk[3] = GETU32(cipherKey + 12); + if (keyBits == 128) { + for (;;) { + temp = rk[3]; + rk[4] = rk[0] ^ (Te4[(temp >> 16) & 0xff] & 0xff000000) ^ + (Te4[(temp >> 8) & 0xff] & 0x00ff0000) ^ + (Te4[(temp)&0xff] & 0x0000ff00) ^ + (Te4[(temp >> 24)] & 0x000000ff) ^ rcon[i]; + rk[5] = rk[1] ^ rk[4]; + rk[6] = rk[2] ^ rk[5]; + rk[7] = rk[3] ^ rk[6]; + if (++i == 10) { + return 10; + } + rk += 4; + } + } + rk[4] = GETU32(cipherKey + 16); + rk[5] = GETU32(cipherKey + 20); + if (keyBits == 192) { + for (;;) { + temp = rk[5]; + rk[6] = rk[0] ^ (Te4[(temp >> 16) & 0xff] & 0xff000000) ^ + (Te4[(temp >> 8) & 0xff] & 0x00ff0000) ^ + (Te4[(temp)&0xff] & 0x0000ff00) ^ + (Te4[(temp >> 24)] & 0x000000ff) ^ rcon[i]; + rk[7] = rk[1] ^ rk[6]; + rk[8] = rk[2] ^ rk[7]; + rk[9] = rk[3] ^ rk[8]; + if (++i == 8) { + return 12; + } + rk[10] = rk[4] ^ rk[9]; + rk[11] = rk[5] ^ rk[10]; + rk += 6; + } + } + rk[6] = GETU32(cipherKey + 24); + rk[7] = GETU32(cipherKey + 28); + if (keyBits == 256) { + for (;;) { + temp = rk[7]; + rk[8] = rk[0] ^ (Te4[(temp >> 16) & 0xff] & 0xff000000) ^ + (Te4[(temp >> 8) & 0xff] & 0x00ff0000) ^ + (Te4[(temp)&0xff] & 0x0000ff00) ^ + (Te4[(temp >> 24)] & 0x000000ff) ^ rcon[i]; + rk[9] = rk[1] ^ rk[8]; + rk[10] = rk[2] ^ rk[9]; + rk[11] = rk[3] ^ rk[10]; + if (++i == 7) { + return 14; + } + temp = rk[11]; + rk[12] = rk[4] ^ (Te4[(temp >> 24)] & 0xff000000) ^ + (Te4[(temp >> 16) & 0xff] & 0x00ff0000) ^ + (Te4[(temp >> 8) & 0xff] & 0x0000ff00) ^ + (Te4[(temp)&0xff] & 0x000000ff); + rk[13] = rk[5] ^ rk[12]; + rk[14] = rk[6] ^ rk[13]; + rk[15] = rk[7] ^ rk[14]; - rk += 8; - } - } - return 0; + rk += 8; + } + } + return 0; } /** @@ -813,590 +611,575 @@ int rijndaelKeySetupEnc(u32 rk[/*4*(Nr + 1)*/], const u8 cipherKey[], int keyBit * * @return the number of rounds for the given cipher key size. */ -int rijndaelKeySetupDec(u32 rk[/*4*(Nr + 1)*/], const u8 cipherKey[], int keyBits) { - int Nr, i, j; - u32 temp; +int +rijndaelKeySetupDec(u32 rk[/*4*(Nr + 1)*/], const u8 cipherKey[], int keyBits) +{ + int Nr, i, j; + u32 temp; - /* expand the cipher key: */ - Nr = rijndaelKeySetupEnc(rk, cipherKey, keyBits); - /* invert the order of the round keys: */ - for (i = 0, j = 4*Nr; i < j; i += 4, j -= 4) { - temp = rk[i ]; rk[i ] = rk[j ]; rk[j ] = temp; - temp = rk[i + 1]; rk[i + 1] = rk[j + 1]; rk[j + 1] = temp; - temp = rk[i + 2]; rk[i + 2] = rk[j + 2]; rk[j + 2] = temp; - temp = rk[i + 3]; rk[i + 3] = rk[j + 3]; rk[j + 3] = temp; - } - /* apply the inverse MixColumn transform to all round keys but the first and the last: */ - for (i = 1; i < Nr; i++) { - rk += 4; - rk[0] = - Td0[Te4[(rk[0] >> 24) ] & 0xff] ^ - Td1[Te4[(rk[0] >> 16) & 0xff] & 0xff] ^ - Td2[Te4[(rk[0] >> 8) & 0xff] & 0xff] ^ - Td3[Te4[(rk[0] ) & 0xff] & 0xff]; - rk[1] = - Td0[Te4[(rk[1] >> 24) ] & 0xff] ^ - Td1[Te4[(rk[1] >> 16) & 0xff] & 0xff] ^ - Td2[Te4[(rk[1] >> 8) & 0xff] & 0xff] ^ - Td3[Te4[(rk[1] ) & 0xff] & 0xff]; - rk[2] = - Td0[Te4[(rk[2] >> 24) ] & 0xff] ^ - Td1[Te4[(rk[2] >> 16) & 0xff] & 0xff] ^ - Td2[Te4[(rk[2] >> 8) & 0xff] & 0xff] ^ - Td3[Te4[(rk[2] ) & 0xff] & 0xff]; - rk[3] = - Td0[Te4[(rk[3] >> 24) ] & 0xff] ^ - Td1[Te4[(rk[3] >> 16) & 0xff] & 0xff] ^ - Td2[Te4[(rk[3] >> 8) & 0xff] & 0xff] ^ - Td3[Te4[(rk[3] ) & 0xff] & 0xff]; - } - return Nr; + /* expand the cipher key: */ + Nr = rijndaelKeySetupEnc(rk, cipherKey, keyBits); + /* invert the order of the round keys: */ + for (i = 0, j = 4 * Nr; i < j; i += 4, j -= 4) { + temp = rk[i]; + rk[i] = rk[j]; + rk[j] = temp; + temp = rk[i + 1]; + rk[i + 1] = rk[j + 1]; + rk[j + 1] = temp; + temp = rk[i + 2]; + rk[i + 2] = rk[j + 2]; + rk[j + 2] = temp; + temp = rk[i + 3]; + rk[i + 3] = rk[j + 3]; + rk[j + 3] = temp; + } + /* apply the inverse MixColumn transform to all round keys but the first and + * the last: */ + for (i = 1; i < Nr; i++) { + rk += 4; + rk[0] = + Td0[Te4[(rk[0] >> 24)] & 0xff] ^ Td1[Te4[(rk[0] >> 16) & 0xff] & 0xff] ^ + Td2[Te4[(rk[0] >> 8) & 0xff] & 0xff] ^ Td3[Te4[(rk[0]) & 0xff] & 0xff]; + rk[1] = + Td0[Te4[(rk[1] >> 24)] & 0xff] ^ Td1[Te4[(rk[1] >> 16) & 0xff] & 0xff] ^ + Td2[Te4[(rk[1] >> 8) & 0xff] & 0xff] ^ Td3[Te4[(rk[1]) & 0xff] & 0xff]; + rk[2] = + Td0[Te4[(rk[2] >> 24)] & 0xff] ^ Td1[Te4[(rk[2] >> 16) & 0xff] & 0xff] ^ + Td2[Te4[(rk[2] >> 8) & 0xff] & 0xff] ^ Td3[Te4[(rk[2]) & 0xff] & 0xff]; + rk[3] = + Td0[Te4[(rk[3] >> 24)] & 0xff] ^ Td1[Te4[(rk[3] >> 16) & 0xff] & 0xff] ^ + Td2[Te4[(rk[3] >> 8) & 0xff] & 0xff] ^ Td3[Te4[(rk[3]) & 0xff] & 0xff]; + } + return Nr; } -void rijndaelEncrypt(const u32 rk[/*4*(Nr + 1)*/], int Nr, const u8 pt[16], u8 ct[16]) { - u32 s0, s1, s2, s3, t0, t1, t2, t3; +void +rijndaelEncrypt(const u32 rk[/*4*(Nr + 1)*/], + int Nr, + const u8 pt[16], + u8 ct[16]) +{ + u32 s0, s1, s2, s3, t0, t1, t2, t3; #ifndef FULL_UNROLL - int r; + int r; #endif /* ?FULL_UNROLL */ - /* - * map byte array block to cipher state - * and add initial round key: - */ - s0 = GETU32(pt ) ^ rk[0]; - s1 = GETU32(pt + 4) ^ rk[1]; - s2 = GETU32(pt + 8) ^ rk[2]; - s3 = GETU32(pt + 12) ^ rk[3]; + /* + * map byte array block to cipher state + * and add initial round key: + */ + s0 = GETU32(pt) ^ rk[0]; + s1 = GETU32(pt + 4) ^ rk[1]; + s2 = GETU32(pt + 8) ^ rk[2]; + s3 = GETU32(pt + 12) ^ rk[3]; #ifdef FULL_UNROLL - /* round 1: */ - t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ Te3[s3 & 0xff] ^ rk[ 4]; - t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ Te3[s0 & 0xff] ^ rk[ 5]; - t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ Te3[s1 & 0xff] ^ rk[ 6]; - t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ Te3[s2 & 0xff] ^ rk[ 7]; - /* round 2: */ - s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ Te3[t3 & 0xff] ^ rk[ 8]; - s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ Te3[t0 & 0xff] ^ rk[ 9]; - s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ Te3[t1 & 0xff] ^ rk[10]; - s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ Te3[t2 & 0xff] ^ rk[11]; - /* round 3: */ - t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ Te3[s3 & 0xff] ^ rk[12]; - t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ Te3[s0 & 0xff] ^ rk[13]; - t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ Te3[s1 & 0xff] ^ rk[14]; - t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ Te3[s2 & 0xff] ^ rk[15]; - /* round 4: */ - s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ Te3[t3 & 0xff] ^ rk[16]; - s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ Te3[t0 & 0xff] ^ rk[17]; - s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ Te3[t1 & 0xff] ^ rk[18]; - s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ Te3[t2 & 0xff] ^ rk[19]; - /* round 5: */ - t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ Te3[s3 & 0xff] ^ rk[20]; - t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ Te3[s0 & 0xff] ^ rk[21]; - t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ Te3[s1 & 0xff] ^ rk[22]; - t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ Te3[s2 & 0xff] ^ rk[23]; - /* round 6: */ - s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ Te3[t3 & 0xff] ^ rk[24]; - s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ Te3[t0 & 0xff] ^ rk[25]; - s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ Te3[t1 & 0xff] ^ rk[26]; - s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ Te3[t2 & 0xff] ^ rk[27]; - /* round 7: */ - t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ Te3[s3 & 0xff] ^ rk[28]; - t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ Te3[s0 & 0xff] ^ rk[29]; - t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ Te3[s1 & 0xff] ^ rk[30]; - t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ Te3[s2 & 0xff] ^ rk[31]; - /* round 8: */ - s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ Te3[t3 & 0xff] ^ rk[32]; - s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ Te3[t0 & 0xff] ^ rk[33]; - s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ Te3[t1 & 0xff] ^ rk[34]; - s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ Te3[t2 & 0xff] ^ rk[35]; - /* round 9: */ - t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ Te3[s3 & 0xff] ^ rk[36]; - t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ Te3[s0 & 0xff] ^ rk[37]; - t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ Te3[s1 & 0xff] ^ rk[38]; - t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ Te3[s2 & 0xff] ^ rk[39]; - if (Nr > 10) { - /* round 10: */ - s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ Te3[t3 & 0xff] ^ rk[40]; - s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ Te3[t0 & 0xff] ^ rk[41]; - s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ Te3[t1 & 0xff] ^ rk[42]; - s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ Te3[t2 & 0xff] ^ rk[43]; - /* round 11: */ - t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ Te3[s3 & 0xff] ^ rk[44]; - t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ Te3[s0 & 0xff] ^ rk[45]; - t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ Te3[s1 & 0xff] ^ rk[46]; - t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ Te3[s2 & 0xff] ^ rk[47]; - if (Nr > 12) { - /* round 12: */ - s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ Te3[t3 & 0xff] ^ rk[48]; - s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ Te3[t0 & 0xff] ^ rk[49]; - s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ Te3[t1 & 0xff] ^ rk[50]; - s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ Te3[t2 & 0xff] ^ rk[51]; - /* round 13: */ - t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ Te3[s3 & 0xff] ^ rk[52]; - t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ Te3[s0 & 0xff] ^ rk[53]; - t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ Te3[s1 & 0xff] ^ rk[54]; - t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ Te3[s2 & 0xff] ^ rk[55]; - } + /* round 1: */ + t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ + Te3[s3 & 0xff] ^ rk[4]; + t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ + Te3[s0 & 0xff] ^ rk[5]; + t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ + Te3[s1 & 0xff] ^ rk[6]; + t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ + Te3[s2 & 0xff] ^ rk[7]; + /* round 2: */ + s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ + Te3[t3 & 0xff] ^ rk[8]; + s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ + Te3[t0 & 0xff] ^ rk[9]; + s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ + Te3[t1 & 0xff] ^ rk[10]; + s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ + Te3[t2 & 0xff] ^ rk[11]; + /* round 3: */ + t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ + Te3[s3 & 0xff] ^ rk[12]; + t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ + Te3[s0 & 0xff] ^ rk[13]; + t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ + Te3[s1 & 0xff] ^ rk[14]; + t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ + Te3[s2 & 0xff] ^ rk[15]; + /* round 4: */ + s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ + Te3[t3 & 0xff] ^ rk[16]; + s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ + Te3[t0 & 0xff] ^ rk[17]; + s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ + Te3[t1 & 0xff] ^ rk[18]; + s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ + Te3[t2 & 0xff] ^ rk[19]; + /* round 5: */ + t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ + Te3[s3 & 0xff] ^ rk[20]; + t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ + Te3[s0 & 0xff] ^ rk[21]; + t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ + Te3[s1 & 0xff] ^ rk[22]; + t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ + Te3[s2 & 0xff] ^ rk[23]; + /* round 6: */ + s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ + Te3[t3 & 0xff] ^ rk[24]; + s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ + Te3[t0 & 0xff] ^ rk[25]; + s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ + Te3[t1 & 0xff] ^ rk[26]; + s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ + Te3[t2 & 0xff] ^ rk[27]; + /* round 7: */ + t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ + Te3[s3 & 0xff] ^ rk[28]; + t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ + Te3[s0 & 0xff] ^ rk[29]; + t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ + Te3[s1 & 0xff] ^ rk[30]; + t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ + Te3[s2 & 0xff] ^ rk[31]; + /* round 8: */ + s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ + Te3[t3 & 0xff] ^ rk[32]; + s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ + Te3[t0 & 0xff] ^ rk[33]; + s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ + Te3[t1 & 0xff] ^ rk[34]; + s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ + Te3[t2 & 0xff] ^ rk[35]; + /* round 9: */ + t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ + Te3[s3 & 0xff] ^ rk[36]; + t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ + Te3[s0 & 0xff] ^ rk[37]; + t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ + Te3[s1 & 0xff] ^ rk[38]; + t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ + Te3[s2 & 0xff] ^ rk[39]; + if (Nr > 10) { + /* round 10: */ + s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ + Te3[t3 & 0xff] ^ rk[40]; + s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ + Te3[t0 & 0xff] ^ rk[41]; + s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ + Te3[t1 & 0xff] ^ rk[42]; + s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ + Te3[t2 & 0xff] ^ rk[43]; + /* round 11: */ + t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ + Te3[s3 & 0xff] ^ rk[44]; + t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ + Te3[s0 & 0xff] ^ rk[45]; + t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ + Te3[s1 & 0xff] ^ rk[46]; + t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ + Te3[s2 & 0xff] ^ rk[47]; + if (Nr > 12) { + /* round 12: */ + s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ + Te3[t3 & 0xff] ^ rk[48]; + s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ + Te3[t0 & 0xff] ^ rk[49]; + s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ + Te3[t1 & 0xff] ^ rk[50]; + s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ + Te3[t2 & 0xff] ^ rk[51]; + /* round 13: */ + t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ + Te3[s3 & 0xff] ^ rk[52]; + t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ + Te3[s0 & 0xff] ^ rk[53]; + t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ + Te3[s1 & 0xff] ^ rk[54]; + t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ + Te3[s2 & 0xff] ^ rk[55]; } - rk += Nr << 2; + } + rk += Nr << 2; #else /* !FULL_UNROLL */ - /* - * Nr - 1 full rounds: - */ - r = Nr >> 1; - for (;;) { - t0 = - Te0[(s0 >> 24) ] ^ - Te1[(s1 >> 16) & 0xff] ^ - Te2[(s2 >> 8) & 0xff] ^ - Te3[(s3 ) & 0xff] ^ - rk[4]; - t1 = - Te0[(s1 >> 24) ] ^ - Te1[(s2 >> 16) & 0xff] ^ - Te2[(s3 >> 8) & 0xff] ^ - Te3[(s0 ) & 0xff] ^ - rk[5]; - t2 = - Te0[(s2 >> 24) ] ^ - Te1[(s3 >> 16) & 0xff] ^ - Te2[(s0 >> 8) & 0xff] ^ - Te3[(s1 ) & 0xff] ^ - rk[6]; - t3 = - Te0[(s3 >> 24) ] ^ - Te1[(s0 >> 16) & 0xff] ^ - Te2[(s1 >> 8) & 0xff] ^ - Te3[(s2 ) & 0xff] ^ - rk[7]; + /* + * Nr - 1 full rounds: + */ + r = Nr >> 1; + for (;;) { + t0 = Te0[(s0 >> 24)] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ + Te3[(s3)&0xff] ^ rk[4]; + t1 = Te0[(s1 >> 24)] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ + Te3[(s0)&0xff] ^ rk[5]; + t2 = Te0[(s2 >> 24)] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ + Te3[(s1)&0xff] ^ rk[6]; + t3 = Te0[(s3 >> 24)] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ + Te3[(s2)&0xff] ^ rk[7]; - rk += 8; - if (--r == 0) { - break; - } - - s0 = - Te0[(t0 >> 24) ] ^ - Te1[(t1 >> 16) & 0xff] ^ - Te2[(t2 >> 8) & 0xff] ^ - Te3[(t3 ) & 0xff] ^ - rk[0]; - s1 = - Te0[(t1 >> 24) ] ^ - Te1[(t2 >> 16) & 0xff] ^ - Te2[(t3 >> 8) & 0xff] ^ - Te3[(t0 ) & 0xff] ^ - rk[1]; - s2 = - Te0[(t2 >> 24) ] ^ - Te1[(t3 >> 16) & 0xff] ^ - Te2[(t0 >> 8) & 0xff] ^ - Te3[(t1 ) & 0xff] ^ - rk[2]; - s3 = - Te0[(t3 >> 24) ] ^ - Te1[(t0 >> 16) & 0xff] ^ - Te2[(t1 >> 8) & 0xff] ^ - Te3[(t2 ) & 0xff] ^ - rk[3]; + rk += 8; + if (--r == 0) { + break; } + + s0 = Te0[(t0 >> 24)] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ + Te3[(t3)&0xff] ^ rk[0]; + s1 = Te0[(t1 >> 24)] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ + Te3[(t0)&0xff] ^ rk[1]; + s2 = Te0[(t2 >> 24)] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ + Te3[(t1)&0xff] ^ rk[2]; + s3 = Te0[(t3 >> 24)] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ + Te3[(t2)&0xff] ^ rk[3]; + } #endif /* ?FULL_UNROLL */ - /* - * apply last round and - * map cipher state to byte array block: - */ - s0 = - (Te4[(t0 >> 24) ] & 0xff000000) ^ - (Te4[(t1 >> 16) & 0xff] & 0x00ff0000) ^ - (Te4[(t2 >> 8) & 0xff] & 0x0000ff00) ^ - (Te4[(t3 ) & 0xff] & 0x000000ff) ^ - rk[0]; - PUTU32(ct , s0); - s1 = - (Te4[(t1 >> 24) ] & 0xff000000) ^ - (Te4[(t2 >> 16) & 0xff] & 0x00ff0000) ^ - (Te4[(t3 >> 8) & 0xff] & 0x0000ff00) ^ - (Te4[(t0 ) & 0xff] & 0x000000ff) ^ - rk[1]; - PUTU32(ct + 4, s1); - s2 = - (Te4[(t2 >> 24) ] & 0xff000000) ^ - (Te4[(t3 >> 16) & 0xff] & 0x00ff0000) ^ - (Te4[(t0 >> 8) & 0xff] & 0x0000ff00) ^ - (Te4[(t1 ) & 0xff] & 0x000000ff) ^ - rk[2]; - PUTU32(ct + 8, s2); - s3 = - (Te4[(t3 >> 24) ] & 0xff000000) ^ - (Te4[(t0 >> 16) & 0xff] & 0x00ff0000) ^ - (Te4[(t1 >> 8) & 0xff] & 0x0000ff00) ^ - (Te4[(t2 ) & 0xff] & 0x000000ff) ^ - rk[3]; - PUTU32(ct + 12, s3); + /* + * apply last round and + * map cipher state to byte array block: + */ + s0 = (Te4[(t0 >> 24)] & 0xff000000) ^ (Te4[(t1 >> 16) & 0xff] & 0x00ff0000) ^ + (Te4[(t2 >> 8) & 0xff] & 0x0000ff00) ^ (Te4[(t3)&0xff] & 0x000000ff) ^ + rk[0]; + PUTU32(ct, s0); + s1 = (Te4[(t1 >> 24)] & 0xff000000) ^ (Te4[(t2 >> 16) & 0xff] & 0x00ff0000) ^ + (Te4[(t3 >> 8) & 0xff] & 0x0000ff00) ^ (Te4[(t0)&0xff] & 0x000000ff) ^ + rk[1]; + PUTU32(ct + 4, s1); + s2 = (Te4[(t2 >> 24)] & 0xff000000) ^ (Te4[(t3 >> 16) & 0xff] & 0x00ff0000) ^ + (Te4[(t0 >> 8) & 0xff] & 0x0000ff00) ^ (Te4[(t1)&0xff] & 0x000000ff) ^ + rk[2]; + PUTU32(ct + 8, s2); + s3 = (Te4[(t3 >> 24)] & 0xff000000) ^ (Te4[(t0 >> 16) & 0xff] & 0x00ff0000) ^ + (Te4[(t1 >> 8) & 0xff] & 0x0000ff00) ^ (Te4[(t2)&0xff] & 0x000000ff) ^ + rk[3]; + PUTU32(ct + 12, s3); } -void rijndaelDecrypt(const u32 rk[/*4*(Nr + 1)*/], int Nr, const u8 ct[16], u8 pt[16]) { - u32 s0, s1, s2, s3, t0, t1, t2, t3; +void +rijndaelDecrypt(const u32 rk[/*4*(Nr + 1)*/], + int Nr, + const u8 ct[16], + u8 pt[16]) +{ + u32 s0, s1, s2, s3, t0, t1, t2, t3; #ifndef FULL_UNROLL - int r; + int r; #endif /* ?FULL_UNROLL */ - /* - * map byte array block to cipher state - * and add initial round key: - */ - s0 = GETU32(ct ) ^ rk[0]; - s1 = GETU32(ct + 4) ^ rk[1]; - s2 = GETU32(ct + 8) ^ rk[2]; - s3 = GETU32(ct + 12) ^ rk[3]; + /* + * map byte array block to cipher state + * and add initial round key: + */ + s0 = GETU32(ct) ^ rk[0]; + s1 = GETU32(ct + 4) ^ rk[1]; + s2 = GETU32(ct + 8) ^ rk[2]; + s3 = GETU32(ct + 12) ^ rk[3]; #ifdef FULL_UNROLL - /* round 1: */ - t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ Td3[s1 & 0xff] ^ rk[ 4]; - t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ Td3[s2 & 0xff] ^ rk[ 5]; - t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ Td3[s3 & 0xff] ^ rk[ 6]; - t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ Td3[s0 & 0xff] ^ rk[ 7]; - /* round 2: */ - s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ Td3[t1 & 0xff] ^ rk[ 8]; - s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ Td3[t2 & 0xff] ^ rk[ 9]; - s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ Td3[t3 & 0xff] ^ rk[10]; - s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ Td3[t0 & 0xff] ^ rk[11]; - /* round 3: */ - t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ Td3[s1 & 0xff] ^ rk[12]; - t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ Td3[s2 & 0xff] ^ rk[13]; - t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ Td3[s3 & 0xff] ^ rk[14]; - t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ Td3[s0 & 0xff] ^ rk[15]; - /* round 4: */ - s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ Td3[t1 & 0xff] ^ rk[16]; - s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ Td3[t2 & 0xff] ^ rk[17]; - s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ Td3[t3 & 0xff] ^ rk[18]; - s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ Td3[t0 & 0xff] ^ rk[19]; - /* round 5: */ - t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ Td3[s1 & 0xff] ^ rk[20]; - t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ Td3[s2 & 0xff] ^ rk[21]; - t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ Td3[s3 & 0xff] ^ rk[22]; - t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ Td3[s0 & 0xff] ^ rk[23]; - /* round 6: */ - s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ Td3[t1 & 0xff] ^ rk[24]; - s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ Td3[t2 & 0xff] ^ rk[25]; - s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ Td3[t3 & 0xff] ^ rk[26]; - s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ Td3[t0 & 0xff] ^ rk[27]; - /* round 7: */ - t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ Td3[s1 & 0xff] ^ rk[28]; - t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ Td3[s2 & 0xff] ^ rk[29]; - t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ Td3[s3 & 0xff] ^ rk[30]; - t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ Td3[s0 & 0xff] ^ rk[31]; - /* round 8: */ - s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ Td3[t1 & 0xff] ^ rk[32]; - s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ Td3[t2 & 0xff] ^ rk[33]; - s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ Td3[t3 & 0xff] ^ rk[34]; - s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ Td3[t0 & 0xff] ^ rk[35]; - /* round 9: */ - t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ Td3[s1 & 0xff] ^ rk[36]; - t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ Td3[s2 & 0xff] ^ rk[37]; - t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ Td3[s3 & 0xff] ^ rk[38]; - t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ Td3[s0 & 0xff] ^ rk[39]; - if (Nr > 10) { - /* round 10: */ - s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ Td3[t1 & 0xff] ^ rk[40]; - s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ Td3[t2 & 0xff] ^ rk[41]; - s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ Td3[t3 & 0xff] ^ rk[42]; - s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ Td3[t0 & 0xff] ^ rk[43]; - /* round 11: */ - t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ Td3[s1 & 0xff] ^ rk[44]; - t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ Td3[s2 & 0xff] ^ rk[45]; - t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ Td3[s3 & 0xff] ^ rk[46]; - t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ Td3[s0 & 0xff] ^ rk[47]; - if (Nr > 12) { - /* round 12: */ - s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ Td3[t1 & 0xff] ^ rk[48]; - s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ Td3[t2 & 0xff] ^ rk[49]; - s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ Td3[t3 & 0xff] ^ rk[50]; - s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ Td3[t0 & 0xff] ^ rk[51]; - /* round 13: */ - t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ Td3[s1 & 0xff] ^ rk[52]; - t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ Td3[s2 & 0xff] ^ rk[53]; - t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ Td3[s3 & 0xff] ^ rk[54]; - t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ Td3[s0 & 0xff] ^ rk[55]; - } + /* round 1: */ + t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ + Td3[s1 & 0xff] ^ rk[4]; + t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ + Td3[s2 & 0xff] ^ rk[5]; + t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ + Td3[s3 & 0xff] ^ rk[6]; + t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ + Td3[s0 & 0xff] ^ rk[7]; + /* round 2: */ + s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ + Td3[t1 & 0xff] ^ rk[8]; + s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ + Td3[t2 & 0xff] ^ rk[9]; + s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ + Td3[t3 & 0xff] ^ rk[10]; + s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ + Td3[t0 & 0xff] ^ rk[11]; + /* round 3: */ + t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ + Td3[s1 & 0xff] ^ rk[12]; + t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ + Td3[s2 & 0xff] ^ rk[13]; + t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ + Td3[s3 & 0xff] ^ rk[14]; + t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ + Td3[s0 & 0xff] ^ rk[15]; + /* round 4: */ + s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ + Td3[t1 & 0xff] ^ rk[16]; + s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ + Td3[t2 & 0xff] ^ rk[17]; + s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ + Td3[t3 & 0xff] ^ rk[18]; + s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ + Td3[t0 & 0xff] ^ rk[19]; + /* round 5: */ + t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ + Td3[s1 & 0xff] ^ rk[20]; + t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ + Td3[s2 & 0xff] ^ rk[21]; + t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ + Td3[s3 & 0xff] ^ rk[22]; + t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ + Td3[s0 & 0xff] ^ rk[23]; + /* round 6: */ + s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ + Td3[t1 & 0xff] ^ rk[24]; + s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ + Td3[t2 & 0xff] ^ rk[25]; + s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ + Td3[t3 & 0xff] ^ rk[26]; + s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ + Td3[t0 & 0xff] ^ rk[27]; + /* round 7: */ + t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ + Td3[s1 & 0xff] ^ rk[28]; + t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ + Td3[s2 & 0xff] ^ rk[29]; + t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ + Td3[s3 & 0xff] ^ rk[30]; + t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ + Td3[s0 & 0xff] ^ rk[31]; + /* round 8: */ + s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ + Td3[t1 & 0xff] ^ rk[32]; + s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ + Td3[t2 & 0xff] ^ rk[33]; + s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ + Td3[t3 & 0xff] ^ rk[34]; + s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ + Td3[t0 & 0xff] ^ rk[35]; + /* round 9: */ + t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ + Td3[s1 & 0xff] ^ rk[36]; + t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ + Td3[s2 & 0xff] ^ rk[37]; + t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ + Td3[s3 & 0xff] ^ rk[38]; + t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ + Td3[s0 & 0xff] ^ rk[39]; + if (Nr > 10) { + /* round 10: */ + s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ + Td3[t1 & 0xff] ^ rk[40]; + s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ + Td3[t2 & 0xff] ^ rk[41]; + s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ + Td3[t3 & 0xff] ^ rk[42]; + s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ + Td3[t0 & 0xff] ^ rk[43]; + /* round 11: */ + t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ + Td3[s1 & 0xff] ^ rk[44]; + t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ + Td3[s2 & 0xff] ^ rk[45]; + t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ + Td3[s3 & 0xff] ^ rk[46]; + t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ + Td3[s0 & 0xff] ^ rk[47]; + if (Nr > 12) { + /* round 12: */ + s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ + Td3[t1 & 0xff] ^ rk[48]; + s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ + Td3[t2 & 0xff] ^ rk[49]; + s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ + Td3[t3 & 0xff] ^ rk[50]; + s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ + Td3[t0 & 0xff] ^ rk[51]; + /* round 13: */ + t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ + Td3[s1 & 0xff] ^ rk[52]; + t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ + Td3[s2 & 0xff] ^ rk[53]; + t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ + Td3[s3 & 0xff] ^ rk[54]; + t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ + Td3[s0 & 0xff] ^ rk[55]; } - rk += Nr << 2; + } + rk += Nr << 2; #else /* !FULL_UNROLL */ - /* - * Nr - 1 full rounds: - */ - r = Nr >> 1; - for (;;) { - t0 = - Td0[(s0 >> 24) ] ^ - Td1[(s3 >> 16) & 0xff] ^ - Td2[(s2 >> 8) & 0xff] ^ - Td3[(s1 ) & 0xff] ^ - rk[4]; - t1 = - Td0[(s1 >> 24) ] ^ - Td1[(s0 >> 16) & 0xff] ^ - Td2[(s3 >> 8) & 0xff] ^ - Td3[(s2 ) & 0xff] ^ - rk[5]; - t2 = - Td0[(s2 >> 24) ] ^ - Td1[(s1 >> 16) & 0xff] ^ - Td2[(s0 >> 8) & 0xff] ^ - Td3[(s3 ) & 0xff] ^ - rk[6]; - t3 = - Td0[(s3 >> 24) ] ^ - Td1[(s2 >> 16) & 0xff] ^ - Td2[(s1 >> 8) & 0xff] ^ - Td3[(s0 ) & 0xff] ^ - rk[7]; - - rk += 8; - if (--r == 0) { - break; - } + /* + * Nr - 1 full rounds: + */ + r = Nr >> 1; + for (;;) { + t0 = Td0[(s0 >> 24)] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ + Td3[(s1)&0xff] ^ rk[4]; + t1 = Td0[(s1 >> 24)] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ + Td3[(s2)&0xff] ^ rk[5]; + t2 = Td0[(s2 >> 24)] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ + Td3[(s3)&0xff] ^ rk[6]; + t3 = Td0[(s3 >> 24)] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ + Td3[(s0)&0xff] ^ rk[7]; - s0 = - Td0[(t0 >> 24) ] ^ - Td1[(t3 >> 16) & 0xff] ^ - Td2[(t2 >> 8) & 0xff] ^ - Td3[(t1 ) & 0xff] ^ - rk[0]; - s1 = - Td0[(t1 >> 24) ] ^ - Td1[(t0 >> 16) & 0xff] ^ - Td2[(t3 >> 8) & 0xff] ^ - Td3[(t2 ) & 0xff] ^ - rk[1]; - s2 = - Td0[(t2 >> 24) ] ^ - Td1[(t1 >> 16) & 0xff] ^ - Td2[(t0 >> 8) & 0xff] ^ - Td3[(t3 ) & 0xff] ^ - rk[2]; - s3 = - Td0[(t3 >> 24) ] ^ - Td1[(t2 >> 16) & 0xff] ^ - Td2[(t1 >> 8) & 0xff] ^ - Td3[(t0 ) & 0xff] ^ - rk[3]; + rk += 8; + if (--r == 0) { + break; } + + s0 = Td0[(t0 >> 24)] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ + Td3[(t1)&0xff] ^ rk[0]; + s1 = Td0[(t1 >> 24)] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ + Td3[(t2)&0xff] ^ rk[1]; + s2 = Td0[(t2 >> 24)] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ + Td3[(t3)&0xff] ^ rk[2]; + s3 = Td0[(t3 >> 24)] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ + Td3[(t0)&0xff] ^ rk[3]; + } #endif /* ?FULL_UNROLL */ - /* - * apply last round and - * map cipher state to byte array block: - */ - s0 = - (Td4[(t0 >> 24) ] & 0xff000000) ^ - (Td4[(t3 >> 16) & 0xff] & 0x00ff0000) ^ - (Td4[(t2 >> 8) & 0xff] & 0x0000ff00) ^ - (Td4[(t1 ) & 0xff] & 0x000000ff) ^ - rk[0]; - PUTU32(pt , s0); - s1 = - (Td4[(t1 >> 24) ] & 0xff000000) ^ - (Td4[(t0 >> 16) & 0xff] & 0x00ff0000) ^ - (Td4[(t3 >> 8) & 0xff] & 0x0000ff00) ^ - (Td4[(t2 ) & 0xff] & 0x000000ff) ^ - rk[1]; - PUTU32(pt + 4, s1); - s2 = - (Td4[(t2 >> 24) ] & 0xff000000) ^ - (Td4[(t1 >> 16) & 0xff] & 0x00ff0000) ^ - (Td4[(t0 >> 8) & 0xff] & 0x0000ff00) ^ - (Td4[(t3 ) & 0xff] & 0x000000ff) ^ - rk[2]; - PUTU32(pt + 8, s2); - s3 = - (Td4[(t3 >> 24) ] & 0xff000000) ^ - (Td4[(t2 >> 16) & 0xff] & 0x00ff0000) ^ - (Td4[(t1 >> 8) & 0xff] & 0x0000ff00) ^ - (Td4[(t0 ) & 0xff] & 0x000000ff) ^ - rk[3]; - PUTU32(pt + 12, s3); + /* + * apply last round and + * map cipher state to byte array block: + */ + s0 = (Td4[(t0 >> 24)] & 0xff000000) ^ (Td4[(t3 >> 16) & 0xff] & 0x00ff0000) ^ + (Td4[(t2 >> 8) & 0xff] & 0x0000ff00) ^ (Td4[(t1)&0xff] & 0x000000ff) ^ + rk[0]; + PUTU32(pt, s0); + s1 = (Td4[(t1 >> 24)] & 0xff000000) ^ (Td4[(t0 >> 16) & 0xff] & 0x00ff0000) ^ + (Td4[(t3 >> 8) & 0xff] & 0x0000ff00) ^ (Td4[(t2)&0xff] & 0x000000ff) ^ + rk[1]; + PUTU32(pt + 4, s1); + s2 = (Td4[(t2 >> 24)] & 0xff000000) ^ (Td4[(t1 >> 16) & 0xff] & 0x00ff0000) ^ + (Td4[(t0 >> 8) & 0xff] & 0x0000ff00) ^ (Td4[(t3)&0xff] & 0x000000ff) ^ + rk[2]; + PUTU32(pt + 8, s2); + s3 = (Td4[(t3 >> 24)] & 0xff000000) ^ (Td4[(t2 >> 16) & 0xff] & 0x00ff0000) ^ + (Td4[(t1 >> 8) & 0xff] & 0x0000ff00) ^ (Td4[(t0)&0xff] & 0x000000ff) ^ + rk[3]; + PUTU32(pt + 12, s3); } #ifdef INTERMEDIATE_VALUE_KAT -void rijndaelEncryptRound(const u32 rk[/*4*(Nr + 1)*/], int Nr, u8 block[16], int rounds) { - int r; - u32 s0, s1, s2, s3, t0, t1, t2, t3; +void +rijndaelEncryptRound(const u32 rk[/*4*(Nr + 1)*/], + int Nr, + u8 block[16], + int rounds) +{ + int r; + u32 s0, s1, s2, s3, t0, t1, t2, t3; - /* - * map byte array block to cipher state - * and add initial round key: - */ - s0 = GETU32(block ) ^ rk[0]; - s1 = GETU32(block + 4) ^ rk[1]; - s2 = GETU32(block + 8) ^ rk[2]; - s3 = GETU32(block + 12) ^ rk[3]; - rk += 4; + /* + * map byte array block to cipher state + * and add initial round key: + */ + s0 = GETU32(block) ^ rk[0]; + s1 = GETU32(block + 4) ^ rk[1]; + s2 = GETU32(block + 8) ^ rk[2]; + s3 = GETU32(block + 12) ^ rk[3]; + rk += 4; - /* - * Nr - 1 full rounds: - */ - for (r = (rounds < Nr ? rounds : Nr - 1); r > 0; r--) { - t0 = - Te0[(s0 >> 24) ] ^ - Te1[(s1 >> 16) & 0xff] ^ - Te2[(s2 >> 8) & 0xff] ^ - Te3[(s3 ) & 0xff] ^ - rk[0]; - t1 = - Te0[(s1 >> 24) ] ^ - Te1[(s2 >> 16) & 0xff] ^ - Te2[(s3 >> 8) & 0xff] ^ - Te3[(s0 ) & 0xff] ^ - rk[1]; - t2 = - Te0[(s2 >> 24) ] ^ - Te1[(s3 >> 16) & 0xff] ^ - Te2[(s0 >> 8) & 0xff] ^ - Te3[(s1 ) & 0xff] ^ - rk[2]; - t3 = - Te0[(s3 >> 24) ] ^ - Te1[(s0 >> 16) & 0xff] ^ - Te2[(s1 >> 8) & 0xff] ^ - Te3[(s2 ) & 0xff] ^ - rk[3]; + /* + * Nr - 1 full rounds: + */ + for (r = (rounds < Nr ? rounds : Nr - 1); r > 0; r--) { + t0 = Te0[(s0 >> 24)] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ + Te3[(s3)&0xff] ^ rk[0]; + t1 = Te0[(s1 >> 24)] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ + Te3[(s0)&0xff] ^ rk[1]; + t2 = Te0[(s2 >> 24)] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ + Te3[(s1)&0xff] ^ rk[2]; + t3 = Te0[(s3 >> 24)] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ + Te3[(s2)&0xff] ^ rk[3]; - s0 = t0; - s1 = t1; - s2 = t2; - s3 = t3; - rk += 4; + s0 = t0; + s1 = t1; + s2 = t2; + s3 = t3; + rk += 4; + } - } + /* + * apply last round and + * map cipher state to byte array block: + */ + if (rounds == Nr) { + t0 = (Te4[(s0 >> 24)] & 0xff000000) ^ + (Te4[(s1 >> 16) & 0xff] & 0x00ff0000) ^ + (Te4[(s2 >> 8) & 0xff] & 0x0000ff00) ^ (Te4[(s3)&0xff] & 0x000000ff) ^ + rk[0]; + t1 = (Te4[(s1 >> 24)] & 0xff000000) ^ + (Te4[(s2 >> 16) & 0xff] & 0x00ff0000) ^ + (Te4[(s3 >> 8) & 0xff] & 0x0000ff00) ^ (Te4[(s0)&0xff] & 0x000000ff) ^ + rk[1]; + t2 = (Te4[(s2 >> 24)] & 0xff000000) ^ + (Te4[(s3 >> 16) & 0xff] & 0x00ff0000) ^ + (Te4[(s0 >> 8) & 0xff] & 0x0000ff00) ^ (Te4[(s1)&0xff] & 0x000000ff) ^ + rk[2]; + t3 = (Te4[(s3 >> 24)] & 0xff000000) ^ + (Te4[(s0 >> 16) & 0xff] & 0x00ff0000) ^ + (Te4[(s1 >> 8) & 0xff] & 0x0000ff00) ^ (Te4[(s2)&0xff] & 0x000000ff) ^ + rk[3]; - /* - * apply last round and - * map cipher state to byte array block: - */ - if (rounds == Nr) { - t0 = - (Te4[(s0 >> 24) ] & 0xff000000) ^ - (Te4[(s1 >> 16) & 0xff] & 0x00ff0000) ^ - (Te4[(s2 >> 8) & 0xff] & 0x0000ff00) ^ - (Te4[(s3 ) & 0xff] & 0x000000ff) ^ - rk[0]; - t1 = - (Te4[(s1 >> 24) ] & 0xff000000) ^ - (Te4[(s2 >> 16) & 0xff] & 0x00ff0000) ^ - (Te4[(s3 >> 8) & 0xff] & 0x0000ff00) ^ - (Te4[(s0 ) & 0xff] & 0x000000ff) ^ - rk[1]; - t2 = - (Te4[(s2 >> 24) ] & 0xff000000) ^ - (Te4[(s3 >> 16) & 0xff] & 0x00ff0000) ^ - (Te4[(s0 >> 8) & 0xff] & 0x0000ff00) ^ - (Te4[(s1 ) & 0xff] & 0x000000ff) ^ - rk[2]; - t3 = - (Te4[(s3 >> 24) ] & 0xff000000) ^ - (Te4[(s0 >> 16) & 0xff] & 0x00ff0000) ^ - (Te4[(s1 >> 8) & 0xff] & 0x0000ff00) ^ - (Te4[(s2 ) & 0xff] & 0x000000ff) ^ - rk[3]; - - s0 = t0; - s1 = t1; - s2 = t2; - s3 = t3; - } + s0 = t0; + s1 = t1; + s2 = t2; + s3 = t3; + } - PUTU32(block , s0); - PUTU32(block + 4, s1); - PUTU32(block + 8, s2); - PUTU32(block + 12, s3); + PUTU32(block, s0); + PUTU32(block + 4, s1); + PUTU32(block + 8, s2); + PUTU32(block + 12, s3); } -void rijndaelDecryptRound(const u32 rk[/*4*(Nr + 1)*/], int Nr, u8 block[16], int rounds) { - int r; - u32 s0, s1, s2, s3, t0, t1, t2, t3; - - /* - * map byte array block to cipher state - * and add initial round key: - */ - s0 = GETU32(block ) ^ rk[0]; - s1 = GETU32(block + 4) ^ rk[1]; - s2 = GETU32(block + 8) ^ rk[2]; - s3 = GETU32(block + 12) ^ rk[3]; - rk += 4; +void +rijndaelDecryptRound(const u32 rk[/*4*(Nr + 1)*/], + int Nr, + u8 block[16], + int rounds) +{ + int r; + u32 s0, s1, s2, s3, t0, t1, t2, t3; - /* - * Nr - 1 full rounds: - */ - for (r = (rounds < Nr ? rounds : Nr) - 1; r > 0; r--) { - t0 = - Td0[(s0 >> 24) ] ^ - Td1[(s3 >> 16) & 0xff] ^ - Td2[(s2 >> 8) & 0xff] ^ - Td3[(s1 ) & 0xff] ^ - rk[0]; - t1 = - Td0[(s1 >> 24) ] ^ - Td1[(s0 >> 16) & 0xff] ^ - Td2[(s3 >> 8) & 0xff] ^ - Td3[(s2 ) & 0xff] ^ - rk[1]; - t2 = - Td0[(s2 >> 24) ] ^ - Td1[(s1 >> 16) & 0xff] ^ - Td2[(s0 >> 8) & 0xff] ^ - Td3[(s3 ) & 0xff] ^ - rk[2]; - t3 = - Td0[(s3 >> 24) ] ^ - Td1[(s2 >> 16) & 0xff] ^ - Td2[(s1 >> 8) & 0xff] ^ - Td3[(s0 ) & 0xff] ^ - rk[3]; + /* + * map byte array block to cipher state + * and add initial round key: + */ + s0 = GETU32(block) ^ rk[0]; + s1 = GETU32(block + 4) ^ rk[1]; + s2 = GETU32(block + 8) ^ rk[2]; + s3 = GETU32(block + 12) ^ rk[3]; + rk += 4; - s0 = t0; - s1 = t1; - s2 = t2; - s3 = t3; - rk += 4; + /* + * Nr - 1 full rounds: + */ + for (r = (rounds < Nr ? rounds : Nr) - 1; r > 0; r--) { + t0 = Td0[(s0 >> 24)] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ + Td3[(s1)&0xff] ^ rk[0]; + t1 = Td0[(s1 >> 24)] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ + Td3[(s2)&0xff] ^ rk[1]; + t2 = Td0[(s2 >> 24)] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ + Td3[(s3)&0xff] ^ rk[2]; + t3 = Td0[(s3 >> 24)] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ + Td3[(s0)&0xff] ^ rk[3]; - } + s0 = t0; + s1 = t1; + s2 = t2; + s3 = t3; + rk += 4; + } - /* - * complete the last round and - * map cipher state to byte array block: - */ - t0 = - (Td4[(s0 >> 24) ] & 0xff000000) ^ - (Td4[(s3 >> 16) & 0xff] & 0x00ff0000) ^ - (Td4[(s2 >> 8) & 0xff] & 0x0000ff00) ^ - (Td4[(s1 ) & 0xff] & 0x000000ff); - t1 = - (Td4[(s1 >> 24) ] & 0xff000000) ^ - (Td4[(s0 >> 16) & 0xff] & 0x00ff0000) ^ - (Td4[(s3 >> 8) & 0xff] & 0x0000ff00) ^ - (Td4[(s2 ) & 0xff] & 0x000000ff); - t2 = - (Td4[(s2 >> 24) ] & 0xff000000) ^ - (Td4[(s1 >> 16) & 0xff] & 0x00ff0000) ^ - (Td4[(s0 >> 8) & 0xff] & 0x0000ff00) ^ - (Td4[(s3 ) & 0xff] & 0x000000ff); - t3 = - (Td4[(s3 >> 24) ] & 0xff000000) ^ - (Td4[(s2 >> 16) & 0xff] & 0x00ff0000) ^ - (Td4[(s1 >> 8) & 0xff] & 0x0000ff00) ^ - (Td4[(s0 ) & 0xff] & 0x000000ff); + /* + * complete the last round and + * map cipher state to byte array block: + */ + t0 = (Td4[(s0 >> 24)] & 0xff000000) ^ (Td4[(s3 >> 16) & 0xff] & 0x00ff0000) ^ + (Td4[(s2 >> 8) & 0xff] & 0x0000ff00) ^ (Td4[(s1)&0xff] & 0x000000ff); + t1 = (Td4[(s1 >> 24)] & 0xff000000) ^ (Td4[(s0 >> 16) & 0xff] & 0x00ff0000) ^ + (Td4[(s3 >> 8) & 0xff] & 0x0000ff00) ^ (Td4[(s2)&0xff] & 0x000000ff); + t2 = (Td4[(s2 >> 24)] & 0xff000000) ^ (Td4[(s1 >> 16) & 0xff] & 0x00ff0000) ^ + (Td4[(s0 >> 8) & 0xff] & 0x0000ff00) ^ (Td4[(s3)&0xff] & 0x000000ff); + t3 = (Td4[(s3 >> 24)] & 0xff000000) ^ (Td4[(s2 >> 16) & 0xff] & 0x00ff0000) ^ + (Td4[(s1 >> 8) & 0xff] & 0x0000ff00) ^ (Td4[(s0)&0xff] & 0x000000ff); - if (rounds == Nr) { - t0 ^= rk[0]; - t1 ^= rk[1]; - t2 ^= rk[2]; - t3 ^= rk[3]; - } + if (rounds == Nr) { + t0 ^= rk[0]; + t1 ^= rk[1]; + t2 ^= rk[2]; + t3 ^= rk[3]; + } - PUTU32(block , t0); - PUTU32(block + 4, t1); - PUTU32(block + 8, t2); - PUTU32(block + 12, t3); + PUTU32(block, t0); + PUTU32(block + 4, t1); + PUTU32(block + 8, t2); + PUTU32(block + 12, t3); } #endif /* INTERMEDIATE_VALUE_KAT */ diff --git a/examples/aes32/rijndael-alg-fst.h b/examples/aes32/rijndael-alg-fst.h index 496bb4d..ba909b5 100644 --- a/examples/aes32/rijndael-alg-fst.h +++ b/examples/aes32/rijndael-alg-fst.h @@ -30,23 +30,40 @@ #define __RIJNDAEL_ALG_FST_H #pragma clang diagnostic pop +#define MAXKC (256 / 32) +#define MAXKB (256 / 8) +#define MAXNR 14 -#define MAXKC (256/32) -#define MAXKB (256/8) -#define MAXNR 14 +typedef unsigned char u8; +typedef unsigned short u16; +typedef unsigned int u32; -typedef unsigned char u8; -typedef unsigned short u16; -typedef unsigned int u32; - -int rijndaelKeySetupEnc(u32 rk[/*4*(Nr + 1)*/], const u8 cipherKey[], int keyBits); -int rijndaelKeySetupDec(u32 rk[/*4*(Nr + 1)*/], const u8 cipherKey[], int keyBits); -void rijndaelEncrypt(const u32 rk[/*4*(Nr + 1)*/], int Nr, const u8 pt[16], u8 ct[16]); -void rijndaelDecrypt(const u32 rk[/*4*(Nr + 1)*/], int Nr, const u8 ct[16], u8 pt[16]); +int +rijndaelKeySetupEnc(u32 rk[/*4*(Nr + 1)*/], const u8 cipherKey[], int keyBits); +int +rijndaelKeySetupDec(u32 rk[/*4*(Nr + 1)*/], const u8 cipherKey[], int keyBits); +void +rijndaelEncrypt(const u32 rk[/*4*(Nr + 1)*/], + int Nr, + const u8 pt[16], + u8 ct[16]); +void +rijndaelDecrypt(const u32 rk[/*4*(Nr + 1)*/], + int Nr, + const u8 ct[16], + u8 pt[16]); #ifdef INTERMEDIATE_VALUE_KAT -void rijndaelEncryptRound(const u32 rk[/*4*(Nr + 1)*/], int Nr, u8 block[16], int rounds); -void rijndaelDecryptRound(const u32 rk[/*4*(Nr + 1)*/], int Nr, u8 block[16], int rounds); +void +rijndaelEncryptRound(const u32 rk[/*4*(Nr + 1)*/], + int Nr, + u8 block[16], + int rounds); +void +rijndaelDecryptRound(const u32 rk[/*4*(Nr + 1)*/], + int Nr, + u8 block[16], + int rounds); #endif /* INTERMEDIATE_VALUE_KAT */ #endif /* __RIJNDAEL_ALG_FST_H */ diff --git a/examples/aesbitsliced/afternm_aes128ctr.c b/examples/aesbitsliced/afternm_aes128ctr.c index a5a9a7a..4a3b098 100644 --- a/examples/aesbitsliced/afternm_aes128ctr.c +++ b/examples/aesbitsliced/afternm_aes128ctr.c @@ -1,13 +1,16 @@ -/* Author: Peter Schwabe, ported from an assembly implementation by Emilia Käsper - * Date: 2009-03-19 - * Public domain */ +/* Author: Peter Schwabe, ported from an assembly implementation by Emilia + * Käsper Date: 2009-03-19 Public domain */ #include "api.h" -#include "int128.h" #include "common.h" #include "consts.h" +#include "int128.h" -int crypto_stream_afternm(unsigned char *out, unsigned long long len, const unsigned char *nonce, const unsigned char *c) +int +crypto_stream_afternm(unsigned char* out, + unsigned long long len, + const unsigned char* nonce, + const unsigned char* c) { int128 xmm0; @@ -31,129 +34,291 @@ int crypto_stream_afternm(unsigned char *out, unsigned long long len, const unsi int128 nonce_stack; unsigned long long lensav; unsigned char bl[128]; - unsigned char *blp; - unsigned char *np; + unsigned char* blp; + unsigned char* np; unsigned char b; uint32 tmp; /* Copy nonce on the stack */ - copy2(&nonce_stack, (const int128 *) (nonce + 0)); - np = (unsigned char *)&nonce_stack; - - enc_block: - - xmm0 = *(int128 *) (np + 0); - copy2(&xmm1, &xmm0); - shufb(&xmm1, SWAP32); - copy2(&xmm2, &xmm1); - copy2(&xmm3, &xmm1); - copy2(&xmm4, &xmm1); - copy2(&xmm5, &xmm1); - copy2(&xmm6, &xmm1); - copy2(&xmm7, &xmm1); - - add_uint32_big(&xmm1, 1); - add_uint32_big(&xmm2, 2); - add_uint32_big(&xmm3, 3); - add_uint32_big(&xmm4, 4); - add_uint32_big(&xmm5, 5); - add_uint32_big(&xmm6, 6); - add_uint32_big(&xmm7, 7); - - shufb(&xmm0, M0); - shufb(&xmm1, M0SWAP); - shufb(&xmm2, M0SWAP); - shufb(&xmm3, M0SWAP); - shufb(&xmm4, M0SWAP); - shufb(&xmm5, M0SWAP); - shufb(&xmm6, M0SWAP); - shufb(&xmm7, M0SWAP); - - bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, xmm8) - - aesround( 1, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c) - aesround( 2, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c) - aesround( 3, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c) - aesround( 4, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c) - aesround( 5, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c) - aesround( 6, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c) - aesround( 7, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c) - aesround( 8, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c) - aesround( 9, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c) - lastround(xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c) - - bitslice(xmm13, xmm10, xmm15, xmm11, xmm14, xmm12, xmm9, xmm8, xmm0) - - if(len < 128) goto partial; - if(len == 128) goto full; - - tmp = load32_bigendian(np + 12); - tmp += 8; - store32_bigendian(np + 12, tmp); - - *(int128 *) (out + 0) = xmm8; - *(int128 *) (out + 16) = xmm9; - *(int128 *) (out + 32) = xmm12; - *(int128 *) (out + 48) = xmm14; - *(int128 *) (out + 64) = xmm11; - *(int128 *) (out + 80) = xmm15; - *(int128 *) (out + 96) = xmm10; - *(int128 *) (out + 112) = xmm13; - - len -= 128; - out += 128; - - goto enc_block; - - partial: - - lensav = len; - len >>= 4; - - tmp = load32_bigendian(np + 12); - tmp += len; - store32_bigendian(np + 12, tmp); - - blp = bl; - *(int128 *)(blp + 0) = xmm8; - *(int128 *)(blp + 16) = xmm9; - *(int128 *)(blp + 32) = xmm12; - *(int128 *)(blp + 48) = xmm14; - *(int128 *)(blp + 64) = xmm11; - *(int128 *)(blp + 80) = xmm15; - *(int128 *)(blp + 96) = xmm10; - *(int128 *)(blp + 112) = xmm13; - - bytes: - - if(lensav == 0) goto end; - - b = blp[0]; /* clang false positive */ - *(unsigned char *)(out + 0) = b; - - blp += 1; - out +=1; - lensav -= 1; - - goto bytes; - - full: - - tmp = load32_bigendian(np + 12); - tmp += 8; - store32_bigendian(np + 12, tmp); - - *(int128 *) (out + 0) = xmm8; - *(int128 *) (out + 16) = xmm9; - *(int128 *) (out + 32) = xmm12; - *(int128 *) (out + 48) = xmm14; - *(int128 *) (out + 64) = xmm11; - *(int128 *) (out + 80) = xmm15; - *(int128 *) (out + 96) = xmm10; - *(int128 *) (out + 112) = xmm13; - - end: - return 0; - + copy2(&nonce_stack, (const int128*)(nonce + 0)); + np = (unsigned char*)&nonce_stack; + +enc_block: + + xmm0 = *(int128*)(np + 0); + copy2(&xmm1, &xmm0); + shufb(&xmm1, SWAP32); + copy2(&xmm2, &xmm1); + copy2(&xmm3, &xmm1); + copy2(&xmm4, &xmm1); + copy2(&xmm5, &xmm1); + copy2(&xmm6, &xmm1); + copy2(&xmm7, &xmm1); + + add_uint32_big(&xmm1, 1); + add_uint32_big(&xmm2, 2); + add_uint32_big(&xmm3, 3); + add_uint32_big(&xmm4, 4); + add_uint32_big(&xmm5, 5); + add_uint32_big(&xmm6, 6); + add_uint32_big(&xmm7, 7); + + shufb(&xmm0, M0); + shufb(&xmm1, M0SWAP); + shufb(&xmm2, M0SWAP); + shufb(&xmm3, M0SWAP); + shufb(&xmm4, M0SWAP); + shufb(&xmm5, M0SWAP); + shufb(&xmm6, M0SWAP); + shufb(&xmm7, M0SWAP); + + bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, xmm8) + + aesround(1, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + c) aesround(2, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + c) aesround(3, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + c) aesround(4, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + c) aesround(5, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + c) aesround(6, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + c) + aesround(7, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + c) aesround(8, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + c) aesround(9, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + c) lastround(xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + c) + + bitslice(xmm13, xmm10, xmm15, xmm11, xmm14, xmm12, xmm9, xmm8, xmm0) + + if (len < 128) goto partial; + if (len == 128) + goto full; + + tmp = load32_bigendian(np + 12); + tmp += 8; + store32_bigendian(np + 12, tmp); + + *(int128*)(out + 0) = xmm8; + *(int128*)(out + 16) = xmm9; + *(int128*)(out + 32) = xmm12; + *(int128*)(out + 48) = xmm14; + *(int128*)(out + 64) = xmm11; + *(int128*)(out + 80) = xmm15; + *(int128*)(out + 96) = xmm10; + *(int128*)(out + 112) = xmm13; + + len -= 128; + out += 128; + + goto enc_block; + +partial: + + lensav = len; + len >>= 4; + + tmp = load32_bigendian(np + 12); + tmp += len; + store32_bigendian(np + 12, tmp); + + blp = bl; + *(int128*)(blp + 0) = xmm8; + *(int128*)(blp + 16) = xmm9; + *(int128*)(blp + 32) = xmm12; + *(int128*)(blp + 48) = xmm14; + *(int128*)(blp + 64) = xmm11; + *(int128*)(blp + 80) = xmm15; + *(int128*)(blp + 96) = xmm10; + *(int128*)(blp + 112) = xmm13; + +bytes: + + if (lensav == 0) + goto end; + + b = blp[0]; /* clang false positive */ + *(unsigned char*)(out + 0) = b; + + blp += 1; + out += 1; + lensav -= 1; + + goto bytes; + +full: + + tmp = load32_bigendian(np + 12); + tmp += 8; + store32_bigendian(np + 12, tmp); + + *(int128*)(out + 0) = xmm8; + *(int128*)(out + 16) = xmm9; + *(int128*)(out + 32) = xmm12; + *(int128*)(out + 48) = xmm14; + *(int128*)(out + 64) = xmm11; + *(int128*)(out + 80) = xmm15; + *(int128*)(out + 96) = xmm10; + *(int128*)(out + 112) = xmm13; + +end: + return 0; } diff --git a/examples/aesbitsliced/beforenm_aes128ctr.c b/examples/aesbitsliced/beforenm_aes128ctr.c index f8623dd..e494e35 100644 --- a/examples/aesbitsliced/beforenm_aes128ctr.c +++ b/examples/aesbitsliced/beforenm_aes128ctr.c @@ -1,13 +1,13 @@ -/* Author: Peter Schwabe, ported from an assembly implementation by Emilia Käsper - * Date: 2009-03-19 - * Public domain */ +/* Author: Peter Schwabe, ported from an assembly implementation by Emilia + * Käsper Date: 2009-03-19 Public domain */ #include "api.h" +#include "common.h" #include "consts.h" #include "int128.h" -#include "common.h" -int crypto_stream_beforenm(unsigned char *c, const unsigned char *k) +int +crypto_stream_beforenm(unsigned char* c, const unsigned char* k) { /* @@ -42,18 +42,183 @@ int crypto_stream_beforenm(unsigned char *c, const unsigned char *k) bitslicekey0(k, c) - keyexpbs1(xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c) - keyexpbs(xmm0, xmm1, xmm4, xmm6, xmm3, xmm7, xmm2, xmm5, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm1);, 2,c) - keyexpbs(xmm0, xmm1, xmm3, xmm2, xmm6, xmm5, xmm4, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm6);, 3,c) - keyexpbs(xmm0, xmm1, xmm6, xmm4, xmm2, xmm7, xmm3, xmm5, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm3);, 4,c) + keyexpbs1(xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + c) keyexpbs(xmm0, + xmm1, + xmm4, + xmm6, + xmm3, + xmm7, + xmm2, + xmm5, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xor_rcon(&xmm1); + , 2, c) keyexpbs(xmm0, + xmm1, + xmm3, + xmm2, + xmm6, + xmm5, + xmm4, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xor_rcon(&xmm6); + , 3, c) keyexpbs(xmm0, + xmm1, + xmm6, + xmm4, + xmm2, + xmm7, + xmm3, + xmm5, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xor_rcon(&xmm3); + , 4, c) - keyexpbs(xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm3);, 5,c) - keyexpbs(xmm0, xmm1, xmm4, xmm6, xmm3, xmm7, xmm2, xmm5, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm5);, 6,c) - keyexpbs(xmm0, xmm1, xmm3, xmm2, xmm6, xmm5, xmm4, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm3);, 7,c) - keyexpbs(xmm0, xmm1, xmm6, xmm4, xmm2, xmm7, xmm3, xmm5, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm7);, 8,c) + keyexpbs(xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xor_rcon(&xmm3); + , 5, c) keyexpbs(xmm0, + xmm1, + xmm4, + xmm6, + xmm3, + xmm7, + xmm2, + xmm5, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xor_rcon(&xmm5); + , 6, c) keyexpbs(xmm0, + xmm1, + xmm3, + xmm2, + xmm6, + xmm5, + xmm4, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xor_rcon(&xmm3); + , 7, c) + keyexpbs(xmm0, + xmm1, + xmm6, + xmm4, + xmm2, + xmm7, + xmm3, + xmm5, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xor_rcon(&xmm7); + , 8, c) - keyexpbs(xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm0); xor_rcon(&xmm1); xor_rcon(&xmm6); xor_rcon(&xmm3);, 9,c) - keyexpbs10(xmm0, xmm1, xmm4, xmm6, xmm3, xmm7, xmm2, xmm5, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c) + keyexpbs(xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xor_rcon(&xmm0); + xor_rcon(&xmm1); + xor_rcon(&xmm6); + xor_rcon(&xmm3); + , 9, c) keyexpbs10(xmm0, + xmm1, + xmm4, + xmm6, + xmm3, + xmm7, + xmm2, + xmm5, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + c) - return 0; + return 0; } diff --git a/examples/aesbitsliced/common.h b/examples/aesbitsliced/common.h index 3923c02..940f386 100644 --- a/examples/aesbitsliced/common.h +++ b/examples/aesbitsliced/common.h @@ -1,788 +1,808 @@ -/* Author: Peter Schwabe, ported from an assembly implementation by Emilia Käsper - Date: 2009-03-19 - Public domain */ +/* Author: Peter Schwabe, ported from an assembly implementation by Emilia + Käsper Date: 2009-03-19 Public domain */ #ifndef COMMON_H #define COMMON_H #include "types.h" #define load32_bigendian crypto_stream_aes128ctr_portable_load32_bigendian -uint32 load32_bigendian(const unsigned char *x); +uint32 +load32_bigendian(const unsigned char* x); #define store32_bigendian crypto_stream_aes128ctr_portable_store32_bigendian -void store32_bigendian(unsigned char *x,uint32 u); +void +store32_bigendian(unsigned char* x, uint32 u); #define load32_littleendian crypto_stream_aes128ctr_portable_load32_littleendian -uint32 load32_littleendian(const unsigned char *x); +uint32 +load32_littleendian(const unsigned char* x); -#define store32_littleendian crypto_stream_aes128ctr_portable_store32_littleendian -void store32_littleendian(unsigned char *x,uint32 u); +#define store32_littleendian \ + crypto_stream_aes128ctr_portable_store32_littleendian +void +store32_littleendian(unsigned char* x, uint32 u); #define load64_littleendian crypto_stream_aes128ctr_portable_load64_littleendian -uint64 load64_littleendian(const unsigned char *x); +uint64 +load64_littleendian(const unsigned char* x); -#define store64_littleendian crypto_stream_aes128ctr_portable_store64_littleendian -void store64_littleendian(unsigned char *x,uint64 u); +#define store64_littleendian \ + crypto_stream_aes128ctr_portable_store64_littleendian +void +store64_littleendian(unsigned char* x, uint64 u); /* Macros required only for key expansion */ -#define keyexpbs1(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7, bskey) \ - rotbyte(&b0);\ - rotbyte(&b1);\ - rotbyte(&b2);\ - rotbyte(&b3);\ - rotbyte(&b4);\ - rotbyte(&b5);\ - rotbyte(&b6);\ - rotbyte(&b7);\ - ;\ - sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7);\ - ;\ - xor_rcon(&b0);\ - shufb(&b0, EXPB0);\ - shufb(&b1, EXPB0);\ - shufb(&b4, EXPB0);\ - shufb(&b6, EXPB0);\ - shufb(&b3, EXPB0);\ - shufb(&b7, EXPB0);\ - shufb(&b2, EXPB0);\ - shufb(&b5, EXPB0);\ - shufb(&b0, EXPB0);\ - ;\ - t0 = *(int128 *)(bskey + 0);\ - t1 = *(int128 *)(bskey + 16);\ - t2 = *(int128 *)(bskey + 32);\ - t3 = *(int128 *)(bskey + 48);\ - t4 = *(int128 *)(bskey + 64);\ - t5 = *(int128 *)(bskey + 80);\ - t6 = *(int128 *)(bskey + 96);\ - t7 = *(int128 *)(bskey + 112);\ - ;\ - xor2(&b0, &t0);\ - xor2(&b1, &t1);\ - xor2(&b4, &t2);\ - xor2(&b6, &t3);\ - xor2(&b3, &t4);\ - xor2(&b7, &t5);\ - xor2(&b2, &t6);\ - xor2(&b5, &t7);\ - ;\ - rshift32_littleendian(&t0, 8);\ - rshift32_littleendian(&t1, 8);\ - rshift32_littleendian(&t2, 8);\ - rshift32_littleendian(&t3, 8);\ - rshift32_littleendian(&t4, 8);\ - rshift32_littleendian(&t5, 8);\ - rshift32_littleendian(&t6, 8);\ - rshift32_littleendian(&t7, 8);\ - ;\ - xor2(&b0, &t0);\ - xor2(&b1, &t1);\ - xor2(&b4, &t2);\ - xor2(&b6, &t3);\ - xor2(&b3, &t4);\ - xor2(&b7, &t5);\ - xor2(&b2, &t6);\ - xor2(&b5, &t7);\ - ;\ - rshift32_littleendian(&t0, 8);\ - rshift32_littleendian(&t1, 8);\ - rshift32_littleendian(&t2, 8);\ - rshift32_littleendian(&t3, 8);\ - rshift32_littleendian(&t4, 8);\ - rshift32_littleendian(&t5, 8);\ - rshift32_littleendian(&t6, 8);\ - rshift32_littleendian(&t7, 8);\ - ;\ - xor2(&b0, &t0);\ - xor2(&b1, &t1);\ - xor2(&b4, &t2);\ - xor2(&b6, &t3);\ - xor2(&b3, &t4);\ - xor2(&b7, &t5);\ - xor2(&b2, &t6);\ - xor2(&b5, &t7);\ - ;\ - rshift32_littleendian(&t0, 8);\ - rshift32_littleendian(&t1, 8);\ - rshift32_littleendian(&t2, 8);\ - rshift32_littleendian(&t3, 8);\ - rshift32_littleendian(&t4, 8);\ - rshift32_littleendian(&t5, 8);\ - rshift32_littleendian(&t6, 8);\ - rshift32_littleendian(&t7, 8);\ - ;\ - xor2(&b0, &t0);\ - xor2(&b1, &t1);\ - xor2(&b4, &t2);\ - xor2(&b6, &t3);\ - xor2(&b3, &t4);\ - xor2(&b7, &t5);\ - xor2(&b2, &t6);\ - xor2(&b5, &t7);\ - ;\ - *(int128 *)(bskey + 128) = b0;\ - *(int128 *)(bskey + 144) = b1;\ - *(int128 *)(bskey + 160) = b4;\ - *(int128 *)(bskey + 176) = b6;\ - *(int128 *)(bskey + 192) = b3;\ - *(int128 *)(bskey + 208) = b7;\ - *(int128 *)(bskey + 224) = b2;\ - *(int128 *)(bskey + 240) = b5;\ - -#define keyexpbs10(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7, bskey) ;\ - toggle(&b0);\ - toggle(&b1);\ - toggle(&b5);\ - toggle(&b6);\ - rotbyte(&b0);\ - rotbyte(&b1);\ - rotbyte(&b2);\ - rotbyte(&b3);\ - rotbyte(&b4);\ - rotbyte(&b5);\ - rotbyte(&b6);\ - rotbyte(&b7);\ - ;\ - sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7);\ - ;\ - xor_rcon(&b1);\ - xor_rcon(&b4);\ - xor_rcon(&b3);\ - xor_rcon(&b7);\ - shufb(&b0, EXPB0);\ - shufb(&b1, EXPB0);\ - shufb(&b4, EXPB0);\ - shufb(&b6, EXPB0);\ - shufb(&b3, EXPB0);\ - shufb(&b7, EXPB0);\ - shufb(&b2, EXPB0);\ - shufb(&b5, EXPB0);\ - ;\ - t0 = *(int128 *)(bskey + 9 * 128 + 0);\ - t1 = *(int128 *)(bskey + 9 * 128 + 16);\ - t2 = *(int128 *)(bskey + 9 * 128 + 32);\ - t3 = *(int128 *)(bskey + 9 * 128 + 48);\ - t4 = *(int128 *)(bskey + 9 * 128 + 64);\ - t5 = *(int128 *)(bskey + 9 * 128 + 80);\ - t6 = *(int128 *)(bskey + 9 * 128 + 96);\ - t7 = *(int128 *)(bskey + 9 * 128 + 112);\ - ;\ - toggle(&t0);\ - toggle(&t1);\ - toggle(&t5);\ - toggle(&t6);\ - ;\ - xor2(&b0, &t0);\ - xor2(&b1, &t1);\ - xor2(&b4, &t2);\ - xor2(&b6, &t3);\ - xor2(&b3, &t4);\ - xor2(&b7, &t5);\ - xor2(&b2, &t6);\ - xor2(&b5, &t7);\ - ;\ - rshift32_littleendian(&t0, 8);\ - rshift32_littleendian(&t1, 8);\ - rshift32_littleendian(&t2, 8);\ - rshift32_littleendian(&t3, 8);\ - rshift32_littleendian(&t4, 8);\ - rshift32_littleendian(&t5, 8);\ - rshift32_littleendian(&t6, 8);\ - rshift32_littleendian(&t7, 8);\ - ;\ - xor2(&b0, &t0);\ - xor2(&b1, &t1);\ - xor2(&b4, &t2);\ - xor2(&b6, &t3);\ - xor2(&b3, &t4);\ - xor2(&b7, &t5);\ - xor2(&b2, &t6);\ - xor2(&b5, &t7);\ - ;\ - rshift32_littleendian(&t0, 8);\ - rshift32_littleendian(&t1, 8);\ - rshift32_littleendian(&t2, 8);\ - rshift32_littleendian(&t3, 8);\ - rshift32_littleendian(&t4, 8);\ - rshift32_littleendian(&t5, 8);\ - rshift32_littleendian(&t6, 8);\ - rshift32_littleendian(&t7, 8);\ - ;\ - xor2(&b0, &t0);\ - xor2(&b1, &t1);\ - xor2(&b4, &t2);\ - xor2(&b6, &t3);\ - xor2(&b3, &t4);\ - xor2(&b7, &t5);\ - xor2(&b2, &t6);\ - xor2(&b5, &t7);\ - ;\ - rshift32_littleendian(&t0, 8);\ - rshift32_littleendian(&t1, 8);\ - rshift32_littleendian(&t2, 8);\ - rshift32_littleendian(&t3, 8);\ - rshift32_littleendian(&t4, 8);\ - rshift32_littleendian(&t5, 8);\ - rshift32_littleendian(&t6, 8);\ - rshift32_littleendian(&t7, 8);\ - ;\ - xor2(&b0, &t0);\ - xor2(&b1, &t1);\ - xor2(&b4, &t2);\ - xor2(&b6, &t3);\ - xor2(&b3, &t4);\ - xor2(&b7, &t5);\ - xor2(&b2, &t6);\ - xor2(&b5, &t7);\ - ;\ - shufb(&b0, M0);\ - shufb(&b1, M0);\ - shufb(&b2, M0);\ - shufb(&b3, M0);\ - shufb(&b4, M0);\ - shufb(&b5, M0);\ - shufb(&b6, M0);\ - shufb(&b7, M0);\ - ;\ - *(int128 *)(bskey + 1280) = b0;\ - *(int128 *)(bskey + 1296) = b1;\ - *(int128 *)(bskey + 1312) = b4;\ - *(int128 *)(bskey + 1328) = b6;\ - *(int128 *)(bskey + 1344) = b3;\ - *(int128 *)(bskey + 1360) = b7;\ - *(int128 *)(bskey + 1376) = b2;\ - *(int128 *)(bskey + 1392) = b5;\ - - -#define keyexpbs(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7, rcon, i, bskey) \ - toggle(&b0);\ - toggle(&b1);\ - toggle(&b5);\ - toggle(&b6);\ - rotbyte(&b0);\ - rotbyte(&b1);\ - rotbyte(&b2);\ - rotbyte(&b3);\ - rotbyte(&b4);\ - rotbyte(&b5);\ - rotbyte(&b6);\ - rotbyte(&b7);\ - ;\ - sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7);\ - ;\ - rcon;\ - shufb(&b0, EXPB0);\ - shufb(&b1, EXPB0);\ - shufb(&b4, EXPB0);\ - shufb(&b6, EXPB0);\ - shufb(&b3, EXPB0);\ - shufb(&b7, EXPB0);\ - shufb(&b2, EXPB0);\ - shufb(&b5, EXPB0);\ - ;\ - t0 = *(int128 *)(bskey + (i-1) * 128 + 0);\ - t1 = *(int128 *)(bskey + (i-1) * 128 + 16);\ - t2 = *(int128 *)(bskey + (i-1) * 128 + 32);\ - t3 = *(int128 *)(bskey + (i-1) * 128 + 48);\ - t4 = *(int128 *)(bskey + (i-1) * 128 + 64);\ - t5 = *(int128 *)(bskey + (i-1) * 128 + 80);\ - t6 = *(int128 *)(bskey + (i-1) * 128 + 96);\ - t7 = *(int128 *)(bskey + (i-1) * 128 + 112);\ - ;\ - toggle(&t0);\ - toggle(&t1);\ - toggle(&t5);\ - toggle(&t6);\ - ;\ - xor2(&b0, &t0);\ - xor2(&b1, &t1);\ - xor2(&b4, &t2);\ - xor2(&b6, &t3);\ - xor2(&b3, &t4);\ - xor2(&b7, &t5);\ - xor2(&b2, &t6);\ - xor2(&b5, &t7);\ - ;\ - rshift32_littleendian(&t0, 8);\ - rshift32_littleendian(&t1, 8);\ - rshift32_littleendian(&t2, 8);\ - rshift32_littleendian(&t3, 8);\ - rshift32_littleendian(&t4, 8);\ - rshift32_littleendian(&t5, 8);\ - rshift32_littleendian(&t6, 8);\ - rshift32_littleendian(&t7, 8);\ - ;\ - xor2(&b0, &t0);\ - xor2(&b1, &t1);\ - xor2(&b4, &t2);\ - xor2(&b6, &t3);\ - xor2(&b3, &t4);\ - xor2(&b7, &t5);\ - xor2(&b2, &t6);\ - xor2(&b5, &t7);\ - ;\ - rshift32_littleendian(&t0, 8);\ - rshift32_littleendian(&t1, 8);\ - rshift32_littleendian(&t2, 8);\ - rshift32_littleendian(&t3, 8);\ - rshift32_littleendian(&t4, 8);\ - rshift32_littleendian(&t5, 8);\ - rshift32_littleendian(&t6, 8);\ - rshift32_littleendian(&t7, 8);\ - ;\ - xor2(&b0, &t0);\ - xor2(&b1, &t1);\ - xor2(&b4, &t2);\ - xor2(&b6, &t3);\ - xor2(&b3, &t4);\ - xor2(&b7, &t5);\ - xor2(&b2, &t6);\ - xor2(&b5, &t7);\ - ;\ - rshift32_littleendian(&t0, 8);\ - rshift32_littleendian(&t1, 8);\ - rshift32_littleendian(&t2, 8);\ - rshift32_littleendian(&t3, 8);\ - rshift32_littleendian(&t4, 8);\ - rshift32_littleendian(&t5, 8);\ - rshift32_littleendian(&t6, 8);\ - rshift32_littleendian(&t7, 8);\ - ;\ - xor2(&b0, &t0);\ - xor2(&b1, &t1);\ - xor2(&b4, &t2);\ - xor2(&b6, &t3);\ - xor2(&b3, &t4);\ - xor2(&b7, &t5);\ - xor2(&b2, &t6);\ - xor2(&b5, &t7);\ - ;\ - *(int128 *)(bskey + i*128 + 0) = b0;\ - *(int128 *)(bskey + i*128 + 16) = b1;\ - *(int128 *)(bskey + i*128 + 32) = b4;\ - *(int128 *)(bskey + i*128 + 48) = b6;\ - *(int128 *)(bskey + i*128 + 64) = b3;\ - *(int128 *)(bskey + i*128 + 80) = b7;\ - *(int128 *)(bskey + i*128 + 96) = b2;\ - *(int128 *)(bskey + i*128 + 112) = b5;\ +#define keyexpbs1( \ + b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7, bskey) \ + rotbyte(&b0); \ + rotbyte(&b1); \ + rotbyte(&b2); \ + rotbyte(&b3); \ + rotbyte(&b4); \ + rotbyte(&b5); \ + rotbyte(&b6); \ + rotbyte(&b7); \ + ; \ + sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7); \ + ; \ + xor_rcon(&b0); \ + shufb(&b0, EXPB0); \ + shufb(&b1, EXPB0); \ + shufb(&b4, EXPB0); \ + shufb(&b6, EXPB0); \ + shufb(&b3, EXPB0); \ + shufb(&b7, EXPB0); \ + shufb(&b2, EXPB0); \ + shufb(&b5, EXPB0); \ + shufb(&b0, EXPB0); \ + ; \ + t0 = *(int128*)(bskey + 0); \ + t1 = *(int128*)(bskey + 16); \ + t2 = *(int128*)(bskey + 32); \ + t3 = *(int128*)(bskey + 48); \ + t4 = *(int128*)(bskey + 64); \ + t5 = *(int128*)(bskey + 80); \ + t6 = *(int128*)(bskey + 96); \ + t7 = *(int128*)(bskey + 112); \ + ; \ + xor2(&b0, &t0); \ + xor2(&b1, &t1); \ + xor2(&b4, &t2); \ + xor2(&b6, &t3); \ + xor2(&b3, &t4); \ + xor2(&b7, &t5); \ + xor2(&b2, &t6); \ + xor2(&b5, &t7); \ + ; \ + rshift32_littleendian(&t0, 8); \ + rshift32_littleendian(&t1, 8); \ + rshift32_littleendian(&t2, 8); \ + rshift32_littleendian(&t3, 8); \ + rshift32_littleendian(&t4, 8); \ + rshift32_littleendian(&t5, 8); \ + rshift32_littleendian(&t6, 8); \ + rshift32_littleendian(&t7, 8); \ + ; \ + xor2(&b0, &t0); \ + xor2(&b1, &t1); \ + xor2(&b4, &t2); \ + xor2(&b6, &t3); \ + xor2(&b3, &t4); \ + xor2(&b7, &t5); \ + xor2(&b2, &t6); \ + xor2(&b5, &t7); \ + ; \ + rshift32_littleendian(&t0, 8); \ + rshift32_littleendian(&t1, 8); \ + rshift32_littleendian(&t2, 8); \ + rshift32_littleendian(&t3, 8); \ + rshift32_littleendian(&t4, 8); \ + rshift32_littleendian(&t5, 8); \ + rshift32_littleendian(&t6, 8); \ + rshift32_littleendian(&t7, 8); \ + ; \ + xor2(&b0, &t0); \ + xor2(&b1, &t1); \ + xor2(&b4, &t2); \ + xor2(&b6, &t3); \ + xor2(&b3, &t4); \ + xor2(&b7, &t5); \ + xor2(&b2, &t6); \ + xor2(&b5, &t7); \ + ; \ + rshift32_littleendian(&t0, 8); \ + rshift32_littleendian(&t1, 8); \ + rshift32_littleendian(&t2, 8); \ + rshift32_littleendian(&t3, 8); \ + rshift32_littleendian(&t4, 8); \ + rshift32_littleendian(&t5, 8); \ + rshift32_littleendian(&t6, 8); \ + rshift32_littleendian(&t7, 8); \ + ; \ + xor2(&b0, &t0); \ + xor2(&b1, &t1); \ + xor2(&b4, &t2); \ + xor2(&b6, &t3); \ + xor2(&b3, &t4); \ + xor2(&b7, &t5); \ + xor2(&b2, &t6); \ + xor2(&b5, &t7); \ + ; \ + *(int128*)(bskey + 128) = b0; \ + *(int128*)(bskey + 144) = b1; \ + *(int128*)(bskey + 160) = b4; \ + *(int128*)(bskey + 176) = b6; \ + *(int128*)(bskey + 192) = b3; \ + *(int128*)(bskey + 208) = b7; \ + *(int128*)(bskey + 224) = b2; \ + *(int128*)(bskey + 240) = b5; + +#define keyexpbs10( \ + b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7, bskey) \ + ; \ + toggle(&b0); \ + toggle(&b1); \ + toggle(&b5); \ + toggle(&b6); \ + rotbyte(&b0); \ + rotbyte(&b1); \ + rotbyte(&b2); \ + rotbyte(&b3); \ + rotbyte(&b4); \ + rotbyte(&b5); \ + rotbyte(&b6); \ + rotbyte(&b7); \ + ; \ + sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7); \ + ; \ + xor_rcon(&b1); \ + xor_rcon(&b4); \ + xor_rcon(&b3); \ + xor_rcon(&b7); \ + shufb(&b0, EXPB0); \ + shufb(&b1, EXPB0); \ + shufb(&b4, EXPB0); \ + shufb(&b6, EXPB0); \ + shufb(&b3, EXPB0); \ + shufb(&b7, EXPB0); \ + shufb(&b2, EXPB0); \ + shufb(&b5, EXPB0); \ + ; \ + t0 = *(int128*)(bskey + 9 * 128 + 0); \ + t1 = *(int128*)(bskey + 9 * 128 + 16); \ + t2 = *(int128*)(bskey + 9 * 128 + 32); \ + t3 = *(int128*)(bskey + 9 * 128 + 48); \ + t4 = *(int128*)(bskey + 9 * 128 + 64); \ + t5 = *(int128*)(bskey + 9 * 128 + 80); \ + t6 = *(int128*)(bskey + 9 * 128 + 96); \ + t7 = *(int128*)(bskey + 9 * 128 + 112); \ + ; \ + toggle(&t0); \ + toggle(&t1); \ + toggle(&t5); \ + toggle(&t6); \ + ; \ + xor2(&b0, &t0); \ + xor2(&b1, &t1); \ + xor2(&b4, &t2); \ + xor2(&b6, &t3); \ + xor2(&b3, &t4); \ + xor2(&b7, &t5); \ + xor2(&b2, &t6); \ + xor2(&b5, &t7); \ + ; \ + rshift32_littleendian(&t0, 8); \ + rshift32_littleendian(&t1, 8); \ + rshift32_littleendian(&t2, 8); \ + rshift32_littleendian(&t3, 8); \ + rshift32_littleendian(&t4, 8); \ + rshift32_littleendian(&t5, 8); \ + rshift32_littleendian(&t6, 8); \ + rshift32_littleendian(&t7, 8); \ + ; \ + xor2(&b0, &t0); \ + xor2(&b1, &t1); \ + xor2(&b4, &t2); \ + xor2(&b6, &t3); \ + xor2(&b3, &t4); \ + xor2(&b7, &t5); \ + xor2(&b2, &t6); \ + xor2(&b5, &t7); \ + ; \ + rshift32_littleendian(&t0, 8); \ + rshift32_littleendian(&t1, 8); \ + rshift32_littleendian(&t2, 8); \ + rshift32_littleendian(&t3, 8); \ + rshift32_littleendian(&t4, 8); \ + rshift32_littleendian(&t5, 8); \ + rshift32_littleendian(&t6, 8); \ + rshift32_littleendian(&t7, 8); \ + ; \ + xor2(&b0, &t0); \ + xor2(&b1, &t1); \ + xor2(&b4, &t2); \ + xor2(&b6, &t3); \ + xor2(&b3, &t4); \ + xor2(&b7, &t5); \ + xor2(&b2, &t6); \ + xor2(&b5, &t7); \ + ; \ + rshift32_littleendian(&t0, 8); \ + rshift32_littleendian(&t1, 8); \ + rshift32_littleendian(&t2, 8); \ + rshift32_littleendian(&t3, 8); \ + rshift32_littleendian(&t4, 8); \ + rshift32_littleendian(&t5, 8); \ + rshift32_littleendian(&t6, 8); \ + rshift32_littleendian(&t7, 8); \ + ; \ + xor2(&b0, &t0); \ + xor2(&b1, &t1); \ + xor2(&b4, &t2); \ + xor2(&b6, &t3); \ + xor2(&b3, &t4); \ + xor2(&b7, &t5); \ + xor2(&b2, &t6); \ + xor2(&b5, &t7); \ + ; \ + shufb(&b0, M0); \ + shufb(&b1, M0); \ + shufb(&b2, M0); \ + shufb(&b3, M0); \ + shufb(&b4, M0); \ + shufb(&b5, M0); \ + shufb(&b6, M0); \ + shufb(&b7, M0); \ + ; \ + *(int128*)(bskey + 1280) = b0; \ + *(int128*)(bskey + 1296) = b1; \ + *(int128*)(bskey + 1312) = b4; \ + *(int128*)(bskey + 1328) = b6; \ + *(int128*)(bskey + 1344) = b3; \ + *(int128*)(bskey + 1360) = b7; \ + *(int128*)(bskey + 1376) = b2; \ + *(int128*)(bskey + 1392) = b5; + +#define keyexpbs(b0, \ + b1, \ + b2, \ + b3, \ + b4, \ + b5, \ + b6, \ + b7, \ + t0, \ + t1, \ + t2, \ + t3, \ + t4, \ + t5, \ + t6, \ + t7, \ + rcon, \ + i, \ + bskey) \ + toggle(&b0); \ + toggle(&b1); \ + toggle(&b5); \ + toggle(&b6); \ + rotbyte(&b0); \ + rotbyte(&b1); \ + rotbyte(&b2); \ + rotbyte(&b3); \ + rotbyte(&b4); \ + rotbyte(&b5); \ + rotbyte(&b6); \ + rotbyte(&b7); \ + ; \ + sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7); \ + ; \ + rcon; \ + shufb(&b0, EXPB0); \ + shufb(&b1, EXPB0); \ + shufb(&b4, EXPB0); \ + shufb(&b6, EXPB0); \ + shufb(&b3, EXPB0); \ + shufb(&b7, EXPB0); \ + shufb(&b2, EXPB0); \ + shufb(&b5, EXPB0); \ + ; \ + t0 = *(int128*)(bskey + (i - 1) * 128 + 0); \ + t1 = *(int128*)(bskey + (i - 1) * 128 + 16); \ + t2 = *(int128*)(bskey + (i - 1) * 128 + 32); \ + t3 = *(int128*)(bskey + (i - 1) * 128 + 48); \ + t4 = *(int128*)(bskey + (i - 1) * 128 + 64); \ + t5 = *(int128*)(bskey + (i - 1) * 128 + 80); \ + t6 = *(int128*)(bskey + (i - 1) * 128 + 96); \ + t7 = *(int128*)(bskey + (i - 1) * 128 + 112); \ + ; \ + toggle(&t0); \ + toggle(&t1); \ + toggle(&t5); \ + toggle(&t6); \ + ; \ + xor2(&b0, &t0); \ + xor2(&b1, &t1); \ + xor2(&b4, &t2); \ + xor2(&b6, &t3); \ + xor2(&b3, &t4); \ + xor2(&b7, &t5); \ + xor2(&b2, &t6); \ + xor2(&b5, &t7); \ + ; \ + rshift32_littleendian(&t0, 8); \ + rshift32_littleendian(&t1, 8); \ + rshift32_littleendian(&t2, 8); \ + rshift32_littleendian(&t3, 8); \ + rshift32_littleendian(&t4, 8); \ + rshift32_littleendian(&t5, 8); \ + rshift32_littleendian(&t6, 8); \ + rshift32_littleendian(&t7, 8); \ + ; \ + xor2(&b0, &t0); \ + xor2(&b1, &t1); \ + xor2(&b4, &t2); \ + xor2(&b6, &t3); \ + xor2(&b3, &t4); \ + xor2(&b7, &t5); \ + xor2(&b2, &t6); \ + xor2(&b5, &t7); \ + ; \ + rshift32_littleendian(&t0, 8); \ + rshift32_littleendian(&t1, 8); \ + rshift32_littleendian(&t2, 8); \ + rshift32_littleendian(&t3, 8); \ + rshift32_littleendian(&t4, 8); \ + rshift32_littleendian(&t5, 8); \ + rshift32_littleendian(&t6, 8); \ + rshift32_littleendian(&t7, 8); \ + ; \ + xor2(&b0, &t0); \ + xor2(&b1, &t1); \ + xor2(&b4, &t2); \ + xor2(&b6, &t3); \ + xor2(&b3, &t4); \ + xor2(&b7, &t5); \ + xor2(&b2, &t6); \ + xor2(&b5, &t7); \ + ; \ + rshift32_littleendian(&t0, 8); \ + rshift32_littleendian(&t1, 8); \ + rshift32_littleendian(&t2, 8); \ + rshift32_littleendian(&t3, 8); \ + rshift32_littleendian(&t4, 8); \ + rshift32_littleendian(&t5, 8); \ + rshift32_littleendian(&t6, 8); \ + rshift32_littleendian(&t7, 8); \ + ; \ + xor2(&b0, &t0); \ + xor2(&b1, &t1); \ + xor2(&b4, &t2); \ + xor2(&b6, &t3); \ + xor2(&b3, &t4); \ + xor2(&b7, &t5); \ + xor2(&b2, &t6); \ + xor2(&b5, &t7); \ + ; \ + *(int128*)(bskey + i * 128 + 0) = b0; \ + *(int128*)(bskey + i * 128 + 16) = b1; \ + *(int128*)(bskey + i * 128 + 32) = b4; \ + *(int128*)(bskey + i * 128 + 48) = b6; \ + *(int128*)(bskey + i * 128 + 64) = b3; \ + *(int128*)(bskey + i * 128 + 80) = b7; \ + *(int128*)(bskey + i * 128 + 96) = b2; \ + *(int128*)(bskey + i * 128 + 112) = b5; /* Macros used in multiple contexts */ -#define bitslicekey0(key, bskey) \ - xmm0 = *(const int128 *) (key + 0);\ - shufb(&xmm0, M0);\ - copy2(&xmm1, &xmm0);\ - copy2(&xmm2, &xmm0);\ - copy2(&xmm3, &xmm0);\ - copy2(&xmm4, &xmm0);\ - copy2(&xmm5, &xmm0);\ - copy2(&xmm6, &xmm0);\ - copy2(&xmm7, &xmm0);\ - ;\ - bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, t);\ - ;\ - *(int128 *) (bskey + 0) = xmm0;\ - *(int128 *) (bskey + 16) = xmm1;\ - *(int128 *) (bskey + 32) = xmm2;\ - *(int128 *) (bskey + 48) = xmm3;\ - *(int128 *) (bskey + 64) = xmm4;\ - *(int128 *) (bskey + 80) = xmm5;\ - *(int128 *) (bskey + 96) = xmm6;\ - *(int128 *) (bskey + 112) = xmm7;\ - - -#define bitslicekey10(key, bskey) \ - xmm0 = *(int128 *) (key + 0);\ - copy2(xmm1, xmm0);\ - copy2(xmm2, xmm0);\ - copy2(xmm3, xmm0);\ - copy2(xmm4, xmm0);\ - copy2(xmm5, xmm0);\ - copy2(xmm6, xmm0);\ - copy2(xmm7, xmm0);\ - ;\ - bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, t);\ - ;\ - toggle(&xmm6);\ - toggle(&xmm5);\ - toggle(&xmm1);\ - toggle(&xmm0);\ - ;\ - *(int128 *) (bskey + 0 + 1280) = xmm0;\ - *(int128 *) (bskey + 16 + 1280) = xmm1;\ - *(int128 *) (bskey + 32 + 1280) = xmm2;\ - *(int128 *) (bskey + 48 + 1280) = xmm3;\ - *(int128 *) (bskey + 64 + 1280) = xmm4;\ - *(int128 *) (bskey + 80 + 1280) = xmm5;\ - *(int128 *) (bskey + 96 + 1280) = xmm6;\ - *(int128 *) (bskey + 112 + 1280) = xmm7;\ - - -#define bitslicekey(i,key,bskey) \ - xmm0 = *(int128 *) (key + 0);\ - shufb(&xmm0, M0);\ - copy2(&xmm1, &xmm0);\ - copy2(&xmm2, &xmm0);\ - copy2(&xmm3, &xmm0);\ - copy2(&xmm4, &xmm0);\ - copy2(&xmm5, &xmm0);\ - copy2(&xmm6, &xmm0);\ - copy2(&xmm7, &xmm0);\ - ;\ - bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, t);\ - ;\ - toggle(&xmm6);\ - toggle(&xmm5);\ - toggle(&xmm1);\ - toggle(&xmm0);\ - ;\ - *(int128 *) (bskey + 0 + 128*i) = xmm0;\ - *(int128 *) (bskey + 16 + 128*i) = xmm1;\ - *(int128 *) (bskey + 32 + 128*i) = xmm2;\ - *(int128 *) (bskey + 48 + 128*i) = xmm3;\ - *(int128 *) (bskey + 64 + 128*i) = xmm4;\ - *(int128 *) (bskey + 80 + 128*i) = xmm5;\ - *(int128 *) (bskey + 96 + 128*i) = xmm6;\ - *(int128 *) (bskey + 112 + 128*i) = xmm7;\ - - -#define bitslice(x0, x1, x2, x3, x4, x5, x6, x7, t) \ - swapmove(x0, x1, 1, BS0, t);\ - swapmove(x2, x3, 1, BS0, t);\ - swapmove(x4, x5, 1, BS0, t);\ - swapmove(x6, x7, 1, BS0, t);\ - ;\ - swapmove(x0, x2, 2, BS1, t);\ - swapmove(x1, x3, 2, BS1, t);\ - swapmove(x4, x6, 2, BS1, t);\ - swapmove(x5, x7, 2, BS1, t);\ - ;\ - swapmove(x0, x4, 4, BS2, t);\ - swapmove(x1, x5, 4, BS2, t);\ - swapmove(x2, x6, 4, BS2, t);\ - swapmove(x3, x7, 4, BS2, t);\ - - -#define swapmove(a, b, n, m, t) \ - copy2(&t, &b);\ - rshift64_littleendian(&t, n);\ - xor2(&t, &a);\ - and2(&t, &m);\ - xor2(&a, &t);\ - lshift64_littleendian(&t, n);\ +#define bitslicekey0(key, bskey) \ + xmm0 = *(const int128*)(key + 0); \ + shufb(&xmm0, M0); \ + copy2(&xmm1, &xmm0); \ + copy2(&xmm2, &xmm0); \ + copy2(&xmm3, &xmm0); \ + copy2(&xmm4, &xmm0); \ + copy2(&xmm5, &xmm0); \ + copy2(&xmm6, &xmm0); \ + copy2(&xmm7, &xmm0); \ + ; \ + bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, t); \ + ; \ + *(int128*)(bskey + 0) = xmm0; \ + *(int128*)(bskey + 16) = xmm1; \ + *(int128*)(bskey + 32) = xmm2; \ + *(int128*)(bskey + 48) = xmm3; \ + *(int128*)(bskey + 64) = xmm4; \ + *(int128*)(bskey + 80) = xmm5; \ + *(int128*)(bskey + 96) = xmm6; \ + *(int128*)(bskey + 112) = xmm7; + +#define bitslicekey10(key, bskey) \ + xmm0 = *(int128*)(key + 0); \ + copy2(xmm1, xmm0); \ + copy2(xmm2, xmm0); \ + copy2(xmm3, xmm0); \ + copy2(xmm4, xmm0); \ + copy2(xmm5, xmm0); \ + copy2(xmm6, xmm0); \ + copy2(xmm7, xmm0); \ + ; \ + bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, t); \ + ; \ + toggle(&xmm6); \ + toggle(&xmm5); \ + toggle(&xmm1); \ + toggle(&xmm0); \ + ; \ + *(int128*)(bskey + 0 + 1280) = xmm0; \ + *(int128*)(bskey + 16 + 1280) = xmm1; \ + *(int128*)(bskey + 32 + 1280) = xmm2; \ + *(int128*)(bskey + 48 + 1280) = xmm3; \ + *(int128*)(bskey + 64 + 1280) = xmm4; \ + *(int128*)(bskey + 80 + 1280) = xmm5; \ + *(int128*)(bskey + 96 + 1280) = xmm6; \ + *(int128*)(bskey + 112 + 1280) = xmm7; + +#define bitslicekey(i, key, bskey) \ + xmm0 = *(int128*)(key + 0); \ + shufb(&xmm0, M0); \ + copy2(&xmm1, &xmm0); \ + copy2(&xmm2, &xmm0); \ + copy2(&xmm3, &xmm0); \ + copy2(&xmm4, &xmm0); \ + copy2(&xmm5, &xmm0); \ + copy2(&xmm6, &xmm0); \ + copy2(&xmm7, &xmm0); \ + ; \ + bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, t); \ + ; \ + toggle(&xmm6); \ + toggle(&xmm5); \ + toggle(&xmm1); \ + toggle(&xmm0); \ + ; \ + *(int128*)(bskey + 0 + 128 * i) = xmm0; \ + *(int128*)(bskey + 16 + 128 * i) = xmm1; \ + *(int128*)(bskey + 32 + 128 * i) = xmm2; \ + *(int128*)(bskey + 48 + 128 * i) = xmm3; \ + *(int128*)(bskey + 64 + 128 * i) = xmm4; \ + *(int128*)(bskey + 80 + 128 * i) = xmm5; \ + *(int128*)(bskey + 96 + 128 * i) = xmm6; \ + *(int128*)(bskey + 112 + 128 * i) = xmm7; + +#define bitslice(x0, x1, x2, x3, x4, x5, x6, x7, t) \ + swapmove(x0, x1, 1, BS0, t); \ + swapmove(x2, x3, 1, BS0, t); \ + swapmove(x4, x5, 1, BS0, t); \ + swapmove(x6, x7, 1, BS0, t); \ + ; \ + swapmove(x0, x2, 2, BS1, t); \ + swapmove(x1, x3, 2, BS1, t); \ + swapmove(x4, x6, 2, BS1, t); \ + swapmove(x5, x7, 2, BS1, t); \ + ; \ + swapmove(x0, x4, 4, BS2, t); \ + swapmove(x1, x5, 4, BS2, t); \ + swapmove(x2, x6, 4, BS2, t); \ + swapmove(x3, x7, 4, BS2, t); + +#define swapmove(a, b, n, m, t) \ + copy2(&t, &b); \ + rshift64_littleendian(&t, n); \ + xor2(&t, &a); \ + and2(&t, &m); \ + xor2(&a, &t); \ + lshift64_littleendian(&t, n); \ xor2(&b, &t); -#define rotbyte(x) \ - shufb(x, ROTB) /* TODO: Make faster */ - +#define rotbyte(x) shufb(x, ROTB) /* TODO: Make faster */ /* Macros used for encryption (and decryption) */ -#define shiftrows(x0, x1, x2, x3, x4, x5, x6, x7, i, M, bskey) \ - xor2(&x0, (const int128 *)(bskey + 128*(i-1) + 0));\ - shufb(&x0, M);\ - xor2(&x1, (const int128 *)(bskey + 128*(i-1) + 16));\ - shufb(&x1, M);\ - xor2(&x2, (const int128 *)(bskey + 128*(i-1) + 32));\ - shufb(&x2, M);\ - xor2(&x3, (const int128 *)(bskey + 128*(i-1) + 48));\ - shufb(&x3, M);\ - xor2(&x4, (const int128 *)(bskey + 128*(i-1) + 64));\ - shufb(&x4, M);\ - xor2(&x5, (const int128 *)(bskey + 128*(i-1) + 80));\ - shufb(&x5, M);\ - xor2(&x6, (const int128 *)(bskey + 128*(i-1) + 96));\ - shufb(&x6, M);\ - xor2(&x7, (const int128 *)(bskey + 128*(i-1) + 112));\ - shufb(&x7, M);\ - - -#define mixcolumns(x0, x1, x2, x3, x4, x5, x6, x7, t0, t1, t2, t3, t4, t5, t6, t7) \ - shufd(&t0, &x0, 0x93);\ - shufd(&t1, &x1, 0x93);\ - shufd(&t2, &x2, 0x93);\ - shufd(&t3, &x3, 0x93);\ - shufd(&t4, &x4, 0x93);\ - shufd(&t5, &x5, 0x93);\ - shufd(&t6, &x6, 0x93);\ - shufd(&t7, &x7, 0x93);\ - ;\ - xor2(&x0, &t0);\ - xor2(&x1, &t1);\ - xor2(&x2, &t2);\ - xor2(&x3, &t3);\ - xor2(&x4, &t4);\ - xor2(&x5, &t5);\ - xor2(&x6, &t6);\ - xor2(&x7, &t7);\ - ;\ - xor2(&t0, &x7);\ - xor2(&t1, &x0);\ - xor2(&t2, &x1);\ - xor2(&t1, &x7);\ - xor2(&t3, &x2);\ - xor2(&t4, &x3);\ - xor2(&t5, &x4);\ - xor2(&t3, &x7);\ - xor2(&t6, &x5);\ - xor2(&t7, &x6);\ - xor2(&t4, &x7);\ - ;\ - shufd(&x0, &x0, 0x4e);\ - shufd(&x1, &x1, 0x4e);\ - shufd(&x2, &x2, 0x4e);\ - shufd(&x3, &x3, 0x4e);\ - shufd(&x4, &x4, 0x4e);\ - shufd(&x5, &x5, 0x4e);\ - shufd(&x6, &x6, 0x4e);\ - shufd(&x7, &x7, 0x4e);\ - ;\ - xor2(&t0, &x0);\ - xor2(&t1, &x1);\ - xor2(&t2, &x2);\ - xor2(&t3, &x3);\ - xor2(&t4, &x4);\ - xor2(&t5, &x5);\ - xor2(&t6, &x6);\ - xor2(&t7, &x7);\ - - -#define aesround(i, b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7, bskey) \ - shiftrows(b0, b1, b2, b3, b4, b5, b6, b7, i, SR, bskey);\ - sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7);\ - mixcolumns(b0, b1, b4, b6, b3, b7, b2, b5, t0, t1, t2, t3, t4, t5, t6, t7);\ - - -#define lastround(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7, bskey) \ - shiftrows(b0, b1, b2, b3, b4, b5, b6, b7, 10, SRM0, bskey);\ - sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7);\ - xor2(&b0,(const int128 *)(bskey + 128*10));\ - xor2(&b1,(const int128 *)(bskey + 128*10+16));\ - xor2(&b4,(const int128 *)(bskey + 128*10+32));\ - xor2(&b6,(const int128 *)(bskey + 128*10+48));\ - xor2(&b3,(const int128 *)(bskey + 128*10+64));\ - xor2(&b7,(const int128 *)(bskey + 128*10+80));\ - xor2(&b2,(const int128 *)(bskey + 128*10+96));\ - xor2(&b5,(const int128 *)(bskey + 128*10+112));\ - - -#define sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, s0, s1, s2, s3) \ - InBasisChange(b0, b1, b2, b3, b4, b5, b6, b7); \ - Inv_GF256(b6, b5, b0, b3, b7, b1, b4, b2, t0, t1, t2, t3, s0, s1, s2, s3); \ - OutBasisChange(b7, b1, b4, b2, b6, b5, b0, b3); \ - - -#define InBasisChange(b0, b1, b2, b3, b4, b5, b6, b7) \ - xor2(&b5, &b6);\ - xor2(&b2, &b1);\ - xor2(&b5, &b0);\ - xor2(&b6, &b2);\ - xor2(&b3, &b0);\ - ;\ - xor2(&b6, &b3);\ - xor2(&b3, &b7);\ - xor2(&b3, &b4);\ - xor2(&b7, &b5);\ - xor2(&b3, &b1);\ - ;\ - xor2(&b4, &b5);\ - xor2(&b2, &b7);\ - xor2(&b1, &b5);\ - -#define OutBasisChange(b0, b1, b2, b3, b4, b5, b6, b7) \ - xor2(&b0, &b6);\ - xor2(&b1, &b4);\ - xor2(&b2, &b0);\ - xor2(&b4, &b6);\ - xor2(&b6, &b1);\ - ;\ - xor2(&b1, &b5);\ - xor2(&b5, &b3);\ - xor2(&b2, &b5);\ - xor2(&b3, &b7);\ - xor2(&b7, &b5);\ - ;\ - xor2(&b4, &b7);\ - -#define Mul_GF4(x0, x1, y0, y1, t0) \ - copy2(&t0, &y0);\ - xor2(&t0, &y1);\ - and2(&t0, &x0);\ - xor2(&x0, &x1);\ - and2(&x0, &y1);\ - and2(&x1, &y0);\ - xor2(&x0, &x1);\ - xor2(&x1, &t0);\ - -#define Mul_GF4_N(x0, x1, y0, y1, t0) \ - copy2(&t0, &y0);\ - xor2(&t0, &y1);\ - and2(&t0, &x0);\ - xor2(&x0, &x1);\ - and2(&x0, &y1);\ - and2(&x1, &y0);\ - xor2(&x1, &x0);\ - xor2(&x0, &t0);\ - -#define Mul_GF4_2(x0, x1, x2, x3, y0, y1, t0, t1) \ - copy2(&t0, = y0);\ - xor2(&t0, &y1);\ - copy2(&t1, &t0);\ - and2(&t0, &x0);\ - and2(&t1, &x2);\ - xor2(&x0, &x1);\ - xor2(&x2, &x3);\ - and2(&x0, &y1);\ - and2(&x2, &y1);\ - and2(&x1, &y0);\ - and2(&x3, &y0);\ - xor2(&x0, &x1);\ - xor2(&x2, &x3);\ - xor2(&x1, &t0);\ - xor2(&x3, &t1);\ - -#define Mul_GF16(x0, x1, x2, x3, y0, y1, y2, y3, t0, t1, t2, t3) \ - copy2(&t0, &x0);\ - copy2(&t1, &x1);\ - Mul_GF4(x0, x1, y0, y1, t2);\ - xor2(&t0, &x2);\ - xor2(&t1, &x3);\ - xor2(&y0, &y2);\ - xor2(&y1, &y3);\ - Mul_GF4_N(t0, t1, y0, y1, t2);\ - Mul_GF4(x2, x3, y2, y3, t3);\ - ;\ - xor2(&x0, &t0);\ - xor2(&x2, &t0);\ - xor2(&x1, &t1);\ - xor2(&x3, &t1);\ - -#define Mul_GF16_2(x0, x1, x2, x3, x4, x5, x6, x7, y0, y1, y2, y3, t0, t1, t2, t3) \ - copy2(&t0, &x0);\ - copy2(&t1, &x1);\ - Mul_GF4(x0, x1, y0, y1, t2);\ - xor2(&t0, &x2);\ - xor2(&t1, &x3);\ - xor2(&y0, &y2);\ - xor2(&y1, &y3);\ - Mul_GF4_N(t0, t1, y0, y1, t3);\ - Mul_GF4(x2, x3, y2, y3, t2);\ - ;\ - xor2(&x0, &t0);\ - xor2(&x2, &t0);\ - xor2(&x1, &t1);\ - xor2(&x3, &t1);\ - ;\ - copy2(&t0, &x4);\ - copy2(&t1, &x5);\ - xor2(&t0, &x6);\ - xor2(&t1, &x7);\ - Mul_GF4_N(t0, t1, y0, y1, t3);\ - Mul_GF4(x6, x7, y2, y3, t2);\ - xor2(&y0, &y2);\ - xor2(&y1, &y3);\ - Mul_GF4(x4, x5, y0, y1, t3);\ - ;\ - xor2(&x4, &t0);\ - xor2(&x6, &t0);\ - xor2(&x5, &t1);\ - xor2(&x7, &t1);\ - -#define Inv_GF16(x0, x1, x2, x3, t0, t1, t2, t3) \ - copy2(&t0, &x1);\ - copy2(&t1, &x0);\ - and2(&t0, &x3);\ - or2(&t1, &x2);\ - copy2(&t2, &x1);\ - copy2(&t3, &x0);\ - or2(&t2, &x2);\ - or2(&t3, &x3);\ - xor2(&t2, &t3);\ - ;\ - xor2(&t0, &t2);\ - xor2(&t1, &t2);\ - ;\ - Mul_GF4_2(x0, x1, x2, x3, t1, t0, t2, t3);\ - - -#define Inv_GF256(x0, x1, x2, x3, x4, x5, x6, x7, t0, t1, t2, t3, s0, s1, s2, s3) \ - copy2(&t3, &x4);\ - copy2(&t2, &x5);\ - copy2(&t1, &x1);\ - copy2(&s1, &x7);\ - copy2(&s0, &x0);\ - ;\ - xor2(&t3, &x6);\ - xor2(&t2, &x7);\ - xor2(&t1, &x3);\ - xor2(&s1, &x6);\ - xor2(&s0, &x2);\ - ;\ - copy2(&s2, &t3);\ - copy2(&t0, &t2);\ - copy2(&s3, &t3);\ - ;\ - or2(&t2, &t1);\ - or2(&t3, &s0);\ - xor2(&s3, &t0);\ - and2(&s2, &s0);\ - and2(&t0, &t1);\ - xor2(&s0, &t1);\ - and2(&s3, &s0);\ - copy2(&s0, &x3);\ - xor2(&s0, &x2);\ - and2(&s1, &s0);\ - xor2(&t3, &s1);\ - xor2(&t2, &s1);\ - copy2(&s1, &x4);\ - xor2(&s1, &x5);\ - copy2(&s0, &x1);\ - copy2(&t1, &s1);\ - xor2(&s0, &x0);\ - or2(&t1, &s0);\ - and2(&s1, &s0);\ - xor2(&t0, &s1);\ - xor2(&t3, &s3);\ - xor2(&t2, &s2);\ - xor2(&t1, &s3);\ - xor2(&t0, &s2);\ - xor2(&t1, &s2);\ - copy2(&s0, &x7);\ - copy2(&s1, &x6);\ - copy2(&s2, &x5);\ - copy2(&s3, &x4);\ - and2(&s0, &x3);\ - and2(&s1, &x2);\ - and2(&s2, &x1);\ - or2(&s3, &x0);\ - xor2(&t3, &s0);\ - xor2(&t2, &s1);\ - xor2(&t1, &s2);\ - xor2(&t0, &s3);\ - ;\ - copy2(&s0, &t3);\ - xor2(&s0, &t2);\ - and2(&t3, &t1);\ - copy2(&s2, &t0);\ - xor2(&s2, &t3);\ - copy2(&s3, &s0);\ - and2(&s3, &s2);\ - xor2(&s3, &t2);\ - copy2(&s1, &t1);\ - xor2(&s1, &t0);\ - xor2(&t3, &t2);\ - and2(&s1, &t3);\ - xor2(&s1, &t0);\ - xor2(&t1, &s1);\ - copy2(&t2, &s2);\ - xor2(&t2, &s1);\ - and2(&t2, &t0);\ - xor2(&t1, &t2);\ - xor2(&s2, &t2);\ - and2(&s2, &s3);\ - xor2(&s2, &s0);\ - ;\ - Mul_GF16_2(x0, x1, x2, x3, x4, x5, x6, x7, s3, s2, s1, t1, s0, t0, t2, t3);\ +#define shiftrows(x0, x1, x2, x3, x4, x5, x6, x7, i, M, bskey) \ + xor2(&x0, (const int128*)(bskey + 128 * (i - 1) + 0)); \ + shufb(&x0, M); \ + xor2(&x1, (const int128*)(bskey + 128 * (i - 1) + 16)); \ + shufb(&x1, M); \ + xor2(&x2, (const int128*)(bskey + 128 * (i - 1) + 32)); \ + shufb(&x2, M); \ + xor2(&x3, (const int128*)(bskey + 128 * (i - 1) + 48)); \ + shufb(&x3, M); \ + xor2(&x4, (const int128*)(bskey + 128 * (i - 1) + 64)); \ + shufb(&x4, M); \ + xor2(&x5, (const int128*)(bskey + 128 * (i - 1) + 80)); \ + shufb(&x5, M); \ + xor2(&x6, (const int128*)(bskey + 128 * (i - 1) + 96)); \ + shufb(&x6, M); \ + xor2(&x7, (const int128*)(bskey + 128 * (i - 1) + 112)); \ + shufb(&x7, M); + +#define mixcolumns( \ + x0, x1, x2, x3, x4, x5, x6, x7, t0, t1, t2, t3, t4, t5, t6, t7) \ + shufd(&t0, &x0, 0x93); \ + shufd(&t1, &x1, 0x93); \ + shufd(&t2, &x2, 0x93); \ + shufd(&t3, &x3, 0x93); \ + shufd(&t4, &x4, 0x93); \ + shufd(&t5, &x5, 0x93); \ + shufd(&t6, &x6, 0x93); \ + shufd(&t7, &x7, 0x93); \ + ; \ + xor2(&x0, &t0); \ + xor2(&x1, &t1); \ + xor2(&x2, &t2); \ + xor2(&x3, &t3); \ + xor2(&x4, &t4); \ + xor2(&x5, &t5); \ + xor2(&x6, &t6); \ + xor2(&x7, &t7); \ + ; \ + xor2(&t0, &x7); \ + xor2(&t1, &x0); \ + xor2(&t2, &x1); \ + xor2(&t1, &x7); \ + xor2(&t3, &x2); \ + xor2(&t4, &x3); \ + xor2(&t5, &x4); \ + xor2(&t3, &x7); \ + xor2(&t6, &x5); \ + xor2(&t7, &x6); \ + xor2(&t4, &x7); \ + ; \ + shufd(&x0, &x0, 0x4e); \ + shufd(&x1, &x1, 0x4e); \ + shufd(&x2, &x2, 0x4e); \ + shufd(&x3, &x3, 0x4e); \ + shufd(&x4, &x4, 0x4e); \ + shufd(&x5, &x5, 0x4e); \ + shufd(&x6, &x6, 0x4e); \ + shufd(&x7, &x7, 0x4e); \ + ; \ + xor2(&t0, &x0); \ + xor2(&t1, &x1); \ + xor2(&t2, &x2); \ + xor2(&t3, &x3); \ + xor2(&t4, &x4); \ + xor2(&t5, &x5); \ + xor2(&t6, &x6); \ + xor2(&t7, &x7); + +#define aesround( \ + i, b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7, bskey) \ + shiftrows(b0, b1, b2, b3, b4, b5, b6, b7, i, SR, bskey); \ + sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7); \ + mixcolumns(b0, b1, b4, b6, b3, b7, b2, b5, t0, t1, t2, t3, t4, t5, t6, t7); + +#define lastround( \ + b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7, bskey) \ + shiftrows(b0, b1, b2, b3, b4, b5, b6, b7, 10, SRM0, bskey); \ + sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7); \ + xor2(&b0, (const int128*)(bskey + 128 * 10)); \ + xor2(&b1, (const int128*)(bskey + 128 * 10 + 16)); \ + xor2(&b4, (const int128*)(bskey + 128 * 10 + 32)); \ + xor2(&b6, (const int128*)(bskey + 128 * 10 + 48)); \ + xor2(&b3, (const int128*)(bskey + 128 * 10 + 64)); \ + xor2(&b7, (const int128*)(bskey + 128 * 10 + 80)); \ + xor2(&b2, (const int128*)(bskey + 128 * 10 + 96)); \ + xor2(&b5, (const int128*)(bskey + 128 * 10 + 112)); + +#define sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, s0, s1, s2, s3) \ + InBasisChange(b0, b1, b2, b3, b4, b5, b6, b7); \ + Inv_GF256(b6, b5, b0, b3, b7, b1, b4, b2, t0, t1, t2, t3, s0, s1, s2, s3); \ + OutBasisChange(b7, b1, b4, b2, b6, b5, b0, b3); + +#define InBasisChange(b0, b1, b2, b3, b4, b5, b6, b7) \ + xor2(&b5, &b6); \ + xor2(&b2, &b1); \ + xor2(&b5, &b0); \ + xor2(&b6, &b2); \ + xor2(&b3, &b0); \ + ; \ + xor2(&b6, &b3); \ + xor2(&b3, &b7); \ + xor2(&b3, &b4); \ + xor2(&b7, &b5); \ + xor2(&b3, &b1); \ + ; \ + xor2(&b4, &b5); \ + xor2(&b2, &b7); \ + xor2(&b1, &b5); + +#define OutBasisChange(b0, b1, b2, b3, b4, b5, b6, b7) \ + xor2(&b0, &b6); \ + xor2(&b1, &b4); \ + xor2(&b2, &b0); \ + xor2(&b4, &b6); \ + xor2(&b6, &b1); \ + ; \ + xor2(&b1, &b5); \ + xor2(&b5, &b3); \ + xor2(&b2, &b5); \ + xor2(&b3, &b7); \ + xor2(&b7, &b5); \ + ; \ + xor2(&b4, &b7); + +#define Mul_GF4(x0, x1, y0, y1, t0) \ + copy2(&t0, &y0); \ + xor2(&t0, &y1); \ + and2(&t0, &x0); \ + xor2(&x0, &x1); \ + and2(&x0, &y1); \ + and2(&x1, &y0); \ + xor2(&x0, &x1); \ + xor2(&x1, &t0); + +#define Mul_GF4_N(x0, x1, y0, y1, t0) \ + copy2(&t0, &y0); \ + xor2(&t0, &y1); \ + and2(&t0, &x0); \ + xor2(&x0, &x1); \ + and2(&x0, &y1); \ + and2(&x1, &y0); \ + xor2(&x1, &x0); \ + xor2(&x0, &t0); + +#define Mul_GF4_2(x0, x1, x2, x3, y0, y1, t0, t1) \ + copy2(&t0, = y0); \ + xor2(&t0, &y1); \ + copy2(&t1, &t0); \ + and2(&t0, &x0); \ + and2(&t1, &x2); \ + xor2(&x0, &x1); \ + xor2(&x2, &x3); \ + and2(&x0, &y1); \ + and2(&x2, &y1); \ + and2(&x1, &y0); \ + and2(&x3, &y0); \ + xor2(&x0, &x1); \ + xor2(&x2, &x3); \ + xor2(&x1, &t0); \ + xor2(&x3, &t1); + +#define Mul_GF16(x0, x1, x2, x3, y0, y1, y2, y3, t0, t1, t2, t3) \ + copy2(&t0, &x0); \ + copy2(&t1, &x1); \ + Mul_GF4(x0, x1, y0, y1, t2); \ + xor2(&t0, &x2); \ + xor2(&t1, &x3); \ + xor2(&y0, &y2); \ + xor2(&y1, &y3); \ + Mul_GF4_N(t0, t1, y0, y1, t2); \ + Mul_GF4(x2, x3, y2, y3, t3); \ + ; \ + xor2(&x0, &t0); \ + xor2(&x2, &t0); \ + xor2(&x1, &t1); \ + xor2(&x3, &t1); + +#define Mul_GF16_2( \ + x0, x1, x2, x3, x4, x5, x6, x7, y0, y1, y2, y3, t0, t1, t2, t3) \ + copy2(&t0, &x0); \ + copy2(&t1, &x1); \ + Mul_GF4(x0, x1, y0, y1, t2); \ + xor2(&t0, &x2); \ + xor2(&t1, &x3); \ + xor2(&y0, &y2); \ + xor2(&y1, &y3); \ + Mul_GF4_N(t0, t1, y0, y1, t3); \ + Mul_GF4(x2, x3, y2, y3, t2); \ + ; \ + xor2(&x0, &t0); \ + xor2(&x2, &t0); \ + xor2(&x1, &t1); \ + xor2(&x3, &t1); \ + ; \ + copy2(&t0, &x4); \ + copy2(&t1, &x5); \ + xor2(&t0, &x6); \ + xor2(&t1, &x7); \ + Mul_GF4_N(t0, t1, y0, y1, t3); \ + Mul_GF4(x6, x7, y2, y3, t2); \ + xor2(&y0, &y2); \ + xor2(&y1, &y3); \ + Mul_GF4(x4, x5, y0, y1, t3); \ + ; \ + xor2(&x4, &t0); \ + xor2(&x6, &t0); \ + xor2(&x5, &t1); \ + xor2(&x7, &t1); + +#define Inv_GF16(x0, x1, x2, x3, t0, t1, t2, t3) \ + copy2(&t0, &x1); \ + copy2(&t1, &x0); \ + and2(&t0, &x3); \ + or2(&t1, &x2); \ + copy2(&t2, &x1); \ + copy2(&t3, &x0); \ + or2(&t2, &x2); \ + or2(&t3, &x3); \ + xor2(&t2, &t3); \ + ; \ + xor2(&t0, &t2); \ + xor2(&t1, &t2); \ + ; \ + Mul_GF4_2(x0, x1, x2, x3, t1, t0, t2, t3); + +#define Inv_GF256( \ + x0, x1, x2, x3, x4, x5, x6, x7, t0, t1, t2, t3, s0, s1, s2, s3) \ + copy2(&t3, &x4); \ + copy2(&t2, &x5); \ + copy2(&t1, &x1); \ + copy2(&s1, &x7); \ + copy2(&s0, &x0); \ + ; \ + xor2(&t3, &x6); \ + xor2(&t2, &x7); \ + xor2(&t1, &x3); \ + xor2(&s1, &x6); \ + xor2(&s0, &x2); \ + ; \ + copy2(&s2, &t3); \ + copy2(&t0, &t2); \ + copy2(&s3, &t3); \ + ; \ + or2(&t2, &t1); \ + or2(&t3, &s0); \ + xor2(&s3, &t0); \ + and2(&s2, &s0); \ + and2(&t0, &t1); \ + xor2(&s0, &t1); \ + and2(&s3, &s0); \ + copy2(&s0, &x3); \ + xor2(&s0, &x2); \ + and2(&s1, &s0); \ + xor2(&t3, &s1); \ + xor2(&t2, &s1); \ + copy2(&s1, &x4); \ + xor2(&s1, &x5); \ + copy2(&s0, &x1); \ + copy2(&t1, &s1); \ + xor2(&s0, &x0); \ + or2(&t1, &s0); \ + and2(&s1, &s0); \ + xor2(&t0, &s1); \ + xor2(&t3, &s3); \ + xor2(&t2, &s2); \ + xor2(&t1, &s3); \ + xor2(&t0, &s2); \ + xor2(&t1, &s2); \ + copy2(&s0, &x7); \ + copy2(&s1, &x6); \ + copy2(&s2, &x5); \ + copy2(&s3, &x4); \ + and2(&s0, &x3); \ + and2(&s1, &x2); \ + and2(&s2, &x1); \ + or2(&s3, &x0); \ + xor2(&t3, &s0); \ + xor2(&t2, &s1); \ + xor2(&t1, &s2); \ + xor2(&t0, &s3); \ + ; \ + copy2(&s0, &t3); \ + xor2(&s0, &t2); \ + and2(&t3, &t1); \ + copy2(&s2, &t0); \ + xor2(&s2, &t3); \ + copy2(&s3, &s0); \ + and2(&s3, &s2); \ + xor2(&s3, &t2); \ + copy2(&s1, &t1); \ + xor2(&s1, &t0); \ + xor2(&t3, &t2); \ + and2(&s1, &t3); \ + xor2(&s1, &t0); \ + xor2(&t1, &s1); \ + copy2(&t2, &s2); \ + xor2(&t2, &s1); \ + and2(&t2, &t0); \ + xor2(&t1, &t2); \ + xor2(&s2, &t2); \ + and2(&s2, &s3); \ + xor2(&s2, &s0); \ + ; \ + Mul_GF16_2(x0, x1, x2, x3, x4, x5, x6, x7, s3, s2, s1, t1, s0, t0, t2, t3); #endif diff --git a/examples/aesbitsliced/common_aes128ctr.c b/examples/aesbitsliced/common_aes128ctr.c index 14a28cc..b35ae00 100644 --- a/examples/aesbitsliced/common_aes128ctr.c +++ b/examples/aesbitsliced/common_aes128ctr.c @@ -1,64 +1,68 @@ #include "common.h" -uint32 load32_bigendian(const unsigned char *x) +uint32 +load32_bigendian(const unsigned char* x) { - return - (uint32) (x[3]) \ - | (((uint32) (x[2])) << 8) \ - | (((uint32) (x[1])) << 16) \ - | (((uint32) (x[0])) << 24) - ; + return (uint32)(x[3]) | (((uint32)(x[2])) << 8) | (((uint32)(x[1])) << 16) | + (((uint32)(x[0])) << 24); } -void store32_bigendian(unsigned char *x,uint32 u) +void +store32_bigendian(unsigned char* x, uint32 u) { - x[3] = u; u >>= 8; - x[2] = u; u >>= 8; - x[1] = u; u >>= 8; + x[3] = u; + u >>= 8; + x[2] = u; + u >>= 8; + x[1] = u; + u >>= 8; x[0] = u; } -uint32 load32_littleendian(const unsigned char *x) +uint32 +load32_littleendian(const unsigned char* x) { - return - (uint32) (x[0]) \ - | (((uint32) (x[1])) << 8) \ - | (((uint32) (x[2])) << 16) \ - | (((uint32) (x[3])) << 24) - ; + return (uint32)(x[0]) | (((uint32)(x[1])) << 8) | (((uint32)(x[2])) << 16) | + (((uint32)(x[3])) << 24); } -void store32_littleendian(unsigned char *x,uint32 u) +void +store32_littleendian(unsigned char* x, uint32 u) { - x[0] = u; u >>= 8; - x[1] = u; u >>= 8; - x[2] = u; u >>= 8; + x[0] = u; + u >>= 8; + x[1] = u; + u >>= 8; + x[2] = u; + u >>= 8; x[3] = u; } - -uint64 load64_littleendian(const unsigned char *x) +uint64 +load64_littleendian(const unsigned char* x) { - return - (uint64) (x[0]) \ - | (((uint64) (x[1])) << 8) \ - | (((uint64) (x[2])) << 16) \ - | (((uint64) (x[3])) << 24) - | (((uint64) (x[4])) << 32) - | (((uint64) (x[5])) << 40) - | (((uint64) (x[6])) << 48) - | (((uint64) (x[7])) << 56) - ; + return (uint64)(x[0]) | (((uint64)(x[1])) << 8) | (((uint64)(x[2])) << 16) | + (((uint64)(x[3])) << 24) | (((uint64)(x[4])) << 32) | + (((uint64)(x[5])) << 40) | (((uint64)(x[6])) << 48) | + (((uint64)(x[7])) << 56); } -void store64_littleendian(unsigned char *x,uint64 u) +void +store64_littleendian(unsigned char* x, uint64 u) { - x[0] = u; u >>= 8; - x[1] = u; u >>= 8; - x[2] = u; u >>= 8; - x[3] = u; u >>= 8; - x[4] = u; u >>= 8; - x[5] = u; u >>= 8; - x[6] = u; u >>= 8; + x[0] = u; + u >>= 8; + x[1] = u; + u >>= 8; + x[2] = u; + u >>= 8; + x[3] = u; + u >>= 8; + x[4] = u; + u >>= 8; + x[5] = u; + u >>= 8; + x[6] = u; + u >>= 8; x[7] = u; } diff --git a/examples/aesbitsliced/consts_aes128ctr.c b/examples/aesbitsliced/consts_aes128ctr.c index f8029b8..fbecb47 100644 --- a/examples/aesbitsliced/consts_aes128ctr.c +++ b/examples/aesbitsliced/consts_aes128ctr.c @@ -1,14 +1,28 @@ #include "consts.h" -const unsigned char ROTB[16] = {0x00, 0x00, 0x00, 0x0c, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x04, 0x00, 0x00, 0x00, 0x08}; -const unsigned char M0[16] = {0x0f, 0x0b, 0x07, 0x03, 0x0e, 0x0a, 0x06, 0x02, 0x0d, 0x09, 0x05, 0x01, 0x0c, 0x08, 0x04, 0x00}; -const unsigned char EXPB0[16] = {0x03, 0x03, 0x03, 0x03, 0x07, 0x07, 0x07, 0x07, 0x0b, 0x0b, 0x0b, 0x0b, 0x0f, 0x0f, 0x0f, 0x0f}; +const unsigned char ROTB[16] = { + 0x00, 0x00, 0x00, 0x0c, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x04, 0x00, 0x00, 0x00, 0x08 +}; +const unsigned char M0[16] = { 0x0f, 0x0b, 0x07, 0x03, 0x0e, 0x0a, 0x06, 0x02, + 0x0d, 0x09, 0x05, 0x01, 0x0c, 0x08, 0x04, 0x00 }; +const unsigned char EXPB0[16] = { 0x03, 0x03, 0x03, 0x03, 0x07, 0x07, + 0x07, 0x07, 0x0b, 0x0b, 0x0b, 0x0b, + 0x0f, 0x0f, 0x0f, 0x0f }; -const unsigned char SWAP32[16] = {0x03, 0x02, 0x01, 0x00, 0x07, 0x06, 0x05, 0x04, 0x0b, 0x0a, 0x09, 0x08, 0x0f, 0x0e, 0x0d, 0x0c}; -const unsigned char M0SWAP[16] = {0x0c, 0x08, 0x04, 0x00, 0x0d, 0x09, 0x05, 0x01, 0x0e, 0x0a, 0x06, 0x02, 0x0f, 0x0b, 0x07, 0x03}; -const unsigned char SR[16] = {0x01, 0x02, 0x03, 0x00, 0x06, 0x07, 0x04, 0x05, 0x0b, 0x08, 0x09, 0x0a, 0x0c, 0x0d, 0x0e, 0x0f}; -const unsigned char SRM0[16] = {0x0f, 0x0a, 0x05, 0x00, 0x0e, 0x09, 0x04, 0x03, 0x0d, 0x08, 0x07, 0x02, 0x0c, 0x0b, 0x06, 0x01}; +const unsigned char SWAP32[16] = { 0x03, 0x02, 0x01, 0x00, 0x07, 0x06, + 0x05, 0x04, 0x0b, 0x0a, 0x09, 0x08, + 0x0f, 0x0e, 0x0d, 0x0c }; +const unsigned char M0SWAP[16] = { 0x0c, 0x08, 0x04, 0x00, 0x0d, 0x09, + 0x05, 0x01, 0x0e, 0x0a, 0x06, 0x02, + 0x0f, 0x0b, 0x07, 0x03 }; +const unsigned char SR[16] = { 0x01, 0x02, 0x03, 0x00, 0x06, 0x07, 0x04, 0x05, + 0x0b, 0x08, 0x09, 0x0a, 0x0c, 0x0d, 0x0e, 0x0f }; +const unsigned char SRM0[16] = { + 0x0f, 0x0a, 0x05, 0x00, 0x0e, 0x09, 0x04, 0x03, + 0x0d, 0x08, 0x07, 0x02, 0x0c, 0x0b, 0x06, 0x01 +}; -const int128 BS0 = {{0x5555555555555555ULL, 0x5555555555555555ULL}}; -const int128 BS1 = {{0x3333333333333333ULL, 0x3333333333333333ULL}}; -const int128 BS2 = {{0x0f0f0f0f0f0f0f0fULL, 0x0f0f0f0f0f0f0f0fULL}}; +const int128 BS0 = { { 0x5555555555555555ULL, 0x5555555555555555ULL } }; +const int128 BS1 = { { 0x3333333333333333ULL, 0x3333333333333333ULL } }; +const int128 BS2 = { { 0x0f0f0f0f0f0f0f0fULL, 0x0f0f0f0f0f0f0f0fULL } }; diff --git a/examples/aesbitsliced/crypto_stream_aes128ctr.h b/examples/aesbitsliced/crypto_stream_aes128ctr.h index 2913319..f3762c6 100644 --- a/examples/aesbitsliced/crypto_stream_aes128ctr.h +++ b/examples/aesbitsliced/crypto_stream_aes128ctr.h @@ -9,49 +9,58 @@ * the crypto_box functions. */ -#include #include "export.h" +#include #ifdef __cplusplus -# if __GNUC__ -# pragma GCC diagnostic ignored "-Wlong-long" -# endif -extern "C" { +#if __GNUC__ +#pragma GCC diagnostic ignored "-Wlong-long" +#endif +extern "C" +{ #endif #define crypto_stream_aes128ctr_KEYBYTES 16U -SODIUM_EXPORT -size_t crypto_stream_aes128ctr_keybytes(void); + SODIUM_EXPORT + size_t crypto_stream_aes128ctr_keybytes(void); #define crypto_stream_aes128ctr_NONCEBYTES 16U -SODIUM_EXPORT -size_t crypto_stream_aes128ctr_noncebytes(void); + SODIUM_EXPORT + size_t crypto_stream_aes128ctr_noncebytes(void); #define crypto_stream_aes128ctr_BEFORENMBYTES 1408U -SODIUM_EXPORT -size_t crypto_stream_aes128ctr_beforenmbytes(void); - -SODIUM_EXPORT -int crypto_stream_aes128ctr(unsigned char *out, unsigned long long outlen, - const unsigned char *n, const unsigned char *k); - -SODIUM_EXPORT -int crypto_stream_aes128ctr_xor(unsigned char *out, const unsigned char *in, - unsigned long long inlen, const unsigned char *n, - const unsigned char *k); - -SODIUM_EXPORT -int crypto_stream_aes128ctr_beforenm(unsigned char *c, const unsigned char *k); - -SODIUM_EXPORT -int crypto_stream_aes128ctr_afternm(unsigned char *out, unsigned long long len, - const unsigned char *nonce, const unsigned char *c); - -SODIUM_EXPORT -int crypto_stream_aes128ctr_xor_afternm(unsigned char *out, const unsigned char *in, - unsigned long long len, - const unsigned char *nonce, - const unsigned char *c); + SODIUM_EXPORT + size_t crypto_stream_aes128ctr_beforenmbytes(void); + + SODIUM_EXPORT + int crypto_stream_aes128ctr(unsigned char* out, + unsigned long long outlen, + const unsigned char* n, + const unsigned char* k); + + SODIUM_EXPORT + int crypto_stream_aes128ctr_xor(unsigned char* out, + const unsigned char* in, + unsigned long long inlen, + const unsigned char* n, + const unsigned char* k); + + SODIUM_EXPORT + int crypto_stream_aes128ctr_beforenm(unsigned char* c, + const unsigned char* k); + + SODIUM_EXPORT + int crypto_stream_aes128ctr_afternm(unsigned char* out, + unsigned long long len, + const unsigned char* nonce, + const unsigned char* c); + + SODIUM_EXPORT + int crypto_stream_aes128ctr_xor_afternm(unsigned char* out, + const unsigned char* in, + unsigned long long len, + const unsigned char* nonce, + const unsigned char* c); #ifdef __cplusplus } diff --git a/examples/aesbitsliced/dut_aesbitsliced.c b/examples/aesbitsliced/dut_aesbitsliced.c index 7a41158..86c9948 100644 --- a/examples/aesbitsliced/dut_aesbitsliced.c +++ b/examples/aesbitsliced/dut_aesbitsliced.c @@ -1,5 +1,5 @@ -#include #include +#include #include #include "api.h" @@ -8,11 +8,12 @@ #define DUDECT_IMPLEMENTATION #include "dudect.h" - -uint8_t do_one_computation(uint8_t *data) { - uint8_t in[16] = {0}; - uint8_t out[16] = {0}; - uint8_t key[16] = {0x01,0}; +uint8_t +do_one_computation(uint8_t* data) +{ + uint8_t in[16] = { 0 }; + uint8_t out[16] = { 0 }; + uint8_t key[16] = { 0x01, 0 }; uint8_t ret = 0; memcpy(in, data, 16); @@ -21,7 +22,9 @@ uint8_t do_one_computation(uint8_t *data) { return ret; } -void prepare_inputs(dudect_config_t *c, uint8_t *input_data, uint8_t *classes) { +void +prepare_inputs(dudect_config_t* c, uint8_t* input_data, uint8_t* classes) +{ randombytes(input_data, c->number_measurements * c->chunk_size); for (size_t i = 0; i < c->number_measurements; i++) { classes[i] = randombit(); @@ -33,15 +36,15 @@ void prepare_inputs(dudect_config_t *c, uint8_t *input_data, uint8_t *classes) { } } - -int main(int argc, char **argv) +int +main(int argc, char** argv) { (void)argc; (void)argv; dudect_config_t config = { - .chunk_size = 16, - .number_measurements = 1e6, + .chunk_size = 16, + .number_measurements = 1e6, }; dudect_ctx_t ctx; diff --git a/examples/aesbitsliced/export.h b/examples/aesbitsliced/export.h index 53fcd7b..f56a562 100644 --- a/examples/aesbitsliced/export.h +++ b/examples/aesbitsliced/export.h @@ -3,30 +3,30 @@ #define sodium_export_H #ifndef __GNUC__ -# ifdef __attribute__ -# undef __attribute__ -# endif -# define __attribute__(a) +#ifdef __attribute__ +#undef __attribute__ +#endif +#define __attribute__(a) #endif #ifdef SODIUM_STATIC -# define SODIUM_EXPORT +#define SODIUM_EXPORT +#else +#if defined(_MSC_VER) +#ifdef SODIUM_DLL_EXPORT +#define SODIUM_EXPORT __declspec(dllexport) +#else +#define SODIUM_EXPORT __declspec(dllimport) +#endif #else -# if defined(_MSC_VER) -# ifdef SODIUM_DLL_EXPORT -# define SODIUM_EXPORT __declspec(dllexport) -# else -# define SODIUM_EXPORT __declspec(dllimport) -# endif -# else -# if defined(__SUNPRO_C) -# define SODIUM_EXPORT __attribute__ __global -# elif defined(_MSG_VER) -# define SODIUM_EXPORT extern __declspec(dllexport) -# else -# define SODIUM_EXPORT __attribute__ ((visibility ("default"))) -# endif -# endif +#if defined(__SUNPRO_C) +#define SODIUM_EXPORT __attribute__ __global +#elif defined(_MSG_VER) +#define SODIUM_EXPORT extern __declspec(dllexport) +#else +#define SODIUM_EXPORT __attribute__((visibility("default"))) +#endif +#endif #endif #endif diff --git a/examples/aesbitsliced/int128.h b/examples/aesbitsliced/int128.h index 3fd2111..4f03895 100644 --- a/examples/aesbitsliced/int128.h +++ b/examples/aesbitsliced/int128.h @@ -6,51 +6,67 @@ #include "common.h" #ifdef __cplusplus -# if __GNUC__ -# pragma GCC diagnostic ignored "-Wlong-long" -# endif +#if __GNUC__ +#pragma GCC diagnostic ignored "-Wlong-long" +#endif #endif -typedef union { +typedef union +{ uint64_t u64[2]; uint32_t u32[4]; - uint8_t u8[16]; + uint8_t u8[16]; } int128; #define xor2 crypto_stream_aes128ctr_portable_xor2 -void xor2(int128 *r, const int128 *x); +void +xor2(int128* r, const int128* x); #define and2 crypto_stream_aes128ctr_portable_and2 -void and2(int128 *r, const int128 *x); +void +and2(int128* r, const int128* x); #define or2 crypto_stream_aes128ctr_portable_or2 -void or2(int128 *r, const int128 *x); +void +or2(int128* r, const int128* x); #define copy2 crypto_stream_aes128ctr_portable_copy2 -void copy2(int128 *r, const int128 *x); +void +copy2(int128* r, const int128* x); #define shufb crypto_stream_aes128ctr_portable_shufb -void shufb(int128 *r, const unsigned char *l); +void +shufb(int128* r, const unsigned char* l); #define shufd crypto_stream_aes128ctr_portable_shufd -void shufd(int128 *r, const int128 *x, const unsigned int c); +void +shufd(int128* r, const int128* x, const unsigned int c); -#define rshift32_littleendian crypto_stream_aes128ctr_portable_rshift32_littleendian -void rshift32_littleendian(int128 *r, const unsigned int n); +#define rshift32_littleendian \ + crypto_stream_aes128ctr_portable_rshift32_littleendian +void +rshift32_littleendian(int128* r, const unsigned int n); -#define rshift64_littleendian crypto_stream_aes128ctr_portable_rshift64_littleendian -void rshift64_littleendian(int128 *r, const unsigned int n); +#define rshift64_littleendian \ + crypto_stream_aes128ctr_portable_rshift64_littleendian +void +rshift64_littleendian(int128* r, const unsigned int n); -#define lshift64_littleendian crypto_stream_aes128ctr_portable_lshift64_littleendian -void lshift64_littleendian(int128 *r, const unsigned int n); +#define lshift64_littleendian \ + crypto_stream_aes128ctr_portable_lshift64_littleendian +void +lshift64_littleendian(int128* r, const unsigned int n); #define toggle crypto_stream_aes128ctr_portable_toggle -void toggle(int128 *r); +void +toggle(int128* r); #define xor_rcon crypto_stream_aes128ctr_portable_xor_rcon -void xor_rcon(int128 *r); +void +xor_rcon(int128* r); #define add_uint32_big crypto_stream_aes128ctr_portable_add_uint32_big -void add_uint32_big(int128 *r, uint32 x); +void +add_uint32_big(int128* r, uint32 x); #endif diff --git a/examples/aesbitsliced/int128_aes128ctr.c b/examples/aesbitsliced/int128_aes128ctr.c index 703de39..487bb28 100644 --- a/examples/aesbitsliced/int128_aes128ctr.c +++ b/examples/aesbitsliced/int128_aes128ctr.c @@ -1,36 +1,41 @@ -#include "int128.h" #include "common.h" +#include "int128.h" -void xor2(int128 *r, const int128 *x) +void +xor2(int128* r, const int128* x) { r->u64[0] ^= x->u64[0]; r->u64[1] ^= x->u64[1]; } -void and2(int128 *r, const int128 *x) +void +and2(int128* r, const int128* x) { r->u64[0] &= x->u64[0]; r->u64[1] &= x->u64[1]; } -void or2(int128 *r, const int128 *x) +void +or2(int128* r, const int128* x) { r->u64[0] |= x->u64[0]; r->u64[1] |= x->u64[1]; } -void copy2(int128 *r, const int128 *x) +void +copy2(int128* r, const int128* x) { r->u64[0] = x->u64[0]; r->u64[1] = x->u64[1]; } -void shufb(int128 *r, const unsigned char *l) +void +shufb(int128* r, const unsigned char* l) { - int128 t; - uint8_t *ct; - uint8_t *cr; + int128 t; + uint8_t* ct; + uint8_t* cr; copy2(&t, r); cr = r->u8; @@ -53,7 +58,8 @@ void shufb(int128 *r, const unsigned char *l) cr[15] = ct[l[15]]; } -void shufd(int128 *r, const int128 *x, const unsigned int c) +void +shufd(int128* r, const int128* x, const unsigned int c) { int128 t; @@ -64,68 +70,74 @@ void shufd(int128 *r, const int128 *x, const unsigned int c) copy2(r, &t); } -void rshift32_littleendian(int128 *r, const unsigned int n) +void +rshift32_littleendian(int128* r, const unsigned int n) { - unsigned char *rp = (unsigned char *)r; + unsigned char* rp = (unsigned char*)r; uint32 t; t = load32_littleendian(rp); t >>= n; store32_littleendian(rp, t); - t = load32_littleendian(rp+4); + t = load32_littleendian(rp + 4); t >>= n; - store32_littleendian(rp+4, t); - t = load32_littleendian(rp+8); + store32_littleendian(rp + 4, t); + t = load32_littleendian(rp + 8); t >>= n; - store32_littleendian(rp+8, t); - t = load32_littleendian(rp+12); + store32_littleendian(rp + 8, t); + t = load32_littleendian(rp + 12); t >>= n; - store32_littleendian(rp+12, t); + store32_littleendian(rp + 12, t); } -void rshift64_littleendian(int128 *r, const unsigned int n) +void +rshift64_littleendian(int128* r, const unsigned int n) { - unsigned char *rp = (unsigned char *)r; + unsigned char* rp = (unsigned char*)r; uint64 t; t = load64_littleendian(rp); t >>= n; store64_littleendian(rp, t); - t = load64_littleendian(rp+8); + t = load64_littleendian(rp + 8); t >>= n; - store64_littleendian(rp+8, t); + store64_littleendian(rp + 8, t); } -void lshift64_littleendian(int128 *r, const unsigned int n) +void +lshift64_littleendian(int128* r, const unsigned int n) { - unsigned char *rp = (unsigned char *)r; + unsigned char* rp = (unsigned char*)r; uint64 t; t = load64_littleendian(rp); t <<= n; store64_littleendian(rp, t); - t = load64_littleendian(rp+8); + t = load64_littleendian(rp + 8); t <<= n; - store64_littleendian(rp+8, t); + store64_littleendian(rp + 8, t); } -void toggle(int128 *r) +void +toggle(int128* r) { r->u64[0] ^= 0xffffffffffffffffULL; r->u64[1] ^= 0xffffffffffffffffULL; } -void xor_rcon(int128 *r) +void +xor_rcon(int128* r) { - unsigned char *rp = (unsigned char *)r; + unsigned char* rp = (unsigned char*)r; uint32 t; - t = load32_littleendian(rp+12); + t = load32_littleendian(rp + 12); t ^= 0xffffffff; - store32_littleendian(rp+12, t); + store32_littleendian(rp + 12, t); } -void add_uint32_big(int128 *r, uint32 x) +void +add_uint32_big(int128* r, uint32 x) { - unsigned char *rp = (unsigned char *)r; + unsigned char* rp = (unsigned char*)r; uint32 t; - t = load32_littleendian(rp+12); + t = load32_littleendian(rp + 12); t += x; - store32_littleendian(rp+12, t); + store32_littleendian(rp + 12, t); } diff --git a/examples/aesbitsliced/stream_aes128ctr.c b/examples/aesbitsliced/stream_aes128ctr.c index 8f4ec72..6edd38a 100644 --- a/examples/aesbitsliced/stream_aes128ctr.c +++ b/examples/aesbitsliced/stream_aes128ctr.c @@ -1,28 +1,26 @@ #include "api.h" -int crypto_stream( - unsigned char *out, - unsigned long long outlen, - const unsigned char *n, - const unsigned char *k - ) +int +crypto_stream(unsigned char* out, + unsigned long long outlen, + const unsigned char* n, + const unsigned char* k) { - unsigned char d[crypto_stream_BEFORENMBYTES]; - crypto_stream_beforenm(d, k); - crypto_stream_afternm(out, outlen, n, d); - return 0; + unsigned char d[crypto_stream_BEFORENMBYTES]; + crypto_stream_beforenm(d, k); + crypto_stream_afternm(out, outlen, n, d); + return 0; } -int crypto_stream_xor( - unsigned char *out, - const unsigned char *in, - unsigned long long inlen, - const unsigned char *n, - const unsigned char *k - ) +int +crypto_stream_xor(unsigned char* out, + const unsigned char* in, + unsigned long long inlen, + const unsigned char* n, + const unsigned char* k) { - unsigned char d[crypto_stream_BEFORENMBYTES]; - crypto_stream_beforenm(d, k); - crypto_stream_xor_afternm(out, in, inlen, n, d); - return 0; + unsigned char d[crypto_stream_BEFORENMBYTES]; + crypto_stream_beforenm(d, k); + crypto_stream_xor_afternm(out, in, inlen, n, d); + return 0; } diff --git a/examples/aesbitsliced/xor_afternm_aes128ctr.c b/examples/aesbitsliced/xor_afternm_aes128ctr.c index 139dbe5..78a34d3 100644 --- a/examples/aesbitsliced/xor_afternm_aes128ctr.c +++ b/examples/aesbitsliced/xor_afternm_aes128ctr.c @@ -1,14 +1,18 @@ -/* Author: Peter Schwabe, ported from an assembly implementation by Emilia Käsper - * Date: 2009-03-19 - * Public domain */ +/* Author: Peter Schwabe, ported from an assembly implementation by Emilia + * Käsper Date: 2009-03-19 Public domain */ -#include #include "api.h" -#include "int128.h" #include "common.h" #include "consts.h" +#include "int128.h" +#include -int crypto_stream_xor_afternm(unsigned char *out, const unsigned char *in, unsigned long long len, const unsigned char *nonce, const unsigned char *c) +int +crypto_stream_xor_afternm(unsigned char* out, + const unsigned char* in, + unsigned long long len, + const unsigned char* nonce, + const unsigned char* c) { int128 xmm0; @@ -32,150 +36,312 @@ int crypto_stream_xor_afternm(unsigned char *out, const unsigned char *in, unsig int128 nonce_stack; unsigned long long lensav; unsigned char bl[128]; - unsigned char *blp; - unsigned char *np; + unsigned char* blp; + unsigned char* np; unsigned char b; uint32 tmp; /* Copy nonce on the stack */ - copy2(&nonce_stack, (const int128 *) (nonce + 0)); - np = (unsigned char *)&nonce_stack; - - enc_block: - - xmm0 = *(int128 *) (np + 0); - copy2(&xmm1, &xmm0); - shufb(&xmm1, SWAP32); - copy2(&xmm2, &xmm1); - copy2(&xmm3, &xmm1); - copy2(&xmm4, &xmm1); - copy2(&xmm5, &xmm1); - copy2(&xmm6, &xmm1); - copy2(&xmm7, &xmm1); - - add_uint32_big(&xmm1, 1); - add_uint32_big(&xmm2, 2); - add_uint32_big(&xmm3, 3); - add_uint32_big(&xmm4, 4); - add_uint32_big(&xmm5, 5); - add_uint32_big(&xmm6, 6); - add_uint32_big(&xmm7, 7); - - shufb(&xmm0, M0); - shufb(&xmm1, M0SWAP); - shufb(&xmm2, M0SWAP); - shufb(&xmm3, M0SWAP); - shufb(&xmm4, M0SWAP); - shufb(&xmm5, M0SWAP); - shufb(&xmm6, M0SWAP); - shufb(&xmm7, M0SWAP); - - bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, xmm8) - - aesround( 1, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c) - aesround( 2, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c) - aesround( 3, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c) - aesround( 4, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c) - aesround( 5, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c) - aesround( 6, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c) - aesround( 7, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c) - aesround( 8, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c) - aesround( 9, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c) - lastround(xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c) - - bitslice(xmm13, xmm10, xmm15, xmm11, xmm14, xmm12, xmm9, xmm8, xmm0) - - if(len < 128) goto partial; - if(len == 128) goto full; - - tmp = load32_bigendian(np + 12); - tmp += 8; - store32_bigendian(np + 12, tmp); - - xor2(&xmm8, (const int128 *)(in + 0)); - xor2(&xmm9, (const int128 *)(in + 16)); - xor2(&xmm12, (const int128 *)(in + 32)); - xor2(&xmm14, (const int128 *)(in + 48)); - xor2(&xmm11, (const int128 *)(in + 64)); - xor2(&xmm15, (const int128 *)(in + 80)); - xor2(&xmm10, (const int128 *)(in + 96)); - xor2(&xmm13, (const int128 *)(in + 112)); - - *(int128 *) (out + 0) = xmm8; - *(int128 *) (out + 16) = xmm9; - *(int128 *) (out + 32) = xmm12; - *(int128 *) (out + 48) = xmm14; - *(int128 *) (out + 64) = xmm11; - *(int128 *) (out + 80) = xmm15; - *(int128 *) (out + 96) = xmm10; - *(int128 *) (out + 112) = xmm13; - - len -= 128; - in += 128; - out += 128; - - goto enc_block; - - partial: - - lensav = len; - len >>= 4; - - tmp = load32_bigendian(np + 12); - tmp += len; - store32_bigendian(np + 12, tmp); - - blp = bl; - *(int128 *)(blp + 0) = xmm8; - *(int128 *)(blp + 16) = xmm9; - *(int128 *)(blp + 32) = xmm12; - *(int128 *)(blp + 48) = xmm14; - *(int128 *)(blp + 64) = xmm11; - *(int128 *)(blp + 80) = xmm15; - *(int128 *)(blp + 96) = xmm10; - *(int128 *)(blp + 112) = xmm13; - - bytes: - - if(lensav == 0) goto end; - - b = blp[0]; /* clang false positive */ - b ^= *(const unsigned char *)(in + 0); - *(unsigned char *)(out + 0) = b; - - blp += 1; - in +=1; - out +=1; - lensav -= 1; - - goto bytes; - - full: - - tmp = load32_bigendian(np + 12); - tmp += 8; - store32_bigendian(np + 12, tmp); - - xor2(&xmm8, (const int128 *)(in + 0)); - xor2(&xmm9, (const int128 *)(in + 16)); - xor2(&xmm12, (const int128 *)(in + 32)); - xor2(&xmm14, (const int128 *)(in + 48)); - xor2(&xmm11, (const int128 *)(in + 64)); - xor2(&xmm15, (const int128 *)(in + 80)); - xor2(&xmm10, (const int128 *)(in + 96)); - xor2(&xmm13, (const int128 *)(in + 112)); - - *(int128 *) (out + 0) = xmm8; - *(int128 *) (out + 16) = xmm9; - *(int128 *) (out + 32) = xmm12; - *(int128 *) (out + 48) = xmm14; - *(int128 *) (out + 64) = xmm11; - *(int128 *) (out + 80) = xmm15; - *(int128 *) (out + 96) = xmm10; - *(int128 *) (out + 112) = xmm13; - - end: - return 0; - + copy2(&nonce_stack, (const int128*)(nonce + 0)); + np = (unsigned char*)&nonce_stack; + +enc_block: + + xmm0 = *(int128*)(np + 0); + copy2(&xmm1, &xmm0); + shufb(&xmm1, SWAP32); + copy2(&xmm2, &xmm1); + copy2(&xmm3, &xmm1); + copy2(&xmm4, &xmm1); + copy2(&xmm5, &xmm1); + copy2(&xmm6, &xmm1); + copy2(&xmm7, &xmm1); + + add_uint32_big(&xmm1, 1); + add_uint32_big(&xmm2, 2); + add_uint32_big(&xmm3, 3); + add_uint32_big(&xmm4, 4); + add_uint32_big(&xmm5, 5); + add_uint32_big(&xmm6, 6); + add_uint32_big(&xmm7, 7); + + shufb(&xmm0, M0); + shufb(&xmm1, M0SWAP); + shufb(&xmm2, M0SWAP); + shufb(&xmm3, M0SWAP); + shufb(&xmm4, M0SWAP); + shufb(&xmm5, M0SWAP); + shufb(&xmm6, M0SWAP); + shufb(&xmm7, M0SWAP); + + bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, xmm8) + + aesround(1, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + c) aesround(2, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + c) aesround(3, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + c) aesround(4, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + c) aesround(5, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + c) aesround(6, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + c) + aesround(7, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + c) aesround(8, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + c) aesround(9, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + c) lastround(xmm8, + xmm9, + xmm10, + xmm11, + xmm12, + xmm13, + xmm14, + xmm15, + xmm0, + xmm1, + xmm2, + xmm3, + xmm4, + xmm5, + xmm6, + xmm7, + c) + + bitslice(xmm13, xmm10, xmm15, xmm11, xmm14, xmm12, xmm9, xmm8, xmm0) + + if (len < 128) goto partial; + if (len == 128) + goto full; + + tmp = load32_bigendian(np + 12); + tmp += 8; + store32_bigendian(np + 12, tmp); + + xor2(&xmm8, (const int128*)(in + 0)); + xor2(&xmm9, (const int128*)(in + 16)); + xor2(&xmm12, (const int128*)(in + 32)); + xor2(&xmm14, (const int128*)(in + 48)); + xor2(&xmm11, (const int128*)(in + 64)); + xor2(&xmm15, (const int128*)(in + 80)); + xor2(&xmm10, (const int128*)(in + 96)); + xor2(&xmm13, (const int128*)(in + 112)); + + *(int128*)(out + 0) = xmm8; + *(int128*)(out + 16) = xmm9; + *(int128*)(out + 32) = xmm12; + *(int128*)(out + 48) = xmm14; + *(int128*)(out + 64) = xmm11; + *(int128*)(out + 80) = xmm15; + *(int128*)(out + 96) = xmm10; + *(int128*)(out + 112) = xmm13; + + len -= 128; + in += 128; + out += 128; + + goto enc_block; + +partial: + + lensav = len; + len >>= 4; + + tmp = load32_bigendian(np + 12); + tmp += len; + store32_bigendian(np + 12, tmp); + + blp = bl; + *(int128*)(blp + 0) = xmm8; + *(int128*)(blp + 16) = xmm9; + *(int128*)(blp + 32) = xmm12; + *(int128*)(blp + 48) = xmm14; + *(int128*)(blp + 64) = xmm11; + *(int128*)(blp + 80) = xmm15; + *(int128*)(blp + 96) = xmm10; + *(int128*)(blp + 112) = xmm13; + +bytes: + + if (lensav == 0) + goto end; + + b = blp[0]; /* clang false positive */ + b ^= *(const unsigned char*)(in + 0); + *(unsigned char*)(out + 0) = b; + + blp += 1; + in += 1; + out += 1; + lensav -= 1; + + goto bytes; + +full: + + tmp = load32_bigendian(np + 12); + tmp += 8; + store32_bigendian(np + 12, tmp); + + xor2(&xmm8, (const int128*)(in + 0)); + xor2(&xmm9, (const int128*)(in + 16)); + xor2(&xmm12, (const int128*)(in + 32)); + xor2(&xmm14, (const int128*)(in + 48)); + xor2(&xmm11, (const int128*)(in + 64)); + xor2(&xmm15, (const int128*)(in + 80)); + xor2(&xmm10, (const int128*)(in + 96)); + xor2(&xmm13, (const int128*)(in + 112)); + + *(int128*)(out + 0) = xmm8; + *(int128*)(out + 16) = xmm9; + *(int128*)(out + 32) = xmm12; + *(int128*)(out + 48) = xmm14; + *(int128*)(out + 64) = xmm11; + *(int128*)(out + 80) = xmm15; + *(int128*)(out + 96) = xmm10; + *(int128*)(out + 112) = xmm13; + +end: + return 0; } diff --git a/examples/donna/curve25519-donna.c b/examples/donna/curve25519-donna.c index ed15d6c..f84268b 100644 --- a/examples/donna/curve25519-donna.c +++ b/examples/donna/curve25519-donna.c @@ -45,8 +45,8 @@ * uses many of the tricks described therein. Only the crecip function is taken * from the sample implementation. */ -#include #include +#include #ifdef _MSC_VER #define inline __inline @@ -65,17 +65,21 @@ typedef int64_t limb; * i.e. the limbs are 26, 25, 26, 25, ... bits wide. */ /* Sum two numbers: output += in */ -static void fsum(limb *output, const limb *in) { +static void +fsum(limb* output, const limb* in) +{ unsigned i; for (i = 0; i < 10; i += 2) { - output[0+i] = output[0+i] + in[0+i]; - output[1+i] = output[1+i] + in[1+i]; + output[0 + i] = output[0 + i] + in[0 + i]; + output[1 + i] = output[1 + i] + in[1 + i]; } } /* Find the difference of two numbers: output = in - output * (note the order of the arguments!). */ -static void fdifference(limb *output, const limb *in) { +static void +fdifference(limb* output, const limb* in) +{ unsigned i; for (i = 0; i < 10; ++i) { output[i] = in[i] - output[i]; @@ -83,7 +87,9 @@ static void fdifference(limb *output, const limb *in) { } /* Multiply a number by a scalar: output = in * scalar */ -static void fscalar_product(limb *output, const limb *in, const limb scalar) { +static void +fscalar_product(limb* output, const limb* in, const limb scalar) +{ unsigned i; for (i = 0; i < 10; ++i) { output[i] = in[i] * scalar; @@ -96,114 +102,118 @@ static void fscalar_product(limb *output, const limb *in, const limb scalar) { * form, the output is not. * * output[x] <= 14 * the largest product of the input limbs. */ -static void fproduct(limb *output, const limb *in2, const limb *in) { - output[0] = ((limb) ((s32) in2[0])) * ((s32) in[0]); - output[1] = ((limb) ((s32) in2[0])) * ((s32) in[1]) + - ((limb) ((s32) in2[1])) * ((s32) in[0]); - output[2] = 2 * ((limb) ((s32) in2[1])) * ((s32) in[1]) + - ((limb) ((s32) in2[0])) * ((s32) in[2]) + - ((limb) ((s32) in2[2])) * ((s32) in[0]); - output[3] = ((limb) ((s32) in2[1])) * ((s32) in[2]) + - ((limb) ((s32) in2[2])) * ((s32) in[1]) + - ((limb) ((s32) in2[0])) * ((s32) in[3]) + - ((limb) ((s32) in2[3])) * ((s32) in[0]); - output[4] = ((limb) ((s32) in2[2])) * ((s32) in[2]) + - 2 * (((limb) ((s32) in2[1])) * ((s32) in[3]) + - ((limb) ((s32) in2[3])) * ((s32) in[1])) + - ((limb) ((s32) in2[0])) * ((s32) in[4]) + - ((limb) ((s32) in2[4])) * ((s32) in[0]); - output[5] = ((limb) ((s32) in2[2])) * ((s32) in[3]) + - ((limb) ((s32) in2[3])) * ((s32) in[2]) + - ((limb) ((s32) in2[1])) * ((s32) in[4]) + - ((limb) ((s32) in2[4])) * ((s32) in[1]) + - ((limb) ((s32) in2[0])) * ((s32) in[5]) + - ((limb) ((s32) in2[5])) * ((s32) in[0]); - output[6] = 2 * (((limb) ((s32) in2[3])) * ((s32) in[3]) + - ((limb) ((s32) in2[1])) * ((s32) in[5]) + - ((limb) ((s32) in2[5])) * ((s32) in[1])) + - ((limb) ((s32) in2[2])) * ((s32) in[4]) + - ((limb) ((s32) in2[4])) * ((s32) in[2]) + - ((limb) ((s32) in2[0])) * ((s32) in[6]) + - ((limb) ((s32) in2[6])) * ((s32) in[0]); - output[7] = ((limb) ((s32) in2[3])) * ((s32) in[4]) + - ((limb) ((s32) in2[4])) * ((s32) in[3]) + - ((limb) ((s32) in2[2])) * ((s32) in[5]) + - ((limb) ((s32) in2[5])) * ((s32) in[2]) + - ((limb) ((s32) in2[1])) * ((s32) in[6]) + - ((limb) ((s32) in2[6])) * ((s32) in[1]) + - ((limb) ((s32) in2[0])) * ((s32) in[7]) + - ((limb) ((s32) in2[7])) * ((s32) in[0]); - output[8] = ((limb) ((s32) in2[4])) * ((s32) in[4]) + - 2 * (((limb) ((s32) in2[3])) * ((s32) in[5]) + - ((limb) ((s32) in2[5])) * ((s32) in[3]) + - ((limb) ((s32) in2[1])) * ((s32) in[7]) + - ((limb) ((s32) in2[7])) * ((s32) in[1])) + - ((limb) ((s32) in2[2])) * ((s32) in[6]) + - ((limb) ((s32) in2[6])) * ((s32) in[2]) + - ((limb) ((s32) in2[0])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[0]); - output[9] = ((limb) ((s32) in2[4])) * ((s32) in[5]) + - ((limb) ((s32) in2[5])) * ((s32) in[4]) + - ((limb) ((s32) in2[3])) * ((s32) in[6]) + - ((limb) ((s32) in2[6])) * ((s32) in[3]) + - ((limb) ((s32) in2[2])) * ((s32) in[7]) + - ((limb) ((s32) in2[7])) * ((s32) in[2]) + - ((limb) ((s32) in2[1])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[1]) + - ((limb) ((s32) in2[0])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[0]); - output[10] = 2 * (((limb) ((s32) in2[5])) * ((s32) in[5]) + - ((limb) ((s32) in2[3])) * ((s32) in[7]) + - ((limb) ((s32) in2[7])) * ((s32) in[3]) + - ((limb) ((s32) in2[1])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[1])) + - ((limb) ((s32) in2[4])) * ((s32) in[6]) + - ((limb) ((s32) in2[6])) * ((s32) in[4]) + - ((limb) ((s32) in2[2])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[2]); - output[11] = ((limb) ((s32) in2[5])) * ((s32) in[6]) + - ((limb) ((s32) in2[6])) * ((s32) in[5]) + - ((limb) ((s32) in2[4])) * ((s32) in[7]) + - ((limb) ((s32) in2[7])) * ((s32) in[4]) + - ((limb) ((s32) in2[3])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[3]) + - ((limb) ((s32) in2[2])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[2]); - output[12] = ((limb) ((s32) in2[6])) * ((s32) in[6]) + - 2 * (((limb) ((s32) in2[5])) * ((s32) in[7]) + - ((limb) ((s32) in2[7])) * ((s32) in[5]) + - ((limb) ((s32) in2[3])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[3])) + - ((limb) ((s32) in2[4])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[4]); - output[13] = ((limb) ((s32) in2[6])) * ((s32) in[7]) + - ((limb) ((s32) in2[7])) * ((s32) in[6]) + - ((limb) ((s32) in2[5])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[5]) + - ((limb) ((s32) in2[4])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[4]); - output[14] = 2 * (((limb) ((s32) in2[7])) * ((s32) in[7]) + - ((limb) ((s32) in2[5])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[5])) + - ((limb) ((s32) in2[6])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[6]); - output[15] = ((limb) ((s32) in2[7])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[7]) + - ((limb) ((s32) in2[6])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[6]); - output[16] = ((limb) ((s32) in2[8])) * ((s32) in[8]) + - 2 * (((limb) ((s32) in2[7])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[7])); - output[17] = ((limb) ((s32) in2[8])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[8]); - output[18] = 2 * ((limb) ((s32) in2[9])) * ((s32) in[9]); +static void +fproduct(limb* output, const limb* in2, const limb* in) +{ + output[0] = ((limb)((s32)in2[0])) * ((s32)in[0]); + output[1] = + ((limb)((s32)in2[0])) * ((s32)in[1]) + ((limb)((s32)in2[1])) * ((s32)in[0]); + output[2] = 2 * ((limb)((s32)in2[1])) * ((s32)in[1]) + + ((limb)((s32)in2[0])) * ((s32)in[2]) + + ((limb)((s32)in2[2])) * ((s32)in[0]); + output[3] = ((limb)((s32)in2[1])) * ((s32)in[2]) + + ((limb)((s32)in2[2])) * ((s32)in[1]) + + ((limb)((s32)in2[0])) * ((s32)in[3]) + + ((limb)((s32)in2[3])) * ((s32)in[0]); + output[4] = ((limb)((s32)in2[2])) * ((s32)in[2]) + + 2 * (((limb)((s32)in2[1])) * ((s32)in[3]) + + ((limb)((s32)in2[3])) * ((s32)in[1])) + + ((limb)((s32)in2[0])) * ((s32)in[4]) + + ((limb)((s32)in2[4])) * ((s32)in[0]); + output[5] = ((limb)((s32)in2[2])) * ((s32)in[3]) + + ((limb)((s32)in2[3])) * ((s32)in[2]) + + ((limb)((s32)in2[1])) * ((s32)in[4]) + + ((limb)((s32)in2[4])) * ((s32)in[1]) + + ((limb)((s32)in2[0])) * ((s32)in[5]) + + ((limb)((s32)in2[5])) * ((s32)in[0]); + output[6] = 2 * (((limb)((s32)in2[3])) * ((s32)in[3]) + + ((limb)((s32)in2[1])) * ((s32)in[5]) + + ((limb)((s32)in2[5])) * ((s32)in[1])) + + ((limb)((s32)in2[2])) * ((s32)in[4]) + + ((limb)((s32)in2[4])) * ((s32)in[2]) + + ((limb)((s32)in2[0])) * ((s32)in[6]) + + ((limb)((s32)in2[6])) * ((s32)in[0]); + output[7] = ((limb)((s32)in2[3])) * ((s32)in[4]) + + ((limb)((s32)in2[4])) * ((s32)in[3]) + + ((limb)((s32)in2[2])) * ((s32)in[5]) + + ((limb)((s32)in2[5])) * ((s32)in[2]) + + ((limb)((s32)in2[1])) * ((s32)in[6]) + + ((limb)((s32)in2[6])) * ((s32)in[1]) + + ((limb)((s32)in2[0])) * ((s32)in[7]) + + ((limb)((s32)in2[7])) * ((s32)in[0]); + output[8] = ((limb)((s32)in2[4])) * ((s32)in[4]) + + 2 * (((limb)((s32)in2[3])) * ((s32)in[5]) + + ((limb)((s32)in2[5])) * ((s32)in[3]) + + ((limb)((s32)in2[1])) * ((s32)in[7]) + + ((limb)((s32)in2[7])) * ((s32)in[1])) + + ((limb)((s32)in2[2])) * ((s32)in[6]) + + ((limb)((s32)in2[6])) * ((s32)in[2]) + + ((limb)((s32)in2[0])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[0]); + output[9] = ((limb)((s32)in2[4])) * ((s32)in[5]) + + ((limb)((s32)in2[5])) * ((s32)in[4]) + + ((limb)((s32)in2[3])) * ((s32)in[6]) + + ((limb)((s32)in2[6])) * ((s32)in[3]) + + ((limb)((s32)in2[2])) * ((s32)in[7]) + + ((limb)((s32)in2[7])) * ((s32)in[2]) + + ((limb)((s32)in2[1])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[1]) + + ((limb)((s32)in2[0])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[0]); + output[10] = 2 * (((limb)((s32)in2[5])) * ((s32)in[5]) + + ((limb)((s32)in2[3])) * ((s32)in[7]) + + ((limb)((s32)in2[7])) * ((s32)in[3]) + + ((limb)((s32)in2[1])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[1])) + + ((limb)((s32)in2[4])) * ((s32)in[6]) + + ((limb)((s32)in2[6])) * ((s32)in[4]) + + ((limb)((s32)in2[2])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[2]); + output[11] = ((limb)((s32)in2[5])) * ((s32)in[6]) + + ((limb)((s32)in2[6])) * ((s32)in[5]) + + ((limb)((s32)in2[4])) * ((s32)in[7]) + + ((limb)((s32)in2[7])) * ((s32)in[4]) + + ((limb)((s32)in2[3])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[3]) + + ((limb)((s32)in2[2])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[2]); + output[12] = ((limb)((s32)in2[6])) * ((s32)in[6]) + + 2 * (((limb)((s32)in2[5])) * ((s32)in[7]) + + ((limb)((s32)in2[7])) * ((s32)in[5]) + + ((limb)((s32)in2[3])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[3])) + + ((limb)((s32)in2[4])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[4]); + output[13] = ((limb)((s32)in2[6])) * ((s32)in[7]) + + ((limb)((s32)in2[7])) * ((s32)in[6]) + + ((limb)((s32)in2[5])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[5]) + + ((limb)((s32)in2[4])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[4]); + output[14] = 2 * (((limb)((s32)in2[7])) * ((s32)in[7]) + + ((limb)((s32)in2[5])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[5])) + + ((limb)((s32)in2[6])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[6]); + output[15] = ((limb)((s32)in2[7])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[7]) + + ((limb)((s32)in2[6])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[6]); + output[16] = ((limb)((s32)in2[8])) * ((s32)in[8]) + + 2 * (((limb)((s32)in2[7])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[7])); + output[17] = + ((limb)((s32)in2[8])) * ((s32)in[9]) + ((limb)((s32)in2[9])) * ((s32)in[8]); + output[18] = 2 * ((limb)((s32)in2[9])) * ((s32)in[9]); } /* Reduce a long form to a short form by taking the input mod 2^255 - 19. * * On entry: |output[i]| < 14*2^54 * On exit: |output[0..8]| < 280*2^54 */ -static void freduce_degree(limb *output) { +static void +freduce_degree(limb* output) +{ /* Each of these shifts and adds ends up multiplying the value by 19. * * For output[0..8], the absolute entry value is < 14*2^54 and we add, at @@ -248,11 +258,11 @@ static inline limb div_by_2_26(const limb v) { /* High word of v; no shift needed. */ - const uint32_t highword = (uint32_t) (((uint64_t) v) >> 32); + const uint32_t highword = (uint32_t)(((uint64_t)v) >> 32); /* Set to all 1s if v was negative; else set to 0s. */ - const int32_t sign = ((int32_t) highword) >> 31; + const int32_t sign = ((int32_t)highword) >> 31; /* Set to 0x3ffffff if v was negative; else set to 0. */ - const int32_t roundoff = ((uint32_t) sign) >> 6; + const int32_t roundoff = ((uint32_t)sign) >> 6; /* Should return v / (1<<26) */ return (v + roundoff) >> 26; } @@ -264,11 +274,11 @@ static inline limb div_by_2_25(const limb v) { /* High word of v; no shift needed*/ - const uint32_t highword = (uint32_t) (((uint64_t) v) >> 32); + const uint32_t highword = (uint32_t)(((uint64_t)v) >> 32); /* Set to all 1s if v was negative; else set to 0s. */ - const int32_t sign = ((int32_t) highword) >> 31; + const int32_t sign = ((int32_t)highword) >> 31; /* Set to 0x1ffffff if v was negative; else set to 0. */ - const int32_t roundoff = ((uint32_t) sign) >> 7; + const int32_t roundoff = ((uint32_t)sign) >> 7; /* Should return v / (1<<25) */ return (v + roundoff) >> 25; } @@ -276,7 +286,9 @@ div_by_2_25(const limb v) /* Reduce all coefficients of the short form input so that |x| < 2^26. * * On entry: |output[i]| < 280*2^54 */ -static void freduce_coefficients(limb *output) { +static void +freduce_coefficients(limb* output) +{ unsigned i; output[10] = 0; @@ -288,7 +300,7 @@ static void freduce_coefficients(limb *output) { * next limb and we can approximate the resulting bound of that limb by * 281*2^54. */ output[i] -= over << 26; - output[i+1] += over; + output[i + 1] += over; /* For the first iteration, |output[i+1]| < 281*2^54, thus |over| < * 281*2^29. When this is added to the next limb, the resulting bound can @@ -296,9 +308,9 @@ static void freduce_coefficients(limb *output) { * * For subsequent iterations of the loop, 281*2^54 remains a conservative * bound and no overflow occurs. */ - over = div_by_2_25(output[i+1]); - output[i+1] -= over << 25; - output[i+2] += over; + over = div_by_2_25(output[i + 1]); + output[i + 1] -= over << 25; + output[i + 2] += over; } /* Now |output[10]| < 281*2^29 and all other coefficients are reduced. */ output[0] += output[10] << 4; @@ -324,9 +336,11 @@ static void freduce_coefficients(limb *output) { * On entry: |in[i]| < 2^27 and |in2[i]| < 2^27. * * output must be distinct to both inputs. The output is reduced degree - * (indeed, one need only provide storage for 10 limbs) and |output[i]| < 2^26. */ + * (indeed, one need only provide storage for 10 limbs) and |output[i]| < 2^26. + */ static void -fmul(limb *output, const limb *in, const limb *in2) { +fmul(limb* output, const limb* in, const limb* in2) +{ limb t[19]; fproduct(t, in, in2); /* |t[i]| < 14*2^54 */ @@ -342,62 +356,64 @@ fmul(limb *output, const limb *in, const limb *in2) { * form, the output is not. * * output[x] <= 14 * the largest product of the input limbs. */ -static void fsquare_inner(limb *output, const limb *in) { - output[0] = ((limb) ((s32) in[0])) * ((s32) in[0]); - output[1] = 2 * ((limb) ((s32) in[0])) * ((s32) in[1]); - output[2] = 2 * (((limb) ((s32) in[1])) * ((s32) in[1]) + - ((limb) ((s32) in[0])) * ((s32) in[2])); - output[3] = 2 * (((limb) ((s32) in[1])) * ((s32) in[2]) + - ((limb) ((s32) in[0])) * ((s32) in[3])); - output[4] = ((limb) ((s32) in[2])) * ((s32) in[2]) + - 4 * ((limb) ((s32) in[1])) * ((s32) in[3]) + - 2 * ((limb) ((s32) in[0])) * ((s32) in[4]); - output[5] = 2 * (((limb) ((s32) in[2])) * ((s32) in[3]) + - ((limb) ((s32) in[1])) * ((s32) in[4]) + - ((limb) ((s32) in[0])) * ((s32) in[5])); - output[6] = 2 * (((limb) ((s32) in[3])) * ((s32) in[3]) + - ((limb) ((s32) in[2])) * ((s32) in[4]) + - ((limb) ((s32) in[0])) * ((s32) in[6]) + - 2 * ((limb) ((s32) in[1])) * ((s32) in[5])); - output[7] = 2 * (((limb) ((s32) in[3])) * ((s32) in[4]) + - ((limb) ((s32) in[2])) * ((s32) in[5]) + - ((limb) ((s32) in[1])) * ((s32) in[6]) + - ((limb) ((s32) in[0])) * ((s32) in[7])); - output[8] = ((limb) ((s32) in[4])) * ((s32) in[4]) + - 2 * (((limb) ((s32) in[2])) * ((s32) in[6]) + - ((limb) ((s32) in[0])) * ((s32) in[8]) + - 2 * (((limb) ((s32) in[1])) * ((s32) in[7]) + - ((limb) ((s32) in[3])) * ((s32) in[5]))); - output[9] = 2 * (((limb) ((s32) in[4])) * ((s32) in[5]) + - ((limb) ((s32) in[3])) * ((s32) in[6]) + - ((limb) ((s32) in[2])) * ((s32) in[7]) + - ((limb) ((s32) in[1])) * ((s32) in[8]) + - ((limb) ((s32) in[0])) * ((s32) in[9])); - output[10] = 2 * (((limb) ((s32) in[5])) * ((s32) in[5]) + - ((limb) ((s32) in[4])) * ((s32) in[6]) + - ((limb) ((s32) in[2])) * ((s32) in[8]) + - 2 * (((limb) ((s32) in[3])) * ((s32) in[7]) + - ((limb) ((s32) in[1])) * ((s32) in[9]))); - output[11] = 2 * (((limb) ((s32) in[5])) * ((s32) in[6]) + - ((limb) ((s32) in[4])) * ((s32) in[7]) + - ((limb) ((s32) in[3])) * ((s32) in[8]) + - ((limb) ((s32) in[2])) * ((s32) in[9])); - output[12] = ((limb) ((s32) in[6])) * ((s32) in[6]) + - 2 * (((limb) ((s32) in[4])) * ((s32) in[8]) + - 2 * (((limb) ((s32) in[5])) * ((s32) in[7]) + - ((limb) ((s32) in[3])) * ((s32) in[9]))); - output[13] = 2 * (((limb) ((s32) in[6])) * ((s32) in[7]) + - ((limb) ((s32) in[5])) * ((s32) in[8]) + - ((limb) ((s32) in[4])) * ((s32) in[9])); - output[14] = 2 * (((limb) ((s32) in[7])) * ((s32) in[7]) + - ((limb) ((s32) in[6])) * ((s32) in[8]) + - 2 * ((limb) ((s32) in[5])) * ((s32) in[9])); - output[15] = 2 * (((limb) ((s32) in[7])) * ((s32) in[8]) + - ((limb) ((s32) in[6])) * ((s32) in[9])); - output[16] = ((limb) ((s32) in[8])) * ((s32) in[8]) + - 4 * ((limb) ((s32) in[7])) * ((s32) in[9]); - output[17] = 2 * ((limb) ((s32) in[8])) * ((s32) in[9]); - output[18] = 2 * ((limb) ((s32) in[9])) * ((s32) in[9]); +static void +fsquare_inner(limb* output, const limb* in) +{ + output[0] = ((limb)((s32)in[0])) * ((s32)in[0]); + output[1] = 2 * ((limb)((s32)in[0])) * ((s32)in[1]); + output[2] = 2 * (((limb)((s32)in[1])) * ((s32)in[1]) + + ((limb)((s32)in[0])) * ((s32)in[2])); + output[3] = 2 * (((limb)((s32)in[1])) * ((s32)in[2]) + + ((limb)((s32)in[0])) * ((s32)in[3])); + output[4] = ((limb)((s32)in[2])) * ((s32)in[2]) + + 4 * ((limb)((s32)in[1])) * ((s32)in[3]) + + 2 * ((limb)((s32)in[0])) * ((s32)in[4]); + output[5] = 2 * (((limb)((s32)in[2])) * ((s32)in[3]) + + ((limb)((s32)in[1])) * ((s32)in[4]) + + ((limb)((s32)in[0])) * ((s32)in[5])); + output[6] = 2 * (((limb)((s32)in[3])) * ((s32)in[3]) + + ((limb)((s32)in[2])) * ((s32)in[4]) + + ((limb)((s32)in[0])) * ((s32)in[6]) + + 2 * ((limb)((s32)in[1])) * ((s32)in[5])); + output[7] = + 2 * + (((limb)((s32)in[3])) * ((s32)in[4]) + ((limb)((s32)in[2])) * ((s32)in[5]) + + ((limb)((s32)in[1])) * ((s32)in[6]) + ((limb)((s32)in[0])) * ((s32)in[7])); + output[8] = ((limb)((s32)in[4])) * ((s32)in[4]) + + 2 * (((limb)((s32)in[2])) * ((s32)in[6]) + + ((limb)((s32)in[0])) * ((s32)in[8]) + + 2 * (((limb)((s32)in[1])) * ((s32)in[7]) + + ((limb)((s32)in[3])) * ((s32)in[5]))); + output[9] = + 2 * + (((limb)((s32)in[4])) * ((s32)in[5]) + ((limb)((s32)in[3])) * ((s32)in[6]) + + ((limb)((s32)in[2])) * ((s32)in[7]) + ((limb)((s32)in[1])) * ((s32)in[8]) + + ((limb)((s32)in[0])) * ((s32)in[9])); + output[10] = 2 * (((limb)((s32)in[5])) * ((s32)in[5]) + + ((limb)((s32)in[4])) * ((s32)in[6]) + + ((limb)((s32)in[2])) * ((s32)in[8]) + + 2 * (((limb)((s32)in[3])) * ((s32)in[7]) + + ((limb)((s32)in[1])) * ((s32)in[9]))); + output[11] = + 2 * + (((limb)((s32)in[5])) * ((s32)in[6]) + ((limb)((s32)in[4])) * ((s32)in[7]) + + ((limb)((s32)in[3])) * ((s32)in[8]) + ((limb)((s32)in[2])) * ((s32)in[9])); + output[12] = ((limb)((s32)in[6])) * ((s32)in[6]) + + 2 * (((limb)((s32)in[4])) * ((s32)in[8]) + + 2 * (((limb)((s32)in[5])) * ((s32)in[7]) + + ((limb)((s32)in[3])) * ((s32)in[9]))); + output[13] = 2 * (((limb)((s32)in[6])) * ((s32)in[7]) + + ((limb)((s32)in[5])) * ((s32)in[8]) + + ((limb)((s32)in[4])) * ((s32)in[9])); + output[14] = 2 * (((limb)((s32)in[7])) * ((s32)in[7]) + + ((limb)((s32)in[6])) * ((s32)in[8]) + + 2 * ((limb)((s32)in[5])) * ((s32)in[9])); + output[15] = 2 * (((limb)((s32)in[7])) * ((s32)in[8]) + + ((limb)((s32)in[6])) * ((s32)in[9])); + output[16] = ((limb)((s32)in[8])) * ((s32)in[8]) + + 4 * ((limb)((s32)in[7])) * ((s32)in[9]); + output[17] = 2 * ((limb)((s32)in[8])) * ((s32)in[9]); + output[18] = 2 * ((limb)((s32)in[9])) * ((s32)in[9]); } /* fsquare sets output = in^2. @@ -408,7 +424,8 @@ static void fsquare_inner(limb *output, const limb *in) { * On exit: The |output| argument is in reduced coefficients form (indeed, one * need only provide storage for 10 limbs) and |out[i]| < 2^26. */ static void -fsquare(limb *output, const limb *in) { +fsquare(limb* output, const limb* in) +{ limb t[19]; fsquare_inner(t, in); /* |t[i]| < 14*2^54 because the largest product of two limbs will be < @@ -422,12 +439,14 @@ fsquare(limb *output, const limb *in) { /* Take a little-endian, 32-byte number and expand it into polynomial form */ static void -fexpand(limb *output, const u8 *input) { -#define F(n,start,shift,mask) \ - output[n] = ((((limb) input[start + 0]) | \ - ((limb) input[start + 1]) << 8 | \ - ((limb) input[start + 2]) << 16 | \ - ((limb) input[start + 3]) << 24) >> shift) & mask; +fexpand(limb* output, const u8* input) +{ +#define F(n, start, shift, mask) \ + output[n] = \ + ((((limb)input[start + 0]) | ((limb)input[start + 1]) << 8 | \ + ((limb)input[start + 2]) << 16 | ((limb)input[start + 3]) << 24) >> \ + shift) & \ + mask; F(0, 0, 0, 0x3ffffff); F(1, 3, 2, 0x1ffffff); F(2, 6, 3, 0x3ffffff); @@ -446,7 +465,9 @@ fexpand(limb *output, const u8 *input) { #endif /* s32_eq returns 0xffffffff iff a == b and zero otherwise. */ -static s32 s32_eq(s32 a, s32 b) { +static s32 +s32_eq(s32 a, s32 b) +{ a = ~(a ^ b); a &= a << 16; a &= a << 8; @@ -458,7 +479,9 @@ static s32 s32_eq(s32 a, s32 b) { /* s32_gte returns 0xffffffff if a >= b and zero otherwise, where a and b are * both non-negative. */ -static s32 s32_gte(s32 a, s32 b) { +static s32 +s32_gte(s32 a, s32 b) +{ a -= b; /* a >= 0 iff a >= b. */ return ~(a >> 31); @@ -469,7 +492,8 @@ static s32 s32_gte(s32 a, s32 b) { * * On entry: |input_limbs[i]| < 2^26 */ static void -fcontract(u8 *output, limb *input_limbs) { +fcontract(u8* output, limb* input_limbs) +{ int i; int j; s32 input[10]; @@ -488,12 +512,12 @@ fcontract(u8 *output, limb *input_limbs) { const s32 mask = input[i] >> 31; const s32 carry = -((input[i] & mask) >> 25); input[i] = input[i] + (carry << 25); - input[i+1] = input[i+1] - carry; + input[i + 1] = input[i + 1] - carry; } else { const s32 mask = input[i] >> 31; const s32 carry = -((input[i] & mask) >> 26); input[i] = input[i] + (carry << 26); - input[i+1] = input[i+1] - carry; + input[i + 1] = input[i + 1] - carry; } } @@ -537,18 +561,18 @@ fcontract(u8 *output, limb *input_limbs) { if ((i & 1) == 1) { const s32 carry = input[i] >> 25; input[i] &= 0x1ffffff; - input[i+1] += carry; + input[i + 1] += carry; } else { const s32 carry = input[i] >> 26; input[i] &= 0x3ffffff; - input[i+1] += carry; + input[i + 1] += carry; } } { const s32 carry = input[9] >> 25; input[9] &= 0x1ffffff; - input[0] += 19*carry; + input[0] += 19 * carry; } } @@ -591,23 +615,23 @@ fcontract(u8 *output, limb *input_limbs) { input[7] <<= 3; input[8] <<= 4; input[9] <<= 6; -#define F(i, s) \ - output[s+0] |= input[i] & 0xff; \ - output[s+1] = (input[i] >> 8) & 0xff; \ - output[s+2] = (input[i] >> 16) & 0xff; \ - output[s+3] = (input[i] >> 24) & 0xff; +#define F(i, s) \ + output[s + 0] |= input[i] & 0xff; \ + output[s + 1] = (input[i] >> 8) & 0xff; \ + output[s + 2] = (input[i] >> 16) & 0xff; \ + output[s + 3] = (input[i] >> 24) & 0xff; output[0] = 0; output[16] = 0; - F(0,0); - F(1,3); - F(2,6); - F(3,9); - F(4,12); - F(5,16); - F(6,19); - F(7,22); - F(8,25); - F(9,28); + F(0, 0); + F(1, 3); + F(2, 6); + F(3, 9); + F(4, 12); + F(5, 16); + F(6, 19); + F(7, 22); + F(8, 25); + F(9, 28); #undef F } @@ -622,18 +646,24 @@ fcontract(u8 *output, limb *input_limbs) { * * On entry and exit, the absolute value of the limbs of all inputs and outputs * are < 2^26. */ -static void fmonty(limb *x2, limb *z2, /* output 2Q */ - limb *x3, limb *z3, /* output Q + Q' */ - limb *x, limb *z, /* input Q */ - limb *xprime, limb *zprime, /* input Q' */ - const limb *qmqp /* input Q - Q' */) { +static void +fmonty(limb* x2, + limb* z2, /* output 2Q */ + limb* x3, + limb* z3, /* output Q + Q' */ + limb* x, + limb* z, /* input Q */ + limb* xprime, + limb* zprime, /* input Q' */ + const limb* qmqp /* input Q - Q' */) +{ limb origx[10], origxprime[10], zzz[19], xx[19], zz[19], xxprime[19], - zzprime[19], zzzprime[19], xxxprime[19]; + zzprime[19], zzzprime[19], xxxprime[19]; memcpy(origx, x, 10 * sizeof(limb)); fsum(x, z); /* |x[i]| < 2^27 */ - fdifference(z, origx); /* does x - z */ + fdifference(z, origx); /* does x - z */ /* |z[i]| < 2^27 */ memcpy(origxprime, xprime, sizeof(limb) * 10); @@ -679,7 +709,7 @@ static void fmonty(limb *x2, limb *z2, /* output 2Q */ freduce_degree(x2); freduce_coefficients(x2); /* |x2[i]| < 2^26 */ - fdifference(zz, xx); // does zz = xx - zz + fdifference(zz, xx); // does zz = xx - zz /* |zz[i]| < 2^27 */ memset(zzz + 10, 0, sizeof(limb) * 9); fscalar_product(zzz, zz, 121665); @@ -707,12 +737,13 @@ static void fmonty(limb *x2, limb *z2, /* output 2Q */ * and all all values in a[0..9],b[0..9] must have magnitude less than * INT32_MAX. */ static void -swap_conditional(limb a[19], limb b[19], limb iswap) { +swap_conditional(limb a[19], limb b[19], limb iswap) +{ unsigned i; - const s32 swap = (s32) -iswap; + const s32 swap = (s32)-iswap; for (i = 0; i < 10; ++i) { - const s32 x = swap & ( ((s32)a[i]) ^ ((s32)b[i]) ); + const s32 x = swap & (((s32)a[i]) ^ ((s32)b[i])); a[i] = ((s32)a[i]) ^ x; b[i] = ((s32)b[i]) ^ x; } @@ -724,10 +755,11 @@ swap_conditional(limb a[19], limb b[19], limb iswap) { * n: a little endian, 32-byte number * q: a point of the curve (short form) */ static void -cmult(limb *resultx, limb *resultz, const u8 *n, const limb *q) { - limb a[19] = {0}, b[19] = {1}, c[19] = {1}, d[19] = {0}; +cmult(limb* resultx, limb* resultz, const u8* n, const limb* q) +{ + limb a[19] = { 0 }, b[19] = { 1 }, c[19] = { 1 }, d[19] = { 0 }; limb *nqpqx = a, *nqpqz = b, *nqx = c, *nqz = d, *t; - limb e[19] = {0}, f[19] = {1}, g[19] = {0}, h[19] = {1}; + limb e[19] = { 0 }, f[19] = { 1 }, g[19] = { 0 }, h[19] = { 1 }; limb *nqpqx2 = e, *nqpqz2 = f, *nqx2 = g, *nqz2 = h; unsigned i, j; @@ -741,11 +773,7 @@ cmult(limb *resultx, limb *resultz, const u8 *n, const limb *q) { swap_conditional(nqx, nqpqx, bit); swap_conditional(nqz, nqpqz, bit); - fmonty(nqx2, nqz2, - nqpqx2, nqpqz2, - nqx, nqz, - nqpqx, nqpqz, - q); + fmonty(nqx2, nqz2, nqpqx2, nqpqz2, nqx, nqz, nqpqx, nqpqz, q); swap_conditional(nqx2, nqpqx2, bit); swap_conditional(nqz2, nqpqz2, bit); @@ -774,7 +802,8 @@ cmult(limb *resultx, limb *resultz, const u8 *n, const limb *q) { // Shamelessly copied from djb's code // ----------------------------------------------------------------------------- static void -crecip(limb *out, const limb *z) { +crecip(limb* out, const limb* z) +{ limb z2[10]; limb z9[10]; limb z11[10]; @@ -787,66 +816,86 @@ crecip(limb *out, const limb *z) { limb t1[10]; int i; - /* 2 */ fsquare(z2,z); - /* 4 */ fsquare(t1,z2); - /* 8 */ fsquare(t0,t1); - /* 9 */ fmul(z9,t0,z); - /* 11 */ fmul(z11,z9,z2); - /* 22 */ fsquare(t0,z11); - /* 2^5 - 2^0 = 31 */ fmul(z2_5_0,t0,z9); - - /* 2^6 - 2^1 */ fsquare(t0,z2_5_0); - /* 2^7 - 2^2 */ fsquare(t1,t0); - /* 2^8 - 2^3 */ fsquare(t0,t1); - /* 2^9 - 2^4 */ fsquare(t1,t0); - /* 2^10 - 2^5 */ fsquare(t0,t1); - /* 2^10 - 2^0 */ fmul(z2_10_0,t0,z2_5_0); - - /* 2^11 - 2^1 */ fsquare(t0,z2_10_0); - /* 2^12 - 2^2 */ fsquare(t1,t0); - /* 2^20 - 2^10 */ for (i = 2;i < 10;i += 2) { fsquare(t0,t1); fsquare(t1,t0); } - /* 2^20 - 2^0 */ fmul(z2_20_0,t1,z2_10_0); - - /* 2^21 - 2^1 */ fsquare(t0,z2_20_0); - /* 2^22 - 2^2 */ fsquare(t1,t0); - /* 2^40 - 2^20 */ for (i = 2;i < 20;i += 2) { fsquare(t0,t1); fsquare(t1,t0); } - /* 2^40 - 2^0 */ fmul(t0,t1,z2_20_0); - - /* 2^41 - 2^1 */ fsquare(t1,t0); - /* 2^42 - 2^2 */ fsquare(t0,t1); - /* 2^50 - 2^10 */ for (i = 2;i < 10;i += 2) { fsquare(t1,t0); fsquare(t0,t1); } - /* 2^50 - 2^0 */ fmul(z2_50_0,t0,z2_10_0); - - /* 2^51 - 2^1 */ fsquare(t0,z2_50_0); - /* 2^52 - 2^2 */ fsquare(t1,t0); - /* 2^100 - 2^50 */ for (i = 2;i < 50;i += 2) { fsquare(t0,t1); fsquare(t1,t0); } - /* 2^100 - 2^0 */ fmul(z2_100_0,t1,z2_50_0); - - /* 2^101 - 2^1 */ fsquare(t1,z2_100_0); - /* 2^102 - 2^2 */ fsquare(t0,t1); - /* 2^200 - 2^100 */ for (i = 2;i < 100;i += 2) { fsquare(t1,t0); fsquare(t0,t1); } - /* 2^200 - 2^0 */ fmul(t1,t0,z2_100_0); - - /* 2^201 - 2^1 */ fsquare(t0,t1); - /* 2^202 - 2^2 */ fsquare(t1,t0); - /* 2^250 - 2^50 */ for (i = 2;i < 50;i += 2) { fsquare(t0,t1); fsquare(t1,t0); } - /* 2^250 - 2^0 */ fmul(t0,t1,z2_50_0); - - /* 2^251 - 2^1 */ fsquare(t1,t0); - /* 2^252 - 2^2 */ fsquare(t0,t1); - /* 2^253 - 2^3 */ fsquare(t1,t0); - /* 2^254 - 2^4 */ fsquare(t0,t1); - /* 2^255 - 2^5 */ fsquare(t1,t0); - /* 2^255 - 21 */ fmul(out,t1,z11); + /* 2 */ fsquare(z2, z); + /* 4 */ fsquare(t1, z2); + /* 8 */ fsquare(t0, t1); + /* 9 */ fmul(z9, t0, z); + /* 11 */ fmul(z11, z9, z2); + /* 22 */ fsquare(t0, z11); + /* 2^5 - 2^0 = 31 */ fmul(z2_5_0, t0, z9); + + /* 2^6 - 2^1 */ fsquare(t0, z2_5_0); + /* 2^7 - 2^2 */ fsquare(t1, t0); + /* 2^8 - 2^3 */ fsquare(t0, t1); + /* 2^9 - 2^4 */ fsquare(t1, t0); + /* 2^10 - 2^5 */ fsquare(t0, t1); + /* 2^10 - 2^0 */ fmul(z2_10_0, t0, z2_5_0); + + /* 2^11 - 2^1 */ fsquare(t0, z2_10_0); + /* 2^12 - 2^2 */ fsquare(t1, t0); + /* 2^20 - 2^10 */ for (i = 2; i < 10; i += 2) { + fsquare(t0, t1); + fsquare(t1, t0); + } + /* 2^20 - 2^0 */ fmul(z2_20_0, t1, z2_10_0); + + /* 2^21 - 2^1 */ fsquare(t0, z2_20_0); + /* 2^22 - 2^2 */ fsquare(t1, t0); + /* 2^40 - 2^20 */ for (i = 2; i < 20; i += 2) { + fsquare(t0, t1); + fsquare(t1, t0); + } + /* 2^40 - 2^0 */ fmul(t0, t1, z2_20_0); + + /* 2^41 - 2^1 */ fsquare(t1, t0); + /* 2^42 - 2^2 */ fsquare(t0, t1); + /* 2^50 - 2^10 */ for (i = 2; i < 10; i += 2) { + fsquare(t1, t0); + fsquare(t0, t1); + } + /* 2^50 - 2^0 */ fmul(z2_50_0, t0, z2_10_0); + + /* 2^51 - 2^1 */ fsquare(t0, z2_50_0); + /* 2^52 - 2^2 */ fsquare(t1, t0); + /* 2^100 - 2^50 */ for (i = 2; i < 50; i += 2) { + fsquare(t0, t1); + fsquare(t1, t0); + } + /* 2^100 - 2^0 */ fmul(z2_100_0, t1, z2_50_0); + + /* 2^101 - 2^1 */ fsquare(t1, z2_100_0); + /* 2^102 - 2^2 */ fsquare(t0, t1); + /* 2^200 - 2^100 */ for (i = 2; i < 100; i += 2) { + fsquare(t1, t0); + fsquare(t0, t1); + } + /* 2^200 - 2^0 */ fmul(t1, t0, z2_100_0); + + /* 2^201 - 2^1 */ fsquare(t0, t1); + /* 2^202 - 2^2 */ fsquare(t1, t0); + /* 2^250 - 2^50 */ for (i = 2; i < 50; i += 2) { + fsquare(t0, t1); + fsquare(t1, t0); + } + /* 2^250 - 2^0 */ fmul(t0, t1, z2_50_0); + + /* 2^251 - 2^1 */ fsquare(t1, t0); + /* 2^252 - 2^2 */ fsquare(t0, t1); + /* 2^253 - 2^3 */ fsquare(t1, t0); + /* 2^254 - 2^4 */ fsquare(t0, t1); + /* 2^255 - 2^5 */ fsquare(t1, t0); + /* 2^255 - 21 */ fmul(out, t1, z11); } int -curve25519_donna(u8 *mypublic, const u8 *secret, const u8 *basepoint) { +curve25519_donna(u8* mypublic, const u8* secret, const u8* basepoint) +{ limb bp[10], x[10], z[11], zmone[10]; uint8_t e[32]; int i; - for (i = 0; i < 32; ++i) e[i] = secret[i]; + for (i = 0; i < 32; ++i) + e[i] = secret[i]; e[0] &= 248; e[31] &= 127; e[31] |= 64; diff --git a/examples/donna/donna.h b/examples/donna/donna.h index 2a7c8f5..53da7c5 100644 --- a/examples/donna/donna.h +++ b/examples/donna/donna.h @@ -1 +1,6 @@ -int curve25519_donna(uint8_t *mypublic, const uint8_t *secret, const uint8_t *basepoint); +#include + +int +curve25519_donna(uint8_t* mypublic, + const uint8_t* secret, + const uint8_t* basepoint); diff --git a/examples/donna/dut_donna.c b/examples/donna/dut_donna.c index 433925d..ae419eb 100644 --- a/examples/donna/dut_donna.c +++ b/examples/donna/dut_donna.c @@ -1,23 +1,27 @@ -#include +#include "donna.h" #include #include +#include #include // memcmp -#include "donna.h" #define DUDECT_IMPLEMENTATION #include "dudect.h" -uint8_t do_one_computation(uint8_t *data) { - uint8_t out[32] = {0}; - const uint8_t secret[32] = {1,2,3}; +uint8_t +do_one_computation(uint8_t* data) +{ + uint8_t out[32] = { 0 }; + const uint8_t secret[32] = { 1, 2, 3 }; uint8_t ret = 0; - const uint8_t basepoint[32] = {9}; + const uint8_t basepoint[32] = { 9 }; curve25519_donna(out, data, basepoint); ret ^= out[0]; return ret; } -void prepare_inputs(dudect_config_t *c, uint8_t *input_data, uint8_t *classes) { +void +prepare_inputs(dudect_config_t* c, uint8_t* input_data, uint8_t* classes) +{ randombytes(input_data, c->number_measurements * c->chunk_size); for (size_t i = 0; i < c->number_measurements; i++) { classes[i] = randombit(); @@ -29,14 +33,15 @@ void prepare_inputs(dudect_config_t *c, uint8_t *input_data, uint8_t *classes) { } } -int main(int argc, char **argv) +int +main(int argc, char** argv) { (void)argc; (void)argv; dudect_config_t config = { - .chunk_size = 32, - .number_measurements = 1e5, + .chunk_size = 32, + .number_measurements = 1e5, }; dudect_ctx_t ctx; diff --git a/examples/donnabad/curve25519-donnabad.c b/examples/donnabad/curve25519-donnabad.c index ca1f9d1..dd333ea 100644 --- a/examples/donnabad/curve25519-donnabad.c +++ b/examples/donnabad/curve25519-donnabad.c @@ -49,8 +49,8 @@ * uses many of the tricks described therein. Only the crecip function is taken * from the sample implementation. */ -#include #include +#include #ifdef _MSC_VER #define inline __inline @@ -69,17 +69,21 @@ typedef int64_t limb; * i.e. the limbs are 26, 25, 26, 25, ... bits wide. */ /* Sum two numbers: output += in */ -static void fsum(limb *output, const limb *in) { +static void +fsum(limb* output, const limb* in) +{ unsigned i; for (i = 0; i < 10; i += 2) { - output[0+i] = output[0+i] + in[0+i]; - output[1+i] = output[1+i] + in[1+i]; + output[0 + i] = output[0 + i] + in[0 + i]; + output[1 + i] = output[1 + i] + in[1 + i]; } } /* Find the difference of two numbers: output = in - output * (note the order of the arguments!). */ -static void fdifference(limb *output, const limb *in) { +static void +fdifference(limb* output, const limb* in) +{ unsigned i; for (i = 0; i < 10; ++i) { output[i] = in[i] - output[i]; @@ -87,7 +91,9 @@ static void fdifference(limb *output, const limb *in) { } /* Multiply a number by a scalar: output = in * scalar */ -static void fscalar_product(limb *output, const limb *in, const limb scalar) { +static void +fscalar_product(limb* output, const limb* in, const limb scalar) +{ unsigned i; for (i = 0; i < 10; ++i) { output[i] = in[i] * scalar; @@ -100,114 +106,118 @@ static void fscalar_product(limb *output, const limb *in, const limb scalar) { * form, the output is not. * * output[x] <= 14 * the largest product of the input limbs. */ -static void fproduct(limb *output, const limb *in2, const limb *in) { - output[0] = ((limb) ((s32) in2[0])) * ((s32) in[0]); - output[1] = ((limb) ((s32) in2[0])) * ((s32) in[1]) + - ((limb) ((s32) in2[1])) * ((s32) in[0]); - output[2] = 2 * ((limb) ((s32) in2[1])) * ((s32) in[1]) + - ((limb) ((s32) in2[0])) * ((s32) in[2]) + - ((limb) ((s32) in2[2])) * ((s32) in[0]); - output[3] = ((limb) ((s32) in2[1])) * ((s32) in[2]) + - ((limb) ((s32) in2[2])) * ((s32) in[1]) + - ((limb) ((s32) in2[0])) * ((s32) in[3]) + - ((limb) ((s32) in2[3])) * ((s32) in[0]); - output[4] = ((limb) ((s32) in2[2])) * ((s32) in[2]) + - 2 * (((limb) ((s32) in2[1])) * ((s32) in[3]) + - ((limb) ((s32) in2[3])) * ((s32) in[1])) + - ((limb) ((s32) in2[0])) * ((s32) in[4]) + - ((limb) ((s32) in2[4])) * ((s32) in[0]); - output[5] = ((limb) ((s32) in2[2])) * ((s32) in[3]) + - ((limb) ((s32) in2[3])) * ((s32) in[2]) + - ((limb) ((s32) in2[1])) * ((s32) in[4]) + - ((limb) ((s32) in2[4])) * ((s32) in[1]) + - ((limb) ((s32) in2[0])) * ((s32) in[5]) + - ((limb) ((s32) in2[5])) * ((s32) in[0]); - output[6] = 2 * (((limb) ((s32) in2[3])) * ((s32) in[3]) + - ((limb) ((s32) in2[1])) * ((s32) in[5]) + - ((limb) ((s32) in2[5])) * ((s32) in[1])) + - ((limb) ((s32) in2[2])) * ((s32) in[4]) + - ((limb) ((s32) in2[4])) * ((s32) in[2]) + - ((limb) ((s32) in2[0])) * ((s32) in[6]) + - ((limb) ((s32) in2[6])) * ((s32) in[0]); - output[7] = ((limb) ((s32) in2[3])) * ((s32) in[4]) + - ((limb) ((s32) in2[4])) * ((s32) in[3]) + - ((limb) ((s32) in2[2])) * ((s32) in[5]) + - ((limb) ((s32) in2[5])) * ((s32) in[2]) + - ((limb) ((s32) in2[1])) * ((s32) in[6]) + - ((limb) ((s32) in2[6])) * ((s32) in[1]) + - ((limb) ((s32) in2[0])) * ((s32) in[7]) + - ((limb) ((s32) in2[7])) * ((s32) in[0]); - output[8] = ((limb) ((s32) in2[4])) * ((s32) in[4]) + - 2 * (((limb) ((s32) in2[3])) * ((s32) in[5]) + - ((limb) ((s32) in2[5])) * ((s32) in[3]) + - ((limb) ((s32) in2[1])) * ((s32) in[7]) + - ((limb) ((s32) in2[7])) * ((s32) in[1])) + - ((limb) ((s32) in2[2])) * ((s32) in[6]) + - ((limb) ((s32) in2[6])) * ((s32) in[2]) + - ((limb) ((s32) in2[0])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[0]); - output[9] = ((limb) ((s32) in2[4])) * ((s32) in[5]) + - ((limb) ((s32) in2[5])) * ((s32) in[4]) + - ((limb) ((s32) in2[3])) * ((s32) in[6]) + - ((limb) ((s32) in2[6])) * ((s32) in[3]) + - ((limb) ((s32) in2[2])) * ((s32) in[7]) + - ((limb) ((s32) in2[7])) * ((s32) in[2]) + - ((limb) ((s32) in2[1])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[1]) + - ((limb) ((s32) in2[0])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[0]); - output[10] = 2 * (((limb) ((s32) in2[5])) * ((s32) in[5]) + - ((limb) ((s32) in2[3])) * ((s32) in[7]) + - ((limb) ((s32) in2[7])) * ((s32) in[3]) + - ((limb) ((s32) in2[1])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[1])) + - ((limb) ((s32) in2[4])) * ((s32) in[6]) + - ((limb) ((s32) in2[6])) * ((s32) in[4]) + - ((limb) ((s32) in2[2])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[2]); - output[11] = ((limb) ((s32) in2[5])) * ((s32) in[6]) + - ((limb) ((s32) in2[6])) * ((s32) in[5]) + - ((limb) ((s32) in2[4])) * ((s32) in[7]) + - ((limb) ((s32) in2[7])) * ((s32) in[4]) + - ((limb) ((s32) in2[3])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[3]) + - ((limb) ((s32) in2[2])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[2]); - output[12] = ((limb) ((s32) in2[6])) * ((s32) in[6]) + - 2 * (((limb) ((s32) in2[5])) * ((s32) in[7]) + - ((limb) ((s32) in2[7])) * ((s32) in[5]) + - ((limb) ((s32) in2[3])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[3])) + - ((limb) ((s32) in2[4])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[4]); - output[13] = ((limb) ((s32) in2[6])) * ((s32) in[7]) + - ((limb) ((s32) in2[7])) * ((s32) in[6]) + - ((limb) ((s32) in2[5])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[5]) + - ((limb) ((s32) in2[4])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[4]); - output[14] = 2 * (((limb) ((s32) in2[7])) * ((s32) in[7]) + - ((limb) ((s32) in2[5])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[5])) + - ((limb) ((s32) in2[6])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[6]); - output[15] = ((limb) ((s32) in2[7])) * ((s32) in[8]) + - ((limb) ((s32) in2[8])) * ((s32) in[7]) + - ((limb) ((s32) in2[6])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[6]); - output[16] = ((limb) ((s32) in2[8])) * ((s32) in[8]) + - 2 * (((limb) ((s32) in2[7])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[7])); - output[17] = ((limb) ((s32) in2[8])) * ((s32) in[9]) + - ((limb) ((s32) in2[9])) * ((s32) in[8]); - output[18] = 2 * ((limb) ((s32) in2[9])) * ((s32) in[9]); +static void +fproduct(limb* output, const limb* in2, const limb* in) +{ + output[0] = ((limb)((s32)in2[0])) * ((s32)in[0]); + output[1] = + ((limb)((s32)in2[0])) * ((s32)in[1]) + ((limb)((s32)in2[1])) * ((s32)in[0]); + output[2] = 2 * ((limb)((s32)in2[1])) * ((s32)in[1]) + + ((limb)((s32)in2[0])) * ((s32)in[2]) + + ((limb)((s32)in2[2])) * ((s32)in[0]); + output[3] = ((limb)((s32)in2[1])) * ((s32)in[2]) + + ((limb)((s32)in2[2])) * ((s32)in[1]) + + ((limb)((s32)in2[0])) * ((s32)in[3]) + + ((limb)((s32)in2[3])) * ((s32)in[0]); + output[4] = ((limb)((s32)in2[2])) * ((s32)in[2]) + + 2 * (((limb)((s32)in2[1])) * ((s32)in[3]) + + ((limb)((s32)in2[3])) * ((s32)in[1])) + + ((limb)((s32)in2[0])) * ((s32)in[4]) + + ((limb)((s32)in2[4])) * ((s32)in[0]); + output[5] = ((limb)((s32)in2[2])) * ((s32)in[3]) + + ((limb)((s32)in2[3])) * ((s32)in[2]) + + ((limb)((s32)in2[1])) * ((s32)in[4]) + + ((limb)((s32)in2[4])) * ((s32)in[1]) + + ((limb)((s32)in2[0])) * ((s32)in[5]) + + ((limb)((s32)in2[5])) * ((s32)in[0]); + output[6] = 2 * (((limb)((s32)in2[3])) * ((s32)in[3]) + + ((limb)((s32)in2[1])) * ((s32)in[5]) + + ((limb)((s32)in2[5])) * ((s32)in[1])) + + ((limb)((s32)in2[2])) * ((s32)in[4]) + + ((limb)((s32)in2[4])) * ((s32)in[2]) + + ((limb)((s32)in2[0])) * ((s32)in[6]) + + ((limb)((s32)in2[6])) * ((s32)in[0]); + output[7] = ((limb)((s32)in2[3])) * ((s32)in[4]) + + ((limb)((s32)in2[4])) * ((s32)in[3]) + + ((limb)((s32)in2[2])) * ((s32)in[5]) + + ((limb)((s32)in2[5])) * ((s32)in[2]) + + ((limb)((s32)in2[1])) * ((s32)in[6]) + + ((limb)((s32)in2[6])) * ((s32)in[1]) + + ((limb)((s32)in2[0])) * ((s32)in[7]) + + ((limb)((s32)in2[7])) * ((s32)in[0]); + output[8] = ((limb)((s32)in2[4])) * ((s32)in[4]) + + 2 * (((limb)((s32)in2[3])) * ((s32)in[5]) + + ((limb)((s32)in2[5])) * ((s32)in[3]) + + ((limb)((s32)in2[1])) * ((s32)in[7]) + + ((limb)((s32)in2[7])) * ((s32)in[1])) + + ((limb)((s32)in2[2])) * ((s32)in[6]) + + ((limb)((s32)in2[6])) * ((s32)in[2]) + + ((limb)((s32)in2[0])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[0]); + output[9] = ((limb)((s32)in2[4])) * ((s32)in[5]) + + ((limb)((s32)in2[5])) * ((s32)in[4]) + + ((limb)((s32)in2[3])) * ((s32)in[6]) + + ((limb)((s32)in2[6])) * ((s32)in[3]) + + ((limb)((s32)in2[2])) * ((s32)in[7]) + + ((limb)((s32)in2[7])) * ((s32)in[2]) + + ((limb)((s32)in2[1])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[1]) + + ((limb)((s32)in2[0])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[0]); + output[10] = 2 * (((limb)((s32)in2[5])) * ((s32)in[5]) + + ((limb)((s32)in2[3])) * ((s32)in[7]) + + ((limb)((s32)in2[7])) * ((s32)in[3]) + + ((limb)((s32)in2[1])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[1])) + + ((limb)((s32)in2[4])) * ((s32)in[6]) + + ((limb)((s32)in2[6])) * ((s32)in[4]) + + ((limb)((s32)in2[2])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[2]); + output[11] = ((limb)((s32)in2[5])) * ((s32)in[6]) + + ((limb)((s32)in2[6])) * ((s32)in[5]) + + ((limb)((s32)in2[4])) * ((s32)in[7]) + + ((limb)((s32)in2[7])) * ((s32)in[4]) + + ((limb)((s32)in2[3])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[3]) + + ((limb)((s32)in2[2])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[2]); + output[12] = ((limb)((s32)in2[6])) * ((s32)in[6]) + + 2 * (((limb)((s32)in2[5])) * ((s32)in[7]) + + ((limb)((s32)in2[7])) * ((s32)in[5]) + + ((limb)((s32)in2[3])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[3])) + + ((limb)((s32)in2[4])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[4]); + output[13] = ((limb)((s32)in2[6])) * ((s32)in[7]) + + ((limb)((s32)in2[7])) * ((s32)in[6]) + + ((limb)((s32)in2[5])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[5]) + + ((limb)((s32)in2[4])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[4]); + output[14] = 2 * (((limb)((s32)in2[7])) * ((s32)in[7]) + + ((limb)((s32)in2[5])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[5])) + + ((limb)((s32)in2[6])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[6]); + output[15] = ((limb)((s32)in2[7])) * ((s32)in[8]) + + ((limb)((s32)in2[8])) * ((s32)in[7]) + + ((limb)((s32)in2[6])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[6]); + output[16] = ((limb)((s32)in2[8])) * ((s32)in[8]) + + 2 * (((limb)((s32)in2[7])) * ((s32)in[9]) + + ((limb)((s32)in2[9])) * ((s32)in[7])); + output[17] = + ((limb)((s32)in2[8])) * ((s32)in[9]) + ((limb)((s32)in2[9])) * ((s32)in[8]); + output[18] = 2 * ((limb)((s32)in2[9])) * ((s32)in[9]); } /* Reduce a long form to a short form by taking the input mod 2^255 - 19. * * On entry: |output[i]| < 14*2^54 * On exit: |output[0..8]| < 280*2^54 */ -static void freduce_degree(limb *output) { +static void +freduce_degree(limb* output) +{ /* Each of these shifts and adds ends up multiplying the value by 19. * * For output[0..8], the absolute entry value is < 14*2^54 and we add, at @@ -252,11 +262,11 @@ static inline limb div_by_2_26(const limb v) { /* High word of v; no shift needed. */ - const uint32_t highword = (uint32_t) (((uint64_t) v) >> 32); + const uint32_t highword = (uint32_t)(((uint64_t)v) >> 32); /* Set to all 1s if v was negative; else set to 0s. */ - const int32_t sign = ((int32_t) highword) >> 31; + const int32_t sign = ((int32_t)highword) >> 31; /* Set to 0x3ffffff if v was negative; else set to 0. */ - const int32_t roundoff = ((uint32_t) sign) >> 6; + const int32_t roundoff = ((uint32_t)sign) >> 6; /* Should return v / (1<<26) */ return (v + roundoff) >> 26; } @@ -268,11 +278,11 @@ static inline limb div_by_2_25(const limb v) { /* High word of v; no shift needed*/ - const uint32_t highword = (uint32_t) (((uint64_t) v) >> 32); + const uint32_t highword = (uint32_t)(((uint64_t)v) >> 32); /* Set to all 1s if v was negative; else set to 0s. */ - const int32_t sign = ((int32_t) highword) >> 31; + const int32_t sign = ((int32_t)highword) >> 31; /* Set to 0x1ffffff if v was negative; else set to 0. */ - const int32_t roundoff = ((uint32_t) sign) >> 7; + const int32_t roundoff = ((uint32_t)sign) >> 7; /* Should return v / (1<<25) */ return (v + roundoff) >> 25; } @@ -280,7 +290,9 @@ div_by_2_25(const limb v) /* Reduce all coefficients of the short form input so that |x| < 2^26. * * On entry: |output[i]| < 280*2^54 */ -static void freduce_coefficients(limb *output) { +static void +freduce_coefficients(limb* output) +{ unsigned i; output[10] = 0; @@ -292,7 +304,7 @@ static void freduce_coefficients(limb *output) { * next limb and we can approximate the resulting bound of that limb by * 281*2^54. */ output[i] -= over << 26; - output[i+1] += over; + output[i + 1] += over; /* For the first iteration, |output[i+1]| < 281*2^54, thus |over| < * 281*2^29. When this is added to the next limb, the resulting bound can @@ -300,9 +312,9 @@ static void freduce_coefficients(limb *output) { * * For subsequent iterations of the loop, 281*2^54 remains a conservative * bound and no overflow occurs. */ - over = div_by_2_25(output[i+1]); - output[i+1] -= over << 25; - output[i+2] += over; + over = div_by_2_25(output[i + 1]); + output[i + 1] -= over << 25; + output[i + 2] += over; } /* Now |output[10]| < 281*2^29 and all other coefficients are reduced. */ output[0] += output[10] << 4; @@ -328,9 +340,11 @@ static void freduce_coefficients(limb *output) { * On entry: |in[i]| < 2^27 and |in2[i]| < 2^27. * * output must be distinct to both inputs. The output is reduced degree - * (indeed, one need only provide storage for 10 limbs) and |output[i]| < 2^26. */ + * (indeed, one need only provide storage for 10 limbs) and |output[i]| < 2^26. + */ static void -fmul(limb *output, const limb *in, const limb *in2) { +fmul(limb* output, const limb* in, const limb* in2) +{ limb t[19]; fproduct(t, in, in2); /* |t[i]| < 14*2^54 */ @@ -346,62 +360,64 @@ fmul(limb *output, const limb *in, const limb *in2) { * form, the output is not. * * output[x] <= 14 * the largest product of the input limbs. */ -static void fsquare_inner(limb *output, const limb *in) { - output[0] = ((limb) ((s32) in[0])) * ((s32) in[0]); - output[1] = 2 * ((limb) ((s32) in[0])) * ((s32) in[1]); - output[2] = 2 * (((limb) ((s32) in[1])) * ((s32) in[1]) + - ((limb) ((s32) in[0])) * ((s32) in[2])); - output[3] = 2 * (((limb) ((s32) in[1])) * ((s32) in[2]) + - ((limb) ((s32) in[0])) * ((s32) in[3])); - output[4] = ((limb) ((s32) in[2])) * ((s32) in[2]) + - 4 * ((limb) ((s32) in[1])) * ((s32) in[3]) + - 2 * ((limb) ((s32) in[0])) * ((s32) in[4]); - output[5] = 2 * (((limb) ((s32) in[2])) * ((s32) in[3]) + - ((limb) ((s32) in[1])) * ((s32) in[4]) + - ((limb) ((s32) in[0])) * ((s32) in[5])); - output[6] = 2 * (((limb) ((s32) in[3])) * ((s32) in[3]) + - ((limb) ((s32) in[2])) * ((s32) in[4]) + - ((limb) ((s32) in[0])) * ((s32) in[6]) + - 2 * ((limb) ((s32) in[1])) * ((s32) in[5])); - output[7] = 2 * (((limb) ((s32) in[3])) * ((s32) in[4]) + - ((limb) ((s32) in[2])) * ((s32) in[5]) + - ((limb) ((s32) in[1])) * ((s32) in[6]) + - ((limb) ((s32) in[0])) * ((s32) in[7])); - output[8] = ((limb) ((s32) in[4])) * ((s32) in[4]) + - 2 * (((limb) ((s32) in[2])) * ((s32) in[6]) + - ((limb) ((s32) in[0])) * ((s32) in[8]) + - 2 * (((limb) ((s32) in[1])) * ((s32) in[7]) + - ((limb) ((s32) in[3])) * ((s32) in[5]))); - output[9] = 2 * (((limb) ((s32) in[4])) * ((s32) in[5]) + - ((limb) ((s32) in[3])) * ((s32) in[6]) + - ((limb) ((s32) in[2])) * ((s32) in[7]) + - ((limb) ((s32) in[1])) * ((s32) in[8]) + - ((limb) ((s32) in[0])) * ((s32) in[9])); - output[10] = 2 * (((limb) ((s32) in[5])) * ((s32) in[5]) + - ((limb) ((s32) in[4])) * ((s32) in[6]) + - ((limb) ((s32) in[2])) * ((s32) in[8]) + - 2 * (((limb) ((s32) in[3])) * ((s32) in[7]) + - ((limb) ((s32) in[1])) * ((s32) in[9]))); - output[11] = 2 * (((limb) ((s32) in[5])) * ((s32) in[6]) + - ((limb) ((s32) in[4])) * ((s32) in[7]) + - ((limb) ((s32) in[3])) * ((s32) in[8]) + - ((limb) ((s32) in[2])) * ((s32) in[9])); - output[12] = ((limb) ((s32) in[6])) * ((s32) in[6]) + - 2 * (((limb) ((s32) in[4])) * ((s32) in[8]) + - 2 * (((limb) ((s32) in[5])) * ((s32) in[7]) + - ((limb) ((s32) in[3])) * ((s32) in[9]))); - output[13] = 2 * (((limb) ((s32) in[6])) * ((s32) in[7]) + - ((limb) ((s32) in[5])) * ((s32) in[8]) + - ((limb) ((s32) in[4])) * ((s32) in[9])); - output[14] = 2 * (((limb) ((s32) in[7])) * ((s32) in[7]) + - ((limb) ((s32) in[6])) * ((s32) in[8]) + - 2 * ((limb) ((s32) in[5])) * ((s32) in[9])); - output[15] = 2 * (((limb) ((s32) in[7])) * ((s32) in[8]) + - ((limb) ((s32) in[6])) * ((s32) in[9])); - output[16] = ((limb) ((s32) in[8])) * ((s32) in[8]) + - 4 * ((limb) ((s32) in[7])) * ((s32) in[9]); - output[17] = 2 * ((limb) ((s32) in[8])) * ((s32) in[9]); - output[18] = 2 * ((limb) ((s32) in[9])) * ((s32) in[9]); +static void +fsquare_inner(limb* output, const limb* in) +{ + output[0] = ((limb)((s32)in[0])) * ((s32)in[0]); + output[1] = 2 * ((limb)((s32)in[0])) * ((s32)in[1]); + output[2] = 2 * (((limb)((s32)in[1])) * ((s32)in[1]) + + ((limb)((s32)in[0])) * ((s32)in[2])); + output[3] = 2 * (((limb)((s32)in[1])) * ((s32)in[2]) + + ((limb)((s32)in[0])) * ((s32)in[3])); + output[4] = ((limb)((s32)in[2])) * ((s32)in[2]) + + 4 * ((limb)((s32)in[1])) * ((s32)in[3]) + + 2 * ((limb)((s32)in[0])) * ((s32)in[4]); + output[5] = 2 * (((limb)((s32)in[2])) * ((s32)in[3]) + + ((limb)((s32)in[1])) * ((s32)in[4]) + + ((limb)((s32)in[0])) * ((s32)in[5])); + output[6] = 2 * (((limb)((s32)in[3])) * ((s32)in[3]) + + ((limb)((s32)in[2])) * ((s32)in[4]) + + ((limb)((s32)in[0])) * ((s32)in[6]) + + 2 * ((limb)((s32)in[1])) * ((s32)in[5])); + output[7] = + 2 * + (((limb)((s32)in[3])) * ((s32)in[4]) + ((limb)((s32)in[2])) * ((s32)in[5]) + + ((limb)((s32)in[1])) * ((s32)in[6]) + ((limb)((s32)in[0])) * ((s32)in[7])); + output[8] = ((limb)((s32)in[4])) * ((s32)in[4]) + + 2 * (((limb)((s32)in[2])) * ((s32)in[6]) + + ((limb)((s32)in[0])) * ((s32)in[8]) + + 2 * (((limb)((s32)in[1])) * ((s32)in[7]) + + ((limb)((s32)in[3])) * ((s32)in[5]))); + output[9] = + 2 * + (((limb)((s32)in[4])) * ((s32)in[5]) + ((limb)((s32)in[3])) * ((s32)in[6]) + + ((limb)((s32)in[2])) * ((s32)in[7]) + ((limb)((s32)in[1])) * ((s32)in[8]) + + ((limb)((s32)in[0])) * ((s32)in[9])); + output[10] = 2 * (((limb)((s32)in[5])) * ((s32)in[5]) + + ((limb)((s32)in[4])) * ((s32)in[6]) + + ((limb)((s32)in[2])) * ((s32)in[8]) + + 2 * (((limb)((s32)in[3])) * ((s32)in[7]) + + ((limb)((s32)in[1])) * ((s32)in[9]))); + output[11] = + 2 * + (((limb)((s32)in[5])) * ((s32)in[6]) + ((limb)((s32)in[4])) * ((s32)in[7]) + + ((limb)((s32)in[3])) * ((s32)in[8]) + ((limb)((s32)in[2])) * ((s32)in[9])); + output[12] = ((limb)((s32)in[6])) * ((s32)in[6]) + + 2 * (((limb)((s32)in[4])) * ((s32)in[8]) + + 2 * (((limb)((s32)in[5])) * ((s32)in[7]) + + ((limb)((s32)in[3])) * ((s32)in[9]))); + output[13] = 2 * (((limb)((s32)in[6])) * ((s32)in[7]) + + ((limb)((s32)in[5])) * ((s32)in[8]) + + ((limb)((s32)in[4])) * ((s32)in[9])); + output[14] = 2 * (((limb)((s32)in[7])) * ((s32)in[7]) + + ((limb)((s32)in[6])) * ((s32)in[8]) + + 2 * ((limb)((s32)in[5])) * ((s32)in[9])); + output[15] = 2 * (((limb)((s32)in[7])) * ((s32)in[8]) + + ((limb)((s32)in[6])) * ((s32)in[9])); + output[16] = ((limb)((s32)in[8])) * ((s32)in[8]) + + 4 * ((limb)((s32)in[7])) * ((s32)in[9]); + output[17] = 2 * ((limb)((s32)in[8])) * ((s32)in[9]); + output[18] = 2 * ((limb)((s32)in[9])) * ((s32)in[9]); } /* fsquare sets output = in^2. @@ -412,7 +428,8 @@ static void fsquare_inner(limb *output, const limb *in) { * On exit: The |output| argument is in reduced coefficients form (indeed, one * need only provide storage for 10 limbs) and |out[i]| < 2^26. */ static void -fsquare(limb *output, const limb *in) { +fsquare(limb* output, const limb* in) +{ limb t[19]; fsquare_inner(t, in); /* |t[i]| < 14*2^54 because the largest product of two limbs will be < @@ -426,12 +443,14 @@ fsquare(limb *output, const limb *in) { /* Take a little-endian, 32-byte number and expand it into polynomial form */ static void -fexpand(limb *output, const u8 *input) { -#define F(n,start,shift,mask) \ - output[n] = ((((limb) input[start + 0]) | \ - ((limb) input[start + 1]) << 8 | \ - ((limb) input[start + 2]) << 16 | \ - ((limb) input[start + 3]) << 24) >> shift) & mask; +fexpand(limb* output, const u8* input) +{ +#define F(n, start, shift, mask) \ + output[n] = \ + ((((limb)input[start + 0]) | ((limb)input[start + 1]) << 8 | \ + ((limb)input[start + 2]) << 16 | ((limb)input[start + 3]) << 24) >> \ + shift) & \ + mask; F(0, 0, 0, 0x3ffffff); F(1, 3, 2, 0x1ffffff); F(2, 6, 3, 0x3ffffff); @@ -450,7 +469,9 @@ fexpand(limb *output, const u8 *input) { #endif /* s32_eq returns 0xffffffff iff a == b and zero otherwise. */ -static s32 s32_eq(s32 a, s32 b) { +static s32 +s32_eq(s32 a, s32 b) +{ a = ~(a ^ b); a &= a << 16; a &= a << 8; @@ -462,7 +483,9 @@ static s32 s32_eq(s32 a, s32 b) { /* s32_gte returns 0xffffffff if a >= b and zero otherwise, where a and b are * both non-negative. */ -static s32 s32_gte(s32 a, s32 b) { +static s32 +s32_gte(s32 a, s32 b) +{ a -= b; /* a >= 0 iff a >= b. */ return ~(a >> 31); @@ -473,7 +496,8 @@ static s32 s32_gte(s32 a, s32 b) { * * On entry: |input_limbs[i]| < 2^26 */ static void -fcontract(u8 *output, limb *input_limbs) { +fcontract(u8* output, limb* input_limbs) +{ int i; int j; s32 input[10]; @@ -492,12 +516,12 @@ fcontract(u8 *output, limb *input_limbs) { const s32 mask = input[i] >> 31; const s32 carry = -((input[i] & mask) >> 25); input[i] = input[i] + (carry << 25); - input[i+1] = input[i+1] - carry; + input[i + 1] = input[i + 1] - carry; } else { const s32 mask = input[i] >> 31; const s32 carry = -((input[i] & mask) >> 26); input[i] = input[i] + (carry << 26); - input[i+1] = input[i+1] - carry; + input[i + 1] = input[i + 1] - carry; } } @@ -541,18 +565,18 @@ fcontract(u8 *output, limb *input_limbs) { if ((i & 1) == 1) { const s32 carry = input[i] >> 25; input[i] &= 0x1ffffff; - input[i+1] += carry; + input[i + 1] += carry; } else { const s32 carry = input[i] >> 26; input[i] &= 0x3ffffff; - input[i+1] += carry; + input[i + 1] += carry; } } { const s32 carry = input[9] >> 25; input[9] &= 0x1ffffff; - input[0] += 19*carry; + input[0] += 19 * carry; } } @@ -595,23 +619,23 @@ fcontract(u8 *output, limb *input_limbs) { input[7] <<= 3; input[8] <<= 4; input[9] <<= 6; -#define F(i, s) \ - output[s+0] |= input[i] & 0xff; \ - output[s+1] = (input[i] >> 8) & 0xff; \ - output[s+2] = (input[i] >> 16) & 0xff; \ - output[s+3] = (input[i] >> 24) & 0xff; +#define F(i, s) \ + output[s + 0] |= input[i] & 0xff; \ + output[s + 1] = (input[i] >> 8) & 0xff; \ + output[s + 2] = (input[i] >> 16) & 0xff; \ + output[s + 3] = (input[i] >> 24) & 0xff; output[0] = 0; output[16] = 0; - F(0,0); - F(1,3); - F(2,6); - F(3,9); - F(4,12); - F(5,16); - F(6,19); - F(7,22); - F(8,25); - F(9,28); + F(0, 0); + F(1, 3); + F(2, 6); + F(3, 9); + F(4, 12); + F(5, 16); + F(6, 19); + F(7, 22); + F(8, 25); + F(9, 28); #undef F } @@ -626,18 +650,24 @@ fcontract(u8 *output, limb *input_limbs) { * * On entry and exit, the absolute value of the limbs of all inputs and outputs * are < 2^26. */ -static void fmonty(limb *x2, limb *z2, /* output 2Q */ - limb *x3, limb *z3, /* output Q + Q' */ - limb *x, limb *z, /* input Q */ - limb *xprime, limb *zprime, /* input Q' */ - const limb *qmqp /* input Q - Q' */) { +static void +fmonty(limb* x2, + limb* z2, /* output 2Q */ + limb* x3, + limb* z3, /* output Q + Q' */ + limb* x, + limb* z, /* input Q */ + limb* xprime, + limb* zprime, /* input Q' */ + const limb* qmqp /* input Q - Q' */) +{ limb origx[10], origxprime[10], zzz[19], xx[19], zz[19], xxprime[19], - zzprime[19], zzzprime[19], xxxprime[19]; + zzprime[19], zzzprime[19], xxxprime[19]; memcpy(origx, x, 10 * sizeof(limb)); fsum(x, z); /* |x[i]| < 2^27 */ - fdifference(z, origx); /* does x - z */ + fdifference(z, origx); /* does x - z */ /* |z[i]| < 2^27 */ memcpy(origxprime, xprime, sizeof(limb) * 10); @@ -683,7 +713,7 @@ static void fmonty(limb *x2, limb *z2, /* output 2Q */ freduce_degree(x2); freduce_coefficients(x2); /* |x2[i]| < 2^26 */ - fdifference(zz, xx); // does zz = xx - zz + fdifference(zz, xx); // does zz = xx - zz /* |zz[i]| < 2^27 */ memset(zzz + 10, 0, sizeof(limb) * 9); fscalar_product(zzz, zz, 121665); @@ -706,7 +736,8 @@ static void fmonty(limb *x2, limb *z2, /* output 2Q */ * */ static void -swap_conditional(limb a[19], limb b[19], limb iswap) { +swap_conditional(limb a[19], limb b[19], limb iswap) +{ unsigned i; #if 0 const s32 swap = (s32) -iswap; @@ -719,7 +750,7 @@ swap_conditional(limb a[19], limb b[19], limb iswap) { #endif // didactical timing leak - if (iswap==1) { + if (iswap == 1) { for (i = 0; i < 10; ++i) { s32 tmp = (s32)a[i]; a[i] = b[i]; @@ -734,10 +765,11 @@ swap_conditional(limb a[19], limb b[19], limb iswap) { * n: a little endian, 32-byte number * q: a point of the curve (short form) */ static void -cmult(limb *resultx, limb *resultz, const u8 *n, const limb *q) { - limb a[19] = {0}, b[19] = {1}, c[19] = {1}, d[19] = {0}; +cmult(limb* resultx, limb* resultz, const u8* n, const limb* q) +{ + limb a[19] = { 0 }, b[19] = { 1 }, c[19] = { 1 }, d[19] = { 0 }; limb *nqpqx = a, *nqpqz = b, *nqx = c, *nqz = d, *t; - limb e[19] = {0}, f[19] = {1}, g[19] = {0}, h[19] = {1}; + limb e[19] = { 0 }, f[19] = { 1 }, g[19] = { 0 }, h[19] = { 1 }; limb *nqpqx2 = e, *nqpqz2 = f, *nqx2 = g, *nqz2 = h; unsigned i, j; @@ -751,11 +783,7 @@ cmult(limb *resultx, limb *resultz, const u8 *n, const limb *q) { swap_conditional(nqx, nqpqx, bit); swap_conditional(nqz, nqpqz, bit); - fmonty(nqx2, nqz2, - nqpqx2, nqpqz2, - nqx, nqz, - nqpqx, nqpqz, - q); + fmonty(nqx2, nqz2, nqpqx2, nqpqz2, nqx, nqz, nqpqx, nqpqz, q); swap_conditional(nqx2, nqpqx2, bit); swap_conditional(nqz2, nqpqz2, bit); @@ -784,7 +812,8 @@ cmult(limb *resultx, limb *resultz, const u8 *n, const limb *q) { // Shamelessly copied from djb's code // ----------------------------------------------------------------------------- static void -crecip(limb *out, const limb *z) { +crecip(limb* out, const limb* z) +{ limb z2[10]; limb z9[10]; limb z11[10]; @@ -797,66 +826,86 @@ crecip(limb *out, const limb *z) { limb t1[10]; int i; - /* 2 */ fsquare(z2,z); - /* 4 */ fsquare(t1,z2); - /* 8 */ fsquare(t0,t1); - /* 9 */ fmul(z9,t0,z); - /* 11 */ fmul(z11,z9,z2); - /* 22 */ fsquare(t0,z11); - /* 2^5 - 2^0 = 31 */ fmul(z2_5_0,t0,z9); - - /* 2^6 - 2^1 */ fsquare(t0,z2_5_0); - /* 2^7 - 2^2 */ fsquare(t1,t0); - /* 2^8 - 2^3 */ fsquare(t0,t1); - /* 2^9 - 2^4 */ fsquare(t1,t0); - /* 2^10 - 2^5 */ fsquare(t0,t1); - /* 2^10 - 2^0 */ fmul(z2_10_0,t0,z2_5_0); - - /* 2^11 - 2^1 */ fsquare(t0,z2_10_0); - /* 2^12 - 2^2 */ fsquare(t1,t0); - /* 2^20 - 2^10 */ for (i = 2;i < 10;i += 2) { fsquare(t0,t1); fsquare(t1,t0); } - /* 2^20 - 2^0 */ fmul(z2_20_0,t1,z2_10_0); - - /* 2^21 - 2^1 */ fsquare(t0,z2_20_0); - /* 2^22 - 2^2 */ fsquare(t1,t0); - /* 2^40 - 2^20 */ for (i = 2;i < 20;i += 2) { fsquare(t0,t1); fsquare(t1,t0); } - /* 2^40 - 2^0 */ fmul(t0,t1,z2_20_0); - - /* 2^41 - 2^1 */ fsquare(t1,t0); - /* 2^42 - 2^2 */ fsquare(t0,t1); - /* 2^50 - 2^10 */ for (i = 2;i < 10;i += 2) { fsquare(t1,t0); fsquare(t0,t1); } - /* 2^50 - 2^0 */ fmul(z2_50_0,t0,z2_10_0); - - /* 2^51 - 2^1 */ fsquare(t0,z2_50_0); - /* 2^52 - 2^2 */ fsquare(t1,t0); - /* 2^100 - 2^50 */ for (i = 2;i < 50;i += 2) { fsquare(t0,t1); fsquare(t1,t0); } - /* 2^100 - 2^0 */ fmul(z2_100_0,t1,z2_50_0); - - /* 2^101 - 2^1 */ fsquare(t1,z2_100_0); - /* 2^102 - 2^2 */ fsquare(t0,t1); - /* 2^200 - 2^100 */ for (i = 2;i < 100;i += 2) { fsquare(t1,t0); fsquare(t0,t1); } - /* 2^200 - 2^0 */ fmul(t1,t0,z2_100_0); - - /* 2^201 - 2^1 */ fsquare(t0,t1); - /* 2^202 - 2^2 */ fsquare(t1,t0); - /* 2^250 - 2^50 */ for (i = 2;i < 50;i += 2) { fsquare(t0,t1); fsquare(t1,t0); } - /* 2^250 - 2^0 */ fmul(t0,t1,z2_50_0); - - /* 2^251 - 2^1 */ fsquare(t1,t0); - /* 2^252 - 2^2 */ fsquare(t0,t1); - /* 2^253 - 2^3 */ fsquare(t1,t0); - /* 2^254 - 2^4 */ fsquare(t0,t1); - /* 2^255 - 2^5 */ fsquare(t1,t0); - /* 2^255 - 21 */ fmul(out,t1,z11); + /* 2 */ fsquare(z2, z); + /* 4 */ fsquare(t1, z2); + /* 8 */ fsquare(t0, t1); + /* 9 */ fmul(z9, t0, z); + /* 11 */ fmul(z11, z9, z2); + /* 22 */ fsquare(t0, z11); + /* 2^5 - 2^0 = 31 */ fmul(z2_5_0, t0, z9); + + /* 2^6 - 2^1 */ fsquare(t0, z2_5_0); + /* 2^7 - 2^2 */ fsquare(t1, t0); + /* 2^8 - 2^3 */ fsquare(t0, t1); + /* 2^9 - 2^4 */ fsquare(t1, t0); + /* 2^10 - 2^5 */ fsquare(t0, t1); + /* 2^10 - 2^0 */ fmul(z2_10_0, t0, z2_5_0); + + /* 2^11 - 2^1 */ fsquare(t0, z2_10_0); + /* 2^12 - 2^2 */ fsquare(t1, t0); + /* 2^20 - 2^10 */ for (i = 2; i < 10; i += 2) { + fsquare(t0, t1); + fsquare(t1, t0); + } + /* 2^20 - 2^0 */ fmul(z2_20_0, t1, z2_10_0); + + /* 2^21 - 2^1 */ fsquare(t0, z2_20_0); + /* 2^22 - 2^2 */ fsquare(t1, t0); + /* 2^40 - 2^20 */ for (i = 2; i < 20; i += 2) { + fsquare(t0, t1); + fsquare(t1, t0); + } + /* 2^40 - 2^0 */ fmul(t0, t1, z2_20_0); + + /* 2^41 - 2^1 */ fsquare(t1, t0); + /* 2^42 - 2^2 */ fsquare(t0, t1); + /* 2^50 - 2^10 */ for (i = 2; i < 10; i += 2) { + fsquare(t1, t0); + fsquare(t0, t1); + } + /* 2^50 - 2^0 */ fmul(z2_50_0, t0, z2_10_0); + + /* 2^51 - 2^1 */ fsquare(t0, z2_50_0); + /* 2^52 - 2^2 */ fsquare(t1, t0); + /* 2^100 - 2^50 */ for (i = 2; i < 50; i += 2) { + fsquare(t0, t1); + fsquare(t1, t0); + } + /* 2^100 - 2^0 */ fmul(z2_100_0, t1, z2_50_0); + + /* 2^101 - 2^1 */ fsquare(t1, z2_100_0); + /* 2^102 - 2^2 */ fsquare(t0, t1); + /* 2^200 - 2^100 */ for (i = 2; i < 100; i += 2) { + fsquare(t1, t0); + fsquare(t0, t1); + } + /* 2^200 - 2^0 */ fmul(t1, t0, z2_100_0); + + /* 2^201 - 2^1 */ fsquare(t0, t1); + /* 2^202 - 2^2 */ fsquare(t1, t0); + /* 2^250 - 2^50 */ for (i = 2; i < 50; i += 2) { + fsquare(t0, t1); + fsquare(t1, t0); + } + /* 2^250 - 2^0 */ fmul(t0, t1, z2_50_0); + + /* 2^251 - 2^1 */ fsquare(t1, t0); + /* 2^252 - 2^2 */ fsquare(t0, t1); + /* 2^253 - 2^3 */ fsquare(t1, t0); + /* 2^254 - 2^4 */ fsquare(t0, t1); + /* 2^255 - 2^5 */ fsquare(t1, t0); + /* 2^255 - 21 */ fmul(out, t1, z11); } int -curve25519_donna(u8 *mypublic, const u8 *secret, const u8 *basepoint) { +curve25519_donna(u8* mypublic, const u8* secret, const u8* basepoint) +{ limb bp[10], x[10], z[11], zmone[10]; uint8_t e[32]; int i; - for (i = 0; i < 32; ++i) e[i] = secret[i]; + for (i = 0; i < 32; ++i) + e[i] = secret[i]; e[0] &= 248; e[31] &= 127; e[31] |= 64; diff --git a/examples/donnabad/donna.h b/examples/donnabad/donna.h index 2a7c8f5..53da7c5 100644 --- a/examples/donnabad/donna.h +++ b/examples/donnabad/donna.h @@ -1 +1,6 @@ -int curve25519_donna(uint8_t *mypublic, const uint8_t *secret, const uint8_t *basepoint); +#include + +int +curve25519_donna(uint8_t* mypublic, + const uint8_t* secret, + const uint8_t* basepoint); diff --git a/examples/donnabad/dut_donnabad.c b/examples/donnabad/dut_donnabad.c index d747c16..be2d7ed 100644 --- a/examples/donnabad/dut_donnabad.c +++ b/examples/donnabad/dut_donnabad.c @@ -1,23 +1,27 @@ -#include +#include "donna.h" #include #include +#include #include // memcmp -#include "donna.h" #define DUDECT_IMPLEMENTATION #include "dudect.h" -uint8_t do_one_computation(uint8_t *data) { - uint8_t out[32] = {0}; - const uint8_t secret[32] = {1,2,3}; +uint8_t +do_one_computation(uint8_t* data) +{ + uint8_t out[32] = { 0 }; + const uint8_t secret[32] = { 1, 2, 3 }; uint8_t ret = 0; - const uint8_t basepoint[32] = {9}; + const uint8_t basepoint[32] = { 9 }; curve25519_donna(out, data, basepoint); ret ^= out[0]; return ret; } -void prepare_inputs(dudect_config_t *c, uint8_t *input_data, uint8_t *classes) { +void +prepare_inputs(dudect_config_t* c, uint8_t* input_data, uint8_t* classes) +{ randombytes(input_data, c->number_measurements * c->chunk_size); for (size_t i = 0; i < c->number_measurements; i++) { classes[i] = randombit(); @@ -29,14 +33,15 @@ void prepare_inputs(dudect_config_t *c, uint8_t *input_data, uint8_t *classes) { } } -int main(int argc, char **argv) +int +main(int argc, char** argv) { (void)argc; (void)argv; dudect_config_t config = { - .chunk_size = 32, - .number_measurements = 1e3, + .chunk_size = 32, + .number_measurements = 1e3, }; dudect_ctx_t ctx; diff --git a/examples/simple/example.c b/examples/simple/example.c index 3b89384..c3626c9 100644 --- a/examples/simple/example.c +++ b/examples/simple/example.c @@ -6,22 +6,28 @@ #include "dudect.h" /* target function to check for constant time */ -int check_tag(uint8_t *x, uint8_t *y, size_t len) { +int +check_tag(uint8_t* x, uint8_t* y, size_t len) +{ return memcmp(x, y, len); } #define SECRET_LEN_BYTES (512) -uint8_t secret[SECRET_LEN_BYTES] = {0, 1, 2, 3, 4, 5, 6, 42}; +uint8_t secret[SECRET_LEN_BYTES] = { 0, 1, 2, 3, 4, 5, 6, 42 }; /* this will be called over and over */ -uint8_t do_one_computation(uint8_t *data) { +uint8_t +do_one_computation(uint8_t* data) +{ /* simulate totally bogus MAC check in non-constant time */ return check_tag(data, secret, SECRET_LEN_BYTES); } /* called once per number_measurements */ -void prepare_inputs(dudect_config_t *c, uint8_t *input_data, uint8_t *classes) { +void +prepare_inputs(dudect_config_t* c, uint8_t* input_data, uint8_t* classes) +{ randombytes(input_data, c->number_measurements * c->chunk_size); for (size_t i = 0; i < c->number_measurements; i++) { /* it is important to randomize the class sequence */ @@ -34,14 +40,16 @@ void prepare_inputs(dudect_config_t *c, uint8_t *input_data, uint8_t *classes) { } } -int run_test(void) { +int +run_test(void) +{ dudect_config_t config = { - .chunk_size = SECRET_LEN_BYTES, - #ifdef MEASUREMENTS_PER_CHUNK - .number_measurements = MEASUREMENTS_PER_CHUNK, - #else - .number_measurements = 500, - #endif + .chunk_size = SECRET_LEN_BYTES, +#ifdef MEASUREMENTS_PER_CHUNK + .number_measurements = MEASUREMENTS_PER_CHUNK, +#else + .number_measurements = 500, +#endif }; dudect_ctx_t ctx; @@ -64,7 +72,9 @@ int run_test(void) { return (int)state; } -int main(int argc, char **argv) { +int +main(int argc, char** argv) +{ (void)argc; (void)argv; diff --git a/src/dudect.h b/src/dudect.h index ee40912..e7a78dc 100644 --- a/src/dudect.h +++ b/src/dudect.h @@ -61,7 +61,8 @@ /* Used for improved C++ compatibility */ #ifdef __cplusplus -extern "C" { +extern "C" +{ #endif #ifndef DUDECT_IMPLEMENTATION @@ -69,9 +70,9 @@ extern "C" { #ifndef DUDECT_H_INCLUDED #define DUDECT_H_INCLUDED +#include #include #include -#include #include #ifdef DUDECT_VISIBLITY_STATIC @@ -80,65 +81,74 @@ extern "C" { #define DUDECT_VISIBILITY extern #endif -/* - The interface of dudect begins here, to be compiled only if DUDECT_IMPLEMENTATION is not defined. - In a multi-file library what follows would be the public-facing dudect.h -*/ + /* + The interface of dudect begins here, to be compiled only if + DUDECT_IMPLEMENTATION is not defined. In a multi-file library what follows + would be the public-facing dudect.h + */ -#define DUDECT_ENOUGH_MEASUREMENTS (10000) /* do not draw any conclusion before we reach this many measurements */ -#define DUDECT_NUMBER_PERCENTILES (100) +#define DUDECT_ENOUGH_MEASUREMENTS \ + (10000) /* do not draw any conclusion before we reach this many measurements \ + */ +#define DUDECT_NUMBER_PERCENTILES (100) /* perform this many tests in total: - 1 first order uncropped test, - DUDECT_NUMBER_PERCENTILES tests - 1 second order test */ -#define DUDECT_TESTS (1+DUDECT_NUMBER_PERCENTILES+1) - -typedef struct { - size_t chunk_size; - size_t number_measurements; -} dudect_config_t; - -typedef struct { - double mean[2]; - double m2[2]; - double n[2]; -} ttest_ctx_t; - -typedef struct { - int64_t *ticks; - int64_t *exec_times; - uint8_t *input_data; - uint8_t *classes; - dudect_config_t *config; - ttest_ctx_t *ttest_ctxs[DUDECT_TESTS]; - int64_t *percentiles; -} dudect_ctx_t; - -typedef enum { - DUDECT_LEAKAGE_FOUND=0, - DUDECT_NO_LEAKAGE_EVIDENCE_YET -} dudect_state_t; - -/* Public API */ - -DUDECT_VISIBILITY int dudect_init(dudect_ctx_t *ctx, dudect_config_t *conf); -DUDECT_VISIBILITY dudect_state_t dudect_main(dudect_ctx_t *c); -DUDECT_VISIBILITY int dudect_free(dudect_ctx_t *ctx); -DUDECT_VISIBILITY void randombytes(uint8_t *x, size_t how_much); -DUDECT_VISIBILITY uint8_t randombit(void); +#define DUDECT_TESTS (1 + DUDECT_NUMBER_PERCENTILES + 1) + + typedef struct + { + size_t chunk_size; + size_t number_measurements; + } dudect_config_t; + + typedef struct + { + double mean[2]; + double m2[2]; + double n[2]; + } ttest_ctx_t; + + typedef struct + { + int64_t* ticks; + int64_t* exec_times; + uint8_t* input_data; + uint8_t* classes; + dudect_config_t* config; + ttest_ctx_t* ttest_ctxs[DUDECT_TESTS]; + int64_t* percentiles; + } dudect_ctx_t; + + typedef enum + { + DUDECT_LEAKAGE_FOUND = 0, + DUDECT_NO_LEAKAGE_EVIDENCE_YET + } dudect_state_t; + + /* Public API */ + + DUDECT_VISIBILITY int dudect_init(dudect_ctx_t* ctx, dudect_config_t* conf); + DUDECT_VISIBILITY dudect_state_t dudect_main(dudect_ctx_t* c); + DUDECT_VISIBILITY int dudect_free(dudect_ctx_t* ctx); + DUDECT_VISIBILITY void randombytes(uint8_t* x, size_t how_much); + DUDECT_VISIBILITY uint8_t randombit(void); /* Public configuration */ /* Implementation details */ #include -#include #include +#include -// kill this -extern void prepare_inputs(dudect_config_t *c, uint8_t *input_data, uint8_t *classes); -extern uint8_t do_one_computation(uint8_t *data); + // kill this + extern void prepare_inputs(dudect_config_t* c, + uint8_t* input_data, + uint8_t* classes); + extern uint8_t do_one_computation(uint8_t* data); #endif /* DUDECT_H_INCLUDED */ @@ -148,7 +158,8 @@ extern uint8_t do_one_computation(uint8_t *data); #include "dudect.h" #define DUDECT_IMPLEMENTATION -/* The implementation of dudect begins here. In a multi-file library what follows would be dudect.c */ +/* The implementation of dudect begins here. In a multi-file library what + * follows would be dudect.c */ #define DUDECT_TRACE (0) @@ -170,7 +181,9 @@ extern uint8_t do_one_computation(uint8_t *data); see https://en.wikipedia.org/wiki/Welch%27s_t-test */ -static void t_push(ttest_ctx_t *ctx, double x, uint8_t clazz) { +static void +t_push(ttest_ctx_t* ctx, double x, uint8_t clazz) +{ assert(clazz == 0 || clazz == 1); ctx->n[clazz]++; /* @@ -182,8 +195,10 @@ static void t_push(ttest_ctx_t *ctx, double x, uint8_t clazz) { ctx->m2[clazz] = ctx->m2[clazz] + delta * (x - ctx->mean[clazz]); } -static double t_compute(ttest_ctx_t *ctx) { - double var[2] = {0.0, 0.0}; +static double +t_compute(ttest_ctx_t* ctx) +{ + double var[2] = { 0.0, 0.0 }; var[0] = ctx->m2[0] / (ctx->n[0] - 1); var[1] = ctx->m2[1] / (ctx->n[1] - 1); double num = (ctx->mean[0] - ctx->mean[1]); @@ -192,17 +207,25 @@ static double t_compute(ttest_ctx_t *ctx) { return t_value; } -static void t_init(ttest_ctx_t *ctx) { - for (int clazz = 0; clazz < 2; clazz ++) { +static void +t_init(ttest_ctx_t* ctx) +{ + for (int clazz = 0; clazz < 2; clazz++) { ctx->mean[clazz] = 0.0; ctx->m2[clazz] = 0.0; ctx->n[clazz] = 0.0; } } -static int cmp(const int64_t *a, const int64_t *b) { return (int)(*a - *b); } +static int +cmp(const int64_t* a, const int64_t* b) +{ + return (int)(*a - *b); +} -static int64_t percentile(int64_t *a_sorted, double which, size_t size) { +static int64_t +percentile(int64_t* a_sorted, double which, size_t size) +{ size_t array_position = (size_t)((double)size * (double)which); assert(array_position < size); return a_sorted[array_position]; @@ -214,17 +237,25 @@ static int64_t percentile(int64_t *a_sorted, double which, size_t size) { the measurements distribution, but there's not more science than that. */ -static void prepare_percentiles(dudect_ctx_t *ctx) { - qsort(ctx->exec_times, ctx->config->number_measurements, sizeof(int64_t), (int (*)(const void *, const void *))cmp); +static void +prepare_percentiles(dudect_ctx_t* ctx) +{ + qsort(ctx->exec_times, + ctx->config->number_measurements, + sizeof(int64_t), + (int (*)(const void*, const void*))cmp); for (size_t i = 0; i < DUDECT_NUMBER_PERCENTILES; i++) { ctx->percentiles[i] = percentile( - ctx->exec_times, 1 - (pow(0.5, 10 * (double)(i + 1) / DUDECT_NUMBER_PERCENTILES)), - ctx->config->number_measurements); + ctx->exec_times, + 1 - (pow(0.5, 10 * (double)(i + 1) / DUDECT_NUMBER_PERCENTILES)), + ctx->config->number_measurements); } } /* this comes from ebacs */ -void randombytes(uint8_t *x, size_t how_much) { +void +randombytes(uint8_t* x, size_t how_much) +{ ssize_t i; static int fd = -1; @@ -256,7 +287,9 @@ void randombytes(uint8_t *x, size_t how_much) { } } -uint8_t randombit(void) { +uint8_t +randombit(void) +{ uint8_t ret = 0; randombytes(&ret, 1); return (ret & 1); @@ -265,16 +298,21 @@ uint8_t randombit(void) { /* Returns current CPU tick count from *T*ime *S*tamp *C*ounter. - To enforce CPU to issue RDTSC instruction where we want it to, we put a `mfence` instruction before - issuing `rdtsc`, which should make all memory load/ store operations, prior to RDTSC, globally visible. + To enforce CPU to issue RDTSC instruction where we want it to, we put a + `mfence` instruction before issuing `rdtsc`, which should make all memory load/ + store operations, prior to RDTSC, globally visible. See https://github.com/oreparaz/dudect/issues/32 - See RDTSC documentation @ https://www.intel.com/content/www/us/en/docs/intrinsics-guide/index.htm#text=rdtsc&ig_expand=4395,5273 - See MFENCE documentation @ https://www.intel.com/content/www/us/en/docs/intrinsics-guide/index.htm#text=mfence&ig_expand=4395,5273,4395 + See RDTSC documentation @ + https://www.intel.com/content/www/us/en/docs/intrinsics-guide/index.htm#text=rdtsc&ig_expand=4395,5273 + See MFENCE documentation @ + https://www.intel.com/content/www/us/en/docs/intrinsics-guide/index.htm#text=mfence&ig_expand=4395,5273,4395 Also see https://stackoverflow.com/a/12634857 */ -static inline int64_t cpucycles(void) { +static inline int64_t +cpucycles(void) +{ _mm_mfence(); return (int64_t)__rdtsc(); } @@ -284,30 +322,38 @@ static inline int64_t cpucycles(void) { #define t_threshold_moderate \ 10 // test failed. Pankaj likes 4.5 but let's be more lenient -static void measure(dudect_ctx_t *ctx) { +static void +measure(dudect_ctx_t* ctx) +{ for (size_t i = 0; i < ctx->config->number_measurements; i++) { ctx->ticks[i] = cpucycles(); do_one_computation(ctx->input_data + i * ctx->config->chunk_size); } - for (size_t i = 0; i < ctx->config->number_measurements-1; i++) { - ctx->exec_times[i] = ctx->ticks[i+1] - ctx->ticks[i]; + for (size_t i = 0; i < ctx->config->number_measurements - 1; i++) { + ctx->exec_times[i] = ctx->ticks[i + 1] - ctx->ticks[i]; } } -static void update_statistics(dudect_ctx_t *ctx) { - for (size_t i = 10 /* discard the first few measurements */; i < (ctx->config->number_measurements-1); i++) { +static void +update_statistics(dudect_ctx_t* ctx) +{ + for (size_t i = 10 /* discard the first few measurements */; + i < (ctx->config->number_measurements - 1); + i++) { int64_t difference = ctx->exec_times[i]; if (difference < 0) { - continue; // the cpu cycle counter overflowed, just throw away the measurement + continue; // the cpu cycle counter overflowed, just throw away the + // measurement } // t-test on the execution time t_push(ctx->ttest_ctxs[0], difference, ctx->classes[i]); // t-test on cropped execution times, for several cropping thresholds. - for (size_t crop_index = 0; crop_index < DUDECT_NUMBER_PERCENTILES; crop_index++) { + for (size_t crop_index = 0; crop_index < DUDECT_NUMBER_PERCENTILES; + crop_index++) { if (difference < ctx->percentiles[crop_index]) { t_push(ctx->ttest_ctxs[crop_index + 1], difference, ctx->classes[i]); } @@ -316,24 +362,32 @@ static void update_statistics(dudect_ctx_t *ctx) { // second-order test (only if we have more than 10000 measurements). // Centered product pre-processing. if (ctx->ttest_ctxs[0]->n[0] > 10000) { - double centered = (double)difference - ctx->ttest_ctxs[0]->mean[ctx->classes[i]]; - t_push(ctx->ttest_ctxs[1 + DUDECT_NUMBER_PERCENTILES], centered * centered, ctx->classes[i]); + double centered = + (double)difference - ctx->ttest_ctxs[0]->mean[ctx->classes[i]]; + t_push(ctx->ttest_ctxs[1 + DUDECT_NUMBER_PERCENTILES], + centered * centered, + ctx->classes[i]); } } } #if DUDECT_TRACE -static void report_test(ttest_ctx_t *x) { +static void +report_test(ttest_ctx_t* x) +{ if (x->n[0] > DUDECT_ENOUGH_MEASUREMENTS) { double tval = t_compute(x); - printf(" abs(t): %4.2f, number measurements: %f\n", tval, x->n[0]+x->n[1]); + printf( + " abs(t): %4.2f, number measurements: %f\n", tval, x->n[0] + x->n[1]); } else { printf(" (not enough measurements: %f + %f)\n", x->n[0], x->n[1]); } } #endif /* DUDECT_TRACE */ -static ttest_ctx_t *max_test(dudect_ctx_t *ctx) { +static ttest_ctx_t* +max_test(dudect_ctx_t* ctx) +{ size_t ret = 0; double max = 0; for (size_t i = 0; i < DUDECT_TESTS; i++) { @@ -348,11 +402,15 @@ static ttest_ctx_t *max_test(dudect_ctx_t *ctx) { return ctx->ttest_ctxs[ret]; } -static dudect_state_t report(dudect_ctx_t *ctx) { +static dudect_state_t +report(dudect_ctx_t* ctx) +{ - #if DUDECT_TRACE +#if DUDECT_TRACE for (size_t i = 0; i < DUDECT_TESTS; i++) { - printf(" bucket %zu has %f measurements\n", i, ctx->ttest_ctxs[i]->n[0] + ctx->ttest_ctxs[i]->n[1]); + printf(" bucket %zu has %f measurements\n", + i, + ctx->ttest_ctxs[i]->n[0] + ctx->ttest_ctxs[i]->n[1]); } printf("t-test on raw measurements\n"); @@ -365,11 +423,11 @@ static dudect_state_t report(dudect_ctx_t *ctx) { printf("t-test for second order leakage\n"); report_test(ctx->ttest_ctxs[1 + DUDECT_NUMBER_PERCENTILES]); - #endif /* DUDECT_TRACE */ +#endif /* DUDECT_TRACE */ - ttest_ctx_t *t = max_test(ctx); + ttest_ctx_t* t = max_test(ctx); double max_t = fabs(t_compute(t)); - double number_traces_max_t = t->n[0] + t->n[1]; + double number_traces_max_t = t->n[0] + t->n[1]; double max_tau = max_t / sqrt(number_traces_max_t); // print the number of measurements of the test that yielded max t. @@ -377,7 +435,8 @@ static dudect_state_t report(dudect_ctx_t *ctx) { // but can happen (different test) printf("meas: %7.2lf M, ", (number_traces_max_t / 1e6)); if (number_traces_max_t < DUDECT_ENOUGH_MEASUREMENTS) { - printf("not enough measurements (%.0f still to go).\n", DUDECT_ENOUGH_MEASUREMENTS-number_traces_max_t); + printf("not enough measurements (%.0f still to go).\n", + DUDECT_ENOUGH_MEASUREMENTS - number_traces_max_t); return DUDECT_NO_LEAKAGE_EVIDENCE_YET; } @@ -398,9 +457,9 @@ static dudect_state_t report(dudect_ctx_t *ctx) { * pretty sensible imho) */ printf("max t: %+7.2f, max tau: %.2e, (5/tau)^2: %.2e.", - max_t, - max_tau, - (double)(5*5)/(double)(max_tau*max_tau)); + max_t, + max_tau, + (double)(5 * 5) / (double)(max_tau * max_tau)); if (max_t > t_threshold_bananas) { printf(" Definitely not constant time.\n"); @@ -416,7 +475,9 @@ static dudect_state_t report(dudect_ctx_t *ctx) { return DUDECT_NO_LEAKAGE_EVIDENCE_YET; } -dudect_state_t dudect_main(dudect_ctx_t *ctx) { +dudect_state_t +dudect_main(dudect_ctx_t* ctx) +{ prepare_inputs(ctx->config, ctx->input_data, ctx->classes); measure(ctx); @@ -435,23 +496,30 @@ dudect_state_t dudect_main(dudect_ctx_t *ctx) { return ret; } -int dudect_init(dudect_ctx_t *ctx, dudect_config_t *conf) +int +dudect_init(dudect_ctx_t* ctx, dudect_config_t* conf) { - ctx->config = (dudect_config_t*) calloc(1, sizeof(*conf)); + ctx->config = (dudect_config_t*)calloc(1, sizeof(*conf)); ctx->config->number_measurements = conf->number_measurements; ctx->config->chunk_size = conf->chunk_size; - ctx->ticks = (int64_t*) calloc(ctx->config->number_measurements, sizeof(int64_t)); - ctx->exec_times = (int64_t*) calloc(ctx->config->number_measurements, sizeof(int64_t)); - ctx->classes = (uint8_t*) calloc(ctx->config->number_measurements, sizeof(uint8_t)); - ctx->input_data = (uint8_t*) calloc(ctx->config->number_measurements * ctx->config->chunk_size, sizeof(uint8_t)); + ctx->ticks = + (int64_t*)calloc(ctx->config->number_measurements, sizeof(int64_t)); + ctx->exec_times = + (int64_t*)calloc(ctx->config->number_measurements, sizeof(int64_t)); + ctx->classes = + (uint8_t*)calloc(ctx->config->number_measurements, sizeof(uint8_t)); + ctx->input_data = + (uint8_t*)calloc(ctx->config->number_measurements * ctx->config->chunk_size, + sizeof(uint8_t)); for (int i = 0; i < DUDECT_TESTS; i++) { - ctx->ttest_ctxs[i] = (ttest_ctx_t *)calloc(1, sizeof(ttest_ctx_t)); + ctx->ttest_ctxs[i] = (ttest_ctx_t*)calloc(1, sizeof(ttest_ctx_t)); assert(ctx->ttest_ctxs[i]); t_init(ctx->ttest_ctxs[i]); } - ctx->percentiles = (int64_t*) calloc(DUDECT_NUMBER_PERCENTILES, sizeof(int64_t)); + ctx->percentiles = + (int64_t*)calloc(DUDECT_NUMBER_PERCENTILES, sizeof(int64_t)); assert(ctx->ticks); assert(ctx->exec_times); @@ -462,7 +530,8 @@ int dudect_init(dudect_ctx_t *ctx, dudect_config_t *conf) return 0; } -int dudect_free(dudect_ctx_t *ctx) +int +dudect_free(dudect_ctx_t* ctx) { for (int i = 0; i < DUDECT_TESTS; i++) { free(ctx->ttest_ctxs[i]); @@ -479,5 +548,5 @@ int dudect_free(dudect_ctx_t *ctx) #endif /* DUDECT_IMPLEMENTATION */ #ifdef __cplusplus -} /* extern "C" */ +} /* extern "C" */ #endif