| /NextBSD/contrib/llvm/lib/Target/ARM/ |
| HD | ARM.td | 34 "Enable VFP2 instructions">; 36 "Enable VFP3 instructions", 39 "Enable NEON instructions", 42 "Enable Thumb2 instructions">; 49 "Enable VFP4 instructions", 57 "Enable divide instructions">; 60 "Enable divide instructions in ARM mode">; 62 "Enable Thumb2 extract and pack instructions">; 64 "Has data barrier (dmb / dsb) instructions">; 77 "Enable support for CRC instructions">; [all …]
|
| /NextBSD/contrib/subversion/subversion/libsvn_fs_x/ |
| HD | reps.c | 175 apr_array_header_t *instructions; member 207 const instruction_t *instructions; member 400 result->instructions = apr_array_make(result_pool, 0, in svn_fs_x__reps_builder_create() 453 APR_ARRAY_PUSH(builder->instructions, instruction_t) = instruction; in add_new_text() 498 if ( builder->instructions->nelts + 2 * contents->len / MATCH_BLOCKSIZE in svn_fs_x__reps_add() 503 rep.first_instruction = (apr_uint32_t)builder->instructions->nelts; in svn_fs_x__reps_add() 555 APR_ARRAY_PUSH(builder->instructions, instruction_t) = instruction; in svn_fs_x__reps_add() 563 rep.instruction_count = (apr_uint32_t)builder->instructions->nelts in svn_fs_x__reps_add() 582 + builder->instructions->nelts * 2 in svn_fs_x__reps_estimate_size() 605 for (instruction = container->instructions + instruction_idx; in get_text() [all …]
|
| /NextBSD/contrib/llvm/lib/Target/Hexagon/ |
| HD | Hexagon.td | 41 // PredRel - Filter class used to relate non-predicated instructions with their 44 // PredNewRel - Filter class used to relate predicated instructions with their 47 // ImmRegRel - Filter class used to relate instructions having reg-reg form 50 // NewValueRel - Filter class used to relate regular store instructions with 53 // NewValueRel - Filter class used to relate load/store instructions having 58 // Generate mapping table to relate non-predicate instructions with their 68 // The key column is the unpredicated instructions. 75 // Generate mapping table to relate predicate-true instructions with their 87 // Generate mapping table to relate predicate-false instructions with their 99 // Generate mapping table to relate predicated instructions with their .new [all …]
|
| /NextBSD/contrib/llvm/lib/Target/X86/ |
| HD | X86.td | 35 "Enable conditional move instructions">; 42 "Enable MMX instructions">; 44 "Enable SSE instructions", 49 "Enable SSE2 instructions", 52 "Enable SSE3 instructions", 55 "Enable SSSE3 instructions", 58 "Enable SSE 4.1 instructions", 61 "Enable SSE 4.2 instructions", 64 "Enable 3DNow! instructions", 67 "Enable 3DNow! Athlon instructions", [all …]
|
| HD | X86InstrFormats.td | 242 // Attributes specific to X86 instructions... 407 // Templates for instructions that use a 16- or 32-bit segmented address as 427 // SI - SSE 1 & 2 scalar instructions 439 // AVX instructions have a 'v' prefix in the mnemonic 457 // AVX instructions have a 'v' prefix in the mnemonic 462 // SIi8 - SSE 1 & 2 scalar instructions - vex form available on AVX512 471 // AVX instructions have a 'v' prefix in the mnemonic 477 // PI - SSE 1 & 2 packed instructions 486 // AVX instructions have a 'v' prefix in the mnemonic 492 // MMXPI - SSE 1 & 2 packed instructions with MMX operands [all …]
|
| /NextBSD/contrib/llvm/lib/Target/PowerPC/ |
| HD | PPC.td | 52 "Enable 64-bit instructions">; 58 "Enable Altivec instructions">; 60 "Enable SPE instructions">; 82 "Enable the fri[mnpz] instructions">; 84 "Enable fc[ft]* (unsigned and single-precision) and lfiwzx instructions">; 88 "Enable the popcnt[dw] instructions">; 92 "Enable extended divide instructions">; 100 "Enable Book E instructions", 106 "Enable E500/E500mc instructions">; 108 "Enable PPC 4xx instructions">; [all …]
|
| /NextBSD/contrib/gcc/config/rs6000/ |
| HD | rs6000.opt | 49 Use PowerPC General Purpose group optional instructions 53 Use PowerPC Graphics group optional instructions 65 Use PowerPC V2.02 floating point rounding instructions 69 Use AltiVec instructions 73 Use 4xx half-word multiply instructions 81 Generate load/store multiple instructions 85 Generate string instructions for block moves 105 Do not generate load/store with update instructions 109 Generate load/store with update instructions 113 Do not generate fused multiply/add instructions [all …]
|
| /NextBSD/contrib/gcc/config/mips/ |
| HD | 4130.md | 4 ;; The processor issues each 8-byte aligned pair of instructions together, 6 ;; want two instructions to issue in parallel, we need to make sure that the 15 ;; can change the addresses of many instructions. 20 ;; through the function looking for pairs of instructions that could 32 ;; (a) dependent instructions are separated by a non-dependent 35 ;; (b) instructions that use the multiplication unit are separated 36 ;; by non-multiplication instructions; and 38 ;; (c) memory access instructions are separated by non-memory 39 ;; instructions. 41 ;; The idea is to keep conflicting instructions apart wherever possible
|
| HD | mips.opt | 32 Use PMC-style 'mad' instructions 40 Use Branch Likely instructions, overriding the architecture default 52 Use trap instructions to check for integer divide by zero 56 Allow hardware floating-point instructions to cover both 32-bit and 64-bit operations 60 Use MIPS-DSP instructions 126 Generate floating-point multiply-add instructions 138 Allow the use of hardware floating-point instructions 150 Use MIPS-3D instructions 182 Do not use MIPS-3D instructions 186 Use paired-single floating-point instructions [all …]
|
| /NextBSD/contrib/binutils/gas/doc/ |
| HD | c-mips.texi | 30 * MIPS autoextend:: Directives for extending MIPS 16 bit instructions 34 generation of MIPS ASE instructions 111 set, 64-bit instructions generate a trap. Also, some 32-bit OSes only 133 provides a number of new instructions which target smartcard and 141 This tells the assembler to accept MIPS-3D instructions. 147 This tells the assembler to accept MDMX instructions. 153 This tells the assembler to accept DSP Release 1 instructions. 160 This tells the assembler to accept DSP Release 2 instructions. 166 This tells the assembler to accept MT instructions. 172 of an mfhi or mflo instruction occurs in the following two instructions. [all …]
|
| HD | c-i386.texi | 67 By default, x86 GAS replaces multiple nop instructions used for 68 alignment within code sections with multi-byte nop instructions such 83 instructions. The following architectures are recognized: 109 This option only affects instructions generated by the assembler. The 116 conjunction with the @option{-march} option, only instructions 181 previous Unix assemblers. Note that instructions with more than one 201 @cindex return instructions, i386 203 @cindex return instructions, x86-64 243 Almost all instructions have the same names in AT&T and Intel format. 245 instructions need two sizes to specify them. They need a size to [all …]
|
| HD | c-ppc.texi | 29 core instruction set, but including a few additional instructions at 31 instructions each variant supports, please see the chip's architecture 53 Generate code for PowerPC 440. BookE and some 405 instructions. 65 Generate code for Motorola SPE instructions. 80 Generate code for processors with AltiVec instructions. 95 Generate code Power/PowerPC common instructions.
|
| /NextBSD/contrib/llvm/lib/Target/AArch64/ |
| HD | AArch64.td | 27 "Enable Advanced SIMD instructions", [FeatureFPARMv8]>; 30 "Enable cryptographic instructions">; 33 "Enable ARMv8 CRC-32 checksum instructions">; 35 /// Cyclone has register move instructions which are "free". 39 /// Cyclone has instructions which zero registers for "free". 41 "Has zero-cycle zeroing instructions">; 48 "Support ARM v8.1a instructions", [FeatureCRC]>;
|
| /NextBSD/usr.bin/bc/ |
| HD | bc.y | 100 static struct tree *instructions = NULL; variable 774 p = realloc(instructions, newsize * sizeof(*p)); 776 free(instructions); 779 instructions = p; 789 instructions[current].index = CONST_STRING; in cs() 790 instructions[current].u.cstr = str; in cs() 799 instructions[current].index = ALLOC_STRING; in as() 800 instructions[current].u.astr = strdup(str); in as() 801 if (instructions[current].u.astr == NULL) in as() 816 instructions[current++].index = arg; in node() [all …]
|
| /NextBSD/contrib/gcc/config/arm/ |
| HD | arm1026ejs.md | 31 ;; instructions is "true", i.e., that all of the instructions are 60 ;; ALU instructions require three cycles to execute, and use the ALU 92 ;; Multiplication instructions loop in the execute stage until the 96 ;; The result of the "smul" and "smulw" instructions is not available 103 ;; The "smlaxy" and "smlawx" instructions require two iterations through 111 ;; The "smlalxy", "mul", and "mla" instructions require two iterations 119 ;; The "muls" and "mlas" instructions loop in the execute stage for 127 ;; Long multiply instructions that produce two registers of 130 ;; word. That fact is not modeled; instead, the instructions are 134 ;; The "umull", "umlal", "smull", and "smlal" instructions all take [all …]
|
| HD | arm1020e.md | 31 ;; instructions is "true", i.e., that all of the instructions are 60 ;; ALU instructions require three cycles to execute, and use the ALU 92 ;; Multiplication instructions loop in the execute stage until the 96 ;; The result of the "smul" and "smulw" instructions is not available 103 ;; The "smlaxy" and "smlawx" instructions require two iterations through 111 ;; The "smlalxy", "mul", and "mla" instructions require two iterations 119 ;; The "muls" and "mlas" instructions loop in the execute stage for 127 ;; Long multiply instructions that produce two registers of 130 ;; word. That fact is not modeled; instead, the instructions are 134 ;; The "umull", "umlal", "smull", and "smlal" instructions all take [all …]
|
| HD | arm926ejs.md | 31 ;; instructions is "true", i.e., that all of the instructions are 52 ;; ALU instructions require three cycles to execute, and use the ALU 78 ;; Multiplication instructions loop in the execute stage until the 117 ;; The models for load/store instructions do not accurately describe 169 ;; Branch instructions are difficult to model accurately. The ARM
|
| /NextBSD/contrib/binutils/opcodes/ |
| HD | arc-ext.c | 49 if (!arc_extension_map.instructions[opcode]) in arcExtMap_instName() 51 *flags = arc_extension_map.instructions[opcode]->flags; in arcExtMap_instName() 52 return arc_extension_map.instructions[opcode]->name; in arcExtMap_instName() 127 insn = arc_extension_map.instructions[i]; in cleanup_ext_map() 193 arc_extension_map.instructions[(int) opcode] = insn; in arcExtMap_add()
|
| HD | ChangeLog-2006 | 97 * m68k-opc.c (m68k_opcodes): Place trap instructions before set 190 cell specific instructions. Add {st,l}x{r,l}{,l} cell specific 191 VMX instructions. 222 set, don't bother printing X. Add new iwmmxt instructions. 275 (OP_MXC): New function to handle cvt* (convert instructions) between 316 * i386-dis.c (dis386): Add support for 4 operand instructions. Add 317 support for amdfam10 SSE4a/ABM instructions. Modify all 319 prefix for non-string instructions. 369 instructions. 393 * mips-dis.c: Disassemble DSP64 instructions for MIPS64R2. [all …]
|
| /NextBSD/contrib/gcc/config/i386/ |
| HD | ppro.md | 51 ;; Since the P6 CPUs execute instructions out-of-order, the most important 57 ;; - Find a less crude way to model complex instructions, in 71 ;; Simple instructions of the register-register form have only one uop. 72 ;; Load instructions are also only one uop. Store instructions decode to 73 ;; two uops, and simple read-modify instructions also take two uops. 74 ;; Simple instructions of the register-memory form have two to three uops. 75 ;; Simple read-modify-write instructions have four uops. The rules for 85 ;; in each cycle, to decode as many instructions per cycle as possible. 96 ;; Most instructions can be decoded on any of the three decoders. 122 ;; Only the irregular instructions have to be modeled here. A load [all …]
|
| HD | pentium.md | 36 ;; and doesn't hurt much on MMX. (Prefixed instructions are not very 93 ;; Floating point instructions can overlap with new issue of integer 94 ;; instructions. We model only first cycle of FP pipeline, as it is 134 ;; Few common long latency instructions 147 ;; latency of these instructions and not modeling the latency 185 ;; Push and pop instructions have 1 cycle latency and special 187 ;; and call instructions. 213 ;; in FP pipeline allowing other instructions to be executed. 225 ;; Long latency FP instructions overlap with integer instructions, 239 ;; Integer instructions. Load/execute/store takes 3 cycles,
|
| /NextBSD/contrib/llvm/lib/Target/Mips/ |
| HD | MSA.txt | 7 (semantically equivalent) instructions to be used in place of the requested 14 example, two instructions might be equally valid for some given IR and one is 31 instructions will be selected instead of vshf.[bhwd]. Unlike the ilv*, 32 and pck* instructions, this is matched from MipsISD::VSHF instead of 68 In future, the compiler may choose between these three instructions 71 between the instructions and the vselect node in one place:
|
| /NextBSD/contrib/binutils/include/opcode/ |
| HD | ChangeLog-9103 | 333 instructions. 338 (PPC_OPCODE_EFS): New opcode type for efs* instructions. 339 (PPC_OPCODE_CLASSIC): New opcode type for Classic PowerPC instructions. 400 instructions. 501 (OPCODE_IS_MEMBER): Allow instructions matching INSN_SB1 512 instructions, respectively. 644 * i386.h (i386_optab): SSE integer converison instructions have 710 * i386.h (i386_optab): Fix 64bit pushf template; Add instructions 715 * i386.h (i386_optab): Add "rex*" instructions; 716 add swapgs; disable jmp/call far direct instructions for [all …]
|
| /NextBSD/contrib/gcc/config/sparc/ |
| HD | sparc.opt | 48 Use hardware quad FP instructions 52 Do not use hardware quad fp instructions 88 Optimize tail call instructions in assembler and linker 125 ;; Generate code that uses the V8 instructions deprecated
|
| /NextBSD/contrib/gcc/ |
| HD | params.def | 89 of a function counted in internal gcc instructions (not in 90 real machine instructions) that is eligible for inlining 96 call other functions, the already inlined instructions are 102 "The maximum number of instructions in a single function eligible for inlining", 114 "The maximum number of instructions when automatically inlining", 119 "The maximum number of instructions inline function can grow to via recursive inlining", 124 "The maximum number of instructions non-inline function can grow to via recursive inlining", 149 /* The maximum number of instructions to consider when looking for an 151 number of instructions is searched, the time savings from filling 157 "The maximum number of instructions to consider to fill a delay slot", [all …]
|