| /trueos/contrib/gcc/config/i386/ |
| HD | athlon.md | 3 ;; The Athlon does contain three pipelined FP units, three integer units and 10 ;; Three DirectPath instructions decoders and only one VectorPath decoder 15 ;; it to the specialized integer (18 entry) and fp (36 entry) schedulers. 23 (and (eq_attr "type" "push") 26 (and (eq_attr "type" "fmov") 27 (and (eq_attr "memory" "load,store") 54 ;; to decode when decoder2 and decoder0 in next cycle 59 ;; and other units, we model decoder as two stage fully pipelined unit 60 ;; and only double decoded instruction may occupy unit in the first cycle. 66 ;; We solve that by specialized vector decoder unit and exclusion set. [all …]
|
| HD | ppro.md | 6 ;; GCC is free software; you can redistribute it and/or modify 22 ;; and Xeon lines of CPUs. The DFA scheduler description in this file is 39 ;; So, the P6 CPUs have out-of-order cores, but the instruction decoder and 62 ;; - Figure out where the p0 and p1 reservations come from. These 67 ;; The ppro_idiv and ppro_fdiv automata are used to model issue 68 ;; latencies of idiv and fdiv type insns. 73 ;; two uops, and simple read-modify instructions also take two uops. 82 ;; decoder 0, and this takes an unspecified number of cycles. 91 ;; decoder1 and decoder2 from being reserved until decoder 0 is 127 ;; reg-reg operation, 1 uop per load on port 2. and 2 uops per store [all …]
|
| HD | k6.md | 7 ;; GCC is free software; you can redistribute it and/or modify 23 ;; that there are only two decoders and they seems to be much slower than 40 ;; completely and it is only used to model lea operation. 51 ;; The fp unit is not pipelined, and it can only do one operation per two 55 ;; the old description, and a lot easier to extend to something more 66 ;; purposes, the long and vector decoder can be modelled as one decoder. 80 ;; Shift instructions and certain arithmetic are issued only on Integer X. 82 (and (eq_attr "cpu" "k6") 83 (and (eq_attr "type" "ishift,ishift1,rotate,rotate1,alu1,negnot,cld") 88 (and (eq_attr "cpu" "k6") [all …]
|
| HD | pentium.md | 6 ;; GCC is free software; you can redistribute it and/or modify 36 ;; and doesn't hurt much on MMX. (Prefixed instructions are not very 43 (and (eq_attr "type" "pop,push") 50 (and (eq_attr "type" "ishift") 53 (and (eq_attr "type" "rotate") 56 (and (eq_attr "type" "ishift1") 59 (and (eq_attr "type" "rotate1") 62 (and (eq_attr "type" "call") 65 (and (eq_attr "type" "callv") 73 ;; Pentium do have U and V pipes. Instruction to both pipes [all …]
|
| HD | geode.md | 7 ;; GCC is free software; you can redistribute it and/or modify 40 ;; There is also memory management unit and execution pipeline for 41 ;; load/store operations. We ignore it and difference between insns 42 ;; using memory and registers. 49 (and (eq_attr "cpu" "geode") 54 (and (eq_attr "cpu" "geode") 59 (and (eq_attr "cpu" "geode") 64 (and (eq_attr "cpu" "geode") 70 (and (eq_attr "cpu" "geode") 75 (and (eq_attr "cpu" "geode") [all …]
|
| /trueos/contrib/dialog/samples/ |
| HD | wheel | 27 --begin `scaley 27` `scalex 98` --infobox "pushd /var/log >/dev/null" 0 0 --and-widget \ 28 --begin `scaley 35` `scalex 95` --infobox "mkdir -p news -m 755" 0 0 --and-widget \ 29 --begin `scaley 45` `scalex 86` --infobox "chown news.news news" 0 0 --and-widget \ 30 --begin `scaley 48` `scalex 78` --infobox "cd /var/log/news" 0 0 --and-widget \ 31 --begin `scaley 51` `scalex 61` --infobox "mkdir -p OLD -m 755" 0 0 --and-widget \ 32 --begin `scaley 52` `scalex 47` --infobox "chown news.news OLD" 0 0 --and-widget \ 33 --begin `scaley 51` `scalex 40` --infobox "cd /var/spool" 0 0 --and-widget \ 34 --begin `scaley 48` `scalex 25` --infobox "mkdir -p news -m 775" 0 0 --and-widget \ 35 --begin `scaley 42` `scalex 13` --infobox "chown news.news news" 0 0 --and-widget \ 36 --begin `scaley 35` `scalex 4` --infobox "cd /var/spool/news" 0 0 --and-widget \ [all …]
|
| /trueos/contrib/gcc/config/mips/ |
| HD | 5500.md | 19 (and (eq_attr "cpu" "r5500") 25 (and (eq_attr "cpu" "r5500") 30 (and (eq_attr "cpu" "r5500") 40 (and (eq_attr "cpu" "r5500") 47 (and (eq_attr "cpu" "r5500") 53 (and (eq_attr "cpu" "r5500") 58 (and (eq_attr "cpu" "r5500") 68 (and (eq_attr "cpu" "r5500") 73 (and (eq_attr "cpu" "r5500") 80 (and (eq_attr "cpu" "r5500") [all …]
|
| HD | 5400.md | 17 (and (eq_attr "cpu" "r5400") 23 (and (eq_attr "cpu" "r5400") 28 (and (eq_attr "cpu" "r5400") 33 (and (eq_attr "cpu" "r5400") 38 (and (eq_attr "cpu" "r5400") 46 (and (eq_attr "cpu" "r5400") 52 (and (eq_attr "cpu" "r5400") 57 (and (eq_attr "cpu" "r5400") 62 (and (eq_attr "cpu" "r5400") 67 (and (eq_attr "cpu" "r5400") [all …]
|
| HD | sb1.md | 7 ;; pipes, and 1 MDMX pipes. It can issue 2 ls insns and 2 exe/fpu/mdmx insns 13 ;; simple alu operations issue to ls1 if it is still available, and their 21 ;; In 32-bit mode, dependent FP can't co-issue with load, and only one FP exe 44 ;; The divide unit is not pipelined, and blocks hi/lo reads and writes. 53 ;; Can only issue to one of the ex and fp pipes at a time. 58 ;; We can use 2 FP pipes only if we have 64-bit FP code, and exceptions are 62 (cond [(and (ne (symbol_ref "TARGET_FLOAT64") (const_int 0)) 78 ;; ??? Try limiting scheduler to 2 long latency operations, and see if this 79 ;; results in a usable DFA, and whether it helps code performance. 93 (and (eq_attr "cpu" "sb1,sb1a") [all …]
|
| HD | 5k.md | 3 ;; and Nigel Stephens (nigel@mips.com) 17 ;; GCC is free software; you can redistribute it and/or modify it 40 (and (eq_attr "cpu" "5kc,5kf") 45 (and (eq_attr "cpu" "5kc,5kf") 50 (and (eq_attr "cpu" "5kc,5kf") 56 (and (eq_attr "cpu" "5kc,5kf") 57 (and (eq_attr "type" "idiv") 62 (and (eq_attr "cpu" "5kc,5kf") 63 (and (eq_attr "type" "idiv") 70 (and (eq_attr "cpu" "5kc,5kf") [all …]
|
| HD | 9000.md | 6 ;; GCC is free software; you can redistribute it and/or modify it 29 ;; F and M pipe units, for instructions that must be issued by a 44 (and (eq_attr "cpu" "r9000") 49 (and (eq_attr "cpu" "r9000") 54 (and (eq_attr "cpu" "r9000") 59 (and (eq_attr "cpu" "r9000") 60 (and (eq_attr "type" "condmove") 64 ;; This applies to both 'mul' and 'mult'. 66 (and (eq_attr "cpu" "r9000") 67 (and (eq_attr "type" "imul,imul3,imadd") [all …]
|
| HD | 24k.md | 3 ;; and David Ung (davidu@mips.com) 15 ;; GCC is free software; you can redistribute it and/or modify it 45 (and (eq_attr "cpu" "24k,24kx") 50 ;; 2. Arithmetic: add, addi, addiu, addiupc, addu, and, andi, clo, clz, 57 (and (eq_attr "cpu" "24k,24kx") 65 (and (eq_attr "cpu" "24k,24kx") 71 (and (eq_attr "cpu" "24k,24kx") 79 (and (eq_attr "cpu" "24k,24kx") 85 (and (eq_attr "cpu" "24k,24kx") 91 (and (eq_attr "cpu" "24k,24kx") [all …]
|
| HD | 7000.md | 6 ;; GCC is free software; you can redistribute it and/or modify it 90 (and (eq_attr "cpu" "r7000") 95 (and (eq_attr "cpu" "r7000") 100 (and (eq_attr "cpu" "r7000") 105 (and (eq_attr "cpu" "r7000") 106 (and (eq_attr "type" "idiv") 111 (and (eq_attr "cpu" "r7000") 112 (and (eq_attr "type" "idiv") 117 (and (eq_attr "cpu" "r7000") 118 (and (eq_attr "type" "imul,imadd") [all …]
|
| HD | sr71k.md | 14 ;; published latencies. Emulation of out-of-order issue and the insn 34 ;; Yes, all the instructions in Table 19-1 only go to ALUX, and all the 48 ;; Floating point stores go to Ld/St and go to MOV in the floating point 51 ;; Floating point loads go to Ld/St and go to LOAD in the floating point 58 ;; RH> What about for RSQRT single and double? 60 ;; The latency/repeat for RECIP and RSQRT are correct. 64 ;; Use four automata to isolate long latency operations, and to 69 ;; feeders for CPU function units and feeders for fpu (CP1 interface) 128 (and (eq_attr "cpu" "sr71000") 135 (and (eq_attr "cpu" "sr71000") [all …]
|
| HD | 4300.md | 6 ;; GCC is free software; you can redistribute it and/or modify it 26 (and (eq_attr "cpu" "r4300") 31 (and (eq_attr "cpu" "r4300") 32 (and (eq_attr "type" "imul,imul3,imadd") 37 (and (eq_attr "cpu" "r4300") 38 (and (eq_attr "type" "imul,imul3,imadd") 43 (and (eq_attr "cpu" "r4300") 44 (and (eq_attr "type" "idiv") 49 (and (eq_attr "cpu" "r4300") 50 (and (eq_attr "type" "idiv") [all …]
|
| HD | 4k.md | 3 ;; and David Ung (davidu@mips.com) 9 ;; 4Kc - pipelined multiplier and translation lookaside buffer (TLB) 10 ;; 4km - pipelined multiplier and block address translator (BAT) 11 ;; 4kp - non-pipelined multiplier and block address translator (BAT) 17 ;; GCC is free software; you can redistribute it and/or modify it 39 (and (eq_attr "cpu" "4kc,4kp") 44 (and (eq_attr "cpu" "4kc,4kp") 49 (and (eq_attr "cpu" "4kc,4kp") 57 (and (eq_attr "cpu" "4kc") 58 (and (eq_attr "type" "idiv") [all …]
|
| /trueos/contrib/gcc/config/arm/ |
| HD | arm-generic.md | 6 ;; GCC is free software; you can redistribute it and/or modify it 28 ; There is room in the buffer for up to two addresses and up to eight words 32 ; It is normally the case that FCLK and MCLK will be in the ratio 2:1, so 33 ; writes will take 2 FCLK cycles per word, if FCLK and MCLK are asynchronous 42 ; The f_mem_r and r_mem_f could also block, but they are to the stack, 51 (and (eq_attr "generic_sched" "yes") 52 (and (eq_attr "model_wbuf" "yes") 57 (and (eq_attr "generic_sched" "yes") 58 (and (eq_attr "model_wbuf" "yes") 63 (and (eq_attr "generic_sched" "yes") [all …]
|
| /trueos/contrib/gcc/doc/ |
| HD | contrib.texi | 21 and iterators. 24 John David Anglin for threading-related fixes and improvements to 25 libstdc++-v3, and the HP-UX port. 32 Abramo and Roberto Bagnara for the SysV68 Motorola 3300 Delta Series 49 Godmar Back for his Java improvements and encouragement. 65 Geoff Berry for his Java object serialization work and various patches. 68 Uros Bizjak for the implementation of x87 math built-in functions and 69 for various middle end and i386 back end improvements and bugfixes. 72 Eric Blake for helping to make GCJ and libgcj conform to the 83 garbage collector}, IA-64 libffi port, and other Java work. [all …]
|
| /trueos/contrib/gcc/config/rs6000/ |
| HD | 6xx.md | 2 ;; and PowerPC 630 processors. 7 ;; GCC is free software; you can redistribute it and/or modify it 30 ;; MCIU used for imul/idiv and moves from/to spr 35 ;; PPC604e is PPC604 with larger caches and a CRU. In the 604 52 (and (eq_attr "type" "load,load_ext,load_ext_u,load_ext_ux,load_ux,load_u") 57 (and (eq_attr "type" "fpload,fpload_ux,fpload_u") 62 (and (eq_attr "type" "store,fpstore,store_ux,store_u,fpstore_ux,fpstore_u") 67 (and (eq_attr "type" "load_l,store_c") 72 (and (eq_attr "type" "load_l,store_c") 77 (and (eq_attr "type" "integer,insert_word") [all …]
|
| HD | rios1.md | 6 ;; GCC is free software; you can redistribute it and/or modify it 29 (and (eq_attr "type" "load,load_ext,load_ext_u,load_ext_ux,load_ux,load_u,\ 35 (and (eq_attr "type" "store,store_ux,store_u") 40 (and (eq_attr "type" "fpload,fpload_ux,fpload_u") 45 (and (eq_attr "type" "fpload,fpload_ux,fpload_u") 50 (and (eq_attr "type" "fpstore,fpstore_ux,fpstore_u") 55 (and (eq_attr "type" "integer,insert_word") 60 (and (eq_attr "type" "two") 65 (and (eq_attr "type" "three") 70 (and (eq_attr "type" "imul,imul_compare") [all …]
|
| HD | rs64.md | 6 ;; GCC is free software; you can redistribute it and/or modify it 30 (and (eq_attr "type" "load,load_ext,load_ext_u,load_ext_ux,load_ux,load_u") 35 (and (eq_attr "type" "store,store_ux,store_u,fpstore,fpstore_ux,fpstore_u") 40 (and (eq_attr "type" "fpload,fpload_ux,fpload_u") 45 (and (eq_attr "type" "load_l,store_c") 50 (and (eq_attr "type" "integer,insert_word") 55 (and (eq_attr "type" "two") 60 (and (eq_attr "type" "three") 65 (and (eq_attr "type" "imul,imul_compare") 70 (and (eq_attr "type" "imul2") [all …]
|
| /trueos/contrib/gcc/config/sparc/ |
| HD | ultra1_2.md | 6 ;; GCC is free software; you can redistribute it and/or modify 21 ;; UltraSPARC-I and II are quad-issue processors. Interesting features 27 ;; - Two integer units. Only one of them can do shifts, and the other 31 ;; setting instruction and one that does set the condition codes. The 61 ;; If we have to schedule an ieu1 specific instruction and we want 74 ;; execute in the first three slots, FPU and branches can go into 82 (and (eq_attr "cpu" "ultrasparc") 87 (and (eq_attr "cpu" "ultrasparc") 92 (and (eq_attr "cpu" "ultrasparc") 97 (and (eq_attr "cpu" "ultrasparc") [all …]
|
| /trueos/contrib/binutils/gas/ |
| HD | CONTRIBUTORS | 3 If you've contributed to gas and your name isn't listed here, it is 5 nickc@redhat.com and I'll correct the situation. 8 the documentation, and info on specific files will go into an AUTHORS 16 gdb-specific debug information and the 68k series machines, most of 17 the preprocessing pass, and extensive changes in messages.c, 21 enhancements and many bug fixes, including merging support for several 24 and b.out backends), adding configuration including heavy testing and 25 verification of cross assemblers and file splits and renaming, 29 sparc opcode file rewrite, decstation, rs6000, and hp300hpux host 30 ports, updated "know" assertions and made them work, much other [all …]
|
| /trueos/contrib/nvi/docs/internals/ |
| HD | autowrite | 11 ^Z Y Y Write file and suspend. 16 ^^ Y Y Write file and jump. 21 # This behavior is identical to :tag, :tagpop, and :tagpush with 23 ^] Y Y Write file and jump. 28 :! Y Y Write file and execute. 29 :! Y N Warn (if warn option) and execute. 33 Commands that are affected by both autowrite and force: 42 # In nvi, :next and :prev commands behave identically to :rewind. 43 :next Y Y Y Write changes and jump. 44 :next Y Y N Write changes and jump. [all …]
|
| /trueos/crypto/openssl/ |
| HD | NEWS | 8 Major changes between OpenSSL 1.0.1n and OpenSSL 1.0.1o [12 Jun 2015] 12 Major changes between OpenSSL 1.0.1m and OpenSSL 1.0.1n [11 Jun 2015] 20 Major changes between OpenSSL 1.0.1l and OpenSSL 1.0.1m [19 Mar 2015] 30 Major changes between OpenSSL 1.0.1k and OpenSSL 1.0.1l [15 Jan 2015] 32 o Build fixes for the Windows and OpenVMS platforms 34 Major changes between OpenSSL 1.0.1j and OpenSSL 1.0.1k [8 Jan 2015] 45 Major changes between OpenSSL 1.0.1i and OpenSSL 1.0.1j [15 Oct 2014] 52 Major changes between OpenSSL 1.0.1h and OpenSSL 1.0.1i [6 Aug 2014] 64 Major changes between OpenSSL 1.0.1g and OpenSSL 1.0.1h [5 Jun 2014] 73 Major changes between OpenSSL 1.0.1f and OpenSSL 1.0.1g [7 Apr 2014] [all …]
|