carl
1f1145d30e
- no longer used files removed
2002-08-11 07:02:00 +00:00
florian
3ec544be88
* fixed powerpc compilation problems
2002-08-11 06:14:40 +00:00
Jonas Maebe
ee7d5f7b2a
* various fixes and optimizations
2002-08-10 17:15:31 +00:00
Jonas Maebe
bedbc15e75
* endianess fix
2002-08-10 17:15:06 +00:00
Jonas Maebe
f1f27e99b4
+ abs, sqr, sqrt implementations
2002-08-10 17:15:00 +00:00
carl
18b102ce95
+ moved target_cpu_string to cpuinfo
...
* renamed asmmode enum.
* assembler reader has now less ifdef's
* move from nppcmem.pas -> ncgmem.pas vec. node.
2002-08-10 14:46:29 +00:00
florian
b136cecacd
* first part of ppc calling conventions fix
2002-08-06 20:55:20 +00:00
Jonas Maebe
e7d92a1aa7
* fixed bug in g_flags2reg()
...
* and yet more constant operation fixes :)
2002-08-06 07:12:05 +00:00
Jonas Maebe
10c4e8eb15
* fixed compilation problems
2002-08-05 08:58:53 +00:00
Jonas Maebe
8ff3e3e1b3
* more misc. fixes, mostly constant-related
2002-08-04 12:57:55 +00:00
Jonas Maebe
54011b5bea
* some misc constant fixes
2002-08-02 11:10:42 +00:00
florian
ed950491a8
* the code generator knows now if parameters are in registers
2002-07-30 20:50:43 +00:00
florian
e7a6cd18dd
* more fixes for the ppc
...
+ wrappers for the tcnvnode.first_* stuff introduced
2002-07-29 21:23:42 +00:00
Jonas Maebe
07089fadb7
+ nppcmem
2002-07-29 09:22:20 +00:00
Jonas Maebe
a1a3283c61
+ tppcvecnode, almost straight copy of the i386 code, can most likely
...
be made generic if all treference type allow a base, index and offset
2002-07-29 09:21:30 +00:00
Jonas Maebe
04cb71275a
+ second_int_to_int implementation which is almost the same as the
...
generic implementation, but it avoids some unnecessary type conversions
2002-07-29 09:20:20 +00:00
florian
37351a4590
- removed debug code which was commited by accident
2002-07-28 21:38:30 +00:00
florian
463b355ba3
* more powerpc fixes
...
+ dummy tcgvecnode
2002-07-28 21:34:31 +00:00
florian
3fe0229843
+ added direct assembler reader for PowerPC
2002-07-28 20:45:22 +00:00
Jonas Maebe
046ee7bf2b
+ 64 bit operations (badly tested), everything is implemented now!
...
* some small fixes
2002-07-28 16:02:49 +00:00
Jonas Maebe
e3cbc3cf22
+ tcg64fppc.a_op64_const_reg_reg() and tcg64fppc.a_op64_reg_reg_reg()
...
* several fixes, most notably in a_load_reg_reg(): it didn't do any
conversion from smaller to larger sizes or vice versa
* some small optimizations
2002-07-28 16:01:59 +00:00
Jonas Maebe
ccf4a67d81
+ second_addboolean(), second_addfloat() and second_addsmallset()
...
(64bit stuff is all that's left to do)
2002-07-27 20:00:59 +00:00
Jonas Maebe
b4cfa73079
* fixed a_loadaddr_ref_reg()
...
* fixed g_flags2reg()
* optimized g_concatcopy()
2002-07-27 19:59:29 +00:00
Jonas Maebe
1b983ee56e
* some typo corrections in the instruction tables
...
* renamed the m* registers to v*
2002-07-27 19:57:18 +00:00
Jonas Maebe
6b83738e7d
+ generic implementation of tcg.g_flags2ref()
...
* tcg.flags2xxx() now also needs a size parameter
2002-07-27 19:53:51 +00:00
florian
1b5435381b
* several PowerPC related fixes to get forward with system unit compilation
2002-07-26 22:22:10 +00:00
florian
eb2ca8d10f
* rewrote the system handling
2002-07-26 21:15:37 +00:00
Jonas Maebe
0d740d4de8
+ intial implementation of add nodes, only integer/enumeration/pointer/...
...
handling is finished
2002-07-26 12:31:56 +00:00
Jonas Maebe
83ce7360ed
* fixed typo in instruction table (_subco_ -> a_subco)
2002-07-26 12:30:51 +00:00
Jonas Maebe
e2c5923e97
* fixed range errors
2002-07-26 11:19:57 +00:00
Jonas Maebe
61a2a3ae37
* fixed bug in shl/shr code
2002-07-26 10:48:34 +00:00
carl
448d40de1a
+ FPURESULTREG -> FPU_RESULT_REG
2002-07-25 17:56:29 +00:00
florian
156fc666c8
* small typo fixed, compiles with 1.0.x again
2002-07-24 14:38:00 +00:00
Jonas Maebe
c1e9bd36a3
* make sure we use rlwi* when possible instead of andi.
2002-07-21 17:00:23 +00:00
Jonas Maebe
33e3b0dda0
+ include ncgset unit
2002-07-21 16:58:59 +00:00
Jonas Maebe
3eb6dc156e
* hopefully final fix for second_int_to_real()
2002-07-21 16:57:22 +00:00
Jonas Maebe
398195464a
* fixed bugs with writing out unconditinal jumps
2002-07-21 16:56:20 +00:00
florian
59abf2555b
* types.pas renamed to defbase.pas because D6 contains a types
...
unit so this would conflicts if D6 programms are compiled
+ Willamette/SSE2 instructions to assembler added
2002-07-20 11:57:52 +00:00
florian
e161eb5e25
+ initial version, a lot of instructions need to be added
2002-07-13 21:50:34 +00:00
florian
15b9d096f9
* some more generic calling stuff fixed
2002-07-13 19:38:43 +00:00
Jonas Maebe
79c0a402ef
* fixed fpu constants in second_int_to_real (fpu values are also stored
...
in big endian)
2002-07-13 06:49:39 +00:00
florian
f0e2409c9f
* fixed to compile with 1.1
2002-07-12 22:02:22 +00:00
Jonas Maebe
4f77082660
* changed motorola syntax of references with symbols to GNU syntax
2002-07-12 10:10:01 +00:00
florian
336808f6c3
* start of the new generic parameter handling
2002-07-11 14:41:27 +00:00
Jonas Maebe
82e23e60f5
* fixed nppccnv and enabled it
...
- removed PPC specific second_int_to_int and use the generic one instead
2002-07-11 07:42:31 +00:00
Jonas Maebe
17f7d6e8e0
* fixed tppcmoddivnode
...
* fixed 64bit parts of tppcshlshrnode
2002-07-11 07:41:27 +00:00
Jonas Maebe
690ba14436
+ tcg64fpc implementation (only a_op64_reg_reg and a_op64_const_reg for
...
now)
* fixed and improved tcgppc.a_load_const_reg
* improved tcgppc.a_op_const_reg, tcgppc.a_cmp_const_reg_label
* A_CMP* -> A_CMPW* (this means that 32bit compares should be done)
2002-07-11 07:38:28 +00:00
Jonas Maebe
5ee1a2ea17
* some available registers fixes
2002-07-11 07:35:36 +00:00
Jonas Maebe
c018cb9116
* fixed mullw entry in instruction list
2002-07-11 07:34:55 +00:00
Jonas Maebe
3e7b6ef7fa
* unarynminus and shlshr node fixed for 32bit and smaller ordinals
...
* small fixes in the assembler writer
* changed scratch registers, because they were used by the linker (r11
and r12) and by the abi under linux (r31)
2002-07-09 19:45:01 +00:00
