| .. |
|
aasmcpu.pas
|
* Put under {$ifndef x86_64} more cases of instructions that do not exist in 64-bit mode.
|
2014-06-11 12:51:38 +00:00 |
|
agx86att.pas
|
Separate out nasm assembler for i8086, i386 and x86_64 cpus, also separte based on target object format
|
2014-01-21 00:26:08 +00:00 |
|
agx86int.pas
|
* Put under {$ifndef x86_64} more cases of instructions that do not exist in 64-bit mode.
|
2014-06-11 12:51:38 +00:00 |
|
agx86nsm.pas
|
* NASM writer: don't force 'near' size for all non-far branches and calls, it disables NASM's capability to optimize branches into short form where possible, and thus only increases size of generated code.
|
2014-06-11 12:32:21 +00:00 |
|
cga.pas
|
+ support for FMA intrinsic: if there is no hardware support, the compiler throws an error.
|
2014-04-13 19:21:54 +00:00 |
|
cgx86.pas
|
* tcgx86.a_op_const_reg_reg: optimize trivial cases before considering more complex ones.
|
2014-06-08 00:08:56 +00:00 |
|
cpubase.pas
|
+ Implemented IEEE 754-compliant checking for unordered results of floating-point compares on x86 targets. Mantis #9362.
|
2014-04-14 12:36:11 +00:00 |
|
hlcgx86.pas
|
|
|
|
itcpugas.pas
|
|
|
|
itx86int.pas
|
|
|
|
ni86mem.pas
|
* isolated segment-related functionality of tabsolutevarsym into i386/i8086-
|
2014-03-30 15:42:53 +00:00 |
|
nx86add.pas
|
* i386: For integer comparisons with zero, emit "test $-1,%reg" instead of "test %reg,%reg". It is more spilling-friendly, because it transforms into "test $-1,spilltemp" and does not require a register.
|
2014-04-20 19:16:58 +00:00 |
|
nx86cal.pas
|
* a_call_ref functionality cannot be implemented efficiently at code generator level, because references need specific preparations at earlier points. Moved this support to tcgcallnode and its x86 descendants, and got rid of all ifdef's around.
|
2014-02-03 13:28:56 +00:00 |
|
nx86cnv.pas
|
* x86: improve x87 qword to float conversion, using single-precision constants saves space and removes need in separate load on FPU stack. No precision loss occurs because 2**64 is representable exactly even in single precision.
|
2014-03-03 20:41:42 +00:00 |
|
nx86con.pas
|
|
|
|
nx86inl.pas
|
+ make use of vfnmsub*/vfmsub*/vfnmadd* instructions if possible
|
2014-05-02 21:07:46 +00:00 |
|
nx86mat.pas
|
* One more fix of operand size, likely harmless because shift amount is taken modulo bit-width anyway.
|
2014-06-14 13:31:48 +00:00 |
|
nx86mem.pas
|
* preserve the segment of the reference in tx86vecnode.update_reference_reg_mul
|
2014-04-05 14:46:35 +00:00 |
|
nx86set.pas
|
* Use GOT-relative constants for i386 PIC jump tables, they don't need runtime relocations. Now almost ABI-compliant on Linux/BSD (Darwin targets unchanged). Also clean up i8086-specific stuff: using tai_const.create_type_sym(aitconst_ptr,...) generates near pointers on i8086, which is the desired goal.
|
2014-03-03 21:06:49 +00:00 |
|
rax86.pas
|
* merged avx2 branch, developed by Torsten Grundke
|
2014-03-20 12:03:52 +00:00 |
|
rax86att.pas
|
|
|
|
rax86int.pas
|
* Put under {$ifndef x86_64} more cases of instructions that do not exist in 64-bit mode.
|
2014-06-11 12:51:38 +00:00 |
|
rgx86.pas
|
* Put under {$ifndef x86_64} more cases of instructions that do not exist in 64-bit mode.
|
2014-06-11 12:51:38 +00:00 |
|
symi86.pas
|
* i8086 and i386-specific code from tabstractprocdef.is_pushleftright moved to
|
2014-04-12 15:34:08 +00:00 |
|
symx86.pas
|
+ symx86 unit, forgot to commit (part of r27397)
|
2014-03-30 22:03:55 +00:00 |
|
x86ins.dat
|
+ prove of concept how FMA4 could be supported in inline assembler
|
2014-03-20 21:25:38 +00:00 |
|
x86reg.dat
|
|
|
