 sergei
|
ae627a4ba8
* tcgx86.a_op_const_reg_reg: optimize trivial cases before considering more complex ones.
|
vor 11 Jahren |
 nickysn
|
daf71e6d88
* fixed multiplication by 64-bit constants on x86_64, mantis #26230
|
vor 11 Jahren |
|
nickysn
|
6e7dd647fd
* Tcgx86.g_concatcopy: optimization for the case where source.segment is
|
vor 11 Jahren |
|
nickysn
|
1eb24f9d98
+ remove the segment prefix in make_simple_ref on i8086, in case it is the same
|
vor 11 Jahren |
|
sergei
|
07e90aaa24
+ Implemented IEEE 754-compliant checking for unordered results of floating-point compares on x86 targets. Mantis #9362.
|
vor 11 Jahren |
 florian
|
18233f2c0c
* fix compilation of i8086 compiler
|
vor 11 Jahren |
|
florian
|
d88d644925
+ support for FMA intrinsic: if there is no hardware support, the compiler throws an error.
|
vor 11 Jahren |
|
nickysn
|
afcb8c3f10
* initialize DS from CS in the interrupt procedure entry code in tiny model,
|
vor 11 Jahren |
|
nickysn
|
7a873f495d
* use the stack segment for SP-relative addresses in Tcgx86.g_concatcopy in
|
vor 11 Jahren |
|
nickysn
|
ee20ecfcf2
* tcgx86.make_simple_ref: in i8086 memory models with DS<>SS, add an SS: segment
|
vor 11 Jahren |
|
nickysn
|
b36a98ce51
+ Tcgx86.g_concatcopy: added support for memory models with DS<>SS
|
vor 11 Jahren |
|
nickysn
|
e9e4cc8bcb
* Tcgx86.g_concatcopy: use boolean variables saved_ds and saved_es to keep track
|
vor 11 Jahren |
|
nickysn
|
096bba90da
* fixed Tcgx86.g_concatcopy not using the correct push instruction size when
|
vor 11 Jahren |
 pierre
|
044891fd94
Avoid range check error in a_op_const_reg_reg
|
vor 11 Jahren |
|
nickysn
|
229ebf33a8
+ support getting the address of far references on i8086 in
|
vor 11 Jahren |
|
florian
|
37b34edbc6
* do not reuse registers when calculating addresses in pic mode
|
vor 11 Jahren |
|
sergei
|
9c1f917e3a
* 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.
|
vor 11 Jahren |
|
sergei
|
0d3f36eebf
- Remove references to global variable 'cg' from methods of tcg and some of its descendants.
|
vor 11 Jahren |
|
florian
|
28a8dc42c6
* fix tcgx86.a_op_reg_reg on x86-64 and make use of it
|
vor 11 Jahren |
|
florian
|
7d7bf1d877
+ make use of SHLX/SHRX on CPUs supporting BMI2
|
vor 11 Jahren |
|
florian
|
48ae2d215a
+ concatcopy variants using sse and avx, only activated if optimization for size is done because at least on an i7-4770 it has shown no benefit
|
vor 11 Jahren |
|
nickysn
|
85dd9e5789
+ added a size parameter to optimize_op_const and do a sign extension of the 'a' parameter up from the specified size, so that things like (i and $ffffffff) get optimized away the same way as (i and -1)
|
vor 11 Jahren |
|
sergei
|
6b1f021fcf
- Removed remaining optimizations that duplicate functionality of tcg.optimize_op_const.
|
vor 11 Jahren |
|
nickysn
|
1f7b14f38e
* fixed the conversion of powerof2 mul/imuls to lea in tcgx86.a_op_const_reg_reg
|
vor 11 Jahren |
|
nickysn
|
cc1faad3d2
+ make use of the 3-operand form of IMUL on i386 and x86_64
|
vor 11 Jahren |
|
nickysn
|
5d75bf4f92
* always use IMUL instead of MUL on i386, when doing a 32->32 multiplication
|
vor 11 Jahren |
|
nickysn
|
820b667e1b
- rm the OP_IMUL to SHL optimization from tcgx86.a_op_const_reg and tcgx86.a_op_const_ref, because it's already done in optimize_op_const, which is called in the beginning of both methods.
|
vor 11 Jahren |
|
florian
|
e210d5f30e
+ cpu_capabilites for x86_64 and i386
|
vor 11 Jahren |
|
sergei
|
f2096de53a
* tcg386.g_proc_exit: instead of recalculating stack size, use current_procinfo.final_localsize which was calculated in g_proc_entry.
|
vor 11 Jahren |
|
sergei
|
179586f589
+ SEH support for Win32. Enable by cycling with OPT=-dTEST_WIN32_SEH.
|
vor 11 Jahren |
