 Jeppe Johansen
|
a1a17447ff
- Fix bug in 64bit softfloat double negation.
|
há 6 anos atrás |
 Jonas Maebe
|
04d295f186
* first step towards supporting 32 bit targets with the LLVM code generator:
|
há 6 anos atrás |
|
Jonas Maebe
|
4d03f3a65e
* fixed LLVM memory leaks
|
há 6 anos atrás |
 florian
|
8c5606b41d
+ support mmx shifting
|
há 7 anos atrás |
|
florian
|
b1dff29cbf
* removed unused units
|
há 8 anos atrás |
 nickysn
|
0b1a54f924
* on 16-bit CPUs, avoid the unnecessary 32-bit conversion of the right shl/shr
|
há 8 anos atrás |
|
Jonas Maebe
|
8ff5371a6c
+ support mmregisters in the generic unaryminus code for floats
|
há 10 anos atrás |
|
Jonas Maebe
|
0fc1fd6ac1
* replaced current_procinfo.currtrue/falselabel with storing the true/false
|
há 10 anos atrás |
|
Jonas Maebe
|
61e4a1b811
+ added tasmlist parameter to getintparaloc() (needed for llvm)
|
há 10 anos atrás |
|
Jonas Maebe
|
67b8aceaee
* synchronized with privatetrunk till r30095
|
há 10 anos atrás |
|
Jonas Maebe
|
9f929695cf
* fixed tcgshlshrnode.pass_generate_code for 64 bit targets
|
há 10 anos atrás |
 sergei
|
5f2821ee5f
* Moved handling of emulated floating point negation into new virtual method tcgunaryminusnode.second_float_emulated. Also changed it into simple flipping of the sign bit, which is how it is defined by IEEE 754, and already done for ARM targets. Relevant ARM-specific parts can now be removed, and m68k code gets significantly better without further efforts.
|
há 11 anos atrás |
|
Jonas Maebe
|
bacd303208
* synchronized with trunk up to r27758
|
há 11 anos atrás |
|
nickysn
|
63f0731913
* 16/8-bit alu fix in tcgunaryminusnode.emit_float_sign_change
|
há 11 anos atrás |
|
nickysn
|
88df284423
* fixed overflow checking for the 64-bit unary minus on 16 and 8-bit platforms
|
há 11 anos atrás |
 Károly Balogh
|
075645d2b3
do not generate zero divisor checks if a divisor is a constant
|
há 11 anos atrás |
|
Jonas Maebe
|
fcbde1d6e9
+ hlcg/llvm support for unary minus
|
há 11 anos atrás |
|
Jonas Maebe
|
81427523ab
* pass a list of (pointers to) the paralocs to hlcgobj.a_call/g_call*, as
|
há 11 anos atrás |
|
Jonas Maebe
|
5729c9c8ef
* fixed shl/shr with constant value>=32 on 64 bit platforms
|
há 11 anos atrás |
|
Jonas Maebe
|
7033ddf679
* converted tshshrnode to thlcgobj
|
há 11 anos atrás |
|
florian
|
07579cf0b7
* cleanup
|
há 11 anos atrás |
|
florian
|
3dddd01309
* do not location_force_reg in tcgshlshrnode.second_integer if left is already in a register
|
há 11 anos atrás |
|
sergei
|
36d0c8a5a7
* x86 and SPARC: fixed handling 64-bit (qwordbool) values in tcgnotnode (partial fix for Mantis #25255).
|
há 11 anos atrás |
|
nickysn
|
988b742d71
* avoid the meaningless conversion of the right shl/shr argument to 32-bit on
|
há 11 anos atrás |
|
florian
|
4d5119bf1c
* fixes several issues which cause warnings by the dfa code when using it to detect uninitialized variables
|
há 11 anos atrás |
|
nickysn
|
cb81153cc4
* fixed tcgnotnode.second_64bit for 8/16-bit cpus after r24678
|
há 12 anos atrás |
|
florian
|
44a7ef600f
* improve register handling for not and unary minus operator by using a new register for the destination
|
há 12 anos atrás |
|
nickysn
|
a5c5b05362
* initial support for system procs that use calling conventions that push left to right on i8086 or i386
|
há 12 anos atrás |
|
nickysn
|
4e640511fb
* tcgunaryminusnode.second_64bit fixed for < 32-bit CPUs
|
há 12 anos atrás |
|
nickysn
|
48bba896a5
* cpu16bitalu fixes in tcgshlshrnode.second_integer
|
há 12 anos atrás |
