 Károly Balogh
|
a57d0d51b4
softfpu: make it possible to optionally inline some of the simple functions. also made it possible to replace the MUL32TO64 function with compiler generated code. If the new defines are enabled, they results in 15-30% speedup on m68k (68020+) softfpu code
|
8 gadi atpakaļ |
|
Károly Balogh
|
f7488f2bd8
m68k: don't try to inline NaN and InF values as consts into FPU operands
|
8 gadi atpakaļ |
|
Károly Balogh
|
b649db5004
rtl-console: support two extra builtin font sizes on Amiga-likes. makes it possible to use the IDE on a 640x256 PAL screen, or have VGA 80x50-alike mode on RTG/high-res screens. switching is possible via an ENV var for now. Also support turning off the SMARTREFRESH attribute of the IDE window
|
8 gadi atpakaļ |
|
Károly Balogh
|
97f876c9d4
rtl-console: added a 8x8 and a 8x14 font data for the video unit used by Amiga-likes. added proper copyright header for the file
|
8 gadi atpakaļ |
|
Károly Balogh
|
b3157aa5ea
m68k: generate MUL helpers for CPUs without 32bit MUL already in pass 1
|
8 gadi atpakaļ |
|
Károly Balogh
|
b755524c38
made it possible to override the decision from the cgs, if a certain muln needs a helper
|
8 gadi atpakaļ |
|
Károly Balogh
|
6ca8b1374b
m68k: quickfix the 68000 builds of the compiler after r36344
|
8 gadi atpakaļ |
 nickysn
|
e8a09fba0d
* use xchg, instead of mov, because it's 1 byte shorter
|
8 gadi atpakaļ |
|
nickysn
|
34113b930b
* removed the checkoverflow: boolean parameter of the software mul helpers;
|
8 gadi atpakaļ |
 Marc Weustink
|
2b2200fcc6
RegisterCustomVariantType: initialize the customvariant array to the correct length before assigning new value
|
8 gadi atpakaļ |
|
nickysn
|
ba665e103c
+ added sqr(shortint) and sqr(smallint) for CPUs with 8/16-bit alu
|
8 gadi atpakaļ |
 michael
|
7c9e5591b6
* Fix bug ID #31583 (patch from Zoran Vucenovic)
|
8 gadi atpakaļ |
 pierre
|
a16326fa95
Avoid MSDOS Memory Control Block (MCB) chain corruption in RegisterTinyHeapBlock
|
8 gadi atpakaļ |
|
Károly Balogh
|
87e8010f05
m68k: support 32x32 to 64bit MUL generation when targeting CPUs which support this instruction
|
8 gadi atpakaļ |
 joost
|
f5d17b2884
* Unified determination of source-package
|
8 gadi atpakaļ |
|
joost
|
569bd1e812
* Only pass prefix and baseinstalldir to fpmake during (de-)installation
|
8 gadi atpakaļ |
|
Károly Balogh
|
63b3304374
m68k: instead of calling a helper, generate inlined code for 64 bit multiplications, when applicable
|
8 gadi atpakaļ |
|
Károly Balogh
|
829b2fc3b3
m68k: enable 68020 codepaths to all CPUs supporting '020 features
|
8 gadi atpakaļ |
|
Károly Balogh
|
e9ff684ff0
m68k: handle operand type correctly for 3 operand mul/div
|
8 gadi atpakaļ |
|
Károly Balogh
|
aa87e911d7
m68k: fixed some copypaste error in a disabled debug string. no functional changes.
|
8 gadi atpakaļ |
|
nickysn
|
78a7445751
+ added Odd(byte) and Odd(shortint) for cpus with 8-bit alu and Odd(word)/
|
8 gadi atpakaļ |
|
nickysn
|
fa645dcaf7
+ added abs(smallint) and abs(shortint) for cpus with 8-bit or 16-bit alu
|
8 gadi atpakaļ |
|
nickysn
|
9946393018
+ added comment about the fptan instruction restrictions on the 8087 and 80287
|
8 gadi atpakaļ |
 florian
|
db49b2f62b
* coalesce spilled registers if possible to avoid mem->mem moves
|
8 gadi atpakaļ |
|
florian
|
4b17ad80d0
* reduce initial size of movelist, 16 is enough for most cases
|
8 gadi atpakaļ |
|
florian
|
5e07957594
+ TLinkedList.RemoveAll
|
8 gadi atpakaļ |
|
florian
|
c83e6991d5
* properly check number of operands
|
8 gadi atpakaļ |
|
Károly Balogh
|
bfc5f67f92
fcl-passrc: attempt to fix the install rule and some of the dependencies
|
8 gadi atpakaļ |
 Mattias Gaertner
|
ae5492af44
fcl-passrc: fixed checking direct class-ancestor cycle
|
8 gadi atpakaļ |
|
Mattias Gaertner
|
e552db6ef6
pastojs: fixed compilation
|
8 gadi atpakaļ |
