| 13 |
Description: |
Description: |
| 14 |
|
|
| 15 |
---------------------------------------------------------------------- |
---------------------------------------------------------------------- |
| 16 |
|
Name: Allen Leung |
| 17 |
|
Date: 2000/11/10 18:00:00 |
| 18 |
|
Tag: leunga-20001110-new-x86-fp |
| 19 |
|
|
| 20 |
|
A new x86 floating point code generator has been added. |
| 21 |
|
By default this is turned off. To turn this on, do: |
| 22 |
|
|
| 23 |
|
CM.autoload "$smlnj/compiler.cm"; |
| 24 |
|
Compiler.Control.MLRISC.getFlag "x86-fast-fp" := true; |
| 25 |
|
|
| 26 |
|
Changes: |
| 27 |
|
|
| 28 |
|
1. Changed FTAN to FPTAN so that the assembly output is correct. |
| 29 |
|
2. Changed the extension callback for FTANGENT to generate: |
| 30 |
|
|
| 31 |
|
fptan |
| 32 |
|
fstp %st(0) |
| 33 |
|
instead of |
| 34 |
|
fptan |
| 35 |
|
fstpl ftempmem |
| 36 |
|
|
| 37 |
|
3. Numerous assembly fixes for x86. |
| 38 |
|
|
| 39 |
|
5. Cleaned up the machine code output module x86/x86MC.sml and added |
| 40 |
|
support for a whole bunch of instructions and addressing modes: |
| 41 |
|
|
| 42 |
|
fadd/fsub/fsubr/fmul/fdiv/fdivr %st, %st(n) |
| 43 |
|
faddp/fsubp/fsubrp/fmulp/fdivp/fdivrp %st, %st(n) |
| 44 |
|
fadd/fsub/fsubr/fmul/fdiv/fdivr %st(n), %st |
| 45 |
|
fiadd/fisub/fisubr/fimul/fidiv/fidivr mem |
| 46 |
|
fxch %st(n) |
| 47 |
|
fld %st(n) |
| 48 |
|
fst %st(n) |
| 49 |
|
fst mem |
| 50 |
|
fstp %st(n) |
| 51 |
|
fucom %st(n) |
| 52 |
|
fucomp %st(n) |
| 53 |
|
|
| 54 |
|
All these are now generated when the fast fp mode is turned on. |
| 55 |
|
|
| 56 |
|
6. Removed the dedicated registers %st(0), ..., %st(7) from X86CpsRegs |
| 57 |
|
|
| 58 |
|
---------------------------------------------------------------------- |
| 59 |
Name: Matthias Blume |
Name: Matthias Blume |
| 60 |
Date: 2000/11/09 11:20:00 JST |
Date: 2000/11/09 11:20:00 JST |
| 61 |
Tag: blume-20001109-scc |
Tag: blume-20001109-scc |
Parent Directory
| 
Revision Log
| 
Patch
