paweld/fpc
1
0
Fork 0
mirror of https://gitlab.com/freepascal.org/fpc/source.git synced 2025-04-13 06:29:32 +02:00
Code Issues Packages Projects Releases Wiki Activity

m68k: fpu capabilities support

Browse Source 
git-svn-id: trunk@42739 -
This commit is contained in:
Károly Balogh 2019-08-19 15:07:22 +00:00
parent 7f6122884f
commit 4936358bee
2 changed files with 30 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
compiler/fpcdefs.inc
View File

@ -209,6 +209,7 @@
{$define cpufpemu}
{$define cpurefshaveindexreg}
{$define cpucapabilities}
{$define fpucapabilities}
{$define cpuneedsmulhelper}
{$define cpuneedsdivhelper}
{$define cpu_uses_separate_address_registers}

29
compiler/m68k/cpuinfo.pas
View File

@ -53,6 +53,8 @@ Type
fpu_soft,
fpu_libgcc,
fpu_68881,
fpu_68040,
fpu_68060,
fpu_coldfire
);
@ -124,6 +126,8 @@ Const
'SOFT',
'LIBGCC',
'68881',
'68040',
'68060',
'COLDFIRE'
);
@ -163,6 +167,15 @@ type
CPUM68K_HAS_BASEDISP { CPU supports addressing with 32bit base displacements }
);
tfpuflags =
(FPUM68K_HAS_HARDWARE, { FPU is actually a hardware implementation, not a software library }
FPUM68K_HAS_EXTENDED, { FPU has 80 bit extended support }
FPUM68K_HAS_TRIGONOMETRY, { FPU supports trigonometric instructions (FSIN/FCOS, etc) }
FPUM68K_HAS_RESULTPRECISION, { FPU supports encoding the result precision into instructions }
FPUM68K_HAS_FLOATIMMEDIATE, { FPU supports floating pont immediate values }
FPUM68K_HAS_FINTRZ { FPU supports the FINT/FINTRZ instruction }
);
const
cpu_capabilities : array[tcputype] of set of tcpuflags =
( { cpu_none } [],
@ -177,12 +190,28 @@ const
{ cpu_cfv4e } [CPUM68K_HAS_TAS,CPUM68K_HAS_BRAL,CPUM68K_HAS_MVSMVZ,CPUM68K_HAS_UNALIGNED,CPUM68K_HAS_32BITMUL,CPUM68K_HAS_16BITDIV,CPUM68K_HAS_32BITDIV,CPUM68K_HAS_REMSREMU]
);
{ on m68k, Motorola provided a software-library, which provides full '881/2 instruction set
compatibility on 040/060 FPU types via emulation for user code compatibility. this is slow
though, so this capabilities list contains the capabilities supported in the hardware itself }
fpu_capabilities : array[tfputype] of set of tfpuflags =
( { fpu_none } [],
{ fpu_soft } [],
{ fpu_libgcc } [],
{ fpu_68881 } [FPUM68K_HAS_HARDWARE,FPUM68K_HAS_EXTENDED,FPUM68K_HAS_FLOATIMMEDIATE,FPUM68K_HAS_TRIGONOMETRY,FPUM68K_HAS_FINTRZ],
{ fpu_68040 } [FPUM68K_HAS_HARDWARE,FPUM68K_HAS_EXTENDED,FPUM68K_HAS_RESULTPRECISION,FPUM68K_HAS_FLOATIMMEDIATE],
{ fpu_68060 } [FPUM68K_HAS_HARDWARE,FPUM68K_HAS_EXTENDED,FPUM68K_HAS_RESULTPRECISION,FPUM68K_HAS_FLOATIMMEDIATE,FPUM68K_HAS_FINTRZ],
{ fpu_coldfire } [FPUM68K_HAS_HARDWARE,FPUM68K_HAS_RESULTPRECISION,FPUM68K_HAS_FINTRZ]
);
{ all CPUs commonly called "coldfire" }
cpu_coldfire = [cpu_isa_a,cpu_isa_a_p,cpu_isa_b,cpu_isa_c,cpu_cfv4e];
{ all CPUs commonly called "68020+" }
cpu_mc68020p = [cpu_mc68020,cpu_mc68040,cpu_mc68060];
{ all FPUs commonly called "68881/2" }
fpu_mc68881 = [fpu_68881,fpu_68040,fpu_68060];
Implementation
end.