- patch from Zwane on December 9. He writes:
> It's safe to deliver ExtINT as Fixed in our setup and just leave a comment > there instead of the panic.
This commit is contained in:
parent
9a5ca5d129
commit
9a204569b7
@ -1,5 +1,5 @@
|
||||
/////////////////////////////////////////////////////////////////////////
|
||||
// $Id: apic.cc,v 1.27 2002-11-19 05:51:52 bdenney Exp $
|
||||
// $Id: apic.cc,v 1.28 2002-12-14 08:48:20 bdenney Exp $
|
||||
/////////////////////////////////////////////////////////////////////////
|
||||
//
|
||||
#define NEED_CPU_REG_SHORTCUTS 1
|
||||
@ -226,10 +226,13 @@ bx_generic_apic_c::deliver (Bit8u dest, Bit8u dest_mode, Bit8u delivery_mode, Bi
|
||||
if (deliver_bitmask & (1<<bit))
|
||||
apic_index[bit]->startup_msg (vector);
|
||||
return true;
|
||||
case 7: // ExtINT we can deliver this as fixed. In theory this requires
|
||||
// extra work from a connected PIC such as an 8259A but we can
|
||||
// skip all of that since all our pins are wired to the IOAPIC
|
||||
break;
|
||||
case 2: // SMI
|
||||
case 3: // reserved
|
||||
case 4: // NMI
|
||||
case 7: // ExtINT (I/O apic only)
|
||||
default:
|
||||
BX_PANIC(("APIC delivery mode %d not implemented", delivery_mode));
|
||||
}
|
||||
|
Loading…
Reference in New Issue
Block a user