mirror of
https://github.com/holub/mame
synced 2025-07-01 16:19:38 +03:00
ncr53c90: better guess at sequence logic for macqd700
This commit is contained in:
parent
11d04180ca
commit
0158cdeab6
@ -446,6 +446,10 @@ void ncr53c90_device::step(bool timeout)
|
||||
case DISC_SEL_ARBITRATION_INIT:
|
||||
// wait until a command is in the fifo
|
||||
if (!fifo_pos) {
|
||||
// this sequence isn't documented for initiator selection, but
|
||||
// it makes macqd700 happy and may be consistent with target
|
||||
// selection sequences
|
||||
seq = 1;
|
||||
// dma starts after bus arbitration/selection is complete
|
||||
check_drq();
|
||||
break;
|
||||
@ -1069,7 +1073,7 @@ void ncr53c90_device::timeout_w(uint8_t data)
|
||||
uint8_t ncr53c90_device::seq_step_r()
|
||||
{
|
||||
LOG("seq_step_r %d (%s)\n", seq, machine().describe_context());
|
||||
return seq + 1; // HACK for Mac Quadra 700; our sequence steps don't correspond exactly to what the real chip does
|
||||
return seq;
|
||||
}
|
||||
|
||||
void ncr53c90_device::sync_period_w(uint8_t data)
|
||||
|
Loading…
Reference in New Issue
Block a user