I've compile-tested this patch but I didn't do any more than that. Are you
able to test whether the mainline atari_scsi module works any better with
a patched aranym?
BTW, the two kernel patches I sent to the mailing lists yesterday are
fixes for NCR5380 driver exception handing (EH) bugs. They should help
with an unpatched aranym, because the fake NCR 5380 will cause SCSI
commands to timeout and hence undergo EH. The eh_ebort_handler call will
likely return FAILED and this will escalate to a call to the
eh_bus_reset_handler. The command may be retried, but the mid-layer should
recover eventually. To observe this, you may need to enable scsi error
logging.
diff --git a/src/include/ncr5380.h b/src/include/ncr5380.h
index f8fc43b..55bb84d 100644
--- a/src/include/ncr5380.h
+++ b/src/include/ncr5380.h
@@ -33,6 +33,7 @@ class NCR5380 {
uae_u8 hd_count;
uae_u8 hd_initiator;
+ uae_u8 hd_mode;
public:
NCR5380(void);
diff --git a/src/ncr5380.cpp b/src/ncr5380.cpp
index 75d8f50..b3177f1 100644
--- a/src/ncr5380.cpp
+++ b/src/ncr5380.cpp
@@ -120,7 +120,7 @@ NCR5380::~NCR5380(void)
void NCR5380::reset(void)
{
- hd_count = hd_status = hd_initiator = 0;
+ hd_count = hd_status = hd_initiator = hd_mode = 0;
D(bug("ncr5380: reset"));
}
@@ -140,6 +140,7 @@ uae_u8 NCR5380::ReadData(uae_u16 control)
data = hd_initiator = ICR_ARBITRATION_PROGRESS;
break;
case MODE_REG:
+ data = hd_mode;
break;
case TARGET_COMMAND_REG:
break;
@@ -172,6 +173,7 @@ void NCR5380::WriteData(uae_u16 control, uae_u8 data)
hd_initiator = data;
break;
case MODE_REG:
+ hd_mode = data;
break;
case TARGET_COMMAND_REG:
break;
--
To unsubscribe from this list: send the line "unsubscribe linux-m68k" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
