summaryrefslogtreecommitdiffstats
path: root/target/linux/ixp4xx/patches/160-nas100d_artop_temp_fix.patch
blob: b7ac4d94c68b01568d4fa395069fbd6d55e7d764 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
From: Alan Cox <alan@redhat.com>
To: Alessandro Zummo <alessandro.zummo@towertech.it>
Cc: Jeff Garzik <jgarzik@redhat.com>, Alan Cox <alan@redhat.com>
Subject: Re: drivers/ata/pata_artop.c
Date: Sun, 15 Oct 2006 14:25:16 -0400
User-Agent: Mutt/1.4.1i

On Sun, Oct 15, 2006 at 07:18:31PM +0200, Alessandro Zummo wrote:
>  In the discovery phase there's a lot of time spent in the detection
>  of the second port.  

The error recovery is a bit determined right now - Tejun's been doing some
work on SRST behaviour and also for the worst cases polled detect so it
should come out ok

>  What's the correct way to inform the driver
>  to avoid checking the second port?  

Set the number of ports to 1 in your own tree for now. The real fix is
not to go poking at pata ports if the ret is 0xFF

---
 drivers/ata/pata_artop.c |    6 ++++++
 1 file changed, 6 insertions(+)

Index: linux-2.6.19/drivers/ata/pata_artop.c
===================================================================
--- linux-2.6.19.orig/drivers/ata/pata_artop.c
+++ linux-2.6.19/drivers/ata/pata_artop.c
@@ -26,6 +26,7 @@
 #include <scsi/scsi_host.h>
 #include <linux/libata.h>
 #include <linux/ata.h>
+#include <asm/mach-types.h>
 
 #define DRV_NAME	"pata_artop"
 #define DRV_VERSION	"0.4.2"
@@ -469,6 +470,11 @@ static int artop_init_one (struct pci_de
 		pci_read_config_byte(pdev, 0x4a, &reg);
 		pci_write_config_byte(pdev, 0x4a, (reg & ~0x01) | 0x80);
 
+		/* NAS100D workaround */
+#ifdef CONFIG_MACH_NAS100D
+		if (machine_is_nas100d())
+			ports = 1;
+#endif
 	}
 
 	BUG_ON(info == NULL);