2012-08-17 08:18:48

by Arnd Hannemann

Subject: [PATCH] pata_atiixp: override cable detection on MSI E350DM-E33

The mainboard MSI E350DM-E33 is advertised with 6 SATA ports.
As it turns out, two of them seem to be driven by on-board
SATA<->PATA converters. If a disk drive is connected to one
of them kernel uses UDMA/33 mode due to cable detection:

[ 34.550823] scsi4 : pata_atiixp
[ 34.555517] scsi5 : pata_atiixp
[ 34.555942] ata5: PATA max UDMA/100 cmd 0x1f0 ctl 0x3f6 bmdma 0xf100 irq 14
[ 34.555948] ata6: PATA max UDMA/100 cmd 0x170 ctl 0x376 bmdma 0xf108 irq 15
...
[ 35.040799] ata5.00: ATA-8: WDC WD20EADS-00R6B0, 01.00A01, max UDMA/133
[ 35.040806] ata5.00: 3907029168 sectors, multi 16: LBA48 NCQ (depth 0/32)
[ 35.040817] ata5.00: limited to UDMA/33 due to 40-wire cable
[ 35.049166] ata5.00: configured for UDMA/33
[ 35.049402] scsi 4:0:0:0: Direct-Access ATA WDC WD20EADS-00R 01.0 PQ: 0 ANSI: 5

This patch forces "short cable" mode on this board, as it seems clear that
the on-board SATA<->PATA "cable" is short.
With this patch the disk is configured for UDMA/100:

[ 5.976756] ata5.00: ATA-8: WDC WD20EADS-00R6B0, 01.00A01, max UDMA/133
[ 5.996434] ata5.00: 3907029168 sectors, multi 16: LBA48 NCQ (depth 0/32)
[ 6.024787] ata5.00: configured for UDMA/100

Testing revealed no transfer issues.

Signed-off-by: Arnd Hannemann <[email protected]>
---
drivers/ata/pata_atiixp.c | 16 ++++++++++++++++
1 file changed, 16 insertions(+)

diff --git a/drivers/ata/pata_atiixp.c b/drivers/ata/pata_atiixp.c
index 361c75c..24e5105 100644
--- a/drivers/ata/pata_atiixp.c
+++ b/drivers/ata/pata_atiixp.c
@@ -20,6 +20,7 @@
#include <linux/delay.h>
#include <scsi/scsi_host.h>
#include <linux/libata.h>
+#include <linux/dmi.h>

#define DRV_NAME "pata_atiixp"
#define DRV_VERSION "0.4.6"
@@ -33,11 +34,26 @@ enum {
ATIIXP_IDE_UDMA_MODE = 0x56
};

+static const struct dmi_system_id attixp_cable_override_dmi_table[] = {
+ {
+ /* Board has onboard PATA<->SATA converters */
+ .ident = "MSI E350DM-E33",
+ .matches = {
+ DMI_MATCH(DMI_BOARD_VENDOR, "MSI"),
+ DMI_MATCH(DMI_BOARD_NAME, "E350DM-E33(MS-7720)"),
+ },
+ },
+ { }
+};
+
static int atiixp_cable_detect(struct ata_port *ap)
{
struct pci_dev *pdev = to_pci_dev(ap->host->dev);
u8 udma;

+ if (dmi_check_system(attixp_cable_override_dmi_table))
+ return ATA_CBL_PATA40_SHORT;
+
/* Hack from drivers/ide/pci. Really we want to know how to do the
raw detection not play follow the bios mode guess */
pci_read_config_byte(pdev, ATIIXP_IDE_UDMA_MODE + ap->port_no, &udma);
--
1.7.9.5

2012-08-17 17:28:14

by Jeff Garzik

Subject: Re: [PATCH] pata_atiixp: override cable detection on MSI E350DM-E33

On 08/17/2012 04:11 AM, Arnd Hannemann wrote:
> The mainboard MSI E350DM-E33 is advertised with 6 SATA ports.
> As it turns out, two of them seem to be driven by on-board
> SATA<->PATA converters. If a disk drive is connected to one
> of them kernel uses UDMA/33 mode due to cable detection:
>
> [ 34.550823] scsi4 : pata_atiixp
> [ 34.555517] scsi5 : pata_atiixp
> [ 34.555942] ata5: PATA max UDMA/100 cmd 0x1f0 ctl 0x3f6 bmdma 0xf100 irq 14
> [ 34.555948] ata6: PATA max UDMA/100 cmd 0x170 ctl 0x376 bmdma 0xf108 irq 15
> ...
> [ 35.040799] ata5.00: ATA-8: WDC WD20EADS-00R6B0, 01.00A01, max UDMA/133
> [ 35.040806] ata5.00: 3907029168 sectors, multi 16: LBA48 NCQ (depth 0/32)
> [ 35.040817] ata5.00: limited to UDMA/33 due to 40-wire cable
> [ 35.049166] ata5.00: configured for UDMA/33
> [ 35.049402] scsi 4:0:0:0: Direct-Access ATA WDC WD20EADS-00R 01.0 PQ: 0 ANSI: 5
>
> This patch forces "short cable" mode on this board, as it seems clear that
> the on-board SATA<->PATA "cable" is short.
> With this patch the disk is configured for UDMA/100:
>
> [ 5.976756] ata5.00: ATA-8: WDC WD20EADS-00R6B0, 01.00A01, max UDMA/133
> [ 5.996434] ata5.00: 3907029168 sectors, multi 16: LBA48 NCQ (depth 0/32)
> [ 6.024787] ata5.00: configured for UDMA/100
>
> Testing revealed no transfer issues.
>
> Signed-off-by: Arnd Hannemann <[email protected]>

applied