[PATCH 2.6.21-rc7] sata_promise: SATAII-150/300 TX4 port numbering fix

[PATCH 2.6.21-rc7] sata_promise: SATAII-150/300 TX4 port numbering fix

[Mikael Pettersson]

There is a known problem with sata_promise on SATAII-150/300 TX4
controller cards: it enumerates drives in an order that differs
from the port numbers printed on the controller cards. However,
Promise's BIOS and Linux driver both get the order right.

I investigated Promise's Linux driver (v1.01.0.23), and found
that it explicitly changes the mapping from logical port number
to ATA engine MMIO address on the SATAII TX4 cards. It does this
on all SATAII TX4 cards, without inspecting revision etc. The
SATAII TX2plus cards continue to use the same mapping that was
used for the first-generation chips.

This patch updates sata_promise to use the new port number to
ATA engine mapping on SATAII TX4 cards, which fixes the drive
enumeration order problem on those cards. Tested on 300 TX4,
300 TX2plus, and SATAII-150 TX2plus chips.

Signed-off-by: Mikael Pettersson <mikpe@it.uu.se>
---
This patch should apply to 2.6.21-rc7 and libata#upstream.
It won't apply to libata#ALL because of the massive changes
for the new init model. I will do a new and cleaner patch for
#ALL once I can get it as a patch in -mm (I don't do git).

--- linux-2.6.21-rc7/drivers/ata/sata_promise.c.~1~	2007-04-23 00:17:35.000000000 +0200
+++ linux-2.6.21-rc7/drivers/ata/sata_promise.c	2007-04-23 00:18:06.000000000 +0200
@@ -989,7 +989,13 @@ static int pdc_ata_init_one (struct pci_
 	switch (board_idx) {
 	case board_40518:
 		hp->flags |= PDC_FLAG_GEN_II;
-		/* Fall through */
+		printk(KERN_INFO DRV_NAME ": applying SATAII-150/300 TX4 port numbering workaround\n");
+		probe_ent->n_ports = 4;
+		pdc_ata_setup_port(&probe_ent->port[2], base + 0x200, base + 0x400);
+		pdc_ata_setup_port(&probe_ent->port[1], base + 0x280, base + 0x500);
+		pdc_ata_setup_port(&probe_ent->port[3], base + 0x300, base + 0x600);
+		pdc_ata_setup_port(&probe_ent->port[0], base + 0x380, base + 0x700);
+		break;
 	case board_20319:
        		probe_ent->n_ports = 4;
 		pdc_ata_setup_port(&probe_ent->port[2], base + 0x300, base + 0x600);

Re: [PATCH 2.6.21-rc7] sata_promise: SATAII-150/300 TX4 port numbering fix

[Jeff Garzik]

Mikael Pettersson wrote:
> There is a known problem with sata_promise on SATAII-150/300 TX4
> controller cards: it enumerates drives in an order that differs
> from the port numbers printed on the controller cards. However,
> Promise's BIOS and Linux driver both get the order right.
> 
> I investigated Promise's Linux driver (v1.01.0.23), and found
> that it explicitly changes the mapping from logical port number
> to ATA engine MMIO address on the SATAII TX4 cards. It does this
> on all SATAII TX4 cards, without inspecting revision etc. The
> SATAII TX2plus cards continue to use the same mapping that was
> used for the first-generation chips.
> 
> This patch updates sata_promise to use the new port number to
> ATA engine mapping on SATAII TX4 cards, which fixes the drive
> enumeration order problem on those cards. Tested on 300 TX4,
> 300 TX2plus, and SATAII-150 TX2plus chips.
> 
> Signed-off-by: Mikael Pettersson <mikpe@it.uu.se>
> ---
> This patch should apply to 2.6.21-rc7 and libata#upstream.
> It won't apply to libata#ALL because of the massive changes
> for the new init model. I will do a new and cleaner patch for
> #ALL once I can get it as a patch in -mm (I don't do git).

ACK; dropped, awaiting rediff and resend against new init model

FWIW:  The new init model was applied to #upstream, and subsequently 
merged into #ALL (which always is a superset of #upstream).  So from 
your description, it sounds like your #upstream may not have been updated.

FWIW2:  When I note that libata-dev.git has been rebased, this means 
that you will need to delete the #upstream branch on your side, and 
download it again.

	Jeff