Re: [PATCH] libata-sff: Allow for wacky systems

Re: [PATCH] libata-sff: Allow for wacky systems

[Valdis.Kletnieks]

[-- Attachment #1: Type: text/plain, Size: 1436 bytes --]

On Mon, 16 Oct 2006 16:24:50 BST, Alan Cox said:
> There are some Linux supported platforms that simply cannot hit the low
> I/O addresses used by ATA legacy mode PCI mappings. These platforms have
> a window for PCI space that is fixed by the board logic and doesn't
> include the neccessary locations.
> 
> Provide a config option so that such platforms faced with a controller
> that they cannot support simply error it and punt
> 
> Signed-off-by: Alan Cox <alan@redhat.com>
> 
> diff -u --new-file --recursive --exclude-from /usr/src/exclude linux.vanilla-
2.6.19-rc1-mm1/drivers/ata/libata-sff.c linux-2.6.19-rc1-mm1/drivers/ata/libata
-sff.c
> --- linux.vanilla-2.6.19-rc1-mm1/drivers/ata/libata-sff.c	2006-10-13 15:0
9:23.000000000 +0100
> +++ linux-2.6.19-rc1-mm1/drivers/ata/libata-sff.c	2006-10-13 17:15:57.000
000000 +0100
> @@ -981,6 +981,15 @@
>  		mask = (1 << 2) | (1 << 0);
>  		if ((tmp8 & mask) != mask)
>  			legacy_mode = (1 << 3);
> +#if defined(CONFIG_NO_ATA_LEGACY)
> +		/* Some platforms with PCI limits cannot address compat
> +		   port space. In that case we punt if their firmware has
> +		   left a device in compatibility mode */
> +		if (legacy_mode) {
> +			printk(KERN_ERR "ata: Compatibility mode ATA is not supported on this platform, skipping.\n");

Would it make sense for the printk to include a hint as to which controller
is on crack, so on boxes with PCI_MULTITHREAD_PROBE it's easier to tell?

[-- Attachment #2: Type: application/pgp-signature, Size: 226 bytes --]

[PATCH] libata-sff: Allow for wacky systems

[Alan Cox]

There are some Linux supported platforms that simply cannot hit the low
I/O addresses used by ATA legacy mode PCI mappings. These platforms have
a window for PCI space that is fixed by the board logic and doesn't
include the neccessary locations.

Provide a config option so that such platforms faced with a controller
that they cannot support simply error it and punt

Signed-off-by: Alan Cox <alan@redhat.com>

diff -u --new-file --recursive --exclude-from /usr/src/exclude linux.vanilla-2.6.19-rc1-mm1/drivers/ata/libata-sff.c linux-2.6.19-rc1-mm1/drivers/ata/libata-sff.c
--- linux.vanilla-2.6.19-rc1-mm1/drivers/ata/libata-sff.c	2006-10-13 15:09:23.000000000 +0100
+++ linux-2.6.19-rc1-mm1/drivers/ata/libata-sff.c	2006-10-13 17:15:57.000000000 +0100
@@ -981,6 +981,15 @@
 		mask = (1 << 2) | (1 << 0);
 		if ((tmp8 & mask) != mask)
 			legacy_mode = (1 << 3);
+#if defined(CONFIG_NO_ATA_LEGACY)
+		/* Some platforms with PCI limits cannot address compat
+		   port space. In that case we punt if their firmware has
+		   left a device in compatibility mode */
+		if (legacy_mode) {
+			printk(KERN_ERR "ata: Compatibility mode ATA is not supported on this platform, skipping.\n");
+			return -EOPNOTSUPP;
+		}
+#endif
 	}
 
 	rc = pci_request_regions(pdev, DRV_NAME);

Re: [PATCH] libata-sff: Allow for wacky systems

[Alan Cox]

Ar Llu, 2006-10-16 am 11:17 -0400, ysgrifennodd Valdis.Kletnieks@vt.edu:
> Would it make sense for the printk to include a hint as to which controller
> is on crack, so on boxes with PCI_MULTITHREAD_PROBE it's easier to tell?

Probably but multithread probe is strictly for lunatics anyway ;)


Alan

Re: [PATCH] libata-sff: Allow for wacky systems

[Jeff Garzik]

Alan Cox wrote:
> There are some Linux supported platforms that simply cannot hit the low
> I/O addresses used by ATA legacy mode PCI mappings. These platforms have
> a window for PCI space that is fixed by the board logic and doesn't
> include the neccessary locations.
> 
> Provide a config option so that such platforms faced with a controller
> that they cannot support simply error it and punt
> 
> Signed-off-by: Alan Cox <alan@redhat.com>

applied

Re: [PATCH] libata-sff: Allow for wacky systems

[David Howells]

Alan Cox <alan@lxorguk.ukuu.org.uk> wrote:

> There are some Linux supported platforms that simply cannot hit the low
> I/O addresses used by ATA legacy mode PCI mappings. These platforms have
> a window for PCI space that is fixed by the board logic and doesn't
> include the neccessary locations.
> 
> Provide a config option so that such platforms faced with a controller
> that they cannot support simply error it and punt

Yep, that works:

warthog>grep ATA_LEGACY .config
CONFIG_NO_ATA_LEGACY=y

...
pata_hpt37x: BIOS has not set timing clocks.                                    
hpt37x: HPT371: Bus clock 33MHz.                                                
ata1: PATA max UDMA/133 cmd 0xE4000820 ctl 0xE400082A bmdma 0xE4000C00 irq 20   
ata2: PATA max UDMA/133 cmd 0xE4000830 ctl 0xE400082E bmdma 0xE4000C08 irq 20   
scsi0 : pata_hpt37x                                                             
ATA: abnormal status 0x7D on port 0xE4000827                                    
scsi1 : pata_hpt37x                                                             
ATA: abnormal status 0x8 on port 0xE4000837                                     
pata_it821x: controller in smart mode.                                          
ata3: PATA max MWDMA2 cmd 0xE4000838 ctl 0xE4000842 bmdma 0xE4000850 irq 22     
ata4: PATA max MWDMA2 cmd 0xE4000848 ctl 0xE4000846 bmdma 0xE4000858 irq 22     
scsi2 : pata_it821x                                                             
scsi3 : pata_it821x                                                             
...


David