A bug in the ATA66 support for Intel 82801AA/82801AB in 2.3.39?

[Andre Hedrick]

I would not be surprized.........

Since I have been asking for the info for more than two months from Intel,
and just found out that they white page booked it.........

http://developer.intel.com/design/chipsets/datashts/29065503.pdf

Given the orginal limited spec had it the other way.......
It is a fix that is simple ....... 

ide-pci.c

#define DEVID_PIIX4E    ((ide_pci_devid_t){PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801AA_1})
#define DEVID_PIIX4U    ((ide_pci_devid_t){PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801AB_1})

change from above to below.

#define DEVID_PIIX4E    ((ide_pci_devid_t){PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801AB_1})
#define DEVID_PIIX4U    ((ide_pci_devid_t){PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801AA_1})

piix.c

        int ultra               = ((dev->device == PCI_DEVICE_ID_INTEL_82371AB) ||
                                   (dev->device == PCI_DEVICE_ID_INTEL_82801AA_1)) ? 1 : 0;
        int ultra66             = (dev->device == PCI_DEVICE_ID_INTEL_82801AB_1)

change from above to below.

        int ultra               = ((dev->device == PCI_DEVICE_ID_INTEL_82371AB) ||
                                   (dev->device == PCI_DEVICE_ID_INTEL_82801AB_1)) ? 1 : 0;
        int ultra66             = (dev->device == PCI_DEVICE_ID_INTEL_82801AA_1)


However.......this blocks ata66 for PIIX controllers regardless....

unsigned int __init ata66_piix (ide_hwif_t *hwif)
{
        if (0)
                return 1;
        return 0;
}


hwif->udma_four = (d->ata66_check) ? d->ata66_check(hwif) : 0;

What changes were made to the code that were not original?
It is blocked regardless because of the lines........above.


On Wed, 19 Jan 2000, H . J . Lu wrote:

> Hi,
> 
> Something is very strange in the ATA66 support for Intel
> 82801AA/82801AB in 2.3.39. According to the Intel doc,
> 82801AA supports ATA66 and 82801AB doesn't support ATA66.
> However, 2.3.39 enables ATA66 for 82801AB. Is that intentional?
> 
> -- 
> H.J. Lu (hjl en gnu.org)
> 

Andre Hedrick
The Linux IDE guy



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo en vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/