Problem with some aec62xx

Problem with some aec62xx

[Benjamin Herrenschmidt]

Hi !

One of you (Bart, Andre ?) has specs to these controllers ? Some
users are reporting me various failures with these on ppc until
they got a bit of code back from earlier 2.4 drivers into the
new (2.4.21+ or 2.6.0) one:

			/* Clear reset and test bits.  */
			pci_read_config_byte(hwif->pci_dev, 0x49, &reg49h);
			pci_write_config_byte(hwif->pci_dev, 0x49, reg49h & ~0x30);
			/* Enable chip interrupt output.  */
			pci_read_config_byte(hwif->pci_dev, 0x4a, &reg4ah);
			pci_write_config_byte(hwif->pci_dev, 0x4a, reg4ah & ~0x01);

Theres also this additional bit, that doesn't seem to be required,
but was there in 2.4.20 (at least in my tree, it's possible that the
version of the driver I had was slightly different as I remember
taking patches on this driver from some users a long time ago).

#define CONFIG_AEC6280_BURST
#ifdef CONFIG_AEC6280_BURST
			/* Must be greater than 0x80 for burst mode.  */
			pci_write_config_byte(hwif->pci_dev, PCI_LATENCY_TIMER, 0x90);
			/* Enable burst mode.  */
			pci_read_config_byte(hwif->pci_dev, 0x4a, &reg4ah);
			pci_write_config_byte(hwif->pci_dev, 0x4a, reg4ah | 0x80);
#endif

Re: Problem with some aec62xx

[Benjamin Herrenschmidt]

On Sun, 2003-12-07 at 13:11, Benjamin Herrenschmidt wrote:
> Hi !
> 
> One of you (Bart, Andre ?) has specs to these controllers ? Some
> users are reporting me various failures with these on ppc until
> they got a bit of code back from earlier 2.4 drivers into the
> new (2.4.21+ or 2.6.0) one:
> 
> 			/* Clear reset and test bits.  */
> 			pci_read_config_byte(hwif->pci_dev, 0x49, &reg49h);
> 			pci_write_config_byte(hwif->pci_dev, 0x49, reg49h & ~0x30);
> 			/* Enable chip interrupt output.  */
> 			pci_read_config_byte(hwif->pci_dev, 0x4a, &reg4ah);
> 			pci_write_config_byte(hwif->pci_dev, 0x4a, reg4ah & ~0x01);

   .../...

And I forgot to mention that the above is for those IDs:

		case PCI_DEVICE_ID_ARTOP_ATP865:
		case PCI_DEVICE_ID_ARTOP_ATP865R:

Ben.