On 27.05.2009 6:08 Uhr, Joseph Smith wrote:
>
> extern struct chip_operations southbridge_intel_i82801xx_ops;
> Index: src/southbridge/intel/i82801xx/i82801xx_ide.c
> ===================================================================
> --- src/southbridge/intel/i82801xx/i82801xx_ide.c (revision 4311)
> +++ src/southbridge/intel/i82801xx/i82801xx_ide.c (working copy)
> @@ -27,29 +27,36 @@
> #include <device/pci_ids.h>
> #include "i82801xx.h"
>
> +typedef struct southbridge_intel_i82801xx_config config_t;
> +
> static void ide_init(struct device *dev)
> {
> + /* Get the chip configuration */
> + config_t *config = dev->chip_info;
> +
> /* TODO: Needs to be tested for compatibility with ICH5(R). */
> /* Enable IDE devices so the Linux IDE driver will work. */
> uint16_t ideTimingConfig;
> - int enable_primary = 1;
> - int enable_secondary = 1;
>
> ideTimingConfig = pci_read_config16(dev, IDE_TIM_PRI);
> ideTimingConfig &= ~IDE_DECODE_ENABLE;
> - if (enable_primary) {
> + if (config->ide0_enable) {
>
You should check if config is not NULL before accessing it or you might
silently read a config_t at the start of memory, which tells you that
all interrupts are configured to zero and both IDEs are disabled. This
was a tricky find for me once.
It will be NULL if the device 1f.2 (or whatever your IDE is) is not
present in Config.lb.
Maybe you want to check
if (!config || config->ide0_enable) { ... } // enable is default if
config is NULL
or you print a message that the user should fix the Config.lb file.
Other than that..
Acked-by: Stefan Reinauer <[email protected]>
All the best,
Stefan
--
coresystems GmbH • Brahmsstr. 16 • D-79104 Freiburg i. Br.
Tel.: +49 761 7668825 • Fax: +49 761 7664613
Email: [email protected] • http://www.coresystems.de/
Registergericht: Amtsgericht Freiburg • HRB 7656
Geschäftsführer: Stefan Reinauer • Ust-IdNr.: DE245674866
--
coreboot mailing list: [email protected]
http://www.coreboot.org/mailman/listinfo/coreboot
