On Sun, Aug 24, 2008 at 6:04 AM, Lennert Buytenhek
<buytenh@xxxxxxxxxxxxxx> wrote:
> For some reason, sata_mv doesn't clear interrupt status during init
> when it's running on an SoC host adapter. If the bootloader has
> touched the SATA controller before starting Linux, Linux can end up
> enabling the SATA interrupt with events pending, which will cause the
> interrupt to be marked as spurious and then be disabled, which then
> breaks all further accesses to the controller.
>
> This patch makes the SoC path clear interrupt status on init like in
> the non-SoC case.
>
> Signed-off-by: Lennert Buytenhek <buytenh@xxxxxxxxxxx>
> ---
> drivers/ata/sata_mv.c | 21 ++++++++++-----------
> 1 files changed, 10 insertions(+), 11 deletions(-)
>
> diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c
> index ad169ff..e829a3a 100644
> --- a/drivers/ata/sata_mv.c
> +++ b/drivers/ata/sata_mv.c
> @@ -3131,19 +3131,18 @@ static int mv_init_host(struct ata_host *host, unsigned int board_idx)
> writelfl(0, hc_mmio + HC_IRQ_CAUSE_OFS);
> }
>
> - if (!IS_SOC(hpriv)) {
> - /* Clear any currently outstanding host interrupt conditions */
> - writelfl(0, mmio + hpriv->irq_cause_ofs);
> + /* Clear any currently outstanding host interrupt conditions */
> + writelfl(0, mmio + hpriv->irq_cause_ofs);
>
> - /* and unmask interrupt generation for host regs */
> - writelfl(hpriv->unmask_all_irqs, mmio + hpriv->irq_mask_ofs);
> + /* and unmask interrupt generation for host regs */
> + writelfl(hpriv->unmask_all_irqs, mmio + hpriv->irq_mask_ofs);
> +
> + /*
> + * enable only global host interrupts for now.
> + * The per-port interrupts get done later as ports are set up.
> + */
> + mv_set_main_irq_mask(host, 0, PCI_ERR);
>
> - /*
> - * enable only global host interrupts for now.
> - * The per-port interrupts get done later as ports are set up.
> - */
> - mv_set_main_irq_mask(host, 0, PCI_ERR);
> - }
> done:
> return rc;
> }
The patch looks fine.
Mark, I think that all the code that clears the cause registers should
be moved from mv_init_host to the reset_hc functions, which already
clears some of those register. for example, register 0x14 (host
controller cause register) get cleared twice, the first time at
mv5_reset_one_hc. also, the reset_hc functions better use the register
define instead of a hardcoded number when possible.
saeed
--
To unsubscribe from this list: send the line "unsubscribe linux-ide" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
