On Tue, Mar 31, 2009 at 9:15 AM, Atsushi Nemoto <anemo@xxxxxxxxxxxxx> wrote:
> * define CHECK_DMA_MASK
> * remove use of wmb()
>
> Suggested-by: Grant Grundler <grundler@xxxxxxxxxx>
Thank you for the attribution!
But I think proper header would be:
Reported-by: Grant Grundler <grundler@xxxxxxxxxx>
But in this case, since i've looked at the code and am under the
illusion I understand it, I'm comfortable with:
Reviewed-by: Grant Grundler <grundler@xxxxxxxxxx>
thanks!
grant
> Signed-off-by: Atsushi Nemoto <anemo@xxxxxxxxxxxxx>
> ---
> This patch is against linux-next 20090331.
>
> drivers/ide/tx4939ide.c | 6 +++---
> 1 files changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/ide/tx4939ide.c b/drivers/ide/tx4939ide.c
> index cc269c0..48186ae 100644
> --- a/drivers/ide/tx4939ide.c
> +++ b/drivers/ide/tx4939ide.c
> @@ -327,15 +327,15 @@ static int tx4939ide_dma_end(ide_drive_t *drive)
> /* read and clear the INTR & ERROR bits */
> dma_stat = tx4939ide_clear_dma_status(base);
>
> - wmb();
> +#define CHECK_DMA_MASK (ATA_DMA_ACTIVE | ATA_DMA_ERR | ATA_DMA_INTR)
>
> /* verify good DMA status */
> - if ((dma_stat & (ATA_DMA_INTR | ATA_DMA_ERR | ATA_DMA_ACTIVE)) == 0 &&
> + if ((dma_stat & CHECK_DMA_MASK) == 0 &&
> (ctl & (TX4939IDE_INT_XFEREND | TX4939IDE_INT_HOST)) ==
> (TX4939IDE_INT_XFEREND | TX4939IDE_INT_HOST))
> /* INT_IDE lost... bug? */
> return 0;
> - return ((dma_stat & (ATA_DMA_INTR | ATA_DMA_ERR | ATA_DMA_ACTIVE)) !=
> + return ((dma_stat & CHECK_DMA_MASK) !=
> ATA_DMA_INTR) ? 0x10 | dma_stat : 0;
> }
>
> --
> 1.5.6.3
>
>
