I told you, we can get rid of that routine :)
On Thu, Jan 24, 2013 at 4:04 PM, Andy Shevchenko
<andriy.shevche...@linux.intel.com> wrote:
I don't see residue - len from interrupt handler. It isn't required?
> @@ -1062,6 +1090,7 @@ dwc_tx_status(struct dma_chan *chan,
> struct dma_tx_state *txstate)
> {
> struct dw_dma_chan *dwc = to_dw_dma_chan(chan);
> + unsigned long flags;
> enum dma_status ret;
>
> ret = dma_cookie_status(chan, cookie, txstate);
> @@ -1071,8 +1100,17 @@ dwc_tx_status(struct dma_chan *chan,
> ret = dma_cookie_status(chan, cookie, txstate);
> }
>
> - if (ret != DMA_SUCCESS)
> - dma_set_residue(txstate, dwc_first_active(dwc)->len);
> + spin_lock_irqsave(&dwc->lock, flags);
> +
> + if (ret != DMA_SUCCESS) {
> + u32 residue = dwc->residue;
If you agree with the explanation that i gave in last mail, you must drop lock
right here. and also take it in this if block.
> + if (test_bit(DW_DMA_IS_SOFT_LLP, &dwc->flags) && residue)
> + residue -= dwc_get_sent(dwc);
> + dma_set_residue(txstate, residue);
> + }
> +
> + spin_unlock_irqrestore(&dwc->lock, flags);
other than that:
Acked-by: Viresh Kumar <viresh.ku...@linaro.org>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
