On Wed, 04 Feb 2009 13:00:42 +0100
Takashi Iwai <tiwai@xxxxxxx> wrote:
> At Wed, 4 Feb 2009 12:48:33 +0100,
> Hans-Christian Egtvedt wrote:
> >
> > +static int at32_abdac_hw_params(struct snd_pcm_substream
> > *substream,
> > + struct snd_pcm_hw_params *hw_params)
> > +{
> > + struct at32_abdac *dac = snd_pcm_substream_chip(substream);
> > + int retval;
> > +
> > + retval = snd_pcm_lib_malloc_pages(substream,
> > + params_buffer_bytes(hw_params));
> > + if (retval)
> > + return retval;
>
> snd_pcm_lib_malloc_pages() returns 1 if the buffer is changed
> (e.g. reallocation). So, this check should be if (retval < 0).
>
Ah, okay, I'll fix this.
> > +static int at32_abdac_trigger(struct snd_pcm_substream *substream,
> > int cmd) +{
> > + struct at32_abdac *dac = snd_pcm_substream_chip(substream);
> > + int retval = -EINVAL;
> > +
> > + switch (cmd) {
> > + case SNDRV_PCM_TRIGGER_PAUSE_RELEASE: /* fall through */
> > + case SNDRV_PCM_TRIGGER_RESUME: /* fall through */
> > + case SNDRV_PCM_TRIGGER_START:
> > + clk_enable(dac->sample_clk);
> > + retval = dw_dma_cyclic_start(dac->dma.chan);
> > + if (retval)
> > + goto out;
> > + dac_writel(dac, CTRL, DAC_BIT(EN));
> > + break;
> > + case SNDRV_PCM_TRIGGER_PAUSE_PUSH: /* fall through */
> > + case SNDRV_PCM_TRIGGER_SUSPEND: /* fall through */
> > + case SNDRV_PCM_TRIGGER_STOP:
> > + set_bit(DMA_STOP, &dac->flags);
> > + dw_dma_cyclic_stop(dac->dma.chan);
> > + dac_writel(dac, DATA, 0);
> > + dac_writel(dac, CTRL, 0);
> > + clk_disable(dac->sample_clk);
> > + break;
>
> No retval is set for these cases?
>
Nope, they are all void return functions and will always complete
successfully.
> > +static int __devinit at32_abdac_probe(struct platform_device *pdev)
> (snip)
> > + retval = -ENOMEM;
> > + card = snd_card_new(SNDRV_DEFAULT_IDX1, SNDRV_DEFAULT_STR1,
> > + THIS_MODULE, sizeof(struct at32_abdac));
>
> The recent version has a newer API, snd_card_create(). This function
> gives the error code. Please use the new function instead.
>
Okay, no problem, will redo with the new API.
--
Best regards,
Hans-Christian Egtvedt
_______________________________________________
Alsa-devel mailing list
Alsa-devel@xxxxxxxxxxxxxxxx
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
