[alsa-devel] [PATCH] [patch 2/2] AC97: add full duplex support for atmel AT91 and AVR.
This patch add full duplex support on AT91 and AVR. It was a bug: we needed to check first if there are some chips opened so we could enable both reception and sending of the data.
Signed-off-by: Sedji Gaouaou sedji.gaouaou@atmel.com --- sound/atmel/ac97c.c | 16 ++++++++++++---- 1 files changed, 12 insertions(+), 4 deletions(-)
diff --git a/sound/atmel/ac97c.c b/sound/atmel/ac97c.c index d48a59a..6f5e22c 100644 --- a/sound/atmel/ac97c.c +++ b/sound/atmel/ac97c.c @@ -339,7 +339,11 @@ static int atmel_ac97c_playback_prepare(struct snd_pcm_substream *substream) ac97c_writel(chip, OCA, word);
/* configure sample format and size */ - word = AC97C_CMR_DMAEN | AC97C_CMR_SIZE_16; + word = ac97c_readl(chip, CAMR); + if (chip->opened <= 1) + word = AC97C_CMR_DMAEN | AC97C_CMR_SIZE_16; + else + word |= AC97C_CMR_DMAEN | AC97C_CMR_SIZE_16;
switch (runtime->format) { case SNDRV_PCM_FORMAT_S16_LE: @@ -426,7 +430,11 @@ static int atmel_ac97c_capture_prepare(struct snd_pcm_substream *substream) ac97c_writel(chip, ICA, word);
/* configure sample format and size */ - word = AC97C_CMR_DMAEN | AC97C_CMR_SIZE_16; + word = ac97c_readl(chip, CAMR); + if (chip->opened <= 1) + word = AC97C_CMR_DMAEN | AC97C_CMR_SIZE_16; + else + word |= AC97C_CMR_DMAEN | AC97C_CMR_SIZE_16;
switch (runtime->format) { case SNDRV_PCM_FORMAT_S16_LE: @@ -506,7 +514,7 @@ atmel_ac97c_playback_trigger(struct snd_pcm_substream *substream, int cmd) } else { ptcr = ATMEL_PDC_TXTEN; } - camr |= AC97C_CMR_CENA; + camr |= AC97C_CMR_CENA | AC97C_CSR_ENDTX; break; case SNDRV_PCM_TRIGGER_PAUSE_PUSH: /* fall through */ case SNDRV_PCM_TRIGGER_SUSPEND: /* fall through */ @@ -551,7 +559,7 @@ atmel_ac97c_capture_trigger(struct snd_pcm_substream *substream, int cmd) } else { ptcr = ATMEL_PDC_RXTEN; } - camr |= AC97C_CMR_CENA; + camr |= AC97C_CMR_CENA | AC97C_CSR_ENDRX; break; case SNDRV_PCM_TRIGGER_PAUSE_PUSH: /* fall through */ case SNDRV_PCM_TRIGGER_SUSPEND: /* fall through */
On Thu, 2010-02-25 at 18:59 +0100, Sedji Gaouaou wrote:
This patch add full duplex support on AT91 and AVR. It was a bug: we needed to check first if there are some chips opened so we could enable both reception and sending of the data.
Signed-off-by: Sedji Gaouaou sedji.gaouaou@atmel.com
Acked-by: Liam Girdwood lrg@slimlogic.co.uk
On Thu, Feb 25, 2010 at 06:59:40PM +0100, Sedji Gaouaou wrote:
This patch add full duplex support on AT91 and AVR. It was a bug: we needed to check first if there are some chips opened so we could enable both reception and sending of the data.
Signed-off-by: Sedji Gaouaou sedji.gaouaou@atmel.com
Both patches
Acked-by: Mark Brown broonie@opensource.wolfsonmicro.com
participants (3)
-
Liam Girdwood
-
Mark Brown
-
Sedji Gaouaou