[alsa-devel] Audio capture not working with AIC23/McBSP1 on OMAP3517
Hi,
I am trying to capture audio through LineIn on my OMAP3517 EVM, using AIC23 codec and McBSP1, and getting the following error:
arecord: pcm_read:1529: read error: Input/output error
Playback is working fine using the same combination.
I have tried using another instance of AIC23 codec which uses McBSP2 and both playback & capture works fine there.
Has anyone faced this issue before? Any suggestions?
Regards, Anuj Aggarwal
On Wed, Aug 26, 2009 at 06:30:52PM +0530, Aggarwal, Anuj wrote:
I am trying to capture audio through LineIn on my OMAP3517 EVM, using AIC23 codec and McBSP1, and getting the following error:
arecord: pcm_read:1529: read error: Input/output error
That suggests that DMA isn't happening - might be worth checking to see that clocks are being generated properly.
Playback is working fine using the same combination.
I have tried using another instance of AIC23 codec which uses McBSP2 and both playback & capture works fine there.
Has anyone faced this issue before? Any suggestions?
Which kernel are you using? There are rather a lot of McBSP changes queued up for 2.6.32. If this ever worked in the past a git bisect might be interesting.
-----Original Message----- From: Mark Brown [mailto:broonie@opensource.wolfsonmicro.com] Sent: Wednesday, August 26, 2009 6:34 PM To: Aggarwal, Anuj Cc: linux-omap@vger.kernel.org; alsa-devel@alsa-project.org Subject: Re: [alsa-devel] Audio capture not working with AIC23/McBSP1 on OMAP3517
On Wed, Aug 26, 2009 at 06:30:52PM +0530, Aggarwal, Anuj wrote:
I am trying to capture audio through LineIn on my OMAP3517 EVM, using AIC23 codec and McBSP1, and getting the following error:
arecord: pcm_read:1529: read error: Input/output error
That suggests that DMA isn't happening - might be worth checking to see that clocks are being generated properly.
I am able to see the bit clock and frame sync coming properly.
Playback is working fine using the same combination.
I have tried using another instance of AIC23 codec which uses McBSP2 and both playback & capture works fine there.
Has anyone faced this issue before? Any suggestions?
Which kernel are you using? There are rather a lot of McBSP changes queued up for 2.6.32. If this ever worked in the past a git bisect might be interesting.
I am on 2.6.31-rc5. I have to check whether the new patches fix anything related to this.
On Wed, 26 Aug 2009 14:03:56 +0100 Mark Brown broonie@opensource.wolfsonmicro.com wrote:
On Wed, Aug 26, 2009 at 06:30:52PM +0530, Aggarwal, Anuj wrote:
I am trying to capture audio through LineIn on my OMAP3517 EVM, using AIC23 codec and McBSP1, and getting the following error:
arecord: pcm_read:1529: read error: Input/output error
That suggests that DMA isn't happening - might be worth checking to see that clocks are being generated properly.
Playback is working fine using the same combination.
What is exactly your configuration: Is the AIC23 master or slave, how is the McBSP1 is connected (it has additional FSR and CLKR pins compared to other ports) and is the pin muxing correct?
I think I have tested the McBSP1 on Beagle but definitely only OMAP as a master since my codec was oscilloscope.
Which kernel are you using? There are rather a lot of McBSP changes queued up for 2.6.32. If this ever worked in the past a git bisect might be interesting.
Those recent patches should not any effect into this since they are mostly playing with the McBSP and DMA interfacing. What comes to my mind if muxing is correct and if codec is master, this might be related to those McBSP1 FSR and CLKR setup.
On Wed, 26 Aug 2009 16:42:39 +0300 Jarkko Nikula jhnikula@gmail.com wrote:
Those recent patches should not any effect into this since they are mostly playing with the McBSP and DMA interfacing. What comes to my mind if muxing is correct and if codec is master, this might be related to those McBSP1 FSR and CLKR setup.
Anuj: I think the issue is here that by default McBSP1 receiver is using the CLKR and FSR pins and if those are not connected, then the capture doesn't work. Can you try a patch below and add following lines into your machine driver?
snd_soc_dai_set_sysclk(cpu_dai, OMAP_MCBSP_CLKR_SRC_CLKX, 0, SND_SOC_CLOCK_IN); snd_soc_dai_set_sysclk(cpu_dai, OMAP_MCBSP_FSR_SRC_FSX, 0, SND_SOC_CLOCK_IN);
Patch is compile tested only, generated against mainline 2.6.31-rc7 but should apply to ALSA tree as well.
Thanks Jarkko, the patch works for me.
I will be submitting my ASoC omap*.c file soon after making all the necessary changes.
Regards, Anuj Aggarwal
-----Original Message----- From: Jarkko Nikula [mailto:jhnikula@gmail.com] Sent: Wednesday, August 26, 2009 8:32 PM To: Aggarwal, Anuj Cc: Mark Brown; alsa-devel@alsa-project.org; linux-omap@vger.kernel.org Subject: Re: [alsa-devel] Audio capture not working with AIC23/McBSP1 on OMAP3517
On Wed, 26 Aug 2009 16:42:39 +0300 Jarkko Nikula jhnikula@gmail.com wrote:
Those recent patches should not any effect into this since they are mostly playing with the McBSP and DMA interfacing. What comes to my mind if muxing is correct and if codec is master, this might be related to those McBSP1 FSR and CLKR setup.
Anuj: I think the issue is here that by default McBSP1 receiver is using the CLKR and FSR pins and if those are not connected, then the capture doesn't work. Can you try a patch below and add following lines into your machine driver?
snd_soc_dai_set_sysclk(cpu_dai, OMAP_MCBSP_CLKR_SRC_CLKX, 0, SND_SOC_CLOCK_IN); snd_soc_dai_set_sysclk(cpu_dai, OMAP_MCBSP_FSR_SRC_FSX, 0, SND_SOC_CLOCK_IN);
Patch is compile tested only, generated against mainline 2.6.31-rc7 but should apply to ALSA tree as well.
-- Jarkko
==================== CUT HERE ==================== From: Jarkko Nikula jhnikula@gmail.com Subject: [PATCH] ASoC: OMAP: Add functionality to set CLKR and FSR sources in McBSP DAI
The McBSP1 port in OMAP3 processors (I believe OMAP2 too but I don't have specifications to check it) have additional CLKR and FSR pins for McBSP1 receiver. Reset default is that receiver is using bit clock and frame sync signal from those pins but it is possible to configure to use also CLKX and FSX pins as well. In fact, other McBSP ports are doing that internally that transmitter and receiver share the CLKX and FSX.
Add functionaly that machine drivers can set the CLKR and FSR sources by using the snd_soc_dai_set_sysclk.
Thanks to "Aggarwal, Anuj" anuj.aggarwal@ti.com for reporting the issue.
Signed-off-by: Jarkko Nikula jhnikula@gmail.com
sound/soc/omap/omap-mcbsp.c | 41 +++++++++++++++++++++++++++++++++++++++++ sound/soc/omap/omap-mcbsp.h | 4 ++++ 2 files changed, 45 insertions(+), 0 deletions(-)
diff --git a/sound/soc/omap/omap-mcbsp.c b/sound/soc/omap/omap-mcbsp.c index a5d46a7..2ea2136 100644 --- a/sound/soc/omap/omap-mcbsp.c +++ b/sound/soc/omap/omap-mcbsp.c @@ -462,6 +462,40 @@ static int omap_mcbsp_dai_set_clks_src(struct omap_mcbsp_data *mcbsp_data, return 0; }
+static int omap_mcbsp_dai_set_rcvr_src(struct omap_mcbsp_data *mcbsp_data,
int clk_id)+{
- int sel_bit, set = 0;
- u16 reg = OMAP2_CONTROL_DEVCONF0;
- if (cpu_class_is_omap1())
return -EINVAL; /* TODO: Can this be implemented for OMAP1? */- if (mcbsp_data->bus_id != 0)
return -EINVAL;- switch (clk_id) {
- case OMAP_MCBSP_CLKR_SRC_CLKX:
set = 1;- case OMAP_MCBSP_CLKR_SRC_CLKR:
sel_bit = 3;break;- case OMAP_MCBSP_FSR_SRC_FSX:
set = 1;- case OMAP_MCBSP_FSR_SRC_FSR:
sel_bit = 4;break;- default:
return -EINVAL;- }
- if (set)
omap_ctrl_writel(omap_ctrl_readl(reg) | (1 << sel_bit), reg);- else
omap_ctrl_writel(omap_ctrl_readl(reg) & ~(1 << sel_bit), reg);- return 0;
+}
static int omap_mcbsp_dai_set_dai_sysclk(struct snd_soc_dai *cpu_dai, int clk_id, unsigned int freq, int dir) @@ -484,6 +518,13 @@ static int omap_mcbsp_dai_set_dai_sysclk(struct snd_soc_dai *cpu_dai, case OMAP_MCBSP_SYSCLK_CLKR_EXT: regs->pcr0 |= SCLKME; break;
- case OMAP_MCBSP_CLKR_SRC_CLKR:
- case OMAP_MCBSP_CLKR_SRC_CLKX:
- case OMAP_MCBSP_FSR_SRC_FSR:
- case OMAP_MCBSP_FSR_SRC_FSX:
err = omap_mcbsp_dai_set_rcvr_src(mcbsp_data, clk_id);break;diff --git a/sound/soc/omap/omap-mcbsp.h b/sound/soc/omap/omap-mcbsp.h index c8147aa..647d2f9 100644 --- a/sound/soc/omap/omap-mcbsp.h +++ b/sound/soc/omap/omap-mcbsp.h @@ -32,6 +32,10 @@ enum omap_mcbsp_clksrg_clk { OMAP_MCBSP_SYSCLK_CLK, /* Internal ICLK */ OMAP_MCBSP_SYSCLK_CLKX_EXT, /* External CLKX pin */ OMAP_MCBSP_SYSCLK_CLKR_EXT, /* External CLKR pin */
- OMAP_MCBSP_CLKR_SRC_CLKR, /* CLKR from CLKR pin */
- OMAP_MCBSP_CLKR_SRC_CLKX, /* CLKR from CLKX pin */
- OMAP_MCBSP_FSR_SRC_FSR, /* FSR from FSR pin */
- OMAP_MCBSP_FSR_SRC_FSX, /* FSR from FSX pin */
};
/* McBSP dividers */
1.6.3.3
On Thu, 27 Aug 2009 15:07:32 +0530 "Aggarwal, Anuj" anuj.aggarwal@ti.com wrote:
Thanks Jarkko, the patch works for me.
This is nice to hear.
Can you, Peter or who has access to 2420 and 2430 TRMs to verify is the DEVCONF0 bits 3 and 4 selecting there also the McBSP1 CLKR and FSR sources like my patch is now assuming for all OMAP2-3.
On Thursday 27 August 2009 14:12:53 ext Jarkko Nikula wrote:
On Thu, 27 Aug 2009 15:07:32 +0530
"Aggarwal, Anuj" anuj.aggarwal@ti.com wrote:
Thanks Jarkko, the patch works for me.
This is nice to hear.
Can you, Peter or who has access to 2420 and 2430 TRMs to verify is the DEVCONF0 bits 3 and 4 selecting there also the McBSP1 CLKR and FSR sources like my patch is now assuming for all OMAP2-3.
On OMAP2: DEVCONF:3 MCBSP1_CLKR DEVCONF:4 MCBSP1_FSR
So it is correct for both OMAP2 and OMAP3.
On Fri, 28 Aug 2009 09:55:01 +0300 Peter Ujfalusi peter.ujfalusi@nokia.com wrote:
Can you, Peter or who has access to 2420 and 2430 TRMs to verify is the DEVCONF0 bits 3 and 4 selecting there also the McBSP1 CLKR and FSR sources like my patch is now assuming for all OMAP2-3.
On OMAP2: DEVCONF:3 MCBSP1_CLKR DEVCONF:4 MCBSP1_FSR
So it is correct for both OMAP2 and OMAP3.
Good :-)
Are you Mark fine with the patch [] where the snd_soc_dai_set_sysclk is used to specify source pin for CLKR and FSR?
On Fri, Aug 28, 2009 at 01:51:03PM +0300, Jarkko Nikula wrote:
Are you Mark fine with the patch [] where the snd_soc_dai_set_sysclk is used to specify source pin for CLKR and FSR?
Yes, that's OK. Could you re-send the patch, I don't think I've got a local copy any more?
The McBSP1 port in OMAP3 processors (I believe OMAP2 too but I don't have specifications to check it) have additional CLKR and FSR pins for McBSP1 receiver. Reset default is that receiver is using bit clock and frame sync signal from those pins but it is possible to configure to use also CLKX and FSX pins as well. In fact, other McBSP ports are doing that internally that transmitter and receiver share the CLKX and FSX.
Add functionaly that machine drivers can set the CLKR and FSR sources by using the snd_soc_dai_set_sysclk.
Thanks to "Aggarwal, Anuj" anuj.aggarwal@ti.com for reporting the issue.
Signed-off-by: Jarkko Nikula jhnikula@gmail.com --- sound/soc/omap/omap-mcbsp.c | 41 +++++++++++++++++++++++++++++++++++++++++ sound/soc/omap/omap-mcbsp.h | 4 ++++ 2 files changed, 45 insertions(+), 0 deletions(-)
diff --git a/sound/soc/omap/omap-mcbsp.c b/sound/soc/omap/omap-mcbsp.c index 0e173e7..3341f49 100644 --- a/sound/soc/omap/omap-mcbsp.c +++ b/sound/soc/omap/omap-mcbsp.c @@ -512,6 +512,40 @@ static int omap_mcbsp_dai_set_clks_src(struct omap_mcbsp_data *mcbsp_data, return 0; }
+static int omap_mcbsp_dai_set_rcvr_src(struct omap_mcbsp_data *mcbsp_data, + int clk_id) +{ + int sel_bit, set = 0; + u16 reg = OMAP2_CONTROL_DEVCONF0; + + if (cpu_class_is_omap1()) + return -EINVAL; /* TODO: Can this be implemented for OMAP1? */ + if (mcbsp_data->bus_id != 0) + return -EINVAL; + + switch (clk_id) { + case OMAP_MCBSP_CLKR_SRC_CLKX: + set = 1; + case OMAP_MCBSP_CLKR_SRC_CLKR: + sel_bit = 3; + break; + case OMAP_MCBSP_FSR_SRC_FSX: + set = 1; + case OMAP_MCBSP_FSR_SRC_FSR: + sel_bit = 4; + break; + default: + return -EINVAL; + } + + if (set) + omap_ctrl_writel(omap_ctrl_readl(reg) | (1 << sel_bit), reg); + else + omap_ctrl_writel(omap_ctrl_readl(reg) & ~(1 << sel_bit), reg); + + return 0; +} + static int omap_mcbsp_dai_set_dai_sysclk(struct snd_soc_dai *cpu_dai, int clk_id, unsigned int freq, int dir) @@ -534,6 +568,13 @@ static int omap_mcbsp_dai_set_dai_sysclk(struct snd_soc_dai *cpu_dai, case OMAP_MCBSP_SYSCLK_CLKR_EXT: regs->pcr0 |= SCLKME; break; + + case OMAP_MCBSP_CLKR_SRC_CLKR: + case OMAP_MCBSP_CLKR_SRC_CLKX: + case OMAP_MCBSP_FSR_SRC_FSR: + case OMAP_MCBSP_FSR_SRC_FSX: + err = omap_mcbsp_dai_set_rcvr_src(mcbsp_data, clk_id); + break; default: err = -ENODEV; } diff --git a/sound/soc/omap/omap-mcbsp.h b/sound/soc/omap/omap-mcbsp.h index c8147aa..647d2f9 100644 --- a/sound/soc/omap/omap-mcbsp.h +++ b/sound/soc/omap/omap-mcbsp.h @@ -32,6 +32,10 @@ enum omap_mcbsp_clksrg_clk { OMAP_MCBSP_SYSCLK_CLK, /* Internal ICLK */ OMAP_MCBSP_SYSCLK_CLKX_EXT, /* External CLKX pin */ OMAP_MCBSP_SYSCLK_CLKR_EXT, /* External CLKR pin */ + OMAP_MCBSP_CLKR_SRC_CLKR, /* CLKR from CLKR pin */ + OMAP_MCBSP_CLKR_SRC_CLKX, /* CLKR from CLKX pin */ + OMAP_MCBSP_FSR_SRC_FSR, /* FSR from FSR pin */ + OMAP_MCBSP_FSR_SRC_FSX, /* FSR from FSX pin */ };
/* McBSP dividers */
On Fri, Aug 28, 2009 at 03:35:35PM +0300, Jarkko Nikula wrote:
Add functionaly that machine drivers can set the CLKR and FSR sources by using the snd_soc_dai_set_sysclk.
Applied, thanks.
participants (4)
-
Aggarwal, Anuj -
Jarkko Nikula -
Mark Brown -
Peter Ujfalusi