[alsa-devel] Audio on omap3+twl4030 codec
Hi all,
I am working on omap3430 and twl4030 combination for audio.
I am wokring on 2.6.26 whre the support was not there for omap3 mcbsp. so i ported some stuff from the 2.6.27 latest git tree and now audio is working on omap3 evm. with the very same kernel when i try with my custom board. I am unable to hear audoi.
I get audio when i boot out of MMC (x-load and u-boot in mmc and fiel system is nfs). But, when i usse nand(x-load and u-boot in nand and file system is on nfs) i dont get audio out meams aplay and arecord seems working but i dont hear any audio.
I probe and checked the Bit clock and Word clock of the I2S which seems are comming when i try a aplay and arecord..
I am sure that hardware has no problem. Can any one tell me what might be the issue and where i have to start debugging the issue.
If this is not the right palce to put my question can any one tell me the correct mailing list.
On Wed, Oct 22, 2008 at 06:17:39PM +0530, naveen krishna ch wrote:
I am working on omap3430 and twl4030 combination for audio.
CCing in several people who've been working on these drivers recently.
I am wokring on 2.6.26 whre the support was not there for omap3 mcbsp. so i ported some stuff from the 2.6.27 latest git tree and now audio is working on omap3 evm. with the very same kernel when i try with my custom board. I am unable to hear audoi.
I get audio when i boot out of MMC (x-load and u-boot in mmc and fiel system is nfs). But, when i usse nand(x-load and u-boot in nand and file system is on nfs) i dont get audio out meams aplay and arecord seems working but i dont hear any audio.
With the same software in all cases?
I probe and checked the Bit clock and Word clock of the I2S which seems are comming when i try a aplay and arecord..
Is there any data at all on the data pin? Have you checked that the clock rates for the two clocks are correct (if not enough bit clocks are generated per frame clock then the DAC will not get enough data to produce a full sample as output)?
I am sure that hardware has no problem. Can any one tell me what might be the issue and where i have to start debugging the issue.
If this is not the right palce to put my question can any one tell me the correct mailing list.
This should be a good place to ask.
I forget to add one more observation that is i have a older omap2 and twl4030 asoc codec driver and driver/ssi mcbsp support in a 2.6.24 kernel.
when i try to use this in my coustum board the audio is working fine all the ways.
On Wed, Oct 22, 2008 at 6:40 PM, Mark Brown broonie@sirena.org.uk wrote:
On Wed, Oct 22, 2008 at 06:17:39PM +0530, naveen krishna ch wrote:
I am working on omap3430 and twl4030 combination for audio.
CCing in several people who've been working on these drivers recently.
I am new so i dont have much information of the other who is working on similar things.
I am wokring on 2.6.26 whre the support was not there for omap3 mcbsp. so i ported some stuff from the 2.6.27 latest git tree and now audio is working on omap3 evm. with the very same kernel when i try with my custom board. I am unable to hear audoi.
I get audio when i boot out of MMC (x-load and u-boot in mmc and fiel
system
is nfs). But, when i usse nand(x-load and u-boot in nand and file system is on
nfs) i
dont get audio out meams aplay and arecord seems working but i dont hear any audio.
With the same software in all cases?
I used the same x-loader, u-boot and kernel and the hardware too in all the cases
I probe and checked the Bit clock and Word clock of the I2S which seems
are
comming when i try a aplay and arecord..
Is there any data at all on the data pin? Have you checked that the clock rates for the two clocks are correct (if not enough bit clocks are generated per frame clock then the DAC will not get enough data to produce a full sample as output)?
yeah there is data seems coming all the times i have probed the clocks and data pins. the data seems comming out of OMAP3 but i have no point here to probe in the twl4030. i used a 22050hz wav format file and the i noticed approx18Khz.
I probed with the same x-load and u-boot and 2.6.24 kernel image setup also.
then also the bit clock and word clock are the same as what i am getting with 2.6.26 kernel image
I am sure that hardware has no problem. Can any one tell me what might be the issue and where i have to start debugging the issue.
If this is not the right palce to put my question can any one tell me the correct mailing list.
This should be a good place to ask.
Can anyone refer me towards the ones who worked on this kind of audio codec driver.
On Wed, 22 Oct 2008 18:54:12 +0530 "ext naveen krishna ch" naveenkrishna.ch@gmail.com wrote:
I forget to add one more observation that is i have a older omap2 and twl4030 asoc codec driver and driver/ssi mcbsp support in a 2.6.24 kernel.
when i try to use this in my coustum board the audio is working fine all the ways.
I am wokring on 2.6.26 whre the support was not there for omap3 mcbsp. so i ported some stuff from the 2.6.27 latest git tree and now audio is working on omap3 evm. with the very same kernel when i try with my custom board. I am unable to hear audoi.
So to summarize:
- 2.6.26 + backported drivers on OMAP3 EVM -> audio works - 2.6.26 + backported drivers on custom board with NAND boot -> no audio - 2.6.26 + backported drivers on custom board with MMC boot -> audio works - 2.6.24 (TI SDP release?) on custom board -> audio works
And I understood also that bootloader images were the same in all of these cases. Did I get these correctly?
Is the rootfile system the same between these cases? I.e. that same mixer settings like audio volumes etc. are used?
Can anyone refer me towards the ones who worked on this kind of audio codec driver.
Worth to ask also on linux-omap mailing list (linux-omap@vger.kernel.org) since TI engineers are reading it.
Jarkko
On Thu, Oct 23, 2008 at 2:34 PM, Jarkko Nikula jarkko.nikula@nokia.comwrote:
On Wed, 22 Oct 2008 18:54:12 +0530 "ext naveen krishna ch" naveenkrishna.ch@gmail.com wrote:
I forget to add one more observation that is i have a older omap2 and twl4030 asoc codec driver and driver/ssi mcbsp support in a 2.6.24 kernel.
when i try to use this in my coustum board the audio is working fine all
the
ways.
I am wokring on 2.6.26 whre the support was not there for omap3
mcbsp.
so i ported some stuff from the 2.6.27 latest git tree and now audio
is
working on omap3 evm. with the very same kernel when i try with my custom board. I am
unable to
hear audoi.
So to summarize:
- 2.6.26 + backported drivers on OMAP3 EVM -> audio works
- 2.6.26 + backported drivers on custom board with NAND boot -> no audio
- 2.6.26 + backported drivers on custom board with MMC boot -> audio
works
- 2.6.24 (TI SDP release?) on custom board -> audio works
And I understood also that bootloader images were the same in all of these cases. Did I get these correctly?
Yeah till here u got every thing correctly.
Is the rootfile system the same between these cases? I.e. that same mixer settings like audio volumes etc. are used?
i Have used the same root file system also that too over nfs all the cases.
Can anyone refer me towards the ones who worked on this kind of audio
codec
driver.
Worth to ask also on linux-omap mailing list (linux-omap@vger.kernel.org) since TI engineers are reading it.
Jarkko
On Thu, 23 Oct 2008 14:50:33 +0530 "ext naveen krishna ch" naveenkrishna.ch@gmail.com wrote:
So to summarize:
- 2.6.26 + backported drivers on OMAP3 EVM -> audio works
- 2.6.26 + backported drivers on custom board with NAND boot -> no audio
- 2.6.26 + backported drivers on custom board with MMC boot -> audio
works
- 2.6.24 (TI SDP release?) on custom board -> audio works
And I understood also that bootloader images were the same in all of these cases. Did I get these correctly?
Yeah till here u got every thing correctly.
Then it looks that different pin muxing is not the reason if the same bootloader images behave differently when they are booted from NAND vs. MMC.
I'm bit out of idea what might then cause that audio doesn't play. Timing issue? Sys.boot pin states affect McBSP muxing or TWL4030, etc?
Is the rootfile system the same between these cases? I.e. that same mixer settings like audio volumes etc. are used?
i Have used the same root file system also that too over nfs all the cases.
Ok. Is it possible you to try with up-to-date linux-omap [1] kernel? There have been lot of changes regarding OMAP3 and TWL4030 since 2.6.26 so problem might have gone and then it's easier for others trying to verify problem on other similar boards.
Jarkko
[1] git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6.git or http://www.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6.git
participants (3)
-
Jarkko Nikula -
Mark Brown -
naveen krishna ch