=== Device Tree === * Tested audio on imx51 babbage board with a full dt kernel, which means I got i2c, sgtl5000 codec, mc13892 regulator, ssi and sdma working on device tree. But due to some unresolved dependency and open issues, only i2c and sdma dt support patches were sent out for review. With a v2 on sdma, there is no more comments. Suppose they are in the good shape for v3.1. * Sent a small patch to enable dt for imx2_wdt driver. * Sent a couple of patches to add device tree support for gpio_keys. * I'm optimistic to expect the dt support for serial, fec, esdhc, gpio, spi, i2c, sdma, imx2_wdt and gpio_keys will be on mainline when v3.1 merge window closes. Plus that gpio_leds already gets dt support, all four imx53 boards, mx53_ard, mx53_evk, mx53_loco, and mx53_smd should be killed by imx53-dt soon. So that we can stop people adding new mx53 boards files and pull them into the dt effort.
=== Plan === * Start looking at Grant's suggestion about directly encoding iomux register, and try to sell it to Sascha. * Add imx53-dt to kill mx53_ard, mx53_evk, mx53_loco and mx53_smd.
Hi Shawn,
=== Device Tree ===
- Tested audio on imx51 babbage board with a full dt kernel, which means I got i2c, sgtl5000 codec, mc13892 regulator, ssi and sdma working on device tree.
Did you get the sgtl5000 to work out of the box? I'd think the cache handling is really broken [1]...
Regards,
Wolfram
[1] http://mailman.alsa-project.org/pipermail/alsa-devel/2011-July/041974.html
On Mon, Jul 18, 2011 at 07:39:51PM +0200, Wolfram Sang wrote:
Hi Shawn,
=== Device Tree ===
- Tested audio on imx51 babbage board with a full dt kernel, which means I got i2c, sgtl5000 codec, mc13892 regulator, ssi and sdma working on device tree.
Did you get the sgtl5000 to work out of the box? I'd think the cache handling
No. I started from a fixed version. It's one of the open issues I meant in my report. It's great to see you start submitting to shoot it. I will give it a test today.
For record, besides the above (solved now :), we have to address a number of issues to get the device tree audio ready for submitting.
* The dynamic device id in device tree makes the device matching using name in ASoC core even harder[1].
* The regulator binding for device tree is still missing. There is a post [2] from Zhuang Haojian. But it's pretty rough and still needs some work.
* There is an argument between merging imx-ssi into fsl-ssi (ssi driver on powerpc) and simply dding dt support for imx-ssi [3].
* We still have not reached an agreement on how to bind dma request line (not the physical channel) in device tree [4].
[1] http://thread.gmane.org/gmane.linux.alsa.devel/87086/focus=87151 [2] http://thread.gmane.org/gmane.linux.ports.arm.kernel/124322/focus=124486 [3] http://lists.ozlabs.org/pipermail/devicetree-discuss/2011-July/006651.html [4] http://lists.ozlabs.org/pipermail/devicetree-discuss/2011-July/006647.html
Regards, Shawn
is really broken [1]...
Regards,
Wolfram
[1] http://mailman.alsa-project.org/pipermail/alsa-devel/2011-July/041974.html
-- Pengutronix e.K. | Wolfram Sang | Industrial Linux Solutions | http://www.pengutronix.de/ |
=== Device Tree ===
- Tested audio on imx51 babbage board with a full dt kernel, which means I got i2c, sgtl5000 codec, mc13892 regulator, ssi and sdma working on device tree.
Did you get the sgtl5000 to work out of the box? I'd think the cache handling
No. I started from a fixed version. It's one of the open issues
Ah, okay. I was curious if the driver does work out of the box for some configuration.
I meant in my report. It's great to see you start submitting to shoot it. I will give it a test today.
Thanks for that and the update here!
Wolfram
linaro-kernel@lists.linaro.org
-
Shawn Guo -
Wolfram Sang