Re: [PATCH v2] ARM: dts: am33xx: Move the cppi41dma node so it's probed early

2014-05-12 Thread Ezequiel Garcia
On 12 May 10:29 AM, George Cherian wrote: On 4/25/2014 2:41 AM, Ezequiel Garcia wrote: The DMA controller is needed for the USB controller to be correctly registered. Therefore, if the DMA node is located at the end an unecessary probe deferral is produced systematically. This is easily

Re: [PATCH v2] ARM: dts: am33xx: Move the cppi41dma node so it's probed early

2014-05-11 Thread George Cherian
On 4/25/2014 2:41 AM, Ezequiel Garcia wrote: The DMA controller is needed for the USB controller to be correctly registered. Therefore, if the DMA node is located at the end an unecessary probe deferral is produced systematically. This is easily fixed by moving the node at the beggining of the

Re: [PATCH v2] ARM: dts: am33xx: Move the cppi41dma node so it's probed early

2014-05-09 Thread George Cherian
On 5/8/2014 10:30 PM, Ezequiel García wrote: Hi George, On 29 April 2014 04:58, George Cherian george.cher...@ti.com wrote: On 4/29/2014 11:49 AM, Yegor Yefremov wrote: On Thu, Apr 24, 2014 at 11:11 PM, Ezequiel Garcia ezequ...@vanguardiasur.com.ar wrote: The DMA controller is needed for the

Re: [PATCH v2] ARM: dts: am33xx: Move the cppi41dma node so it's probed early

2014-05-09 Thread Sebastian Andrzej Siewior
On 05/09/2014 08:22 AM, George Cherian wrote: Just by remodelling the dt the whole problem can be solved. I am still not convinced why we should not be doing it? Because neither ways its not the exact representation of the H/W. Ha. Now I am confused. First I assumed that the musb_am335x module

Re: [PATCH v2] ARM: dts: am33xx: Move the cppi41dma node so it's probed early

2014-05-09 Thread George Cherian
On 5/9/2014 12:55 PM, Sebastian Andrzej Siewior wrote: On 05/09/2014 08:22 AM, George Cherian wrote: Just by remodelling the dt the whole problem can be solved. I am still not convinced why we should not be doing it? Because neither ways its not the exact representation of the H/W. Ha. Now I

Re: [PATCH v2] ARM: dts: am33xx: Move the cppi41dma node so it's probed early

2014-05-09 Thread Ezequiel García
On 09 May 09:25 AM, Sebastian Andrzej Siewior wrote: On 05/09/2014 08:22 AM, George Cherian wrote: Just by remodelling the dt the whole problem can be solved. I am still not convinced why we should not be doing it? Because neither ways its not the exact representation of the H/W. Ha. Now

Re: [PATCH v2] ARM: dts: am33xx: Move the cppi41dma node so it's probed early

2014-05-08 Thread Ezequiel García
Hi George, On 29 April 2014 04:58, George Cherian george.cher...@ti.com wrote: On 4/29/2014 11:49 AM, Yegor Yefremov wrote: On Thu, Apr 24, 2014 at 11:11 PM, Ezequiel Garcia ezequ...@vanguardiasur.com.ar wrote: The DMA controller is needed for the USB controller to be correctly registered.

Re: [PATCH v2] ARM: dts: am33xx: Move the cppi41dma node so it's probed early

2014-04-29 Thread Yegor Yefremov
On Thu, Apr 24, 2014 at 11:11 PM, Ezequiel Garcia ezequ...@vanguardiasur.com.ar wrote: The DMA controller is needed for the USB controller to be correctly registered. Therefore, if the DMA node is located at the end an unecessary probe deferral is produced systematically. This is easily fixed

Re: [PATCH v2] ARM: dts: am33xx: Move the cppi41dma node so it's probed early

2014-04-29 Thread George Cherian
On 4/29/2014 11:49 AM, Yegor Yefremov wrote: On Thu, Apr 24, 2014 at 11:11 PM, Ezequiel Garcia ezequ...@vanguardiasur.com.ar wrote: The DMA controller is needed for the USB controller to be correctly registered. Therefore, if the DMA node is located at the end an unecessary probe deferral is

Re: [PATCH v2] ARM: dts: am33xx: Move the cppi41dma node so it's probed early

2014-04-29 Thread Sebastian Andrzej Siewior
On 04/29/2014 09:58 AM, George Cherian wrote: This is easily fixed by moving the node at the beggining of the child list, so it's probed first. This will give issues on module removal. Since we use device_for_each_child in remove patch, it will try to remove cppi dma controller, while the

Re: [PATCH v2] ARM: dts: am33xx: Move the cppi41dma node so it's probed early

2014-04-29 Thread George Cherian
Hi Sebastian, On 4/29/2014 1:36 PM, Sebastian Andrzej Siewior wrote: On 04/29/2014 09:58 AM, George Cherian wrote: This is easily fixed by moving the node at the beggining of the child list, so it's probed first. This will give issues on module removal. Since we use device_for_each_child in

Re: [PATCH v2] ARM: dts: am33xx: Move the cppi41dma node so it's probed early

2014-04-29 Thread Sebastian Andrzej Siewior
On 04/29/2014 10:27 AM, George Cherian wrote: Hi Sebastian, Hi George, On 4/29/2014 1:36 PM, Sebastian Andrzej Siewior wrote: On 04/29/2014 09:58 AM, George Cherian wrote: This is easily fixed by moving the node at the beggining of the child list, so it's probed first. This will give

Re: [PATCH v2] ARM: dts: am33xx: Move the cppi41dma node so it's probed early

2014-04-29 Thread Ezequiel García
On 29 April 2014 06:09, Sebastian Andrzej Siewior bige...@linutronix.de wrote: On 04/29/2014 10:27 AM, George Cherian wrote: Hi Sebastian, Hi George, On 4/29/2014 1:36 PM, Sebastian Andrzej Siewior wrote: On 04/29/2014 09:58 AM, George Cherian wrote: This is easily fixed by moving the node

[PATCH v2] ARM: dts: am33xx: Move the cppi41dma node so it's probed early

2014-04-24 Thread Ezequiel Garcia
The DMA controller is needed for the USB controller to be correctly registered. Therefore, if the DMA node is located at the end an unecessary probe deferral is produced systematically. This is easily fixed by moving the node at the beggining of the child list, so it's probed first.