From: Kishon Vijay Abraham I <hidden> Date: 2012-09-19 10:33:39
This patch series contains a patch to explicitly add reg property
to the ocp2scp dt node and a patch to add the subnode for ocp2scp
(omap-usb2).
Since the original patch series that adds ocp2scp driver and ocp2scp
data are merged in arm-soc, I'm sending this patch series based on the
same tree.
git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc.git
But Benoit already pointed out that there's going to be merge conflicts
as he has sent a bunch of dt data patches to linux-omap (devel-dt) tree.
So now I'm not sure how that can be actually avoided.
Kishon Vijay Abraham I (2):
ARM: dts: omap4: add *reg* property for ocp2scp
ARM: dts: omap: Add omap-usb2 dt data
.../devicetree/bindings/bus/omap-ocp2scp.txt | 14 ++++++++++++++
arch/arm/boot/dts/omap4.dtsi | 8 +++++++-
2 files changed, 21 insertions(+), 1 deletion(-)
--
1.7.9.5
From: Kishon Vijay Abraham I <hidden> Date: 2012-09-19 10:33:25
*reg* property for ocp2scp was previously obtained from ti,hwmods property.
But that is now explicitly added to the dt node. Also updated the
documentation with *reg* info.
Signed-off-by: Kishon Vijay Abraham I <redacted>
---
.../devicetree/bindings/bus/omap-ocp2scp.txt | 14 ++++++++++++++
arch/arm/boot/dts/omap4.dtsi | 3 ++-
2 files changed, 16 insertions(+), 1 deletion(-)
@@ -2,9 +2,23 @@ properties: - compatible : Should be "ti,omap-ocp2scp"+- reg : Address and length of the register set for the device - #address-cells, #size-cells : Must be present if the device has sub-nodes - ranges : the child address space are mapped 1:1 onto the parent address space - ti,hwmods : must be "ocp2scp_usb_phy" Sub-nodes: All the devices connected to ocp2scp are described using sub-node to ocp2scp++ocp2scp at 4a0ad000 {+ compatible = "ti,omap-ocp2scp";+ reg = <0x4a0ad000 0x1f>;+ #address-cells = <1>;+ #size-cells = <1>;+ ranges;+ ti,hwmods = "ocp2scp_usb_phy";+ subnode1 {+ };+ subnode2 {+ };+};
From: Kishon Vijay Abraham I <hidden> Date: 2012-09-19 10:33:27
Add omap-usb2 data node in omap4 device tree file. Since omap-usb2 is
connected to ocp2scp, omap-usb2 dt data is added as a child node
of ocp2scp.
Acked-by: Felipe Balbi <redacted>
Signed-off-by: Kishon Vijay Abraham I <redacted>
---
arch/arm/boot/dts/omap4.dtsi | 5 +++++
1 file changed, 5 insertions(+)
@@ -303,6 +303,11 @@#size-cells=<1>;ranges;ti,hwmods="ocp2scp_usb_phy";+usb2phyat4a0ad080{+compatible="ti,omap-usb2";+reg=<0x4a0ad0800x58>,+<0x4a0023000x4>;/* TO BE REMOVED: SCM */+};};};};
Hi Tony,
On 09/19/2012 12:32 PM, Kishon Vijay Abraham I wrote:
This patch series contains a patch to explicitly add reg property
to the ocp2scp dt node and a patch to add the subnode for ocp2scp
(omap-usb2).
Since the original patch series that adds ocp2scp driver and ocp2scp
data are merged in arm-soc, I'm sending this patch series based on the
same tree.
git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc.git
But Benoit already pointed out that there's going to be merge conflicts
as he has sent a bunch of dt data patches to linux-omap (devel-dt) tree.
So now I'm not sure how that can be actually avoided.
What would you suggest?
I merged arm-soc/drivers/ocp2scp into my dts branch to get a proper base
for that series, but as expected, it conflicts :-(
So I fixed that and pulled this series on top of it.
It is something acceptable for a pull request? We already did that in
the past, but at that time it was not conflicting.
Thanks,
Benoit
Hi Benoit,
On Wednesday 19 September 2012 04:02 PM, Kishon Vijay Abraham I wrote:
Add omap-usb2 data node in omap4 device tree file. Since omap-usb2 is
connected to ocp2scp, omap-usb2 dt data is added as a child node
of ocp2scp.
Acked-by: Felipe Balbi <redacted>
Signed-off-by: Kishon Vijay Abraham I <redacted>
This patch is also missing in mainline :-(
Thanks
Kishon
@@ -303,6 +303,11 @@#size-cells=<1>;ranges;ti,hwmods="ocp2scp_usb_phy";+usb2phyat4a0ad080{+compatible="ti,omap-usb2";+reg=<0x4a0ad0800x58>,+<0x4a0023000x4>;/* TO BE REMOVED: SCM */+};};};};
Hi Benoit,
On Wednesday 19 September 2012 04:02 PM, Kishon Vijay Abraham I wrote:
quoted
Add omap-usb2 data node in omap4 device tree file. Since omap-usb2 is
connected to ocp2scp, omap-usb2 dt data is added as a child node
of ocp2scp.
Acked-by: Felipe Balbi <redacted>
Signed-off-by: Kishon Vijay Abraham I <redacted>
@@ -303,6 +303,11 @@#size-cells=<1>;ranges;ti,hwmods="ocp2scp_usb_phy";+usb2phyat4a0ad080{+compatible="ti,omap-usb2";+reg=<0x4a0ad0800x58>,+<0x4a0023000x4>;/* TO BE REMOVED: SCM */
Rob and I did not agree to use that temp hack in the case of DT, so you
were supposed to repost with a proper driver for the SCM part that
control the USB.
Regards,
Benoit
Hi,
On Thursday 10 January 2013 03:34 PM, Benoit Cousson wrote:
On 01/10/2013 10:31 AM, kishon wrote:
quoted
Hi Benoit,
On Wednesday 19 September 2012 04:02 PM, Kishon Vijay Abraham I wrote:
quoted
Add omap-usb2 data node in omap4 device tree file. Since omap-usb2 is
connected to ocp2scp, omap-usb2 dt data is added as a child node
of ocp2scp.
Acked-by: Felipe Balbi <redacted>
Signed-off-by: Kishon Vijay Abraham I <redacted>
@@ -303,6 +303,11 @@#size-cells=<1>;ranges;ti,hwmods="ocp2scp_usb_phy";+usb2phyat4a0ad080{+compatible="ti,omap-usb2";+reg=<0x4a0ad0800x58>,+<0x4a0023000x4>;/* TO BE REMOVED: SCM */
Rob and I did not agree to use that temp hack in the case of DT, so you
were supposed to repost with a proper driver for the SCM part that
control the USB.
From: Felipe Balbi <hidden> Date: 2013-01-10 12:46:21
On Thu, Jan 10, 2013 at 03:46:55PM +0530, kishon wrote:
Hi,
On Thursday 10 January 2013 03:34 PM, Benoit Cousson wrote:
quoted
On 01/10/2013 10:31 AM, kishon wrote:
quoted
Hi Benoit,
On Wednesday 19 September 2012 04:02 PM, Kishon Vijay Abraham I wrote:
quoted
Add omap-usb2 data node in omap4 device tree file. Since omap-usb2 is
connected to ocp2scp, omap-usb2 dt data is added as a child node
of ocp2scp.
Acked-by: Felipe Balbi <redacted>
Signed-off-by: Kishon Vijay Abraham I <redacted>
@@ -303,6 +303,11 @@#size-cells=<1>;ranges;ti,hwmods="ocp2scp_usb_phy";+usb2phyat4a0ad080{+compatible="ti,omap-usb2";+reg=<0x4a0ad0800x58>,+<0x4a0023000x4>;/* TO BE REMOVED: SCM */
Rob and I did not agree to use that temp hack in the case of DT, so you
were supposed to repost with a proper driver for the SCM part that
control the USB.
From: Tony Lindgren <tony@atomide.com> Date: 2013-01-10 18:35:17
* Felipe Balbi [off-list ref] [130110 04:49]:
On Thu, Jan 10, 2013 at 03:46:55PM +0530, kishon wrote:
quoted
On Thursday 10 January 2013 03:34 PM, Benoit Cousson wrote:
quoted
Rob and I did not agree to use that temp hack in the case of DT, so you
were supposed to repost with a proper driver for the SCM part that
control the USB.
Ok. Missed that. Will work on that.
Didn't Tony say we should access SCM directly since it's a one time
thing ?
Tony ?
The SCM parts should be a separate driver that eventually will be
a child of the minimal SCM core driver. The reason why it should be
separate is because the SCM registers are in a different block as
pointed out by Paul.
There's a good chance that we'll be able to eventually handle the USB
SCM registers directly via pinctrl-single.c using pinconf-bits + generic
pinconf binding. That is pending the changes from Haojian to implement
generic pinconf for pinctrl-single.c.
As the USB SCM registers can get quite complex, I suggest creating a
minimal driver to handle that, and then eventually that driver might
just use the generic pinconf API for setting the SCM registers for
things like bias voltage, pulls and comparators.
Regards,
Tony