On 1/14/20 8:20 AM, Christophe ROULLIER wrote:
> On 1/13/20 6:30 PM, Marek Vasut wrote:
>> On 1/13/20 5:14 PM, Christophe ROULLIER wrote:
>> [...]
>>>>> Christophe you can complete my answer if it is not enough clear.
>>>> I would expect that you should describe _all_ the clock which are routed
>>>> into the ethernet IP in the DT and if the driver doesn't need some of
>>>> those clock, the driver should just not enable such clock.
>>> Hi Marek,
>> Hi,
>>> You are right, we will add by default all clocks used by stm32mp157, and
>>> modify driver to decide if we need to enable it or not.
>>> So ACK to add ETHCK_K in stm32mp157c.dtsi
>> Do you want me to send the kernel patch ? I have it somewhere here too,
>> I can roll it out tomorrow-ish.
> Hi Marek,
> I have not changed yet, the kernel and uboot driver code, to manage this 
> behavior. I will implement and push new code soon.

The kernel code already does what is needed. You only need similar patch
to the DT, cfr attachment.
>From a84e5d5680c2b1e3e08cba67aaff0b5417f80790 Mon Sep 17 00:00:00 2001
From: Marek Vasut <ma...@denx.de>
Date: Tue, 3 Dec 2019 00:50:25 +0100
Subject: [PATCH] ARM: dts: stm32: Add missing ETHCK clock to ethernet node

Add missing 'eth-ck' clock to the ethernet node. These clock are used to
generate external clock signal for the PHY in case 'st,eth_ref_clk_sel'
is specified.

Signed-off-by: Marek Vasut <ma...@denx.de>
Cc: Patrick Delaunay <patrick.delau...@st.com>
Cc: Patrice Chotard <patrice.chot...@st.com>
 arch/arm/boot/dts/stm32mp157c.dtsi | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/arm/boot/dts/stm32mp157c.dtsi b/arch/arm/boot/dts/stm32mp157c.dtsi
index f20e0101b577..9e7e6328f97c 100644
--- a/arch/arm/boot/dts/stm32mp157c.dtsi
+++ b/arch/arm/boot/dts/stm32mp157c.dtsi
@@ -1380,11 +1380,13 @@
 			clock-names = "stmmaceth",
+				      "eth-ck",
 			clocks = <&rcc ETHMAC>,
 				 <&rcc ETHTX>,
 				 <&rcc ETHRX>,
+				 <&rcc ETHCK_K>,
 				 <&rcc ETHSTP>,
 				 <&rcc SYSCFG>;
 			st,syscon = <&syscfg 0x4>;

Reply via email to