diff mbox

ARM: dts: keystone-k2l: fix broken Ethernet due to disabled OSR

Message ID 59ba5ef37f17ff1c10d9d52ca06ef341b2d476eb.1490782833.git.nsekhar@ti.com (mailing list archive)
State New, archived
Headers show

Commit Message

Sekhar Nori March 29, 2017, 10:32 a.m. UTC
From: Murali Karicheri <m-karicheri2@ti.com>

Ethernet networking on K2L has been broken since v4.11-rc1. This was
caused by commit 32a34441a9bd ("ARM: keystone: dts: fix netcp clocks
and add names"). This commit inadvertently moves on-chip static RAM
clock to the end of list of clocks provided for netcp. Since keystone
PM domain support does not have a list of recognized con_ids, only the
first clock in the list comes under runtime PM management. This means
the OSR (On-chip Static RAM) clock remains disabled and that broke
networking on K2L.

The OSR is used by QMSS on K2L as an external linking RAM. However this
is a standalone RAM that can be used for non-QMSS usage (as well as from
DSP side). So add a SRAM device node for the same and add the OSR clock
to the node.

Remove the now redundant OSR clock node from netcp.

To manage all clocks defined for netCP's use by runtime PM needs keystone
generic power domain (genpd) driver support which is under works.
Meanwhile, this patch restores K2L networking and is correct irrespective
of any future genpd work since OSR is an independent module and not part
of NetCP anyway.

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Acked-by: Tero Kristo <t-kristo@ti.com>
[nsekhar@ti.com: commit message updates, port to latest mainline]
Signed-off-by: Sekhar Nori <nsekhar@ti.com>
---
 arch/arm/boot/dts/keystone-k2l-netcp.dtsi | 4 ++--
 arch/arm/boot/dts/keystone-k2l.dtsi       | 8 ++++++++
 2 files changed, 10 insertions(+), 2 deletions(-)

Comments

Santosh Shilimkar March 29, 2017, 4:13 p.m. UTC | #1
Hi Arnd, Olof,

On 3/29/17 5:32 AM, Sekhar Nori wrote:
> From: Murali Karicheri <m-karicheri2@ti.com>
>
> Ethernet networking on K2L has been broken since v4.11-rc1. This was
> caused by commit 32a34441a9bd ("ARM: keystone: dts: fix netcp clocks
> and add names"). This commit inadvertently moves on-chip static RAM
> clock to the end of list of clocks provided for netcp. Since keystone
> PM domain support does not have a list of recognized con_ids, only the
> first clock in the list comes under runtime PM management. This means
> the OSR (On-chip Static RAM) clock remains disabled and that broke
> networking on K2L.
>
> The OSR is used by QMSS on K2L as an external linking RAM. However this
> is a standalone RAM that can be used for non-QMSS usage (as well as from
> DSP side). So add a SRAM device node for the same and add the OSR clock
> to the node.
>
> Remove the now redundant OSR clock node from netcp.
>
> To manage all clocks defined for netCP's use by runtime PM needs keystone
> generic power domain (genpd) driver support which is under works.
> Meanwhile, this patch restores K2L networking and is correct irrespective
> of any future genpd work since OSR is an independent module and not part
> of NetCP anyway.
>
> Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
> Acked-by: Tero Kristo <t-kristo@ti.com>
> [nsekhar@ti.com: commit message updates, port to latest mainline]
> Signed-off-by: Sekhar Nori <nsekhar@ti.com>
> ---
Acked-by: Santosh Shilimkar <ssantosh@kernel.org>

Can you please pick this up for rcx fixes ?

Regards,
Santosh
Arnd Bergmann March 31, 2017, 10:13 a.m. UTC | #2
On Wed, Mar 29, 2017 at 6:13 PM, santosh.shilimkar@oracle.com
<santosh.shilimkar@oracle.com> wrote:
>> Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
>> Acked-by: Tero Kristo <t-kristo@ti.com>
>> [nsekhar@ti.com: commit message updates, port to latest mainline]
>> Signed-off-by: Sekhar Nori <nsekhar@ti.com>
>> ---
>
> Acked-by: Santosh Shilimkar <ssantosh@kernel.org>
>
> Can you please pick this up for rcx fixes ?

Applied, thanks!

       Arnd
Sekhar Nori May 26, 2017, 7:44 a.m. UTC | #3
Hi Arnd,

On Friday 31 March 2017 03:43 PM, Arnd Bergmann wrote:
> On Wed, Mar 29, 2017 at 6:13 PM, santosh.shilimkar@oracle.com
> <santosh.shilimkar@oracle.com> wrote:
>>> Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
>>> Acked-by: Tero Kristo <t-kristo@ti.com>
>>> [nsekhar@ti.com: commit message updates, port to latest mainline]
>>> Signed-off-by: Sekhar Nori <nsekhar@ti.com>
>>> ---
>>
>> Acked-by: Santosh Shilimkar <ssantosh@kernel.org>
>>
>> Can you please pick this up for rcx fixes ?
> 
> Applied, thanks!

Looks like this fix never made to to mainline. Do you have the patch
with you still? It will have to be marked for v4.11 stable and pushed.

Thanks,
Sekhar
Arnd Bergmann May 26, 2017, 8:12 a.m. UTC | #4
On Fri, May 26, 2017 at 9:44 AM, Sekhar Nori <nsekhar@ti.com> wrote:
> Hi Arnd,
>
> On Friday 31 March 2017 03:43 PM, Arnd Bergmann wrote:
>> On Wed, Mar 29, 2017 at 6:13 PM, santosh.shilimkar@oracle.com
>> <santosh.shilimkar@oracle.com> wrote:
>>>> Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
>>>> Acked-by: Tero Kristo <t-kristo@ti.com>
>>>> [nsekhar@ti.com: commit message updates, port to latest mainline]
>>>> Signed-off-by: Sekhar Nori <nsekhar@ti.com>
>>>> ---
>>>
>>> Acked-by: Santosh Shilimkar <ssantosh@kernel.org>
>>>
>>> Can you please pick this up for rcx fixes ?
>>
>> Applied, thanks!
>
> Looks like this fix never made to to mainline. Do you have the patch
> with you still? It will have to be marked for v4.11 stable and pushed.

Sorry about the mistake, I don't know what happened here but I
found it in my git reflog and applied it to the latest fixes branch
with the stable tag added.

Thanks for checking this and the reminder.

       Arnd
Sekhar Nori May 26, 2017, 8:16 a.m. UTC | #5
On Friday 26 May 2017 01:42 PM, Arnd Bergmann wrote:
> On Fri, May 26, 2017 at 9:44 AM, Sekhar Nori <nsekhar@ti.com> wrote:
>> Hi Arnd,
>>
>> On Friday 31 March 2017 03:43 PM, Arnd Bergmann wrote:
>>> On Wed, Mar 29, 2017 at 6:13 PM, santosh.shilimkar@oracle.com
>>> <santosh.shilimkar@oracle.com> wrote:
>>>>> Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
>>>>> Acked-by: Tero Kristo <t-kristo@ti.com>
>>>>> [nsekhar@ti.com: commit message updates, port to latest mainline]
>>>>> Signed-off-by: Sekhar Nori <nsekhar@ti.com>
>>>>> ---
>>>>
>>>> Acked-by: Santosh Shilimkar <ssantosh@kernel.org>
>>>>
>>>> Can you please pick this up for rcx fixes ?
>>>
>>> Applied, thanks!
>>
>> Looks like this fix never made to to mainline. Do you have the patch
>> with you still? It will have to be marked for v4.11 stable and pushed.
> 
> Sorry about the mistake, I don't know what happened here but I
> found it in my git reflog and applied it to the latest fixes branch
> with the stable tag added.
> 
> Thanks for checking this and the reminder.

No problem, Arnd. Thanks for taking care of it.

Regards,
Sekhar
diff mbox

Patch

diff --git a/arch/arm/boot/dts/keystone-k2l-netcp.dtsi b/arch/arm/boot/dts/keystone-k2l-netcp.dtsi
index b6f26824e83a..66f615a74118 100644
--- a/arch/arm/boot/dts/keystone-k2l-netcp.dtsi
+++ b/arch/arm/boot/dts/keystone-k2l-netcp.dtsi
@@ -137,8 +137,8 @@  netcp: netcp@26000000 {
 	/* NetCP address range */
 	ranges = <0 0x26000000 0x1000000>;
 
-	clocks = <&clkpa>, <&clkcpgmac>, <&chipclk12>, <&clkosr>;
-	clock-names = "pa_clk", "ethss_clk", "cpts", "osr_clk";
+	clocks = <&clkpa>, <&clkcpgmac>, <&chipclk12>;
+	clock-names = "pa_clk", "ethss_clk", "cpts";
 	dma-coherent;
 
 	ti,navigator-dmas = <&dma_gbe 0>,
diff --git a/arch/arm/boot/dts/keystone-k2l.dtsi b/arch/arm/boot/dts/keystone-k2l.dtsi
index b58e7ebc0919..148650406cf7 100644
--- a/arch/arm/boot/dts/keystone-k2l.dtsi
+++ b/arch/arm/boot/dts/keystone-k2l.dtsi
@@ -232,6 +232,14 @@ 
 			};
 		};
 
+		osr: sram@70000000 {
+			compatible = "mmio-sram";
+			reg = <0x70000000 0x10000>;
+			#address-cells = <1>;
+			#size-cells = <1>;
+			clocks = <&clkosr>;
+		};
+
 		dspgpio0: keystone_dsp_gpio@02620240 {
 			compatible = "ti,keystone-dsp-gpio";
 			gpio-controller;