diff mbox

[2/2] ARM: vexpress: Add interrupt-affinity

Message ID 1430416448-25203-1-git-send-email-r.schwebel@pengutronix.de (mailing list archive)
State New, archived
Headers show

Commit Message

Robert Schwebel April 30, 2015, 5:54 p.m. UTC
Without this property, we get this boot warning:

[    0.459361] CPU PMU: Failed to parse <no-node>/interrupt-affinity[0]

Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
---
 arch/arm/boot/dts/vexpress-v2p-ca9.dts | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

Comments

Robert Schwebel April 30, 2015, 5:55 p.m. UTC | #1
On Thu, Apr 30, 2015 at 07:54:08PM +0200, Robert Schwebel wrote:
> Without this property, we get this boot warning:
> 
> [    0.459361] CPU PMU: Failed to parse <no-node>/interrupt-affinity[0]
> 
> Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>

However, with this patch applied, the messages above disappear, but I
still get this:

[    0.495280] CPU PMU: Failed to find logical CPU for cpu

on a kernel configured with CONFIG_SMP=n. nr_cpu_ids is 1 there, while 
the loop still loops over all available cores. Ths makes the logic fail.

rsc
Sudeep Holla May 1, 2015, 1:44 p.m. UTC | #2
On 30/04/15 18:55, Robert Schwebel wrote:
> On Thu, Apr 30, 2015 at 07:54:08PM +0200, Robert Schwebel wrote:
>> Without this property, we get this boot warning:
>>
>> [    0.459361] CPU PMU: Failed to parse <no-node>/interrupt-affinity[0]
>>
>> Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>

May be you can also mention the commit introducing the change in the log
something like:

"
Commit 9fd85eb502a7 ("ARM: pmu: add support for interrupt-affinity
property") added optional optional "interrupt-affinity" property, to
specify CPU affinity for each SPI listed in the interrupts property.

Without this property, we get this boot warning:
	CPU PMU: Failed to parse <no-node>/interrupt-affinity[0]

This patch add interrupt-affinity to the PMU node in CA9x4 device tree.
"

Otherwise:
Acked-by: Sudeep Holla <sudeep.holla@arm.com>

>
> However, with this patch applied, the messages above disappear, but I
> still get this:
>
> [    0.495280] CPU PMU: Failed to find logical CPU for cpu
>
> on a kernel configured with CONFIG_SMP=n. nr_cpu_ids is 1 there, while
> the loop still loops over all available cores. Ths makes the logic fail.

This is a different issue. You must also see warning from devtree.c when
parsing CPUs complaining about more CPU nodes found in DT. IIUC the
functionality still works fine.

Regards,
Sudeep
diff mbox

Patch

diff --git a/arch/arm/boot/dts/vexpress-v2p-ca9.dts b/arch/arm/boot/dts/vexpress-v2p-ca9.dts
index a411274..9d71493 100644
--- a/arch/arm/boot/dts/vexpress-v2p-ca9.dts
+++ b/arch/arm/boot/dts/vexpress-v2p-ca9.dts
@@ -33,28 +33,28 @@ 
 		#address-cells = <1>;
 		#size-cells = <0>;
 
-		cpu@0 {
+		A9_0: cpu@0 {
 			device_type = "cpu";
 			compatible = "arm,cortex-a9";
 			reg = <0>;
 			next-level-cache = <&L2>;
 		};
 
-		cpu@1 {
+		A9_1: cpu@1 {
 			device_type = "cpu";
 			compatible = "arm,cortex-a9";
 			reg = <1>;
 			next-level-cache = <&L2>;
 		};
 
-		cpu@2 {
+		A9_2: cpu@2 {
 			device_type = "cpu";
 			compatible = "arm,cortex-a9";
 			reg = <2>;
 			next-level-cache = <&L2>;
 		};
 
-		cpu@3 {
+		A9_3: cpu@3 {
 			device_type = "cpu";
 			compatible = "arm,cortex-a9";
 			reg = <3>;
@@ -182,6 +182,11 @@ 
 			     <0 61 4>,
 			     <0 62 4>,
 			     <0 63 4>;
+		interrupt-affinity = <&A9_0>,
+		                     <&A9_1>,
+		                     <&A9_2>,
+		                     <&A9_3>;
+
 	};
 
 	dcc {