Hi Leo,

On 12/09/18 11:17, leo.yan@linaro.org wrote:

Hi Suzuki,

On Tue, Sep 11, 2018 at 11:17:05AM +0100, Suzuki K Poulose wrote:

quoted

Switch to updated coresight bindings for hw ports

Cc: Andy Gross <redacted>
Cc: David Brown <redacted>
Cc: Ivan T. Ivanov <redacted>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
---
  arch/arm64/boot/dts/qcom/msm8916.dtsi | 98 ++++++++++++++++++++++-------------
  1 file changed, 63 insertions(+), 35 deletions(-)

diff --git a/arch/arm64/boot/dts/qcom/msm8916.dtsi b/arch/arm64/boot/dts/qcom/msm8916.dtsi
index 7b32b89..5bed52b 100644
--- a/arch/arm64/boot/dts/qcom/msm8916.dtsi
+++ b/arch/arm64/boot/dts/qcom/msm8916.dtsi

@@ -1099,10 +1099,11 @@
  			clocks = <&rpmcc RPM_QDSS_CLK>, <&rpmcc RPM_QDSS_A_CLK>;
  			clock-names = "apb_pclk", "atclk";
  
-			port {
-				tpiu_in: endpoint {
-					slave-mode;
-					remote-endpoint = <&replicator_out1>;
+			out-ports {

Here should be in-ports for tpiu node?

BTW, if upper comment is valid, this means the DT binding doc also
gives wrong example for tpiu node [1].

You are right on both counts.

I tested this patch with upper changing to 'in-ports', etm with perf
mode can work well on DB410c board.

Thanks a lot for the testing. I will fix both the problems.

Suzuki