[PATCH V2 00/13] Add multipd remoteproc support

[PATCH V2 00/13] Add multipd remoteproc support

[Manikanta Mylavarapu]

APSS brings Q6 out of reset and then Q6 brings
WCSS block (wifi radio's) out of reset.

				   ---------------
			      -->  |WiFi 2G radio|
			      |	   --------------
			      |
--------	-------	      |
| APSS | --->   |QDSP6|  -----|
---------	-------       |
                              |
      			      |
			      |   --------------
			      --> |WiFi 5G radio|
				  --------------

Problem here is if any radio crashes, subsequently other
radio also should crash because Q6 crashed. Let's say
2G radio crashed, Q6 should pass this info to APSS. Only
Q6 processor interrupts registered with APSS. Obviously
Q6 should crash and raise fatal interrupt to APSS. Due
to this 5G radio also crashed. But no issue in 5G radio,
because of 2G radio crash 5G radio also impacted.

In multi pd model, this problem is resolved. Here WCSS
functionality (WiFi radio's) moved out from Q6 root pd
to a separate user pd. Due to this, radio's independently
pass their status info to APPS with out crashing Q6. So
other radio's won't be impacted.

Pd means protection domain. It's similar to process in Linux.
Here QDSP6 processor runs each wifi radio functionality on a
separate process. One process can't access other process
resources, so this is termed as PD i.e protection domain.

						---------
					    	|WiFi    |
					    --> |2G radio|
					    | 	---------
------	Start Q6     		-------     |
|    |	------------------>     |     |     |
|    |  Start WCSS PD1 (2G)   	|     |	    |
|APSS|	----------------------->|QDSP6|-----|
|    |	Start WCSS PD1 (5G)	|     |
|    |	----------------------->|     |-----|
------		     		-------     |
					    |
					    |	-----------
					    |-->|WiFi	  |
						|5G radio |
						-----------
According to linux terminology, here consider Q6 as root
i.e it provide all services, WCSS (wifi radio's) as user
i.e it uses services provided by root.

Since Q6 root & WCSS user pd's able to communicate with
APSS individually, multipd remoteproc driver registers
each PD with rproc framework. Here clients (Wifi host drivers)
intrested on WCSS PD rproc, so multipd driver start's root
pd in the context of WCSS user pd rproc start. Similarly
on down path, root pd will be stopped after wcss user pd
stopped.

Here WCSS(user) PD is dependent on Q6(root) PD, so first
q6 pd should be up before wcss pd. After wcss pd goes down,
q6 pd should be turned off.

rproc->ops->start(userpd_rproc) {
	/* Boot root pd rproc */
	rproc_boot(upd_dev->parent);
	---
	/* user pd rproc start sequence */
	---
	---
}
With this way we ensure that root pd brought up before userpd.

rproc->ops->stop(userpd_rproc) {
	---
	---
	/* user pd rproc stop sequence */
	---
	---
	/* Shutdown root pd rproc */
	rproc_shutdown(upd_dev->parent);
}
After userpd rproc stops, root pd rproc will be stopped.
IPQ5018, IPQ9574 supports multipd remoteproc driver.

DTS patch [13/13] is based on below series
https://lore.kernel.org/linux-arm-msm/20230519090219.15925-1-quic_devipriy@quicinc.com/
	
[V2]
	- This patch is depends on the below series
https://lore.kernel.org/linux-arm-msm/20230519090219.15925-1-quic_devipriy@quicinc.com/
https://lore.kernel.org/linux-arm-msm/20230519125409.497439-1-quic_srichara@quicinc.com/
	- Fixed all comments and rebased for TOT.
	- Strip off SCM code changes from https://lore.kernel.org/linux-arm-msm/1678164097-13247-9-git-send-email-quic_mmanikan@quicinc.com/
	  and made new patchset [08/13]
	- Strip off qcom_q6v5 changes from https://lore.kernel.org/linux-arm-msm/1678164097-13247-9-git-send-email-quic_mmanikan@quicinc.com/
	  and made new patchset [09/13]
	- since clocks handled by QDSP6 firmware
	  Dropped https://lore.kernel.org/linux-arm-msm/1678164097-13247-6-git-send-email-quic_mmanikan@quicinc.com/,
	  https://lore.kernel.org/linux-arm-msm/1678164097-13247-7-git-send-email-quic_mmanikan@quicinc.com/.
 	  Added [04/13], [05/13], [06/13], [07/13] patches.

Manikanta Mylavarapu (13):
  dt-bindings: remoteproc: qcom: Add support for multipd model
  dt-bindings: mailbox: qcom: Add IPQ5018 APCS compatible
  dt-bindings: arm: qcom: Document the Qualcomm rdp432-c1 board
  dt-bindings: clock: qcom: gcc-ipq5018: remove q6 clocks macros
  dt-bindings: clock: qcom: gcc-ipq9574: remove q6 bring up clock macros
  clk: qcom: ipq5018: remove q6 bring up clocks
  clk: qcom: ipq9574: remove q6 bring up clocks
  firmware: qcom_scm: ipq5018: Add WCSS AHB pd support
  remoteproc: qcom: q6v5: Add multipd interrupts support
  remoteproc: qcom: Add Hexagon based multipd rproc driver
  arm64: dtsi: qcom: ipq5018: enable nodes required for multipd
  arm64: dts: qcom: ipq5018: Add RDP432-c1 board support
  arm64: dtsi: qcom: ipq9574: Add nodes to bring up multipd

 .../devicetree/bindings/arm/qcom.yaml         |   1 +
 .../mailbox/qcom,apcs-kpss-global.yaml        |   1 +
 .../bindings/remoteproc/qcom,multipd-pil.yaml | 265 +++++++
 arch/arm64/boot/dts/qcom/Makefile             |   1 +
 .../arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts |  49 ++
 arch/arm64/boot/dts/qcom/ipq5018.dtsi         | 140 ++++
 arch/arm64/boot/dts/qcom/ipq9574.dtsi         | 118 +++
 drivers/clk/qcom/gcc-ipq5018.c                | 414 -----------
 drivers/clk/qcom/gcc-ipq9574.c                | 326 ---------
 drivers/firmware/qcom_scm.c                   | 114 +++
 drivers/firmware/qcom_scm.h                   |   6 +
 drivers/remoteproc/Kconfig                    |  20 +
 drivers/remoteproc/Makefile                   |   1 +
 drivers/remoteproc/qcom_common.h              |   9 +
 drivers/remoteproc/qcom_q6v5.c                |  37 +-
 drivers/remoteproc/qcom_q6v5.h                |  11 +
 drivers/remoteproc/qcom_q6v5_mpd.c            | 677 ++++++++++++++++++
 drivers/soc/qcom/mdt_loader.c                 | 332 +++++++++
 include/dt-bindings/clock/qcom,gcc-ipq5018.h  |  21 -
 include/dt-bindings/clock/qcom,ipq9574-gcc.h  |  18 -
 include/linux/firmware/qcom/qcom_scm.h        |   3 +
 include/linux/soc/qcom/mdt_loader.h           |  19 +
 22 files changed, 1801 insertions(+), 782 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
 create mode 100644 arch/arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts
 create mode 100644 drivers/remoteproc/qcom_q6v5_mpd.c


base-commit: dbd91ef4e91c1ce3a24429f5fb3876b7a0306733
--
2.17.1

[PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Manikanta Mylavarapu]

Add new binding document for multipd model remoteproc.
IPQ5018, IPQ9574 follows multipd model.

Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
---
Changes in V2:
	- Fixed all comments and rebased for TOT.
	- Changed to BSD-3-Clause based on internal open source team
          suggestion.
	- Added firmware-name.

 .../bindings/remoteproc/qcom,multipd-pil.yaml | 265 ++++++++++++++++++
 1 file changed, 265 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml

diff --git a/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml b/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
new file mode 100644
index 000000000000..3257f27dc569
--- /dev/null
+++ b/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
@@ -0,0 +1,265 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/remoteproc/qcom,multipd-pil.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm Multipd Secure Peripheral Image Loader
+
+maintainers:
+  - Bjorn Andersson <andersson@kernel.org>
+  - Mathieu Poirier <mathieu.poirier@linaro.org>
+
+description:
+  Multipd Peripheral Image Loader loads firmware and boots Q6 pd, WCSS pd
+  remoteproc's on the Qualcomm IPQ5018, IPQ9574 SoC. Pd means protection
+  domain. It's similar to process in Linux. Here QDSP6 processor runs each
+  wifi radio functionality on a separate process. One process can't access
+  other process resources, so this is termed as PD i.e protection domain.
+
+properties:
+  compatible:
+    enum:
+      - qcom,ipq5018-q6-mpd
+      - qcom,ipq9574-q6-mpd
+
+  reg:
+    maxItems: 1
+
+  firmware-name:
+    $ref: /schemas/types.yaml#/definitions/string-array
+    description: Firmware name of the Hexagon core
+
+  interrupts-extended:
+    items:
+      - description: Watchdog interrupt
+      - description: Fatal interrupt
+      - description: Ready interrupt
+      - description: Handover interrupt
+      - description: Stop acknowledge interrupt
+
+  interrupt-names:
+    items:
+      - const: wdog
+      - const: fatal
+      - const: ready
+      - const: handover
+      - const: stop-ack
+
+  qcom,smem-states:
+    $ref: /schemas/types.yaml#/definitions/phandle-array
+    description: States used by the AP to signal the remote processor
+    items:
+      - description: Shutdown Q6
+      - description: Stop Q6
+
+  qcom,smem-state-names:
+    description:
+      Names of the states used by the AP to signal the remote processor
+    items:
+      - const: shutdown
+      - const: stop
+
+  memory-region:
+    items:
+      - description: Q6 pd reserved region
+
+  glink-edge:
+    $ref: /schemas/remoteproc/qcom,glink-edge.yaml#
+    description:
+      Qualcomm G-Link subnode which represents communication edge, channels
+      and devices related to the Modem.
+
+patternProperties:
+  "^pd-1|pd-2|pd-3":
+    type: object
+    description:
+      In Multipd model, WCSS pd depends on Q6 pd i.e Q6 pd should be up before
+      WCSS. It can be achieved by keeping wcss pd node as subnode of Q6
+      device node.
+
+    properties:
+      compatible:
+        enum:
+          - qcom,ipq5018-wcss-ahb-mpd
+          - qcom,ipq9574-wcss-ahb-mpd
+          - qcom,ipq5018-wcss-pcie-mpd
+
+      firmware-name:
+        $ref: /schemas/types.yaml#/definitions/string-array
+        items:
+          - description: Firmware name of the Hexagon core
+
+      interrupts-extended:
+        items:
+          - description: Fatal interrupt
+          - description: Ready interrupt
+          - description: Spawn acknowledge interrupt
+          - description: Stop acknowledge interrupt
+
+      interrupt-names:
+        items:
+          - const: fatal
+          - const: ready
+          - const: spawn-ack
+          - const: stop-ack
+
+      qcom,smem-states:
+        $ref: /schemas/types.yaml#/definitions/phandle-array
+        description: States used by the AP to signal the remote processor
+        items:
+          - description: Shutdown WCSS pd
+          - description: Stop WCSS pd
+          - description: Spawn WCSS pd
+
+      qcom,smem-state-names:
+        description:
+          Names of the states used by the AP to signal the remote processor
+        items:
+          - const: shutdown
+          - const: stop
+          - const: spawn
+
+    required:
+      - compatible
+      - firmware-name
+      - interrupts-extended
+      - interrupt-names
+      - qcom,smem-states
+      - qcom,smem-state-names
+
+    unevaluatedProperties: false
+
+required:
+  - compatible
+  - firmware-name
+  - reg
+  - interrupts-extended
+  - interrupt-names
+  - qcom,smem-states
+  - qcom,smem-state-names
+  - memory-region
+
+allOf:
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - qcom,ipq5018-q6-mpd
+    then:
+      properties:
+        firmware-name:
+          items:
+            - const: IPQ5018/q6_fw.mdt
+            - const: IPQ5018/m3_fw.mdt
+            - const: qcn6122/m3_fw.mdt
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - qcom,ipq9574-q6-mpd
+    then:
+      properties:
+        firmware-name:
+          items:
+            - const: IPQ9574/q6_fw.mdt
+            - const: IPQ9574/m3_fw.mdt
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    q6v5_wcss: remoteproc@cd00000 {
+      compatible = "qcom,ipq5018-q6-mpd";
+      reg = <0x0cd00000 0x4040>;
+      firmware-name = "IPQ5018/q6_fw.mdt",
+                      "IPQ5018/m3_fw.mdt",
+                      "qcn6122/m3_fw.mdt";
+      interrupts-extended = <&intc GIC_SPI 291 IRQ_TYPE_EDGE_RISING>,
+                            <&wcss_smp2p_in 0 0>,
+                            <&wcss_smp2p_in 1 0>,
+                            <&wcss_smp2p_in 2 0>,
+                            <&wcss_smp2p_in 3 0>;
+      interrupt-names = "wdog",
+                        "fatal",
+                        "ready",
+                        "handover",
+                        "stop-ack";
+
+      qcom,smem-states = <&wcss_smp2p_out 0>,
+                         <&wcss_smp2p_out 1>;
+      qcom,smem-state-names = "shutdown",
+                              "stop";
+
+      memory-region = <&q6_region>;
+
+      glink-edge {
+        interrupts = <GIC_SPI 179 IRQ_TYPE_EDGE_RISING>;
+        label = "rtr";
+        qcom,remote-pid = <1>;
+        mboxes = <&apcs_glb 8>;
+      };
+
+      pd-1 {
+        compatible = "qcom,ipq5018-wcss-ahb-mpd";
+        firmware-name = "IPQ5018/q6_fw.mdt";
+        interrupts-extended = <&wcss_smp2p_in 8 0>,
+                              <&wcss_smp2p_in 9 0>,
+                              <&wcss_smp2p_in 12 0>,
+                              <&wcss_smp2p_in 11 0>;
+        interrupt-names = "fatal",
+                          "ready",
+                          "spawn-ack",
+                          "stop-ack";
+        qcom,smem-states = <&wcss_smp2p_out 8>,
+                           <&wcss_smp2p_out 9>,
+                           <&wcss_smp2p_out 10>;
+        qcom,smem-state-names = "shutdown",
+                                "stop",
+                                "spawn";
+      };
+
+      pd-2 {
+        compatible = "qcom,ipq5018-wcss-pcie-mpd";
+        firmware-name = "IPQ5018/q6_fw.mdt";
+        interrupts-extended = <&wcss_smp2p_in 16 0>,
+                              <&wcss_smp2p_in 17 0>,
+                              <&wcss_smp2p_in 20 0>,
+                              <&wcss_smp2p_in 19 0>;
+        interrupt-names = "fatal",
+                          "ready",
+                          "spawn-ack",
+                          "stop-ack";
+
+        qcom,smem-states = <&wcss_smp2p_out 16>,
+                           <&wcss_smp2p_out 17>,
+                           <&wcss_smp2p_out 18>;
+        qcom,smem-state-names = "shutdown",
+                                "stop",
+                                "spawn";
+      };
+
+      pd-3 {
+        compatible = "qcom,ipq5018-wcss-pcie-mpd";
+        firmware-name = "IPQ5018/q6_fw.mdt";
+        interrupts-extended = <&wcss_smp2p_in 24 0>,
+                              <&wcss_smp2p_in 25 0>,
+                              <&wcss_smp2p_in 28 0>,
+                              <&wcss_smp2p_in 27 0>;
+        interrupt-names = "fatal",
+                          "ready",
+                          "spawn-ack",
+                          "stop-ack";
+
+        qcom,smem-states = <&wcss_smp2p_out 24>,
+                           <&wcss_smp2p_out 25>,
+                           <&wcss_smp2p_out 26>;
+        qcom,smem-state-names = "shutdown",
+                                "stop",
+                                "spawn";
+       };
+    };
--
2.17.1

[PATCH V2 02/13] dt-bindings: mailbox: qcom: Add IPQ5018 APCS compatible

[Manikanta Mylavarapu]

Add compatible for the Qualcomm IPQ5018 APCS block.

Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
---
Changes in V2:
	- As suggested by Kathirvan & Krzysztof, used fallback
          mechanism similar to IPQ5332

 .../devicetree/bindings/mailbox/qcom,apcs-kpss-global.yaml       | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/mailbox/qcom,apcs-kpss-global.yaml b/Documentation/devicetree/bindings/mailbox/qcom,apcs-kpss-global.yaml
index 32d7bbc98cac..d2e25ff6db7f 100644
--- a/Documentation/devicetree/bindings/mailbox/qcom,apcs-kpss-global.yaml
+++ b/Documentation/devicetree/bindings/mailbox/qcom,apcs-kpss-global.yaml
@@ -18,6 +18,7 @@ properties:
     oneOf:
       - items:
           - enum:
+              - qcom,ipq5018-apcs-apps-global
               - qcom,ipq5332-apcs-apps-global
               - qcom,ipq8074-apcs-apps-global
               - qcom,ipq9574-apcs-apps-global
--
2.17.1

[PATCH V2 03/13] dt-bindings: arm: qcom: Document the Qualcomm rdp432-c1 board

[Manikanta Mylavarapu]

Document the Qualcomm rdp432-c1 board based on IPQ5018 Soc.

Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
---
Changes in V2:
	- Renamed mp03.5-c1 to rdp432-c1

 Documentation/devicetree/bindings/arm/qcom.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/arm/qcom.yaml b/Documentation/devicetree/bindings/arm/qcom.yaml
index 33a9a97d7b5b..53ae2b6b2da6 100644
--- a/Documentation/devicetree/bindings/arm/qcom.yaml
+++ b/Documentation/devicetree/bindings/arm/qcom.yaml
@@ -337,6 +337,7 @@ properties:
       - items:
           - enum:
               - qcom,ipq5018-rdp432-c2
+              - qcom,ipq5018-rdp432-c1
           - const: qcom,ipq5018

       - items:
--
2.17.1

[PATCH V2 04/13] dt-bindings: clock: qcom: gcc-ipq5018: remove q6 clocks macros

[Manikanta Mylavarapu]

Since Q6 firmware takes care of bring up clocks in multipd
model, remove bring up clock macros.

Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
---
 include/dt-bindings/clock/qcom,gcc-ipq5018.h | 21 --------------------
 1 file changed, 21 deletions(-)

diff --git a/include/dt-bindings/clock/qcom,gcc-ipq5018.h b/include/dt-bindings/clock/qcom,gcc-ipq5018.h
index f3de2fdfeea1..89d81fa1a758 100644
--- a/include/dt-bindings/clock/qcom,gcc-ipq5018.h
+++ b/include/dt-bindings/clock/qcom,gcc-ipq5018.h
@@ -74,15 +74,8 @@
 #define GCC_PCIE1_AXI_S_CLK				65
 #define GCC_PCIE1_PIPE_CLK				66
 #define GCC_PRNG_AHB_CLK				67
-#define GCC_Q6_AXIM_CLK					68
-#define GCC_Q6_AXIM2_CLK				69
-#define GCC_Q6_AXIS_CLK					70
-#define GCC_Q6_AHB_CLK					71
-#define GCC_Q6_AHB_S_CLK				72
 #define GCC_Q6_TSCTR_1TO2_CLK				73
 #define GCC_Q6SS_ATBM_CLK				74
-#define GCC_Q6SS_PCLKDBG_CLK				75
-#define GCC_Q6SS_TRIG_CLK				76
 #define GCC_QDSS_AT_CLK					77
 #define GCC_QDSS_CFG_AHB_CLK				78
 #define GCC_QDSS_DAP_AHB_CLK				79
@@ -109,7 +102,6 @@
 #define GCC_SYS_NOC_PCIE1_AXI_CLK			100
 #define GCC_SYS_NOC_QDSS_STM_AXI_CLK			101
 #define GCC_SYS_NOC_USB0_AXI_CLK			102
-#define GCC_SYS_NOC_WCSS_AHB_CLK			103
 #define GCC_UBI0_AXI_CLK				104
 #define GCC_UBI0_CFG_CLK				105
 #define GCC_UBI0_CORE_CLK				106
@@ -127,19 +119,6 @@
 #define GCC_USB0_MOCK_UTMI_CLK				118
 #define GCC_USB0_PHY_CFG_AHB_CLK			119
 #define GCC_USB0_SLEEP_CLK				120
-#define GCC_WCSS_ACMT_CLK				121
-#define GCC_WCSS_AHB_S_CLK				122
-#define GCC_WCSS_AXI_M_CLK				123
-#define GCC_WCSS_AXI_S_CLK				124
-#define GCC_WCSS_DBG_IFC_APB_BDG_CLK			125
-#define GCC_WCSS_DBG_IFC_APB_CLK			126
-#define GCC_WCSS_DBG_IFC_ATB_BDG_CLK			127
-#define GCC_WCSS_DBG_IFC_ATB_CLK			128
-#define GCC_WCSS_DBG_IFC_DAPBUS_BDG_CLK			129
-#define GCC_WCSS_DBG_IFC_DAPBUS_CLK			130
-#define GCC_WCSS_DBG_IFC_NTS_BDG_CLK			131
-#define GCC_WCSS_DBG_IFC_NTS_CLK			132
-#define GCC_WCSS_ECAHB_CLK				133
 #define GCC_XO_CLK					134
 #define GCC_XO_CLK_SRC					135
 #define GMAC0_RX_CLK_SRC				136
--
2.17.1

[PATCH V2 05/13] dt-bindings: clock: qcom: gcc-ipq9574: remove q6 bring up clock macros

[Manikanta Mylavarapu]

In multipd model Q6 firmware takes care of bringup clocks,
so remove them.

Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
---
 include/dt-bindings/clock/qcom,ipq9574-gcc.h | 18 ------------------
 1 file changed, 18 deletions(-)

diff --git a/include/dt-bindings/clock/qcom,ipq9574-gcc.h b/include/dt-bindings/clock/qcom,ipq9574-gcc.h
index 8bf26d62c54f..d97a8a68fd33 100644
--- a/include/dt-bindings/clock/qcom,ipq9574-gcc.h
+++ b/include/dt-bindings/clock/qcom,ipq9574-gcc.h
@@ -132,16 +132,8 @@
 #define GCC_NSSNOC_SNOC_1_CLK				123
 #define GCC_QDSS_ETR_USB_CLK				124
 #define WCSS_AHB_CLK_SRC				125
-#define GCC_Q6_AHB_CLK					126
-#define GCC_Q6_AHB_S_CLK				127
-#define GCC_WCSS_ECAHB_CLK				128
-#define GCC_WCSS_ACMT_CLK				129
-#define GCC_SYS_NOC_WCSS_AHB_CLK			130
 #define WCSS_AXI_M_CLK_SRC				131
-#define GCC_ANOC_WCSS_AXI_M_CLK				132
 #define QDSS_AT_CLK_SRC					133
-#define GCC_Q6SS_ATBM_CLK				134
-#define GCC_WCSS_DBG_IFC_ATB_CLK			135
 #define GCC_NSSNOC_ATB_CLK				136
 #define GCC_QDSS_AT_CLK					137
 #define GCC_SYS_NOC_AT_CLK				138
@@ -154,27 +146,18 @@
 #define QDSS_TRACECLKIN_CLK_SRC				145
 #define GCC_QDSS_TRACECLKIN_CLK				146
 #define QDSS_TSCTR_CLK_SRC				147
-#define GCC_Q6_TSCTR_1TO2_CLK				148
-#define GCC_WCSS_DBG_IFC_NTS_CLK			149
 #define GCC_QDSS_TSCTR_DIV2_CLK				150
 #define GCC_QDSS_TS_CLK					151
 #define GCC_QDSS_TSCTR_DIV4_CLK				152
 #define GCC_NSS_TS_CLK					153
 #define GCC_QDSS_TSCTR_DIV8_CLK				154
 #define GCC_QDSS_TSCTR_DIV16_CLK			155
-#define GCC_Q6SS_PCLKDBG_CLK				156
-#define GCC_Q6SS_TRIG_CLK				157
-#define GCC_WCSS_DBG_IFC_APB_CLK			158
-#define GCC_WCSS_DBG_IFC_DAPBUS_CLK			159
 #define GCC_QDSS_DAP_CLK				160
 #define GCC_QDSS_APB2JTAG_CLK				161
 #define GCC_QDSS_TSCTR_DIV3_CLK				162
 #define QPIC_IO_MACRO_CLK_SRC				163
 #define GCC_QPIC_IO_MACRO_CLK                           164
 #define Q6_AXI_CLK_SRC					165
-#define GCC_Q6_AXIM_CLK					166
-#define GCC_WCSS_Q6_TBU_CLK				167
-#define GCC_MEM_NOC_Q6_AXI_CLK				168
 #define Q6_AXIM2_CLK_SRC				169
 #define NSSNOC_MEMNOC_BFDCD_CLK_SRC			170
 #define GCC_NSSNOC_MEMNOC_CLK				171
@@ -199,7 +182,6 @@
 #define GCC_UNIPHY2_SYS_CLK				190
 #define GCC_CMN_12GPLL_SYS_CLK				191
 #define GCC_NSSNOC_XO_DCD_CLK				192
-#define GCC_Q6SS_BOOT_CLK				193
 #define UNIPHY_SYS_CLK_SRC				194
 #define NSS_TS_CLK_SRC					195
 #define GCC_ANOC_PCIE0_1LANE_M_CLK			196
--
2.17.1

[PATCH V2 06/13] clk: qcom: ipq5018: remove q6 bring up clocks

[Manikanta Mylavarapu]

Since Q6 firmware takes care of it's bring up clocks
in multipd model, remove from gcc driver.

Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
---
 drivers/clk/qcom/gcc-ipq5018.c | 414 ---------------------------------
 1 file changed, 414 deletions(-)

diff --git a/drivers/clk/qcom/gcc-ipq5018.c b/drivers/clk/qcom/gcc-ipq5018.c
index 7ac93b1d6b3f..0dab7d94a3ae 100644
--- a/drivers/clk/qcom/gcc-ipq5018.c
+++ b/drivers/clk/qcom/gcc-ipq5018.c
@@ -2225,159 +2225,6 @@ static struct clk_branch gcc_prng_ahb_clk = {
 	},
 };

-static struct clk_branch gcc_q6_ahb_clk = {
-	.halt_reg = 0x59138,
-	.clkr = {
-		.enable_reg = 0x59138,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_q6_ahb_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&wcss_ahb_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_q6_ahb_s_clk = {
-	.halt_reg = 0x5914C,
-	.clkr = {
-		.enable_reg = 0x5914C,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_q6_ahb_s_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&wcss_ahb_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_q6_axim_clk = {
-	.halt_reg = 0x5913C,
-	.clkr = {
-		.enable_reg = 0x5913C,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_q6_axim_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&q6_axi_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_q6_axim2_clk = {
-	.halt_reg = 0x59150,
-	.clkr = {
-		.enable_reg = 0x59150,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_q6_axim2_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&q6_axi_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_q6_axis_clk = {
-	.halt_reg = 0x59154,
-	.clkr = {
-		.enable_reg = 0x59154,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_q6_axis_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&system_noc_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_q6_tsctr_1to2_clk = {
-	.halt_reg = 0x59148,
-	.clkr = {
-		.enable_reg = 0x59148,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_q6_tsctr_1to2_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_tsctr_div2_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_q6ss_atbm_clk = {
-	.halt_reg = 0x59144,
-	.clkr = {
-		.enable_reg = 0x59144,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_q6ss_atbm_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_at_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_q6ss_pclkdbg_clk = {
-	.halt_reg = 0x59140,
-	.clkr = {
-		.enable_reg = 0x59140,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_q6ss_pclkdbg_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_dap_sync_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_q6ss_trig_clk = {
-	.halt_reg = 0x59128,
-	.clkr = {
-		.enable_reg = 0x59128,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_q6ss_trig_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_dap_sync_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
 static struct clk_branch gcc_qdss_at_clk = {
 	.halt_reg = 0x29024,
 	.clkr = {
@@ -2803,23 +2650,6 @@ static struct clk_branch gcc_sys_noc_usb0_axi_clk = {
 	},
 };

-static struct clk_branch gcc_sys_noc_wcss_ahb_clk = {
-	.halt_reg = 0x26034,
-	.clkr = {
-		.enable_reg = 0x26034,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_sys_noc_wcss_ahb_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&wcss_ahb_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
 static struct clk_branch gcc_ubi0_axi_clk = {
 	.halt_reg = 0x68200,
 	.halt_check = BRANCH_HALT_DELAY,
@@ -3134,227 +2964,6 @@ static struct clk_branch gcc_usb0_pipe_clk = {
 	},
 };

-static struct clk_branch gcc_wcss_acmt_clk = {
-	.halt_reg = 0x59064,
-	.clkr = {
-		.enable_reg = 0x59064,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_wcss_acmt_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&wcss_ahb_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_ahb_s_clk = {
-	.halt_reg = 0x59034,
-	.clkr = {
-		.enable_reg = 0x59034,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_wcss_ahb_s_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&wcss_ahb_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_axi_m_clk = {
-	.halt_reg = 0x5903C,
-	.clkr = {
-		.enable_reg = 0x5903C,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_wcss_axi_m_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&system_noc_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_axi_s_clk = {
-	.halt_reg = 0x59068,
-	.clkr = {
-		.enable_reg = 0x59068,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_wi_s_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&system_noc_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_dbg_ifc_apb_bdg_clk = {
-	.halt_reg = 0x59050,
-	.clkr = {
-		.enable_reg = 0x59050,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_wcss_dbg_ifc_apb_bdg_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_dap_sync_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_dbg_ifc_apb_clk = {
-	.halt_reg = 0x59040,
-	.clkr = {
-		.enable_reg = 0x59040,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_wcss_dbg_ifc_apb_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_dap_sync_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_dbg_ifc_atb_bdg_clk = {
-	.halt_reg = 0x59054,
-	.clkr = {
-		.enable_reg = 0x59054,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_wcss_dbg_ifc_atb_bdg_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_at_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_dbg_ifc_atb_clk = {
-	.halt_reg = 0x59044,
-	.clkr = {
-		.enable_reg = 0x59044,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_wcss_dbg_ifc_atb_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_at_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_dbg_ifc_dapbus_bdg_clk = {
-	.halt_reg = 0x59060,
-	.clkr = {
-		.enable_reg = 0x59060,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_wcss_dbg_ifc_dapbus_bdg_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_dap_sync_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_dbg_ifc_dapbus_clk = {
-	.halt_reg = 0x5905C,
-	.clkr = {
-		.enable_reg = 0x5905C,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_wcss_dbg_ifc_dapbus_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_dap_sync_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_dbg_ifc_nts_bdg_clk = {
-	.halt_reg = 0x59058,
-	.clkr = {
-		.enable_reg = 0x59058,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_wcss_dbg_ifc_nts_bdg_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_tsctr_div2_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_dbg_ifc_nts_clk = {
-	.halt_reg = 0x59048,
-	.clkr = {
-		.enable_reg = 0x59048,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_wcss_dbg_ifc_nts_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_tsctr_div2_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_ecahb_clk = {
-	.halt_reg = 0x59038,
-	.clkr = {
-		.enable_reg = 0x59038,
-		.enable_mask = BIT(0),
-		.hw.init = &(struct clk_init_data) {
-			.name = "gcc_wcss_ecahb_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&wcss_ahb_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
 static struct clk_hw *gcc_ipq5018_hws[] = {
 	&gpll0_out_main_div2.hw,
 	&pcnoc_clk_src.hw,
@@ -3444,15 +3053,6 @@ static struct clk_regmap *gcc_ipq5018_clks[] = {
 	[GCC_PCIE1_AXI_S_BRIDGE_CLK] = &gcc_pcie1_axi_s_bridge_clk.clkr,
 	[GCC_PCIE1_AXI_S_CLK] = &gcc_pcie1_axi_s_clk.clkr,
 	[GCC_PRNG_AHB_CLK] = &gcc_prng_ahb_clk.clkr,
-	[GCC_Q6_AXIM_CLK] = &gcc_q6_axim_clk.clkr,
-	[GCC_Q6_AXIM2_CLK] = &gcc_q6_axim2_clk.clkr,
-	[GCC_Q6_AXIS_CLK] = &gcc_q6_axis_clk.clkr,
-	[GCC_Q6_AHB_CLK] = &gcc_q6_ahb_clk.clkr,
-	[GCC_Q6_AHB_S_CLK] = &gcc_q6_ahb_s_clk.clkr,
-	[GCC_Q6_TSCTR_1TO2_CLK] = &gcc_q6_tsctr_1to2_clk.clkr,
-	[GCC_Q6SS_ATBM_CLK] = &gcc_q6ss_atbm_clk.clkr,
-	[GCC_Q6SS_PCLKDBG_CLK] = &gcc_q6ss_pclkdbg_clk.clkr,
-	[GCC_Q6SS_TRIG_CLK] = &gcc_q6ss_trig_clk.clkr,
 	[GCC_QDSS_AT_CLK] = &gcc_qdss_at_clk.clkr,
 	[GCC_QDSS_CFG_AHB_CLK] = &gcc_qdss_cfg_ahb_clk.clkr,
 	[GCC_QDSS_DAP_AHB_CLK] = &gcc_qdss_dap_ahb_clk.clkr,
@@ -3479,7 +3079,6 @@ static struct clk_regmap *gcc_ipq5018_clks[] = {
 	[GCC_SYS_NOC_PCIE1_AXI_CLK] = &gcc_sys_noc_pcie1_axi_clk.clkr,
 	[GCC_SYS_NOC_QDSS_STM_AXI_CLK] = &gcc_sys_noc_qdss_stm_axi_clk.clkr,
 	[GCC_SYS_NOC_USB0_AXI_CLK] = &gcc_sys_noc_usb0_axi_clk.clkr,
-	[GCC_SYS_NOC_WCSS_AHB_CLK] = &gcc_sys_noc_wcss_ahb_clk.clkr,
 	[GCC_UBI0_AXI_CLK] = &gcc_ubi0_axi_clk.clkr,
 	[GCC_UBI0_CFG_CLK] = &gcc_ubi0_cfg_clk.clkr,
 	[GCC_UBI0_CORE_CLK] = &gcc_ubi0_core_clk.clkr,
@@ -3497,19 +3096,6 @@ static struct clk_regmap *gcc_ipq5018_clks[] = {
 	[GCC_USB0_MOCK_UTMI_CLK] = &gcc_usb0_mock_utmi_clk.clkr,
 	[GCC_USB0_PHY_CFG_AHB_CLK] = &gcc_usb0_phy_cfg_ahb_clk.clkr,
 	[GCC_USB0_SLEEP_CLK] = &gcc_usb0_sleep_clk.clkr,
-	[GCC_WCSS_ACMT_CLK] = &gcc_wcss_acmt_clk.clkr,
-	[GCC_WCSS_AHB_S_CLK] = &gcc_wcss_ahb_s_clk.clkr,
-	[GCC_WCSS_AXI_M_CLK] = &gcc_wcss_axi_m_clk.clkr,
-	[GCC_WCSS_AXI_S_CLK] = &gcc_wcss_axi_s_clk.clkr,
-	[GCC_WCSS_DBG_IFC_APB_BDG_CLK] = &gcc_wcss_dbg_ifc_apb_bdg_clk.clkr,
-	[GCC_WCSS_DBG_IFC_APB_CLK] = &gcc_wcss_dbg_ifc_apb_clk.clkr,
-	[GCC_WCSS_DBG_IFC_ATB_BDG_CLK] = &gcc_wcss_dbg_ifc_atb_bdg_clk.clkr,
-	[GCC_WCSS_DBG_IFC_ATB_CLK] = &gcc_wcss_dbg_ifc_atb_clk.clkr,
-	[GCC_WCSS_DBG_IFC_DAPBUS_BDG_CLK] = &gcc_wcss_dbg_ifc_dapbus_bdg_clk.clkr,
-	[GCC_WCSS_DBG_IFC_DAPBUS_CLK] = &gcc_wcss_dbg_ifc_dapbus_clk.clkr,
-	[GCC_WCSS_DBG_IFC_NTS_BDG_CLK] = &gcc_wcss_dbg_ifc_nts_bdg_clk.clkr,
-	[GCC_WCSS_DBG_IFC_NTS_CLK] = &gcc_wcss_dbg_ifc_nts_clk.clkr,
-	[GCC_WCSS_ECAHB_CLK] = &gcc_wcss_ecahb_clk.clkr,
 	[GCC_XO_CLK] = &gcc_xo_clk.clkr,
 	[GCC_XO_CLK_SRC] = &gcc_xo_clk_src.clkr,
 	[GMAC0_RX_CLK_SRC] = &gmac0_rx_clk_src.clkr,
--
2.17.1

[PATCH V2 07/13] clk: qcom: ipq9574: remove q6 bring up clocks

[Manikanta Mylavarapu]

In multipd model Q6 firmware takes care of bringup clocks,
so remove them from gcc driver.

Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
---
 drivers/clk/qcom/gcc-ipq9574.c | 326 ---------------------------------
 1 file changed, 326 deletions(-)

diff --git a/drivers/clk/qcom/gcc-ipq9574.c b/drivers/clk/qcom/gcc-ipq9574.c
index 1245ac52a17c..e4b0304dd43e 100644
--- a/drivers/clk/qcom/gcc-ipq9574.c
+++ b/drivers/clk/qcom/gcc-ipq9574.c
@@ -2608,24 +2608,6 @@ static struct clk_rcg2 system_noc_bfdcd_clk_src = {
 	},
 };

-static struct clk_branch gcc_q6ss_boot_clk = {
-	.halt_reg = 0x25080,
-	.halt_check = BRANCH_HALT_SKIP,
-	.clkr = {
-		.enable_reg = 0x25080,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_q6ss_boot_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&system_noc_bfdcd_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
 static struct clk_branch gcc_nssnoc_snoc_clk = {
 	.halt_reg = 0x17028,
 	.clkr = {
@@ -2696,91 +2678,6 @@ static struct clk_rcg2 wcss_ahb_clk_src = {
 	},
 };

-static struct clk_branch gcc_q6_ahb_clk = {
-	.halt_reg = 0x25014,
-	.clkr = {
-		.enable_reg = 0x25014,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_q6_ahb_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&wcss_ahb_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_q6_ahb_s_clk = {
-	.halt_reg = 0x25018,
-	.clkr = {
-		.enable_reg = 0x25018,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_q6_ahb_s_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&wcss_ahb_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_ecahb_clk = {
-	.halt_reg = 0x25058,
-	.clkr = {
-		.enable_reg = 0x25058,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_wcss_ecahb_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&wcss_ahb_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_acmt_clk = {
-	.halt_reg = 0x2505c,
-	.clkr = {
-		.enable_reg = 0x2505c,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_wcss_acmt_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&wcss_ahb_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_sys_noc_wcss_ahb_clk = {
-	.halt_reg = 0x2e030,
-	.clkr = {
-		.enable_reg = 0x2e030,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_sys_noc_wcss_ahb_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&wcss_ahb_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
 static const struct freq_tbl ftbl_wcss_axi_m_clk_src[] = {
 	F(24000000, P_XO, 1, 0, 0),
 	F(133333333, P_GPLL0, 6, 0, 0),
@@ -2801,23 +2698,6 @@ static struct clk_rcg2 wcss_axi_m_clk_src = {
 	},
 };

-static struct clk_branch gcc_anoc_wcss_axi_m_clk = {
-	.halt_reg = 0x2e0a8,
-	.clkr = {
-		.enable_reg = 0x2e0a8,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_anoc_wcss_axi_m_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&wcss_axi_m_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
 static const struct freq_tbl ftbl_qdss_at_clk_src[] = {
 	F(240000000, P_GPLL4, 5, 0, 0),
 	{ }
@@ -2836,40 +2716,6 @@ static struct clk_rcg2 qdss_at_clk_src = {
 	},
 };

-static struct clk_branch gcc_q6ss_atbm_clk = {
-	.halt_reg = 0x2501c,
-	.clkr = {
-		.enable_reg = 0x2501c,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_q6ss_atbm_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_at_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_dbg_ifc_atb_clk = {
-	.halt_reg = 0x2503c,
-	.clkr = {
-		.enable_reg = 0x2503c,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_wcss_dbg_ifc_atb_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_at_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
 static struct clk_branch gcc_nssnoc_atb_clk = {
 	.halt_reg = 0x17014,
 	.clkr = {
@@ -3106,40 +2952,6 @@ static struct clk_fixed_factor qdss_tsctr_div2_clk_src = {
 	},
 };

-static struct clk_branch gcc_q6_tsctr_1to2_clk = {
-	.halt_reg = 0x25020,
-	.clkr = {
-		.enable_reg = 0x25020,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_q6_tsctr_1to2_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_tsctr_div2_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_dbg_ifc_nts_clk = {
-	.halt_reg = 0x25040,
-	.clkr = {
-		.enable_reg = 0x25040,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_wcss_dbg_ifc_nts_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_tsctr_div2_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
 static struct clk_branch gcc_qdss_tsctr_div2_clk = {
 	.halt_reg = 0x2d044,
 	.clkr = {
@@ -3314,74 +3126,6 @@ static struct clk_branch gcc_qdss_tsctr_div16_clk = {
 	},
 };

-static struct clk_branch gcc_q6ss_pclkdbg_clk = {
-	.halt_reg = 0x25024,
-	.clkr = {
-		.enable_reg = 0x25024,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_q6ss_pclkdbg_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_dap_sync_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_q6ss_trig_clk = {
-	.halt_reg = 0x25068,
-	.clkr = {
-		.enable_reg = 0x25068,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_q6ss_trig_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_dap_sync_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_dbg_ifc_apb_clk = {
-	.halt_reg = 0x25038,
-	.clkr = {
-		.enable_reg = 0x25038,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_wcss_dbg_ifc_apb_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_dap_sync_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_dbg_ifc_dapbus_clk = {
-	.halt_reg = 0x25044,
-	.clkr = {
-		.enable_reg = 0x25044,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_wcss_dbg_ifc_dapbus_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&qdss_dap_sync_clk_src.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
 static struct clk_branch gcc_qdss_dap_clk = {
 	.halt_reg = 0x2d058,
 	.clkr = {
@@ -3503,58 +3247,6 @@ static struct clk_rcg2 q6_axi_clk_src = {
 	},
 };

-static struct clk_branch gcc_q6_axim_clk = {
-	.halt_reg = 0x2500c,
-	.clkr = {
-		.enable_reg = 0x2500c,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_q6_axim_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&q6_axi_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_wcss_q6_tbu_clk = {
-	.halt_reg = 0x12050,
-	.halt_check = BRANCH_HALT_DELAY,
-	.clkr = {
-		.enable_reg = 0xb00c,
-		.enable_mask = BIT(6),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_wcss_q6_tbu_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&q6_axi_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
-static struct clk_branch gcc_mem_noc_q6_axi_clk = {
-	.halt_reg = 0x19010,
-	.clkr = {
-		.enable_reg = 0x19010,
-		.enable_mask = BIT(0),
-		.hw.init = &(const struct clk_init_data) {
-			.name = "gcc_mem_noc_q6_axi_clk",
-			.parent_hws = (const struct clk_hw *[]) {
-				&q6_axi_clk_src.clkr.hw
-			},
-			.num_parents = 1,
-			.flags = CLK_SET_RATE_PARENT,
-			.ops = &clk_branch2_ops,
-		},
-	},
-};
-
 static const struct freq_tbl ftbl_q6_axim2_clk_src[] = {
 	F(342857143, P_GPLL4, 3.5, 0, 0),
 	{ }
@@ -4102,16 +3794,8 @@ static struct clk_regmap *gcc_ipq9574_clks[] = {
 	[GCC_NSSNOC_SNOC_1_CLK] = &gcc_nssnoc_snoc_1_clk.clkr,
 	[GCC_QDSS_ETR_USB_CLK] = &gcc_qdss_etr_usb_clk.clkr,
 	[WCSS_AHB_CLK_SRC] = &wcss_ahb_clk_src.clkr,
-	[GCC_Q6_AHB_CLK] = &gcc_q6_ahb_clk.clkr,
-	[GCC_Q6_AHB_S_CLK] = &gcc_q6_ahb_s_clk.clkr,
-	[GCC_WCSS_ECAHB_CLK] = &gcc_wcss_ecahb_clk.clkr,
-	[GCC_WCSS_ACMT_CLK] = &gcc_wcss_acmt_clk.clkr,
-	[GCC_SYS_NOC_WCSS_AHB_CLK] = &gcc_sys_noc_wcss_ahb_clk.clkr,
 	[WCSS_AXI_M_CLK_SRC] = &wcss_axi_m_clk_src.clkr,
-	[GCC_ANOC_WCSS_AXI_M_CLK] = &gcc_anoc_wcss_axi_m_clk.clkr,
 	[QDSS_AT_CLK_SRC] = &qdss_at_clk_src.clkr,
-	[GCC_Q6SS_ATBM_CLK] = &gcc_q6ss_atbm_clk.clkr,
-	[GCC_WCSS_DBG_IFC_ATB_CLK] = &gcc_wcss_dbg_ifc_atb_clk.clkr,
 	[GCC_NSSNOC_ATB_CLK] = &gcc_nssnoc_atb_clk.clkr,
 	[GCC_QDSS_AT_CLK] = &gcc_qdss_at_clk.clkr,
 	[GCC_SYS_NOC_AT_CLK] = &gcc_sys_noc_at_clk.clkr,
@@ -4124,27 +3808,18 @@ static struct clk_regmap *gcc_ipq9574_clks[] = {
 	[QDSS_TRACECLKIN_CLK_SRC] = &qdss_traceclkin_clk_src.clkr,
 	[GCC_QDSS_TRACECLKIN_CLK] = &gcc_qdss_traceclkin_clk.clkr,
 	[QDSS_TSCTR_CLK_SRC] = &qdss_tsctr_clk_src.clkr,
-	[GCC_Q6_TSCTR_1TO2_CLK] = &gcc_q6_tsctr_1to2_clk.clkr,
-	[GCC_WCSS_DBG_IFC_NTS_CLK] = &gcc_wcss_dbg_ifc_nts_clk.clkr,
 	[GCC_QDSS_TSCTR_DIV2_CLK] = &gcc_qdss_tsctr_div2_clk.clkr,
 	[GCC_QDSS_TS_CLK] = &gcc_qdss_ts_clk.clkr,
 	[GCC_QDSS_TSCTR_DIV4_CLK] = &gcc_qdss_tsctr_div4_clk.clkr,
 	[GCC_NSS_TS_CLK] = &gcc_nss_ts_clk.clkr,
 	[GCC_QDSS_TSCTR_DIV8_CLK] = &gcc_qdss_tsctr_div8_clk.clkr,
 	[GCC_QDSS_TSCTR_DIV16_CLK] = &gcc_qdss_tsctr_div16_clk.clkr,
-	[GCC_Q6SS_PCLKDBG_CLK] = &gcc_q6ss_pclkdbg_clk.clkr,
-	[GCC_Q6SS_TRIG_CLK] = &gcc_q6ss_trig_clk.clkr,
-	[GCC_WCSS_DBG_IFC_APB_CLK] = &gcc_wcss_dbg_ifc_apb_clk.clkr,
-	[GCC_WCSS_DBG_IFC_DAPBUS_CLK] = &gcc_wcss_dbg_ifc_dapbus_clk.clkr,
 	[GCC_QDSS_DAP_CLK] = &gcc_qdss_dap_clk.clkr,
 	[GCC_QDSS_APB2JTAG_CLK] = &gcc_qdss_apb2jtag_clk.clkr,
 	[GCC_QDSS_TSCTR_DIV3_CLK] = &gcc_qdss_tsctr_div3_clk.clkr,
 	[QPIC_IO_MACRO_CLK_SRC] = &qpic_io_macro_clk_src.clkr,
 	[GCC_QPIC_IO_MACRO_CLK] = &gcc_qpic_io_macro_clk.clkr,
 	[Q6_AXI_CLK_SRC] = &q6_axi_clk_src.clkr,
-	[GCC_Q6_AXIM_CLK] = &gcc_q6_axim_clk.clkr,
-	[GCC_WCSS_Q6_TBU_CLK] = &gcc_wcss_q6_tbu_clk.clkr,
-	[GCC_MEM_NOC_Q6_AXI_CLK] = &gcc_mem_noc_q6_axi_clk.clkr,
 	[Q6_AXIM2_CLK_SRC] = &q6_axim2_clk_src.clkr,
 	[NSSNOC_MEMNOC_BFDCD_CLK_SRC] = &nssnoc_memnoc_bfdcd_clk_src.clkr,
 	[GCC_NSSNOC_MEMNOC_CLK] = &gcc_nssnoc_memnoc_clk.clkr,
@@ -4168,7 +3843,6 @@ static struct clk_regmap *gcc_ipq9574_clks[] = {
 	[GCC_UNIPHY1_SYS_CLK] = &gcc_uniphy1_sys_clk.clkr,
 	[GCC_UNIPHY2_SYS_CLK] = &gcc_uniphy2_sys_clk.clkr,
 	[GCC_CMN_12GPLL_SYS_CLK] = &gcc_cmn_12gpll_sys_clk.clkr,
-	[GCC_Q6SS_BOOT_CLK] = &gcc_q6ss_boot_clk.clkr,
 	[UNIPHY_SYS_CLK_SRC] = &uniphy_sys_clk_src.clkr,
 	[NSS_TS_CLK_SRC] = &nss_ts_clk_src.clkr,
 	[GCC_ANOC_PCIE0_1LANE_M_CLK] = &gcc_anoc_pcie0_1lane_m_clk.clkr,
--
2.17.1

[PATCH V2 08/13] firmware: qcom_scm: ipq5018: Add WCSS AHB pd support

[Manikanta Mylavarapu]

Add support to power up, down & load userpd firmware.

Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
---
Changes in V2:
	- This patch is generated by Strip off SCM code changes from here
	  https://lore.kernel.org/linux-arm-msm/1678164097-13247-9-git-send-email-quic_mmanikan@quicinc.com/

 drivers/firmware/qcom_scm.c            | 114 +++++++++++++++++++++++++
 drivers/firmware/qcom_scm.h            |   6 ++
 include/linux/firmware/qcom/qcom_scm.h |   3 +
 3 files changed, 123 insertions(+)

diff --git a/drivers/firmware/qcom_scm.c b/drivers/firmware/qcom_scm.c
index fde33acd46b7..c617e9e671ec 100644
--- a/drivers/firmware/qcom_scm.c
+++ b/drivers/firmware/qcom_scm.c
@@ -643,6 +643,120 @@ int qcom_scm_pas_shutdown(u32 peripheral)
 }
 EXPORT_SYMBOL(qcom_scm_pas_shutdown);

+/**
+ * qti_scm_int_radio_powerup - Bring up WCSS AHB userpd
+ *
+ * @peripheral:	peripheral id
+ *
+ * Return 0 on success.
+ */
+int qti_scm_int_radio_powerup(u32 peripheral)
+{
+	int ret;
+	struct qcom_scm_desc desc = {
+		.svc = QCOM_SCM_PD_LOAD_SVC_ID,
+		.cmd = QCOM_SCM_INT_RAD_PWR_UP_CMD_ID,
+		.arginfo = QCOM_SCM_ARGS(1),
+		.args[0] = peripheral,
+		.owner = ARM_SMCCC_OWNER_SIP,
+	};
+	struct qcom_scm_res res;
+
+	ret = qcom_scm_clk_enable();
+	if (ret)
+		return ret;
+
+	ret = qcom_scm_bw_enable();
+	if (ret)
+		return ret;
+
+	ret = qcom_scm_call(__scm->dev, &desc, &res);
+	qcom_scm_bw_disable();
+	qcom_scm_clk_disable();
+
+	return ret ? : res.result[0];
+}
+EXPORT_SYMBOL(qti_scm_int_radio_powerup);
+
+/**
+ * qti_scm_int_radio_powerdown() - Shut down WCSS AHB userpd
+ *
+ * @peripheral: peripheral id
+ *
+ * Returns 0 on success.
+ */
+int qti_scm_int_radio_powerdown(u32 peripheral)
+{
+	int ret;
+	struct qcom_scm_desc desc = {
+		.svc = QCOM_SCM_PD_LOAD_SVC_ID,
+		.cmd = QCOM_SCM_INT_RAD_PWR_DN_CMD_ID,
+		.arginfo = QCOM_SCM_ARGS(1),
+		.args[0] = peripheral,
+		.owner = ARM_SMCCC_OWNER_SIP,
+	};
+	struct qcom_scm_res res;
+
+	ret = qcom_scm_clk_enable();
+	if (ret)
+		return ret;
+
+	ret = qcom_scm_bw_enable();
+	if (ret)
+		return ret;
+
+	ret = qcom_scm_call(__scm->dev, &desc, &res);
+	qcom_scm_bw_disable();
+	qcom_scm_clk_disable();
+
+	return ret ? : res.result[0];
+}
+EXPORT_SYMBOL(qti_scm_int_radio_powerdown);
+
+/**
+ * qti_scm_pdseg_memcpy_v2() - copy userpd PIL segments data to dma blocks
+ *
+ * @peripheral:		peripheral id
+ * @phno:		program header no
+ * @dma:		handle of dma region
+ * @seg_cnt:		no of dma blocks
+ *
+ * Returns 0 if trustzone successfully loads userpd PIL segments from dma
+ * blocks to DDR
+ */
+int qti_scm_pdseg_memcpy_v2(u32 peripheral, int phno, dma_addr_t dma,
+			    int seg_cnt)
+{
+	int ret;
+	struct qcom_scm_desc desc = {
+		.svc = QCOM_SCM_PD_LOAD_SVC_ID,
+		.cmd = QCOM_SCM_PD_LOAD_V2_CMD_ID,
+		.arginfo = QCOM_SCM_ARGS(4, QCOM_SCM_VAL, QCOM_SCM_VAL,
+						QCOM_SCM_RW, QCOM_SCM_VAL),
+		.args[0] = peripheral,
+		.args[1] = phno,
+		.args[2] = dma,
+		.args[3] = seg_cnt,
+		.owner = ARM_SMCCC_OWNER_SIP,
+	};
+	struct qcom_scm_res res;
+
+	ret = qcom_scm_clk_enable();
+	if (ret)
+		return ret;
+
+	ret = qcom_scm_bw_enable();
+	if (ret)
+		return ret;
+
+	ret = qcom_scm_call(__scm->dev, &desc, &res);
+	qcom_scm_bw_disable();
+	qcom_scm_clk_disable();
+
+	return ret ? : res.result[0];
+}
+EXPORT_SYMBOL(qti_scm_pdseg_memcpy_v2);
+
 /**
  * qcom_scm_pas_supported() - Check if the peripheral authentication service is
  *			      available for the given peripherial
diff --git a/drivers/firmware/qcom_scm.h b/drivers/firmware/qcom_scm.h
index e6e512bd57d1..99e3ab2f1986 100644
--- a/drivers/firmware/qcom_scm.h
+++ b/drivers/firmware/qcom_scm.h
@@ -132,6 +132,12 @@ extern int scm_legacy_call(struct device *dev, const struct qcom_scm_desc *desc,
 #define QCOM_SCM_SMMU_CONFIG_ERRATA1		0x03
 #define QCOM_SCM_SMMU_CONFIG_ERRATA1_CLIENT_ALL	0x02

+#define QCOM_SCM_PD_LOAD_SVC_ID			0x2
+#define QCOM_SCM_PD_LOAD_CMD_ID			0x16
+#define QCOM_SCM_PD_LOAD_V2_CMD_ID		0x19
+#define QCOM_SCM_INT_RAD_PWR_UP_CMD_ID		0x17
+#define QCOM_SCM_INT_RAD_PWR_DN_CMD_ID		0x18
+
 #define QCOM_SCM_SVC_WAITQ			0x24
 #define QCOM_SCM_WAITQ_RESUME			0x02
 #define QCOM_SCM_WAITQ_GET_WQ_CTX		0x03
diff --git a/include/linux/firmware/qcom/qcom_scm.h b/include/linux/firmware/qcom/qcom_scm.h
index 250ea4efb7cb..488d6eccb5a4 100644
--- a/include/linux/firmware/qcom/qcom_scm.h
+++ b/include/linux/firmware/qcom/qcom_scm.h
@@ -81,6 +81,9 @@ extern int qcom_scm_pas_mem_setup(u32 peripheral, phys_addr_t addr,
 extern int qcom_scm_pas_auth_and_reset(u32 peripheral);
 extern int qcom_scm_pas_shutdown(u32 peripheral);
 extern bool qcom_scm_pas_supported(u32 peripheral);
+int qti_scm_int_radio_powerup(u32 peripheral);
+int qti_scm_int_radio_powerdown(u32 peripheral);
+int qti_scm_pdseg_memcpy_v2(u32 peripheral, int phno, dma_addr_t dma, int seg_cnt);

 extern int qcom_scm_io_readl(phys_addr_t addr, unsigned int *val);
 extern int qcom_scm_io_writel(phys_addr_t addr, unsigned int val);
--
2.17.1

[PATCH V2 09/13] remoteproc: qcom: q6v5: Add multipd interrupts support

[Manikanta Mylavarapu]

In multipd model, root & user pd remoteproc's interrupts are
different. User pd needs additional interrupts like spawn.
Instead of going with qcom_q6v5_init(), we defined a new
function to register userpd rproc interrupts in mpd driver.
Since userpd rproc uses some of common interrupts like fatal,
ready, static is removed from ISR handler and used in userpd
interrupt registration.

Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
---
Changes in V2:
	- This patch is generated by Strip off qcom_q6v5 changes from here
	  https://lore.kernel.org/linux-arm-msm/1678164097-13247-9-git-send-email-quic_mmanikan@quicinc.com/

 drivers/remoteproc/qcom_q6v5.c | 37 +++++++++++++++++++++++++++++++---
 drivers/remoteproc/qcom_q6v5.h | 11 ++++++++++
 2 files changed, 45 insertions(+), 3 deletions(-)

diff --git a/drivers/remoteproc/qcom_q6v5.c b/drivers/remoteproc/qcom_q6v5.c
index 192c7aa0e39e..d535f3365865 100644
--- a/drivers/remoteproc/qcom_q6v5.c
+++ b/drivers/remoteproc/qcom_q6v5.c
@@ -118,7 +118,7 @@ static irqreturn_t q6v5_wdog_interrupt(int irq, void *data)
 	return IRQ_HANDLED;
 }

-static irqreturn_t q6v5_fatal_interrupt(int irq, void *data)
+irqreturn_t q6v5_fatal_interrupt(int irq, void *data)
 {
 	struct qcom_q6v5 *q6v5 = data;
 	size_t len;
@@ -139,7 +139,7 @@ static irqreturn_t q6v5_fatal_interrupt(int irq, void *data)
 	return IRQ_HANDLED;
 }

-static irqreturn_t q6v5_ready_interrupt(int irq, void *data)
+irqreturn_t q6v5_ready_interrupt(int irq, void *data)
 {
 	struct qcom_q6v5 *q6v5 = data;

@@ -183,7 +183,16 @@ static irqreturn_t q6v5_handover_interrupt(int irq, void *data)
 	return IRQ_HANDLED;
 }

-static irqreturn_t q6v5_stop_interrupt(int irq, void *data)
+irqreturn_t q6v5_spawn_interrupt(int irq, void *data)
+{
+	struct qcom_q6v5 *q6v5 = data;
+
+	complete(&q6v5->spawn_done);
+
+	return IRQ_HANDLED;
+}
+
+irqreturn_t q6v5_stop_interrupt(int irq, void *data)
 {
 	struct qcom_q6v5 *q6v5 = data;

@@ -220,6 +229,28 @@ int qcom_q6v5_request_stop(struct qcom_q6v5 *q6v5, struct qcom_sysmon *sysmon)
 }
 EXPORT_SYMBOL_GPL(qcom_q6v5_request_stop);

+/**
+ * qcom_q6v5_request_spawn() - request the remote processor to spawn
+ * @q6v5:      reference to qcom_q6v5 context
+ *
+ * Return: 0 on success, negative errno on failure
+ */
+int qcom_q6v5_request_spawn(struct qcom_q6v5 *q6v5)
+{
+	int ret;
+
+	ret = qcom_smem_state_update_bits(q6v5->spawn_state,
+					  BIT(q6v5->spawn_bit), BIT(q6v5->spawn_bit));
+
+	ret = wait_for_completion_timeout(&q6v5->spawn_done, 5 * HZ);
+
+	qcom_smem_state_update_bits(q6v5->spawn_state,
+				    BIT(q6v5->spawn_bit), 0);
+
+	return ret == 0 ? -ETIMEDOUT : 0;
+}
+EXPORT_SYMBOL_GPL(qcom_q6v5_request_spawn);
+
 /**
  * qcom_q6v5_panic() - panic handler to invoke a stop on the remote
  * @q6v5:	reference to qcom_q6v5 context
diff --git a/drivers/remoteproc/qcom_q6v5.h b/drivers/remoteproc/qcom_q6v5.h
index 5a859c41896e..4e1bb1a68284 100644
--- a/drivers/remoteproc/qcom_q6v5.h
+++ b/drivers/remoteproc/qcom_q6v5.h
@@ -18,21 +18,27 @@ struct qcom_q6v5 {

 	struct qcom_smem_state *state;
 	struct qmp *qmp;
+	struct qcom_smem_state *shutdown_state;
+	struct qcom_smem_state *spawn_state;

 	struct icc_path *path;

 	unsigned stop_bit;
+	unsigned shutdown_bit;
+	unsigned spawn_bit;

 	int wdog_irq;
 	int fatal_irq;
 	int ready_irq;
 	int handover_irq;
 	int stop_irq;
+	int spawn_irq;

 	bool handover_issued;

 	struct completion start_done;
 	struct completion stop_done;
+	struct completion spawn_done;

 	int crash_reason;

@@ -50,7 +56,12 @@ void qcom_q6v5_deinit(struct qcom_q6v5 *q6v5);
 int qcom_q6v5_prepare(struct qcom_q6v5 *q6v5);
 int qcom_q6v5_unprepare(struct qcom_q6v5 *q6v5);
 int qcom_q6v5_request_stop(struct qcom_q6v5 *q6v5, struct qcom_sysmon *sysmon);
+int qcom_q6v5_request_spawn(struct qcom_q6v5 *q6v5);
 int qcom_q6v5_wait_for_start(struct qcom_q6v5 *q6v5, int timeout);
 unsigned long qcom_q6v5_panic(struct qcom_q6v5 *q6v5);
+irqreturn_t q6v5_fatal_interrupt(int irq, void *data);
+irqreturn_t q6v5_ready_interrupt(int irq, void *data);
+irqreturn_t q6v5_spawn_interrupt(int irq, void *data);
+irqreturn_t q6v5_stop_interrupt(int irq, void *data);

 #endif
--
2.17.1

[PATCH V2 10/13] remoteproc: qcom: Add Hexagon based multipd rproc driver

[Manikanta Mylavarapu]

It adds support to bring up remoteproc's on multipd model.
Pd means protection domain. It's similar to process in Linux.
Here QDSP6 processor runs each wifi radio functionality on a
separate process. One process can't access other process
resources, so this is termed as PD i.e protection domain.

Here we have two pd's called root and user pd. We can correlate
Root pd as root and user pd as user in linux. Root pd has more
privileges than user pd. Root will provide services to user pd.

From remoteproc driver perspective, root pd corresponds to QDSP6
processor bring up and user pd corresponds to Wifi radio (WCSS)
bring up.

Here WCSS(user) PD is dependent on Q6(root) PD, so first
q6 pd should be up before wcss pd. After wcss pd goes down,
q6 pd should be turned off.

IPQ5018, IPQ9574 supports multipd remoteproc driver.

Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
---
Changes in V2:
	- Common functionalities moved to seperate patches
	- qcom_get_pd_asid() moved to mpd driver
	- Last DMA block alone memset to zero
	- Added diagram to show how userpd data is organized and sent to
	  trustzone
	- Rewritten commit message since most of the content available
	  in cover page
	- Removed 'remote_id' becuase it's not required for bring up.

 drivers/remoteproc/Kconfig          |  20 +
 drivers/remoteproc/Makefile         |   1 +
 drivers/remoteproc/qcom_common.h    |   9 +
 drivers/remoteproc/qcom_q6v5_mpd.c  | 677 ++++++++++++++++++++++++++++
 drivers/soc/qcom/mdt_loader.c       | 332 ++++++++++++++
 include/linux/soc/qcom/mdt_loader.h |  19 +
 6 files changed, 1058 insertions(+)
 create mode 100644 drivers/remoteproc/qcom_q6v5_mpd.c

diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig
index a850e9f486dd..44af5c36f67e 100644
--- a/drivers/remoteproc/Kconfig
+++ b/drivers/remoteproc/Kconfig
@@ -234,6 +234,26 @@ config QCOM_Q6V5_PAS
 	  CDSP (Compute DSP), MPSS (Modem Peripheral SubSystem), and
 	  SLPI (Sensor Low Power Island).

+config QCOM_Q6V5_MPD
+	tristate "Qualcomm Hexagon based MPD model Peripheral Image Loader"
+	depends on OF && ARCH_QCOM
+	depends on QCOM_SMEM
+	depends on RPMSG_QCOM_SMD || RPMSG_QCOM_SMD=n
+	depends on RPMSG_QCOM_GLINK_SMEM || RPMSG_QCOM_GLINK_SMEM=n
+	depends on QCOM_SYSMON || QCOM_SYSMON=n
+	depends on RPMSG_QCOM_GLINK || RPMSG_QCOM_GLINK=n
+	depends on QCOM_AOSS_QMP || QCOM_AOSS_QMP=n
+	select MFD_SYSCON
+	select QCOM_MDT_LOADER
+	select QCOM_PIL_INFO
+	select QCOM_Q6V5_COMMON
+	select QCOM_RPROC_COMMON
+	select QCOM_SCM
+	help
+	  Say y here to support the Qualcomm Secure Peripheral Image Loader
+	  for the Hexagon based MultiPD model remote processors on e.g. IPQ5018.
+	  This is trustZone wireless subsystem.
+
 config QCOM_Q6V5_WCSS
 	tristate "Qualcomm Hexagon based WCSS Peripheral Image Loader"
 	depends on OF && ARCH_QCOM
diff --git a/drivers/remoteproc/Makefile b/drivers/remoteproc/Makefile
index 91314a9b43ce..b64051080ec1 100644
--- a/drivers/remoteproc/Makefile
+++ b/drivers/remoteproc/Makefile
@@ -25,6 +25,7 @@ obj-$(CONFIG_QCOM_PIL_INFO)		+= qcom_pil_info.o
 obj-$(CONFIG_QCOM_RPROC_COMMON)		+= qcom_common.o
 obj-$(CONFIG_QCOM_Q6V5_COMMON)		+= qcom_q6v5.o
 obj-$(CONFIG_QCOM_Q6V5_ADSP)		+= qcom_q6v5_adsp.o
+obj-$(CONFIG_QCOM_Q6V5_MPD)		+= qcom_q6v5_mpd.o
 obj-$(CONFIG_QCOM_Q6V5_MSS)		+= qcom_q6v5_mss.o
 obj-$(CONFIG_QCOM_Q6V5_PAS)		+= qcom_q6v5_pas.o
 obj-$(CONFIG_QCOM_Q6V5_WCSS)		+= qcom_q6v5_wcss.o
diff --git a/drivers/remoteproc/qcom_common.h b/drivers/remoteproc/qcom_common.h
index 9ef4449052a9..9d9e08aac6ef 100644
--- a/drivers/remoteproc/qcom_common.h
+++ b/drivers/remoteproc/qcom_common.h
@@ -76,4 +76,13 @@ static inline bool qcom_sysmon_shutdown_acked(struct qcom_sysmon *sysmon)
 }
 #endif

+#if IS_ENABLED(CONFIG_QCOM_Q6V5_MPD)
+s8 qcom_get_pd_asid(struct device_node *node);
+#else
+static inline s8 qcom_get_pd_asid(struct device_node *node)
+{
+	return 0;
+}
+#endif
+
 #endif
diff --git a/drivers/remoteproc/qcom_q6v5_mpd.c b/drivers/remoteproc/qcom_q6v5_mpd.c
new file mode 100644
index 000000000000..959143960350
--- /dev/null
+++ b/drivers/remoteproc/qcom_q6v5_mpd.c
@@ -0,0 +1,677 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2016-2018 Linaro Ltd.
+ * Copyright (C) 2014 Sony Mobile Communications AB
+ * Copyright (c) 2012-2018, 2021 The Linux Foundation. All rights reserved.
+ */
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/firmware/qcom/qcom_scm.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/iopoll.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of_address.h>
+#include <linux/of_device.h>
+#include <linux/of_reserved_mem.h>
+#include <linux/platform_device.h>
+#include <linux/reset.h>
+#include <linux/soc/qcom/mdt_loader.h>
+#include <linux/soc/qcom/smem.h>
+#include <linux/soc/qcom/smem_state.h>
+#include "qcom_common.h"
+#include "qcom_q6v5.h"
+
+#include "remoteproc_internal.h"
+
+#define WCSS_CRASH_REASON		421
+#define WCSS_SMEM_HOST			1
+
+#define WCNSS_PAS_ID			6
+#define MPD_WCNSS_PAS_ID        0xD
+
+#define BUF_SIZE			35
+
+#define MAX_FIRMWARE			3
+/**
+ * enum state - state of a wcss (private)
+ * @WCSS_NORMAL: subsystem is operating normally
+ * @WCSS_SHUTDOWN: subsystem has been shutdown
+ *
+ */
+enum q6_wcss_state {
+	WCSS_NORMAL,
+	WCSS_SHUTDOWN,
+};
+
+enum {
+	Q6_IPQ,
+	WCSS_AHB_IPQ,
+	WCSS_PCIE_IPQ,
+};
+
+struct q6_wcss {
+	struct device *dev;
+	struct qcom_rproc_glink glink_subdev;
+	struct qcom_rproc_ssr ssr_subdev;
+	struct qcom_q6v5 q6;
+	phys_addr_t mem_phys;
+	phys_addr_t mem_reloc;
+	void *mem_region;
+	size_t mem_size;
+	int crash_reason_smem;
+	s8 pd_asid;
+	enum q6_wcss_state state;
+	const struct wcss_data *desc;
+	const char **firmware;
+};
+
+struct wcss_data {
+	int (*init_irq)(struct qcom_q6v5 *q6, struct platform_device *pdev,
+			struct rproc *rproc, int crash_reason,
+			const char *load_state,
+			void (*handover)(struct qcom_q6v5 *q6));
+	int crash_reason_smem;
+	u32 version;
+	const char *ssr_name;
+	const struct rproc_ops *ops;
+	bool glink_subdev_required;
+	bool reset_seq;
+	u32 pasid;
+	int (*mdt_load_sec)(struct device *dev, const struct firmware *fw,
+			    const char *fw_name, int pas_id, void *mem_region,
+			    phys_addr_t mem_phys, size_t mem_size,
+			    phys_addr_t *reloc_base);
+	int (*powerup_scm)(u32 peripheral);
+	int (*powerdown_scm)(u32 peripheral);
+};
+
+/**
+ * qcom_get_pd_asid() - get the pd asid number from DT node
+ * @node:	device tree node
+ *
+ * Returns asid if node name has 'pd' string
+ */
+s8 qcom_get_pd_asid(struct device_node *node)
+{
+	char *str;
+	s8 pd_asid;
+
+	if (!node)
+		return -EINVAL;
+
+	str = strstr(node->name, "pd");
+	if (!str)
+		return 0;
+
+	str += strlen("pd") + 1;
+	return kstrtos8(str, 10, &pd_asid) ? -EINVAL : pd_asid;
+}
+EXPORT_SYMBOL(qcom_get_pd_asid);
+
+static int q6_wcss_start(struct rproc *rproc)
+{
+	struct q6_wcss *wcss = rproc->priv;
+	int ret;
+	struct device_node *upd_np;
+	struct platform_device *upd_pdev;
+	struct rproc *upd_rproc;
+	struct q6_wcss *upd_wcss;
+	const struct wcss_data *desc = wcss->desc;
+
+	qcom_q6v5_prepare(&wcss->q6);
+
+	ret = qcom_scm_pas_auth_and_reset(desc->pasid);
+	if (ret) {
+		dev_err(wcss->dev, "wcss_reset failed\n");
+		return ret;
+	}
+
+	ret = qcom_q6v5_wait_for_start(&wcss->q6, 5 * HZ);
+	if (ret == -ETIMEDOUT)
+		dev_err(wcss->dev, "start timed out\n");
+
+	/* On rootpd restart still user pd wcss state's
+	 * initialized to WCSS_SHUTDOWN and it leads to
+	 * user pd FW load (user pd fw load should happen
+	 * only on user pd restart, not on root pd restart).
+	 * So bring userpd wcss state to default value.
+	 */
+	for_each_available_child_of_node(wcss->dev->of_node, upd_np) {
+		upd_pdev = of_find_device_by_node(upd_np);
+		if (!upd_pdev)
+			continue;
+		upd_rproc = platform_get_drvdata(upd_pdev);
+		upd_wcss = upd_rproc->priv;
+		upd_wcss->state = WCSS_NORMAL;
+	}
+	return ret;
+}
+
+static int q6_wcss_spawn_pd(struct rproc *rproc)
+{
+	int ret;
+	struct q6_wcss *wcss = rproc->priv;
+
+	ret = qcom_q6v5_request_spawn(&wcss->q6);
+	if (ret == -ETIMEDOUT) {
+		pr_err("%s spawn timedout\n", rproc->name);
+		return ret;
+	}
+
+	ret = qcom_q6v5_wait_for_start(&wcss->q6, msecs_to_jiffies(10000));
+	if (ret == -ETIMEDOUT) {
+		pr_err("%s start timedout\n", rproc->name);
+		wcss->q6.running = false;
+		return ret;
+	}
+	wcss->q6.running = true;
+	return ret;
+}
+
+static int wcss_ahb_pcie_pd_start(struct rproc *rproc)
+{
+	struct q6_wcss *wcss = rproc->priv;
+	const struct wcss_data *desc = wcss->desc;
+	int ret;
+
+	if (!desc->reset_seq)
+		return 0;
+
+	if (desc->powerup_scm) {
+		ret = desc->powerup_scm(desc->pasid);
+		if (ret) {
+			dev_err(wcss->dev, "failed to power up pd\n");
+			return ret;
+		}
+	}
+
+	ret = q6_wcss_spawn_pd(rproc);
+	if (ret)
+		return ret;
+
+	wcss->state = WCSS_NORMAL;
+	return ret;
+}
+
+static int q6_wcss_stop(struct rproc *rproc)
+{
+	struct q6_wcss *wcss = rproc->priv;
+	const struct wcss_data *desc = wcss->desc;
+	int ret;
+
+	ret = qcom_scm_pas_shutdown(desc->pasid);
+	if (ret) {
+		dev_err(wcss->dev, "not able to shutdown\n");
+		return ret;
+	}
+	qcom_q6v5_unprepare(&wcss->q6);
+
+	return 0;
+}
+
+static int wcss_ahb_pcie_pd_stop(struct rproc *rproc)
+{
+	struct q6_wcss *wcss = rproc->priv;
+	struct rproc *rpd_rproc = dev_get_drvdata(wcss->dev->parent);
+	const struct wcss_data *desc = wcss->desc;
+	int ret;
+
+	if (!desc->reset_seq)
+		goto shut_down_rpd;
+
+	if (rproc->state != RPROC_CRASHED && wcss->q6.stop_bit) {
+		ret = qcom_q6v5_request_stop(&wcss->q6, NULL);
+		if (ret) {
+			dev_err(&rproc->dev, "pd not stopped\n");
+			return ret;
+		}
+	}
+
+	if (desc->powerdown_scm) {
+		ret = desc->powerdown_scm(desc->pasid);
+		if (ret) {
+			dev_err(wcss->dev, "failed to power down pd\n");
+			return ret;
+		}
+	}
+
+shut_down_rpd:
+	rproc_shutdown(rpd_rproc);
+
+	wcss->state = WCSS_SHUTDOWN;
+	return 0;
+}
+
+static void *q6_wcss_da_to_va(struct rproc *rproc, u64 da, size_t len,
+			      bool *is_iomem)
+{
+	struct q6_wcss *wcss = rproc->priv;
+	int offset;
+
+	offset = da - wcss->mem_reloc;
+	if (offset < 0 || offset + len > wcss->mem_size)
+		return NULL;
+
+	return wcss->mem_region + offset;
+}
+
+static int q6_wcss_load(struct rproc *rproc, const struct firmware *fw)
+{
+	struct q6_wcss *wcss = rproc->priv;
+	const struct firmware *fw_hdl;
+	int ret;
+	const struct wcss_data *desc = wcss->desc;
+	int loop;
+
+	ret = qcom_mdt_load(wcss->dev, fw, rproc->firmware,
+			    desc->pasid, wcss->mem_region,
+			    wcss->mem_phys, wcss->mem_size,
+			    &wcss->mem_reloc);
+	if (ret)
+		return ret;
+
+	for (loop = 1; loop < MAX_FIRMWARE; loop++) {
+		if (!wcss->firmware[loop])
+			continue;
+
+		ret = request_firmware(&fw_hdl, wcss->firmware[loop],
+				       wcss->dev);
+		if (ret)
+			continue;
+
+		ret = qcom_mdt_load_no_init(wcss->dev, fw_hdl,
+					    wcss->firmware[loop], 0,
+					    wcss->mem_region,
+					    wcss->mem_phys,
+					    wcss->mem_size,
+					    &wcss->mem_reloc);
+
+		release_firmware(fw_hdl);
+
+		if (ret) {
+			dev_err(wcss->dev,
+				"can't load %s ret:%d\n", wcss->firmware[loop], ret);
+			return ret;
+		}
+	}
+	return 0;
+}
+
+/* This function load's userpd firmware. Since Userpd depends on rootpd
+ * first bring up root pd and then load. User pd firmware load is required
+ * only during user pd restart because root pd loads user pd FW pil segments
+ * during it's bringup.
+ */
+static int wcss_ahb_pcie_pd_load(struct rproc *rproc, const struct firmware *fw)
+{
+	struct q6_wcss *wcss = rproc->priv, *wcss_rpd;
+	struct rproc *rpd_rproc = dev_get_drvdata(wcss->dev->parent);
+	const struct wcss_data *desc = wcss->desc;
+	int ret;
+
+	wcss_rpd = rpd_rproc->priv;
+
+	/* Boot rootpd rproc */
+	ret = rproc_boot(rpd_rproc);
+	if (ret || wcss->state == WCSS_NORMAL)
+		return ret;
+
+	return desc->mdt_load_sec(wcss->dev, fw, rproc->firmware,
+				  desc->pasid, wcss->mem_region,
+				  wcss->mem_phys, wcss->mem_size,
+				  &wcss->mem_reloc);
+}
+
+static unsigned long q6_wcss_panic(struct rproc *rproc)
+{
+	struct q6_wcss *wcss = rproc->priv;
+
+	return qcom_q6v5_panic(&wcss->q6);
+}
+
+static const struct rproc_ops wcss_ahb_pcie_ipq5018_ops = {
+	.start = wcss_ahb_pcie_pd_start,
+	.stop = wcss_ahb_pcie_pd_stop,
+	.load = wcss_ahb_pcie_pd_load,
+};
+
+static const struct rproc_ops q6_wcss_ipq5018_ops = {
+	.start = q6_wcss_start,
+	.stop = q6_wcss_stop,
+	.da_to_va = q6_wcss_da_to_va,
+	.load = q6_wcss_load,
+	.get_boot_addr = rproc_elf_get_boot_addr,
+	.panic = q6_wcss_panic,
+};
+
+static int q6_alloc_memory_region(struct q6_wcss *wcss)
+{
+	struct reserved_mem *rmem = NULL;
+	struct device_node *node;
+	struct device *dev = wcss->dev;
+	const struct wcss_data *desc = wcss->desc;
+
+	if (desc->version == Q6_IPQ) {
+		node = of_parse_phandle(dev->of_node, "memory-region", 0);
+		if (node)
+			rmem = of_reserved_mem_lookup(node);
+
+		of_node_put(node);
+
+		if (!rmem) {
+			dev_err(dev, "unable to acquire memory-region\n");
+			return -EINVAL;
+		}
+	} else {
+		struct rproc *rpd_rproc = dev_get_drvdata(dev->parent);
+		struct q6_wcss *rpd_wcss = rpd_rproc->priv;
+
+		wcss->mem_phys = rpd_wcss->mem_phys;
+		wcss->mem_reloc = rpd_wcss->mem_reloc;
+		wcss->mem_size = rpd_wcss->mem_size;
+		wcss->mem_region = rpd_wcss->mem_region;
+		return 0;
+	}
+
+	wcss->mem_phys = rmem->base;
+	wcss->mem_reloc = rmem->base;
+	wcss->mem_size = rmem->size;
+	wcss->mem_region = devm_ioremap_wc(dev, wcss->mem_phys, wcss->mem_size);
+	if (!wcss->mem_region) {
+		dev_err(dev, "unable to map memory region: %pa+%pa\n",
+			&rmem->base, &rmem->size);
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
+static int q6_get_inbound_irq(struct qcom_q6v5 *q6,
+			      struct platform_device *pdev,
+			      const char *int_name,
+			      irqreturn_t (*handler)(int irq, void *data))
+{
+	int ret, irq;
+	char *interrupt, *tmp = (char *)int_name;
+	struct q6_wcss *wcss = q6->rproc->priv;
+
+	irq = platform_get_irq_byname(pdev, int_name);
+	if (irq < 0) {
+		if (irq != -EPROBE_DEFER)
+			dev_err(&pdev->dev,
+				"failed to retrieve %s IRQ: %d\n",
+					int_name, irq);
+		return irq;
+	}
+
+	if (!strcmp(int_name, "fatal")) {
+		q6->fatal_irq = irq;
+	} else if (!strcmp(int_name, "stop-ack")) {
+		q6->stop_irq = irq;
+		tmp = "stop_ack";
+	} else if (!strcmp(int_name, "ready")) {
+		q6->ready_irq = irq;
+	} else if (!strcmp(int_name, "handover")) {
+		q6->handover_irq  = irq;
+	} else if (!strcmp(int_name, "spawn-ack")) {
+		q6->spawn_irq = irq;
+		tmp = "spawn_ack";
+	} else {
+		dev_err(&pdev->dev, "unknown interrupt\n");
+		return -EINVAL;
+	}
+
+	interrupt = devm_kzalloc(&pdev->dev, BUF_SIZE, GFP_KERNEL);
+	if (!interrupt)
+		return -ENOMEM;
+
+	snprintf(interrupt, BUF_SIZE, "q6v5_wcss_userpd%d_%s", wcss->pd_asid, tmp);
+
+	ret = devm_request_threaded_irq(&pdev->dev, irq,
+					NULL, handler,
+					IRQF_TRIGGER_RISING | IRQF_ONESHOT,
+					interrupt, q6);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to acquire %s irq\n", interrupt);
+		return ret;
+	}
+	return 0;
+}
+
+static int q6_get_outbound_irq(struct qcom_q6v5 *q6,
+			       struct platform_device *pdev,
+			       const char *int_name)
+{
+	struct qcom_smem_state *tmp_state;
+	unsigned  bit;
+
+	tmp_state = qcom_smem_state_get(&pdev->dev, int_name, &bit);
+	if (IS_ERR(tmp_state)) {
+		dev_err(&pdev->dev, "failed to acquire %s state\n", int_name);
+		return PTR_ERR(tmp_state);
+	}
+
+	if (!strcmp(int_name, "stop")) {
+		q6->state = tmp_state;
+		q6->stop_bit = bit;
+	} else if (!strcmp(int_name, "spawn")) {
+		q6->spawn_state = tmp_state;
+		q6->spawn_bit = bit;
+	}
+
+	return 0;
+}
+
+static int init_irq(struct qcom_q6v5 *q6,
+		    struct platform_device *pdev, struct rproc *rproc,
+		    int crash_reason, const char *load_state,
+		    void (*handover)(struct qcom_q6v5 *q6))
+{
+	int ret;
+
+	q6->rproc = rproc;
+	q6->dev = &pdev->dev;
+	q6->crash_reason = crash_reason;
+	q6->handover = handover;
+
+	init_completion(&q6->start_done);
+	init_completion(&q6->stop_done);
+	init_completion(&q6->spawn_done);
+
+	ret = q6_get_inbound_irq(q6, pdev, "fatal",
+				 q6v5_fatal_interrupt);
+	if (ret)
+		return ret;
+
+	ret = q6_get_inbound_irq(q6, pdev, "ready",
+				 q6v5_ready_interrupt);
+	if (ret)
+		return ret;
+
+	ret = q6_get_inbound_irq(q6, pdev, "stop-ack",
+				 q6v5_stop_interrupt);
+	if (ret)
+		return ret;
+
+	ret = q6_get_inbound_irq(q6, pdev, "spawn-ack",
+				 q6v5_spawn_interrupt);
+	if (ret)
+		return ret;
+
+	ret = q6_get_outbound_irq(q6, pdev, "stop");
+	if (ret)
+		return ret;
+
+	ret = q6_get_outbound_irq(q6, pdev, "spawn");
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int q6_wcss_probe(struct platform_device *pdev)
+{
+	const struct wcss_data *desc;
+	struct q6_wcss *wcss;
+	struct rproc *rproc;
+	int ret;
+	char *subdev_name;
+	const char **firmware;
+
+	desc = of_device_get_match_data(&pdev->dev);
+	if (!desc)
+		return -EINVAL;
+
+	firmware = devm_kcalloc(&pdev->dev, MAX_FIRMWARE,
+				sizeof(*firmware), GFP_KERNEL);
+	if (!firmware)
+		return -ENOMEM;
+
+	ret = of_property_read_string_array(pdev->dev.of_node, "firmware-name",
+					    firmware, MAX_FIRMWARE);
+	if (ret < 0)
+		return ret;
+
+	rproc = rproc_alloc(&pdev->dev, pdev->name, desc->ops,
+			    firmware[0], sizeof(*wcss));
+	if (!rproc) {
+		dev_err(&pdev->dev, "failed to allocate rproc\n");
+		return -ENOMEM;
+	}
+	wcss = rproc->priv;
+	wcss->dev = &pdev->dev;
+	wcss->desc = desc;
+	wcss->firmware = firmware;
+
+	ret = q6_alloc_memory_region(wcss);
+	if (ret)
+		goto free_rproc;
+
+	wcss->pd_asid = qcom_get_pd_asid(wcss->dev->of_node);
+	if (wcss->pd_asid < 0)
+		goto free_rproc;
+
+	if (desc->init_irq) {
+		ret = desc->init_irq(&wcss->q6, pdev, rproc,
+				     desc->crash_reason_smem, NULL, NULL);
+		if (ret)
+			goto free_rproc;
+	}
+
+	if (desc->glink_subdev_required)
+		qcom_add_glink_subdev(rproc, &wcss->glink_subdev, desc->ssr_name);
+
+	subdev_name = (char *)(desc->ssr_name ? desc->ssr_name : pdev->name);
+	qcom_add_ssr_subdev(rproc, &wcss->ssr_subdev, subdev_name);
+
+	rproc->auto_boot = false;
+	ret = rproc_add(rproc);
+	if (ret)
+		goto free_rproc;
+
+	platform_set_drvdata(pdev, rproc);
+
+	ret = of_platform_populate(wcss->dev->of_node, NULL,
+				   NULL, wcss->dev);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to populate wcss pd nodes\n");
+		goto free_rproc;
+	}
+	return 0;
+
+free_rproc:
+	rproc_free(rproc);
+
+	return ret;
+}
+
+static int q6_wcss_remove(struct platform_device *pdev)
+{
+	struct rproc *rproc = platform_get_drvdata(pdev);
+	struct q6_wcss *wcss = rproc->priv;
+
+	qcom_q6v5_deinit(&wcss->q6);
+
+	rproc_del(rproc);
+	rproc_free(rproc);
+
+	return 0;
+}
+
+static const struct wcss_data q6_ipq5018_res_init = {
+	.init_irq = qcom_q6v5_init,
+	.crash_reason_smem = WCSS_CRASH_REASON,
+	.ssr_name = "q6wcss",
+	.ops = &q6_wcss_ipq5018_ops,
+	.version = Q6_IPQ,
+	.glink_subdev_required = true,
+	.pasid = MPD_WCNSS_PAS_ID,
+};
+
+static const struct wcss_data q6_ipq9574_res_init = {
+	.init_irq = qcom_q6v5_init,
+	.crash_reason_smem = WCSS_CRASH_REASON,
+	.ssr_name = "q6wcss",
+	.ops = &q6_wcss_ipq5018_ops,
+	.version = Q6_IPQ,
+	.glink_subdev_required = true,
+	.pasid = WCNSS_PAS_ID,
+};
+
+static const struct wcss_data wcss_ahb_ipq5018_res_init = {
+	.init_irq = init_irq,
+	.crash_reason_smem = WCSS_CRASH_REASON,
+	.ops = &wcss_ahb_pcie_ipq5018_ops,
+	.version = WCSS_AHB_IPQ,
+	.pasid = MPD_WCNSS_PAS_ID,
+	.reset_seq = true,
+	.mdt_load_sec = qcom_mdt_load_pd_seg,
+	.powerup_scm = qti_scm_int_radio_powerup,
+	.powerdown_scm = qti_scm_int_radio_powerdown,
+};
+
+static const struct wcss_data wcss_ahb_ipq9574_res_init = {
+	.crash_reason_smem = WCSS_CRASH_REASON,
+	.ops = &wcss_ahb_pcie_ipq5018_ops,
+	.version = WCSS_AHB_IPQ,
+	.pasid = WCNSS_PAS_ID,
+	.mdt_load_sec = qcom_mdt_load,
+};
+
+static const struct wcss_data wcss_pcie_ipq5018_res_init = {
+	.init_irq = init_irq,
+	.crash_reason_smem = WCSS_CRASH_REASON,
+	.ops = &wcss_ahb_pcie_ipq5018_ops,
+	.version = WCSS_PCIE_IPQ,
+	.reset_seq = true,
+	.mdt_load_sec = qcom_mdt_load_pd_seg,
+	.pasid = MPD_WCNSS_PAS_ID,
+};
+
+static const struct of_device_id q6_wcss_of_match[] = {
+	{ .compatible = "qcom,ipq5018-q6-mpd", .data = &q6_ipq5018_res_init },
+	{ .compatible = "qcom,ipq9574-q6-mpd", .data = &q6_ipq9574_res_init },
+	{ .compatible = "qcom,ipq5018-wcss-ahb-mpd",
+		.data = &wcss_ahb_ipq5018_res_init },
+	{ .compatible = "qcom,ipq9574-wcss-ahb-mpd",
+		.data = &wcss_ahb_ipq9574_res_init },
+	{ .compatible = "qcom,ipq5018-wcss-pcie-mpd",
+		.data = &wcss_pcie_ipq5018_res_init },
+	{ },
+};
+MODULE_DEVICE_TABLE(of, q6_wcss_of_match);
+
+static struct platform_driver q6_wcss_driver = {
+	.probe = q6_wcss_probe,
+	.remove = q6_wcss_remove,
+	.driver = {
+		.name = "qcom-q6-mpd",
+		.of_match_table = q6_wcss_of_match,
+	},
+};
+module_platform_driver(q6_wcss_driver);
+
+MODULE_DESCRIPTION("Hexagon WCSS Multipd Peripheral Image Loader");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/soc/qcom/mdt_loader.c b/drivers/soc/qcom/mdt_loader.c
index 33dd8c315eb7..ab7f60cceafc 100644
--- a/drivers/soc/qcom/mdt_loader.c
+++ b/drivers/soc/qcom/mdt_loader.c
@@ -16,6 +16,56 @@
 #include <linux/sizes.h>
 #include <linux/slab.h>
 #include <linux/soc/qcom/mdt_loader.h>
+#include <linux/dma-mapping.h>
+
+#include "../../remoteproc/qcom_common.h"
+#define PDSEG_PAS_ID    0xD
+
+/**
+ * struct region - structure passed to TrustZone
+ * @addr:	address of dma region, where dma blocks/chunks address resides
+ * @blk_size:	size of each block
+ */
+struct region {
+	u64 addr;
+	unsigned int blk_size;
+};
+
+/**
+ * struct pdseg_dma_mem_info
+ * @tz_addr:			reference to structure passed to trustzone
+ * @blocks:			no of blocks
+ * @tz_dma:			dma handle of tz_addr
+ * @dma_blk_arr_addr_phys:	dma handle of dma_blk_arr_addr
+ * @dma_blk_arr_addr:		VA of dma array, where each index points to
+ *				dma block PA
+ * @pt:				stores VA of each block
+ *
+ *
+	   ----	     ----	----
+ DMA       |  |	     |  |	|  |
+ blocks	   ----      ----	----
+	    |          |   	  |
+	    |-----|    |   |-------
+		  |    |   |
+ Array of 	--------------
+ pointers	|    |	 |   | dma_blk_arr_addr_phys
+		--------------
+			|
+			|
+ Address referred   tz_addr->addr
+ by trustzone
+ to get dma blocks
+ */
+
+struct pdseg_dma_mem_info {
+	struct region *tz_addr;
+	int blocks;
+	dma_addr_t tz_dma;
+	dma_addr_t dma_blk_arr_addr_phys;
+	u64 *dma_blk_arr_addr;
+	void **pt;
+};

 static bool mdt_phdr_valid(const struct elf32_phdr *phdr)
 {
@@ -358,6 +408,261 @@ static int __qcom_mdt_load(struct device *dev, const struct firmware *fw,
 	return ret;
 }

+static int allocate_dma_mem(struct device *dev,
+			    struct pdseg_dma_mem_info *pd_dma,
+			    int max_size)
+{
+	dma_addr_t dma_tmp = 0;
+	int i;
+
+	pd_dma->blocks = DIV_ROUND_UP(max_size, PAGE_SIZE);
+
+	/* Allocate dma memory for structure passed to trust zone */
+	pd_dma->tz_addr = dma_alloc_coherent(dev, sizeof(struct region),
+					     &pd_dma->tz_dma, GFP_DMA);
+	if (!pd_dma->tz_addr) {
+		pr_err("Error in dma alloc\n");
+		return -ENOMEM;
+	}
+
+	/* Allocate dma memory to store array of blocks PA */
+	pd_dma->dma_blk_arr_addr =
+			dma_alloc_coherent(dev, (pd_dma->blocks * sizeof(u64)),
+					   &pd_dma->dma_blk_arr_addr_phys, GFP_DMA);
+	if (!pd_dma->dma_blk_arr_addr) {
+		pr_err("Error in dma alloc\n");
+		goto free_tz_dma_alloc;
+	}
+
+	/* Assign dma block array PA to trustzone structure addr variable */
+	memcpy(&pd_dma->tz_addr->addr, &pd_dma->dma_blk_arr_addr_phys,
+	       sizeof(dma_addr_t));
+
+	/* Allocate memory to store array of blocks VA */
+	pd_dma->pt = kzalloc(pd_dma->blocks * sizeof(void *), GFP_KERNEL);
+	if (!pd_dma->pt) {
+		pr_err("Error in memory alloc\n");
+		goto free_dma_blk_arr_alloc;
+	}
+
+	for (i = 0; i < pd_dma->blocks; i++) {
+		/* Allocate dma memory for blocks with PAGE_SIZE each */
+		pd_dma->pt[i] = dma_alloc_coherent(dev, PAGE_SIZE,
+						   &dma_tmp, GFP_DMA);
+		if (!pd_dma->pt[i]) {
+			pr_err("Error in dma alloc i:%d - blocks:%d\n", i,
+			       pd_dma->blocks);
+			goto free_mem_alloc;
+		}
+
+		/* Assign dma block PA to dma_blk_arr_addr */
+		memcpy(&pd_dma->dma_blk_arr_addr[i], &dma_tmp,
+		       sizeof(dma_addr_t));
+	}
+	pd_dma->tz_addr->blk_size = PAGE_SIZE;
+	return 0;
+
+free_mem_alloc:
+	i = 0;
+	while (i < pd_dma->blocks && pd_dma->pt[i]) {
+		memcpy(&dma_tmp, &pd_dma->dma_blk_arr_addr[i],
+		       sizeof(dma_addr_t));
+		dma_free_coherent(dev, PAGE_SIZE, pd_dma->pt[i], dma_tmp);
+		i++;
+	}
+	kfree(pd_dma->pt);
+free_dma_blk_arr_alloc:
+	dma_free_coherent(dev, (pd_dma->blocks * sizeof(u64)),
+			  pd_dma->dma_blk_arr_addr,
+			  pd_dma->dma_blk_arr_addr_phys);
+free_tz_dma_alloc:
+	dma_free_coherent(dev, sizeof(struct region), pd_dma->tz_addr,
+			  pd_dma->tz_dma);
+
+	return -ENOMEM;
+}
+
+static void free_dma_mem(struct device *dev, struct pdseg_dma_mem_info *pd_dma)
+{
+	int i;
+	dma_addr_t dma_tmp = 0;
+
+	for (i = 0; i < pd_dma->blocks; i++) {
+		memcpy(&dma_tmp, &pd_dma->dma_blk_arr_addr[i],
+		       sizeof(dma_addr_t));
+		dma_free_coherent(dev, PAGE_SIZE, pd_dma->pt[i],
+				  dma_tmp);
+	}
+
+	dma_free_coherent(dev, (pd_dma->blocks * sizeof(u64)),
+			  pd_dma->dma_blk_arr_addr,
+			  pd_dma->dma_blk_arr_addr_phys);
+
+	dma_free_coherent(dev, sizeof(struct region), pd_dma->tz_addr,
+			  pd_dma->tz_dma);
+	kfree(pd_dma->pt);
+}
+
+static int memcpy_pdseg_to_dma_blk(const char *fw_name, struct device *dev,
+				   int ph_no, struct pdseg_dma_mem_info *pd_dma)
+{
+	const struct firmware *seg_fw;
+	int ret, offset_tmp = 0, tmp = 0;
+	size_t size = 0;
+
+	ret = request_firmware(&seg_fw, fw_name, dev);
+	if (ret) {
+		dev_err(dev, "failed to load %s\n", fw_name);
+		return ret;
+	}
+	size = seg_fw->size < PAGE_SIZE ?
+		seg_fw->size : PAGE_SIZE;
+	while (tmp < pd_dma->blocks && size) {
+		/* Clear last block and copy data */
+		if ((tmp + 1) == pd_dma->blocks)
+			memset_io(pd_dma->pt[tmp], 0, PAGE_SIZE);
+		memcpy_toio(pd_dma->pt[tmp], seg_fw->data + offset_tmp, size);
+		tmp++;
+		offset_tmp += size;
+		if ((seg_fw->size - offset_tmp) < PAGE_SIZE)
+			size = seg_fw->size - offset_tmp;
+	}
+	release_firmware(seg_fw);
+	ret = qti_scm_pdseg_memcpy_v2(PDSEG_PAS_ID, ph_no, pd_dma->tz_dma,
+				      tmp);
+	if (ret) {
+		dev_err(dev, "pd seg memcpy scm failed\n");
+		return ret;
+	}
+	return ret;
+}
+
+static int __qcom_mdt_load_pd_seg(struct device *dev, const struct firmware *fw,
+				  const char *fw_name, int pas_id, void *mem_region,
+				  phys_addr_t mem_phys, size_t mem_size,
+				  phys_addr_t *reloc_base, bool pas_init)
+{
+	const struct elf32_phdr *phdrs;
+	const struct elf32_phdr *phdr;
+	const struct elf32_hdr *ehdr;
+	phys_addr_t mem_reloc;
+	phys_addr_t min_addr = PHYS_ADDR_MAX;
+	ssize_t offset;
+	bool relocate = false;
+	int ret = 0;
+	int i;
+	u8 pd_asid;
+	int max_size = 0;
+	struct pdseg_dma_mem_info pd_dma = {0};
+	char *firmware_name;
+	size_t fw_name_len = strlen(fw_name);
+
+	if (!fw || !mem_region || !mem_phys || !mem_size)
+		return -EINVAL;
+
+	firmware_name = kstrdup(fw_name, GFP_KERNEL);
+	if (!firmware_name)
+		return -ENOMEM;
+
+	pd_asid = qcom_get_pd_asid(dev->of_node);
+
+	ehdr = (struct elf32_hdr *)fw->data;
+	phdrs = (struct elf32_phdr *)(ehdr + 1);
+
+	for (i = 0; i < ehdr->e_phnum; i++) {
+		phdr = &phdrs[i];
+
+		if (!mdt_phdr_valid(phdr))
+			continue;
+		/*
+		 * While doing PD specific reloading, load only that PD
+		 * specific writeable entries. Skip others
+		 */
+		if ((QCOM_MDT_PF_ASID(phdr->p_flags) != pd_asid) ||
+		    ((phdr->p_flags & PF_W) == 0))
+			continue;
+
+		if (phdr->p_flags & QCOM_MDT_RELOCATABLE)
+			relocate = true;
+
+		if (phdr->p_paddr < min_addr)
+			min_addr = phdr->p_paddr;
+
+		if (max_size < phdr->p_memsz)
+			max_size = phdr->p_memsz;
+	}
+
+	/**
+	 * During userpd PIL segments reloading, Q6 is live. Due to
+	 * this we can't access memory region of PIL segments. So
+	 * create DMA chunks/blocks to store PIL segments data.
+	 */
+	ret = allocate_dma_mem(dev, &pd_dma, max_size);
+	if (ret)
+		goto out;
+
+	if (relocate) {
+		/*
+		 * The image is relocatable, so offset each segment based on
+		 * the lowest segment address.
+		 */
+		mem_reloc = min_addr;
+	} else {
+		/*
+		 * Image is not relocatable, so offset each segment based on
+		 * the allocated physical chunk of memory.
+		 */
+		mem_reloc = mem_phys;
+	}
+
+	for (i = 0; i < ehdr->e_phnum; i++) {
+		phdr = &phdrs[i];
+
+		if (!mdt_phdr_valid(phdr))
+			continue;
+
+		/*
+		 * While doing PD specific reloading, load only that PD
+		 * specific writeable entries. Skip others
+		 */
+		if ((QCOM_MDT_PF_ASID(phdr->p_flags) != pd_asid) ||
+		    ((phdr->p_flags & PF_W) == 0))
+			continue;
+
+		offset = phdr->p_paddr - mem_reloc;
+		if (offset < 0 || offset + phdr->p_memsz > mem_size) {
+			dev_err(dev, "segment outside memory range\n");
+			ret = -EINVAL;
+			break;
+		}
+
+		if (phdr->p_filesz > phdr->p_memsz) {
+			dev_err(dev,
+				"refusing to load segment %d with p_filesz > p_memsz\n",
+				i);
+			ret = -EINVAL;
+			break;
+		}
+
+		if (phdr->p_filesz) {
+			snprintf(firmware_name + fw_name_len - 3, 4, "b%02d", i);
+
+			/* copy PIL segments data to dma blocks */
+			ret = memcpy_pdseg_to_dma_blk(firmware_name, dev, i, &pd_dma);
+			if (ret)
+				goto free_dma;
+		}
+	}
+free_dma:
+	free_dma_mem(dev, &pd_dma);
+
+out:
+	if (reloc_base)
+		*reloc_base = mem_reloc;
+
+	return ret;
+}
+
 /**
  * qcom_mdt_load() - load the firmware which header is loaded as fw
  * @dev:	device handle to associate resources with
@@ -410,5 +715,32 @@ int qcom_mdt_load_no_init(struct device *dev, const struct firmware *fw,
 }
 EXPORT_SYMBOL_GPL(qcom_mdt_load_no_init);

+/**
+ * qcom_mdt_load_pd_seg() - load userpd specific PIL segements
+ * @dev:	device handle to associate resources with
+ * @fw:		firmware object for the mdt file
+ * @firmware:	name of the firmware, for construction of segment file names
+ * @pas_id:	PAS identifier
+ * @mem_region:	allocated memory region to load firmware into
+ * @mem_phys:	physical address of allocated memory region
+ * @mem_size:	size of the allocated memory region
+ * @reloc_base:	adjusted physical address after relocation
+ *
+ * Here userpd PIL segements are stitched with rootpd firmware.
+ * This function reloads userpd specific PIL segments during SSR
+ * of userpd.
+ *
+ * Returns 0 on success, negative errno otherwise.
+ */
+int qcom_mdt_load_pd_seg(struct device *dev, const struct firmware *fw,
+			 const char *firmware, int pas_id, void *mem_region,
+			 phys_addr_t mem_phys, size_t mem_size,
+			 phys_addr_t *reloc_base)
+{
+	return __qcom_mdt_load_pd_seg(dev, fw, firmware, pas_id, mem_region, mem_phys,
+				      mem_size, reloc_base, true);
+}
+EXPORT_SYMBOL_GPL(qcom_mdt_load_pd_seg);
+
 MODULE_DESCRIPTION("Firmware parser for Qualcomm MDT format");
 MODULE_LICENSE("GPL v2");
diff --git a/include/linux/soc/qcom/mdt_loader.h b/include/linux/soc/qcom/mdt_loader.h
index 9e8e60421192..57021236dfc9 100644
--- a/include/linux/soc/qcom/mdt_loader.h
+++ b/include/linux/soc/qcom/mdt_loader.h
@@ -7,6 +7,11 @@
 #define QCOM_MDT_TYPE_MASK	(7 << 24)
 #define QCOM_MDT_TYPE_HASH	(2 << 24)
 #define QCOM_MDT_RELOCATABLE	BIT(27)
+#define QCOM_MDT_ASID_MASK      0xfu
+#define QCOM_MDT_PF_ASID_SHIFT  16
+#define QCOM_MDT_PF_ASID_MASK   (QCOM_MDT_ASID_MASK << QCOM_MDT_PF_ASID_SHIFT)
+#define QCOM_MDT_PF_ASID(x)     \
+			(((x) >> QCOM_MDT_PF_ASID_SHIFT) & QCOM_MDT_ASID_MASK)

 struct device;
 struct firmware;
@@ -27,6 +32,10 @@ int qcom_mdt_load_no_init(struct device *dev, const struct firmware *fw,
 			  const char *fw_name, int pas_id, void *mem_region,
 			  phys_addr_t mem_phys, size_t mem_size,
 			  phys_addr_t *reloc_base);
+int qcom_mdt_load_pd_seg(struct device *dev, const struct firmware *fw,
+			 const char *firmware, int pas_id, void *mem_region,
+			 phys_addr_t mem_phys, size_t mem_size,
+			 phys_addr_t *reloc_base);
 void *qcom_mdt_read_metadata(const struct firmware *fw, size_t *data_len,
 			     const char *fw_name, struct device *dev);

@@ -62,6 +71,16 @@ static inline int qcom_mdt_load_no_init(struct device *dev,
 	return -ENODEV;
 }

+static inline int qcom_mdt_load_pd_seg(struct device *dev,
+				       const struct firmware *fw,
+				       const char *fw_name, int pas_id,
+				       void *mem_region, phys_addr_t mem_phys,
+				       size_t mem_size,
+				       phys_addr_t *reloc_base)
+{
+	return -ENODEV;
+}
+
 static inline void *qcom_mdt_read_metadata(const struct firmware *fw,
 					   size_t *data_len, const char *fw_name,
 					   struct device *dev)
--
2.17.1

[PATCH V2 11/13] arm64: dtsi: qcom: ipq5018: enable nodes required for multipd

[Manikanta Mylavarapu]

Enable nodes required for multipd remoteproc bring up

Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
---
Changes in V2:
	- Corrected syntax like alignmnet and kept nodes in sorted order.
	- Covered entire TCSR region.
	- Added 'firmware-name' property.

 arch/arm64/boot/dts/qcom/ipq5018.dtsi | 140 ++++++++++++++++++++++++++
 1 file changed, 140 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/ipq5018.dtsi b/arch/arm64/boot/dts/qcom/ipq5018.dtsi
index 9f13d2dcdfd5..3772d54d89e4 100644
--- a/arch/arm64/boot/dts/qcom/ipq5018.dtsi
+++ b/arch/arm64/boot/dts/qcom/ipq5018.dtsi
@@ -85,6 +85,18 @@
 			reg = <0x0 0x4ac00000 0x0 0x200000>;
 			no-map;
 		};
+
+		q6_region: wcnss@4b000000 {
+			reg = <0x0 0x4b000000 0x0 0x1700000>;
+			no-map;
+		};
+
+		smem@4ab00000 {
+			compatible = "qcom,smem";
+			reg = <0x0 0x4ab00000 0x0 0x100000>;
+			hwlocks = <&tcsr_mutex 0>;
+			no-map;
+		};
 	};

 	soc: soc@0 {
@@ -128,6 +140,12 @@
 			#power-domain-cells = <1>;
 		};

+		tcsr_mutex: hwlock@1905000 {
+			compatible = "qcom,tcsr-mutex";
+			reg = <0x01905000 0x20000>;
+			#hwlock-cells = <1>;
+		};
+
 		sdhc_1: mmc@7804000 {
 			compatible = "qcom,ipq5018-sdhci", "qcom,sdhci-msm-v5";
 			reg = <0x7804000 0x1000>;
@@ -181,6 +199,14 @@
 			};
 		};

+		apcs_glb: mailbox@b111000 {
+			compatible = "qcom,ipq5018-apcs-apps-global",
+				     "qcom,ipq6018-apcs-apps-global";
+			reg = <0x0b111000 0x1000>;
+			#clock-cells = <1>;
+			#mbox-cells = <1>;
+		};
+
 		timer@b120000 {
 			compatible = "arm,armv7-timer-mem";
 			reg = <0x0b120000 0x1000>;
@@ -238,6 +264,96 @@
 				status = "disabled";
 			};
 		};
+
+		q6v5_wcss: remoteproc@cd00000 {
+			compatible = "qcom,ipq5018-q6-mpd";
+			reg = <0x0cd00000 0x4040>;
+			firmware-name = "IPQ5018/q6_fw.mdt",
+					"IPQ5018/m3_fw.mdt",
+					"qcn6122/m3_fw.mdt";
+			interrupts-extended = <&intc GIC_SPI 291 IRQ_TYPE_EDGE_RISING>,
+					      <&wcss_smp2p_in 0 0>,
+					      <&wcss_smp2p_in 1 0>,
+					      <&wcss_smp2p_in 2 0>,
+					      <&wcss_smp2p_in 3 0>;
+			interrupt-names = "wdog",
+					  "fatal",
+					  "ready",
+					  "handover",
+					  "stop-ack";
+
+			qcom,smem-states = <&wcss_smp2p_out 0>,
+					   <&wcss_smp2p_out 1>;
+			qcom,smem-state-names = "shutdown",
+						"stop";
+			memory-region = <&q6_region>;
+
+			glink-edge {
+				interrupts = <GIC_SPI 179 IRQ_TYPE_EDGE_RISING>;
+				label = "rtr";
+				qcom,remote-pid = <1>;
+				mboxes = <&apcs_glb 8>;
+			};
+
+			pd-1 {
+				compatible = "qcom,ipq5018-wcss-ahb-mpd";
+				firmware-name = "IPQ5018/q6_fw.mdt";
+				interrupts-extended = <&wcss_smp2p_in 8 0>,
+						      <&wcss_smp2p_in 9 0>,
+						      <&wcss_smp2p_in 12 0>,
+						      <&wcss_smp2p_in 11 0>;
+				interrupt-names = "fatal",
+						  "ready",
+						  "spawn-ack",
+						  "stop-ack";
+				qcom,smem-states = <&wcss_smp2p_out 8>,
+						   <&wcss_smp2p_out 9>,
+						   <&wcss_smp2p_out 10>;
+				qcom,smem-state-names = "shutdown",
+							"stop",
+							"spawn";
+			};
+
+			pd-2 {
+				compatible = "qcom,ipq5018-wcss-pcie-mpd";
+				interrupts-extended = <&wcss_smp2p_in 16 0>,
+						      <&wcss_smp2p_in 17 0>,
+						      <&wcss_smp2p_in 20 0>,
+						      <&wcss_smp2p_in 19 0>;
+				interrupt-names = "fatal",
+						  "ready",
+						  "spawn-ack",
+						  "stop-ack";
+
+				qcom,smem-states = <&wcss_smp2p_out 16>,
+						   <&wcss_smp2p_out 17>,
+						   <&wcss_smp2p_out 18>;
+				qcom,smem-state-names = "shutdown",
+							"stop",
+							"spawn";
+				status = "disabled";
+			};
+
+			pd-3 {
+				compatible = "qcom,ipq5018-wcss-pcie-mpd";
+				interrupts-extended = <&wcss_smp2p_in 24 0>,
+						      <&wcss_smp2p_in 25 0>,
+						      <&wcss_smp2p_in 28 0>,
+						      <&wcss_smp2p_in 27 0>;
+				interrupt-names = "fatal",
+						  "ready",
+						  "spawn-ack",
+						  "stop-ack";
+
+				qcom,smem-states = <&wcss_smp2p_out 24>,
+						   <&wcss_smp2p_out 25>,
+						   <&wcss_smp2p_out 26>;
+				qcom,smem-state-names = "shutdown",
+							"stop",
+							"spawn";
+				status = "disabled";
+			};
+		};
 	};

 	timer {
@@ -247,4 +363,28 @@
 			     <GIC_PPI 4 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
 			     <GIC_PPI 1 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>;
 	};
+
+	wcss: wcss-smp2p {
+		compatible = "qcom,smp2p";
+		qcom,smem = <435>, <428>;
+
+		interrupt-parent = <&intc>;
+		interrupts = <GIC_SPI 177 IRQ_TYPE_EDGE_RISING>;
+
+		mboxes = <&apcs_glb 9>;
+
+		qcom,local-pid = <0>;
+		qcom,remote-pid = <1>;
+
+		wcss_smp2p_out: master-kernel {
+			qcom,entry-name = "master-kernel";
+			#qcom,smem-state-cells = <1>;
+		};
+
+		wcss_smp2p_in: slave-kernel {
+			qcom,entry-name = "slave-kernel";
+			interrupt-controller;
+			#interrupt-cells = <2>;
+		};
+	};
 };
--
2.17.1

[PATCH V2 12/13] arm64: dts: qcom: ipq5018: Add RDP432-c1 board support

[Manikanta Mylavarapu]

Add initial device tree support for the RDP432-C1 board.

Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
---
Changes in V2:
	- Renamed mp03.5-c1 to RDP432-c1
	- Removed boot-args
	- Resolved dt-binding error's

 arch/arm64/boot/dts/qcom/Makefile             |  1 +
 .../arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts | 49 +++++++++++++++++++
 2 files changed, 50 insertions(+)
 create mode 100644 arch/arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts

diff --git a/arch/arm64/boot/dts/qcom/Makefile b/arch/arm64/boot/dts/qcom/Makefile
index 259bd57c6064..bb9164de75b0 100644
--- a/arch/arm64/boot/dts/qcom/Makefile
+++ b/arch/arm64/boot/dts/qcom/Makefile
@@ -3,6 +3,7 @@ dtb-$(CONFIG_ARCH_QCOM)	+= apq8016-sbc.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= apq8094-sony-xperia-kitakami-karin_windy.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= apq8096-db820c.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= apq8096-ifc6640.dtb
+dtb-$(CONFIG_ARCH_QCOM)	+= ipq5018-rdp432-c1.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= ipq5018-rdp432-c2.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= ipq5332-mi01.2.dtb
 dtb-$(CONFIG_ARCH_QCOM)	+= ipq5332-rdp468.dtb
diff --git a/arch/arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts b/arch/arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts
new file mode 100644
index 000000000000..7fe28d9859b0
--- /dev/null
+++ b/arch/arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts
@@ -0,0 +1,49 @@
+// SPDX-License-Identifier: GPL-2.0+ OR BSD-3-Clause
+/*
+ * IPQ5018 RDP432-C1 board device tree source
+ *
+ * Copyright (c) 2023, The Linux Foundation. All rights reserved.
+ */
+
+/dts-v1/;
+
+#include "ipq5018.dtsi"
+
+/ {
+	model = "Qualcomm Technologies, Inc. IPQ5018/AP-RDP432.1-C1";
+	compatible = "qcom,ipq5018-rdp432-c1", "qcom,ipq5018";
+
+	aliases {
+		serial0 = &blsp1_uart1;
+	};
+
+	chosen {
+		stdout-path = "serial0:115200n8";
+	};
+};
+
+&blsp1_uart1 {
+	pinctrl-0 = <&uart1_pins>;
+	pinctrl-names = "default";
+	status = "okay";
+};
+
+&q6v5_wcss {
+	pd-2 {
+		firmware-name = "IPQ5018/q6_fw.mdt";
+		status = "okay";
+	};
+
+	pd-3 {
+		firmware-name = "IPQ5018/q6_fw.mdt";
+		status = "okay";
+	};
+};
+
+&sleep_clk {
+	clock-frequency = <32000>;
+};
+
+&xo_board_clk {
+	clock-frequency = <24000000>;
+};
--
2.17.1

[PATCH V2 13/13] arm64: dtsi: qcom: ipq9574: Add nodes to bring up multipd

[Manikanta Mylavarapu]

Enable nodes required for multipd remoteproc bring up.

Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
---
Changes in V2:
	- Corrected syntax like alignmnet and kept nodes in sorted order.
	- Added 'firmware-name' property.

 arch/arm64/boot/dts/qcom/ipq9574.dtsi | 118 ++++++++++++++++++++++++++
 1 file changed, 118 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/ipq9574.dtsi b/arch/arm64/boot/dts/qcom/ipq9574.dtsi
index 0e04549c69a5..ff0da53ba05f 100644
--- a/arch/arm64/boot/dts/qcom/ipq9574.dtsi
+++ b/arch/arm64/boot/dts/qcom/ipq9574.dtsi
@@ -160,6 +160,11 @@
 			no-map;
 		};

+		q6_region: wcnss@4ab00000 {
+			reg = <0x0 0x4ab00000 0x0 0x2b00000>;
+			no-map;
+		};
+
 		smem@4aa00000 {
 			compatible = "qcom,smem";
 			reg = <0x0 0x4aa00000 0x0 0x00100000>;
@@ -697,6 +702,95 @@
 			};
 		};

+		q6v5_wcss: remoteproc@cd00000 {
+			compatible = "qcom,ipq9574-q6-mpd";
+			reg = <0x0cd00000 0x4040>;
+			firmware-name = "IPQ9574/q6_fw.mdt",
+					"IPQ9574/m3_fw.mdt";
+			interrupts-extended = <&intc GIC_SPI 325 IRQ_TYPE_EDGE_RISING>,
+					      <&wcss_smp2p_in 0 0>,
+					      <&wcss_smp2p_in 1 0>,
+					      <&wcss_smp2p_in 2 0>,
+					      <&wcss_smp2p_in 3 0>;
+			interrupt-names = "wdog",
+					  "fatal",
+					  "ready",
+					  "handover",
+					  "stop-ack";
+
+			qcom,smem-states = <&wcss_smp2p_out 0>,
+					   <&wcss_smp2p_out 1>;
+			qcom,smem-state-names = "shutdown",
+						"stop";
+			memory-region = <&q6_region>;
+
+			glink-edge {
+				interrupts = <GIC_SPI 321 IRQ_TYPE_EDGE_RISING>;
+				label = "rtr";
+				qcom,remote-pid = <1>;
+				mboxes = <&apcs_glb 8>;
+			};
+
+			pd-1 {
+				compatible = "qcom,ipq9574-wcss-ahb-mpd";
+				firmware-name = "IPQ9574/q6_fw.mdt";
+				interrupts-extended = <&wcss_smp2p_in 8 0>,
+						      <&wcss_smp2p_in 9 0>,
+						      <&wcss_smp2p_in 12 0>,
+						      <&wcss_smp2p_in 11 0>;
+				interrupt-names = "fatal",
+						  "ready",
+						  "spawn-ack",
+						  "stop-ack";
+				qcom,smem-states = <&wcss_smp2p_out 8>,
+						   <&wcss_smp2p_out 9>,
+						   <&wcss_smp2p_out 10>;
+				qcom,smem-state-names = "shutdown",
+							"stop",
+							"spawn";
+			};
+
+			pd-2 {
+				compatible = "qcom,ipq5018-wcss-pcie-mpd";
+				interrupts-extended = <&wcss_smp2p_in 16 0>,
+						      <&wcss_smp2p_in 17 0>,
+						      <&wcss_smp2p_in 20 0>,
+						      <&wcss_smp2p_in 19 0>;
+				interrupt-names = "fatal",
+						  "ready",
+						  "spawn-ack",
+						  "stop-ack";
+
+				qcom,smem-states = <&wcss_smp2p_out 16>,
+						   <&wcss_smp2p_out 17>,
+						   <&wcss_smp2p_out 18>;
+				qcom,smem-state-names = "shutdown",
+							"stop",
+							"spawn";
+				status = "disabled";
+			};
+
+			pd-3 {
+				compatible = "qcom,ipq5018-wcss-pcie-mpd";
+				interrupts-extended = <&wcss_smp2p_in 24 0>,
+						      <&wcss_smp2p_in 25 0>,
+						      <&wcss_smp2p_in 28 0>,
+						      <&wcss_smp2p_in 27 0>;
+				interrupt-names = "fatal",
+						  "ready",
+						  "spawn-ack",
+						  "stop-ack";
+
+				qcom,smem-states = <&wcss_smp2p_out 24>,
+						   <&wcss_smp2p_out 25>,
+						   <&wcss_smp2p_out 26>;
+				qcom,smem-state-names = "shutdown",
+							"stop",
+							"spawn";
+				status = "disabled";
+			};
+		};
+
 		pcie1: pci@10000000 {
 			compatible = "qcom,pcie-ipq9574";
 			reg =  <0x10000000 0xf1d>,
@@ -966,4 +1060,28 @@
 			     <GIC_PPI 4 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
 			     <GIC_PPI 1 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>;
 	};
+
+	wcss: wcss-smp2p {
+		compatible = "qcom,smp2p";
+		qcom,smem = <435>, <428>;
+
+		interrupt-parent = <&intc>;
+		interrupts = <GIC_SPI 322 IRQ_TYPE_EDGE_RISING>;
+
+		mboxes = <&apcs_glb 9>;
+
+		qcom,local-pid = <0>;
+		qcom,remote-pid = <1>;
+
+		wcss_smp2p_out: master-kernel {
+			qcom,entry-name = "master-kernel";
+			#qcom,smem-state-cells = <1>;
+		};
+
+		wcss_smp2p_in: slave-kernel {
+			qcom,entry-name = "slave-kernel";
+			interrupt-controller;
+			#interrupt-cells = <2>;
+		};
+	};
 };
--
2.17.1

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Krzysztof Kozlowski]

On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
> Add new binding document for multipd model remoteproc.
> IPQ5018, IPQ9574 follows multipd model.
> 
> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
> ---
> Changes in V2:
> 	- Fixed all comments and rebased for TOT.
> 	- Changed to BSD-3-Clause based on internal open source team
>           suggestion.
> 	- Added firmware-name.
> 
>  .../bindings/remoteproc/qcom,multipd-pil.yaml | 265 ++++++++++++++++++
>  1 file changed, 265 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
> 
> diff --git a/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml b/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
> new file mode 100644
> index 000000000000..3257f27dc569
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
> @@ -0,0 +1,265 @@
> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
> +%YAML 1.2
> +---
> +$id: http://devicetree.org/schemas/remoteproc/qcom,multipd-pil.yaml#
> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> +
> +title: Qualcomm Multipd Secure Peripheral Image Loader
> +
> +maintainers:
> +  - Bjorn Andersson <andersson@kernel.org>
> +  - Mathieu Poirier <mathieu.poirier@linaro.org>
> +
> +description:
> +  Multipd Peripheral Image Loader loads firmware and boots Q6 pd, WCSS pd
> +  remoteproc's on the Qualcomm IPQ5018, IPQ9574 SoC. Pd means protection
> +  domain. It's similar to process in Linux. Here QDSP6 processor runs each
> +  wifi radio functionality on a separate process. One process can't access
> +  other process resources, so this is termed as PD i.e protection domain.
> +
> +properties:
> +  compatible:
> +    enum:
> +      - qcom,ipq5018-q6-mpd
> +      - qcom,ipq9574-q6-mpd
> +
> +  reg:
> +    maxItems: 1
> +
> +  firmware-name:
> +    $ref: /schemas/types.yaml#/definitions/string-array
> +    description: Firmware name of the Hexagon core
> +
> +  interrupts-extended:
> +    items:
> +      - description: Watchdog interrupt
> +      - description: Fatal interrupt
> +      - description: Ready interrupt
> +      - description: Handover interrupt
> +      - description: Stop acknowledge interrupt


interrupts instead. The same in required:. I replied also to your
comment on your comment. :)

Best regards,
Krzysztof

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Krzysztof Kozlowski]

On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
> Add new binding document for multipd model remoteproc.
> IPQ5018, IPQ9574 follows multipd model.
> 
> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
> ---
> Changes in V2:
> 	- Fixed all comments and rebased for TOT.
> 	- Changed to BSD-3-Clause based on internal open source team
>           suggestion.
> 	- Added firmware-name.
> 
>  .../bindings/remoteproc/qcom,multipd-pil.yaml | 265 ++++++++++++++++++
>  1 file changed, 265 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
> 
> diff --git a/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml b/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
> new file mode 100644
> index 000000000000..3257f27dc569
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
> @@ -0,0 +1,265 @@
> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
> +%YAML 1.2
> +---
> +$id: http://devicetree.org/schemas/remoteproc/qcom,multipd-pil.yaml#
> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> +
> +title: Qualcomm Multipd Secure Peripheral Image Loader
> +
> +maintainers:
> +  - Bjorn Andersson <andersson@kernel.org>
> +  - Mathieu Poirier <mathieu.poirier@linaro.org>
> +
> +description:
> +  Multipd Peripheral Image Loader loads firmware and boots Q6 pd, WCSS pd

... boots Q6 Protection Domain (PD), WCSS PD ...

> +  remoteproc's on the Qualcomm IPQ5018, IPQ9574 SoC.

> Pd means protection
> +  domain. 

so you can skip this sentence as it is explained already.

> It's similar to process in Linux. Here QDSP6 processor runs each
> +  wifi radio functionality on a separate process. One process can't access
> +  other process resources, so this is termed as PD i.e protection domain.
> +
> +properties:
> +  compatible:
> +    enum:
> +      - qcom,ipq5018-q6-mpd
> +      - qcom,ipq9574-q6-mpd
> +
> +  reg:
> +    maxItems: 1
> +
> +  firmware-name:
> +    $ref: /schemas/types.yaml#/definitions/string-array
> +    description: Firmware name of the Hexagon core

No need for ref and description. Instead maxItems.

> +
> +  interrupts-extended:
> +    items:
> +      - description: Watchdog interrupt
> +      - description: Fatal interrupt
> +      - description: Ready interrupt
> +      - description: Handover interrupt
> +      - description: Stop acknowledge interrupt
> +
> +  interrupt-names:
> +    items:
> +      - const: wdog
> +      - const: fatal
> +      - const: ready
> +      - const: handover
> +      - const: stop-ack
> +
> +  qcom,smem-states:
> +    $ref: /schemas/types.yaml#/definitions/phandle-array
> +    description: States used by the AP to signal the remote processor
> +    items:
> +      - description: Shutdown Q6
> +      - description: Stop Q6
> +
> +  qcom,smem-state-names:
> +    description:
> +      Names of the states used by the AP to signal the remote processor
> +    items:
> +      - const: shutdown
> +      - const: stop
> +
> +  memory-region:
> +    items:
> +      - description: Q6 pd reserved region
> +
> +  glink-edge:
> +    $ref: /schemas/remoteproc/qcom,glink-edge.yaml#
> +    description:
> +      Qualcomm G-Link subnode which represents communication edge, channels
> +      and devices related to the Modem.
> +
> +patternProperties:
> +  "^pd-1|pd-2|pd-3":
> +    type: object
> +    description:
> +      In Multipd model, WCSS pd depends on Q6 pd i.e Q6 pd should be up before
> +      WCSS. It can be achieved by keeping wcss pd node as subnode of Q6
> +      device node.

That's not enough. Your description does not say what is this, why you
have two protection domains for same compatible. What's more, it a bit
deviates from hardware description.

> +
> +    properties:
> +      compatible:
> +        enum:
> +          - qcom,ipq5018-wcss-ahb-mpd
> +          - qcom,ipq9574-wcss-ahb-mpd
> +          - qcom,ipq5018-wcss-pcie-mpd

Keep rather alphabetical order (so both 5018 together).

I also do not understand these at all. Why adding bus type to
compatible? This rarely is allowed (unless it is PCIe controller within
soc).

> +
> +      firmware-name:
> +        $ref: /schemas/types.yaml#/definitions/string-array
> +        items:
> +          - description: Firmware name of the Hexagon core

same comments

> +
> +      interrupts-extended:
> +        items:
> +          - description: Fatal interrupt
> +          - description: Ready interrupt
> +          - description: Spawn acknowledge interrupt
> +          - description: Stop acknowledge interrupt

ditto

> +
> +      interrupt-names:
> +        items:
> +          - const: fatal
> +          - const: ready
> +          - const: spawn-ack
> +          - const: stop-ack
> +
> +      qcom,smem-states:
> +        $ref: /schemas/types.yaml#/definitions/phandle-array
> +        description: States used by the AP to signal the remote processor
> +        items:
> +          - description: Shutdown WCSS pd
> +          - description: Stop WCSS pd
> +          - description: Spawn WCSS pd
> +
> +      qcom,smem-state-names:
> +        description:
> +          Names of the states used by the AP to signal the remote processor
> +        items:
> +          - const: shutdown
> +          - const: stop
> +          - const: spawn
> +
> +    required:
> +      - compatible
> +      - firmware-name
> +      - interrupts-extended
> +      - interrupt-names
> +      - qcom,smem-states
> +      - qcom,smem-state-names
> +
> +    unevaluatedProperties: false
> +
> +required:
> +  - compatible
> +  - firmware-name
> +  - reg
> +  - interrupts-extended
> +  - interrupt-names
> +  - qcom,smem-states
> +  - qcom,smem-state-names
> +  - memory-region
> +
> +allOf:
> +  - if:
> +      properties:
> +        compatible:
> +          contains:
> +            enum:
> +              - qcom,ipq5018-q6-mpd
> +    then:
> +      properties:
> +        firmware-name:
> +          items:
> +            - const: IPQ5018/q6_fw.mdt
> +            - const: IPQ5018/m3_fw.mdt
> +            - const: qcn6122/m3_fw.mdt

No, names are not part of bindings. Also paths do not look correct. Open
linux-firmware package and verify these are good...

> +
> +  - if:
> +      properties:
> +        compatible:
> +          contains:
> +            enum:
> +              - qcom,ipq9574-q6-mpd
> +    then:
> +      properties:
> +        firmware-name:
> +          items:
> +            - const: IPQ9574/q6_fw.mdt
> +            - const: IPQ9574/m3_fw.mdt

Drop.

> +
> +unevaluatedProperties: false

This changed... why?


Best regards,
Krzysztof

Re: [PATCH V2 02/13] dt-bindings: mailbox: qcom: Add IPQ5018 APCS compatible

[Krzysztof Kozlowski]

On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
> Add compatible for the Qualcomm IPQ5018 APCS block.
> 
> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
> ---


Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>

Best regards,
Krzysztof

Re: [PATCH V2 03/13] dt-bindings: arm: qcom: Document the Qualcomm rdp432-c1 board

[Krzysztof Kozlowski]

On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
> Document the Qualcomm rdp432-c1 board based on IPQ5018 Soc.
> 
> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
> ---
> Changes in V2:
> 	- Renamed mp03.5-c1 to rdp432-c1
> 
>  Documentation/devicetree/bindings/arm/qcom.yaml | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/Documentation/devicetree/bindings/arm/qcom.yaml b/Documentation/devicetree/bindings/arm/qcom.yaml
> index 33a9a97d7b5b..53ae2b6b2da6 100644
> --- a/Documentation/devicetree/bindings/arm/qcom.yaml
> +++ b/Documentation/devicetree/bindings/arm/qcom.yaml
> @@ -337,6 +337,7 @@ properties:
>        - items:
>            - enum:
>                - qcom,ipq5018-rdp432-c2
> +              - qcom,ipq5018-rdp432-c1

Then keep it before c2, to have it sorted.

Best regards,
Krzysztof

Re: [PATCH V2 04/13] dt-bindings: clock: qcom: gcc-ipq5018: remove q6 clocks macros

[Krzysztof Kozlowski]

On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
> Since Q6 firmware takes care of bring up clocks in multipd
> model, remove bring up clock macros.
> 
> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
> ---
>  include/dt-bindings/clock/qcom,gcc-ipq5018.h | 21 --------------------
>  1 file changed, 21 deletions(-)
> 

I am fine with this if it still compiles... I have doubts about it,
unless of some depedencies (you mentioned three !)... but then it gets
complicated.

Keep patches doing same logical change in same patchset. This dependency
dance in recent submissions is making things tricky and prolonging your
upstreaming process significantly.

Best regards,
Krzysztof

Re: [PATCH V2 06/13] clk: qcom: ipq5018: remove q6 bring up clocks

[Krzysztof Kozlowski]

On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
> Since Q6 firmware takes care of it's bring up clocks
> in multipd model, remove from gcc driver.
> 
> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
> ---
>  drivers/clk/qcom/gcc-ipq5018.c | 414 ---------------------------------
>  1 file changed, 414 deletions(-)
> 

So here is explanation - your patchset is not bisectable.

What's more, I really wonder now why these clocks were added if firmware
handles them?

Best regards,
Krzysztof

Re: [PATCH V2 11/13] arm64: dtsi: qcom: ipq5018: enable nodes required for multipd

[Krzysztof Kozlowski]

On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
> Enable nodes required for multipd remoteproc bring up

Please use subject prefixes matching the subsystem. You can get them for
example with `git log --oneline -- DIRECTORY_OR_FILE` on the directory
your patch is touching.

It is "dts". Not "dtsi".

> 
> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
> ---
> Changes in V2:
> 	- Corrected syntax like alignmnet and kept nodes in sorted order.
> 	- Covered entire TCSR region.
> 	- Added 'firmware-name' property.
> 
>  arch/arm64/boot/dts/qcom/ipq5018.dtsi | 140 ++++++++++++++++++++++++++
>  1 file changed, 140 insertions(+)
> 
> diff --git a/arch/arm64/boot/dts/qcom/ipq5018.dtsi b/arch/arm64/boot/dts/qcom/ipq5018.dtsi
> index 9f13d2dcdfd5..3772d54d89e4 100644
> --- a/arch/arm64/boot/dts/qcom/ipq5018.dtsi
> +++ b/arch/arm64/boot/dts/qcom/ipq5018.dtsi
> @@ -85,6 +85,18 @@
>  			reg = <0x0 0x4ac00000 0x0 0x200000>;
>  			no-map;
>  		};
> +
> +		q6_region: wcnss@4b000000 {
> +			reg = <0x0 0x4b000000 0x0 0x1700000>;
> +			no-map;
> +		};
> +
> +		smem@4ab00000 {
> +			compatible = "qcom,smem";
> +			reg = <0x0 0x4ab00000 0x0 0x100000>;
> +			hwlocks = <&tcsr_mutex 0>;
> +			no-map;
> +		};
>  	};
> 
>  	soc: soc@0 {
> @@ -128,6 +140,12 @@
>  			#power-domain-cells = <1>;
>  		};
> 
> +		tcsr_mutex: hwlock@1905000 {
> +			compatible = "qcom,tcsr-mutex";
> +			reg = <0x01905000 0x20000>;
> +			#hwlock-cells = <1>;
> +		};
> +
>  		sdhc_1: mmc@7804000 {
>  			compatible = "qcom,ipq5018-sdhci", "qcom,sdhci-msm-v5";
>  			reg = <0x7804000 0x1000>;
> @@ -181,6 +199,14 @@
>  			};
>  		};
> 
> +		apcs_glb: mailbox@b111000 {
> +			compatible = "qcom,ipq5018-apcs-apps-global",
> +				     "qcom,ipq6018-apcs-apps-global";
> +			reg = <0x0b111000 0x1000>;
> +			#clock-cells = <1>;
> +			#mbox-cells = <1>;
> +		};
> +
>  		timer@b120000 {
>  			compatible = "arm,armv7-timer-mem";
>  			reg = <0x0b120000 0x1000>;
> @@ -238,6 +264,96 @@
>  				status = "disabled";
>  			};
>  		};
> +
> +		q6v5_wcss: remoteproc@cd00000 {
> +			compatible = "qcom,ipq5018-q6-mpd";
> +			reg = <0x0cd00000 0x4040>;
> +			firmware-name = "IPQ5018/q6_fw.mdt",
> +					"IPQ5018/m3_fw.mdt",
> +					"qcn6122/m3_fw.mdt";
> +			interrupts-extended = <&intc GIC_SPI 291 IRQ_TYPE_EDGE_RISING>,
> +					      <&wcss_smp2p_in 0 0>,
> +					      <&wcss_smp2p_in 1 0>,
> +					      <&wcss_smp2p_in 2 0>,
> +					      <&wcss_smp2p_in 3 0>;
> +			interrupt-names = "wdog",
> +					  "fatal",
> +					  "ready",
> +					  "handover",
> +					  "stop-ack";
> +
> +			qcom,smem-states = <&wcss_smp2p_out 0>,
> +					   <&wcss_smp2p_out 1>;
> +			qcom,smem-state-names = "shutdown",
> +						"stop";
> +			memory-region = <&q6_region>;
> +
> +			glink-edge {
> +				interrupts = <GIC_SPI 179 IRQ_TYPE_EDGE_RISING>;
> +				label = "rtr";
> +				qcom,remote-pid = <1>;
> +				mboxes = <&apcs_glb 8>;
> +			};
> +
> +			pd-1 {
> +				compatible = "qcom,ipq5018-wcss-ahb-mpd";
> +				firmware-name = "IPQ5018/q6_fw.mdt";
> +				interrupts-extended = <&wcss_smp2p_in 8 0>,
> +						      <&wcss_smp2p_in 9 0>,
> +						      <&wcss_smp2p_in 12 0>,
> +						      <&wcss_smp2p_in 11 0>;

What "0" stands for?

> +				interrupt-names = "fatal",
> +						  "ready",
Best regards,
Krzysztof

Re: [PATCH V2 12/13] arm64: dts: qcom: ipq5018: Add RDP432-c1 board support

[Krzysztof Kozlowski]

On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
> Add initial device tree support for the RDP432-C1 board.
> 
> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
> ---
> Changes in V2:
> 	- Renamed mp03.5-c1 to RDP432-c1
> 	- Removed boot-args
> 	- Resolved dt-binding error's
> 
>  arch/arm64/boot/dts/qcom/Makefile             |  1 +
>  .../arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts | 49 +++++++++++++++++++
>  2 files changed, 50 insertions(+)
>  create mode 100644 arch/arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts
> 
> diff --git a/arch/arm64/boot/dts/qcom/Makefile b/arch/arm64/boot/dts/qcom/Makefile
> index 259bd57c6064..bb9164de75b0 100644
> --- a/arch/arm64/boot/dts/qcom/Makefile
> +++ b/arch/arm64/boot/dts/qcom/Makefile
> @@ -3,6 +3,7 @@ dtb-$(CONFIG_ARCH_QCOM)	+= apq8016-sbc.dtb
>  dtb-$(CONFIG_ARCH_QCOM)	+= apq8094-sony-xperia-kitakami-karin_windy.dtb
>  dtb-$(CONFIG_ARCH_QCOM)	+= apq8096-db820c.dtb
>  dtb-$(CONFIG_ARCH_QCOM)	+= apq8096-ifc6640.dtb
> +dtb-$(CONFIG_ARCH_QCOM)	+= ipq5018-rdp432-c1.dtb
>  dtb-$(CONFIG_ARCH_QCOM)	+= ipq5018-rdp432-c2.dtb
>  dtb-$(CONFIG_ARCH_QCOM)	+= ipq5332-mi01.2.dtb
>  dtb-$(CONFIG_ARCH_QCOM)	+= ipq5332-rdp468.dtb
> diff --git a/arch/arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts b/arch/arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts
> new file mode 100644
> index 000000000000..7fe28d9859b0
> --- /dev/null
> +++ b/arch/arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts
> @@ -0,0 +1,49 @@
> +// SPDX-License-Identifier: GPL-2.0+ OR BSD-3-Clause
> +/*
> + * IPQ5018 RDP432-C1 board device tree source
> + *
> + * Copyright (c) 2023, The Linux Foundation. All rights reserved.
> + */
> +
> +/dts-v1/;
> +
> +#include "ipq5018.dtsi"
> +
> +/ {
> +	model = "Qualcomm Technologies, Inc. IPQ5018/AP-RDP432.1-C1";
> +	compatible = "qcom,ipq5018-rdp432-c1", "qcom,ipq5018";
> +
> +	aliases {
> +		serial0 = &blsp1_uart1;
> +	};
> +
> +	chosen {
> +		stdout-path = "serial0:115200n8";
> +	};
> +};
> +
> +&blsp1_uart1 {
> +	pinctrl-0 = <&uart1_pins>;
> +	pinctrl-names = "default";
> +	status = "okay";
> +};
> +
> +&q6v5_wcss {
> +	pd-2 {
> +		firmware-name = "IPQ5018/q6_fw.mdt";

Does not look like correct paths. Please open firmware and look there.
Or open other DTS files and check example. Missing qcom, lowercase,
that's minimum of fixes needed.

Best regards,
Krzysztof

Re: [PATCH V2 04/13] dt-bindings: clock: qcom: gcc-ipq5018: remove q6 clocks macros

[Robert Marko]

On 30. 05. 2023. 13:01, Krzysztof Kozlowski wrote:

> On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
>> Since Q6 firmware takes care of bring up clocks in multipd
>> model, remove bring up clock macros.
>>
>> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
>> ---
>>   include/dt-bindings/clock/qcom,gcc-ipq5018.h | 21 --------------------
>>   1 file changed, 21 deletions(-)
>>
> I am fine with this if it still compiles... I have doubts about it,
> unless of some depedencies (you mentioned three !)... but then it gets
> complicated.
>
> Keep patches doing same logical change in same patchset. This dependency
> dance in recent submissions is making things tricky and prolonging your
> upstreaming process significantly.

Considering that the basic IPQ5018 patchset that this series depends on
has not yet been merged, why not just drop these there instead?

Regards,
Robert

>
> Best regards,
> Krzysztof
>
>
>

Re: [PATCH V2 03/13] dt-bindings: arm: qcom: Document the Qualcomm rdp432-c1 board

[Manikanta Mylavarapu]

On 5/30/2023 4:29 PM, Krzysztof Kozlowski wrote:
> On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
>> Document the Qualcomm rdp432-c1 board based on IPQ5018 Soc.
>>
>> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
>> ---
>> Changes in V2:
>> 	- Renamed mp03.5-c1 to rdp432-c1
>>
>>   Documentation/devicetree/bindings/arm/qcom.yaml | 1 +
>>   1 file changed, 1 insertion(+)
>>
>> diff --git a/Documentation/devicetree/bindings/arm/qcom.yaml b/Documentation/devicetree/bindings/arm/qcom.yaml
>> index 33a9a97d7b5b..53ae2b6b2da6 100644
>> --- a/Documentation/devicetree/bindings/arm/qcom.yaml
>> +++ b/Documentation/devicetree/bindings/arm/qcom.yaml
>> @@ -337,6 +337,7 @@ properties:
>>         - items:
>>             - enum:
>>                 - qcom,ipq5018-rdp432-c2
>> +              - qcom,ipq5018-rdp432-c1
> 
> Then keep it before c2, to have it sorted.
> 
> Best regards,
> Krzysztof
> 

Sure, I will move c1 first.

Thanks & Regards,
Manikanta.

Re: [PATCH V2 11/13] arm64: dtsi: qcom: ipq5018: enable nodes required for multipd

[Manikanta Mylavarapu]

On 5/30/2023 4:35 PM, Krzysztof Kozlowski wrote:
> On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
>> Enable nodes required for multipd remoteproc bring up
> 
> Please use subject prefixes matching the subsystem. You can get them for
> example with `git log --oneline -- DIRECTORY_OR_FILE` on the directory
> your patch is touching.
> 
> It is "dts". Not "dtsi".
> 
Sure, I will update to 'dtsi'.

>>
>> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
>> ---
>> Changes in V2:
>> 	- Corrected syntax like alignmnet and kept nodes in sorted order.
>> 	- Covered entire TCSR region.
>> 	- Added 'firmware-name' property.
>>
>>   arch/arm64/boot/dts/qcom/ipq5018.dtsi | 140 ++++++++++++++++++++++++++
>>   1 file changed, 140 insertions(+)
>>
>> diff --git a/arch/arm64/boot/dts/qcom/ipq5018.dtsi b/arch/arm64/boot/dts/qcom/ipq5018.dtsi
>> index 9f13d2dcdfd5..3772d54d89e4 100644
>> --- a/arch/arm64/boot/dts/qcom/ipq5018.dtsi
>> +++ b/arch/arm64/boot/dts/qcom/ipq5018.dtsi
>> @@ -85,6 +85,18 @@
>>   			reg = <0x0 0x4ac00000 0x0 0x200000>;
>>   			no-map;
>>   		};
>> +
>> +		q6_region: wcnss@4b000000 {
>> +			reg = <0x0 0x4b000000 0x0 0x1700000>;
>> +			no-map;
>> +		};
>> +
>> +		smem@4ab00000 {
>> +			compatible = "qcom,smem";
>> +			reg = <0x0 0x4ab00000 0x0 0x100000>;
>> +			hwlocks = <&tcsr_mutex 0>;
>> +			no-map;
>> +		};
>>   	};
>>
>>   	soc: soc@0 {
>> @@ -128,6 +140,12 @@
>>   			#power-domain-cells = <1>;
>>   		};
>>
>> +		tcsr_mutex: hwlock@1905000 {
>> +			compatible = "qcom,tcsr-mutex";
>> +			reg = <0x01905000 0x20000>;
>> +			#hwlock-cells = <1>;
>> +		};
>> +
>>   		sdhc_1: mmc@7804000 {
>>   			compatible = "qcom,ipq5018-sdhci", "qcom,sdhci-msm-v5";
>>   			reg = <0x7804000 0x1000>;
>> @@ -181,6 +199,14 @@
>>   			};
>>   		};
>>
>> +		apcs_glb: mailbox@b111000 {
>> +			compatible = "qcom,ipq5018-apcs-apps-global",
>> +				     "qcom,ipq6018-apcs-apps-global";
>> +			reg = <0x0b111000 0x1000>;
>> +			#clock-cells = <1>;
>> +			#mbox-cells = <1>;
>> +		};
>> +
>>   		timer@b120000 {
>>   			compatible = "arm,armv7-timer-mem";
>>   			reg = <0x0b120000 0x1000>;
>> @@ -238,6 +264,96 @@
>>   				status = "disabled";
>>   			};
>>   		};
>> +
>> +		q6v5_wcss: remoteproc@cd00000 {
>> +			compatible = "qcom,ipq5018-q6-mpd";
>> +			reg = <0x0cd00000 0x4040>;
>> +			firmware-name = "IPQ5018/q6_fw.mdt",
>> +					"IPQ5018/m3_fw.mdt",
>> +					"qcn6122/m3_fw.mdt";
>> +			interrupts-extended = <&intc GIC_SPI 291 IRQ_TYPE_EDGE_RISING>,
>> +					      <&wcss_smp2p_in 0 0>,
>> +					      <&wcss_smp2p_in 1 0>,
>> +					      <&wcss_smp2p_in 2 0>,
>> +					      <&wcss_smp2p_in 3 0>;
>> +			interrupt-names = "wdog",
>> +					  "fatal",
>> +					  "ready",
>> +					  "handover",
>> +					  "stop-ack";
>> +
>> +			qcom,smem-states = <&wcss_smp2p_out 0>,
>> +					   <&wcss_smp2p_out 1>;
>> +			qcom,smem-state-names = "shutdown",
>> +						"stop";
>> +			memory-region = <&q6_region>;
>> +
>> +			glink-edge {
>> +				interrupts = <GIC_SPI 179 IRQ_TYPE_EDGE_RISING>;
>> +				label = "rtr";
>> +				qcom,remote-pid = <1>;
>> +				mboxes = <&apcs_glb 8>;
>> +			};
>> +
>> +			pd-1 {
>> +				compatible = "qcom,ipq5018-wcss-ahb-mpd";
>> +				firmware-name = "IPQ5018/q6_fw.mdt";
>> +				interrupts-extended = <&wcss_smp2p_in 8 0>,
>> +						      <&wcss_smp2p_in 9 0>,
>> +						      <&wcss_smp2p_in 12 0>,
>> +						      <&wcss_smp2p_in 11 0>;
> 
> What "0" stands for?
> 
0 means IRQ_NONE. These are software interrupts (register write),
so we have configured to 0. I will replace 0 with IRQ_NONE.

Thanks & Regards,
Manikanta.

>> +				interrupt-names = "fatal",
>> +						  "ready",
> Best regards,
> Krzysztof
> 

Re: [PATCH V2 11/13] arm64: dtsi: qcom: ipq5018: enable nodes required for multipd

[Manikanta Mylavarapu]

On 6/5/2023 11:05 AM, Manikanta Mylavarapu wrote:
> 
> 
> On 5/30/2023 4:35 PM, Krzysztof Kozlowski wrote:
>> On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
>>> Enable nodes required for multipd remoteproc bring up
>>
>> Please use subject prefixes matching the subsystem. You can get them for
>> example with `git log --oneline -- DIRECTORY_OR_FILE` on the directory
>> your patch is touching.
>>
>> It is "dts". Not "dtsi".
>>
> Sure, I will update to 'dtsi'.
> 
Sorry please discard my previous reply.
I will update to dts.

>>>
>>> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
>>> ---
>>> Changes in V2:
>>>     - Corrected syntax like alignmnet and kept nodes in sorted order.
>>>     - Covered entire TCSR region.
>>>     - Added 'firmware-name' property.
>>>
>>>   arch/arm64/boot/dts/qcom/ipq5018.dtsi | 140 ++++++++++++++++++++++++++
>>>   1 file changed, 140 insertions(+)
>>>
>>> diff --git a/arch/arm64/boot/dts/qcom/ipq5018.dtsi 
>>> b/arch/arm64/boot/dts/qcom/ipq5018.dtsi
>>> index 9f13d2dcdfd5..3772d54d89e4 100644
>>> --- a/arch/arm64/boot/dts/qcom/ipq5018.dtsi
>>> +++ b/arch/arm64/boot/dts/qcom/ipq5018.dtsi
>>> @@ -85,6 +85,18 @@
>>>               reg = <0x0 0x4ac00000 0x0 0x200000>;
>>>               no-map;
>>>           };
>>> +
>>> +        q6_region: wcnss@4b000000 {
>>> +            reg = <0x0 0x4b000000 0x0 0x1700000>;
>>> +            no-map;
>>> +        };
>>> +
>>> +        smem@4ab00000 {
>>> +            compatible = "qcom,smem";
>>> +            reg = <0x0 0x4ab00000 0x0 0x100000>;
>>> +            hwlocks = <&tcsr_mutex 0>;
>>> +            no-map;
>>> +        };
>>>       };
>>>
>>>       soc: soc@0 {
>>> @@ -128,6 +140,12 @@
>>>               #power-domain-cells = <1>;
>>>           };
>>>
>>> +        tcsr_mutex: hwlock@1905000 {
>>> +            compatible = "qcom,tcsr-mutex";
>>> +            reg = <0x01905000 0x20000>;
>>> +            #hwlock-cells = <1>;
>>> +        };
>>> +
>>>           sdhc_1: mmc@7804000 {
>>>               compatible = "qcom,ipq5018-sdhci", "qcom,sdhci-msm-v5";
>>>               reg = <0x7804000 0x1000>;
>>> @@ -181,6 +199,14 @@
>>>               };
>>>           };
>>>
>>> +        apcs_glb: mailbox@b111000 {
>>> +            compatible = "qcom,ipq5018-apcs-apps-global",
>>> +                     "qcom,ipq6018-apcs-apps-global";
>>> +            reg = <0x0b111000 0x1000>;
>>> +            #clock-cells = <1>;
>>> +            #mbox-cells = <1>;
>>> +        };
>>> +
>>>           timer@b120000 {
>>>               compatible = "arm,armv7-timer-mem";
>>>               reg = <0x0b120000 0x1000>;
>>> @@ -238,6 +264,96 @@
>>>                   status = "disabled";
>>>               };
>>>           };
>>> +
>>> +        q6v5_wcss: remoteproc@cd00000 {
>>> +            compatible = "qcom,ipq5018-q6-mpd";
>>> +            reg = <0x0cd00000 0x4040>;
>>> +            firmware-name = "IPQ5018/q6_fw.mdt",
>>> +                    "IPQ5018/m3_fw.mdt",
>>> +                    "qcn6122/m3_fw.mdt";
>>> +            interrupts-extended = <&intc GIC_SPI 291 
>>> IRQ_TYPE_EDGE_RISING>,
>>> +                          <&wcss_smp2p_in 0 0>,
>>> +                          <&wcss_smp2p_in 1 0>,
>>> +                          <&wcss_smp2p_in 2 0>,
>>> +                          <&wcss_smp2p_in 3 0>;
>>> +            interrupt-names = "wdog",
>>> +                      "fatal",
>>> +                      "ready",
>>> +                      "handover",
>>> +                      "stop-ack";
>>> +
>>> +            qcom,smem-states = <&wcss_smp2p_out 0>,
>>> +                       <&wcss_smp2p_out 1>;
>>> +            qcom,smem-state-names = "shutdown",
>>> +                        "stop";
>>> +            memory-region = <&q6_region>;
>>> +
>>> +            glink-edge {
>>> +                interrupts = <GIC_SPI 179 IRQ_TYPE_EDGE_RISING>;
>>> +                label = "rtr";
>>> +                qcom,remote-pid = <1>;
>>> +                mboxes = <&apcs_glb 8>;
>>> +            };
>>> +
>>> +            pd-1 {
>>> +                compatible = "qcom,ipq5018-wcss-ahb-mpd";
>>> +                firmware-name = "IPQ5018/q6_fw.mdt";
>>> +                interrupts-extended = <&wcss_smp2p_in 8 0>,
>>> +                              <&wcss_smp2p_in 9 0>,
>>> +                              <&wcss_smp2p_in 12 0>,
>>> +                              <&wcss_smp2p_in 11 0>;
>>
>> What "0" stands for?
>>
> 0 means IRQ_NONE. These are software interrupts (register write),
> so we have configured to 0. I will replace 0 with IRQ_NONE.
> 
> Thanks & Regards,
> Manikanta.
> 
>>> +                interrupt-names = "fatal",
>>> +                          "ready",
>> Best regards,
>> Krzysztof
>>

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Manikanta Mylavarapu]

On 5/30/2023 4:16 PM, Krzysztof Kozlowski wrote:
> On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
>> Add new binding document for multipd model remoteproc.
>> IPQ5018, IPQ9574 follows multipd model.
>>
>> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
>> ---
>> Changes in V2:
>> 	- Fixed all comments and rebased for TOT.
>> 	- Changed to BSD-3-Clause based on internal open source team
>>            suggestion.
>> 	- Added firmware-name.
>>
>>   .../bindings/remoteproc/qcom,multipd-pil.yaml | 265 ++++++++++++++++++
>>   1 file changed, 265 insertions(+)
>>   create mode 100644 Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
>>
>> diff --git a/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml b/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
>> new file mode 100644
>> index 000000000000..3257f27dc569
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
>> @@ -0,0 +1,265 @@
>> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
>> +%YAML 1.2
>> +---
>> +$id: http://devicetree.org/schemas/remoteproc/qcom,multipd-pil.yaml#
>> +$schema: http://devicetree.org/meta-schemas/core.yaml#
>> +
>> +title: Qualcomm Multipd Secure Peripheral Image Loader
>> +
>> +maintainers:
>> +  - Bjorn Andersson <andersson@kernel.org>
>> +  - Mathieu Poirier <mathieu.poirier@linaro.org>
>> +
>> +description:
>> +  Multipd Peripheral Image Loader loads firmware and boots Q6 pd, WCSS pd
>> +  remoteproc's on the Qualcomm IPQ5018, IPQ9574 SoC. Pd means protection
>> +  domain. It's similar to process in Linux. Here QDSP6 processor runs each
>> +  wifi radio functionality on a separate process. One process can't access
>> +  other process resources, so this is termed as PD i.e protection domain.
>> +
>> +properties:
>> +  compatible:
>> +    enum:
>> +      - qcom,ipq5018-q6-mpd
>> +      - qcom,ipq9574-q6-mpd
>> +
>> +  reg:
>> +    maxItems: 1
>> +
>> +  firmware-name:
>> +    $ref: /schemas/types.yaml#/definitions/string-array
>> +    description: Firmware name of the Hexagon core
>> +
>> +  interrupts-extended:
>> +    items:
>> +      - description: Watchdog interrupt
>> +      - description: Fatal interrupt
>> +      - description: Ready interrupt
>> +      - description: Handover interrupt
>> +      - description: Stop acknowledge interrupt
> 
> 
> interrupts instead. The same in required:. I replied also to your
> comment on your comment. :)
> 
> Best regards,
> Krzysztof
> 

I will use interrupts instead of interrupts-extended in dt-bindings.

Thanks & Regards,
Manikanta.

Re: [PATCH V2 04/13] dt-bindings: clock: qcom: gcc-ipq5018: remove q6 clocks macros

[Manikanta Mylavarapu]

On 6/2/2023 12:25 AM, Robert Marko wrote:
> On 30. 05. 2023. 13:01, Krzysztof Kozlowski wrote:
> 
>> On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
>>> Since Q6 firmware takes care of bring up clocks in multipd
>>> model, remove bring up clock macros.
>>>
>>> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
>>> ---
>>>   include/dt-bindings/clock/qcom,gcc-ipq5018.h | 21 --------------------
>>>   1 file changed, 21 deletions(-)
>>>
>> I am fine with this if it still compiles... I have doubts about it,
>> unless of some depedencies (you mentioned three !)... but then it gets
>> complicated.
>>
>> Keep patches doing same logical change in same patchset. This dependency
>> dance in recent submissions is making things tricky and prolonging your
>> upstreaming process significantly.
> 
> Considering that the basic IPQ5018 patchset that this series depends on
> has not yet been merged, why not just drop these there instead?
> 
> Regards,
> Robert
> 
Yeah i agree. I will drop this patch in this series.

Thanks & Regards,
Manikanta.

>>
>> Best regards,
>> Krzysztof
>>
>>
>>

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Manikanta Mylavarapu]

On 5/30/2023 4:28 PM, Krzysztof Kozlowski wrote:
> On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
>> Add new binding document for multipd model remoteproc.
>> IPQ5018, IPQ9574 follows multipd model.
>>
>> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
>> ---
>> Changes in V2:
>> 	- Fixed all comments and rebased for TOT.
>> 	- Changed to BSD-3-Clause based on internal open source team
>>            suggestion.
>> 	- Added firmware-name.
>>
>>   .../bindings/remoteproc/qcom,multipd-pil.yaml | 265 ++++++++++++++++++
>>   1 file changed, 265 insertions(+)
>>   create mode 100644 Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
>>
>> diff --git a/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml b/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
>> new file mode 100644
>> index 000000000000..3257f27dc569
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
>> @@ -0,0 +1,265 @@
>> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
>> +%YAML 1.2
>> +---
>> +$id: http://devicetree.org/schemas/remoteproc/qcom,multipd-pil.yaml#
>> +$schema: http://devicetree.org/meta-schemas/core.yaml#
>> +
>> +title: Qualcomm Multipd Secure Peripheral Image Loader
>> +
>> +maintainers:
>> +  - Bjorn Andersson <andersson@kernel.org>
>> +  - Mathieu Poirier <mathieu.poirier@linaro.org>
>> +
>> +description:
>> +  Multipd Peripheral Image Loader loads firmware and boots Q6 pd, WCSS pd
> 
> ... boots Q6 Protection Domain (PD), WCSS PD ...
> 
I will replace 'PD' with 'protection domain' wherever applicable.

>> +  remoteproc's on the Qualcomm IPQ5018, IPQ9574 SoC.
> 
>> Pd means protection
>> +  domain.
> 
> so you can skip this sentence as it is explained already.
> 
Sure. I will skip.

>> It's similar to process in Linux. Here QDSP6 processor runs each
>> +  wifi radio functionality on a separate process. One process can't access
>> +  other process resources, so this is termed as PD i.e protection domain.
>> +
>> +properties:
>> +  compatible:
>> +    enum:
>> +      - qcom,ipq5018-q6-mpd
>> +      - qcom,ipq9574-q6-mpd
>> +
>> +  reg:
>> +    maxItems: 1
>> +
>> +  firmware-name:
>> +    $ref: /schemas/types.yaml#/definitions/string-array
>> +    description: Firmware name of the Hexagon core
> 
> No need for ref and description. Instead maxItems.
> 
Ok. I will remove ref, description and add maxItems.

>> +
>> +  interrupts-extended:
>> +    items:
>> +      - description: Watchdog interrupt
>> +      - description: Fatal interrupt
>> +      - description: Ready interrupt
>> +      - description: Handover interrupt
>> +      - description: Stop acknowledge interrupt
>> +
>> +  interrupt-names:
>> +    items:
>> +      - const: wdog
>> +      - const: fatal
>> +      - const: ready
>> +      - const: handover
>> +      - const: stop-ack
>> +
>> +  qcom,smem-states:
>> +    $ref: /schemas/types.yaml#/definitions/phandle-array
>> +    description: States used by the AP to signal the remote processor
>> +    items:
>> +      - description: Shutdown Q6
>> +      - description: Stop Q6
>> +
>> +  qcom,smem-state-names:
>> +    description:
>> +      Names of the states used by the AP to signal the remote processor
>> +    items:
>> +      - const: shutdown
>> +      - const: stop
>> +
>> +  memory-region:
>> +    items:
>> +      - description: Q6 pd reserved region
>> +
>> +  glink-edge:
>> +    $ref: /schemas/remoteproc/qcom,glink-edge.yaml#
>> +    description:
>> +      Qualcomm G-Link subnode which represents communication edge, channels
>> +      and devices related to the Modem.
>> +
>> +patternProperties:
>> +  "^pd-1|pd-2|pd-3":
>> +    type: object
>> +    description:
>> +      In Multipd model, WCSS pd depends on Q6 pd i.e Q6 pd should be up before
>> +      WCSS. It can be achieved by keeping wcss pd node as subnode of Q6
>> +      device node.
> 
> That's not enough. Your description does not say what is this, why you
> have two protection domains for same compatible. What's more, it a bit
> deviates from hardware description.
> 
>> +
>> +    properties:
>> +      compatible:
>> +        enum:
>> +          - qcom,ipq5018-wcss-ahb-mpd
>> +          - qcom,ipq9574-wcss-ahb-mpd
>> +          - qcom,ipq5018-wcss-pcie-mpd
> 
> Keep rather alphabetical order (so both 5018 together).
> 
Sure, i will update.

> I also do not understand these at all. Why adding bus type to
> compatible? This rarely is allowed (unless it is PCIe controller within
> soc).
> 
>> +
>> +      firmware-name:
>> +        $ref: /schemas/types.yaml#/definitions/string-array
>> +        items:
>> +          - description: Firmware name of the Hexagon core
> 
> same comments
> 
Ok. I will remove ref, description and add maxItems.

>> +
>> +      interrupts-extended:
>> +        items:
>> +          - description: Fatal interrupt
>> +          - description: Ready interrupt
>> +          - description: Spawn acknowledge interrupt
>> +          - description: Stop acknowledge interrupt
> 
> ditto
> 
I will use 'interrupts' here.

Thanks & Regards,
Manikanta.

>> +
>> +      interrupt-names:
>> +        items:
>> +          - const: fatal
>> +          - const: ready
>> +          - const: spawn-ack
>> +          - const: stop-ack
>> +
>> +      qcom,smem-states:
>> +        $ref: /schemas/types.yaml#/definitions/phandle-array
>> +        description: States used by the AP to signal the remote processor
>> +        items:
>> +          - description: Shutdown WCSS pd
>> +          - description: Stop WCSS pd
>> +          - description: Spawn WCSS pd
>> +
>> +      qcom,smem-state-names:
>> +        description:
>> +          Names of the states used by the AP to signal the remote processor
>> +        items:
>> +          - const: shutdown
>> +          - const: stop
>> +          - const: spawn
>> +
>> +    required:
>> +      - compatible
>> +      - firmware-name
>> +      - interrupts-extended
>> +      - interrupt-names
>> +      - qcom,smem-states
>> +      - qcom,smem-state-names
>> +
>> +    unevaluatedProperties: false
>> +
>> +required:
>> +  - compatible
>> +  - firmware-name
>> +  - reg
>> +  - interrupts-extended
>> +  - interrupt-names
>> +  - qcom,smem-states
>> +  - qcom,smem-state-names
>> +  - memory-region
>> +
>> +allOf:
>> +  - if:
>> +      properties:
>> +        compatible:
>> +          contains:
>> +            enum:
>> +              - qcom,ipq5018-q6-mpd
>> +    then:
>> +      properties:
>> +        firmware-name:
>> +          items:
>> +            - const: IPQ5018/q6_fw.mdt
>> +            - const: IPQ5018/m3_fw.mdt
>> +            - const: qcn6122/m3_fw.mdt
> 
> No, names are not part of bindings. Also paths do not look correct. Open
> linux-firmware package and verify these are good...
> 
>> +
>> +  - if:
>> +      properties:
>> +        compatible:
>> +          contains:
>> +            enum:
>> +              - qcom,ipq9574-q6-mpd
>> +    then:
>> +      properties:
>> +        firmware-name:
>> +          items:
>> +            - const: IPQ9574/q6_fw.mdt
>> +            - const: IPQ9574/m3_fw.mdt
> 
> Drop.
> 
>> +
>> +unevaluatedProperties: false
> 
> This changed... why?
> 
> 
> Best regards,
> Krzysztof
> 

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Manikanta Mylavarapu]

On 5/30/2023 4:28 PM, Krzysztof Kozlowski wrote:
> On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
>> Add new binding document for multipd model remoteproc.
>> IPQ5018, IPQ9574 follows multipd model.
>>
>> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
>> ---
>> Changes in V2:
>> 	- Fixed all comments and rebased for TOT.
>> 	- Changed to BSD-3-Clause based on internal open source team
>>            suggestion.
>> 	- Added firmware-name.
>>
>>   .../bindings/remoteproc/qcom,multipd-pil.yaml | 265 ++++++++++++++++++
>>   1 file changed, 265 insertions(+)
>>   create mode 100644 Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
>>
>> diff --git a/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml b/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
>> new file mode 100644
>> index 000000000000..3257f27dc569
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
>> @@ -0,0 +1,265 @@
>> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
>> +%YAML 1.2
>> +---
>> +$id: http://devicetree.org/schemas/remoteproc/qcom,multipd-pil.yaml#
>> +$schema: http://devicetree.org/meta-schemas/core.yaml#
>> +
>> +title: Qualcomm Multipd Secure Peripheral Image Loader
>> +
>> +maintainers:
>> +  - Bjorn Andersson <andersson@kernel.org>
>> +  - Mathieu Poirier <mathieu.poirier@linaro.org>
>> +
>> +description:
>> +  Multipd Peripheral Image Loader loads firmware and boots Q6 pd, WCSS pd
> 
> ... boots Q6 Protection Domain (PD), WCSS PD ...
> 
>> +  remoteproc's on the Qualcomm IPQ5018, IPQ9574 SoC.
> 
>> Pd means protection
>> +  domain.
> 
> so you can skip this sentence as it is explained already.
> 
>> It's similar to process in Linux. Here QDSP6 processor runs each
>> +  wifi radio functionality on a separate process. One process can't access
>> +  other process resources, so this is termed as PD i.e protection domain.
>> +
>> +properties:
>> +  compatible:
>> +    enum:
>> +      - qcom,ipq5018-q6-mpd
>> +      - qcom,ipq9574-q6-mpd
>> +
>> +  reg:
>> +    maxItems: 1
>> +
>> +  firmware-name:
>> +    $ref: /schemas/types.yaml#/definitions/string-array
>> +    description: Firmware name of the Hexagon core
> 
> No need for ref and description. Instead maxItems.
> 
>> +
>> +  interrupts-extended:
>> +    items:
>> +      - description: Watchdog interrupt
>> +      - description: Fatal interrupt
>> +      - description: Ready interrupt
>> +      - description: Handover interrupt
>> +      - description: Stop acknowledge interrupt
>> +
>> +  interrupt-names:
>> +    items:
>> +      - const: wdog
>> +      - const: fatal
>> +      - const: ready
>> +      - const: handover
>> +      - const: stop-ack
>> +
>> +  qcom,smem-states:
>> +    $ref: /schemas/types.yaml#/definitions/phandle-array
>> +    description: States used by the AP to signal the remote processor
>> +    items:
>> +      - description: Shutdown Q6
>> +      - description: Stop Q6
>> +
>> +  qcom,smem-state-names:
>> +    description:
>> +      Names of the states used by the AP to signal the remote processor
>> +    items:
>> +      - const: shutdown
>> +      - const: stop
>> +
>> +  memory-region:
>> +    items:
>> +      - description: Q6 pd reserved region
>> +
>> +  glink-edge:
>> +    $ref: /schemas/remoteproc/qcom,glink-edge.yaml#
>> +    description:
>> +      Qualcomm G-Link subnode which represents communication edge, channels
>> +      and devices related to the Modem.
>> +
>> +patternProperties:
>> +  "^pd-1|pd-2|pd-3":
>> +    type: object
>> +    description:
>> +      In Multipd model, WCSS pd depends on Q6 pd i.e Q6 pd should be up before
>> +      WCSS. It can be achieved by keeping wcss pd node as subnode of Q6
>> +      device node.
> 
> That's not enough. Your description does not say what is this, why you
> have two protection domains for same compatible. What's more, it a bit
> deviates from hardware description.
>
WCSS means 'wireless connectivity sub system', in simple words it's a
wifi radio block.

IPQ5018 SOC has both internal (AHB) wifi radio/WCSS and external (PCIE)
wifi radio/WCSS. In Q6, Root protection domain will provide services to
both internal (AHB) and external (PCIE) wifi radio's protection domain.
So we have two protection domains for IPQ5018, one is for internal(AHB) 
and other is for external(PCIE) wifi radio.

>> +
>> +    properties:
>> +      compatible:
>> +        enum:
>> +          - qcom,ipq5018-wcss-ahb-mpd
>> +          - qcom,ipq9574-wcss-ahb-mpd
>> +          - qcom,ipq5018-wcss-pcie-mpd
> 
> Keep rather alphabetical order (so both 5018 together).
> 
> I also do not understand these at all. Why adding bus type to
> compatible? This rarely is allowed (unless it is PCIe controller within
> soc).
> 
IPQ5018 SOC has in-built PCIE controller. Here QDSP6 will bring up
external(PCIE) and internal (AHB) wifi radio's. To separate AHB, PCIE 
radio's properties, i have added bus type to compatible.

>> +
>> +      firmware-name:
>> +        $ref: /schemas/types.yaml#/definitions/string-array
>> +        items:
>> +          - description: Firmware name of the Hexagon core
> 
> same comments
> 
>> +
>> +      interrupts-extended:
>> +        items:
>> +          - description: Fatal interrupt
>> +          - description: Ready interrupt
>> +          - description: Spawn acknowledge interrupt
>> +          - description: Stop acknowledge interrupt
> 
> ditto
> 
>> +
>> +      interrupt-names:
>> +        items:
>> +          - const: fatal
>> +          - const: ready
>> +          - const: spawn-ack
>> +          - const: stop-ack
>> +
>> +      qcom,smem-states:
>> +        $ref: /schemas/types.yaml#/definitions/phandle-array
>> +        description: States used by the AP to signal the remote processor
>> +        items:
>> +          - description: Shutdown WCSS pd
>> +          - description: Stop WCSS pd
>> +          - description: Spawn WCSS pd
>> +
>> +      qcom,smem-state-names:
>> +        description:
>> +          Names of the states used by the AP to signal the remote processor
>> +        items:
>> +          - const: shutdown
>> +          - const: stop
>> +          - const: spawn
>> +
>> +    required:
>> +      - compatible
>> +      - firmware-name
>> +      - interrupts-extended
>> +      - interrupt-names
>> +      - qcom,smem-states
>> +      - qcom,smem-state-names
>> +
>> +    unevaluatedProperties: false
>> +
>> +required:
>> +  - compatible
>> +  - firmware-name
>> +  - reg
>> +  - interrupts-extended
>> +  - interrupt-names
>> +  - qcom,smem-states
>> +  - qcom,smem-state-names
>> +  - memory-region
>> +
>> +allOf:
>> +  - if:
>> +      properties:
>> +        compatible:
>> +          contains:
>> +            enum:
>> +              - qcom,ipq5018-q6-mpd
>> +    then:
>> +      properties:
>> +        firmware-name:
>> +          items:
>> +            - const: IPQ5018/q6_fw.mdt
>> +            - const: IPQ5018/m3_fw.mdt
>> +            - const: qcn6122/m3_fw.mdt
> 
> No, names are not part of bindings. Also paths do not look correct. Open
> linux-firmware package and verify these are good...
> 
>> +
>> +  - if:
>> +      properties:
>> +        compatible:
>> +          contains:
>> +            enum:
>> +              - qcom,ipq9574-q6-mpd
>> +    then:
>> +      properties:
>> +        firmware-name:
>> +          items:
>> +            - const: IPQ9574/q6_fw.mdt
>> +            - const: IPQ9574/m3_fw.mdt
> 
> Drop.
> 
>> +
>> +unevaluatedProperties: false
> 
> This changed... why?
> 
> 
'unevaluatedProperties' is similar to 'additionalProperties' except
that it recognize properties declared in subschemas as well.

Thanks & Regards,
Manikanta.

> Best regards,
> Krzysztof
> 

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Krzysztof Kozlowski]

On 05/06/2023 14:02, Manikanta Mylavarapu wrote:
>>> +  memory-region:
>>> +    items:
>>> +      - description: Q6 pd reserved region
>>> +
>>> +  glink-edge:
>>> +    $ref: /schemas/remoteproc/qcom,glink-edge.yaml#
>>> +    description:
>>> +      Qualcomm G-Link subnode which represents communication edge, channels
>>> +      and devices related to the Modem.
>>> +
>>> +patternProperties:
>>> +  "^pd-1|pd-2|pd-3":
>>> +    type: object
>>> +    description:
>>> +      In Multipd model, WCSS pd depends on Q6 pd i.e Q6 pd should be up before
>>> +      WCSS. It can be achieved by keeping wcss pd node as subnode of Q6
>>> +      device node.
>>
>> That's not enough. Your description does not say what is this, why you
>> have two protection domains for same compatible. What's more, it a bit
>> deviates from hardware description.
>>
> WCSS means 'wireless connectivity sub system', in simple words it's a
> wifi radio block.
> 
> IPQ5018 SOC has both internal (AHB) wifi radio/WCSS and external (PCIE)
> wifi radio/WCSS. In Q6, Root protection domain will provide services to
> both internal (AHB) and external (PCIE) wifi radio's protection domain.
> So we have two protection domains for IPQ5018, one is for internal(AHB) 
> and other is for external(PCIE) wifi radio.

So it is now in email, but not in the code...
> 
>>> +
>>> +    properties:
>>> +      compatible:
>>> +        enum:
>>> +          - qcom,ipq5018-wcss-ahb-mpd
>>> +          - qcom,ipq9574-wcss-ahb-mpd
>>> +          - qcom,ipq5018-wcss-pcie-mpd
>>
>> Keep rather alphabetical order (so both 5018 together).
>>
>> I also do not understand these at all. Why adding bus type to
>> compatible? This rarely is allowed (unless it is PCIe controller within
>> soc).
>>
> IPQ5018 SOC has in-built PCIE controller. Here QDSP6 will bring up
> external(PCIE) and internal (AHB) wifi radio's. To separate AHB, PCIE 
> radio's properties, i have added bus type to compatible.

It's the same device - WCSS - right? We do not create multiple nodes and
compatibles for the same devices. Bus suffixes are almost never parts of
compatibles.


>>
>> Drop.
>>
>>> +
>>> +unevaluatedProperties: false
>>
>> This changed... why?
>>
>>
> 'unevaluatedProperties' is similar to 'additionalProperties' except
> that it recognize properties declared in subschemas as well.

You don't have to explain me what are unevaluatedProperties or
additionalProperties. Let's assume that I know them. What you should
explain is why you changed it. Where is the reference to other schema?


Best regards,
Krzysztof

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Manikanta Mylavarapu]

On 6/6/2023 11:44 AM, Krzysztof Kozlowski wrote:
> On 05/06/2023 14:02, Manikanta Mylavarapu wrote:
>>>> +  memory-region:
>>>> +    items:
>>>> +      - description: Q6 pd reserved region
>>>> +
>>>> +  glink-edge:
>>>> +    $ref: /schemas/remoteproc/qcom,glink-edge.yaml#
>>>> +    description:
>>>> +      Qualcomm G-Link subnode which represents communication edge, channels
>>>> +      and devices related to the Modem.
>>>> +
>>>> +patternProperties:
>>>> +  "^pd-1|pd-2|pd-3":
>>>> +    type: object
>>>> +    description:
>>>> +      In Multipd model, WCSS pd depends on Q6 pd i.e Q6 pd should be up before
>>>> +      WCSS. It can be achieved by keeping wcss pd node as subnode of Q6
>>>> +      device node.
>>>
>>> That's not enough. Your description does not say what is this, why you
>>> have two protection domains for same compatible. What's more, it a bit
>>> deviates from hardware description.
>>>
>> WCSS means 'wireless connectivity sub system', in simple words it's a
>> wifi radio block.
>>
>> IPQ5018 SOC has both internal (AHB) wifi radio/WCSS and external (PCIE)
>> wifi radio/WCSS. In Q6, Root protection domain will provide services to
>> both internal (AHB) and external (PCIE) wifi radio's protection domain.
>> So we have two protection domains for IPQ5018, one is for internal(AHB)
>> and other is for external(PCIE) wifi radio.
> 
> So it is now in email, but not in the code...
>>
I will add this info, corresponding block diagram in driver code.

>>>> +
>>>> +    properties:
>>>> +      compatible:
>>>> +        enum:
>>>> +          - qcom,ipq5018-wcss-ahb-mpd
>>>> +          - qcom,ipq9574-wcss-ahb-mpd
>>>> +          - qcom,ipq5018-wcss-pcie-mpd
>>>
>>> Keep rather alphabetical order (so both 5018 together).
>>>
>>> I also do not understand these at all. Why adding bus type to
>>> compatible? This rarely is allowed (unless it is PCIe controller within
>>> soc).
>>>
>> IPQ5018 SOC has in-built PCIE controller. Here QDSP6 will bring up
>> external(PCIE) and internal (AHB) wifi radio's. To separate AHB, PCIE
>> radio's properties, i have added bus type to compatible.
> 
> It's the same device - WCSS - right? We do not create multiple nodes and
> compatibles for the same devices. Bus suffixes are almost never parts of
> compatibles.
> 
> 
No it's not the same device. WCSS on inside IPQ5018 and WCSS attached 
via pcie to IPQ5018. Here QDSP6 managing both WCSS's.

So for better clarity i will use attached SOC ID in compatible.
Below are the new compatible's.

- qcom,ipq5018-wcss-mpd //IPQ5018 internal radio
- qcom,ipq9574-wcss-mpd	//IPQ9574 internal radio
- qcom,qcn6122-wcss-mpd //IPQ5018 attached radio

>>>
>>> Drop.
>>>
>>>> +
>>>> +unevaluatedProperties: false
>>>
>>> This changed... why?
>>>
>>>
>> 'unevaluatedProperties' is similar to 'additionalProperties' except
>> that it recognize properties declared in subschemas as well.
> 
> You don't have to explain me what are unevaluatedProperties or
> additionalProperties. Let's assume that I know them. What you should
> explain is why you changed it. Where is the reference to other schema?
> 
I will go with previously used 'additionalProperties' itself and add
'unevaluatedProperties' in glink-edge.

Thanks & Regards,
Manikanta.
> 
> Best regards,
> Krzysztof
> 

Re: [PATCH V2 06/13] clk: qcom: ipq5018: remove q6 bring up clocks

[Manikanta Mylavarapu]

On 5/30/2023 4:33 PM, Krzysztof Kozlowski wrote:
> On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
>> Since Q6 firmware takes care of it's bring up clocks
>> in multipd model, remove from gcc driver.
>>
>> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
>> ---
>>   drivers/clk/qcom/gcc-ipq5018.c | 414 ---------------------------------
>>   1 file changed, 414 deletions(-)
>>
> 
> So here is explanation - your patchset is not bisectable.
> 
> What's more, I really wonder now why these clocks were added if firmware
> handles them?
> 
> Best regards,
> Krzysztof
> 

I will drop this patch now.
It will be taken care in basic IPQ5018 patchset series.

Thanks & Regards,
Manikanta.

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Kalle Valo]

Manikanta Mylavarapu <quic_mmanikan@quicinc.com> writes:

>>>>> +
>>>>> +    properties:
>>>>> +      compatible:
>>>>> +        enum:
>>>>> +          - qcom,ipq5018-wcss-ahb-mpd
>>>>> +          - qcom,ipq9574-wcss-ahb-mpd
>>>>> +          - qcom,ipq5018-wcss-pcie-mpd
>>>>
>>>> Keep rather alphabetical order (so both 5018 together).
>>>>
>>>> I also do not understand these at all. Why adding bus type to
>>>> compatible? This rarely is allowed (unless it is PCIe controller within
>>>> soc).
>>>>
>>> IPQ5018 SOC has in-built PCIE controller. Here QDSP6 will bring up
>>> external(PCIE) and internal (AHB) wifi radio's. To separate AHB, PCIE
>>> radio's properties, i have added bus type to compatible.
>>
>> It's the same device - WCSS - right? We do not create multiple nodes and
>> compatibles for the same devices. Bus suffixes are almost never parts of
>> compatibles.
>
>
> No it's not the same device. WCSS on inside IPQ5018 and WCSS attached
> via pcie to IPQ5018. Here QDSP6 managing both WCSS's.
>
> So for better clarity i will use attached SOC ID in compatible.
> Below are the new compatible's.
>
> - qcom,ipq5018-wcss-mpd //IPQ5018 internal radio
> - qcom,ipq9574-wcss-mpd	//IPQ9574 internal radio
> - qcom,qcn6122-wcss-mpd //IPQ5018 attached radio

What mandates that there's just one QCN6122 device attached to PCI?
Assuming fixed PCI configurations like that makes me worried.

-- 
https://patchwork.kernel.org/project/linux-wireless/list/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Manikanta Mylavarapu]

On 6/6/2023 7:19 PM, Kalle Valo wrote:
> Manikanta Mylavarapu <quic_mmanikan@quicinc.com> writes:
> 
>>>>>> +
>>>>>> +    properties:
>>>>>> +      compatible:
>>>>>> +        enum:
>>>>>> +          - qcom,ipq5018-wcss-ahb-mpd
>>>>>> +          - qcom,ipq9574-wcss-ahb-mpd
>>>>>> +          - qcom,ipq5018-wcss-pcie-mpd
>>>>>
>>>>> Keep rather alphabetical order (so both 5018 together).
>>>>>
>>>>> I also do not understand these at all. Why adding bus type to
>>>>> compatible? This rarely is allowed (unless it is PCIe controller within
>>>>> soc).
>>>>>
>>>> IPQ5018 SOC has in-built PCIE controller. Here QDSP6 will bring up
>>>> external(PCIE) and internal (AHB) wifi radio's. To separate AHB, PCIE
>>>> radio's properties, i have added bus type to compatible.
>>>
>>> It's the same device - WCSS - right? We do not create multiple nodes and
>>> compatibles for the same devices. Bus suffixes are almost never parts of
>>> compatibles.
>>
>>
>> No it's not the same device. WCSS on inside IPQ5018 and WCSS attached
>> via pcie to IPQ5018. Here QDSP6 managing both WCSS's.
>>
>> So for better clarity i will use attached SOC ID in compatible.
>> Below are the new compatible's.
>>
>> - qcom,ipq5018-wcss-mpd //IPQ5018 internal radio
>> - qcom,ipq9574-wcss-mpd	//IPQ9574 internal radio
>> - qcom,qcn6122-wcss-mpd //IPQ5018 attached radio
> 
> What mandates that there's just one QCN6122 device attached to PCI?
> Assuming fixed PCI configurations like that makes me worried.
> 

IPQ5018 always has one internal radio, attached pcie radio's depends on 
no of pcie ports. IPQ5018 has 2 pcie ports, so it supports max two 
qcn6122 devices. One compatible (qcom,qcn6122-wcss-mpd) itself support's 
number of pcie devices controlled by QDSP6.

Thanks & Regards,
Manikanta.

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Krzysztof Kozlowski]

On 07/06/2023 10:10, Manikanta Mylavarapu wrote:
> 
> 
> On 6/6/2023 7:19 PM, Kalle Valo wrote:
>> Manikanta Mylavarapu <quic_mmanikan@quicinc.com> writes:
>>
>>>>>>> +
>>>>>>> +    properties:
>>>>>>> +      compatible:
>>>>>>> +        enum:
>>>>>>> +          - qcom,ipq5018-wcss-ahb-mpd
>>>>>>> +          - qcom,ipq9574-wcss-ahb-mpd
>>>>>>> +          - qcom,ipq5018-wcss-pcie-mpd
>>>>>>
>>>>>> Keep rather alphabetical order (so both 5018 together).
>>>>>>
>>>>>> I also do not understand these at all. Why adding bus type to
>>>>>> compatible? This rarely is allowed (unless it is PCIe controller within
>>>>>> soc).
>>>>>>
>>>>> IPQ5018 SOC has in-built PCIE controller. Here QDSP6 will bring up
>>>>> external(PCIE) and internal (AHB) wifi radio's. To separate AHB, PCIE
>>>>> radio's properties, i have added bus type to compatible.
>>>>
>>>> It's the same device - WCSS - right? We do not create multiple nodes and
>>>> compatibles for the same devices. Bus suffixes are almost never parts of
>>>> compatibles.
>>>
>>>
>>> No it's not the same device. WCSS on inside IPQ5018 and WCSS attached
>>> via pcie to IPQ5018. Here QDSP6 managing both WCSS's.
>>>
>>> So for better clarity i will use attached SOC ID in compatible.
>>> Below are the new compatible's.
>>>
>>> - qcom,ipq5018-wcss-mpd //IPQ5018 internal radio
>>> - qcom,ipq9574-wcss-mpd	//IPQ9574 internal radio
>>> - qcom,qcn6122-wcss-mpd //IPQ5018 attached radio
>>
>> What mandates that there's just one QCN6122 device attached to PCI?
>> Assuming fixed PCI configurations like that makes me worried.
>>
> 
> IPQ5018 always has one internal radio, attached pcie radio's depends on 
> no of pcie ports. IPQ5018 has 2 pcie ports, so it supports max two 
> qcn6122 devices. One compatible (qcom,qcn6122-wcss-mpd) itself support's 
> number of pcie devices controlled by QDSP6.

So this is hot-pluggable (or at least board-pluggable), then should not
be a part of static DTS.

Some concepts of virtual-processes is anyway far away from hardware
description, thus does not fit into DTS. Adding now to the equation PCIe
with variable number of such processes, brings us even further.

This is not a DT property. Remember - DT describes hardware.

Best regards,
Krzysztof

Re: [PATCH V2 12/13] arm64: dts: qcom: ipq5018: Add RDP432-c1 board support

[Manikanta Mylavarapu]

On 5/30/2023 4:36 PM, Krzysztof Kozlowski wrote:
> On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
>> Add initial device tree support for the RDP432-C1 board.
>>
>> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
>> ---
>> Changes in V2:
>> 	- Renamed mp03.5-c1 to RDP432-c1
>> 	- Removed boot-args
>> 	- Resolved dt-binding error's
>>
>>   arch/arm64/boot/dts/qcom/Makefile             |  1 +
>>   .../arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts | 49 +++++++++++++++++++
>>   2 files changed, 50 insertions(+)
>>   create mode 100644 arch/arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts
>>
>> diff --git a/arch/arm64/boot/dts/qcom/Makefile b/arch/arm64/boot/dts/qcom/Makefile
>> index 259bd57c6064..bb9164de75b0 100644
>> --- a/arch/arm64/boot/dts/qcom/Makefile
>> +++ b/arch/arm64/boot/dts/qcom/Makefile
>> @@ -3,6 +3,7 @@ dtb-$(CONFIG_ARCH_QCOM)	+= apq8016-sbc.dtb
>>   dtb-$(CONFIG_ARCH_QCOM)	+= apq8094-sony-xperia-kitakami-karin_windy.dtb
>>   dtb-$(CONFIG_ARCH_QCOM)	+= apq8096-db820c.dtb
>>   dtb-$(CONFIG_ARCH_QCOM)	+= apq8096-ifc6640.dtb
>> +dtb-$(CONFIG_ARCH_QCOM)	+= ipq5018-rdp432-c1.dtb
>>   dtb-$(CONFIG_ARCH_QCOM)	+= ipq5018-rdp432-c2.dtb
>>   dtb-$(CONFIG_ARCH_QCOM)	+= ipq5332-mi01.2.dtb
>>   dtb-$(CONFIG_ARCH_QCOM)	+= ipq5332-rdp468.dtb
>> diff --git a/arch/arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts b/arch/arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts
>> new file mode 100644
>> index 000000000000..7fe28d9859b0
>> --- /dev/null
>> +++ b/arch/arm64/boot/dts/qcom/ipq5018-rdp432-c1.dts
>> @@ -0,0 +1,49 @@
>> +// SPDX-License-Identifier: GPL-2.0+ OR BSD-3-Clause
>> +/*
>> + * IPQ5018 RDP432-C1 board device tree source
>> + *
>> + * Copyright (c) 2023, The Linux Foundation. All rights reserved.
>> + */
>> +
>> +/dts-v1/;
>> +
>> +#include "ipq5018.dtsi"
>> +
>> +/ {
>> +	model = "Qualcomm Technologies, Inc. IPQ5018/AP-RDP432.1-C1";
>> +	compatible = "qcom,ipq5018-rdp432-c1", "qcom,ipq5018";
>> +
>> +	aliases {
>> +		serial0 = &blsp1_uart1;
>> +	};
>> +
>> +	chosen {
>> +		stdout-path = "serial0:115200n8";
>> +	};
>> +};
>> +
>> +&blsp1_uart1 {
>> +	pinctrl-0 = <&uart1_pins>;
>> +	pinctrl-names = "default";
>> +	status = "okay";
>> +};
>> +
>> +&q6v5_wcss {
>> +	pd-2 {
>> +		firmware-name = "IPQ5018/q6_fw.mdt";
> 
> Does not look like correct paths. Please open firmware and look there.
> Or open other DTS files and check example. Missing qcom, lowercase,
> that's minimum of fixes needed.
> 
> Best regards,
> Krzysztof
> 

IPQ5018 firmware files are available here
https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/tree/ath11k/IPQ5018/hw1.0
I will use 'ath11k/IPQ5018/hw1.0' path.

Thanks & Regards,
Manikanta.

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Manikanta Mylavarapu]

On 5/30/2023 4:28 PM, Krzysztof Kozlowski wrote:
> On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
>> Add new binding document for multipd model remoteproc.
>> IPQ5018, IPQ9574 follows multipd model.
>>
>> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
>> ---
>> Changes in V2:
>> 	- Fixed all comments and rebased for TOT.
>> 	- Changed to BSD-3-Clause based on internal open source team
>>            suggestion.
>> 	- Added firmware-name.
>>
>>   .../bindings/remoteproc/qcom,multipd-pil.yaml | 265 ++++++++++++++++++
>>   1 file changed, 265 insertions(+)
>>   create mode 100644 Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
>>
>> diff --git a/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml b/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
>> new file mode 100644
>> index 000000000000..3257f27dc569
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/remoteproc/qcom,multipd-pil.yaml
>> @@ -0,0 +1,265 @@
>> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
>> +%YAML 1.2
>> +---
>> +$id: http://devicetree.org/schemas/remoteproc/qcom,multipd-pil.yaml#
>> +$schema: http://devicetree.org/meta-schemas/core.yaml#
>> +
>> +title: Qualcomm Multipd Secure Peripheral Image Loader
>> +
>> +maintainers:
>> +  - Bjorn Andersson <andersson@kernel.org>
>> +  - Mathieu Poirier <mathieu.poirier@linaro.org>
>> +
>> +description:
>> +  Multipd Peripheral Image Loader loads firmware and boots Q6 pd, WCSS pd
> 
> ... boots Q6 Protection Domain (PD), WCSS PD ...
> 
>> +  remoteproc's on the Qualcomm IPQ5018, IPQ9574 SoC.
> 
>> Pd means protection
>> +  domain.
> 
> so you can skip this sentence as it is explained already.
> 
>> It's similar to process in Linux. Here QDSP6 processor runs each
>> +  wifi radio functionality on a separate process. One process can't access
>> +  other process resources, so this is termed as PD i.e protection domain.
>> +
>> +properties:
>> +  compatible:
>> +    enum:
>> +      - qcom,ipq5018-q6-mpd
>> +      - qcom,ipq9574-q6-mpd
>> +
>> +  reg:
>> +    maxItems: 1
>> +
>> +  firmware-name:
>> +    $ref: /schemas/types.yaml#/definitions/string-array
>> +    description: Firmware name of the Hexagon core
> 
> No need for ref and description. Instead maxItems.
> 
>> +
>> +  interrupts-extended:
>> +    items:
>> +      - description: Watchdog interrupt
>> +      - description: Fatal interrupt
>> +      - description: Ready interrupt
>> +      - description: Handover interrupt
>> +      - description: Stop acknowledge interrupt
>> +
>> +  interrupt-names:
>> +    items:
>> +      - const: wdog
>> +      - const: fatal
>> +      - const: ready
>> +      - const: handover
>> +      - const: stop-ack
>> +
>> +  qcom,smem-states:
>> +    $ref: /schemas/types.yaml#/definitions/phandle-array
>> +    description: States used by the AP to signal the remote processor
>> +    items:
>> +      - description: Shutdown Q6
>> +      - description: Stop Q6
>> +
>> +  qcom,smem-state-names:
>> +    description:
>> +      Names of the states used by the AP to signal the remote processor
>> +    items:
>> +      - const: shutdown
>> +      - const: stop
>> +
>> +  memory-region:
>> +    items:
>> +      - description: Q6 pd reserved region
>> +
>> +  glink-edge:
>> +    $ref: /schemas/remoteproc/qcom,glink-edge.yaml#
>> +    description:
>> +      Qualcomm G-Link subnode which represents communication edge, channels
>> +      and devices related to the Modem.
>> +
>> +patternProperties:
>> +  "^pd-1|pd-2|pd-3":
>> +    type: object
>> +    description:
>> +      In Multipd model, WCSS pd depends on Q6 pd i.e Q6 pd should be up before
>> +      WCSS. It can be achieved by keeping wcss pd node as subnode of Q6
>> +      device node.
> 
> That's not enough. Your description does not say what is this, why you
> have two protection domains for same compatible. What's more, it a bit
> deviates from hardware description.
> 
>> +
>> +    properties:
>> +      compatible:
>> +        enum:
>> +          - qcom,ipq5018-wcss-ahb-mpd
>> +          - qcom,ipq9574-wcss-ahb-mpd
>> +          - qcom,ipq5018-wcss-pcie-mpd
> 
> Keep rather alphabetical order (so both 5018 together).
> 
> I also do not understand these at all. Why adding bus type to
> compatible? This rarely is allowed (unless it is PCIe controller within
> soc).
> 
>> +
>> +      firmware-name:
>> +        $ref: /schemas/types.yaml#/definitions/string-array
>> +        items:
>> +          - description: Firmware name of the Hexagon core
> 
> same comments
> 
>> +
>> +      interrupts-extended:
>> +        items:
>> +          - description: Fatal interrupt
>> +          - description: Ready interrupt
>> +          - description: Spawn acknowledge interrupt
>> +          - description: Stop acknowledge interrupt
> 
> ditto
> 
>> +
>> +      interrupt-names:
>> +        items:
>> +          - const: fatal
>> +          - const: ready
>> +          - const: spawn-ack
>> +          - const: stop-ack
>> +
>> +      qcom,smem-states:
>> +        $ref: /schemas/types.yaml#/definitions/phandle-array
>> +        description: States used by the AP to signal the remote processor
>> +        items:
>> +          - description: Shutdown WCSS pd
>> +          - description: Stop WCSS pd
>> +          - description: Spawn WCSS pd
>> +
>> +      qcom,smem-state-names:
>> +        description:
>> +          Names of the states used by the AP to signal the remote processor
>> +        items:
>> +          - const: shutdown
>> +          - const: stop
>> +          - const: spawn
>> +
>> +    required:
>> +      - compatible
>> +      - firmware-name
>> +      - interrupts-extended
>> +      - interrupt-names
>> +      - qcom,smem-states
>> +      - qcom,smem-state-names
>> +
>> +    unevaluatedProperties: false
>> +
>> +required:
>> +  - compatible
>> +  - firmware-name
>> +  - reg
>> +  - interrupts-extended
>> +  - interrupt-names
>> +  - qcom,smem-states
>> +  - qcom,smem-state-names
>> +  - memory-region
>> +
>> +allOf:
>> +  - if:
>> +      properties:
>> +        compatible:
>> +          contains:
>> +            enum:
>> +              - qcom,ipq5018-q6-mpd
>> +    then:
>> +      properties:
>> +        firmware-name:
>> +          items:
>> +            - const: IPQ5018/q6_fw.mdt
>> +            - const: IPQ5018/m3_fw.mdt
>> +            - const: qcn6122/m3_fw.mdt
> 
> No, names are not part of bindings. Also paths do not look correct. Open
> linux-firmware package and verify these are good...
> 
I will remove names from bindings and use firmware path as per
https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/tree/ath11k/IPQ5018/hw1.0

>> +
>> +  - if:
>> +      properties:
>> +        compatible:
>> +          contains:
>> +            enum:
>> +              - qcom,ipq9574-q6-mpd
>> +    then:
>> +      properties:
>> +        firmware-name:
>> +          items:
>> +            - const: IPQ9574/q6_fw.mdt
>> +            - const: IPQ9574/m3_fw.mdt
> 
> Drop.
> 
I will drop.

Thanks & Regards,
Manikanta.

>> +
>> +unevaluatedProperties: false
> 
> This changed... why?
> 
> 
> Best regards,
> Krzysztof
> 

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Manikanta Mylavarapu]

On 6/7/2023 1:57 PM, Krzysztof Kozlowski wrote:
> On 07/06/2023 10:10, Manikanta Mylavarapu wrote:
>>
>>
>> On 6/6/2023 7:19 PM, Kalle Valo wrote:
>>> Manikanta Mylavarapu <quic_mmanikan@quicinc.com> writes:
>>>
>>>>>>>> +
>>>>>>>> +    properties:
>>>>>>>> +      compatible:
>>>>>>>> +        enum:
>>>>>>>> +          - qcom,ipq5018-wcss-ahb-mpd
>>>>>>>> +          - qcom,ipq9574-wcss-ahb-mpd
>>>>>>>> +          - qcom,ipq5018-wcss-pcie-mpd
>>>>>>>
>>>>>>> Keep rather alphabetical order (so both 5018 together).
>>>>>>>
>>>>>>> I also do not understand these at all. Why adding bus type to
>>>>>>> compatible? This rarely is allowed (unless it is PCIe controller within
>>>>>>> soc).
>>>>>>>
>>>>>> IPQ5018 SOC has in-built PCIE controller. Here QDSP6 will bring up
>>>>>> external(PCIE) and internal (AHB) wifi radio's. To separate AHB, PCIE
>>>>>> radio's properties, i have added bus type to compatible.
>>>>>
>>>>> It's the same device - WCSS - right? We do not create multiple nodes and
>>>>> compatibles for the same devices. Bus suffixes are almost never parts of
>>>>> compatibles.
>>>>
>>>>
>>>> No it's not the same device. WCSS on inside IPQ5018 and WCSS attached
>>>> via pcie to IPQ5018. Here QDSP6 managing both WCSS's.
>>>>
>>>> So for better clarity i will use attached SOC ID in compatible.
>>>> Below are the new compatible's.
>>>>
>>>> - qcom,ipq5018-wcss-mpd //IPQ5018 internal radio
>>>> - qcom,ipq9574-wcss-mpd	//IPQ9574 internal radio
>>>> - qcom,qcn6122-wcss-mpd //IPQ5018 attached radio
>>>
>>> What mandates that there's just one QCN6122 device attached to PCI?
>>> Assuming fixed PCI configurations like that makes me worried.
>>>
>>
>> IPQ5018 always has one internal radio, attached pcie radio's depends on
>> no of pcie ports. IPQ5018 has 2 pcie ports, so it supports max two
>> qcn6122 devices. One compatible (qcom,qcn6122-wcss-mpd) itself support's
>> number of pcie devices controlled by QDSP6.
> 
> So this is hot-pluggable (or at least board-pluggable), then should not
> be a part of static DTS.
> 
> Some concepts of virtual-processes is anyway far away from hardware
> description, thus does not fit into DTS. Adding now to the equation PCIe
> with variable number of such processes, brings us even further.
> 
> This is not a DT property. Remember - DT describes hardware.
> 
> Best regards,
> Krzysztof
> 

In the multipd architecture based Socs, There is one Q6 DSP which runs 
the OS/kernel and there are one or more instances of WCSS radios
(It can be either internal or pcie attached).
These WCSS cores are controlled by the Q6 (Q6 DSP brings wcss radios out 
of reset/ shuts it down etc). Q6 forms the 'root Protection domain' and 
the wcss radios are termed as the 'user protection domain'.
The compatible's that is being added here are to manage the 'root 
domain' and 'user domain'.
Not sure if using the words 'pcie'/'ahb' made it confusing.
So, 'qcom,ipq5018-q6-mpd' and 'qcom,ipq5018-wcss-mpd'.

There will be multiple instances of 'qcom,ipq5018-wcss-mpd' in DT based 
on number of wcss radios connected on that board and only one instance 
of 'qcom,ipq5018-q6-mpd'.

Is this approach ok ?

Thanks & Regards,
Manikanta.

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Manikanta Mylavarapu]

On 6/14/2023 5:13 PM, Manikanta Mylavarapu wrote:
> 
> 
> On 6/7/2023 1:57 PM, Krzysztof Kozlowski wrote:
>> On 07/06/2023 10:10, Manikanta Mylavarapu wrote:
>>>
>>>
>>> On 6/6/2023 7:19 PM, Kalle Valo wrote:
>>>> Manikanta Mylavarapu <quic_mmanikan@quicinc.com> writes:
>>>>
>>>>>>>>> +
>>>>>>>>> +    properties:
>>>>>>>>> +      compatible:
>>>>>>>>> +        enum:
>>>>>>>>> +          - qcom,ipq5018-wcss-ahb-mpd
>>>>>>>>> +          - qcom,ipq9574-wcss-ahb-mpd
>>>>>>>>> +          - qcom,ipq5018-wcss-pcie-mpd
>>>>>>>>
>>>>>>>> Keep rather alphabetical order (so both 5018 together).
>>>>>>>>
>>>>>>>> I also do not understand these at all. Why adding bus type to
>>>>>>>> compatible? This rarely is allowed (unless it is PCIe controller 
>>>>>>>> within
>>>>>>>> soc).
>>>>>>>>
>>>>>>> IPQ5018 SOC has in-built PCIE controller. Here QDSP6 will bring up
>>>>>>> external(PCIE) and internal (AHB) wifi radio's. To separate AHB, 
>>>>>>> PCIE
>>>>>>> radio's properties, i have added bus type to compatible.
>>>>>>
>>>>>> It's the same device - WCSS - right? We do not create multiple 
>>>>>> nodes and
>>>>>> compatibles for the same devices. Bus suffixes are almost never 
>>>>>> parts of
>>>>>> compatibles.
>>>>>
>>>>>
>>>>> No it's not the same device. WCSS on inside IPQ5018 and WCSS attached
>>>>> via pcie to IPQ5018. Here QDSP6 managing both WCSS's.
>>>>>
>>>>> So for better clarity i will use attached SOC ID in compatible.
>>>>> Below are the new compatible's.
>>>>>
>>>>> - qcom,ipq5018-wcss-mpd //IPQ5018 internal radio
>>>>> - qcom,ipq9574-wcss-mpd    //IPQ9574 internal radio
>>>>> - qcom,qcn6122-wcss-mpd //IPQ5018 attached radio
>>>>
>>>> What mandates that there's just one QCN6122 device attached to PCI?
>>>> Assuming fixed PCI configurations like that makes me worried.
>>>>
>>>
>>> IPQ5018 always has one internal radio, attached pcie radio's depends on
>>> no of pcie ports. IPQ5018 has 2 pcie ports, so it supports max two
>>> qcn6122 devices. One compatible (qcom,qcn6122-wcss-mpd) itself support's
>>> number of pcie devices controlled by QDSP6.
>>
>> So this is hot-pluggable (or at least board-pluggable), then should not
>> be a part of static DTS.
>>
>> Some concepts of virtual-processes is anyway far away from hardware
>> description, thus does not fit into DTS. Adding now to the equation PCIe
>> with variable number of such processes, brings us even further.
>>
>> This is not a DT property. Remember - DT describes hardware.
>>
>> Best regards,
>> Krzysztof
>>
> 
> In the multipd architecture based Socs, There is one Q6 DSP which runs 
> the OS/kernel and there are one or more instances of WCSS radios
> (It can be either internal or pcie attached).
> These WCSS cores are controlled by the Q6 (Q6 DSP brings wcss radios out 
> of reset/ shuts it down etc). Q6 forms the 'root Protection domain' and 
> the wcss radios are termed as the 'user protection domain'.
> The compatible's that is being added here are to manage the 'root 
> domain' and 'user domain'.
> Not sure if using the words 'pcie'/'ahb' made it confusing.
> So, 'qcom,ipq5018-q6-mpd' and 'qcom,ipq5018-wcss-mpd'.
> 
> There will be multiple instances of 'qcom,ipq5018-wcss-mpd' in DT based 
> on number of wcss radios connected on that board and only one instance 
> of 'qcom,ipq5018-q6-mpd'.
> 
> Is this approach ok ?
> 
> Thanks & Regards,
> Manikanta.


I didn't aligned previous reply properly. Now i corrected it
and re sending.

In the multipd architecture based Socs, There is one Q6 DSP which
runs the OS/kernel and there are one or more instances of WCSS radios
(It can be either internal or pcie attached).
These WCSS cores are controlled by the Q6 (Q6 DSP brings wcss radios
out of reset/ shuts it down etc).
Q6 forms the 'root Protection domain' and the wcss radios are termed
as the 'user protection domain'. The compatible's that is being
added here are to manage the 'root domain' and 'user domain'.
Not sure if using the words 'pcie'/'ahb' made it confusing.
So, 'qcom,ipq5018-q6-mpd' and 'qcom,ipq5018-wcss-mpd'.

There will be multiple instances of 'qcom,ipq5018-wcss-mpd' in DT
based on number of wcss radios connected on that board and only
one instance of 'qcom,ipq5018-q6-mpd'.

Is this approach ok ?

Thanks & Regards,
Manikanta.

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Krzysztof Kozlowski]

On 14/06/2023 13:43, Manikanta Mylavarapu wrote:
>>>>>>>>> +    properties:
>>>>>>>>> +      compatible:
>>>>>>>>> +        enum:
>>>>>>>>> +          - qcom,ipq5018-wcss-ahb-mpd
>>>>>>>>> +          - qcom,ipq9574-wcss-ahb-mpd
>>>>>>>>> +          - qcom,ipq5018-wcss-pcie-mpd
>>>>>>>>
>>>>>>>> Keep rather alphabetical order (so both 5018 together).
>>>>>>>>
>>>>>>>> I also do not understand these at all. Why adding bus type to
>>>>>>>> compatible? This rarely is allowed (unless it is PCIe controller within
>>>>>>>> soc).
>>>>>>>>
>>>>>>> IPQ5018 SOC has in-built PCIE controller. Here QDSP6 will bring up
>>>>>>> external(PCIE) and internal (AHB) wifi radio's. To separate AHB, PCIE
>>>>>>> radio's properties, i have added bus type to compatible.
>>>>>>
>>>>>> It's the same device - WCSS - right? We do not create multiple nodes and
>>>>>> compatibles for the same devices. Bus suffixes are almost never parts of
>>>>>> compatibles.
>>>>>
>>>>>
>>>>> No it's not the same device. WCSS on inside IPQ5018 and WCSS attached
>>>>> via pcie to IPQ5018. Here QDSP6 managing both WCSS's.
>>>>>
>>>>> So for better clarity i will use attached SOC ID in compatible.
>>>>> Below are the new compatible's.
>>>>>
>>>>> - qcom,ipq5018-wcss-mpd //IPQ5018 internal radio
>>>>> - qcom,ipq9574-wcss-mpd	//IPQ9574 internal radio
>>>>> - qcom,qcn6122-wcss-mpd //IPQ5018 attached radio
>>>>
>>>> What mandates that there's just one QCN6122 device attached to PCI?
>>>> Assuming fixed PCI configurations like that makes me worried.
>>>>
>>>
>>> IPQ5018 always has one internal radio, attached pcie radio's depends on
>>> no of pcie ports. IPQ5018 has 2 pcie ports, so it supports max two
>>> qcn6122 devices. One compatible (qcom,qcn6122-wcss-mpd) itself support's
>>> number of pcie devices controlled by QDSP6.
>>
>> So this is hot-pluggable (or at least board-pluggable), then should not
>> be a part of static DTS.
>>
>> Some concepts of virtual-processes is anyway far away from hardware
>> description, thus does not fit into DTS. Adding now to the equation PCIe
>> with variable number of such processes, brings us even further.
>>
>> This is not a DT property. Remember - DT describes hardware.
>>
>> Best regards,
>> Krzysztof
>>
> 
> In the multipd architecture based Socs, There is one Q6 DSP which runs 
> the OS/kernel and there are one or more instances of WCSS radios
> (It can be either internal or pcie attached).
> These WCSS cores are controlled by the Q6 (Q6 DSP brings wcss radios out 
> of reset/ shuts it down etc). Q6 forms the 'root Protection domain' and 
> the wcss radios are termed as the 'user protection domain'.
> The compatible's that is being added here are to manage the 'root 
> domain' and 'user domain'.
> Not sure if using the words 'pcie'/'ahb' made it confusing.
> So, 'qcom,ipq5018-q6-mpd' and 'qcom,ipq5018-wcss-mpd'.
> 
> There will be multiple instances of 'qcom,ipq5018-wcss-mpd' in DT based 
> on number of wcss radios connected on that board and only one instance 
> of 'qcom,ipq5018-q6-mpd'.
> 

I don't understand why the user protection domains need a specific
compatible. Why do they need compatible at all?

Not mentioning that amount of your domains on Q6 is actually fixed per
SoC and probably should not be in DT at all.

Qualcomm puts so many so weird stuff into DT which is not a hardware
description. I understand that everything is there a firmware, but then
make it discoverable for example...

Best regards,
Krzysztof

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Manikanta Mylavarapu]

On 6/14/2023 7:29 PM, Krzysztof Kozlowski wrote:
> On 14/06/2023 13:43, Manikanta Mylavarapu wrote:
>>>>>>>>>> +    properties:
>>>>>>>>>> +      compatible:
>>>>>>>>>> +        enum:
>>>>>>>>>> +          - qcom,ipq5018-wcss-ahb-mpd
>>>>>>>>>> +          - qcom,ipq9574-wcss-ahb-mpd
>>>>>>>>>> +          - qcom,ipq5018-wcss-pcie-mpd
>>>>>>>>>
>>>>>>>>> Keep rather alphabetical order (so both 5018 together).
>>>>>>>>>
>>>>>>>>> I also do not understand these at all. Why adding bus type to
>>>>>>>>> compatible? This rarely is allowed (unless it is PCIe controller within
>>>>>>>>> soc).
>>>>>>>>>
>>>>>>>> IPQ5018 SOC has in-built PCIE controller. Here QDSP6 will bring up
>>>>>>>> external(PCIE) and internal (AHB) wifi radio's. To separate AHB, PCIE
>>>>>>>> radio's properties, i have added bus type to compatible.
>>>>>>>
>>>>>>> It's the same device - WCSS - right? We do not create multiple nodes and
>>>>>>> compatibles for the same devices. Bus suffixes are almost never parts of
>>>>>>> compatibles.
>>>>>>
>>>>>>
>>>>>> No it's not the same device. WCSS on inside IPQ5018 and WCSS attached
>>>>>> via pcie to IPQ5018. Here QDSP6 managing both WCSS's.
>>>>>>
>>>>>> So for better clarity i will use attached SOC ID in compatible.
>>>>>> Below are the new compatible's.
>>>>>>
>>>>>> - qcom,ipq5018-wcss-mpd //IPQ5018 internal radio
>>>>>> - qcom,ipq9574-wcss-mpd	//IPQ9574 internal radio
>>>>>> - qcom,qcn6122-wcss-mpd //IPQ5018 attached radio
>>>>>
>>>>> What mandates that there's just one QCN6122 device attached to PCI?
>>>>> Assuming fixed PCI configurations like that makes me worried.
>>>>>
>>>>
>>>> IPQ5018 always has one internal radio, attached pcie radio's depends on
>>>> no of pcie ports. IPQ5018 has 2 pcie ports, so it supports max two
>>>> qcn6122 devices. One compatible (qcom,qcn6122-wcss-mpd) itself support's
>>>> number of pcie devices controlled by QDSP6.
>>>
>>> So this is hot-pluggable (or at least board-pluggable), then should not
>>> be a part of static DTS.
>>>
>>> Some concepts of virtual-processes is anyway far away from hardware
>>> description, thus does not fit into DTS. Adding now to the equation PCIe
>>> with variable number of such processes, brings us even further.
>>>
>>> This is not a DT property. Remember - DT describes hardware.
>>>
>>> Best regards,
>>> Krzysztof
>>>
>>
>> In the multipd architecture based Socs, There is one Q6 DSP which runs
>> the OS/kernel and there are one or more instances of WCSS radios
>> (It can be either internal or pcie attached).
>> These WCSS cores are controlled by the Q6 (Q6 DSP brings wcss radios out
>> of reset/ shuts it down etc). Q6 forms the 'root Protection domain' and
>> the wcss radios are termed as the 'user protection domain'.
>> The compatible's that is being added here are to manage the 'root
>> domain' and 'user domain'.
>> Not sure if using the words 'pcie'/'ahb' made it confusing.
>> So, 'qcom,ipq5018-q6-mpd' and 'qcom,ipq5018-wcss-mpd'.
>>
>> There will be multiple instances of 'qcom,ipq5018-wcss-mpd' in DT based
>> on number of wcss radios connected on that board and only one instance
>> of 'qcom,ipq5018-q6-mpd'.
>>
> 
> I don't understand why the user protection domains need a specific
> compatible. Why do they need compatible at all?
> 
> Not mentioning that amount of your domains on Q6 is actually fixed per
> SoC and probably should not be in DT at all.
> 
   root domain is fixed per soc (One Q6 DSP, one per soc)
   user domain(s) are variable (based on number of wcss radios attached)

   The sequence to initialize, bring up, tear down the Q6 and wcss radios
   are completely different. So in order to differentiate them, we will
   need different compatibles. So this is a new rproc driver/architecture
   which has a parent/child topology (Q6 DSP -> Master/parent controls
   the WCSS (child)).

> Qualcomm puts so many so weird stuff into DT which is not a hardware
> description. I understand that everything is there a firmware, but then
> make it discoverable for example...
> 

    Hmm, in order to init/bring up these domains, clks, resets
    and some more register access are required. Q6 FW does not have
    access to all these resources. Also, HLOS rproc driver is needed to
    listen on all the events related to  these domains just like any
    other rproc driver.

Regards,
Manikanta

Re: [PATCH V2 13/13] arm64: dtsi: qcom: ipq9574: Add nodes to bring up multipd

[Krzysztof Kozlowski]

On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
> Enable nodes required for multipd remoteproc bring up.
> 
> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
> ---
> Changes in V2:
> 	- Corrected syntax like alignmnet and kept nodes in sorted order.
> 	- Added 'firmware-name' property.
> 
>  arch/arm64/boot/dts/qcom/ipq9574.dtsi | 118 ++++++++++++++++++++++++++
>  1 file changed, 118 insertions(+)
> 
> diff --git a/arch/arm64/boot/dts/qcom/ipq9574.dtsi b/arch/arm64/boot/dts/qcom/ipq9574.dtsi
> index 0e04549c69a5..ff0da53ba05f 100644
> --- a/arch/arm64/boot/dts/qcom/ipq9574.dtsi
> +++ b/arch/arm64/boot/dts/qcom/ipq9574.dtsi
> @@ -160,6 +160,11 @@
>  			no-map;
>  		};
> 
> +		q6_region: wcnss@4ab00000 {
> +			reg = <0x0 0x4ab00000 0x0 0x2b00000>;
> +			no-map;
> +		};
> +
>  		smem@4aa00000 {
>  			compatible = "qcom,smem";
>  			reg = <0x0 0x4aa00000 0x0 0x00100000>;
> @@ -697,6 +702,95 @@
>  			};
>  		};
> 
> +		q6v5_wcss: remoteproc@cd00000 {
> +			compatible = "qcom,ipq9574-q6-mpd";
> +			reg = <0x0cd00000 0x4040>;
> +			firmware-name = "IPQ9574/q6_fw.mdt",
> +					"IPQ9574/m3_fw.mdt";

Here and...

> +			interrupts-extended = <&intc GIC_SPI 325 IRQ_TYPE_EDGE_RISING>,
> +					      <&wcss_smp2p_in 0 0>,
> +					      <&wcss_smp2p_in 1 0>,
> +					      <&wcss_smp2p_in 2 0>,
> +					      <&wcss_smp2p_in 3 0>;
> +			interrupt-names = "wdog",
> +					  "fatal",
> +					  "ready",
> +					  "handover",
> +					  "stop-ack";
> +
> +			qcom,smem-states = <&wcss_smp2p_out 0>,
> +					   <&wcss_smp2p_out 1>;
> +			qcom,smem-state-names = "shutdown",
> +						"stop";
> +			memory-region = <&q6_region>;
> +
> +			glink-edge {
> +				interrupts = <GIC_SPI 321 IRQ_TYPE_EDGE_RISING>;
> +				label = "rtr";
> +				qcom,remote-pid = <1>;
> +				mboxes = <&apcs_glb 8>;
> +			};
> +
> +			pd-1 {
> +				compatible = "qcom,ipq9574-wcss-ahb-mpd";
> +				firmware-name = "IPQ9574/q6_fw.mdt";

... here - why do you have firmware in both places?

> +				interrupts-extended = <&wcss_smp2p_in 8 0>,
> +						      <&wcss_smp2p_in 9 0>,
> +						      <&wcss_smp2p_in 12 0>,
> +						      <&wcss_smp2p_in 11 0>;
> +				interrupt-names = "fatal",
> +						  "ready",
> +						  "spawn-ack",
> +						  "stop-ack";
> +				qcom,smem-states = <&wcss_smp2p_out 8>,
> +						   <&wcss_smp2p_out 9>,
> +						   <&wcss_smp2p_out 10>;
> +				qcom,smem-state-names = "shutdown",
> +							"stop",
> +							"spawn";
> +			};
> +
> +			pd-2 {
> +				compatible = "qcom,ipq5018-wcss-pcie-mpd";

This compatible is confusing for this device.

Best regards,
Krzysztof

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Krzysztof Kozlowski]

On 21/06/2023 13:39, Manikanta Mylavarapu wrote:
>>> on number of wcss radios connected on that board and only one instance
>>> of 'qcom,ipq5018-q6-mpd'.
>>>
>>
>> I don't understand why the user protection domains need a specific
>> compatible. Why do they need compatible at all?
>>
>> Not mentioning that amount of your domains on Q6 is actually fixed per
>> SoC and probably should not be in DT at all.
>>
>    root domain is fixed per soc (One Q6 DSP, one per soc)
>    user domain(s) are variable (based on number of wcss radios attached)
> 
>    The sequence to initialize, bring up, tear down the Q6 and wcss radios
>    are completely different. So in order to differentiate them, we will
>    need different compatibles. So this is a new rproc driver/architecture
>    which has a parent/child topology (Q6 DSP -> Master/parent controls
>    the WCSS (child)).

I understand you need different properties, but I don't see yet the
benefit of creating here artificial compatibles. Look at your ipq9574
DTSI change - it does not use even ipq9574 compatibles for 66% of its
children.

Maybe you should have there just property describing type of device or
bringup?

Given SoC cannot come with different amount of children (PD) and
different amount of radios. You even fix the firmware name, so
boards/customers cannot use anything else. It's fixed and the only
variable element here is disabling some of the blocks per board, if they
do not have physical interface (e.g. radio).

You even hard-code the number of PD by using "pd-[123]", without unit
address, so you do not expect it will grow.

Unless you want to say that these are devices? But your binding does not
really suggest it...


Best regards,
Krzysztof

Re: [PATCH V2 10/13] remoteproc: qcom: Add Hexagon based multipd rproc driver

[Krzysztof Kozlowski]

On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
> It adds support to bring up remoteproc's on multipd model.
> Pd means protection domain. It's similar to process in Linux.
> Here QDSP6 processor runs each wifi radio functionality on a
> separate process. One process can't access other process
> resources, so this is termed as PD i.e protection domain.
> 
> Here we have two pd's called root and user pd. We can correlate
> Root pd as root and user pd as user in linux. Root pd has more
> privileges than user pd. Root will provide services to user pd.
> 
> From remoteproc driver perspective, root pd corresponds to QDSP6
> processor bring up and user pd corresponds to Wifi radio (WCSS)
> bring up.
> 
> Here WCSS(user) PD is dependent on Q6(root) PD, so first
> q6 pd should be up before wcss pd. After wcss pd goes down,
> q6 pd should be turned off.
> 
> IPQ5018, IPQ9574 supports multipd remoteproc driver.
> 
> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
> ---
> Changes in V2:
> 	- Common functionalities moved to seperate patches
> 	- qcom_get_pd_asid() moved to mpd driver
> 	- Last DMA block alone memset to zero
> 	- Added diagram to show how userpd data is organized and sent to
> 	  trustzone
> 	- Rewritten commit message since most of the content available
> 	  in cover page
> 	- Removed 'remote_id' becuase it's not required for bring up.
> 
>  drivers/remoteproc/Kconfig          |  20 +
>  drivers/remoteproc/Makefile         |   1 +
>  drivers/remoteproc/qcom_common.h    |   9 +
>  drivers/remoteproc/qcom_q6v5_mpd.c  | 677 ++++++++++++++++++++++++++++
>  drivers/soc/qcom/mdt_loader.c       | 332 ++++++++++++++
>  include/linux/soc/qcom/mdt_loader.h |  19 +
>  6 files changed, 1058 insertions(+)
>  create mode 100644 drivers/remoteproc/qcom_q6v5_mpd.c
> 
> diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig
> index a850e9f486dd..44af5c36f67e 100644
> --- a/drivers/remoteproc/Kconfig
> +++ b/drivers/remoteproc/Kconfig
> @@ -234,6 +234,26 @@ config QCOM_Q6V5_PAS
>  	  CDSP (Compute DSP), MPSS (Modem Peripheral SubSystem), and
>  	  SLPI (Sensor Low Power Island).
> 
> +config QCOM_Q6V5_MPD
> +	tristate "Qualcomm Hexagon based MPD model Peripheral Image Loader"
> +	depends on OF && ARCH_QCOM
> +	depends on QCOM_SMEM
> +	depends on RPMSG_QCOM_SMD || RPMSG_QCOM_SMD=n
> +	depends on RPMSG_QCOM_GLINK_SMEM || RPMSG_QCOM_GLINK_SMEM=n
> +	depends on QCOM_SYSMON || QCOM_SYSMON=n
> +	depends on RPMSG_QCOM_GLINK || RPMSG_QCOM_GLINK=n
> +	depends on QCOM_AOSS_QMP || QCOM_AOSS_QMP=n
> +	select MFD_SYSCON

Do you really need this?

> +	select QCOM_MDT_LOADER
> +	select QCOM_PIL_INFO
> +	select QCOM_Q6V5_COMMON
> +	select QCOM_RPROC_COMMON
> +	select QCOM_SCM
> +	help
> +	  Say y here to support the Qualcomm Secure Peripheral Image Loader
> +	  for the Hexagon based MultiPD model remote processors on e.g. IPQ5018.
> +	  This is trustZone wireless subsystem.
> +

...

> +	int (*powerup_scm)(u32 peripheral);
> +	int (*powerdown_scm)(u32 peripheral);
> +};
> +
> +/**
> + * qcom_get_pd_asid() - get the pd asid number from DT node

From node name? NAK. It does not work like that. Node names can change
and you did not define this number as part of ABI.

Probably you wanted unit address.

> + * @node:	device tree node
> + *
> + * Returns asid if node name has 'pd' string
> + */
> +s8 qcom_get_pd_asid(struct device_node *node)
> +{
> +	char *str;
> +	s8 pd_asid;
> +
> +	if (!node)
> +		return -EINVAL;
> +
> +	str = strstr(node->name, "pd");
> +	if (!str)
> +		return 0;
> +
> +	str += strlen("pd") + 1;
> +	return kstrtos8(str, 10, &pd_asid) ? -EINVAL : pd_asid;
> +}
> +EXPORT_SYMBOL(qcom_get_pd_asid);
> +

...

> +
> +static int q6_wcss_spawn_pd(struct rproc *rproc)
> +{
> +	int ret;
> +	struct q6_wcss *wcss = rproc->priv;
> +
> +	ret = qcom_q6v5_request_spawn(&wcss->q6);
> +	if (ret == -ETIMEDOUT) {
> +		pr_err("%s spawn timedout\n", rproc->name);

dev_err

> +		return ret;
> +	}
> +
> +	ret = qcom_q6v5_wait_for_start(&wcss->q6, msecs_to_jiffies(10000));
> +	if (ret == -ETIMEDOUT) {
> +		pr_err("%s start timedout\n", rproc->name);

dev_err

> +		wcss->q6.running = false;
> +		return ret;
> +	}
> +	wcss->q6.running = true;
> +	return ret;
> +}
> +
> +static int wcss_ahb_pcie_pd_start(struct rproc *rproc)
> +{
> +	struct q6_wcss *wcss = rproc->priv;
> +	const struct wcss_data *desc = wcss->desc;
> +	int ret;
> +
> +	if (!desc->reset_seq)
> +		return 0;
> +
> +	if (desc->powerup_scm) {
> +		ret = desc->powerup_scm(desc->pasid);
> +		if (ret) {
> +			dev_err(wcss->dev, "failed to power up pd\n");
> +			return ret;
> +		}
> +	}
> +
> +	ret = q6_wcss_spawn_pd(rproc);
> +	if (ret)
> +		return ret;
> +
> +	wcss->state = WCSS_NORMAL;
> +	return ret;
> +}
> +
> +static int q6_wcss_stop(struct rproc *rproc)
> +{
> +	struct q6_wcss *wcss = rproc->priv;
> +	const struct wcss_data *desc = wcss->desc;
> +	int ret;
> +
> +	ret = qcom_scm_pas_shutdown(desc->pasid);
> +	if (ret) {
> +		dev_err(wcss->dev, "not able to shutdown\n");
> +		return ret;
> +	}
> +	qcom_q6v5_unprepare(&wcss->q6);
> +
> +	return 0;
> +}
> +
> +static int wcss_ahb_pcie_pd_stop(struct rproc *rproc)
> +{
> +	struct q6_wcss *wcss = rproc->priv;
> +	struct rproc *rpd_rproc = dev_get_drvdata(wcss->dev->parent);
> +	const struct wcss_data *desc = wcss->desc;
> +	int ret;
> +
> +	if (!desc->reset_seq)
> +		goto shut_down_rpd;
> +
> +	if (rproc->state != RPROC_CRASHED && wcss->q6.stop_bit) {
> +		ret = qcom_q6v5_request_stop(&wcss->q6, NULL);
> +		if (ret) {
> +			dev_err(&rproc->dev, "pd not stopped\n");
> +			return ret;
> +		}
> +	}
> +
> +	if (desc->powerdown_scm) {
> +		ret = desc->powerdown_scm(desc->pasid);
> +		if (ret) {
> +			dev_err(wcss->dev, "failed to power down pd\n");
> +			return ret;
> +		}
> +	}
> +
> +shut_down_rpd:
> +	rproc_shutdown(rpd_rproc);
> +
> +	wcss->state = WCSS_SHUTDOWN;
> +	return 0;
> +}
> +
> +static void *q6_wcss_da_to_va(struct rproc *rproc, u64 da, size_t len,
> +			      bool *is_iomem)
> +{
> +	struct q6_wcss *wcss = rproc->priv;
> +	int offset;
> +
> +	offset = da - wcss->mem_reloc;
> +	if (offset < 0 || offset + len > wcss->mem_size)
> +		return NULL;
> +
> +	return wcss->mem_region + offset;
> +}
> +
> +static int q6_wcss_load(struct rproc *rproc, const struct firmware *fw)
> +{
> +	struct q6_wcss *wcss = rproc->priv;
> +	const struct firmware *fw_hdl;
> +	int ret;
> +	const struct wcss_data *desc = wcss->desc;
> +	int loop;
> +
> +	ret = qcom_mdt_load(wcss->dev, fw, rproc->firmware,
> +			    desc->pasid, wcss->mem_region,
> +			    wcss->mem_phys, wcss->mem_size,
> +			    &wcss->mem_reloc);
> +	if (ret)
> +		return ret;
> +
> +	for (loop = 1; loop < MAX_FIRMWARE; loop++) {
> +		if (!wcss->firmware[loop])
> +			continue;
> +
> +		ret = request_firmware(&fw_hdl, wcss->firmware[loop],
> +				       wcss->dev);
> +		if (ret)
> +			continue;
> +
> +		ret = qcom_mdt_load_no_init(wcss->dev, fw_hdl,
> +					    wcss->firmware[loop], 0,
> +					    wcss->mem_region,
> +					    wcss->mem_phys,
> +					    wcss->mem_size,
> +					    &wcss->mem_reloc);
> +
> +		release_firmware(fw_hdl);
> +
> +		if (ret) {
> +			dev_err(wcss->dev,
> +				"can't load %s ret:%d\n", wcss->firmware[loop], ret);
> +			return ret;
> +		}
> +	}
> +	return 0;
> +}
> +
> +/* This function load's userpd firmware. Since Userpd depends on rootpd

Use Linux coding style comments.

> + * first bring up root pd and then load. User pd firmware load is required
> + * only during user pd restart because root pd loads user pd FW pil segments
> + * during it's bringup.
> + */
> +static int wcss_ahb_pcie_pd_load(struct rproc *rproc, const struct firmware *fw)
> +{
> +	struct q6_wcss *wcss = rproc->priv, *wcss_rpd;
> +	struct rproc *rpd_rproc = dev_get_drvdata(wcss->dev->parent);
> +	const struct wcss_data *desc = wcss->desc;
> +	int ret;
> +
> +	wcss_rpd = rpd_rproc->priv;
> +
> +	/* Boot rootpd rproc */

That's not helpful - i tcopies the function name.

> +	ret = rproc_boot(rpd_rproc);
> +	if (ret || wcss->state == WCSS_NORMAL)
> +		return ret;
> +
> +	return desc->mdt_load_sec(wcss->dev, fw, rproc->firmware,
> +				  desc->pasid, wcss->mem_region,
> +				  wcss->mem_phys, wcss->mem_size,
> +				  &wcss->mem_reloc);
> +}
> +

...

> +
> +static int q6_get_inbound_irq(struct qcom_q6v5 *q6,
> +			      struct platform_device *pdev,
> +			      const char *int_name,
> +			      irqreturn_t (*handler)(int irq, void *data))
> +{
> +	int ret, irq;
> +	char *interrupt, *tmp = (char *)int_name;
> +	struct q6_wcss *wcss = q6->rproc->priv;
> +
> +	irq = platform_get_irq_byname(pdev, int_name);
> +	if (irq < 0) {
> +		if (irq != -EPROBE_DEFER)
> +			dev_err(&pdev->dev,
> +				"failed to retrieve %s IRQ: %d\n",
> +					int_name, irq);

return dev_err_probe

> +		return irq;
> +	}
> +
> +	if (!strcmp(int_name, "fatal")) {
> +		q6->fatal_irq = irq;
> +	} else if (!strcmp(int_name, "stop-ack")) {
> +		q6->stop_irq = irq;
> +		tmp = "stop_ack";
> +	} else if (!strcmp(int_name, "ready")) {
> +		q6->ready_irq = irq;
> +	} else if (!strcmp(int_name, "handover")) {
> +		q6->handover_irq  = irq;
> +	} else if (!strcmp(int_name, "spawn-ack")) {
> +		q6->spawn_irq = irq;
> +		tmp = "spawn_ack";
> +	} else {
> +		dev_err(&pdev->dev, "unknown interrupt\n");
> +		return -EINVAL;
> +	}

This is over complicated method of getting interrupts. First, you can
pass here address of variable with interrupt to assign (*irq_field =
irq). Second, drop the names and get by index. Entries are fixed, not
flexible.

> +
> +	interrupt = devm_kzalloc(&pdev->dev, BUF_SIZE, GFP_KERNEL);
> +	if (!interrupt)
> +		return -ENOMEM;
> +
> +	snprintf(interrupt, BUF_SIZE, "q6v5_wcss_userpd%d_%s", wcss->pd_asid, tmp);
> +
> +	ret = devm_request_threaded_irq(&pdev->dev, irq,
> +					NULL, handler,
> +					IRQF_TRIGGER_RISING | IRQF_ONESHOT,
> +					interrupt, q6);
> +	if (ret) {
> +		dev_err(&pdev->dev, "failed to acquire %s irq\n", interrupt);
> +		return ret;
> +	}
> +	return 0;
> +}
> +
> +static int q6_get_outbound_irq(struct qcom_q6v5 *q6,
> +			       struct platform_device *pdev,
> +			       const char *int_name)
> +{
> +	struct qcom_smem_state *tmp_state;
> +	unsigned  bit;
> +
> +	tmp_state = qcom_smem_state_get(&pdev->dev, int_name, &bit);
> +	if (IS_ERR(tmp_state)) {
> +		dev_err(&pdev->dev, "failed to acquire %s state\n", int_name);
> +		return PTR_ERR(tmp_state);
> +	}
> +
> +	if (!strcmp(int_name, "stop")) {
> +		q6->state = tmp_state;
> +		q6->stop_bit = bit;
> +	} else if (!strcmp(int_name, "spawn")) {
> +		q6->spawn_state = tmp_state;
> +		q6->spawn_bit = bit;
> +	}
> +
> +	return 0;
> +}
> +
> +static int init_irq(struct qcom_q6v5 *q6,
> +		    struct platform_device *pdev, struct rproc *rproc,
> +		    int crash_reason, const char *load_state,
> +		    void (*handover)(struct qcom_q6v5 *q6))
> +{
> +	int ret;
> +
> +	q6->rproc = rproc;
> +	q6->dev = &pdev->dev;
> +	q6->crash_reason = crash_reason;
> +	q6->handover = handover;
> +
> +	init_completion(&q6->start_done);
> +	init_completion(&q6->stop_done);
> +	init_completion(&q6->spawn_done);
> +
> +	ret = q6_get_inbound_irq(q6, pdev, "fatal",
> +				 q6v5_fatal_interrupt);
> +	if (ret)
> +		return ret;
> +
> +	ret = q6_get_inbound_irq(q6, pdev, "ready",
> +				 q6v5_ready_interrupt);
> +	if (ret)
> +		return ret;
> +
> +	ret = q6_get_inbound_irq(q6, pdev, "stop-ack",
> +				 q6v5_stop_interrupt);
> +	if (ret)
> +		return ret;
> +
> +	ret = q6_get_inbound_irq(q6, pdev, "spawn-ack",
> +				 q6v5_spawn_interrupt);
> +	if (ret)
> +		return ret;
> +
> +	ret = q6_get_outbound_irq(q6, pdev, "stop");
> +	if (ret)
> +		return ret;
> +
> +	ret = q6_get_outbound_irq(q6, pdev, "spawn");
> +	if (ret)
> +		return ret;
> +
> +	return 0;
> +}
> +
> +static int q6_wcss_probe(struct platform_device *pdev)
> +{
> +	const struct wcss_data *desc;
> +	struct q6_wcss *wcss;
> +	struct rproc *rproc;
> +	int ret;
> +	char *subdev_name;
> +	const char **firmware;
> +
> +	desc = of_device_get_match_data(&pdev->dev);
> +	if (!desc)
> +		return -EINVAL;
> +
> +	firmware = devm_kcalloc(&pdev->dev, MAX_FIRMWARE,
> +				sizeof(*firmware), GFP_KERNEL);
> +	if (!firmware)
> +		return -ENOMEM;
> +
> +	ret = of_property_read_string_array(pdev->dev.of_node, "firmware-name",
> +					    firmware, MAX_FIRMWARE);
> +	if (ret < 0)
> +		return ret;
> +
> +	rproc = rproc_alloc(&pdev->dev, pdev->name, desc->ops,
> +			    firmware[0], sizeof(*wcss));
> +	if (!rproc) {
> +		dev_err(&pdev->dev, "failed to allocate rproc\n");

ENOMEM do not print errors. Why do you have to print something here?

> +		return -ENOMEM;
> +	}
> +	wcss = rproc->priv;
> +	wcss->dev = &pdev->dev;
> +	wcss->desc = desc;
> +	wcss->firmware = firmware;
> +
> +	ret = q6_alloc_memory_region(wcss);
> +	if (ret)
> +		goto free_rproc;
> +
> +	wcss->pd_asid = qcom_get_pd_asid(wcss->dev->of_node);
> +	if (wcss->pd_asid < 0)
> +		goto free_rproc;
> +
> +	if (desc->init_irq) {
> +		ret = desc->init_irq(&wcss->q6, pdev, rproc,
> +				     desc->crash_reason_smem, NULL, NULL);
> +		if (ret)
> +			goto free_rproc;
> +	}
> +
> +	if (desc->glink_subdev_required)
> +		qcom_add_glink_subdev(rproc, &wcss->glink_subdev, desc->ssr_name);
> +
> +	subdev_name = (char *)(desc->ssr_name ? desc->ssr_name : pdev->name);
> +	qcom_add_ssr_subdev(rproc, &wcss->ssr_subdev, subdev_name);
> +
> +	rproc->auto_boot = false;
> +	ret = rproc_add(rproc);
> +	if (ret)
> +		goto free_rproc;
> +
> +	platform_set_drvdata(pdev, rproc);
> +
> +	ret = of_platform_populate(wcss->dev->of_node, NULL,
> +				   NULL, wcss->dev);

It is the same probe used for the children, right? So whom do they populate?

> +	if (ret) {
> +		dev_err(&pdev->dev, "failed to populate wcss pd nodes\n");

dev_err_probe

> +		goto free_rproc;
> +	}
> +	return 0;
> +
> +free_rproc:
> +	rproc_free(rproc);
> +
> +	return ret;
> +}
> +
> +static int q6_wcss_remove(struct platform_device *pdev)
> +{
> +	struct rproc *rproc = platform_get_drvdata(pdev);
> +	struct q6_wcss *wcss = rproc->priv;
> +
> +	qcom_q6v5_deinit(&wcss->q6);
> +
> +	rproc_del(rproc);
> +	rproc_free(rproc);
> +
> +	return 0;
> +}



Best regards,
Krzysztof

Re: [PATCH V2 13/13] arm64: dtsi: qcom: ipq9574: Add nodes to bring up multipd

[Manikanta Mylavarapu]

On 6/27/2023 1:14 PM, Manikanta Mylavarapu wrote:
> 
> 
> On 6/24/2023 12:31 PM, Krzysztof Kozlowski wrote:
>> On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
>>> Enable nodes required for multipd remoteproc bring up.
>>>
>>> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
>>> ---
>>> Changes in V2:
>>>     - Corrected syntax like alignmnet and kept nodes in sorted order.
>>>     - Added 'firmware-name' property.
>>>
>>>   arch/arm64/boot/dts/qcom/ipq9574.dtsi | 118 ++++++++++++++++++++++++++
>>>   1 file changed, 118 insertions(+)
>>>
>>> diff --git a/arch/arm64/boot/dts/qcom/ipq9574.dtsi 
>>> b/arch/arm64/boot/dts/qcom/ipq9574.dtsi
>>> index 0e04549c69a5..ff0da53ba05f 100644
>>> --- a/arch/arm64/boot/dts/qcom/ipq9574.dtsi
>>> +++ b/arch/arm64/boot/dts/qcom/ipq9574.dtsi
>>> @@ -160,6 +160,11 @@
>>>               no-map;
>>>           };
>>>
>>> +        q6_region: wcnss@4ab00000 {
>>> +            reg = <0x0 0x4ab00000 0x0 0x2b00000>;
>>> +            no-map;
>>> +        };
>>> +
>>>           smem@4aa00000 {
>>>               compatible = "qcom,smem";
>>>               reg = <0x0 0x4aa00000 0x0 0x00100000>;
>>> @@ -697,6 +702,95 @@
>>>               };
>>>           };
>>>
>>> +        q6v5_wcss: remoteproc@cd00000 {
>>> +            compatible = "qcom,ipq9574-q6-mpd";
>>> +            reg = <0x0cd00000 0x4040>;
>>> +            firmware-name = "IPQ9574/q6_fw.mdt",
>>> +                    "IPQ9574/m3_fw.mdt";
>>
>> Here and...
>>
>>> +            interrupts-extended = <&intc GIC_SPI 325 
>>> IRQ_TYPE_EDGE_RISING>,
>>> +                          <&wcss_smp2p_in 0 0>,
>>> +                          <&wcss_smp2p_in 1 0>,
>>> +                          <&wcss_smp2p_in 2 0>,
>>> +                          <&wcss_smp2p_in 3 0>;
>>> +            interrupt-names = "wdog",
>>> +                      "fatal",
>>> +                      "ready",
>>> +                      "handover",
>>> +                      "stop-ack";
>>> +
>>> +            qcom,smem-states = <&wcss_smp2p_out 0>,
>>> +                       <&wcss_smp2p_out 1>;
>>> +            qcom,smem-state-names = "shutdown",
>>> +                        "stop";
>>> +            memory-region = <&q6_region>;
>>> +
>>> +            glink-edge {
>>> +                interrupts = <GIC_SPI 321 IRQ_TYPE_EDGE_RISING>;
>>> +                label = "rtr";
>>> +                qcom,remote-pid = <1>;
>>> +                mboxes = <&apcs_glb 8>;
>>> +            };
>>> +
>>> +            pd-1 {
>>> +                compatible = "qcom,ipq9574-wcss-ahb-mpd";
>>> +                firmware-name = "IPQ9574/q6_fw.mdt";
>>
>> ... here - why do you have firmware in both places?
>>

	In multipd model, Q6 & WCSS uses different firmware.
	I will correct the firmware-name. Thanks for catching.

>>> +                interrupts-extended = <&wcss_smp2p_in 8 0>,
>>> +                              <&wcss_smp2p_in 9 0>,
>>> +                              <&wcss_smp2p_in 12 0>,
>>> +                              <&wcss_smp2p_in 11 0>;
>>> +                interrupt-names = "fatal",
>>> +                          "ready",
>>> +                          "spawn-ack",
>>> +                          "stop-ack";
>>> +                qcom,smem-states = <&wcss_smp2p_out 8>,
>>> +                           <&wcss_smp2p_out 9>,
>>> +                           <&wcss_smp2p_out 10>;
>>> +                qcom,smem-state-names = "shutdown",
>>> +                            "stop",
>>> +                            "spawn";
>>> +            };
>>> +
>>> +            pd-2 {
>>> +                compatible = "qcom,ipq5018-wcss-pcie-mpd";
>>
>> This compatible is confusing for this device.
>>
	I will clean up all SOC specific compatibles and have
	only device specific compatibles for Q6 & WCSS radio's
	as i mentioned on other thread.

Thanks & Regards,
Manikanta.

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Manikanta Mylavarapu]

On 6/24/2023 12:49 PM, Krzysztof Kozlowski wrote:
> On 21/06/2023 13:39, Manikanta Mylavarapu wrote:
>>>> on number of wcss radios connected on that board and only one instance
>>>> of 'qcom,ipq5018-q6-mpd'.
>>>>
>>>
>>> I don't understand why the user protection domains need a specific
>>> compatible. Why do they need compatible at all?
>>>
>>> Not mentioning that amount of your domains on Q6 is actually fixed per
>>> SoC and probably should not be in DT at all.
>>>
>>     root domain is fixed per soc (One Q6 DSP, one per soc)
>>     user domain(s) are variable (based on number of wcss radios attached)
>>
>>     The sequence to initialize, bring up, tear down the Q6 and wcss radios
>>     are completely different. So in order to differentiate them, we will
>>     need different compatibles. So this is a new rproc driver/architecture
>>     which has a parent/child topology (Q6 DSP -> Master/parent controls
>>     the WCSS (child)).
> 
> I understand you need different properties, but I don't see yet the
> benefit of creating here artificial compatibles. Look at your ipq9574
> DTSI change - it does not use even ipq9574 compatibles for 66% of its
> children.
> 
> Maybe you should have there just property describing type of device or
> bringup?
> 

	Yeah i got your point. Indeed the requirement here is to
	have device specific compatibles, so will have just two
	compatible one for Q6-MPD and another for WCSS-MPD device's


	"qcom,q6-mpd" --> For Q6-MPD device
	"qcom,wcss-mpd" --> For WCSS-MPD device

	Is this approach fine ?

> Given SoC cannot come with different amount of children (PD) and
> different amount of radios. You even fix the firmware name, so
> boards/customers cannot use anything else. It's fixed and the only
> variable element here is disabling some of the blocks per board, if they
> do not have physical interface (e.g. radio).
> 
> You even hard-code the number of PD by using "pd-[123]", without unit
> address, so you do not expect it will grow.
> 
> Unless you want to say that these are devices? But your binding does not
> really suggest it...
> 
> 
> 	Yes, as i mentioned above the requirement is to have device
	specific bindings. We will remove "PD-X" from soc dtsi and
	add it in board dts file.

	So soc dts would have "Q6-MPD" master node & board dts have
	"WCSS-MPD" child nodes based on number of radio's connected
	on board.

	Is this fine ?

Thanks & Regards,
Manikanta.

Re: [PATCH V2 10/13] remoteproc: qcom: Add Hexagon based multipd rproc driver

[Manikanta Mylavarapu]

On 6/27/2023 6:09 PM, Manikanta Mylavarapu wrote:
> 
> 
> On 6/24/2023 12:58 PM, Krzysztof Kozlowski wrote:
>> On 22/05/2023 00:28, Manikanta Mylavarapu wrote:
>>> It adds support to bring up remoteproc's on multipd model.
>>> Pd means protection domain. It's similar to process in Linux.
>>> Here QDSP6 processor runs each wifi radio functionality on a
>>> separate process. One process can't access other process
>>> resources, so this is termed as PD i.e protection domain.
>>>
>>> Here we have two pd's called root and user pd. We can correlate
>>> Root pd as root and user pd as user in linux. Root pd has more
>>> privileges than user pd. Root will provide services to user pd.
>>>
>>>  From remoteproc driver perspective, root pd corresponds to QDSP6
>>> processor bring up and user pd corresponds to Wifi radio (WCSS)
>>> bring up.
>>>
>>> Here WCSS(user) PD is dependent on Q6(root) PD, so first
>>> q6 pd should be up before wcss pd. After wcss pd goes down,
>>> q6 pd should be turned off.
>>>
>>> IPQ5018, IPQ9574 supports multipd remoteproc driver.
>>>
>>> Signed-off-by: Manikanta Mylavarapu <quic_mmanikan@quicinc.com>
>>> ---
>>> Changes in V2:
>>>     - Common functionalities moved to seperate patches
>>>     - qcom_get_pd_asid() moved to mpd driver
>>>     - Last DMA block alone memset to zero
>>>     - Added diagram to show how userpd data is organized and sent to
>>>       trustzone
>>>     - Rewritten commit message since most of the content available
>>>       in cover page
>>>     - Removed 'remote_id' becuase it's not required for bring up.
>>>
>>>   drivers/remoteproc/Kconfig          |  20 +
>>>   drivers/remoteproc/Makefile         |   1 +
>>>   drivers/remoteproc/qcom_common.h    |   9 +
>>>   drivers/remoteproc/qcom_q6v5_mpd.c  | 677 ++++++++++++++++++++++++++++
>>>   drivers/soc/qcom/mdt_loader.c       | 332 ++++++++++++++
>>>   include/linux/soc/qcom/mdt_loader.h |  19 +
>>>   6 files changed, 1058 insertions(+)
>>>   create mode 100644 drivers/remoteproc/qcom_q6v5_mpd.c
>>>
>>> diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig
>>> index a850e9f486dd..44af5c36f67e 100644
>>> --- a/drivers/remoteproc/Kconfig
>>> +++ b/drivers/remoteproc/Kconfig
>>> @@ -234,6 +234,26 @@ config QCOM_Q6V5_PAS
>>>         CDSP (Compute DSP), MPSS (Modem Peripheral SubSystem), and
>>>         SLPI (Sensor Low Power Island).
>>>
>>> +config QCOM_Q6V5_MPD
>>> +    tristate "Qualcomm Hexagon based MPD model Peripheral Image Loader"
>>> +    depends on OF && ARCH_QCOM
>>> +    depends on QCOM_SMEM
>>> +    depends on RPMSG_QCOM_SMD || RPMSG_QCOM_SMD=n
>>> +    depends on RPMSG_QCOM_GLINK_SMEM || RPMSG_QCOM_GLINK_SMEM=n
>>> +    depends on QCOM_SYSMON || QCOM_SYSMON=n
>>> +    depends on RPMSG_QCOM_GLINK || RPMSG_QCOM_GLINK=n
>>> +    depends on QCOM_AOSS_QMP || QCOM_AOSS_QMP=n
>>> +    select MFD_SYSCON
>>
>> Do you really need this?
>>
	It's not required. I will remove. Thanks for catching.

>>> +    select QCOM_MDT_LOADER
>>> +    select QCOM_PIL_INFO
>>> +    select QCOM_Q6V5_COMMON
>>> +    select QCOM_RPROC_COMMON
>>> +    select QCOM_SCM
>>> +    help
>>> +      Say y here to support the Qualcomm Secure Peripheral Image Loader
>>> +      for the Hexagon based MultiPD model remote processors on e.g. 
>>> IPQ5018.
>>> +      This is trustZone wireless subsystem.
>>> +
>>
>> ...
>>
	I didn't understand. Can you please elaborate your comment?

>>> +    int (*powerup_scm)(u32 peripheral);
>>> +    int (*powerdown_scm)(u32 peripheral);
>>> +};
>>> +
>>> +/**
>>> + * qcom_get_pd_asid() - get the pd asid number from DT node
>>
>>  From node name? NAK. It does not work like that. Node names can change
>> and you did not define this number as part of ABI.
>>
>> Probably you wanted unit address.
>>

	Yeah i got your point. Each of the WCSS PD's are internally
	represented in Q6 with their corresponding "spawn" bit numbers.
	I will use same and remove the "PD-" hardcodings.

	Is this fine ?

>>> + * @node:    device tree node
>>> + *
>>> + * Returns asid if node name has 'pd' string
>>> + */
>>> +s8 qcom_get_pd_asid(struct device_node *node)
>>> +{
>>> +    char *str;
>>> +    s8 pd_asid;
>>> +
>>> +    if (!node)
>>> +        return -EINVAL;
>>> +
>>> +    str = strstr(node->name, "pd");
>>> +    if (!str)
>>> +        return 0;
>>> +
>>> +    str += strlen("pd") + 1;
>>> +    return kstrtos8(str, 10, &pd_asid) ? -EINVAL : pd_asid;
>>> +}
>>> +EXPORT_SYMBOL(qcom_get_pd_asid);
>>> +
>>
>> ...
>>
	
	I didn't understand. Can you please elaborate your comment?

>>> +
>>> +static int q6_wcss_spawn_pd(struct rproc *rproc)
>>> +{
>>> +    int ret;
>>> +    struct q6_wcss *wcss = rproc->priv;
>>> +
>>> +    ret = qcom_q6v5_request_spawn(&wcss->q6);
>>> +    if (ret == -ETIMEDOUT) {
>>> +        pr_err("%s spawn timedout\n", rproc->name);
>>
>> dev_err
>>

	Sure, I will change to dev_err.

>>> +        return ret;
>>> +    }
>>> +
>>> +    ret = qcom_q6v5_wait_for_start(&wcss->q6, msecs_to_jiffies(10000));
>>> +    if (ret == -ETIMEDOUT) {
>>> +        pr_err("%s start timedout\n", rproc->name);
>>
>> dev_err
>>

	Sure, I will change to dev_err.

>>> +        wcss->q6.running = false;
>>> +        return ret;
>>> +    }
>>> +    wcss->q6.running = true;
>>> +    return ret;
>>> +}
>>> +
>>> +static int wcss_ahb_pcie_pd_start(struct rproc *rproc)
>>> +{
>>> +    struct q6_wcss *wcss = rproc->priv;
>>> +    const struct wcss_data *desc = wcss->desc;
>>> +    int ret;
>>> +
>>> +    if (!desc->reset_seq)
>>> +        return 0;
>>> +
>>> +    if (desc->powerup_scm) {
>>> +        ret = desc->powerup_scm(desc->pasid);
>>> +        if (ret) {
>>> +            dev_err(wcss->dev, "failed to power up pd\n");
>>> +            return ret;
>>> +        }
>>> +    }
>>> +
>>> +    ret = q6_wcss_spawn_pd(rproc);
>>> +    if (ret)
>>> +        return ret;
>>> +
>>> +    wcss->state = WCSS_NORMAL;
>>> +    return ret;
>>> +}
>>> +
>>> +static int q6_wcss_stop(struct rproc *rproc)
>>> +{
>>> +    struct q6_wcss *wcss = rproc->priv;
>>> +    const struct wcss_data *desc = wcss->desc;
>>> +    int ret;
>>> +
>>> +    ret = qcom_scm_pas_shutdown(desc->pasid);
>>> +    if (ret) {
>>> +        dev_err(wcss->dev, "not able to shutdown\n");
>>> +        return ret;
>>> +    }
>>> +    qcom_q6v5_unprepare(&wcss->q6);
>>> +
>>> +    return 0;
>>> +}
>>> +
>>> +static int wcss_ahb_pcie_pd_stop(struct rproc *rproc)
>>> +{
>>> +    struct q6_wcss *wcss = rproc->priv;
>>> +    struct rproc *rpd_rproc = dev_get_drvdata(wcss->dev->parent);
>>> +    const struct wcss_data *desc = wcss->desc;
>>> +    int ret;
>>> +
>>> +    if (!desc->reset_seq)
>>> +        goto shut_down_rpd;
>>> +
>>> +    if (rproc->state != RPROC_CRASHED && wcss->q6.stop_bit) {
>>> +        ret = qcom_q6v5_request_stop(&wcss->q6, NULL);
>>> +        if (ret) {
>>> +            dev_err(&rproc->dev, "pd not stopped\n");
>>> +            return ret;
>>> +        }
>>> +    }
>>> +
>>> +    if (desc->powerdown_scm) {
>>> +        ret = desc->powerdown_scm(desc->pasid);
>>> +        if (ret) {
>>> +            dev_err(wcss->dev, "failed to power down pd\n");
>>> +            return ret;
>>> +        }
>>> +    }
>>> +
>>> +shut_down_rpd:
>>> +    rproc_shutdown(rpd_rproc);
>>> +
>>> +    wcss->state = WCSS_SHUTDOWN;
>>> +    return 0;
>>> +}
>>> +
>>> +static void *q6_wcss_da_to_va(struct rproc *rproc, u64 da, size_t len,
>>> +                  bool *is_iomem)
>>> +{
>>> +    struct q6_wcss *wcss = rproc->priv;
>>> +    int offset;
>>> +
>>> +    offset = da - wcss->mem_reloc;
>>> +    if (offset < 0 || offset + len > wcss->mem_size)
>>> +        return NULL;
>>> +
>>> +    return wcss->mem_region + offset;
>>> +}
>>> +
>>> +static int q6_wcss_load(struct rproc *rproc, const struct firmware *fw)
>>> +{
>>> +    struct q6_wcss *wcss = rproc->priv;
>>> +    const struct firmware *fw_hdl;
>>> +    int ret;
>>> +    const struct wcss_data *desc = wcss->desc;
>>> +    int loop;
>>> +
>>> +    ret = qcom_mdt_load(wcss->dev, fw, rproc->firmware,
>>> +                desc->pasid, wcss->mem_region,
>>> +                wcss->mem_phys, wcss->mem_size,
>>> +                &wcss->mem_reloc);
>>> +    if (ret)
>>> +        return ret;
>>> +
>>> +    for (loop = 1; loop < MAX_FIRMWARE; loop++) {
>>> +        if (!wcss->firmware[loop])
>>> +            continue;
>>> +
>>> +        ret = request_firmware(&fw_hdl, wcss->firmware[loop],
>>> +                       wcss->dev);
>>> +        if (ret)
>>> +            continue;
>>> +
>>> +        ret = qcom_mdt_load_no_init(wcss->dev, fw_hdl,
>>> +                        wcss->firmware[loop], 0,
>>> +                        wcss->mem_region,
>>> +                        wcss->mem_phys,
>>> +                        wcss->mem_size,
>>> +                        &wcss->mem_reloc);
>>> +
>>> +        release_firmware(fw_hdl);
>>> +
>>> +        if (ret) {
>>> +            dev_err(wcss->dev,
>>> +                "can't load %s ret:%d\n", wcss->firmware[loop], ret);
>>> +            return ret;
>>> +        }
>>> +    }
>>> +    return 0;
>>> +}
>>> +
>>> +/* This function load's userpd firmware. Since Userpd depends on rootpd
>>
>> Use Linux coding style comments.
>>

	Sure, I will update.

>>> + * first bring up root pd and then load. User pd firmware load is 
>>> required
>>> + * only during user pd restart because root pd loads user pd FW pil 
>>> segments
>>> + * during it's bringup.
>>> + */
>>> +static int wcss_ahb_pcie_pd_load(struct rproc *rproc, const struct 
>>> firmware *fw)
>>> +{
>>> +    struct q6_wcss *wcss = rproc->priv, *wcss_rpd;
>>> +    struct rproc *rpd_rproc = dev_get_drvdata(wcss->dev->parent);
>>> +    const struct wcss_data *desc = wcss->desc;
>>> +    int ret;
>>> +
>>> +    wcss_rpd = rpd_rproc->priv;
>>> +
>>> +    /* Boot rootpd rproc */
>>
>> That's not helpful - i tcopies the function name.
>>

	sure, i will remove comment.

>>> +    ret = rproc_boot(rpd_rproc);
>>> +    if (ret || wcss->state == WCSS_NORMAL)
>>> +        return ret;
>>> +
>>> +    return desc->mdt_load_sec(wcss->dev, fw, rproc->firmware,
>>> +                  desc->pasid, wcss->mem_region,
>>> +                  wcss->mem_phys, wcss->mem_size,
>>> +                  &wcss->mem_reloc);
>>> +}
>>> +

...

	I didn't understand. Can you please elaborate your comment?

>>> +static int q6_get_inbound_irq(struct qcom_q6v5 *q6,
>>> +                  struct platform_device *pdev,
>>> +                  const char *int_name,
>>> +                  irqreturn_t (*handler)(int irq, void *data))
>>> +{
>>> +    int ret, irq;
>>> +    char *interrupt, *tmp = (char *)int_name;
>>> +    struct q6_wcss *wcss = q6->rproc->priv;
>>> +
>>> +    irq = platform_get_irq_byname(pdev, int_name);
>>> +    if (irq < 0) {
>>> +        if (irq != -EPROBE_DEFER)
>>> +            dev_err(&pdev->dev,
>>> +                "failed to retrieve %s IRQ: %d\n",
>>> +                    int_name, irq);
>>
>> return dev_err_probe
>>

	I will use dev_err_probe.

>>> +        return irq;
>>> +    }
>>> +
>>> +    if (!strcmp(int_name, "fatal")) {
>>> +        q6->fatal_irq = irq;
>>> +    } else if (!strcmp(int_name, "stop-ack")) {
>>> +        q6->stop_irq = irq;
>>> +        tmp = "stop_ack";
>>> +    } else if (!strcmp(int_name, "ready")) {
>>> +        q6->ready_irq = irq;
>>> +    } else if (!strcmp(int_name, "handover")) {
>>> +        q6->handover_irq  = irq;
>>> +    } else if (!strcmp(int_name, "spawn-ack")) {
>>> +        q6->spawn_irq = irq;
>>> +        tmp = "spawn_ack";
>>> +    } else {
>>> +        dev_err(&pdev->dev, "unknown interrupt\n");
>>> +        return -EINVAL;
>>> +    }
>>
>> This is over complicated method of getting interrupts. First, you can
>> pass here address of variable with interrupt to assign (*irq_field =
>> irq). Second, drop the names and get by index. Entries are fixed, not
>> flexible.
>>
	Sure, I will do it.

>>> +
>>> +    interrupt = devm_kzalloc(&pdev->dev, BUF_SIZE, GFP_KERNEL);
>>> +    if (!interrupt)
>>> +        return -ENOMEM;
>>> +
>>> +    snprintf(interrupt, BUF_SIZE, "q6v5_wcss_userpd%d_%s", 
>>> wcss->pd_asid, tmp);
>>> +
>>> +    ret = devm_request_threaded_irq(&pdev->dev, irq,
>>> +                    NULL, handler,
>>> +                    IRQF_TRIGGER_RISING | IRQF_ONESHOT,
>>> +                    interrupt, q6);
>>> +    if (ret) {
>>> +        dev_err(&pdev->dev, "failed to acquire %s irq\n", interrupt);
>>> +        return ret;
>>> +    }
>>> +    return 0;
>>> +}
>>> +
>>> +static int q6_get_outbound_irq(struct qcom_q6v5 *q6,
>>> +                   struct platform_device *pdev,
>>> +                   const char *int_name)
>>> +{
>>> +    struct qcom_smem_state *tmp_state;
>>> +    unsigned  bit;
>>> +
>>> +    tmp_state = qcom_smem_state_get(&pdev->dev, int_name, &bit);
>>> +    if (IS_ERR(tmp_state)) {
>>> +        dev_err(&pdev->dev, "failed to acquire %s state\n", int_name);
>>> +        return PTR_ERR(tmp_state);
>>> +    }
>>> +
>>> +    if (!strcmp(int_name, "stop")) {
>>> +        q6->state = tmp_state;
>>> +        q6->stop_bit = bit;
>>> +    } else if (!strcmp(int_name, "spawn")) {
>>> +        q6->spawn_state = tmp_state;
>>> +        q6->spawn_bit = bit;
>>> +    }
>>> +
>>> +    return 0;
>>> +}
>>> +
>>> +static int init_irq(struct qcom_q6v5 *q6,
>>> +            struct platform_device *pdev, struct rproc *rproc,
>>> +            int crash_reason, const char *load_state,
>>> +            void (*handover)(struct qcom_q6v5 *q6))
>>> +{
>>> +    int ret;
>>> +
>>> +    q6->rproc = rproc;
>>> +    q6->dev = &pdev->dev;
>>> +    q6->crash_reason = crash_reason;
>>> +    q6->handover = handover;
>>> +
>>> +    init_completion(&q6->start_done);
>>> +    init_completion(&q6->stop_done);
>>> +    init_completion(&q6->spawn_done);
>>> +
>>> +    ret = q6_get_inbound_irq(q6, pdev, "fatal",
>>> +                 q6v5_fatal_interrupt);
>>> +    if (ret)
>>> +        return ret;
>>> +
>>> +    ret = q6_get_inbound_irq(q6, pdev, "ready",
>>> +                 q6v5_ready_interrupt);
>>> +    if (ret)
>>> +        return ret;
>>> +
>>> +    ret = q6_get_inbound_irq(q6, pdev, "stop-ack",
>>> +                 q6v5_stop_interrupt);
>>> +    if (ret)
>>> +        return ret;
>>> +
>>> +    ret = q6_get_inbound_irq(q6, pdev, "spawn-ack",
>>> +                 q6v5_spawn_interrupt);
>>> +    if (ret)
>>> +        return ret;
>>> +
>>> +    ret = q6_get_outbound_irq(q6, pdev, "stop");
>>> +    if (ret)
>>> +        return ret;
>>> +
>>> +    ret = q6_get_outbound_irq(q6, pdev, "spawn");
>>> +    if (ret)
>>> +        return ret;
>>> +
>>> +    return 0;
>>> +}
>>> +
>>> +static int q6_wcss_probe(struct platform_device *pdev)
>>> +{
>>> +    const struct wcss_data *desc;
>>> +    struct q6_wcss *wcss;
>>> +    struct rproc *rproc;
>>> +    int ret;
>>> +    char *subdev_name;
>>> +    const char **firmware;
>>> +
>>> +    desc = of_device_get_match_data(&pdev->dev);
>>> +    if (!desc)
>>> +        return -EINVAL;
>>> +
>>> +    firmware = devm_kcalloc(&pdev->dev, MAX_FIRMWARE,
>>> +                sizeof(*firmware), GFP_KERNEL);
>>> +    if (!firmware)
>>> +        return -ENOMEM;
>>> +
>>> +    ret = of_property_read_string_array(pdev->dev.of_node, 
>>> "firmware-name",
>>> +                        firmware, MAX_FIRMWARE);
>>> +    if (ret < 0)
>>> +        return ret;
>>> +
>>> +    rproc = rproc_alloc(&pdev->dev, pdev->name, desc->ops,
>>> +                firmware[0], sizeof(*wcss));
>>> +    if (!rproc) {
>>> +        dev_err(&pdev->dev, "failed to allocate rproc\n");
>>
>> ENOMEM do not print errors. Why do you have to print something here?
>>

	Yeah, I will remove print.

>>> +        return -ENOMEM;
>>> +    }
>>> +    wcss = rproc->priv;
>>> +    wcss->dev = &pdev->dev;
>>> +    wcss->desc = desc;
>>> +    wcss->firmware = firmware;
>>> +
>>> +    ret = q6_alloc_memory_region(wcss);
>>> +    if (ret)
>>> +        goto free_rproc;
>>> +
>>> +    wcss->pd_asid = qcom_get_pd_asid(wcss->dev->of_node);
>>> +    if (wcss->pd_asid < 0)
>>> +        goto free_rproc;
>>> +
>>> +    if (desc->init_irq) {
>>> +        ret = desc->init_irq(&wcss->q6, pdev, rproc,
>>> +                     desc->crash_reason_smem, NULL, NULL);
>>> +        if (ret)
>>> +            goto free_rproc;
>>> +    }
>>> +
>>> +    if (desc->glink_subdev_required)
>>> +        qcom_add_glink_subdev(rproc, &wcss->glink_subdev, 
>>> desc->ssr_name);
>>> +
>>> +    subdev_name = (char *)(desc->ssr_name ? desc->ssr_name : 
>>> pdev->name);
>>> +    qcom_add_ssr_subdev(rproc, &wcss->ssr_subdev, subdev_name);
>>> +
>>> +    rproc->auto_boot = false;
>>> +    ret = rproc_add(rproc);
>>> +    if (ret)
>>> +        goto free_rproc;
>>> +
>>> +    platform_set_drvdata(pdev, rproc);
>>> +
>>> +    ret = of_platform_populate(wcss->dev->of_node, NULL,
>>> +                   NULL, wcss->dev);
>>
>> It is the same probe used for the children, right? So whom do they 
>> populate?
>>

	I will add match table in 2nd param, so that childs, whose
	compatible matches with match table will be populated.

	Is this fine ?
>>> +    if (ret) {
>>> +        dev_err(&pdev->dev, "failed to populate wcss pd nodes\n");
>>
>> dev_err_probe
>>
	 Sure, i will use dev_err_probe

>>> +        goto free_rproc;
>>> +    }
>>> +    return 0;
>>> +
>>> +free_rproc:
>>> +    rproc_free(rproc);
>>> +
>>> +    return ret;
>>> +}
>>> +
>>> +static int q6_wcss_remove(struct platform_device *pdev)
>>> +{
>>> +    struct rproc *rproc = platform_get_drvdata(pdev);
>>> +    struct q6_wcss *wcss = rproc->priv;
>>> +
>>> +    qcom_q6v5_deinit(&wcss->q6);
>>> +
>>> +    rproc_del(rproc);
>>> +    rproc_free(rproc);
>>> +
>>> +    return 0;
>>> +}
>>
>>
>>

Thanks & Regards,
Manikanta.

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Krzysztof Kozlowski]

On 30/06/2023 09:12, Manikanta Mylavarapu wrote:
> 
> 
> On 6/24/2023 12:49 PM, Krzysztof Kozlowski wrote:
>> On 21/06/2023 13:39, Manikanta Mylavarapu wrote:
>>>>> on number of wcss radios connected on that board and only one instance
>>>>> of 'qcom,ipq5018-q6-mpd'.
>>>>>
>>>>
>>>> I don't understand why the user protection domains need a specific
>>>> compatible. Why do they need compatible at all?
>>>>
>>>> Not mentioning that amount of your domains on Q6 is actually fixed per
>>>> SoC and probably should not be in DT at all.
>>>>
>>>     root domain is fixed per soc (One Q6 DSP, one per soc)
>>>     user domain(s) are variable (based on number of wcss radios attached)
>>>
>>>     The sequence to initialize, bring up, tear down the Q6 and wcss radios
>>>     are completely different. So in order to differentiate them, we will
>>>     need different compatibles. So this is a new rproc driver/architecture
>>>     which has a parent/child topology (Q6 DSP -> Master/parent controls
>>>     the WCSS (child)).
>>
>> I understand you need different properties, but I don't see yet the
>> benefit of creating here artificial compatibles. Look at your ipq9574
>> DTSI change - it does not use even ipq9574 compatibles for 66% of its
>> children.
>>
>> Maybe you should have there just property describing type of device or
>> bringup?
>>
> 
> 	Yeah i got your point. Indeed the requirement here is to
> 	have device specific compatibles, so will have just two
> 	compatible one for Q6-MPD and another for WCSS-MPD device's
> 
> 
> 	"qcom,q6-mpd" --> For Q6-MPD device
> 	"qcom,wcss-mpd" --> For WCSS-MPD device
> 
> 	Is this approach fine ?

Can you fix your reply style, so it is like on every mailing list? Some
weird indentation does no help to read it.

I was proposing to drop compatibles entirely. Making compatibles generic
is not solving any of my concerns.

I don't understand what do you want to achieve here and very limited
description of the hardware in the binding does not help.

> 
>> Given SoC cannot come with different amount of children (PD) and
>> different amount of radios. You even fix the firmware name, so
>> boards/customers cannot use anything else. It's fixed and the only
>> variable element here is disabling some of the blocks per board, if they
>> do not have physical interface (e.g. radio).
>>
>> You even hard-code the number of PD by using "pd-[123]", without unit
>> address, so you do not expect it will grow.
>>
>> Unless you want to say that these are devices? But your binding does not
>> really suggest it...
>>
>>
>> 	Yes, as i mentioned above the requirement is to have device

What requirement? You did not describe anything. Binding describes
hardware, not your requirements.

Binding said nothing about devices.

> 	specific bindings. We will remove "PD-X" from soc dtsi and
> 	add it in board dts file.

Why? How is it related to the bindings? What does it solve? Instead of
doing some changes you should explain why.

> 
> 	So soc dts would have "Q6-MPD" master node & board dts have
> 	"WCSS-MPD" child nodes based on number of radio's connected
> 	on board.
> 
> 	Is this fine ?
> 

Why?

Best regards,
Krzysztof

Re: [PATCH V2 10/13] remoteproc: qcom: Add Hexagon based multipd rproc driver

[Krzysztof Kozlowski]

On 30/06/2023 12:29, Manikanta Mylavarapu wrote:
> 
> 
> On 6/27/2023 6:09 PM, Manikanta Mylavarapu wrote:
>>
>>

Do you see the lines above? You quote my reply and claim it is you who
wrote it.

Please fix your mailer, it's making discussions unnecessary difficult.


>>>> +      Say y here to support the Qualcomm Secure Peripheral Image Loader
>>>> +      for the Hexagon based MultiPD model remote processors on e.g. 
>>>> IPQ5018.
>>>> +      This is trustZone wireless subsystem.
>>>> +
>>>
>>> ...
>>>
> 	I didn't understand. Can you please elaborate your comment?

https://en.wikipedia.org/wiki/Ellipsis

> 
>>>> +    int (*powerup_scm)(u32 peripheral);
>>>> +    int (*powerdown_scm)(u32 peripheral);
>>>> +};
>>>> +
>>>> +/**
>>>> + * qcom_get_pd_asid() - get the pd asid number from DT node
>>>
>>>  From node name? NAK. It does not work like that. Node names can change
>>> and you did not define this number as part of ABI.
>>>
>>> Probably you wanted unit address.
>>>
> 
> 	Yeah i got your point. Each of the WCSS PD's are internally
> 	represented in Q6 with their corresponding "spawn" bit numbers.
> 	I will use same and remove the "PD-" hardcodings.
> 
> 	Is this fine ?

I don't get what you are going to use, so post a patch. Talk is cheap.


Best regards,
Krzysztof

Re: [PATCH V2 10/13] remoteproc: qcom: Add Hexagon based multipd rproc driver

[Manikanta Mylavarapu]

On 7/1/2023 4:25 PM, Krzysztof Kozlowski wrote:
> On 30/06/2023 12:29, Manikanta Mylavarapu wrote:
>>
>>
>> On 6/27/2023 6:09 PM, Manikanta Mylavarapu wrote:
>>>
>>>
> 
> Do you see the lines above? You quote my reply and claim it is you who
> wrote it.
> 
> Please fix your mailer, it's making discussions unnecessary difficult.
> 
> 

Sure, i will correct it.

>>>>> +      Say y here to support the Qualcomm Secure Peripheral Image Loader
>>>>> +      for the Hexagon based MultiPD model remote processors on e.g.
>>>>> IPQ5018.
>>>>> +      This is trustZone wireless subsystem.
>>>>> +
>>>>
>>>> ...
>>>>
>> 	I didn't understand. Can you please elaborate your comment?
> 
> https://en.wikipedia.org/wiki/Ellipsis
> 
Got it.

>>
>>>>> +    int (*powerup_scm)(u32 peripheral);
>>>>> +    int (*powerdown_scm)(u32 peripheral);
>>>>> +};
>>>>> +
>>>>> +/**
>>>>> + * qcom_get_pd_asid() - get the pd asid number from DT node
>>>>
>>>>   From node name? NAK. It does not work like that. Node names can change
>>>> and you did not define this number as part of ABI.
>>>>
>>>> Probably you wanted unit address.
>>>>
>>
>> 	Yeah i got your point. Each of the WCSS PD's are internally
>> 	represented in Q6 with their corresponding "spawn" bit numbers.
>> 	I will use same and remove the "PD-" hardcodings.
>>
>> 	Is this fine ?
> 
> I don't get what you are going to use, so post a patch. Talk is cheap.
> 
> 
Sure, i posted a patch. Patch available here
https://lore.kernel.org/linux-arm-msm/20230718120501.3205661-10-quic_mmanikan@quicinc.com/

Thanks & Regards,
Manikanta.

> Best regards,
> Krzysztof
> 

Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

[Manikanta Mylavarapu]

On 7/1/2023 4:21 PM, Krzysztof Kozlowski wrote:
> On 30/06/2023 09:12, Manikanta Mylavarapu wrote:
>>
>>
>> On 6/24/2023 12:49 PM, Krzysztof Kozlowski wrote:
>>> On 21/06/2023 13:39, Manikanta Mylavarapu wrote:
>>>>>> on number of wcss radios connected on that board and only one instance
>>>>>> of 'qcom,ipq5018-q6-mpd'.
>>>>>>
>>>>>
>>>>> I don't understand why the user protection domains need a specific
>>>>> compatible. Why do they need compatible at all?
>>>>>
>>>>> Not mentioning that amount of your domains on Q6 is actually fixed per
>>>>> SoC and probably should not be in DT at all.
>>>>>
>>>>      root domain is fixed per soc (One Q6 DSP, one per soc)
>>>>      user domain(s) are variable (based on number of wcss radios attached)
>>>>
>>>>      The sequence to initialize, bring up, tear down the Q6 and wcss radios
>>>>      are completely different. So in order to differentiate them, we will
>>>>      need different compatibles. So this is a new rproc driver/architecture
>>>>      which has a parent/child topology (Q6 DSP -> Master/parent controls
>>>>      the WCSS (child)).
>>>
>>> I understand you need different properties, but I don't see yet the
>>> benefit of creating here artificial compatibles. Look at your ipq9574
>>> DTSI change - it does not use even ipq9574 compatibles for 66% of its
>>> children.
>>>
>>> Maybe you should have there just property describing type of device or
>>> bringup?
>>>
>>
>> 	Yeah i got your point. Indeed the requirement here is to
>> 	have device specific compatibles, so will have just two
>> 	compatible one for Q6-MPD and another for WCSS-MPD device's
>>
>>
>> 	"qcom,q6-mpd" --> For Q6-MPD device
>> 	"qcom,wcss-mpd" --> For WCSS-MPD device
>>
>> 	Is this approach fine ?
> 
> Can you fix your reply style, so it is like on every mailing list? Some
> weird indentation does no help to read it.
>

Sure, i will change my reply style and don't use any indentation.
Sorry for inconvenience.

> I was proposing to drop compatibles entirely. Making compatibles generic
> is not solving any of my concerns.
> 
> I don't understand what do you want to achieve here and very limited
> description of the hardware in the binding does not help.
> 

Sure, i remove user pd compatibles. In root pd probe itself, user pd
remoteproc's are taken care. Also I updated diagram in cover page, it
gives enough information.

>>
>>> Given SoC cannot come with different amount of children (PD) and
>>> different amount of radios. You even fix the firmware name, so
>>> boards/customers cannot use anything else. It's fixed and the only
>>> variable element here is disabling some of the blocks per board, if they
>>> do not have physical interface (e.g. radio).
>>>
>>> You even hard-code the number of PD by using "pd-[123]", without unit
>>> address, so you do not expect it will grow.
>>>
>>> Unless you want to say that these are devices? But your binding does not
>>> really suggest it...
>>>
>>>
>>> 	Yes, as i mentioned above the requirement is to have device
> 
> What requirement? You did not describe anything. Binding describes
> hardware, not your requirements.
> 
> Binding said nothing about devices.
> 

Yeah got your point. So we removed userpd compatibles from dt-bindings.


>> 	specific bindings. We will remove "PD-X" from soc dtsi and
>> 	add it in board dts file.
> 
> Why? How is it related to the bindings? What does it solve? Instead of
> doing some changes you should explain why.
> 
>>
>> 	So soc dts would have "Q6-MPD" master node & board dts have
>> 	"WCSS-MPD" child nodes based on number of radio's connected
>> 	on board.
>>
>> 	Is this fine ?
>>
> 
> Why?
> 

"PD-X" controls user pd WCSS bring up. WCSS blocks may vary based on
number of radio's connected on board but QDSP6 is always present.
So i will keep Q6 node in soc dtsi file and move 'PD-X' node to board
dts file.

Thanks & Regards,
Manikanta.

> Best regards,
> Krzysztof
>