Skip to contentSkip to conversation

dt-bindings: soc: qcom: Add remote-pid binding for GLINK SMEM

PATCH Superseded Sibi Sankar 4 participants 24 messages Dec 17, 2018 – Dec 19, 2018

Subsystems (4)

v2 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/7 dt-bindings: soc: qcom: Add remote-pid binding for GLINK SMEM

Add missing qcom,remote-pid dt binding required for GLINK SMEM which specifies the remote endpoint of the GLINK edge.

---
 Documentation/devicetree/bindings/soc/qcom/qcom,glink.txt | 5 +++++
 1 file changed, 5 insertions(+)
Documentation/devicetree/bindings/soc/qcom/qcom,glink.txt +5 -0
index 0b8cc533ca83..59ae603ba520 100644
@@ -21,6 +21,11 @@ edge.
 	Definition: should specify the IRQ used by the remote processor to
 		    signal this processor about communication related events
 
+- qcom,remote-pid:
+	Usage: required for glink-smem
+	Value type: <u32>
+	Definition: specifies the identfier of the remote endpoint of this edge
+
 - qcom,rpm-msg-ram:
 	Usage: required for glink-rpm
 	Value type: <prop-encoded-array>
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project
#1 lore raw

PATCH v2 2/7 dt-bindings: remoteproc: qcom: Add clock bindings for Q6V5

Add missing clock bindings for Q6V5 MSS on SDM845 SoCs.

---
 .../devicetree/bindings/remoteproc/qcom,q6v5.txt       | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +7 -3
index 9ff5b0309417..780adc043b37 100644
@@ -39,13 +39,17 @@ on the Qualcomm Hexagon core.
 - clocks:
 	Usage: required
 	Value type: <phandle>
-	Definition: reference to the iface, bus and mem clocks to be held on
-		    behalf of the booting of the Hexagon core
+	Definition: reference to the list of 4 clocks for the modem sub-system
+		    reference to the list of 8 clocks for the modem sub-system
+		    on SDM845 SoCs
 
 - clock-names:
 	Usage: required
 	Value type: <stringlist>
-	Definition: must be "iface", "bus", "mem"
+	Definition: must be "iface", "bus", "mem", "xo" for the modem sub-system
+		    must be "iface", "bus", "mem", "gpll0_mss", "snoc_axi",
+		    "mnoc_axi", "prng", "xo" for the modem sub-system on SDM845
+		    SoCs
 
 - resets:
 	Usage: required
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project
#2 lore raw

Re: [PATCH v2 2/7] dt-bindings: remoteproc: qcom: Add clock bindings for Q6V5

Hi,

Quoted 7 lines · Sibi Sankar

Add missing clock bindings for Q6V5 MSS on SDM845 SoCs.

---
 .../devicetree/bindings/remoteproc/qcom,q6v5.txt       | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
Fixes 9f058fa2efb1 ("remoteproc: qcom: Add support for mss remoteproc

on msm8996")

Fixes fb22022ff63d ("dt-bindings: remoteproc: Add Q6v5 Modem PIL

binding for SDM845")

...it probably doesn't matter too much but if we wanted to be really careful we could split into two patches, one for the msm8996 and one for sdm845. I don't think people care that much about stable backports of bindings though (someone can feel free to correct me)...

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt on the Qualcomm Hexagon core. · line 44

Quoted 13 lines
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +3 -2
index 9ff5b0309417..780adc043b37 100644
@@ -39,13 +39,17 @@ on the Qualcomm Hexagon core.
 - clocks:
        Usage: required
        Value type: <phandle>
-       Definition: reference to the iface, bus and mem clocks to be held on
-                   behalf of the booting of the Hexagon core
+       Definition: reference to the list of 4 clocks for the modem sub-system
+                   reference to the list of 8 clocks for the modem sub-system
+                   on SDM845 SoCs

The above is confusing because you don't list the SoCs that are supposed to use the 4 clocks. How about instead:

Definition: reference to the clocks that match clock-names

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt on the Qualcomm Hexagon core. · line 52

Quoted 8 lines
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +4 -1
 - clock-names:
        Usage: required
        Value type: <stringlist>
-       Definition: must be "iface", "bus", "mem"
+       Definition: must be "iface", "bus", "mem", "xo" for the modem sub-system
+                   must be "iface", "bus", "mem", "gpll0_mss", "snoc_axi",
+                   "mnoc_axi", "prng", "xo" for the modem sub-system on SDM845
+                   SoCs

Same here where it's confusing. ...but also, it it correct? As far as I can tell you're missing msm8996. It's better to just be explicit and list each one, ideally without all the prose.

Definition: The clocks needed depend on the compatible string:

qcom,sdm845-mss-pil: "xo", "prng", "iface", "snoc_axi", "bus", "mem", "gpll0_mss", "mnoc_axi"
qcom,msm8996-mss-pil: "xo", "pnoc", "iface", "bus", "mem", "gpll0_mss_clk" qcom,msm8974-mss-pil: "xo", "iface", "bus", "mem" qcom,msm8916-mss-pil: "xo", "iface", "bus", "mem" qcom,q6v5-pil: "xo", "iface", "bus", "mem"

...as far as I can tell this binding is supposed to account for "qcom,ipq8074-wcss-pil" too but it seems that one doesn't have clock-names.

-Doug

#3 lore raw

Re: [PATCH v2 2/7] dt-bindings: remoteproc: qcom: Add clock bindings for Q6V5

Hi Doug,
Thanks for the review :)

Quoted 13 lines · Doug Anderson

Hi,

Quoted 8 lines · Sibi Sankar

Add missing clock bindings for Q6V5 MSS on SDM845 SoCs.

---
 .../devicetree/bindings/remoteproc/qcom,q6v5.txt       | 10 
+++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
Fixes 9f058fa2efb1 ("remoteproc: qcom: Add support for mss remoteproc
Quoted 9 lines

on msm8996")

Fixes fb22022ff63d ("dt-bindings: remoteproc: Add Q6v5 Modem PIL

binding for SDM845")

...it probably doesn't matter too much but if we wanted to be really careful we could split into two patches, one for the msm8996 and one for sdm845. I don't think people care that much about stable backports of bindings though (someone can feel free to correct me)...

I did think of splitting this up but it doesn't actually fix 9f058fa2efb1 yet. I noticed a few missing clocks for mss on 8996 when I did a diff with the corresponding CAF tree. Hence couldn't add bindings for it. Will add them once I validate mss on 8996 with the necessary changes.

Quoted 21 lines
Quoted 13 lines
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +0 -1
index 9ff5b0309417..780adc043b37 100644
@@ -39,13 +39,17 @@ on the Qualcomm Hexagon core.
 - clocks:
        Usage: required
        Value type: <phandle>
-       Definition: reference to the iface, bus and mem clocks to be 

held on

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +1 -1
-                   behalf of the booting of the Hexagon core
+       Definition: reference to the list of 4 clocks for the modem 
Quoted 4 lines

sub-system
+ reference to the list of 8 clocks for the modem
sub-system
+ on SDM845 SoCs

The above is confusing because you don't list the SoCs that are supposed to use the 4 clocks. How about instead:

Quoted 2 lines

Definition: reference to the clocks that match clock-names

AFAIK, only the exceptions are captured. I am fine with both, I'll wait for Bjorn/Rob's preference.

Quoted 13 lines
Quoted 6 lines

- clock-names:
Usage: required
Value type: <stringlist>

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +1 -1
-       Definition: must be "iface", "bus", "mem"
+       Definition: must be "iface", "bus", "mem", "xo" for the modem 
Quoted 6 lines

sub-system
+ must be "iface", "bus", "mem", "gpll0_mss",
"snoc_axi",
+ "mnoc_axi", "prng", "xo" for the modem sub-system
on SDM845
+ SoCs

Quoted 6 lines

Same here where it's confusing. ...but also, it it correct? As far as I can tell you're missing msm8996. It's better to just be explicit and list each one, ideally without all the prose.

Definition: The clocks needed depend on the compatible string:

ditto

Quoted 3 lines

qcom,sdm845-mss-pil: "xo", "prng", "iface", "snoc_axi", "bus", "mem", "gpll0_mss", "mnoc_axi"
qcom,msm8996-mss-pil: "xo", "pnoc", "iface", "bus", "mem", "gpll0_mss_clk"

ditto

Quoted 8 lines

qcom,msm8974-mss-pil: "xo", "iface", "bus", "mem"
qcom,msm8916-mss-pil: "xo", "iface", "bus", "mem"
qcom,q6v5-pil: "xo", "iface", "bus", "mem"

...as far as I can tell this binding is supposed to account for "qcom,ipq8074-wcss-pil" too but it seems that one doesn't have clock-names.

Yeah the lack of clocks have to be documented
for ipq8074-wcss-pil.. will do it in v3

Quoted 2 lines

-Doug

Signature · 3 lines
-- Sibi Sankar --
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project.
#4 lore raw

Re: [PATCH v2 2/7] dt-bindings: remoteproc: qcom: Add clock bindings for Q6V5

Hi,

Quoted 16 lines · Sibi Sankar
Quoted 16 lines
Quoted 8 lines
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +0 -1
@@ -39,13 +39,17 @@ on the Qualcomm Hexagon core.
 - clocks:
        Usage: required
        Value type: <phandle>
-       Definition: reference to the iface, bus and mem clocks to be

held on

(quoted diff) +1 -1
-                   behalf of the booting of the Hexagon core
+       Definition: reference to the list of 4 clocks for the modem
Quoted 4 lines

sub-system
+ reference to the list of 8 clocks for the modem
sub-system
+ on SDM845 SoCs

The above is confusing because you don't list the SoCs that are supposed to use the 4 clocks. How about instead:

Quoted 5 lines
Quoted 2 lines

Definition: reference to the clocks that match clock-names

AFAIK, only the exceptions are captured. I am fine with both, I'll wait for Bjorn/Rob's preference.

Sure, waiting for Bjron / Rob to chime in sounds good. If you are going to document just the exception, though, IMO at least reword it. AKA:

reference to the list of 8 clocks for the modem sub-system on SDM845 SoCs; reference to the list of 4 clocks for the modem sub-system on all other SoCs.

...the key being the "on all other" to make it clear. ...but I still like my original suggestion of using less prose and more lists, so hopefully Bjorn / Rob are OK with that.

-Doug

maintainer
#5 lore raw

Re: [PATCH v2 2/7] dt-bindings: remoteproc: qcom: Add clock bindings for Q6V5

In reply to Sibi Sankar
Quoted 19 lines · Sibi Sankar

Add missing clock bindings for Q6V5 MSS on SDM845 SoCs.

---
 .../devicetree/bindings/remoteproc/qcom,q6v5.txt       | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +2 -2
index 9ff5b0309417..780adc043b37 100644
@@ -39,13 +39,17 @@ on the Qualcomm Hexagon core.
 - clocks:
 	Usage: required
 	Value type: <phandle>
-	Definition: reference to the iface, bus and mem clocks to be held on
-		    behalf of the booting of the Hexagon core
+	Definition: reference to the list of 4 clocks for the modem sub-system
+		    reference to the list of 8 clocks for the modem sub-system

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt on the Qualcomm Hexagon core. · line 52

Quoted 10 lines
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +5 -1
+		    on SDM845 SoCs
 
 - clock-names:
 	Usage: required
 	Value type: <stringlist>
-	Definition: must be "iface", "bus", "mem"
+	Definition: must be "iface", "bus", "mem", "xo" for the modem sub-system
+		    must be "iface", "bus", "mem", "gpll0_mss", "snoc_axi",
+		    "mnoc_axi", "prng", "xo" for the modem sub-system on SDM845
+		    SoCs

This seems to me a list of all clocks you need enabled, not what clocks actually go to the modem. Specifically, shouldn't the *noc_axi clocks be managed by the interconnect driver?

Rob

#6 lore raw

Re: [PATCH v2 2/7] dt-bindings: remoteproc: qcom: Add clock bindings for Q6V5

Hi Rob,
Thanks for the review!

Quoted 24 lines · Rob Herring
Quoted 23 lines · Sibi Sankar

Add missing clock bindings for Q6V5 MSS on SDM845 SoCs.

---
 .../devicetree/bindings/remoteproc/qcom,q6v5.txt       | 10 
+++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +1 -2
index 9ff5b0309417..780adc043b37 100644
@@ -39,13 +39,17 @@ on the Qualcomm Hexagon core.
 - clocks:
 	Usage: required
 	Value type: <phandle>
-	Definition: reference to the iface, bus and mem clocks to be held on
-		    behalf of the booting of the Hexagon core
+	Definition: reference to the list of 4 clocks for the modem 

sub-system

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +2 -0
+		    reference to the list of 8 clocks for the modem sub-system
+		    on SDM845 SoCs
Quoted 13 lines

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt on the Qualcomm Hexagon core. · line 52

Quoted 9 lines

- clock-names:

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +1 -1
 	Usage: required
 	Value type: <stringlist>
-	Definition: must be "iface", "bus", "mem"
+	Definition: must be "iface", "bus", "mem", "xo" for the modem 

sub-system

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +3 -0
+		    must be "iface", "bus", "mem", "gpll0_mss", "snoc_axi",
+		    "mnoc_axi", "prng", "xo" for the modem sub-system on SDM845
+		    SoCs

This seems to me a list of all clocks you need enabled, not what clocks actually go to the modem. Specifically, shouldn't the *noc_axi clocks be managed by the interconnect driver?

clocks = ...,

       <&gcc GCC_MSS_SNOC_AXI_CLK>,
       <&gcc GCC_MSS_MFAB_AXIS_CLK>,
         ...;

clock-names = ..., "snoc_axi", "mnoc_axi",...;

snoc_axi and mnoc_axi maps to above GCC clks and
both of them fall under the MSS functional group

Quoted 2 lines

Rob

Signature · 3 lines
-- Sibi Sankar --
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project.
#7 lore raw

PATCH v2 3/7 dt-bindings: remoteproc: qcom: Fixup regulator dependencies

In reply to Sibi Sankar

Fixup regulator supply dependencies for Q6V5 MSS on MSM996 SoCs.

---
 Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +3 -1
index 780adc043b37..98894e6ad456 100644
@@ -76,7 +76,9 @@ on the Qualcomm Hexagon core.
 	Usage: required
 	Value type: <phandle>
 	Definition: reference to the regulators to be held on behalf of the
-		    booting of the Hexagon core
+		    booting of the Hexagon core on MSM8916 SoCs
+		    reference to the pll-supply regulator to be held on behalf
+		    of the booting of the Hexagon core on MSM8996 SoCs
 
 - qcom,smem-states:
 	Usage: required
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project
#8 lore raw

Re: [PATCH v2 3/7] dt-bindings: remoteproc: qcom: Fixup regulator dependencies

Hi,

Quoted 13 lines · Sibi Sankar

Fixup regulator supply dependencies for Q6V5 MSS on MSM996 SoCs.

---
 Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +0 -0
index 780adc043b37..98894e6ad456 100644

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt on the Qualcomm Hexagon core. · line 81

Quoted 7 lines
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +3 -1
@@ -76,7 +76,9 @@ on the Qualcomm Hexagon core.
        Usage: required
        Value type: <phandle>
        Definition: reference to the regulators to be held on behalf of the
-                   booting of the Hexagon core
+                   booting of the Hexagon core on MSM8916 SoCs
+                   reference to the pll-supply regulator to be held on behalf
+                   of the booting of the Hexagon core on MSM8996 SoCs

The prose gets in the way and doesn't add anything. I also don't understand what you're saying for msm8996. You're saying that "pll-supply" is required there but none of the others? That doesn't seem to be true in the code I have in front of me, but maybe I'm missing some patch. For me, I'd write:

For the compatible strings below the following supplies are required:
"qcom,q6v5-pil"
"qcom,msm8916-mss-pil",
"qcom,msm8974-mss-pil"
- cx-supply:
- mss-supply:
- mx-supply:
- pll-supply:
Usage: required
Value type: <phandle>
Definition: reference to the regulators to be held on behalf of the
booting of the Hexagon core

...and if msm8996 actually needs "pll-supply", you could add in...

For the compatible strings below the following supplies are required:
"qcom,msm8996-mss-pil"
- pll-supply:
Usage: required
Value type: <phandle>
Definition: reference to the regulators to be held on behalf of the
booting of the Hexagon core

#9 lore raw

Re: [PATCH v2 3/7] dt-bindings: remoteproc: qcom: Fixup regulator dependencies

Hi Doug,
Thanks for the review :)

Quoted 23 lines · Doug Anderson

Hi,

Quoted 20 lines · Sibi Sankar

Fixup regulator supply dependencies for Q6V5 MSS on MSM996 SoCs.

---
 Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +0 -0
index 780adc043b37..98894e6ad456 100644
@@ -76,7 +76,9 @@ on the Qualcomm Hexagon core.
        Usage: required
        Value type: <phandle>
        Definition: reference to the regulators to be held on behalf 

of the

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +2 -1
-                   booting of the Hexagon core
+                   booting of the Hexagon core on MSM8916 SoCs
+                   reference to the pll-supply regulator to be held 
Quoted 9 lines
Quoted 2 lines

on behalf
+ of the booting of the Hexagon core on MSM8996 SoCs

The prose gets in the way and doesn't add anything. I also don't understand what you're saying for msm8996. You're saying that "pll-supply" is required there but none of the others? That doesn't seem to be true in the code I have in front of me, but maybe I'm missing some patch. For me, I'd write:

AFAIK, only the exceptions are captured. But your suggestion seems more simple/complete. Perhaps I'll replace SoCs instead of compatibles? Anyway I'll wait for Bjorn/Rob's preference.

Quoted 23 lines

For the compatible strings below the following supplies are required:
"qcom,q6v5-pil"
"qcom,msm8916-mss-pil",
"qcom,msm8974-mss-pil"
- cx-supply:
- mss-supply:
- mx-supply:
- pll-supply:
Usage: required
Value type: <phandle>
Definition: reference to the regulators to be held on behalf of the
booting of the Hexagon core

...and if msm8996 actually needs "pll-supply", you could add in...

For the compatible strings below the following supplies are required:
"qcom,msm8996-mss-pil"
- pll-supply:
Usage: required
Value type: <phandle>
Definition: reference to the regulators to be held on behalf of the
booting of the Hexagon core

Signature · 3 lines
-- Sibi Sankar --
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project.
#10 lore raw

PATCH v2 4/7 dt-bindings: remoteproc: qcom: Add power-domain bindings for Q6V5

In reply to Sibi Sankar

Add power-domain bindings for Q6V5 MSS on MSM8996 and SDM845 SoCs.

---

v2:
* Add load_state power-domain
* List cx and mx power-domains for MSM8996

 .../devicetree/bindings/remoteproc/qcom,q6v5.txt | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +16 -0
index 98894e6ad456..50695cd86397 100644
@@ -80,6 +80,22 @@ on the Qualcomm Hexagon core.
 		    reference to the pll-supply regulator to be held on behalf
 		    of the booting of the Hexagon core on MSM8996 SoCs
 
+- power-domains:
+	Usage: required
+	Value type: <phandle>
+	Definition: reference to the list of 2 power-domains for the modem
+		    sub-system on MSM8996 SoCs
+		    reference to the list of 4 power-domains for the modem
+		    sub-system on SDM845 SoCs
+
+- power-domain-names:
+	Usage: required
+	Value type: <stringlist>
+	Definition: must be "cx", "mx" for the modem sub-system on MSM8996
+		    SoCs
+		    must be "cx", "mx", "mss", "load_state" for the modem
+		    sub-system on SDM845 SoCs
+
 - qcom,smem-states:
 	Usage: required
 	Value type: <phandle>
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project
#11 lore raw

Re: [PATCH v2 4/7] dt-bindings: remoteproc: qcom: Add power-domain bindings for Q6V5

Hi,

Quoted 19 lines · Sibi Sankar

Add power-domain bindings for Q6V5 MSS on MSM8996 and SDM845 SoCs.

---

v2:
* Add load_state power-domain
* List cx and mx power-domains for MSM8996

 .../devicetree/bindings/remoteproc/qcom,q6v5.txt | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +0 -0
index 98894e6ad456..50695cd86397 100644

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt on the Qualcomm Hexagon core. · line 87

Quoted 8 lines
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +5 -0
@@ -80,6 +80,22 @@ on the Qualcomm Hexagon core.
                    reference to the pll-supply regulator to be held on behalf
                    of the booting of the Hexagon core on MSM8996 SoCs
 
+- power-domains:
+       Usage: required
+       Value type: <phandle>
+       Definition: reference to the list of 2 power-domains for the modem
+                   sub-system on MSM8996 SoCs

This is truly required for msm8996 SoCs? The code I'm looking at doesn't try to get these power domains for 8996 so presumably you're breaking backward compatibility with old device tree files by making this required now. I don't personally know how widespread msm8996 usage is w/ upstream, so I'd let Bjorn comment on whether he thinks this is OK.

As with the other patches in this series, I personally prefer less prose and more lists / tables of exactly what is required for which compatible string.

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt on the Qualcomm Hexagon core. · line 97

Quoted 10 lines
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +10 -0
+                   reference to the list of 4 power-domains for the modem
+                   sub-system on SDM845 SoCs
+
+- power-domain-names:
+       Usage: required
+       Value type: <stringlist>
+       Definition: must be "cx", "mx" for the modem sub-system on MSM8996
+                   SoCs
+                   must be "cx", "mx", "mss", "load_state" for the modem
+                   sub-system on SDM845 SoCs

I haven't see a patch for using "load_state". Can you point at it? I guess this was "aop" in your last version?

-Doug

#12 lore raw

Re: [PATCH v2 4/7] dt-bindings: remoteproc: qcom: Add power-domain bindings for Q6V5

Hi Doug,
Thanks for the review :)

Quoted 27 lines · Doug Anderson

Hi,

Quoted 24 lines · Sibi Sankar

Add power-domain bindings for Q6V5 MSS on MSM8996 and SDM845 SoCs.

---

v2:
* Add load_state power-domain
* List cx and mx power-domains for MSM8996

 .../devicetree/bindings/remoteproc/qcom,q6v5.txt | 16 
++++++++++++++++
 1 file changed, 16 insertions(+)
Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +0 -0
index 98894e6ad456..50695cd86397 100644
@@ -80,6 +80,22 @@ on the Qualcomm Hexagon core.
                    reference to the pll-supply regulator to be held 

on behalf
of the booting of the Hexagon core on MSM8996 SoCs

Quoted 14 lines
Quoted 6 lines

+- power-domains:

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +3 -0
+       Usage: required
+       Value type: <phandle>
+       Definition: reference to the list of 2 power-domains for the 

modem
+ sub-system on MSM8996 SoCs

This is truly required for msm8996 SoCs? The code I'm looking at doesn't try to get these power domains for 8996 so presumably you're breaking backward compatibility with old device tree files by making this required now. I don't personally know how widespread msm8996 usage is w/ upstream, so I'd let Bjorn comment on whether he thinks this is OK.

This is one of the reasons why the dt node for mss on 8996 has not been posted/merged upstream. Hence backward compatibility is not broken yet in mainline :) .. However it will break on official linaro integration releases (old dt + new kernel)

Quoted 19 lines

As with the other patches in this series, I personally prefer less prose and more lists / tables of exactly what is required for which compatible string.

Quoted 11 lines

+ reference to the list of 4 power-domains for the
modem

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +6 -0
+                   sub-system on SDM845 SoCs
+
+- power-domain-names:
+       Usage: required
+       Value type: <stringlist>
+       Definition: must be "cx", "mx" for the modem sub-system on 

MSM8996

Documentation/devicetree/bindings/remoteproc/qcom,q6v5.txt +2 -0
+                   SoCs
+                   must be "cx", "mx", "mss", "load_state" for the 
Quoted 2 lines

modem
+ sub-system on SDM845 SoCs

Quoted 3 lines

I haven't see a patch for using "load_state". Can you point at it? I guess this was "aop" in your last version?

using load_state was Bjorn's suggestion and seemed more appropriate than aop

Quoted 2 lines

-Doug

Signature · 3 lines
-- Sibi Sankar --
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project.
#13 lore raw

PATCH v2 5/7 arm64: dts: qcom: sdm845: Add Q6V5 MSS node

In reply to Sibi Sankar

This patch adds Q6V5 MSS remoteproc node for SDM845 SoCs.

---

v2:
* Fixed style changes
* Added missing clocks in the dt-bindings
* Split mss remoteproc node into a number of patches

This patch depends on the mpss and mba memory regions and pdc reset node.
https://patchwork.kernel.org/patch/10662089/
https://patchwork.kernel.org/patch/10657325/

 arch/arm64/boot/dts/qcom/sdm845.dtsi | 52 ++++++++++++++++++++++++++++
 1 file changed, 52 insertions(+)
arch/arm64/boot/dts/qcom/sdm845.dtsi +52 -0
index 5da9fa1feb8a..33ff8668828f 100644
@@ -1366,6 +1366,58 @@
 			};
 		};
 
+		remoteproc@4080000 {
+			compatible = "qcom,sdm845-mss-pil";
+			reg = <0x04080000 0x408>, <0x04180000 0x48>;
+			reg-names = "qdsp6", "rmb";
+
+			interrupts-extended =
+				<&intc GIC_SPI 266 IRQ_TYPE_EDGE_RISING>,
+				<&modem_smp2p_in 0 IRQ_TYPE_EDGE_RISING>,
+				<&modem_smp2p_in 1 IRQ_TYPE_EDGE_RISING>,
+				<&modem_smp2p_in 2 IRQ_TYPE_EDGE_RISING>,
+				<&modem_smp2p_in 3 IRQ_TYPE_EDGE_RISING>;
+			interrupt-names = "wdog", "fatal", "ready",
+					  "handover", "stop-ack";
+
+			clocks = <&gcc GCC_MSS_CFG_AHB_CLK>,
+				 <&gcc GCC_MSS_Q6_MEMNOC_AXI_CLK>,
+				 <&gcc GCC_BOOT_ROM_AHB_CLK>,
+				 <&gcc GCC_MSS_GPLL0_DIV_CLK_SRC>,
+				 <&gcc GCC_MSS_SNOC_AXI_CLK>,
+				 <&gcc GCC_MSS_MFAB_AXIS_CLK>,
+				 <&gcc GCC_PRNG_AHB_CLK>,
+				 <&rpmhcc RPMH_CXO_CLK>;
+			clock-names = "iface", "bus", "mem", "gpll0_mss",
+				      "snoc_axi", "mnoc_axi", "prng", "xo";
+
+			qcom,smem-states = <&modem_smp2p_out 0>;
+			qcom,smem-state-names = "stop";
+
+			resets = <&aoss_reset AOSS_CC_MSS_RESTART>,
+				 <&pdc_reset PDC_MODEM_SYNC_RESET>;
+			reset-names = "mss_restart", "pdc_reset";
+
+			qcom,halt-regs = <&tcsr_mutex_regs
+						0x23000 0x25000 0x24000>;
+
+			mba {
+				memory-region = <&mba_region>;
+			};
+
+			mpss {
+				memory-region = <&mpss_region>;
+			};
+
+			glink-edge {
+				interrupts = <GIC_SPI 449 IRQ_TYPE_EDGE_RISING>;
+				label = "modem";
+				qcom,remote-pid = <1>;
+				mboxes = <&apss_shared 12>;
+				mbox-names = "mpss_smem";
+			};
+		};
+
 		usb_1_hsphy: phy@88e2000 {
 			compatible = "qcom,sdm845-qusb2-phy";
 			reg = <0x88e2000 0x400>;
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project
#14 lore raw

Re: [PATCH v2 5/7] arm64: dts: qcom: sdm845: Add Q6V5 MSS node

Hi,

Quoted 10 lines · Sibi Sankar

This patch adds Q6V5 MSS remoteproc node for SDM845 SoCs.

---

v2:
* Fixed style changes
* Added missing clocks in the dt-bindings
* Split mss remoteproc node into a number of patches

I know there was some off-list suggestion to split this into a number of patches, but to actually make that useful to anyone we'd actually need to _also_ post up patches to make the driver probe / work without these power domains. ...and as per other discussions it's kinda "lucky" that it happens to work without them and Bjorn wasn't supportive of making this optional.

So I'd actually fold patch 6 into patch 5 and focus on getting the "aoss_qmp_pd" landed sooner rather than later.

Keeping the "shutdown-ack" as a separate patch makes sense though since the bindings currently list that as "optional" and I guess things work OK w/out it.

Once patch #6 is folded into patch #5 feel free to add my Reviewed-by tag.

#15 lore raw

Re: [PATCH v2 5/7] arm64: dts: qcom: sdm845: Add Q6V5 MSS node

Hi Doug,
Thanks for the review :)

Quoted 21 lines · Doug Anderson

Hi,

Quoted 10 lines · Sibi Sankar

This patch adds Q6V5 MSS remoteproc node for SDM845 SoCs.

---

v2:
* Fixed style changes
* Added missing clocks in the dt-bindings
* Split mss remoteproc node into a number of patches

I know there was some off-list suggestion to split this into a number of patches, but to actually make that useful to anyone we'd actually need to _also_ post up patches to make the driver probe / work without these power domains. ...and as per other discussions it's kinda "lucky" that it happens to work without them and Bjorn wasn't supportive of making this optional.

Quoted 3 lines

So I'd actually fold patch 6 into patch 5 and focus on getting the "aoss_qmp_pd" landed sooner rather than later.

I'll fold them in v3

Quoted 7 lines

Keeping the "shutdown-ack" as a separate patch makes sense though since the bindings currently list that as "optional" and I guess things work OK w/out it.

Once patch #6 is folded into patch #5 feel free to add my Reviewed-by tag.

okay

Signature · 3 lines
-- Sibi Sankar --
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project.
#16 lore raw

Re: [PATCH v2 5/7] arm64: dts: qcom: sdm845: Add Q6V5 MSS node

In reply to Sibi Sankar
Thank you for the patch! Yet something to improve:

Hi Sibi,

Thank you for the patch! Yet something to improve:

[auto build test ERROR on agross/for-next]
[also build test ERROR on v4.20-rc7 next-20181218]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]

url:    https://github.com/0day-ci/linux/commits/Sibi-Sankar/dt-bindings-soc-qcom-Add-remote-pid-binding-for-GLINK-SMEM/20181217-205522
base:   https://git.kernel.org/pub/scm/linux/kernel/git/agross/linux.git for-next
config: arm64-defconfig (attached as .config)
compiler: aarch64-linux-gnu-gcc (Debian 7.2.0-11) 7.2.0

reproduce:
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross
# save the attached .config to linux build tree
GCC_VERSION=7.2.0 make.cross ARCH=arm64

All errors (new ones prefixed by >>):

>> Error: arch/arm64/boot/dts/qcom/sdm845.dtsi:1380.18-19 syntax error
>> FATAL ERROR: Unable to parse input tree

---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all Intel Corporation

#17 lore raw

PATCH v2 6/7 arm64: dts: qcom: sdm845: Add power-domain for Q6V5 MSS node

In reply to Sibi Sankar

Add power-domains cx, mx, mss and load_state for Q6V5 MSS node.

---

This patch depends on the following bindings:
https://patchwork.kernel.org/patch/10725801/ - rpmhpd dt bindings
https://patchwork.kernel.org/patch/10725793/ - rpmhpd dt node
https://patchwork.kernel.org/patch/10678301/ - AOP QMP dt bindings

 arch/arm64/boot/dts/qcom/sdm845.dtsi | 6 ++++++
 1 file changed, 6 insertions(+)
arch/arm64/boot/dts/qcom/sdm845.dtsi +6 -0
index 33ff8668828f..56f5f55db9e2 100644
@@ -1401,6 +1401,12 @@
 			qcom,halt-regs = <&tcsr_mutex_regs
 						0x23000 0x25000 0x24000>;
 
+			power-domains = <&aoss_qmp_pd AOSS_QMP_LS_MODEM>,
+					<&rpmhpd SDM845_CX>,
+					<&rpmhpd SDM845_MX>,
+					<&rpmhpd SDM845_MSS>;
+			power-domain-names = "load_state", "cx", "mx", "mss";
+
 			mba {
 				memory-region = <&mba_region>;
 			};
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project
#18 lore raw

Re: [PATCH v2 6/7] arm64: dts: qcom: sdm845: Add power-domain for Q6V5 MSS node

Hi,

Quoted 6 lines · Sibi Sankar

Add power-domains cx, mx, mss and load_state for Q6V5 MSS node.

---

Quoted 7 lines

This patch depends on the following bindings:
https://patchwork.kernel.org/patch/10725801/ - rpmhpd dt bindings
https://patchwork.kernel.org/patch/10725793/ - rpmhpd dt node
https://patchwork.kernel.org/patch/10678301/ - AOP QMP dt bindings

 arch/arm64/boot/dts/qcom/sdm845.dtsi | 6 ++++++
 1 file changed, 6 insertions(+)

As per my comments on patch #5, I think this patch (AKA patch #6) should be folded in there.

arch/arm64/boot/dts/qcom/sdm845.dtsi line 1408

Quoted 13 lines
arch/arm64/boot/dts/qcom/sdm845.dtsi +5 -0
index 33ff8668828f..56f5f55db9e2 100644
@@ -1401,6 +1401,12 @@
                        qcom,halt-regs = <&tcsr_mutex_regs
                                                0x23000 0x25000 0x24000>;
 
+                       power-domains = <&aoss_qmp_pd AOSS_QMP_LS_MODEM>,
+                                       <&rpmhpd SDM845_CX>,
+                                       <&rpmhpd SDM845_MX>,
+                                       <&rpmhpd SDM845_MSS>;
+                       power-domain-names = "load_state", "cx", "mx", "mss";

I guess you changed this to "load_state" from "aop" before? Is there code that actually uses this?

-Doug

-Doug

#19 lore raw

Re: [PATCH v2 6/7] arm64: dts: qcom: sdm845: Add power-domain for Q6V5 MSS node

Hi Doug,
Thanks for the review :)

Quoted 14 lines · Doug Anderson

Hi,

Quoted 11 lines · Sibi Sankar

Add power-domains cx, mx, mss and load_state for Q6V5 MSS node.

---

This patch depends on the following bindings:
https://patchwork.kernel.org/patch/10725801/ - rpmhpd dt bindings
https://patchwork.kernel.org/patch/10725793/ - rpmhpd dt node
https://patchwork.kernel.org/patch/10678301/ - AOP QMP dt bindings

Quoted 6 lines
Quoted 2 lines
 arch/arm64/boot/dts/qcom/sdm845.dtsi | 6 ++++++
 1 file changed, 6 insertions(+)

As per my comments on patch #5, I think this patch (AKA patch #6) should be folded in there.

okay

Quoted 10 lines
Quoted 10 lines
arch/arm64/boot/dts/qcom/sdm845.dtsi +0 -0
index 33ff8668828f..56f5f55db9e2 100644
@@ -1401,6 +1401,12 @@
                        qcom,halt-regs = <&tcsr_mutex_regs
                                                0x23000 0x25000 
 0x24000>;
 
Quoted 11 lines

arch/arm64/boot/dts/qcom/sdm845.dtsi line 1411

Quoted 7 lines
arch/arm64/boot/dts/qcom/sdm845.dtsi +5 -0
+                       power-domains = <&aoss_qmp_pd 
 AOSS_QMP_LS_MODEM>,
+                                       <&rpmhpd SDM845_CX>,
+                                       <&rpmhpd SDM845_MX>,
+                                       <&rpmhpd SDM845_MSS>;
+                       power-domain-names = "load_state", "cx", "mx", 
 "mss";

I guess you changed this to "load_state" from "aop" before? Is there code that actually uses this?

Bjorn said he will be posting the patch
for handling power-domains for mss..

Quoted 4 lines
(quoted diff) +0 -2
-Doug
 
-Doug
Signature · 3 lines
-- Sibi Sankar --
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project.
#20 lore raw

PATCH v2 7/7 arm64: dts: qcom: sdm845: Add shutdown-ack for Q6V5 MSS node

In reply to Sibi Sankar

Add shutdown-ack smp2p interrupt for Q6V5 MSS node.

---

This patch depends on:
https://patchwork.kernel.org/patch/10691213/ - shutdown-ack dt bindings

 arch/arm64/boot/dts/qcom/sdm845.dtsi | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/qcom/sdm845.dtsi +4 -2
index 56f5f55db9e2..db17216a5bce 100644
@@ -1376,9 +1376,11 @@
 				<&modem_smp2p_in 0 IRQ_TYPE_EDGE_RISING>,
 				<&modem_smp2p_in 1 IRQ_TYPE_EDGE_RISING>,
 				<&modem_smp2p_in 2 IRQ_TYPE_EDGE_RISING>,
-				<&modem_smp2p_in 3 IRQ_TYPE_EDGE_RISING>;
+				<&modem_smp2p_in 3 IRQ_TYPE_EDGE_RISING>,
+				<&modem_smp2p_in 7 IRQ_TYPE_EDGE_RISING>;
 			interrupt-names = "wdog", "fatal", "ready",
-					  "handover", "stop-ack";
+					  "handover", "stop-ack",
+					  "shutdown-ack";
 
 			clocks = <&gcc GCC_MSS_CFG_AHB_CLK>,
 				 <&gcc GCC_MSS_Q6_MEMNOC_AXI_CLK>,
Signature · 2 lines
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project
Reviewed-by on v2 7/7
#21 lore raw

Re: [PATCH v2 7/7] arm64: dts: qcom: sdm845: Add shutdown-ack for Q6V5 MSS node

Hi,

Quoted 11 lines · Sibi Sankar

Add shutdown-ack smp2p interrupt for Q6V5 MSS node.

---

This patch depends on:
https://patchwork.kernel.org/patch/10691213/ - shutdown-ack dt bindings

 arch/arm64/boot/dts/qcom/sdm845.dtsi | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
#22 lore raw
In reply to Sibi Sankar

Hi,

Quoted 8 lines · Sibi Sankar

Add missing qcom,remote-pid dt binding required for GLINK SMEM which specifies the remote endpoint of the GLINK edge.

---
 Documentation/devicetree/bindings/soc/qcom/qcom,glink.txt | 5 +++++
 1 file changed, 5 insertions(+)
Fixes 2b41d6c8e696 ("dt-bindings: soc: qcom: Extend GLINK to cover SMEM")

Documentation/devicetree/bindings/soc/qcom/qcom,glink.txt edge. · line 27

Quoted 12 lines
Documentation/devicetree/bindings/soc/qcom/qcom,glink.txt +4 -0
index 0b8cc533ca83..59ae603ba520 100644
@@ -21,6 +21,11 @@ edge.
        Definition: should specify the IRQ used by the remote processor to
                    signal this processor about communication related events
 
+- qcom,remote-pid:
+       Usage: required for glink-smem
+       Value type: <u32>
+       Definition: specifies the identfier of the remote endpoint of this edge

s/identfier/identifier/

Other than the typo this seems right to me. Feel free to add my Reviewed-by tag when that's fixed.

-Doug

#23 lore raw

Hi Doug,
Thanks for the review :)

Quoted 25 lines · Doug Anderson

Hi,

Quoted 8 lines · Sibi Sankar

Add missing qcom,remote-pid dt binding required for GLINK SMEM which specifies the remote endpoint of the GLINK edge.

---
 Documentation/devicetree/bindings/soc/qcom/qcom,glink.txt | 5 +++++
 1 file changed, 5 insertions(+)
Fixes 2b41d6c8e696 ("dt-bindings: soc: qcom: Extend GLINK to cover SMEM")
Quoted 10 lines
Documentation/devicetree/bindings/soc/qcom/qcom,glink.txt +0 -0
index 0b8cc533ca83..59ae603ba520 100644
@@ -21,6 +21,11 @@ edge.
        Definition: should specify the IRQ used by the remote 

processor to
signal this processor about communication related
events

Quoted 8 lines
Quoted 5 lines

+- qcom,remote-pid:

Documentation/devicetree/bindings/soc/qcom/qcom,glink.txt +3 -0
+       Usage: required for glink-smem
+       Value type: <u32>
+       Definition: specifies the identfier of the remote endpoint of 

this edge

s/identfier/identifier/

missed this, will correct it in v3.

Quoted 6 lines

Other than the typo this seems right to me. Feel free to add my Reviewed-by tag when that's fixed.

-Doug

Signature · 3 lines
-- Sibi Sankar --
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project.