Re: [PATCH,RESEND2] regulator: fixed, gpio: dt: regulator-name is required property

[Mark Brown]

On Wed, Feb 14, 2018 at 11:00:12AM +0100, Harald Geyer wrote:
> Mark Brown writes:

> > No, a content free ping is when you just send a reply saying something
> > to the effect of "hey, what's going on with this?".

> Well, I did quote the entire patch and gave a summary of the status
> as far as I know it. So I really don't understand why I got the
> "no content free pings" response ...

Which just boils down to a "what's going on with this" message.

> > As I said in my
> > reply best case the answer is going to be "I have no idea, you need to
> > resend since I'll need the patch to do anything with it" and worst case
> > it just won't get seen at all if the thing really did get buried
> > somehow.

> So you are saying you don't want pings as followups to patches at all, if
> the patch is old enough that it probably got lost?

> I guess that would make sense too, but again it's not obvious to read
> the canned response that way.

No, I'm saying don't send pings at all.  Resend patches if you think
they've got lost, and as normal when you're sending a patch it should
start off a new thread.  The goal is to send something that can be
directly acted on when it's seen rather than requring another round of
mails.


signature.asc
Description: PGP signature

Re: [PATCH,RESEND2] regulator: fixed, gpio: dt: regulator-name is required property

[Mark Brown]

On Wed, Feb 14, 2018 at 11:00:12AM +0100, Harald Geyer wrote:
> Mark Brown writes:

> > No, a content free ping is when you just send a reply saying something
> > to the effect of "hey, what's going on with this?".

> Well, I did quote the entire patch and gave a summary of the status
> as far as I know it. So I really don't understand why I got the
> "no content free pings" response ...

Which just boils down to a "what's going on with this" message.

> > As I said in my
> > reply best case the answer is going to be "I have no idea, you need to
> > resend since I'll need the patch to do anything with it" and worst case
> > it just won't get seen at all if the thing really did get buried
> > somehow.

> So you are saying you don't want pings as followups to patches at all, if
> the patch is old enough that it probably got lost?

> I guess that would make sense too, but again it's not obvious to read
> the canned response that way.

No, I'm saying don't send pings at all.  Resend patches if you think
they've got lost, and as normal when you're sending a patch it should
start off a new thread.  The goal is to send something that can be
directly acted on when it's seen rather than requring another round of
mails.


signature.asc
Description: PGP signature

Re: [PATCH,RESEND2] regulator: fixed, gpio: dt: regulator-name is required property

[Harald Geyer]

Mark Brown writes:
> On Tue, Feb 13, 2018 at 02:43:08PM +, Harald Geyer wrote:
> 
> > BTW, it took me a bit to figure out that by "content free ping" you are
> > refering to the quoting of the patch. Maybe you can reword your canned
> > response to make it easier to understand.
> 
> No, a content free ping is when you just send a reply saying something
> to the effect of "hey, what's going on with this?".

Well, I did quote the entire patch and gave a summary of the status
as far as I know it. So I really don't understand why I got the
"no content free pings" response ...

> As I said in my
> reply best case the answer is going to be "I have no idea, you need to
> resend since I'll need the patch to do anything with it" and worst case
> it just won't get seen at all if the thing really did get buried
> somehow.

So you are saying you don't want pings as followups to patches at all, if
the patch is old enough that it probably got lost?

I guess that would make sense too, but again it's not obvious to read
the canned response that way.

Re: [PATCH,RESEND2] regulator: fixed, gpio: dt: regulator-name is required property

[Harald Geyer]

Mark Brown writes:
> On Tue, Feb 13, 2018 at 02:43:08PM +, Harald Geyer wrote:
> 
> > BTW, it took me a bit to figure out that by "content free ping" you are
> > refering to the quoting of the patch. Maybe you can reword your canned
> > response to make it easier to understand.
> 
> No, a content free ping is when you just send a reply saying something
> to the effect of "hey, what's going on with this?".

Well, I did quote the entire patch and gave a summary of the status
as far as I know it. So I really don't understand why I got the
"no content free pings" response ...

> As I said in my
> reply best case the answer is going to be "I have no idea, you need to
> resend since I'll need the patch to do anything with it" and worst case
> it just won't get seen at all if the thing really did get buried
> somehow.

So you are saying you don't want pings as followups to patches at all, if
the patch is old enough that it probably got lost?

I guess that would make sense too, but again it's not obvious to read
the canned response that way.

Re: [PATCH,RESEND2] regulator: fixed, gpio: dt: regulator-name is required property

[Mark Brown]

On Tue, Feb 13, 2018 at 02:43:08PM +, Harald Geyer wrote:

> BTW, it took me a bit to figure out that by "content free ping" you are
> refering to the quoting of the patch. Maybe you can reword your canned
> response to make it easier to understand.

No, a content free ping is when you just send a reply saying something
to the effect of "hey, what's going on with this?".  As I said in my
reply best case the answer is going to be "I have no idea, you need to
resend since I'll need the patch to do anything with it" and worst case
it just won't get seen at all if the thing really did get buried
somehow.  This is unfortunately something that happens upstream, things
fall through the cracks sometimes for various reasons (eg, if a new
patch gets sent in the middle of a thread sometimes it'll get deleted
along with the reset of the thread).  


signature.asc
Description: PGP signature

Re: [PATCH,RESEND2] regulator: fixed, gpio: dt: regulator-name is required property

[Mark Brown]

On Tue, Feb 13, 2018 at 02:43:08PM +, Harald Geyer wrote:

> BTW, it took me a bit to figure out that by "content free ping" you are
> refering to the quoting of the patch. Maybe you can reword your canned
> response to make it easier to understand.

No, a content free ping is when you just send a reply saying something
to the effect of "hey, what's going on with this?".  As I said in my
reply best case the answer is going to be "I have no idea, you need to
resend since I'll need the patch to do anything with it" and worst case
it just won't get seen at all if the thing really did get buried
somehow.  This is unfortunately something that happens upstream, things
fall through the cracks sometimes for various reasons (eg, if a new
patch gets sent in the middle of a thread sometimes it'll get deleted
along with the reset of the thread).  


signature.asc
Description: PGP signature

[PATCH,RESEND2] regulator: fixed, gpio: dt: regulator-name is required property

[Harald Geyer]

These two drivers fail to probe if no name is provided. For details see:
https://www.spinics.net/lists/kernel/msg2457515.html

Signed-off-by: Harald Geyer 
Acked-by: Rob Herring 
---
This is already the second resend for this trivial device probing bug
fix. It was written to your specification, got an ACK from devicetree
and has no unaddressed review comments, so I can't see what might be
holding it back. If there is still anything wrong with it, I'd appreciate
actual feedback instead of a canned response or silence.

BTW, it took me a bit to figure out that by "content free ping" you are
refering to the quoting of the patch. Maybe you can reword your canned
response to make it easier to understand.

The ACK was added in the (first) resend.

 Documentation/devicetree/bindings/regulator/fixed-regulator.txt | 1 +
 Documentation/devicetree/bindings/regulator/gpio-regulator.txt  | 2 ++
 2 files changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/regulator/fixed-regulator.txt 
b/Documentation/devicetree/bindings/regulator/fixed-regulator.txt
index 4fae41d..0c2a6c8 100644
--- a/Documentation/devicetree/bindings/regulator/fixed-regulator.txt
+++ b/Documentation/devicetree/bindings/regulator/fixed-regulator.txt
@@ -2,6 +2,7 @@ Fixed Voltage regulators
 
 Required properties:
 - compatible: Must be "regulator-fixed";
+- regulator-name: Defined in regulator.txt as optional, but required here.
 
 Optional properties:
 - gpio: gpio to use for enable control
diff --git a/Documentation/devicetree/bindings/regulator/gpio-regulator.txt 
b/Documentation/devicetree/bindings/regulator/gpio-regulator.txt
index dd1ed78..1f49615 100644
--- a/Documentation/devicetree/bindings/regulator/gpio-regulator.txt
+++ b/Documentation/devicetree/bindings/regulator/gpio-regulator.txt
@@ -2,6 +2,8 @@ GPIO controlled regulators
 
 Required properties:
 - compatible   : Must be "regulator-gpio".
+- regulator-name   : Defined in regulator.txt as optional, but required
+ here.
 - states   : Selection of available voltages and GPIO configs.
   if there are no states, then use a fixed regulator
 
-- 
2.1.4

[PATCH,RESEND2] regulator: fixed, gpio: dt: regulator-name is required property

[Harald Geyer]

These two drivers fail to probe if no name is provided. For details see:
https://www.spinics.net/lists/kernel/msg2457515.html

Signed-off-by: Harald Geyer 
Acked-by: Rob Herring 
---
This is already the second resend for this trivial device probing bug
fix. It was written to your specification, got an ACK from devicetree
and has no unaddressed review comments, so I can't see what might be
holding it back. If there is still anything wrong with it, I'd appreciate
actual feedback instead of a canned response or silence.

BTW, it took me a bit to figure out that by "content free ping" you are
refering to the quoting of the patch. Maybe you can reword your canned
response to make it easier to understand.

The ACK was added in the (first) resend.

 Documentation/devicetree/bindings/regulator/fixed-regulator.txt | 1 +
 Documentation/devicetree/bindings/regulator/gpio-regulator.txt  | 2 ++
 2 files changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/regulator/fixed-regulator.txt 
b/Documentation/devicetree/bindings/regulator/fixed-regulator.txt
index 4fae41d..0c2a6c8 100644
--- a/Documentation/devicetree/bindings/regulator/fixed-regulator.txt
+++ b/Documentation/devicetree/bindings/regulator/fixed-regulator.txt
@@ -2,6 +2,7 @@ Fixed Voltage regulators
 
 Required properties:
 - compatible: Must be "regulator-fixed";
+- regulator-name: Defined in regulator.txt as optional, but required here.
 
 Optional properties:
 - gpio: gpio to use for enable control
diff --git a/Documentation/devicetree/bindings/regulator/gpio-regulator.txt 
b/Documentation/devicetree/bindings/regulator/gpio-regulator.txt
index dd1ed78..1f49615 100644
--- a/Documentation/devicetree/bindings/regulator/gpio-regulator.txt
+++ b/Documentation/devicetree/bindings/regulator/gpio-regulator.txt
@@ -2,6 +2,8 @@ GPIO controlled regulators
 
 Required properties:
 - compatible   : Must be "regulator-gpio".
+- regulator-name   : Defined in regulator.txt as optional, but required
+ here.
 - states   : Selection of available voltages and GPIO configs.
   if there are no states, then use a fixed regulator
 
-- 
2.1.4