Re: [RFC PATCH 13/13] of: Remove unused variable overlay_symbols

[Frank Rowand]

On 10/27/16 09:53, Frank Rowand wrote:
> On 10/27/16 09:27, Frank Rowand wrote:
>> On 10/27/16 07:41, Pantelis Antoniou wrote:
>>> Hi Frank,
>>>
>>>
 On Oct 25, 2016, at 23:59 , frowand.l...@gmail.com wrote:

 From: Frank Rowand 

 This unused variable is a reminder that symbols in overlays are
 not available to subsequent overlays.  If such a feature is
 desired then there are several ways it could be implemented.

>>>
>>> Please don’t apply that. There’s a patch that actually imports
>>> the symbol table from overlays that subsequent operations
>>> work.
>>>
>>> Please see:
>>>
>>> https://patchwork.kernel.org/patch/9104701/
>>
>> Thanks for the pointer!  When the import symbols patch is applied
>> then the comment in my patch header becomes incorrect.  I will
>> change the patch comment to act is if the import symbols patch
>> is in place.
>>
>> But the node pointer that my patch removes is still not used
>> for anything, even if the import symbols patch is applied.
>>
>> Am I missing something?
> 
> I was missing a later patch in the symbol import patch set that
> updated resolver.c to use the imported symbols.  I'll go look
> at that.

Crap.  I misread the file name in the patch that I thought was
updating resolver.c.  It was actually overlay.c.  So I am back
to my question of:

   Am I missing something?

Re: [RFC PATCH 13/13] of: Remove unused variable overlay_symbols

[Frank Rowand]

On 10/27/16 09:53, Frank Rowand wrote:
> On 10/27/16 09:27, Frank Rowand wrote:
>> On 10/27/16 07:41, Pantelis Antoniou wrote:
>>> Hi Frank,
>>>
>>>
 On Oct 25, 2016, at 23:59 , frowand.l...@gmail.com wrote:

 From: Frank Rowand 

 This unused variable is a reminder that symbols in overlays are
 not available to subsequent overlays.  If such a feature is
 desired then there are several ways it could be implemented.

>>>
>>> Please don’t apply that. There’s a patch that actually imports
>>> the symbol table from overlays that subsequent operations
>>> work.
>>>
>>> Please see:
>>>
>>> https://patchwork.kernel.org/patch/9104701/
>>
>> Thanks for the pointer!  When the import symbols patch is applied
>> then the comment in my patch header becomes incorrect.  I will
>> change the patch comment to act is if the import symbols patch
>> is in place.
>>
>> But the node pointer that my patch removes is still not used
>> for anything, even if the import symbols patch is applied.
>>
>> Am I missing something?
> 
> I was missing a later patch in the symbol import patch set that
> updated resolver.c to use the imported symbols.  I'll go look
> at that.

Crap.  I misread the file name in the patch that I thought was
updating resolver.c.  It was actually overlay.c.  So I am back
to my question of:

   Am I missing something?

Re: [RFC PATCH 13/13] of: Remove unused variable overlay_symbols

[Frank Rowand]

On 10/27/16 09:27, Frank Rowand wrote:
> On 10/27/16 07:41, Pantelis Antoniou wrote:
>> Hi Frank,
>>
>>
>>> On Oct 25, 2016, at 23:59 , frowand.l...@gmail.com wrote:
>>>
>>> From: Frank Rowand 
>>>
>>> This unused variable is a reminder that symbols in overlays are
>>> not available to subsequent overlays.  If such a feature is
>>> desired then there are several ways it could be implemented.
>>>
>>
>> Please don’t apply that. There’s a patch that actually imports
>> the symbol table from overlays that subsequent operations
>> work.
>>
>> Please see:
>>
>> https://patchwork.kernel.org/patch/9104701/
> 
> Thanks for the pointer!  When the import symbols patch is applied
> then the comment in my patch header becomes incorrect.  I will
> change the patch comment to act is if the import symbols patch
> is in place.
> 
> But the node pointer that my patch removes is still not used
> for anything, even if the import symbols patch is applied.
> 
> Am I missing something?

I was missing a later patch in the symbol import patch set that
updated resolver.c to use the imported symbols.  I'll go look
at that.

Re: [RFC PATCH 13/13] of: Remove unused variable overlay_symbols

[Frank Rowand]

On 10/27/16 09:27, Frank Rowand wrote:
> On 10/27/16 07:41, Pantelis Antoniou wrote:
>> Hi Frank,
>>
>>
>>> On Oct 25, 2016, at 23:59 , frowand.l...@gmail.com wrote:
>>>
>>> From: Frank Rowand 
>>>
>>> This unused variable is a reminder that symbols in overlays are
>>> not available to subsequent overlays.  If such a feature is
>>> desired then there are several ways it could be implemented.
>>>
>>
>> Please don’t apply that. There’s a patch that actually imports
>> the symbol table from overlays that subsequent operations
>> work.
>>
>> Please see:
>>
>> https://patchwork.kernel.org/patch/9104701/
> 
> Thanks for the pointer!  When the import symbols patch is applied
> then the comment in my patch header becomes incorrect.  I will
> change the patch comment to act is if the import symbols patch
> is in place.
> 
> But the node pointer that my patch removes is still not used
> for anything, even if the import symbols patch is applied.
> 
> Am I missing something?

I was missing a later patch in the symbol import patch set that
updated resolver.c to use the imported symbols.  I'll go look
at that.

Re: [RFC PATCH 13/13] of: Remove unused variable overlay_symbols

[Frank Rowand]

On 10/27/16 07:41, Pantelis Antoniou wrote:
> Hi Frank,
> 
> 
>> On Oct 25, 2016, at 23:59 , frowand.l...@gmail.com wrote:
>>
>> From: Frank Rowand 
>>
>> This unused variable is a reminder that symbols in overlays are
>> not available to subsequent overlays.  If such a feature is
>> desired then there are several ways it could be implemented.
>>
> 
> Please don’t apply that. There’s a patch that actually imports
> the symbol table from overlays that subsequent operations
> work.
> 
> Please see:
> 
> https://patchwork.kernel.org/patch/9104701/

Thanks for the pointer!  When the import symbols patch is applied
then the comment in my patch header becomes incorrect.  I will
change the patch comment to act is if the import symbols patch
is in place.

But the node pointer that my patch removes is still not used
for anything, even if the import symbols patch is applied.

Am I missing something?


> 
>> Signed-off-by: Frank Rowand 
>> ---
>> drivers/of/resolver.c | 5 +
>> 1 file changed, 1 insertion(+), 4 deletions(-)
>>
>> diff --git a/drivers/of/resolver.c b/drivers/of/resolver.c
>> index 3f7cf569c7ea..b48d16200ccd 100644
>> --- a/drivers/of/resolver.c
>> +++ b/drivers/of/resolver.c
>> @@ -272,7 +272,7 @@ static int adjust_local_phandle_references(struct 
>> device_node *local_fixups,
>> int of_resolve_phandles(struct device_node *overlay)
>> {
>>  struct device_node *child, *local_fixups, *refnode;
>> -struct device_node *tree_symbols, *overlay_symbols, *overlay_fixups;
>> +struct device_node *tree_symbols, *overlay_fixups;
>>  struct property *prop;
>>  const char *refpath;
>>  phandle phandle, phandle_delta;
>> @@ -302,12 +302,9 @@ int of_resolve_phandles(struct device_node *overlay)
>>  if (err)
>>  goto err_out;
>>
>> -overlay_symbols = NULL;
>>  overlay_fixups = NULL;
>>
>>  for_each_child_of_node(overlay, child) {
>> -if (!of_node_cmp(child->name, "__symbols__"))
>> -overlay_symbols = child;
>>  if (!of_node_cmp(child->name, "__fixups__"))
>>  overlay_fixups = child;
>>  }
>> -- 
>> 1.9.1
>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe devicetree" in
>> the body of a message to majord...@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 
> Regards
> 
> — Pantelis
> 
> 

Re: [RFC PATCH 13/13] of: Remove unused variable overlay_symbols

[Frank Rowand]

On 10/27/16 07:41, Pantelis Antoniou wrote:
> Hi Frank,
> 
> 
>> On Oct 25, 2016, at 23:59 , frowand.l...@gmail.com wrote:
>>
>> From: Frank Rowand 
>>
>> This unused variable is a reminder that symbols in overlays are
>> not available to subsequent overlays.  If such a feature is
>> desired then there are several ways it could be implemented.
>>
> 
> Please don’t apply that. There’s a patch that actually imports
> the symbol table from overlays that subsequent operations
> work.
> 
> Please see:
> 
> https://patchwork.kernel.org/patch/9104701/

Thanks for the pointer!  When the import symbols patch is applied
then the comment in my patch header becomes incorrect.  I will
change the patch comment to act is if the import symbols patch
is in place.

But the node pointer that my patch removes is still not used
for anything, even if the import symbols patch is applied.

Am I missing something?


> 
>> Signed-off-by: Frank Rowand 
>> ---
>> drivers/of/resolver.c | 5 +
>> 1 file changed, 1 insertion(+), 4 deletions(-)
>>
>> diff --git a/drivers/of/resolver.c b/drivers/of/resolver.c
>> index 3f7cf569c7ea..b48d16200ccd 100644
>> --- a/drivers/of/resolver.c
>> +++ b/drivers/of/resolver.c
>> @@ -272,7 +272,7 @@ static int adjust_local_phandle_references(struct 
>> device_node *local_fixups,
>> int of_resolve_phandles(struct device_node *overlay)
>> {
>>  struct device_node *child, *local_fixups, *refnode;
>> -struct device_node *tree_symbols, *overlay_symbols, *overlay_fixups;
>> +struct device_node *tree_symbols, *overlay_fixups;
>>  struct property *prop;
>>  const char *refpath;
>>  phandle phandle, phandle_delta;
>> @@ -302,12 +302,9 @@ int of_resolve_phandles(struct device_node *overlay)
>>  if (err)
>>  goto err_out;
>>
>> -overlay_symbols = NULL;
>>  overlay_fixups = NULL;
>>
>>  for_each_child_of_node(overlay, child) {
>> -if (!of_node_cmp(child->name, "__symbols__"))
>> -overlay_symbols = child;
>>  if (!of_node_cmp(child->name, "__fixups__"))
>>  overlay_fixups = child;
>>  }
>> -- 
>> 1.9.1
>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe devicetree" in
>> the body of a message to majord...@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 
> Regards
> 
> — Pantelis
> 
> 

Re: [RFC PATCH 13/13] of: Remove unused variable overlay_symbols

[Pantelis Antoniou]

Hi Frank,


> On Oct 25, 2016, at 23:59 , frowand.l...@gmail.com wrote:
> 
> From: Frank Rowand 
> 
> This unused variable is a reminder that symbols in overlays are
> not available to subsequent overlays.  If such a feature is
> desired then there are several ways it could be implemented.
> 

Please don’t apply that. There’s a patch that actually imports
the symbol table from overlays that subsequent operations
work.

Please see:

https://patchwork.kernel.org/patch/9104701/

> Signed-off-by: Frank Rowand 
> ---
> drivers/of/resolver.c | 5 +
> 1 file changed, 1 insertion(+), 4 deletions(-)
> 
> diff --git a/drivers/of/resolver.c b/drivers/of/resolver.c
> index 3f7cf569c7ea..b48d16200ccd 100644
> --- a/drivers/of/resolver.c
> +++ b/drivers/of/resolver.c
> @@ -272,7 +272,7 @@ static int adjust_local_phandle_references(struct 
> device_node *local_fixups,
> int of_resolve_phandles(struct device_node *overlay)
> {
>   struct device_node *child, *local_fixups, *refnode;
> - struct device_node *tree_symbols, *overlay_symbols, *overlay_fixups;
> + struct device_node *tree_symbols, *overlay_fixups;
>   struct property *prop;
>   const char *refpath;
>   phandle phandle, phandle_delta;
> @@ -302,12 +302,9 @@ int of_resolve_phandles(struct device_node *overlay)
>   if (err)
>   goto err_out;
> 
> - overlay_symbols = NULL;
>   overlay_fixups = NULL;
> 
>   for_each_child_of_node(overlay, child) {
> - if (!of_node_cmp(child->name, "__symbols__"))
> - overlay_symbols = child;
>   if (!of_node_cmp(child->name, "__fixups__"))
>   overlay_fixups = child;
>   }
> -- 
> 1.9.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe devicetree" in
> the body of a message to majord...@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Regards

— Pantelis

Re: [RFC PATCH 13/13] of: Remove unused variable overlay_symbols

[Pantelis Antoniou]

Hi Frank,


> On Oct 25, 2016, at 23:59 , frowand.l...@gmail.com wrote:
> 
> From: Frank Rowand 
> 
> This unused variable is a reminder that symbols in overlays are
> not available to subsequent overlays.  If such a feature is
> desired then there are several ways it could be implemented.
> 

Please don’t apply that. There’s a patch that actually imports
the symbol table from overlays that subsequent operations
work.

Please see:

https://patchwork.kernel.org/patch/9104701/

> Signed-off-by: Frank Rowand 
> ---
> drivers/of/resolver.c | 5 +
> 1 file changed, 1 insertion(+), 4 deletions(-)
> 
> diff --git a/drivers/of/resolver.c b/drivers/of/resolver.c
> index 3f7cf569c7ea..b48d16200ccd 100644
> --- a/drivers/of/resolver.c
> +++ b/drivers/of/resolver.c
> @@ -272,7 +272,7 @@ static int adjust_local_phandle_references(struct 
> device_node *local_fixups,
> int of_resolve_phandles(struct device_node *overlay)
> {
>   struct device_node *child, *local_fixups, *refnode;
> - struct device_node *tree_symbols, *overlay_symbols, *overlay_fixups;
> + struct device_node *tree_symbols, *overlay_fixups;
>   struct property *prop;
>   const char *refpath;
>   phandle phandle, phandle_delta;
> @@ -302,12 +302,9 @@ int of_resolve_phandles(struct device_node *overlay)
>   if (err)
>   goto err_out;
> 
> - overlay_symbols = NULL;
>   overlay_fixups = NULL;
> 
>   for_each_child_of_node(overlay, child) {
> - if (!of_node_cmp(child->name, "__symbols__"))
> - overlay_symbols = child;
>   if (!of_node_cmp(child->name, "__fixups__"))
>   overlay_fixups = child;
>   }
> -- 
> 1.9.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe devicetree" in
> the body of a message to majord...@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Regards

— Pantelis

[RFC PATCH 13/13] of: Remove unused variable overlay_symbols

[frowand . list]

From: Frank Rowand 

This unused variable is a reminder that symbols in overlays are
not available to subsequent overlays.  If such a feature is
desired then there are several ways it could be implemented.

Signed-off-by: Frank Rowand 
---
 drivers/of/resolver.c | 5 +
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/of/resolver.c b/drivers/of/resolver.c
index 3f7cf569c7ea..b48d16200ccd 100644
--- a/drivers/of/resolver.c
+++ b/drivers/of/resolver.c
@@ -272,7 +272,7 @@ static int adjust_local_phandle_references(struct 
device_node *local_fixups,
 int of_resolve_phandles(struct device_node *overlay)
 {
struct device_node *child, *local_fixups, *refnode;
-   struct device_node *tree_symbols, *overlay_symbols, *overlay_fixups;
+   struct device_node *tree_symbols, *overlay_fixups;
struct property *prop;
const char *refpath;
phandle phandle, phandle_delta;
@@ -302,12 +302,9 @@ int of_resolve_phandles(struct device_node *overlay)
if (err)
goto err_out;
 
-   overlay_symbols = NULL;
overlay_fixups = NULL;
 
for_each_child_of_node(overlay, child) {
-   if (!of_node_cmp(child->name, "__symbols__"))
-   overlay_symbols = child;
if (!of_node_cmp(child->name, "__fixups__"))
overlay_fixups = child;
}
-- 
1.9.1

[RFC PATCH 13/13] of: Remove unused variable overlay_symbols

[frowand . list]

From: Frank Rowand 

This unused variable is a reminder that symbols in overlays are
not available to subsequent overlays.  If such a feature is
desired then there are several ways it could be implemented.

Signed-off-by: Frank Rowand 
---
 drivers/of/resolver.c | 5 +
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/of/resolver.c b/drivers/of/resolver.c
index 3f7cf569c7ea..b48d16200ccd 100644
--- a/drivers/of/resolver.c
+++ b/drivers/of/resolver.c
@@ -272,7 +272,7 @@ static int adjust_local_phandle_references(struct 
device_node *local_fixups,
 int of_resolve_phandles(struct device_node *overlay)
 {
struct device_node *child, *local_fixups, *refnode;
-   struct device_node *tree_symbols, *overlay_symbols, *overlay_fixups;
+   struct device_node *tree_symbols, *overlay_fixups;
struct property *prop;
const char *refpath;
phandle phandle, phandle_delta;
@@ -302,12 +302,9 @@ int of_resolve_phandles(struct device_node *overlay)
if (err)
goto err_out;
 
-   overlay_symbols = NULL;
overlay_fixups = NULL;
 
for_each_child_of_node(overlay, child) {
-   if (!of_node_cmp(child->name, "__symbols__"))
-   overlay_symbols = child;
if (!of_node_cmp(child->name, "__fixups__"))
overlay_fixups = child;
}
-- 
1.9.1