[PATCH] checks: Update #{size,address}-cells check for 'dma-ranges'

[PATCH] checks: Update #{size,address}-cells check for 'dma-ranges'

[Qun-Wei Lin]

The "dma-ranges" property value is a sequence of
	child-address  parent-address  child-size

The size of each field is determined by taking the child's
"#address-cells" value, the parent's "#address-cells" value,
and the child's "#size-cells" value.

However, in the following example, it gives a false alarm:
+-----------------------------------+---------------------------------------+
| ranges.dts                        | dma-ranges.dts                        |
+-----------------------------------+---------------------------------------+
| /dts-v1/;                         | /dts-v1/;                             |
|                                   |                                       |
| /{                                | /{                                    |
|   #address-cells = <1>;           |   #address-cells = <1>;               |
|                                   |                                       |
|   parent {                        |   parent {                            |
|     #address-cells = <1>;         |     #address-cells = <1>;             |
|     #size-cells = <1>;            |     #size-cells = <1>;                |
|     ranges = <0x0 0xe000 0x1000>; |     dma-ranges = <0x0 0xe000 0x1000>; |
|     child {                       |     child {                           |
|       ...                         |       ...                             |
|     };                            |     };                                |
|   };                              |   };                                  |
| };                                | };                                    |
+-----------------------------------+---------------------------------------+
| no warning                        | Warning (avoid_unnecessary_addr_size) |
+-----------------------------------+---------------------------------------+

Same as "ranges", it should not be reported in this check.

Signed-off-by: Qun-Wei Lin <qun-wei.lin-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>
---
 checks.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/checks.c b/checks.c
index afa8f23..607e0f2 100644
--- a/checks.c
+++ b/checks.c
@@ -1232,7 +1232,7 @@ static void check_avoid_unnecessary_addr_size(struct check *c, struct dt_info *d
 	if (!node->parent || node->addr_cells < 0 || node->size_cells < 0)
 		return;
 
-	if (get_property(node, "ranges") || !node->children)
+	if (get_property(node, "ranges") || get_property(node, "dma-ranges") || !node->children)
 		return;
 
 	for_each_child(node, child) {
@@ -1242,7 +1242,7 @@ static void check_avoid_unnecessary_addr_size(struct check *c, struct dt_info *d
 	}
 
 	if (!has_reg)
-		FAIL(c, dti, node, "unnecessary #address-cells/#size-cells without \"ranges\" or child \"reg\" property");
+		FAIL(c, dti, node, "unnecessary #address-cells/#size-cells without \"ranges\", \"dma-ranges\" or child \"reg\" property");
 }
 WARNING(avoid_unnecessary_addr_size, check_avoid_unnecessary_addr_size, NULL, &avoid_default_addr_size);
 
-- 
2.18.0

Re: [PATCH] checks: Update #{size,address}-cells check for 'dma-ranges'

[Rob Herring]

On Thu, Jan 12, 2023 at 6:59 AM Qun-Wei Lin <qun-wei.lin-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org> wrote:
>
> The "dma-ranges" property value is a sequence of
>         child-address  parent-address  child-size
>
> The size of each field is determined by taking the child's
> "#address-cells" value, the parent's "#address-cells" value,
> and the child's "#size-cells" value.
>
> However, in the following example, it gives a false alarm:
> +-----------------------------------+---------------------------------------+
> | ranges.dts                        | dma-ranges.dts                        |
> +-----------------------------------+---------------------------------------+
> | /dts-v1/;                         | /dts-v1/;                             |
> |                                   |                                       |
> | /{                                | /{                                    |
> |   #address-cells = <1>;           |   #address-cells = <1>;               |
> |                                   |                                       |
> |   parent {                        |   parent {                            |
> |     #address-cells = <1>;         |     #address-cells = <1>;             |
> |     #size-cells = <1>;            |     #size-cells = <1>;                |
> |     ranges = <0x0 0xe000 0x1000>; |     dma-ranges = <0x0 0xe000 0x1000>; |
> |     child {                       |     child {                           |
> |       ...                         |       ...                             |
> |     };                            |     };                                |
> |   };                              |   };                                  |
> | };                                | };                                    |
> +-----------------------------------+---------------------------------------+
> | no warning                        | Warning (avoid_unnecessary_addr_size) |
> +-----------------------------------+---------------------------------------+
>
> Same as "ranges", it should not be reported in this check.

Except that dma-ranges only makes sense if there is a ranges property.

Rob

Re: [PATCH] checks: Update #{size,address}-cells check for 'dma-ranges'

[David Gibson]

[-- Attachment #1: Type: text/plain, Size: 2684 bytes --]

On Thu, Jan 12, 2023 at 01:47:03PM -0600, Rob Herring wrote:
> On Thu, Jan 12, 2023 at 6:59 AM Qun-Wei Lin <qun-wei.lin-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org> wrote:
> >
> > The "dma-ranges" property value is a sequence of
> >         child-address  parent-address  child-size
> >
> > The size of each field is determined by taking the child's
> > "#address-cells" value, the parent's "#address-cells" value,
> > and the child's "#size-cells" value.
> >
> > However, in the following example, it gives a false alarm:
> > +-----------------------------------+---------------------------------------+
> > | ranges.dts                        | dma-ranges.dts                        |
> > +-----------------------------------+---------------------------------------+
> > | /dts-v1/;                         | /dts-v1/;                             |
> > |                                   |                                       |
> > | /{                                | /{                                    |
> > |   #address-cells = <1>;           |   #address-cells = <1>;               |
> > |                                   |                                       |
> > |   parent {                        |   parent {                            |
> > |     #address-cells = <1>;         |     #address-cells = <1>;             |
> > |     #size-cells = <1>;            |     #size-cells = <1>;                |
> > |     ranges = <0x0 0xe000 0x1000>; |     dma-ranges = <0x0 0xe000 0x1000>; |
> > |     child {                       |     child {                           |
> > |       ...                         |       ...                             |
> > |     };                            |     };                                |
> > |   };                              |   };                                  |
> > | };                                | };                                    |
> > +-----------------------------------+---------------------------------------+
> > | no warning                        | Warning (avoid_unnecessary_addr_size) |
> > +-----------------------------------+---------------------------------------+
> >
> > Same as "ranges", it should not be reported in this check.
> 
> Except that dma-ranges only makes sense if there is a ranges property.

Hm, does it?  It would certainly be an odd situation, but I can
imagine a sort of bus that can DMA, but which has no-addressable
registers on it.

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH] checks: Update #{size,address}-cells check for 'dma-ranges'

[David Gibson]

[-- Attachment #1: Type: text/plain, Size: 3552 bytes --]

On Thu, Jan 12, 2023 at 08:56:54PM +0800, Qun-Wei Lin wrote:
> The "dma-ranges" property value is a sequence of
> 	child-address  parent-address  child-size
> 
> The size of each field is determined by taking the child's
> "#address-cells" value, the parent's "#address-cells" value,
> and the child's "#size-cells" value.
> 
> However, in the following example, it gives a false alarm:
> +-----------------------------------+---------------------------------------+
> | ranges.dts                        | dma-ranges.dts                        |
> +-----------------------------------+---------------------------------------+
> | /dts-v1/;                         | /dts-v1/;                             |
> |                                   |                                       |
> | /{                                | /{                                    |
> |   #address-cells = <1>;           |   #address-cells = <1>;               |
> |                                   |                                       |
> |   parent {                        |   parent {                            |
> |     #address-cells = <1>;         |     #address-cells = <1>;             |
> |     #size-cells = <1>;            |     #size-cells = <1>;                |
> |     ranges = <0x0 0xe000 0x1000>; |     dma-ranges = <0x0 0xe000 0x1000>; |
> |     child {                       |     child {                           |
> |       ...                         |       ...                             |
> |     };                            |     };                                |
> |   };                              |   };                                  |
> | };                                | };                                    |
> +-----------------------------------+---------------------------------------+
> | no warning                        | Warning (avoid_unnecessary_addr_size) |
> +-----------------------------------+---------------------------------------+
> 
> Same as "ranges", it should not be reported in this check.
> 
> Signed-off-by: Qun-Wei Lin <qun-wei.lin-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org>

Applied, thanks.  An additional testcase to verify this behaviour
would be appreciated.

> ---
>  checks.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/checks.c b/checks.c
> index afa8f23..607e0f2 100644
> --- a/checks.c
> +++ b/checks.c
> @@ -1232,7 +1232,7 @@ static void check_avoid_unnecessary_addr_size(struct check *c, struct dt_info *d
>  	if (!node->parent || node->addr_cells < 0 || node->size_cells < 0)
>  		return;
>  
> -	if (get_property(node, "ranges") || !node->children)
> +	if (get_property(node, "ranges") || get_property(node, "dma-ranges") || !node->children)
>  		return;
>  
>  	for_each_child(node, child) {
> @@ -1242,7 +1242,7 @@ static void check_avoid_unnecessary_addr_size(struct check *c, struct dt_info *d
>  	}
>  
>  	if (!has_reg)
> -		FAIL(c, dti, node, "unnecessary #address-cells/#size-cells without \"ranges\" or child \"reg\" property");
> +		FAIL(c, dti, node, "unnecessary #address-cells/#size-cells without \"ranges\", \"dma-ranges\" or child \"reg\" property");
>  }
>  WARNING(avoid_unnecessary_addr_size, check_avoid_unnecessary_addr_size, NULL, &avoid_default_addr_size);
>  

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]