[PATCH mmc-next v3 1/2] dt-bindings: mmc: add alias example

[PATCH mmc-next v3 1/2] dt-bindings: mmc: add alias example

[Matthias Schiffer]

As for I2C and SPI, it now is possible to reserve a fixed index for
mmc/mmcblk devices.

Signed-off-by: Matthias Schiffer <matthias.schiffer@ew.tq-group.com>
---

v3: new patch

 Documentation/devicetree/bindings/mmc/mmc-controller.yaml | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/Documentation/devicetree/bindings/mmc/mmc-controller.yaml b/Documentation/devicetree/bindings/mmc/mmc-controller.yaml
index b96da0c7f819..22ed4a36c65d 100644
--- a/Documentation/devicetree/bindings/mmc/mmc-controller.yaml
+++ b/Documentation/devicetree/bindings/mmc/mmc-controller.yaml
@@ -367,6 +367,14 @@ examples:
     };
 
   - |
+    /*
+     * Optionally define an alias to reserve a fixed index for the
+     * mmc and mmcblk devices
+     */
+    aliases {
+        mmc0 = &mmc3;
+    };
+
     mmc3: mmc@1c12000 {
         #address-cells = <1>;
         #size-cells = <0>;
-- 
2.17.1

[PATCH mmc-next v3 2/2] mmc: allow setting slot index via device tree alias

[Matthias Schiffer]

As with GPIO, UART and others, allow specifying the device index via the
aliases node in the device tree.

On embedded devices, there is often a combination of removable (e.g.
SD card) and non-removable MMC devices (e.g. eMMC).
Therefore the index might change depending on

* host of removable device
* removable card present or not

This makes it difficult to hardcode the root device, if it is on the
non-removable device. E.g. if SD card is present eMMC will be mmcblk1,
if SD card is not present at boot, eMMC will be mmcblk0.

Alternative solutions like PARTUUIDs do not cover the case where multiple
mmcblk devices contain the same image. This is a common issue on devices
that can boot both from eMMC (for regular boot) and SD cards (as a
temporary boot medium for development). When a firmware image is
installed to eMMC after a test boot via SD card, there will be no
reliable way to refer to a specific device using (PART)UUIDs oder
LABELs.

The demand for this feature has led to multiple attempts to implement
it, dating back at least to 2012 (see
https://www.spinics.net/lists/linux-mmc/msg26586.html for a previous
discussion from 2014).

All indices defined in the aliases node will be reserved for use by the
respective MMC device, moving the indices of devices that don't have an
alias up into the non-reserved range. If the aliases node is not found,
the driver will act as before.

This is a rebased and slightly cleaned up version of
https://www.spinics.net/lists/linux-mmc/msg26588.html .

Based-on-patch-by: Sascha Hauer <s.hauer@pengutronix.de>
Link: https://lkml.org/lkml/2020/8/5/194
Signed-off-by: Matthias Schiffer <matthias.schiffer@ew.tq-group.com>
---

v3:
- remove unneeded mmcblock changes
- remove most helper functions to simplify code
- extended commit message

v2:
- fix missing symbols for modular mmcblock


 drivers/mmc/core/host.c | 26 +++++++++++++++++++++++++-
 1 file changed, 25 insertions(+), 1 deletion(-)

diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c
index c8fae6611b73..96b2ca1f1b06 100644
--- a/drivers/mmc/core/host.c
+++ b/drivers/mmc/core/host.c
@@ -376,6 +376,20 @@ int mmc_of_parse_voltage(struct device_node *np, u32 *mask)
 }
 EXPORT_SYMBOL(mmc_of_parse_voltage);
 
+/**
+ * mmc_first_nonreserved_index() - get the first index that is not reserved
+ */
+static int mmc_first_nonreserved_index(void)
+{
+	int max;
+
+	max = of_alias_get_highest_id("mmc");
+	if (max < 0)
+		return 0;
+
+	return max + 1;
+}
+
 /**
  *	mmc_alloc_host - initialise the per-host structure.
  *	@extra: sizeof private data structure
@@ -387,6 +401,7 @@ struct mmc_host *mmc_alloc_host(int extra, struct device *dev)
 {
 	int err;
 	struct mmc_host *host;
+	int alias_id, min_idx, max_idx;
 
 	host = kzalloc(sizeof(struct mmc_host) + extra, GFP_KERNEL);
 	if (!host)
@@ -395,7 +410,16 @@ struct mmc_host *mmc_alloc_host(int extra, struct device *dev)
 	/* scanning will be enabled when we're ready */
 	host->rescan_disable = 1;
 
-	err = ida_simple_get(&mmc_host_ida, 0, 0, GFP_KERNEL);
+	alias_id = of_alias_get_id(dev->of_node, "mmc");
+	if (alias_id >= 0) {
+		min_idx = alias_id;
+		max_idx = alias_id + 1;
+	} else {
+		min_idx = mmc_first_nonreserved_index();
+		max_idx = 0;
+	}
+
+	err = ida_simple_get(&mmc_host_ida, min_idx, max_idx, GFP_KERNEL);
 	if (err < 0) {
 		kfree(host);
 		return NULL;
-- 
2.17.1

Re: [PATCH mmc-next v3 1/2] dt-bindings: mmc: add alias example

[Rob Herring]

On Tue, Aug 25, 2020 at 03:44:40PM +0200, Matthias Schiffer wrote:
> As for I2C and SPI, it now is possible to reserve a fixed index for
> mmc/mmcblk devices.
> 
> Signed-off-by: Matthias Schiffer <matthias.schiffer@ew.tq-group.com>
> ---
> 
> v3: new patch
> 
>  Documentation/devicetree/bindings/mmc/mmc-controller.yaml | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/mmc/mmc-controller.yaml b/Documentation/devicetree/bindings/mmc/mmc-controller.yaml
> index b96da0c7f819..22ed4a36c65d 100644
> --- a/Documentation/devicetree/bindings/mmc/mmc-controller.yaml
> +++ b/Documentation/devicetree/bindings/mmc/mmc-controller.yaml
> @@ -367,6 +367,14 @@ examples:
>      };
>  
>    - |
> +    /*
> +     * Optionally define an alias to reserve a fixed index for the
> +     * mmc and mmcblk devices
> +     */
> +    aliases {
> +        mmc0 = &mmc3;
> +    };

This will break if we improve schemas because this node is actually 
/example-1/aliases.

So please drop. If you want, I'd really like to have a defined set (i.e. 
a schema) of alias names. This would require deleting a bunch on some 
platforms that just made up a bunch of them.

> +
>      mmc3: mmc@1c12000 {
>          #address-cells = <1>;
>          #size-cells = <0>;
> -- 
> 2.17.1
> 

Re: [PATCH mmc-next v3 1/2] dt-bindings: mmc: add alias example

[Matthias Schiffer]

On Fri, 2020-08-28 at 16:24 -0600, Rob Herring wrote:
> On Tue, Aug 25, 2020 at 03:44:40PM +0200, Matthias Schiffer wrote:
> > As for I2C and SPI, it now is possible to reserve a fixed index for
> > mmc/mmcblk devices.
> > 
> > Signed-off-by: Matthias Schiffer <matthias.schiffer@ew.tq-group.com
> > >
> > ---
> > 
> > v3: new patch
> > 
> >  Documentation/devicetree/bindings/mmc/mmc-controller.yaml | 8
> > ++++++++
> >  1 file changed, 8 insertions(+)
> > 
> > diff --git a/Documentation/devicetree/bindings/mmc/mmc-
> > controller.yaml b/Documentation/devicetree/bindings/mmc/mmc-
> > controller.yaml
> > index b96da0c7f819..22ed4a36c65d 100644
> > --- a/Documentation/devicetree/bindings/mmc/mmc-controller.yaml
> > +++ b/Documentation/devicetree/bindings/mmc/mmc-controller.yaml
> > @@ -367,6 +367,14 @@ examples:
> >      };
> >  
> >    - |
> > +    /*
> > +     * Optionally define an alias to reserve a fixed index for the
> > +     * mmc and mmcblk devices
> > +     */
> > +    aliases {
> > +        mmc0 = &mmc3;
> > +    };
> 
> This will break if we improve schemas because this node is actually 
> /example-1/aliases.
> 
> So please drop. If you want, I'd really like to have a defined set
> (i.e. 
> a schema) of alias names. This would require deleting a bunch on
> some 
> platforms that just made up a bunch of them.

Ulf suggested that I add some kind of documentation about the new mmc
alias support to the binding docs.

As long as we don't have a proper schema for aliases, should I just add
an explanation to the toplevel description of
Documentation/devicetree/bindings/mmc/mmc-controller.yaml, or maybe a
comment?


> 
> > +
> >      mmc3: mmc@1c12000 {
> >          #address-cells = <1>;
> >          #size-cells = <0>;
> > -- 
> > 2.17.1
> > 

Re: [PATCH mmc-next v3 1/2] dt-bindings: mmc: add alias example

[Ulf Hansson]

On Mon, 31 Aug 2020 at 09:58, Matthias Schiffer
<matthias.schiffer@ew.tq-group.com> wrote:
>
> On Fri, 2020-08-28 at 16:24 -0600, Rob Herring wrote:
> > On Tue, Aug 25, 2020 at 03:44:40PM +0200, Matthias Schiffer wrote:
> > > As for I2C and SPI, it now is possible to reserve a fixed index for
> > > mmc/mmcblk devices.
> > >
> > > Signed-off-by: Matthias Schiffer <matthias.schiffer@ew.tq-group.com
> > > >
> > > ---
> > >
> > > v3: new patch
> > >
> > >  Documentation/devicetree/bindings/mmc/mmc-controller.yaml | 8
> > > ++++++++
> > >  1 file changed, 8 insertions(+)
> > >
> > > diff --git a/Documentation/devicetree/bindings/mmc/mmc-
> > > controller.yaml b/Documentation/devicetree/bindings/mmc/mmc-
> > > controller.yaml
> > > index b96da0c7f819..22ed4a36c65d 100644
> > > --- a/Documentation/devicetree/bindings/mmc/mmc-controller.yaml
> > > +++ b/Documentation/devicetree/bindings/mmc/mmc-controller.yaml
> > > @@ -367,6 +367,14 @@ examples:
> > >      };
> > >
> > >    - |
> > > +    /*
> > > +     * Optionally define an alias to reserve a fixed index for the
> > > +     * mmc and mmcblk devices
> > > +     */
> > > +    aliases {
> > > +        mmc0 = &mmc3;
> > > +    };
> >
> > This will break if we improve schemas because this node is actually
> > /example-1/aliases.
> >
> > So please drop. If you want, I'd really like to have a defined set
> > (i.e.
> > a schema) of alias names. This would require deleting a bunch on
> > some
> > platforms that just made up a bunch of them.
>
> Ulf suggested that I add some kind of documentation about the new mmc
> alias support to the binding docs.
>
> As long as we don't have a proper schema for aliases, should I just add
> an explanation to the toplevel description of
> Documentation/devicetree/bindings/mmc/mmc-controller.yaml, or maybe a
> comment?

I would prefer to add an explanation in the toplevel description,
although I have no strong opinion.

This is similar to what we already do for serial devices, see
Documentation/devicetree/bindings/serial/serial.yaml.

>
>
> >
> > > +
> > >      mmc3: mmc@1c12000 {
> > >          #address-cells = <1>;
> > >          #size-cells = <0>;
> > > --
> > > 2.17.1
> > >
>

Kind regards
Uffe