* [PATCH v3 0/2] Input: cros-ec-keyb: Better matrixless support
@ 2022-05-03 4:22 Stephen Boyd
2022-05-03 4:22 ` [PATCH v3 1/2] dt-bindings: google,cros-ec-keyb: Introduce switches only compatible Stephen Boyd
2022-05-03 4:22 ` [PATCH v3 2/2] Input: cros-ec-keyb - skip keyboard registration w/o cros-ec-keyb compatible Stephen Boyd
0 siblings, 2 replies; 5+ messages in thread
From: Stephen Boyd @ 2022-05-03 4:22 UTC (permalink / raw)
To: Dmitry Torokhov
Cc: linux-kernel, patches, chrome-platform, Krzysztof Kozlowski,
Rob Herring, devicetree, Benson Leung, Guenter Roeck,
Douglas Anderson, Hsin-Yi Wang, Joseph S. Barrera III
This is a followup to my previous patch[1] that skips keyboard registration
when the matrix properties aren't present. This adds a compatible string
for this scenario so we can ease existing DTBs over to the new design.
Changes from v2 (https://lore.kernel.org/r/[email protected]):
* Drop rows/cols check now that compatible schema handles it
* Make binding require rows,cols,keymap for cros-ec-keyb compatible
Changes from v1 (https://lore.kernel.org/r/[email protected]):
* Better enforcement of properties in DT binding
* Skip registration by means of adding compatible to device id list
Stephen Boyd (2):
dt-bindings: google,cros-ec-keyb: Introduce switches only compatible
Input: cros-ec-keyb - skip keyboard registration w/o cros-ec-keyb
compatible
.../bindings/input/google,cros-ec-keyb.yaml | 87 +++++++++++++++++--
drivers/input/keyboard/cros_ec_keyb.c | 20 +++--
2 files changed, 92 insertions(+), 15 deletions(-)
Cc: Krzysztof Kozlowski <[email protected]>
Cc: Rob Herring <[email protected]>
Cc: <[email protected]>
Cc: Benson Leung <[email protected]>
Cc: Guenter Roeck <[email protected]>
Cc: Douglas Anderson <[email protected]>
Cc: Hsin-Yi Wang <[email protected]>
Cc: "Joseph S. Barrera III" <[email protected]>
[1] https://lore.kernel.org/all/[email protected]/
base-commit: 4352e23a7ff2f8a4ff229dd1283ed2f2b708ec51
--
https://chromeos.dev
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH v3 1/2] dt-bindings: google,cros-ec-keyb: Introduce switches only compatible
2022-05-03 4:22 [PATCH v3 0/2] Input: cros-ec-keyb: Better matrixless support Stephen Boyd
@ 2022-05-03 4:22 ` Stephen Boyd
2022-05-03 15:09 ` Doug Anderson
2022-05-03 4:22 ` [PATCH v3 2/2] Input: cros-ec-keyb - skip keyboard registration w/o cros-ec-keyb compatible Stephen Boyd
1 sibling, 1 reply; 5+ messages in thread
From: Stephen Boyd @ 2022-05-03 4:22 UTC (permalink / raw)
To: Dmitry Torokhov
Cc: linux-kernel, patches, chrome-platform, Krzysztof Kozlowski,
Rob Herring, devicetree, Benson Leung, Guenter Roeck,
Douglas Anderson, Hsin-Yi Wang, Joseph S. Barrera III
If the ChromeOS board is a detachable, this cros-ec-keyb device won't
have a matrix keyboard but it may have some button switches, e.g. volume
buttons and power buttons. The driver still registers a keyboard though
and that leads to userspace confusion around where the keyboard is.
We tried to work around this in commit 4352e23a7ff2 ("Input:
cros-ec-keyb - only register keyboard if rows/columns exist") but that
led to another problem where removing the rows/columns properties breaks
the binding[1]. Technically before that commit the rows/columns
properties were required, otherwise the driver would fail to probe.
Removing the properties from devicetrees makes the driver fail to probe
unless the corresponding driver patch is present. Furthermore, this
makes requiring matrix keyboard properties for devices that really have
a keyboard impossible because the compatible drives the schema and now
the properties are optional.
Add a more specific compatible for this type of device that indicates to
the OS that there are only switches and no matrix keyboard present. If
only the switches compatible is present, then the matrix keyboard
properties are denied. Otherwise, as long as the "google,cros-ec-keyb"
copmatible is present, the matrix keyboard properties are required.
This more clearly describes what the driver was expecting, i.e. that the
kernel driver would fail to probe if the linux,{rows,cols,keymap}
properties are missing and the "google,cros-ec-keyb" compatible is
present. If the new "google,cros-ec-keyb-switches" compatible is present
then the matrix keyboard properties are likely to be ignored by any
modern driver, but they're still allowed per the binding if
"google,cros-ec-keyb" is present to conform to the original binding and
to keep the driver probing when used with newer DTBs.
This lets us gracefully migrate devices that only have switches over to
the new compatible string and properly enforce the "google,cros-ec-keyb"
binding at the same time. Eventually we'll be able to use only
"google,cros-ec-keyb-switches" or "google,cros-ec-keyb" when the driver
support has baked for a release or two, but we'll always support the
combined compatible list so that DTBs can be mixed and matched with
kernels, no dependency between the two binaries is required.
Link: https://lore.kernel.org/r/CAD=FV=Wey2P_=3Lp6M8GEaoyCn1XcYFhfJwfx43a5f_8H0obwg@mail.gmail.com [1]
Cc: Krzysztof Kozlowski <[email protected]>
Cc: Rob Herring <[email protected]>
Cc: <[email protected]>
Cc: Benson Leung <[email protected]>
Cc: Guenter Roeck <[email protected]>
Cc: Douglas Anderson <[email protected]>
Cc: Hsin-Yi Wang <[email protected]>
Cc: "Joseph S. Barrera III" <[email protected]>
Signed-off-by: Stephen Boyd <[email protected]>
---
.../bindings/input/google,cros-ec-keyb.yaml | 87 +++++++++++++++++--
1 file changed, 81 insertions(+), 6 deletions(-)
diff --git a/Documentation/devicetree/bindings/input/google,cros-ec-keyb.yaml b/Documentation/devicetree/bindings/input/google,cros-ec-keyb.yaml
index e8f137abb03c..6f4e0242ad5a 100644
--- a/Documentation/devicetree/bindings/input/google,cros-ec-keyb.yaml
+++ b/Documentation/devicetree/bindings/input/google,cros-ec-keyb.yaml
@@ -15,14 +15,19 @@ description: |
Google's ChromeOS EC Keyboard is a simple matrix keyboard
implemented on a separate EC (Embedded Controller) device. It provides
a message for reading key scans from the EC. These are then converted
- into keycodes for processing by the kernel.
-
-allOf:
- - $ref: "/schemas/input/matrix-keymap.yaml#"
+ into keycodes for processing by the kernel. This device also supports
+ switches/buttons like power and volume buttons.
properties:
compatible:
- const: google,cros-ec-keyb
+ oneOf:
+ - items:
+ - const: google,cros-ec-keyb-switches
+ - items:
+ - const: google,cros-ec-keyb-switches
+ - const: google,cros-ec-keyb
+ - items:
+ - const: google,cros-ec-keyb
google,needs-ghost-filter:
description:
@@ -41,15 +46,32 @@ properties:
where the lower 16 bits are reserved. This property is specified only
when the keyboard has a custom design for the top row keys.
+dependencies:
+ function-row-phsymap: [ 'linux,keymap' ]
+ google,needs-ghost-filter: [ 'linux,keymap' ]
+
required:
- compatible
+if:
+ properties:
+ compatible:
+ contains:
+ const: google,cros-ec-keyb
+then:
+ allOf:
+ - $ref: "/schemas/input/matrix-keymap.yaml#"
+ required:
+ - keypad,num-rows
+ - keypad,num-columns
+ - linux,keymap
+
unevaluatedProperties: false
examples:
- |
#include <dt-bindings/input/input.h>
- cros-ec-keyb {
+ keyboard-controller {
compatible = "google,cros-ec-keyb";
keypad,num-rows = <8>;
keypad,num-columns = <13>;
@@ -113,3 +135,56 @@ examples:
/* UP LEFT */
0x070b0067 0x070c0069>;
};
+
+ - |
+ keyboard-controller {
+ compatible = "google,cros-ec-keyb-switches", "google,cros-ec-keyb";
+ /* Matrix keymap properties are allowed but ignored */
+ keypad,num-rows = <8>;
+ keypad,num-columns = <13>;
+ linux,keymap = <
+ /* CAPSLCK F1 B F10 */
+ 0x0001003a 0x0002003b 0x00030030 0x00040044
+ /* N = R_ALT ESC */
+ 0x00060031 0x0008000d 0x000a0064 0x01010001
+ /* F4 G F7 H */
+ 0x0102003e 0x01030022 0x01040041 0x01060023
+ /* ' F9 BKSPACE L_CTRL */
+ 0x01080028 0x01090043 0x010b000e 0x0200001d
+ /* TAB F3 T F6 */
+ 0x0201000f 0x0202003d 0x02030014 0x02040040
+ /* ] Y 102ND [ */
+ 0x0205001b 0x02060015 0x02070056 0x0208001a
+ /* F8 GRAVE F2 5 */
+ 0x02090042 0x03010029 0x0302003c 0x03030006
+ /* F5 6 - \ */
+ 0x0304003f 0x03060007 0x0308000c 0x030b002b
+ /* R_CTRL A D F */
+ 0x04000061 0x0401001e 0x04020020 0x04030021
+ /* S K J ; */
+ 0x0404001f 0x04050025 0x04060024 0x04080027
+ /* L ENTER Z C */
+ 0x04090026 0x040b001c 0x0501002c 0x0502002e
+ /* V X , M */
+ 0x0503002f 0x0504002d 0x05050033 0x05060032
+ /* L_SHIFT / . SPACE */
+ 0x0507002a 0x05080035 0x05090034 0x050B0039
+ /* 1 3 4 2 */
+ 0x06010002 0x06020004 0x06030005 0x06040003
+ /* 8 7 0 9 */
+ 0x06050009 0x06060008 0x0608000b 0x0609000a
+ /* L_ALT DOWN RIGHT Q */
+ 0x060a0038 0x060b006c 0x060c006a 0x07010010
+ /* E R W I */
+ 0x07020012 0x07030013 0x07040011 0x07050017
+ /* U R_SHIFT P O */
+ 0x07060016 0x07070036 0x07080019 0x07090018
+ /* UP LEFT */
+ 0x070b0067 0x070c0069>;
+ };
+ - |
+ /* No matrix keyboard, just buttons/switches */
+ keyboard-controller {
+ compatible = "google,cros-ec-keyb-switches";
+ };
+...
--
https://chromeos.dev
^ permalink raw reply related [flat|nested] 5+ messages in thread
* [PATCH v3 2/2] Input: cros-ec-keyb - skip keyboard registration w/o cros-ec-keyb compatible
2022-05-03 4:22 [PATCH v3 0/2] Input: cros-ec-keyb: Better matrixless support Stephen Boyd
2022-05-03 4:22 ` [PATCH v3 1/2] dt-bindings: google,cros-ec-keyb: Introduce switches only compatible Stephen Boyd
@ 2022-05-03 4:22 ` Stephen Boyd
2022-05-03 15:09 ` Doug Anderson
1 sibling, 1 reply; 5+ messages in thread
From: Stephen Boyd @ 2022-05-03 4:22 UTC (permalink / raw)
To: Dmitry Torokhov
Cc: linux-kernel, patches, chrome-platform, Krzysztof Kozlowski,
Rob Herring, devicetree, Benson Leung, Guenter Roeck,
Douglas Anderson, Hsin-Yi Wang, Joseph S. Barrera III
In commit 4352e23a7ff2 ("Input: cros-ec-keyb - only register keyboard if
rows/columns exist") we skipped registration of the keyboard if the
row/columns property didn't exist, but that has a slight problem for
existing DTBs. The DTBs have the rows/columns properties, so removing
the properties to indicate only switches exist makes this keyboard
driver fail to probe, resulting in broken power and volume buttons. Ease
the migration of existing DTBs by skipping keyboard registration if the
google,cros-ec-keyb-switches compatible exists.
The end result is that new DTBs can either choose to remove the matrix
keymap properties or leave them in place and add this new compatible
indicating the matrix keyboard properties should be ignored. Existing
DTBs will continue to work, but they will keep registering the keyboard
that does nothing. To fix that problem we can add this extra compatible
to existing devicetrees and the keyboard will stop being registered.
Finally, if google,cros-ec-keyb is missing then this driver won't even
attempt to register the matrix keyboard. Of course, this driver won't
probe until this patch is applied in that scenario, but that's OK. This
last case is likely only going to be used by new devicetrees created
after this commit.
Cc: Krzysztof Kozlowski <[email protected]>
Cc: Rob Herring <[email protected]>
Cc: <[email protected]>
Cc: Benson Leung <[email protected]>
Cc: Guenter Roeck <[email protected]>
Cc: Douglas Anderson <[email protected]>
Cc: Hsin-Yi Wang <[email protected]>
Cc: "Joseph S. Barrera III" <[email protected]>
Fixes: 4352e23a7ff2 ("Input: cros-ec-keyb - only register keyboard if rows/columns exist")
Signed-off-by: Stephen Boyd <[email protected]>
---
drivers/input/keyboard/cros_ec_keyb.c | 20 +++++++++++---------
1 file changed, 11 insertions(+), 9 deletions(-)
diff --git a/drivers/input/keyboard/cros_ec_keyb.c b/drivers/input/keyboard/cros_ec_keyb.c
index eef909e52e23..04c550aaf897 100644
--- a/drivers/input/keyboard/cros_ec_keyb.c
+++ b/drivers/input/keyboard/cros_ec_keyb.c
@@ -536,14 +536,11 @@ static int cros_ec_keyb_register_matrix(struct cros_ec_keyb *ckdev)
u32 *physmap;
u32 key_pos;
unsigned int row, col, scancode, n_physmap;
+ bool register_keyboard;
- /*
- * No rows and columns? There isn't a matrix but maybe there are
- * switches to register in cros_ec_keyb_register_bs() because
- * this is a detachable device.
- */
- if (!device_property_present(dev, "keypad,num-rows") &&
- !device_property_present(dev, "keypad,num-cols"))
+ /* Skip matrix registration if no keyboard */
+ register_keyboard = device_get_match_data(dev);
+ if (!register_keyboard)
return 0;
err = matrix_keypad_parse_properties(dev, &ckdev->rows, &ckdev->cols);
@@ -718,8 +715,13 @@ static int cros_ec_keyb_remove(struct platform_device *pdev)
#ifdef CONFIG_OF
static const struct of_device_id cros_ec_keyb_of_match[] = {
- { .compatible = "google,cros-ec-keyb" },
- {},
+ {
+ /* Must be first */
+ .compatible = "google,cros-ec-keyb",
+ .data = (void *)true
+ },
+ { .compatible = "google,cros-ec-keyb-switches" },
+ {}
};
MODULE_DEVICE_TABLE(of, cros_ec_keyb_of_match);
#endif
--
https://chromeos.dev
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH v3 1/2] dt-bindings: google,cros-ec-keyb: Introduce switches only compatible
2022-05-03 4:22 ` [PATCH v3 1/2] dt-bindings: google,cros-ec-keyb: Introduce switches only compatible Stephen Boyd
@ 2022-05-03 15:09 ` Doug Anderson
0 siblings, 0 replies; 5+ messages in thread
From: Doug Anderson @ 2022-05-03 15:09 UTC (permalink / raw)
To: Stephen Boyd
Cc: Dmitry Torokhov, LKML, patches, chrome-platform,
Krzysztof Kozlowski, Rob Herring,
open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS,
Benson Leung, Guenter Roeck, Hsin-Yi Wang, Joseph S. Barrera III
Hi,
On Mon, May 2, 2022 at 9:22 PM Stephen Boyd <[email protected]> wrote:
>
> @@ -113,3 +135,56 @@ examples:
> /* UP LEFT */
> 0x070b0067 0x070c0069>;
> };
> +
> + - |
> + keyboard-controller {
> + compatible = "google,cros-ec-keyb-switches", "google,cros-ec-keyb";
> + /* Matrix keymap properties are allowed but ignored */
nit: s/allowed/required/
This now looks reasonable and well explained to me.
Reviewed-by: Douglas Anderson <[email protected]>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v3 2/2] Input: cros-ec-keyb - skip keyboard registration w/o cros-ec-keyb compatible
2022-05-03 4:22 ` [PATCH v3 2/2] Input: cros-ec-keyb - skip keyboard registration w/o cros-ec-keyb compatible Stephen Boyd
@ 2022-05-03 15:09 ` Doug Anderson
0 siblings, 0 replies; 5+ messages in thread
From: Doug Anderson @ 2022-05-03 15:09 UTC (permalink / raw)
To: Stephen Boyd
Cc: Dmitry Torokhov, LKML, patches, chrome-platform,
Krzysztof Kozlowski, Rob Herring,
open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS,
Benson Leung, Guenter Roeck, Hsin-Yi Wang, Joseph S. Barrera III
Hi,
On Mon, May 2, 2022 at 9:22 PM Stephen Boyd <[email protected]> wrote:
>
> In commit 4352e23a7ff2 ("Input: cros-ec-keyb - only register keyboard if
> rows/columns exist") we skipped registration of the keyboard if the
> row/columns property didn't exist, but that has a slight problem for
> existing DTBs. The DTBs have the rows/columns properties, so removing
> the properties to indicate only switches exist makes this keyboard
> driver fail to probe, resulting in broken power and volume buttons. Ease
> the migration of existing DTBs by skipping keyboard registration if the
> google,cros-ec-keyb-switches compatible exists.
>
> The end result is that new DTBs can either choose to remove the matrix
> keymap properties or leave them in place and add this new compatible
> indicating the matrix keyboard properties should be ignored. Existing
> DTBs will continue to work, but they will keep registering the keyboard
> that does nothing. To fix that problem we can add this extra compatible
> to existing devicetrees and the keyboard will stop being registered.
> Finally, if google,cros-ec-keyb is missing then this driver won't even
> attempt to register the matrix keyboard. Of course, this driver won't
> probe until this patch is applied in that scenario, but that's OK. This
> last case is likely only going to be used by new devicetrees created
> after this commit.
>
> Cc: Krzysztof Kozlowski <[email protected]>
> Cc: Rob Herring <[email protected]>
> Cc: <[email protected]>
> Cc: Benson Leung <[email protected]>
> Cc: Guenter Roeck <[email protected]>
> Cc: Douglas Anderson <[email protected]>
> Cc: Hsin-Yi Wang <[email protected]>
> Cc: "Joseph S. Barrera III" <[email protected]>
> Fixes: 4352e23a7ff2 ("Input: cros-ec-keyb - only register keyboard if rows/columns exist")
> Signed-off-by: Stephen Boyd <[email protected]>
> ---
> drivers/input/keyboard/cros_ec_keyb.c | 20 +++++++++++---------
> 1 file changed, 11 insertions(+), 9 deletions(-)
>
> diff --git a/drivers/input/keyboard/cros_ec_keyb.c b/drivers/input/keyboard/cros_ec_keyb.c
> index eef909e52e23..04c550aaf897 100644
> --- a/drivers/input/keyboard/cros_ec_keyb.c
> +++ b/drivers/input/keyboard/cros_ec_keyb.c
> @@ -536,14 +536,11 @@ static int cros_ec_keyb_register_matrix(struct cros_ec_keyb *ckdev)
> u32 *physmap;
> u32 key_pos;
> unsigned int row, col, scancode, n_physmap;
> + bool register_keyboard;
>
> - /*
> - * No rows and columns? There isn't a matrix but maybe there are
> - * switches to register in cros_ec_keyb_register_bs() because
> - * this is a detachable device.
> - */
> - if (!device_property_present(dev, "keypad,num-rows") &&
> - !device_property_present(dev, "keypad,num-cols"))
> + /* Skip matrix registration if no keyboard */
> + register_keyboard = device_get_match_data(dev);
> + if (!register_keyboard)
> return 0;
I'm a little on the fence about the local variable. It could have been
shorter as:
/* Skip matrix registration if no keyboard */
if (!device_get_match_data(dev))
...but I guess the "register_keyboard" maybe makes it more a little
more obvious?
In any case, I'm happy either way:
Reviewed-by: Douglas Anderson <[email protected]>
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2022-05-03 15:09 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2022-05-03 4:22 [PATCH v3 0/2] Input: cros-ec-keyb: Better matrixless support Stephen Boyd
2022-05-03 4:22 ` [PATCH v3 1/2] dt-bindings: google,cros-ec-keyb: Introduce switches only compatible Stephen Boyd
2022-05-03 15:09 ` Doug Anderson
2022-05-03 4:22 ` [PATCH v3 2/2] Input: cros-ec-keyb - skip keyboard registration w/o cros-ec-keyb compatible Stephen Boyd
2022-05-03 15:09 ` Doug Anderson
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox