detect kbd backlight on Google Link (original Chromebook Pixel)
Subsystems (2)
Hi,
Hopefully this is a trivial and obviously correct patch that will be easy to accept. Tested on my very own Chromebook Pixel.
Please advise if this is not the correct mailing list or procedure for submitting this patch, and/or if there is anything else that I need to do.
Thanks,
-Brandon
PATCH 1/1 platform/chrome: cros_kbd_led_backlight: support Google Link
On the Google Pixel chromebook (aka Google Link), the keyboard led backlight has the ACPI device id "GGL0002". Let's add it to the configuration array.
--- drivers/platform/chrome/cros_kbd_led_backlight.c | 1 + 1 file changed, 1 insertion(+)
drivers/platform/chrome/cros_kbd_led_backlight.c +1 -0
@@ -268,6 +268,7 @@ static int keyboard_led_probe(struct platform_device *pdev)#ifdef CONFIG_ACPI static const struct acpi_device_id keyboard_led_acpi_match[] = { {"GOOG0002" , (kernel_ulong_t)&keyboard_led_drvdata_acpi }, + {"GGL0002" , (kernel_ulong_t)&keyboard_led_drvdata_acpi }, { } };MODULE_DEVICE_TABLE (acpi, keyboard_led_acpi_match);
Signature · 1 line
2.43.0
Tzung-Bi Shih
maintainerRe: [PATCH] platform/chrome: cros_kbd_led_backlight: support Google Link
Quoted 3 lines · Brandon Casey
On the Google Pixel chromebook (aka Google Link), the keyboard led backlight has the ACPI device id "GGL0002". Let's add it to the configuration array.
Why it can't be matched by "GOOG0002"? [1] Have you tried to update your AP firmware?
Re: [PATCH] platform/chrome: cros_kbd_led_backlight: support Google Link
Quoted 6 lines · Tzung-Bi Shih
Quoted 3 lines · Brandon Casey
On the Google Pixel chromebook (aka Google Link), the keyboard led backlight has the ACPI device id "GGL0002". Let's add it to the configuration array.
Why it can't be matched by "GOOG0002"? [1]
I'm not sure I understand this question. On my device the backlight has ACPI id "GGL0002", not "GOOG0002"?
Quoted 1 line
Have you tried to update your AP firmware?
I have not explicitly updated any firmware. It has whatever firmware Chromeos has installed.
-Brandon
Tzung-Bi Shih
maintainerRe: [PATCH] platform/chrome: cros_kbd_led_backlight: support Google Link
Quoted 8 lines · Brandon Casey
Quoted 5 lines · Tzung-Bi Shih
Quoted 3 lines · Brandon Casey
On the Google Pixel chromebook (aka Google Link), the keyboard led backlight has the ACPI device id "GGL0002". Let's add it to the configuration array.
Quoted 7 lines
Quoted 1 line
Why it can't be matched by "GOOG0002"? [1]
I'm not sure I understand this question. On my device the backlight has ACPI id "GGL0002", not "GOOG0002"?
Quoted 1 line
Have you tried to update your AP firmware?
I have not explicitly updated any firmware. It has whatever firmware Chromeos has installed.
IIUC, if your AP firmware includes patch [1] (submitted in 2015), it should report "GOOG0002" instead of "GGL0002".
Re: [PATCH] platform/chrome: cros_kbd_led_backlight: support Google Link
Quoted 13 lines · Tzung-Bi Shih
Quoted 12 lines · Brandon Casey
Quoted 6 lines · Tzung-Bi Shih
Quoted 3 lines · Brandon Casey
On the Google Pixel chromebook (aka Google Link), the keyboard led backlight has the ACPI device id "GGL0002". Let's add it to the configuration array.
Why it can't be matched by "GOOG0002"? [1]
I'm not sure I understand this question. On my device the backlight has ACPI id "GGL0002", not "GOOG0002"?
Quoted 8 lines
Quoted 3 lines
Quoted 1 line
Have you tried to update your AP firmware?
I have not explicitly updated any firmware. It has whatever firmware Chromeos has installed.
IIUC, if your AP firmware includes patch [1] (submitted in 2015), it should report "GOOG0002" instead of "GGL0002".
I can only assume that Google did not ever ship a firmware update for Link that included that patch. I believe this device has all ChromeOS updates applied, so it represents a stock system, though no longer supported, and aside from being in developer mode and having a Linux partition.
My firmware also reports the acpi device as GGL0001.
$ cat /sys/bus/acpi/devices/GGL000*/hid
GGL0001
GGL0002 $ cat /sys/bus/acpi/devices/GGL000*/path
\CRHW
\_SB_.KBLTIs there any reason why we shouldn't do the same thing here for the keyboard backlight that was done in drivers/platform/chrome/chromeos_acpi.c and support both the legacy ID and the official ACPI ID? I see your commit 703e77134 which added GOOG0016 in addition to GGL0001.
-Brandon
Tzung-Bi Shih
maintainerRe: [PATCH] platform/chrome: cros_kbd_led_backlight: support Google Link
Quoted 18 lines · Brandon Casey
Quoted 15 lines · Tzung-Bi Shih
Quoted 14 lines · Brandon Casey
Quoted 6 lines · Tzung-Bi Shih
Quoted 3 lines · Brandon Casey
On the Google Pixel chromebook (aka Google Link), the keyboard led backlight has the ACPI device id "GGL0002". Let's add it to the configuration array.
Why it can't be matched by "GOOG0002"? [1]
I'm not sure I understand this question. On my device the backlight has ACPI id "GGL0002", not "GOOG0002"?
Quoted 1 line
Have you tried to update your AP firmware?
Quoted 8 lines
Quoted 6 lines
Quoted 1 line
I have not explicitly updated any firmware. It has whatever firmware Chromeos has installed.
IIUC, if your AP firmware includes patch [1] (submitted in 2015), it should report "GOOG0002" instead of "GGL0002".
I can only assume that Google did not ever ship a firmware update for Link that included that patch. I believe this device has all ChromeOS updates applied, so it represents a stock system, though no longer supported, and aside from being in developer mode and having a Linux partition.
Please try to wrap your message to fit to 80-column. See [2].
[2]: https://people.kernel.org/tglx/notes-about-netiquette
Quoted 11 lines
My firmware also reports the acpi device as GGL0001.
$ cat /sys/bus/acpi/devices/GGL000*/hid GGL0001 GGL0002$ cat /sys/bus/acpi/devices/GGL000*/path \CRHW \_SB_.KBLTIs there any reason why we shouldn't do the same thing here for the keyboard backlight that was done in
drivers/platform/chrome/chromeos_acpi.cand support both the legacy ID and the official ACPI ID? I see your commit703e77134which added GOOG0016 in addition to GGL0001.
- The most important reason: "GGL0002" is assigned to other purpose.
- For some reason, your machine is using a quite old firmware. It makes less sense to me for supporting a should-be-deprecated PNP ID by [1].
- The case is different from 703e77134. We prefer HID to PNP ID.
Re: [PATCH] platform/chrome: cros_kbd_led_backlight: support Google Link
Quoted 9 lines · Tzung-Bi Shih
Quoted 5 lines · Brandon Casey
I can only assume that Google did not ever ship a firmware update for Link that included that patch. I believe this device has all ChromeOS updates applied, so it represents a stock system, though no longer supported, and aside from being in developer mode and having a Linux partition.
Please try to wrap your message to fit to 80-column. See [2].
Sorry about that. I didn't realize my email client was doing that.
Quoted 6 lines
Quoted 4 lines
Is there any reason why we shouldn't do the same thing here for the keyboard backlight that was done in
drivers/platform/chrome/chromeos_acpi.cand support both the legacy ID and the official ACPI ID? I see your commit703e77134which added GOOG0016 in addition to GGL0001.- The most important reason: "GGL0002" is assigned to other purpose.
Ah, so this seems to be the crux of it :-(
Quoted 2 lines
- For some reason, your machine is using a quite old firmware. It makes less sense to me for supporting a should-be-deprecated PNP ID by [1].
You seem to be suggesting that my machine is in some odd/abnormal state. Do you know whether Google has actually ever pushed an updated firmware for this device (LINK) that included the patch referenced in [1]? If not, then shouldn't my machine be considered to be in the "normal" state, not the oddity?
-Brandon
Tzung-Bi Shih
maintainerRe: [PATCH] platform/chrome: cros_kbd_led_backlight: support Google Link
Quoted 5 lines · Brandon Casey
You seem to be suggesting that my machine is in some odd/abnormal state. Do you know whether Google has actually ever pushed an updated firmware for this device (LINK) that included the patch referenced in [1]? If not, then shouldn't my machine be considered to be in the "normal" state, not the oddity?
No, I don't know.