Skip to contentSkip to conversation

usb: core: Add quirk for Gigabyte Aorus Waterforce X II AIO coolers

PATCH Superseded Lyude Paul 2 participants 4 messages Feb 3 – Feb 5 USB

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore

Lyude Paul

author
#0 lore raw

cover usb: core: Add quirk for Gigabyte Aorus Waterforce X II AIO coolers

This appears to be the final bug that I've discovered on my newly built desktop - the liquid cooler I installed doesn't appear to reset properly during resume, and causes the kernel to print some errors (see the commit message for more info). As far as I can tell, it seems to just be a simple case of not supporting resets.

It's worth mentioning what the USB topology for this device actually looks like:

/: Bus 007.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/12p, 480M

    ID 1d6b:0002 Linux Foundation 2.0 root hub
    |__ Port 010: Dev 006, If 0, Class=Hub, Driver=hub/4p, 480M
        ID 05e3:0608 Genesys Logic, Inc. Hub
        |__ Port 004: Dev 015, If 0, Class=Human Interface Device, Driver=usbhid, 480M
            ID 0414:7a5e Giga-Byte Technology Co., Ltd

To be clear, I believe the second hub (Genesys Logic) is built into the liquid cooler itself. Since there seem to be a number of known issues with Genesys Logic hubs (at least according to our quirks file) I did give a try at applying various quirks to it, but none of them seemed to make any difference.

Of course, if anyone thinks there's more things I should try here feel free to let me know :).

Lyude Paul (1):
usb: core: Add quirk for Gigabyte Aorus Waterforce X II AIO coolers

 drivers/usb/core/quirks.c | 3 +++
 1 file changed, 3 insertions(+)

base-commit: de0674d9bc69699c497477d45172493393ae9007

Signature · 1 line
2.52.0
author
#1 lore raw

PATCH 1/1 usb: core: Add quirk for Gigabyte Aorus Waterforce X II AIO coolers

The "Aorus Waterforce" is one of those fancy gaming water coolers that comes with fun LEDs and an entire dedicated LCD screen. A minor issue I've noticed with this cooler on Linux, is that we fail to actually reset the device when coming out of standby, ish:

  [  281.652184] usb 7-10.4: reset high-speed USB device number 7 using xhci_hcd
  [  287.179843] usb 7-10.4: PM: dpm_run_callback(): usb_dev_resume returns -5
  [  287.179854] usb 7-10.4: PM: failed to resume async: error -5

Generally this isn't fatal, as even when the water cooler doesn't resume it will still reconnect itself moments later:

15:48:12 GoldenWind kernel: usb 7-10.4: USB disconnect, device number 14 15:48:12 GoldenWind kernel: usb 7-10.4: new high-speed USB device number
15 using xhci_hcd
15:48:14 GoldenWind kernel: usb 7-10.4: New USB device found,

     idVendor=0414, idProduct=7a5e, bcdDevice= 2.00

15:48:14 GoldenWind kernel: usb 7-10.4: New USB device strings: Mfr=1,
Product=2, SerialNumber=0
15:48:14 GoldenWind kernel: usb 7-10.4: Product: Castor3 15:48:14 GoldenWind kernel: usb 7-10.4: Manufacturer: ITE Tech. 15:48:14 GoldenWind kernel: hid-generic 0003:0414:7A5E.000D:
hiddev98,hidraw2: USB HID v1.01 Device [ITE Tech. Castor3] on usb-0000:74:00.0-10.4/input0

My guess here is that the device simply doesn't support resets properly - since that seems to be the only quirk I've found that actually works. Note that the device still ends up disconnecting and reconnecting after resume, but this seems to be normal behavior.

So, let's fix this by adding the USB_QUIRK_RESET quirk for this USB device. This seems to work perfectly fine and prevents us from getting the aforementioned errors on resume.

---
 drivers/usb/core/quirks.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/usb/core/quirks.c +3 -0
index c4d85089d19b1..9a81994ee725b 100644
@@ -584,6 +584,9 @@ static const struct usb_device_id usb_quirk_list[] = {
 	/* INTEL VALUE SSD */
 	{ USB_DEVICE(0x8086, 0xf1a5), .driver_info = USB_QUIRK_RESET_RESUME },
 
+	/* Gigabyte ITE Tech. Castor3 (Aorus Waterforce X II AIO coolers) */
+	{ USB_DEVICE(0x0414, 0x7a5e), .driver_info = USB_QUIRK_RESET },
+
 	{ }  /* terminating entry must be last */
 };
 
Signature · 1 line
2.52.0
#2 lore raw
Quoted 44 lines · Lyude Paul

The "Aorus Waterforce" is one of those fancy gaming water coolers that comes with fun LEDs and an entire dedicated LCD screen. A minor issue I've noticed with this cooler on Linux, is that we fail to actually reset the device when coming out of standby, ish:

  [  281.652184] usb 7-10.4: reset high-speed USB device number 7 using xhci_hcd
  [  287.179843] usb 7-10.4: PM: dpm_run_callback(): usb_dev_resume returns -5
  [  287.179854] usb 7-10.4: PM: failed to resume async: error -5

Generally this isn't fatal, as even when the water cooler doesn't resume it will still reconnect itself moments later:

15:48:12 GoldenWind kernel: usb 7-10.4: USB disconnect, device number 14 15:48:12 GoldenWind kernel: usb 7-10.4: new high-speed USB device number
15 using xhci_hcd
15:48:14 GoldenWind kernel: usb 7-10.4: New USB device found,

     idVendor=0414, idProduct=7a5e, bcdDevice= 2.00

15:48:14 GoldenWind kernel: usb 7-10.4: New USB device strings: Mfr=1,
Product=2, SerialNumber=0
15:48:14 GoldenWind kernel: usb 7-10.4: Product: Castor3 15:48:14 GoldenWind kernel: usb 7-10.4: Manufacturer: ITE Tech. 15:48:14 GoldenWind kernel: hid-generic 0003:0414:7A5E.000D:
hiddev98,hidraw2: USB HID v1.01 Device [ITE Tech. Castor3] on usb-0000:74:00.0-10.4/input0

My guess here is that the device simply doesn't support resets properly - since that seems to be the only quirk I've found that actually works. Note that the device still ends up disconnecting and reconnecting after resume, but this seems to be normal behavior.

So, let's fix this by adding the USB_QUIRK_RESET quirk for this USB device. This seems to work perfectly fine and prevents us from getting the aforementioned errors on resume.

---
 drivers/usb/core/quirks.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/usb/core/quirks.c +0 -0
index c4d85089d19b1..9a81994ee725b 100644

drivers/usb/core/quirks.c static const struct usb_device_id usb_quirk_list[] = { · line 588

Quoted 5 lines
drivers/usb/core/quirks.c +2 -0
@@ -584,6 +584,9 @@ static const struct usb_device_id usb_quirk_list[] = {
 	/* INTEL VALUE SSD */
 	{ USB_DEVICE(0x8086, 0xf1a5), .driver_info = USB_QUIRK_RESET_RESUME },
 
+	/* Gigabyte ITE Tech. Castor3 (Aorus Waterforce X II AIO coolers) */
+	{ USB_DEVICE(0x0414, 0x7a5e), .driver_info = USB_QUIRK_RESET },

Please see the comments for this structure on how to place this device in the list.

thanks,

greg k-h

#3 lore raw
Quoted 10 lines · Greg Kroah-Hartman
Quoted 6 lines

USB_QUIRK_RESET_RESUME },

(quoted diff) +1 -0
+	/* Gigabyte ITE Tech. Castor3 (Aorus Waterforce X II AIO

coolers) */

(quoted diff) +1 -0
+	{ USB_DEVICE(0x0414, 0x7a5e), .driver_info =

USB_QUIRK_RESET },

Please see the comments for this structure on how to place this device
in the list.

aha-whoops, my bad :). I'll respin a new version in a bit

Quoted 4 lines

thanks,

greg k-h