https://bugzilla.kernel.org/show_bug.cgi?id=215167 --- Comment #48 from Oleksandr Natalenko (oleksandr@xxxxxxxxxxxxxx) --- (In reply to Chaim Eliyah from comment #46) > | Reloading `btusb` module several times makes it working. > > Any specific steps for this? I have not been able to get it to work. Issue > persists despite reboots incl. hard power off/on. This happened suddenly. No specific steps, just `modprobe -r btusb && modprobe btusb`. I'm not facing the issue regularly, for me it happened only once recently after I powered off the system completely for a couple of days. The issue does not occur across reboots or suspends/resumes. It also doesn't occur after each power off (at least, so far). I've adopted the patch from c#38 like below just to be on the safe side: ``` commit 2875e3ab82d1777a06a356167c52a745827b2ad1 Author: Oleksandr Natalenko <oleksandr@xxxxxxxxxxxxxx> Date: Sun Feb 26 00:17:09 2023 +0100 Bluetooth: btusb: work around command 0xfc05 tx timeout Instead of realoding `btusb` module to tackle this: ``` Bluetooth: hci0: command 0xfc05 tx timeout Bluetooth: hci0: Reading Intel version command failed (-110) ``` increase `btusb_qca_cmd_timeout()` sleep duration while hoping for the best. This looks like an ugly hack. Link: https://bugzilla.kernel.org/show_bug.cgi?id=215167 Signed-off-by: Oleksandr Natalenko <oleksandr@xxxxxxxxxxxxxx> diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index 2ad4efdd9e40..7aff6c468c0d 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -901,7 +901,7 @@ static void btusb_qca_cmd_timeout(struct hci_dev *hdev) } gpiod_set_value_cansleep(reset_gpio, 0); - msleep(200); + usleep_range(USEC_PER_SEC / 2, USEC_PER_SEC); gpiod_set_value_cansleep(reset_gpio, 1); return; ``` Let me know if you are able to test it. -- You may reply to this email to add a comment. You are receiving this mail because: You are the assignee for the bug.
