[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20241023112828.320157-2-hadess@hadess.net>
Date: Wed, 23 Oct 2024 13:24:37 +0200
From: Bastien Nocera <hadess@...ess.net>
To: linux-input@...r.kernel.org
Cc: linux-kernel@...r.kernel.org,
Jiri Kosina <jikos@...nel.org>,
Benjamin Tissoires <benjamin.tissoires@...hat.com>,
Bastien Nocera <hadess@...ess.net>
Subject: [PATCH 1/2] HID: steelseries: Fix battery requests stopping after some time
In some cases, the headset receiver will answer one of our requests with
garbage, or not at all. This is a problem when we only request battery
information once we've received a battery response, as we might never
get to request battery information again.
If the data from the receiver could not be parsed, and there's no
pending battery requests, schedule a new request.
Signed-off-by: Bastien Nocera <hadess@...ess.net>
---
drivers/hid/hid-steelseries.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/hid/hid-steelseries.c b/drivers/hid/hid-steelseries.c
index 7e83fee1ffa0..16138f7dae17 100644
--- a/drivers/hid/hid-steelseries.c
+++ b/drivers/hid/hid-steelseries.c
@@ -603,8 +603,11 @@ static int steelseries_headset_raw_event(struct hid_device *hdev,
hid_dbg(sd->hdev,
"Parsing raw event for Arctis 1 headset (%*ph)\n", size, read_buf);
if (size < ARCTIS_1_BATTERY_RESPONSE_LEN ||
- memcmp (read_buf, arctis_1_battery_request, sizeof(arctis_1_battery_request)))
+ memcmp(read_buf, arctis_1_battery_request, sizeof(arctis_1_battery_request))) {
+ if (!delayed_work_pending(&sd->battery_work))
+ goto request_battery;
return 0;
+ }
if (read_buf[2] == 0x01) {
connected = false;
capacity = 100;
@@ -631,6 +634,7 @@ static int steelseries_headset_raw_event(struct hid_device *hdev,
power_supply_changed(sd->battery);
}
+request_battery:
spin_lock_irqsave(&sd->lock, flags);
if (!sd->removed)
schedule_delayed_work(&sd->battery_work,
--
2.47.0
Powered by blists - more mailing lists