Bluetooth: btrtl: Restore old logic to assume firmware is already loaded

commit 00df214b1faae520880cc5c57e206f21239ef741 upstream.

Realtek bluetooth may not work after reboot:
[   12.446130] Bluetooth: hci0: RTL: rtl: unknown IC info, lmp subver a99e, hci rev 826c, hci ver 0008

This is a regression introduced by commit 26503ad2 ("Bluetooth:
btrtl: split the device initialization into smaller parts"). The new
logic errors out early when no matching IC info can be found, in this
case it means the firmware is already loaded.

So let's assume the firmware is already loaded when we can't find
matching IC info, like the old logic did.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=201921
Fixes: 26503ad2 ("Bluetooth: btrtl: split the device initialization into smaller parts")
Cc: stable@vger.kernel.org # 4.19+
Signed-off-by: NKai-Heng Feng <kai.heng.feng@canonical.com>
Signed-off-by: NMarcel Holtmann <marcel@holtmann.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/bluetooth/btrtl.c

@@ -544,10 +544,9 @@ struct btrtl_device_info *btrtl_initialize(struct hci_dev *hdev,
 					    hdev->bus);
 
 	if (!btrtl_dev->ic_info) {
-		rtl_dev_err(hdev, "rtl: unknown IC info, lmp subver %04x, hci rev %04x, hci ver %04x",
+		rtl_dev_info(hdev, "rtl: unknown IC info, lmp subver %04x, hci rev %04x, hci ver %04x",
 			    lmp_subver, hci_rev, hci_ver);
-		ret = -EINVAL;
-		goto err_free;
+		return btrtl_dev;
 	}
 
 	if (btrtl_dev->ic_info->has_rom_version) {
@@ -602,6 +601,11 @@ int btrtl_download_firmware(struct hci_dev *hdev,
 	 * standard btusb. Once that firmware is uploaded, the subver changes
 	 * to a different value.
 	 */
+	if (!btrtl_dev->ic_info) {
+		rtl_dev_info(hdev, "rtl: assuming no firmware upload needed\n");
+		return 0;
+	}
+
 	switch (btrtl_dev->ic_info->lmp_subver) {
 	case RTL_ROM_LMP_8723A:
 	case RTL_ROM_LMP_3499: