linux-kernel - Re: [PATCH 3/3] Bluetooth: btrtl: Skip initialization if firmware is already loaded

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-Id: <5D6AFF40-BBE2-4B42-91D2-C89C52ED06A0@canonical.com>
Date:   Sat, 26 Jan 2019 01:46:01 +0800
From:   Kai-Heng Feng <kai.heng.feng@...onical.com>
To:     Daniel Drake <drake@...lessm.com>
Cc:     Marcel Holtmann <marcel@...tmann.org>,
        Johan Hedberg <johan.hedberg@...il.com>,
        Linux Bluetooth mailing list 
        <linux-bluetooth@...r.kernel.org>,
        Linux Kernel <linux-kernel@...r.kernel.org>,
        Martin Blumenstingl <martin.blumenstingl@...glemail.com>
Subject: Re: [PATCH 3/3] Bluetooth: btrtl: Skip initialization if firmware is
 already loaded



> On Jan 25, 2019, at 08:55, Daniel Drake <drake@...lessm.com> wrote:
> 
> On Thu, Jan 24, 2019 at 11:23 PM Kai-Heng Feng
> <kai.heng.feng@...onical.com> wrote:
>> 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
>> 
>> The power is not cut during system reboot, so the firmware is kept in
>> Bluetooth controller.
>> 
>> Realtek bluetooth doesn't have the ability to check firmware loading
>> status. but the version queried by HCI_OP_READ_LOCAL_VERSION will be
>> different if firmware is already loaded. Realtek's own fork, rtk_btusb
>> also use this method to detect the loading status.
>> 
>> So let's assume the firmware is already loaded when we can't find
>> matching IC info.
> 
> This logic was already present in the driver - but it looks like this
> regressed at this point:
> 
> commit 26503ad25de8c7c93a2037f919c2e49a62cf65f1
> Author: Martin Blumenstingl <martin.blumenstingl@...glemail.com>
> Date:   Thu Aug 2 16:57:13 2018 +0200
> 
>    Bluetooth: btrtl: split the device initialization into smaller parts

Thanks, didn’t find out it’s a regression.

> 
> After your patch it is effectively there in two places now, since it
> is also in btrtl_download_firmware() (although not really effective
> after the above commit). I wonder if that can be cleaned up to avoid
> duplication.

Put the additional check to btrtl_download_firmware() should be sufficient.

> 
> Regarding the other patches that move away from the style of returning
> either a useful value or an error, is this purely a stylistic thing or
> is it needed for your 3rd patch? I don't have strong feelings either
> way but I have the impression that the currently implemented approach
> is a common style within kernel code and I don't see benefit in
> splitting off a separate out parameter.

Ok. I’ll send a v2 without the refactoring part.

Kai-Heng

> 
> Daniel
> 
> 
>> Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=201921
>> Signed-off-by: Kai-Heng Feng <kai.heng.feng@...onical.com>
>> ---
>> drivers/bluetooth/btrtl.c | 10 ++++++----
>> 1 file changed, 6 insertions(+), 4 deletions(-)
>> 
>> diff --git a/drivers/bluetooth/btrtl.c b/drivers/bluetooth/btrtl.c
>> index c36f500d8313..e2f89d57dd14 100644
>> --- a/drivers/bluetooth/btrtl.c
>> +++ b/drivers/bluetooth/btrtl.c
>> @@ -546,9 +546,10 @@ int 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",
>> -                           lmp_subver, hci_rev, hci_ver);
>> -               return -EINVAL;
>> +               rtl_dev_info(hdev, "rtl: unknown IC info, lmp subver %04x, hci rev %04x, hci ver %04x",
>> +                            lmp_subver, hci_rev, hci_ver);
>> +               rtl_dev_info(hdev, "rtl: firmware may be already loaded, or it's an unsupported IC.");
>> +               return 0;
>>        }
>> 
>>        if (btrtl_dev->ic_info->has_rom_version) {
>> @@ -621,7 +622,8 @@ int btrtl_setup_realtek(struct hci_dev *hdev)
>>        if (ret)
>>                return ret;
>> 
>> -       ret = btrtl_download_firmware(hdev, &btrtl_dev);
>> +       if (btrtl_dev.ic_info)
>> +               ret = btrtl_download_firmware(hdev, &btrtl_dev);
>> 
>>        return ret;
>> }
>> --
>> 2.17.1
>>