Date:   Tue, 9 Apr 2019 09:36:35 -0700
From:   Matthias Kaehlcke <mka@chromium.org>
To:     Balakrishna Godavarthi <bgodavar@codeaurora.org>
Cc:     c-hbandi@codeaurora.org, Marcel Holtmann <marcel@holtmann.org>,
        Johan Hedberg <johan.hedberg@gmail.com>,
        linux-bluetooth@vger.kernel.org, linux-kernel@vger.kernel.org,
        Hemantg <hemantg@codeaurora.org>
Subject: Re: [PATCH 2/2] Bluetooth: hci_qca: wcn3990: Drop baudrate change
 vendor event
Message-ID: <20190409163635.GU112750@google.com>
References: <20190307182009.GB138592@google.com>
 <20190307233039.GA69116@google.com>
 <c985a6712f91930d7c710ca4abb42e80@codeaurora.org>
 <a03acbdc6bfad172fdd65f428ffbe4cf@codeaurora.org>
 <20190401171202.GH112750@google.com>
 <1bebaf0766988a9f27cbbcb9bb35e26f@codeaurora.org>
 <20190402180501.GK112750@google.com>
 <1200f5c32e029c603667d66a4c62a634@codeaurora.org>
 <20190403161438.GN112750@google.com>
 <a3bde5419145e9dccc2886d2d29c9d3f@codeaurora.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <a3bde5419145e9dccc2886d2d29c9d3f@codeaurora.org>
User-Agent: Mutt/1.10.1 (2018-07-13)
Sender: linux-bluetooth-owner@vger.kernel.org
Precedence: bulk

On Thu, Apr 04, 2019 at 08:22:15PM +0530, Balakrishna Godavarthi wrote:
> Hi Matthias,
> 
> On 2019-04-03 21:44, Matthias Kaehlcke wrote:
> > On Wed, Apr 03, 2019 at 11:53:26AM +0530, Balakrishna Godavarthi wrote:
> > > + Harish to update his findings on wcn3998.
> > > Mean time i will perform a regression to detect the failure.
> > 
> > On my system it typically reproduces within a few dozen
> > iterations. Make sure your tree doesn't contain any custom BT patches,
> > even if they just add logging or fix the timeout during
> > initialization. Since this problem is timing sensitive it might get
> > masked. My tree is based on 4.19 LTS with all QCA BT related changes
> > on top:
> > https://chromium.googlesource.com/chromiumos/third_party/kernel/+/refs/heads/chromeos-4.19
> > 
> > I found some problems during initialization easier to reproduce
> > when binding and unbinding the device through sysfs, instead of
> > doing hciconfig up/down, this resembles more the initialization at
> > boot time.
> > 
> [Bala]: I am able to replicate this issue.
>         are you seeing the below error message getting logged on the console
> in the issue case or an different error message.
>         Bluetooth: hci0: QCA TLV response size mismatch

Yes, that's the error message. The size doesn't match because the
driver expects a TLV response but gets a baud rate response.