Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH v1] i2c: tegra: Fix i2c_writesl() to use writel() instead
 of writesl()
From:   Sowjanya Komatineni <skomatineni@nvidia.com>
To:     Thierry Reding <thierry.reding@gmail.com>
CC:     <jonathanh@nvidia.com>, <digetx@gmail.com>,
        <linux-tegra@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
        <linux-i2c@vger.kernel.org>
References: <1603166634-13639-1-git-send-email-skomatineni@nvidia.com>
 <20201020074846.GA1877013@ulmo>
 <538d8436-260d-40a8-b0a3-a822a0f9c909@nvidia.com>
Message-ID: <d880a7b1-28e2-1bfc-2ba7-a7139455b890@nvidia.com>
Date:   Tue, 20 Oct 2020 09:39:10 -0700
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.10.0
MIME-Version: 1.0
In-Reply-To: <538d8436-260d-40a8-b0a3-a822a0f9c909@nvidia.com>
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: quoted-printable
Content-Language: en-US
Precedence: bulk


On 10/20/20 9:37 AM, Sowjanya Komatineni wrote:
>
> On 10/20/20 12:48 AM, Thierry Reding wrote:
>> On Mon, Oct 19, 2020 at 09:03:54PM -0700, Sowjanya Komatineni wrote:
>>> VI I2C don't have DMA support and uses PIO mode all the time.
>>>
>>> Current driver uses writesl() to fill TX FIFO based on available
>>> empty slots and with this seeing strange silent hang during any I2C
>>> register access after filling TX FIFO with 8 words.
>>>
>>> Using writel() followed by i2c_readl() in a loop to write all words
>>> to TX FIFO instead of using writesl() helps for large transfers in
>>> PIO mode.
>>>
>>> So, this patch updates i2c_writesl() API to use writel() in a loop
>>> instead of writesl().
>>>
>>> Signed-off-by: Sowjanya Komatineni <skomatineni@nvidia.com>
>>> ---
>>> =A0 drivers/i2c/busses/i2c-tegra.c | 9 ++++++---
>>> =A0 1 file changed, 6 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/drivers/i2c/busses/i2c-tegra.c=20
>>> b/drivers/i2c/busses/i2c-tegra.c
>>> index 6f08c0c..274bf3a 100644
>>> --- a/drivers/i2c/busses/i2c-tegra.c
>>> +++ b/drivers/i2c/busses/i2c-tegra.c
>>> @@ -333,10 +333,13 @@ static u32 i2c_readl(struct tegra_i2c_dev=20
>>> *i2c_dev, unsigned int reg)
>>> =A0=A0=A0=A0=A0 return readl_relaxed(i2c_dev->base +=20
>>> tegra_i2c_reg_addr(i2c_dev, reg));
>>> =A0 }
>>> =A0 -static void i2c_writesl(struct tegra_i2c_dev *i2c_dev, void *data,
>>> +static void i2c_writesl(struct tegra_i2c_dev *i2c_dev, u32 *data,
>>> =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 unsigned int reg, unsigned int =
len)
>>> =A0 {
>>> -=A0=A0=A0 writesl(i2c_dev->base + tegra_i2c_reg_addr(i2c_dev, reg), da=
ta,=20
>>> len);
>>> +=A0=A0=A0 while (len--) {
>>> +=A0=A0=A0=A0=A0=A0=A0 writel(*data++, i2c_dev->base + tegra_i2c_reg_ad=
dr(i2c_dev,=20
>>> reg));
>>> +=A0=A0=A0=A0=A0=A0=A0 i2c_readl(i2c_dev, I2C_INT_STATUS);
>>> +=A0=A0=A0 }
>>> =A0 }
>>> =A0 =A0 static void i2c_readsl(struct tegra_i2c_dev *i2c_dev, void *dat=
a,
>>> @@ -811,7 +814,7 @@ static int tegra_i2c_fill_tx_fifo(struct=20
>>> tegra_i2c_dev *i2c_dev)
>>> =A0=A0=A0=A0=A0=A0=A0=A0=A0 i2c_dev->msg_buf_remaining =3D buf_remainin=
g;
>>> =A0=A0=A0=A0=A0=A0=A0=A0=A0 i2c_dev->msg_buf =3D buf + words_to_transfe=
r *=20
>>> BYTES_PER_FIFO_WORD;
>>> =A0 -=A0=A0=A0=A0=A0=A0=A0 i2c_writesl(i2c_dev, buf, I2C_TX_FIFO, words=
_to_transfer);
>>> +=A0=A0=A0=A0=A0=A0=A0 i2c_writesl(i2c_dev, (u32 *)buf, I2C_TX_FIFO,=20
>>> words_to_transfer);
>> I've thought a bit more about this and I wonder if we're simply reading
>> out the wrong value for tx_fifo_avail and therefore end up overflowing
>> the TX FIFO. Have you checked what the value is for tx_fifo_avail when
>> this silent hang occurs? Given that this is specific to the VI I2C I'm
>> wondering if this is perhaps a hardware bug where we read the wrong TX
>> FIFO available count.
>>
>> Thierry
>
> Yes FIFO status shows all 8 slots available.

To be clear, TX slots availability in FIFO status before filling shows 8=20
words empty.

With writesl() when silent hang is seen, couldn't access any registers=20
and with debug messages I see silent hang happens immediate during=20
accessing any of i2c controller register after writesl() call

>
> Also, HW wise VI I2C is similar to other I2C and FIFO depth is also 8=20
> words. Confirmed from HW designers as well.
>
> Using writesl() causes silent hang after filling some words in FIFO=20
> and most of time after filling 8 words and sometime after filling it=20
> with around 6 words.
>
> I am not sure if this issue is specific to VI I2C alone as other I2C=20
> mostly use DMA for more than 8 words and I am not sure if we ever used=20
> other I2C in PIO mode for ~8 words transfer for slave devices we have=20
> on the platform.
>
>
> Thanks
>
> Sowjanya
>