From mboxrd@z Thu Jan  1 00:00:00 1970
From: "Sricharan" <sricharan@codeaurora.org>
Subject: RE: [PATCH 1/2] i2c: qup: Cleared the error bits in ISR
Date: Wed, 11 May 2016 21:27:40 +0530
Message-ID: <000701d1ab9d$df9b3100$9ed19300$@codeaurora.org>
References: <1462797871-8595-1-git-send-email-absahu@codeaurora.org> <1462797871-8595-2-git-send-email-absahu@codeaurora.org>
Mime-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
Return-path: <linux-i2c-owner@vger.kernel.org>
Received: from smtp.codeaurora.org ([198.145.29.96]:48603 "EHLO
	smtp.codeaurora.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751401AbcEKP5y (ORCPT
	<rfc822;linux-i2c@vger.kernel.org>); Wed, 11 May 2016 11:57:54 -0400
In-Reply-To: <1462797871-8595-2-git-send-email-absahu@codeaurora.org>
Content-Language: en-us
Sender: linux-i2c-owner@vger.kernel.org
List-Id: linux-i2c@vger.kernel.org
To: 'Abhishek Sahu' <absahu@codeaurora.org>, agross@codeaurora.org
Cc: architt@codeaurora.org, linux-arm-msm@vger.kernel.org, ntelkar@codeaurora.org, linux-kernel@vger.kernel.org, andy.gross@linaro.org, linux-i2c@vger.kernel.org, dmaengine@vger.kernel.org, linux-arm-kernel@lists.infradead.org, wsa@the-dreams.de

Hi,

> 1. Current QCOM I2C driver hangs when sending data to address 0x03-0x07
> in some scenarios. The QUP controller generates invalid write in this
case,
> since these addresses are reserved for different bus formats.
> 
> 2. Also, the error handling is done by I2C QUP ISR in the case of DMA
mode.
> The state need to be RESET in case of any error for clearing the available
> data in FIFO, which otherwise leaves the BAM DMA controller in hang state.
> 
> This patch fixes the above two issues by clearing the error bits from I2C
and
> QUP status in ISR in case of I2C error, QUP error and resets the QUP state
to
> clear the FIFO data.
> 
> Signed-off-by: Abhishek Sahu <absahu@codeaurora.org>
> ---
>  drivers/i2c/busses/i2c-qup.c | 14 ++++++++------
>  1 file changed, 8 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/i2c/busses/i2c-qup.c b/drivers/i2c/busses/i2c-qup.c
index
> 23eaabb..8c2f1bc 100644
> --- a/drivers/i2c/busses/i2c-qup.c
> +++ b/drivers/i2c/busses/i2c-qup.c
> @@ -213,14 +213,16 @@ static irqreturn_t qup_i2c_interrupt(int irq, void
> *dev)
>  	bus_err &= I2C_STATUS_ERROR_MASK;
>  	qup_err &= QUP_STATUS_ERROR_FLAGS;
> 
> -	if (qup_err) {
> -		/* Clear Error interrupt */
> +	/* Clear the error bits in QUP_ERROR_FLAGS */
> +	if (qup_err)
>  		writel(qup_err, qup->base + QUP_ERROR_FLAGS);
> -		goto done;
> -	}
> 
> -	if (bus_err) {
> -		/* Clear Error interrupt */
> +	/* Clear the error bits in QUP_I2C_STATUS */
> +	if (bus_err)
> +		writel(bus_err, qup->base + QUP_I2C_STATUS);
> +
> +	/* Reset the QUP State in case of error */
> +	if (qup_err || bus_err) {
>  		writel(QUP_RESET_STATE, qup->base + QUP_STATE);
>  		goto done;
>  	}
       In qup_i2c_xfer and qup_i2c_xfer_v2 state is set to RESET at the end,
when
       there is no error. So would it be fine if we do it there
unconditionally ?

Regards,
 Sricharan