From mboxrd@z Thu Jan 1 00:00:00 1970 From: Uwe =?iso-8859-1?Q?Kleine-K=F6nig?= Subject: Re: [2/5] i2c: davinci: query STP always when NACK is received Date: Mon, 24 Nov 2014 21:02:35 +0100 Message-ID: <20141124200235.GV4431@pengutronix.de> References: <1416477788-5544-3-git-send-email-grygorii.strashko@ti.com> <20141120221953.GI27002@pengutronix.de> <546F34B9.1000206@ti.com> <20141121131008.GQ27002@pengutronix.de> <546F5B51.1030006@ti.com> <20141123203314.GF4431@pengutronix.de> <547333EB.5050403@ti.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: <547333EB.5050403-l0cyMroinI0@public.gmane.org> Sender: linux-i2c-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Grygorii Strashko , Wolfram Sang Cc: linux-i2c-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Sekhar Nori , Kevin Hilman , Santosh Shilimkar , Murali Karicheri List-Id: linux-i2c@vger.kernel.org Hello Grygorii, On Mon, Nov 24, 2014 at 03:34:35PM +0200, Grygorii Strashko wrote: > On 11/23/2014 10:33 PM, Uwe Kleine-K=F6nig wrote: > > A call to .master_xfer with a message sequence implicitly expects A= CKs > > from the slave and doesn't tell anything about what should be done = on a > > NAK. So IMHO you must not send a P when the slave responds with a N= AK, > > but error out and let the sender decide if it wants to reply with P= or > > Sr. >=20 > Sry, but what should be done is defined by I2C/SMbus specs? Does it? > For SMBus devices, the specification states (http://smbus.org/specs/) > "4.2.Acknowledge (ACK) and not acknowledge (NACK)": > - "The slave device detects an invalid command or invalid data. In th= is=20 > case the slave device must not acknowledge the received byte. The m= aster > upon detection of this condition must generate a STOP condition and > retry the transaction" > For I2C devices, the specification states [http://www.nxp.com/documen= ts/user_manual/UM10204.pdf]: > "3.1.6 Acknowledge (ACK) and Not Acknowledge (NACK)" > "When SDA remains HIGH during this ninth clock pulse, this is defined= as the Not > Acknowledge signal. The master can then generate either a STOP condit= ion to > abort the transfer, or a repeated START condition to start a new tran= sfer." Yes, that's exactly what I meant. The master has the choice, and the driver should not eliminate options here. =20 > Let take a look on i2c/smbus xfer: > i2c_lock_adapter(adap) > adap->algo->master_xfer/smbus_xfer() > i2c_unlock_adapter(adap); > |- rt_mutex_unlock(&adapter->bus_lock); > |- task switch > =20 > So, there is no guarantee that next xfer will address the same I2C cl= ient device, > which, in turn, may lead to BB detection (will lead to BB detection i= f previous > transfer has been not acknowledged by SMbus client device). That's a valid concern. =20 > Small summary, I2C core + Davinci I2C driver provide ability to use r= epeated > start (Sr) only within one I2C transaction - which is a number of wri= te/read > operations specified by i2c_msg array. NACK always interrupts transac= tion > with -EREMOTEIO. >=20 > Also, the I2C core doesn't provide ability to manually send P. Hmm, Wolfram, what do you think? Best regards Uwe --=20 Pengutronix e.K. | Uwe Kleine-K=F6nig = | Industrial Linux Solutions | http://www.pengutronix.de/= |