From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wolfram Sang Subject: Re: [PATCH V7 3/6] i2c: qup: Transfer each i2c_msg in i2c_msgs without a stop bit Date: Sun, 24 Jan 2016 12:29:16 +0100 Message-ID: <20160124112915.GA1775@katana> References: <1453197766-18976-1-git-send-email-sricharan@codeaurora.org> <1453197766-18976-4-git-send-email-sricharan@codeaurora.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="EeQfGwPcQSOJBaQU" Return-path: Received: from sauhun.de ([89.238.76.85]:44882 "EHLO pokefinder.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751380AbcAXL3Z (ORCPT ); Sun, 24 Jan 2016 06:29:25 -0500 Content-Disposition: inline In-Reply-To: <1453197766-18976-4-git-send-email-sricharan@codeaurora.org> Sender: linux-i2c-owner@vger.kernel.org List-Id: linux-i2c@vger.kernel.org To: Sricharan R Cc: devicetree@vger.kernel.org, linux-arm-msm@vger.kernel.org, agross@codeaurora.org, linux-kernel@vger.kernel.org, linux-i2c@vger.kernel.org, iivanov@mm-sol.com, galak@codeaurora.org, dmaengine@vger.kernel.org, linux-arm-kernel@lists.infradead.org, andy.gross@linaro.org, ntelkar@codeaurora.org, architt@codeaurora.org --EeQfGwPcQSOJBaQU Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Hi, > "If this is the last message in a group, it is followed by a STOP. > Otherwise it is followed by the next @i2c_msg transaction segment, > beginning with a (repeated) START" This is correct. > So the expectation is that there is no 'STOP' bit inbetween individual > i2c_msg segments with repeated 'START'. The QUP i2c hardware has no way > to inform that there should not be a 'STOP' at the end of transaction. > The only way to implement this is to coalesce all the i2c_msg in i2c_msgs > in to one transaction and transfer them. Adding the support for the same. So, there will not be a REP_START condition on the bus? I am sorry to say that I can't accept this. A REP_START is a REP_START and nothing else. There are devices which will get confused if there is no real REP_START condition. Without knowing the HW in detail, can't you implement I2C_M_NOSTART and let the touchscreen driver use it via regmap? That would be the proper way (from what I understand). Regards, Wolfram --EeQfGwPcQSOJBaQU Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQIcBAEBAgAGBQJWpLWLAAoJEBQN5MwUoCm2l+AQAIgY4+l/7rEBOcC6kUTlPoJo 5hpuGyigAoiJfHir9TaO6LkjKnM4ouhbVjV1AOfHVhwR7q3CvHcpKV2tWxJ6ymJr snrzRr2akAoB0pVssZLxyE4OOJjEBx+ZVt9/cE+UhWYdWIlEI8JlbELcVFxKGMZv 4+b3Ee52k3SJRX5XQODDlZqoDxF5mBjBASZtjmTlI7IMFY12kqRZIdE8I9m7yGRO KL+rkl9ZfZPHiSRfdwTRDyVr0iU88t72+FSo7a2qk0/1YOyRkdtXaOo3YRMlCWxq Bdjl18aGchequPsvRbTlP3RV9jinrmx4E3P6DSmqapwTnA8rZl+SrFRbcnqvq01y 4lQe8J/2vJCnCXTCzhF8o4zm1GD5eBLwdHGIUumOXH85LvZspU0nMgKOkRhgHtW+ +XGoy4d9MxjoyM/1aqa3SPsa+Aew1FKJT32vchlxwUAdiMIPyupqP6tK5778HfKY QG78Eb/jJEtLZkJCU8XwkH68kgOGeY/6e6mPcBPbFt7ek5L3SMr30Nk2KXz3RFYN OIRj+TUYLwIjoPUs01bmgBMMEJx9BwAzlkM/SxT1Jv8tw+nGD/AJmyIDDhX0sYz/ KJAcaiX6txLgEkUmQSr7HvuMkQShG70lzt631wlI3NwzgehN9CYHyZtKpeK7hBpg NYojpudFdMxXS+tj3vGT =+daC -----END PGP SIGNATURE----- --EeQfGwPcQSOJBaQU--