From mboxrd@z Thu Jan 1 00:00:00 1970 From: wsa@the-dreams.de (Wolfram Sang) Date: Mon, 15 Aug 2016 00:34:37 +0200 Subject: [PATCH] i2c: at91: fix support of the "alternative command" feature In-Reply-To: <20160803145826.11927-1-ludovic.desroches@atmel.com> References: <20160803145826.11927-1-ludovic.desroches@atmel.com> Message-ID: <20160814223437.GG10501@katana> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Wed, Aug 03, 2016 at 04:58:26PM +0200, Ludovic Desroches wrote: > From: Cyrille Pitchen > > The "alternative command" feature was introduced with sama5d2 SoCs. > > Its purpose is to let the hardware i2c controller automatically send the > STOP condition on the i2c bus at the end of a data transfer. > Without this feature, the i2c driver has to write the 'STOP' bit into the > Control Register so the hardware i2c controller is triggered to send the > STOP condition on the bus. > > Using the "alternative command" feature requires to set the transfer data > length into the 8bit DATAL field of the Alternative Command Register. > Hence only data transfers up to 255 bytes can take advantage of the > "alternative command" feature. For greater data transfer sizes, the driver > should use the previous implementation, when the "alternative command" > support was not implemented yet. > > Signed-off-by: Cyrille Pitchen > Signed-off-by: Ludovic Desroches Applied to for-current, thanks! -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: not available URL: