Hi Niklas, thanks for the submission. I was finally able to test this change. On Wed, May 04, 2016 at 02:38:23PM +0200, Niklas Söderlund wrote: > Make it possible to transfer i2c message buffers via DMA. > Start/Stop/Sending_Slave_Address and some data is still handled using > the old state machine, it is sending the bulk of the data that is done > via DMA. > > The first byte of a transmission and the last two bytes of reception are > sent/received using PIO. This is needed for the HW to have access to the > first byte before DMA transmit and to be able to set the STOP condition > for DMA reception. > > Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@xxxxxxxxxxxx> Tested-by: Wolfram Sang <wsa+renesas@xxxxxxxxxxxxxxxxxxxx> I did regression tests on my Salvator-X trying to trigger previously known issues. Nothing bad happened. This could be expected since START and STOP is done in PIO mode, but one never knows :) Also did verify that DMA is triggered for bigger transfers. Did you have time to re-measure the threshold? Also, did you try booting without DMA and on Gen2? We had a bit of hazzle with !DMA with the sh_mobile-driver. Boot test and basic i2cdetect will suffice. Patch looks good, only minor nits: > + /* Do not use DMA for messages shorter then 8 bytes */ > + if (msg->len < 8) > + return; > + > + if (IS_ERR(chan)) > + return; Make this one if (a || b)? > @@ -516,6 +738,7 @@ static int rcar_i2c_master_xfer(struct i2c_adapter *adap, > out: > pm_runtime_put(dev); > > + Unrelated change ;) Thanks, Wolfram
