From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wolfram Sang Subject: [RFT] i2c: sh_mobile: avoid unnecessary register read Date: Thu, 9 Nov 2017 23:20:53 +0100 Message-ID: <20171109222053.27093-1-wsa+renesas@sang-engineering.com> Return-path: Sender: linux-renesas-soc-owner@vger.kernel.org To: linux-i2c@vger.kernel.org Cc: linux-renesas-soc@vger.kernel.org, Jacopo Mondi , Yoshihiro Shimoda , Wolfram Sang List-Id: linux-i2c@vger.kernel.org There is no data when the first WAIT interrupt arrives. No need to read something then. Signed-off-by: Wolfram Sang --- I am not super happy with (real pos >= 0) done twice, but I didn't find a better solution yet. The compiler will make this cheap anyhow, I guess. Jacopo: can you please test this on top of all other patches? drivers/i2c/busses/i2c-sh_mobile.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/i2c/busses/i2c-sh_mobile.c b/drivers/i2c/busses/i2c-sh_mobile.c index 80561ffbcf7b46..40a66d466c3c49 100644 --- a/drivers/i2c/busses/i2c-sh_mobile.c +++ b/drivers/i2c/busses/i2c-sh_mobile.c @@ -433,8 +433,9 @@ static int sh_mobile_i2c_isr_rx(struct sh_mobile_i2c_data *pd) break; } data = i2c_op(pd, OP_RX_STOP_DATA, 0); - } else + } else if (real_pos >= 0) { data = i2c_op(pd, OP_RX, 0); + } if (real_pos >= 0) pd->msg->buf[real_pos] = data; -- 2.11.0