On Wed, Jul 11, 2018 at 10:10 AM Joe Perches <joe@xxxxxxxxxxx> wrote:
>
> On Wed, 2018-07-11 at 09:53 -0700, Jae Hyun Yoo wrote:
> > On 7/10/2018 10:42 PM, Brendan Higgins wrote:
> > > On Mon, Jul 2, 2018 at 2:14 PM Jae Hyun Yoo <jae.hyun.yoo@xxxxxxxxxxxxxxx> wrote:
> > > > There are some log printing without a newline character. This
> > > > patch adds the missing newline characters.
> []
> > > > diff --git a/drivers/i2c/busses/i2c-aspeed.c b/drivers/i2c/busses/i2c-aspeed.c
> []
> > > > @@ -431,7 +431,7 @@ static bool aspeed_i2c_master_irq(struct aspeed_i2c_bus *bus)
> > > > */
> > > > if (bus->master_state == ASPEED_I2C_MASTER_START) {
> > > > if (unlikely(!(irq_status & ASPEED_I2CD_INTR_TX_ACK))) {
> > > > - pr_devel("no slave present at %02x", msg->addr);
> > > > + pr_devel("no slave present at %02x\n", msg->addr);
> > >
> > > Unless something changed in the last couple versions of the kernel, this is the
> > > only line that actually changes anything. dev_* inserts a newline for every
> > > call.
>
> Not true.
>
> Any printk without KERN_CONT inserts a newline
> if the last character
> emitted is not a newline.
>
> dev_<level> uses can also be followed by pr_cont.
>
> So this patch does reduce the possibility of
> interleaved messages from multiple processes.
My mistake. Thanks for pointing that out.
Jae, forget what I said earlier. This looks good to me.
