Actually I found that there is some ambiguity with spi_message_init - it exists and is used by every spi-device-driver...
So I have started to think of why not call it: spi_message_(un)compile or spi_message_(un)optimize To make it totally clear that this is an optimization. So for now I have used the first. I will give measurements for timing-"differences" when using the stock spi-bcm2835 on my use-case with reduced traffic by moving to only 125khz CAN-bus-speed, to show there is no regression. Martin On 19.11.2013, at 16:02, Mark Brown wrote: > On Tue, Nov 19, 2013 at 02:11:12PM +0100, Martin Sperl wrote: > >> So I will call it: >> int spi_init_message(struct spi_device*,struct spi_message, >> unsigned long flags); >> int spi_done_message(struct spi_device*,struct spi_message); > >> (or you want to call it spi_message_init/done instead?) > > The latter might be better for consistency with spi_message_add_tail() > and similar. > >> I will post a patch for this soon. > > Excellent. -- To unsubscribe from this list: send the line "unsubscribe linux-spi" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
