Hello, On 8/22/19 9:04 AM, Ahmad Fatoum wrote: > > > On 8/5/19 11:11 AM, Sascha Hauer wrote: >> On Wed, Jul 31, 2019 at 12:21:41PM +0200, Ahmad Fatoum wrote: >>> Both the PBL and simple console only return number of input bytes, not >>> number of bytes actually written out. These differ, because each LF is >>> converted to CRLF pairs. >>> >>> The behavior of not counting actual written out characters is more sensible, >>> because otherwise callers interested in finding out if all bytes have been >>> written (e.g. to avoid incomplete writes with ratp) would need to keep count >>> of all line feeds in the string. >>> Therefore change the normal console to behave like its less featureful >>> brethren. >> >> According to "man puts" puts() returns a non negative number on success. >> I can't find a place where it's claimed that we have to return the >> number of characters. > > console_puts in common/console_simple.c and pbl/console.c already return the > number of characters. Same for all the console_device puts members in the > drivers. > >> I also can't find a place where the return value >> of any puts like function in barebox is ever evaluated. Given that it >> might make more sense to just return zero in the absense of an error. > > I too think there isn't, but it would be less surprising if common/console.c > would behave like other puts's in barebox. Sorry, I mixed it up. Ye, it's arguable whether they should return number of input bytes written or zero. But no other puts in barebox counts new lines twice, so I think the patch has its merit, even if a follow up patch in future has puts functions return 0 instead. > >> >> Sascha >> > -- Pengutronix e.K. | | Industrial Linux Solutions | http://www.pengutronix.de/ | Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 | _______________________________________________ barebox mailing list barebox@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/barebox
