On Thu, 2018-03-01 at 20:59 +0000, Joshua Scott wrote: > Hi, > > While decreasing or removing the delay may work at 4 Mbps, this will > break behaviour at lower baud rates. Currently you adding difficulties to most of our users (Bluetooth) since they are working at >= 3Mbps. > When adding the delay, I tested down to 9600 baud (the default on many > user-facing devices). The combination of 1000 retries of 10 µs was > enough to allow the "resize" command (busybox) to function correctly. > Without the delay, characters were going missing in the middle of a > command sequence, resulting in a messed up terminal window on the > other end of the serial cable (minicom). Smaller delays do work for > higher baud rates, but support for 9600 and other common low baud > rates should be maintained. There is no 1st or 2nd class of users, they all should be considered equal. Without your patch only you complained, now, complained almost everyone, except you. See the point? > A couple of alternatives to removing the delay: > > * Moving from 1,000 retries to 10,000 retries of 1 µs each > * Having the delay occur only below a certain baud rate In any case delay must be _as little as possible_. Please, fix your stuff. Otherwise I would insist to revert (Btw, why you didn't include me in the first place? I'm a designated reviewer of the driver). -- Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx> Intel Finland Oy -- To unsubscribe from this list: send the line "unsubscribe linux-serial" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
