[SOLVED] QSerialPort and Data Loss on Linux

[Leonardo]

Hello, everyone. I have a TTL printer and I send data to it using an USB-TTL converter. It works perfectly on Windows, but on Linux it's like some data is missing. I have some gaps in the printing output. As I said, it works fine on Windows, so I am assuming it's not really a problem related to Qt, but most likely to Linux. I've decided to ask here anyway, because I need to ensure a consistent behavior across platforms, and maybe some more experienced Qt programmer have already faced this before. I don't think QSerialPort itself can do much for me, but I don't know what else to try. I've been reading a lot about data loss in serial communication, but I haven't found anything applicable to Qt. Here's my code:

port = new QSerialPort("/dev/ttyUSB0", this);
port->setBaudRate(38400);
port->setParity(QSerialPort::Parity::NoParity);
port->setDataBits(QSerialPort::DataBits::Data8);
port->setStopBits(QSerialPort::StopBits::OneStop);
port->setFlowControl(QSerialPort::FlowControl::SoftwareControl);
if(port->open(QIODevice::OpenModeFlag::ReadWrite)){
....

Could anyone help me?

Thank you.

[mcosta]

Hi,

to exclude some programming error you should verify that all your data are correctly sent to the device.

If the transmission is ok, probably the problem is inside the device driver; i suggest to read the device manual to see how to set-up the driver. (usually drivers are shipped with test programs that allows to test the configuration).

Have you checked if the driver is correctly loaded?

[Leonardo]

Hi. I don't think it's a programming error, because it's working on Windows, but I've logged to a text file everything sent to the port anyway, right before

 port->write(data);
 port->waitForBytesWritten(-1);

I did it both on Windows and Linux and the output was the same. So the binary data my application produces is consistent.

The driver also loads without raising any error or warning.

[ 6.510918] USB Serial support registered for generic
[ 6.522729] usbserial: USB Serial Driver core
[ 6.551413] usbcore: registered new interface driver cp210x
[ 6.568382] USB Serial support registered for cp210x
[ 6.584142] cp210x 3-1.4:1.0: cp210x converter detected
[ 6.672059] usb 3-1.4: reset full-speed USB device number 4 using musb-hdrc
[ 6.782817] usb 3-1.4: cp210x converter now attached to ttyUSB0

I've tried updating it to the latest one I could find compatible with my kernel, but it didn't seem to have had any effect.

Based on what I've been reading, I feel it's somehow related to these parameters:

# stty -a -F /dev/ttyUSB0
speed 115200 baud;stty: /dev/ttyUSB0
line = 0;
intr = ^C; quit = ^; erase = ^?; kill = ^U; eof = ^D; eol = <undef>; eol2 = <undef>; swtch = <undef>; start = ^Q; stop = ^S; susp = ^Z; rprnt = ^R; werase = ^W;
lnext = ^V; flush = ^O; min = 1; time = 0;
-parenb -parodd cs8 hupcl -cstopb cread clocal -crtscts
-ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl ixon -ixoff -iuclc -ixany -imaxbel -iutf8
opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 ff0
isig icanon iexten echo echoe echok -echonl -noflsh -xcase -tostop -echoprt echoctl echoke

But I have no idea what they mean. :-(

[hpollak]

Have you tried to call flush() after the write?

[mcosta]

@Leonardo Sorry no Idea.

I only see the port support baudrate at 115200 and you're using 38400 (but I think is not an error)

[kuzulis]

Also maybe the reason is in SoftwareControl, because this mode (and HardwareControl) are not tested properly.. Maybe you can use "strace" to check for data that are really written to driver.

[Leonardo]

Hello, everyone. After some hours of debugging, I could finally find the problem. Long story short: apparently, the Linux driver for CP210x USB-TTL converter does not implement software flow control, but only hardware flow control. I've moved to the latter and it's working like a charm now. Thank you all for your help.