skoehler

The Gentoo bug for this problem is https://bugs.gentoo.org/show_bug.cgi?id=571526 However, they don't seem to make any progress.

Am 09.02.2016 um 14:55 schrieb Karol Herbst: > funny, it works for me on gentoo No, it doesn't. x11-drivers/nvidia-drivers-358.16-r4 is the last version that doesn't produce the libGLX.so not found...

Am 10.02.2016 um 15:35 schrieb Karol Herbst: > sure it does > > |$ equery files nvidia-drivers | grep -i libGLX.so > /usr/lib32/opengl/nvidia/lib/libGLX.so > /usr/lib32/opengl/nvidia/lib/libGLX.so.0 > /usr/lib64/opengl/nvidia/extensions/libglx.so > /usr/lib64/opengl/nvidia/extensions/libglx.so.361.18 >...

I'm using Arch Linux. The problem is not that I'm missing some packages. I have lm_sensors and I have the equivalent of i2c -tools and libi2c-dev installed (/usr/include/linux/i2c-dev.h exists) but...

Find attached the data that I captured with my logic analyzer at 8MHz. pyserial was instruction to open the COM port at 57600 baud, no parity, 8 data bits, 1...

I forgot to mention that the same setup (rx connected to tx pin) works just fine with other programs, like hterm. I could send a 6MB text file and hterm...

Connect the RX and TX pins your favorite FTDI adapter and execute the script below. It will fail within seconds. You can see data corruption like on the screenshot below....

I assume the problem here is that setting the timeout basically causes the port to be reconfigured. It's impressive: you have one big function that sets all kinds of parameters...

Re-configuring the WHOLE port (baudrate etc.) to set the timeouts is rubbish. It's a design flaw in pyserial. It has so many, like the read semantics (see my other bug...

My personal workaround was to set the timeout to a fixed value (say 1 second) and then have a wrapper function that simply retries 5 times when the total timeout...