Table of Contents
Linux, Serial, C, and Intel vs. ARM
I'm working on some mcu stuff and I kept running into the same problems when writing code to talk to it on Linux. I've been writing code on my desktop, moving it to either my Chromebook or an RPi.
So I'm writing it down sos I remember. I thought it might be because of the ch341 UART adaptor I had on my shitty arduino clone, but I'm running into it elsewhere.
- In c_iflag, IRCNL is set on be default on ARM, converting CR's into NL's. Annoying when you write stuff designed to be understood by humans and machines.
- cfsetispeed and cfsetospeed don't seem to work. cfsetspeed does.
- O_NONBLOCK never seems to work right. Leave it alone and use c_cc[VMIN || VTIME] instead, because that's smarter anyway.
There will probably be more here.
Get Out Of Infinality Part 1
This is what I do to my systems to get them up to date with freetype 2.7, which invalidates1) the infinality patches. This assumes you have multilib installed. Remove references to lib32- packages if you don't.
Step 0: Update system and get rid of orphans to make sure things are clean. Make sure you have an alternative wm like windowmaker installed so freetype2 problems aren't deadly.
vis.nu Cleanup
One of the problems with offering services to friends and acquaintences for free is that when they don't use your stuff anymore, they neglect to tell you.
There's about 18 years of disused email accounts, websites, and all sorts of weirdness in the vis.nu servers, and starting today I'm cleaning it all up. I'm trying to make sure that I'm not getting rid of things that people are using, but rather than erring on the side of caution, I'm erring on the side of personal simplicity.
I'll be saving things for a while, but not likely permanently. If I step on something you want, contact me via mail or FB or whatever and I'll send you a copy. If you want it to stay on my servers, we can discuss that.
New IXGBE taurball
Updated for the most recent lts, and changes in the build process.
Mystery PanDigital Digital Picture Frame
Not much to it. Got it at goodwill, found a header on it with serial output at 115200 TTL. I *think* I've found the input, but it hasn't been sensitive to input yet. What it does print out, however, is this:
MP620 MaskRom V002 SDRAM 8Mx16/16Mx16 Load NAND NAND ID : 000000D3 AP found AP size : 0038AA10 AP check sum : FC394CD8 load OK ../source/main.c 1159 Enable PLL1 spread spectrum CPU Initiale to 96MHz -I- CPU Initial to 0077083f dwMainProcSize 00527000 Network buffer pool 8094f000 00072000 disp buffer1 80773980 000ed800 disp buffer2 80861180 000ed800 Heap buffer 809c0980 0063f680 !!!Free_CacheWin 0