Trammell Hudson

It appears that we can find the address from `/sys`. On qemu: ``` # hd /sys/bus/pci/devices/0000\:00\:00.0/config 00000000 86 80 c0 29 00 00 00 00 00 00 00 06 00...

On skylake (x1 gen4) the address is `0xf8000000`, but there doesn't seem to be the RCBA pointer there. ``` sudo xxd -e -g4 /sys/bus/pci/devices/0000\:00\:00.0/config 00000000: 19048086 20900006 06000008 00000000 ..........

[sysklogd](https://github.com/troglobit/sysklogd) is a fairly small logger and seems to work. This is enough in `/etc/syslogd` to send it to outside machines: ``` *.* /dev/console *.* @safeboot:9999 ; RFC5424 ``` busybox...

especially since these micro-vm's have no way to login to run htop or see what is going on.

my guess is undefined behaviour due to this wildcard: ``` EXPORTED_SYMBOLS := $(wildcard $(srcdir)/.exported_symbols.Base $(srcdir)/.exported_symbols.$(EXPORTED_UC)_[0-9_]*[0-9]) ```

 I used http://www.pentacom.jp/pentacom/bitfontmaker2/ to draw the font and output the bitmap. It also supports TTF export.

Thanks for the feedback on the patch. I'll make the style corrections and push an updated version.

Thanks for doing the performance testing on the patch, @jfeather-amd . Can you describe where the slowdowns seem to be? In the non-vlan, non-echo, non-validating case (the default), my latency...

I've re-run tests on the X3 cards with better isolation and pinning the `eflatency` task to a single CPU; the results show no change in the `min`, `50%`, `95%` and...

Maybe calling firmware version in the constructor is breaking the test due to the serial port mocking? I'll try moving it to the `src/server.ts` calls to `new EBB` instead.