<wpwrak>
hmm. the FLL works a specified. with quite precisely the -558 ppm error they designed in. if they had been a little bit more reckless, it could run with +125 ppm. well, i can detune the crystal to still make it work, but ...
<DocScrutinizer05>
I wonder what for you need such precision in system clock
<DocScrutinizer05>
ok, on 2GHz it's already 500kHz
<DocScrutinizer05>
err 1MHz
<wpwrak>
for USB. USB is picky.
<wpwrak>
and it's also nice to have a decent RTC
<wpwrak>
though i can correct any constant error there by software,
rz2k has joined #qi-hardware
xiangfu has joined #qi-hardware
<DocScrutinizer05>
exactly
<DocScrutinizer05>
anyway I doubt you want to use a sysclock based RTC
<DocScrutinizer05>
prolly way too greedy to do the CMOS clock
<DocScrutinizer05>
there's a lot of reasons why all such RTC use a true 32kHz crystal
<wpwrak>
i am using a 32 kHz crystal :)
<wpwrak>
the FLL then generates a 48 MHz clock from it
<wpwrak>
but if i have to detune the 32 kHz clock to get the 48 MHz right, then i also get incorrect RTC readings, so i'll have to correct these "back"
pabspabspabs is now known as pabs3
xiangfu has quit [Ping timeout: 272 seconds]
xiangfu has joined #qi-hardware
atommann has joined #qi-hardware
atommann has quit [Ping timeout: 250 seconds]
atommann has joined #qi-hardware
<wpwrak>
hmm, wrestled the beast down to 504 uA "suspended". that's about 500 uA more than i'd like to see ... i wonder who is stealing all that power. still haven't equipped all the components, so the number of suspects is small ...
xiangfu has quit [Ping timeout: 272 seconds]
xiangfu has joined #qi-hardware
<wpwrak>
grmbl. qi-hw servers down again :-( i wonder what's killing them so often ... moribund hardware ? "visitors" ? the global conspiracy of ddos'ing script kiddies ?
wolfspraul has joined #qi-hardware
qi-bot has joined #qi-hardware
jekhor has joined #qi-hardware
<wpwrak>
wolfspraul: thanks ! ;-)
<wolfspraul>
I need to move to a new physical machine though
<wolfspraul>
It's only stable for minutes now - enough...
<wpwrak>
wolfspraul: btw, /srv/www/downloads.qi-hardware.com seem to fail mounting quite regularly. i wonder if setting it to bg may help
<wpwrak>
oh, so the hw is dying
qi-bot has quit [Ping timeout: 272 seconds]
qi-bot has joined #qi-hardware
qi-bot has quit [Ping timeout: 245 seconds]
atommann has quit [Quit: Leaving]
atommann has joined #qi-hardware
atommann has quit [Ping timeout: 260 seconds]
atommann has joined #qi-hardware
<kyak>
must.. resist... moving... to cloud.. :)
atommann has quit [Ping timeout: 250 seconds]
qi-bot has joined #qi-hardware
atommann has joined #qi-hardware
mirko has quit [Ping timeout: 272 seconds]
mirko has joined #qi-hardware
xiangfu has quit [Ping timeout: 240 seconds]
jekhor has quit [Ping timeout: 245 seconds]
qi-bot6 has joined #qi-hardware
qi-bot has quit [Read error: Connection reset by peer]
qi-bot6 is now known as qi-bot
jekhor has joined #qi-hardware
atommann has quit [Ping timeout: 245 seconds]
rodgort has quit [Quit: ERC Version 5.3 (IRC client for Emacs)]
qi-bot has quit [Ping timeout: 250 seconds]
jekhor has quit [Ping timeout: 260 seconds]
rodgort has joined #qi-hardware
atommann has joined #qi-hardware
mirko has quit [Ping timeout: 256 seconds]
mirko has joined #qi-hardware
atommann has quit [Quit: Leaving]
jekhor has joined #qi-hardware
jekhor has quit [Ping timeout: 260 seconds]
kilae has joined #qi-hardware
kristianpaul has quit [Quit: leaving]
pcercuei has joined #qi-hardware
kristianpaul has joined #qi-hardware
valhalla has quit [Ping timeout: 260 seconds]
valhalla has joined #qi-hardware
wolfspraul has quit [Ping timeout: 255 seconds]
wolfspraul has joined #qi-hardware
<wpwrak>
hmm .... <discussion@lists.en.qi-hardware.com>... Deferred: Connection timed out with turandot.qi-hardware.com.
jekhor has joined #qi-hardware
<larsc>
pcercuei: I was informed that you haven't signed up for the christmas party yet
<pcercuei>
hmm yes
<wolfspraul>
wpwrak: I'm setting up a new machine
<larsc>
pcercuei: is that a "Yes I'll be there"?
<pcercuei>
yes
<wpwrak>
wolfspraul: great :) thanks !
<DocScrutinizer05>
wolfspraul: elcos gone dry and popped up?
rz2k has quit []
<DocScrutinizer05>
(the usual fatal failure pattern of old PC boards)
<DocScrutinizer05>
except when fan gets stuck and whole PC overheats
Ornoterm1s has quit [Quit: leaving]
atommann has joined #qi-hardware
Ornotermes has joined #qi-hardware
atommann has quit [Quit: Leaving]
Ornotermes has quit [Quit: leaving]
Ornotermes has joined #qi-hardware
<pcercuei>
what possibilities do I have to debug a kernel which won't start at all?
<larsc>
jtag!
<pcercuei>
I don't have jtag, unfortunately
<pcercuei>
anything that can be used on serial? KGDB?
<pcercuei>
from what I can see KGDB can be used on a running kernel, not to debug a kernel that fails to boot
<larsc>
strategically placed gpio toggles
jekhor has quit [Read error: Connection reset by peer]
jekhor has joined #qi-hardware
<DocScrutinizer05>
exactly. Use the LEDs or whatever you got, to signal checkpoints your code reaches
<DocScrutinizer05>
[2014-07-28 Mon 01:48:37] <wpwrak> well, in a bit, i'll make my code return that register value. then i'll know for sure what's cooking there. for now, all i have is a lone LED :)
<DocScrutinizer05>
[2014-07-28 Mon 01:49:41] <whitequark> hook a scope there - then shift it out
<whitequark>
so, somewhat relevant
<DocScrutinizer05>
very convenient: a piezo beeper
<whitequark>
I found an old ar9285 wifi card
<whitequark>
and turns out it has JTAG and UART pins on the sides, and open-source firmware for a very close device in same family, and the host interface allows to arbitrarily dump RAM
<whitequark>
I want to run something there
<whitequark>
96k SRAM, 1M of architectural SPI (i.e. SoC-integrated) flash
<whitequark>
and if I understand it correctly, ar9k (PHY+MAC) series is basically the same IP core as their earlier ar5k (PHY) series and some CPU with a PCI interface that twiddles it
<whitequark>
so that FOSS firmware just contains a port of mac80211 stack (!)
<DocScrutinizer05>
sounds about correct
<whitequark>
only problem is, i have no idea how to program the PCI-e interface
<whitequark>
no FOSS drivers, firmware or even docs exist that mention PCI or PCI-e host iface
<DocScrutinizer05>
umm, on card side?
<whitequark>
yes
<DocScrutinizer05>
yeah, prolly pretty unusual platform for FOSS
<whitequark>
it doesn't make it easier that they use some incredibly obscure CPU architecture I've never even heard of
<whitequark>
tensilica
<whitequark>
*tensilica xtensa
<DocScrutinizer05>
errr
<whitequark>
apparently it is synthesizable and customizable, and tailored for DSP
<whitequark>
but I've seen prcisely 0 DSP operations in FOSS firmware
<whitequark>
which makes me wonder, why not just put a cortex-m there or at least keep using SH-2, as they used to