<lekernel> wpwrak_, maybe halt works. what does the board do when you type the command? do the LEDs go briefly dim lit, and then completely off?
<wpwrak_> hmm, i'll have to try when the current test run finishes. 512 cycles and counting ...
<GitHub111> [rtems-yaffs2] sebhub pushed 2 new commits to master: https://github.com/milkymist/rtems-yaffs2/compare/5e31a3f...5848a7e
<GitHub111> [rtems-yaffs2/master] C++ compatibility. - Sebastian Huber
<GitHub111> [rtems-yaffs2/master] Merge branch 'master' of github.com:milkymist/rtems-yaffs2 - Sebastian Huber
<qi-bot> The Firmware build was successfull, see images here: http://fidelio.qi-hardware.com/~xiangfu/build-milkymist/milkymist-firmware-09142011-1325/
<qi-bot> The Firmware build was successfull, see images here: http://fidelio.qi-hardware.com/~xiangfu/build-milkymist/milkymist-firmware-09142011-1700/
<qi-bot> The Firmware build was successfull, see images here: http://fidelio.qi-hardware.com/~xiangfu/build-milkymist/milkymist-firmware-09142011-2034/
<mwalle> wpwrak: without looking at the flickernoise source, i guess it does the reconfiguration via icap (and writing the standby bitstream offset before, lekernel?)
<mwalle> which should be the same as using jtag with cfg_in
<wpwrak> that would be convenient. because that's something i already know how to do :)
<mwalle> btw i tried to port your script to urjtags reconfigure command
<mwalle> it works every second time
<wpwrak> i'm still fighting with the repeatability of the corruption. i had it strike after ~4 hours, then after ~13.5 hours, then after ~2 hours, and now 24 hours have gone by without anything happening :-( (that's always with power cycling. 24 hours = ~1100 cycles)
<wpwrak> (every second time) oh, interesting :)
<wpwrak> that would be  pld reconfigure <address>  ?
<mwalle> the first time the command seems to set the address, and the second time triggers something
<mwalle> wpwrak: yep
<wpwrak> when you try my script, does it work each time ? or also only every 2nd time ?
<mwalle> iirc each time ;)
<wpwrak> then it should be easy to debug :)
<wpwrak> A works, B doens't. find the difference :)
<mwalle> wpwrak: yeah but no time too look further at it atm ;)
<lekernel> mwalle, yes, it does (but the icap code is in the rtems bsp)
<lekernel> also, a software shutdown should make sure that no flash operation is in progress when the standby bitstream is reloaded - using "pld reconfigure" won't
<qi-bot> The Firmware build was successfull, see images here: http://fidelio.qi-hardware.com/~xiangfu/build-milkymist/milkymist-firmware-09152011-0005/