<pitillo>
morning. Two release optimizations up, rpi3 and pine64
<pitillo>
upgraded one rpi3 to 3.7 and it's running very well
<pitillo>
currently waiting for cubieboard2 and samsung-chromebook optimizations
<beerman>
( ͡° ͜ʖ ͡°)
<pitillo>
jaeger: with lasts changes made by sepen and some fixes I've on the issues list, the bootstrap runs flawlesly
<beerman>
nice work
<pitillo>
all optimizations are based on the generic release to avoid stage build
<pitillo>
but crux-arm-release allows to do a lot of veeeery interesting things (cut down ports, changes to libc and probably more than these one I can think atm with the coffee xD)
<jaeger>
nice progress :)
<pitillo>
getting an extrange bahavior with prt-get + fakeroot (makecommand sudo -H -u pkgmk fakeroot pkgmk). I'm getting pkgmk:pkgmk permissions on footprints instead of root:root
<pitillo>
and with ps, which isn't possible to get a process list (Ssps: Unable to get system boot time)
<pitillo>
I can see both problems in arm and arm64
<jaeger>
odd
Stalevar has quit [Ping timeout: 272 seconds]
Stalevar has joined #crux-arm
<pitillo>
both are very odd
<jaeger>
does the latter have anything to do with /var/run vs. /run?
<pitillo>
checking
<pitillo>
may be both
<pitillo>
I have /var/run and /run in place...
<jaeger>
Currently /var/run should be a symlink to /run
<pitillo>
no symlink...
<jaeger>
It's missed during an upgrade and caused some issues with things caring about utmp/wtmp, at least
<pitillo>
interesting... I need to review filesystem port
<pitillo>
here I have a live 3.7 and a 3.7 jail and both have the same behavior, two different places and no symlink
<jaeger>
It's something pkgadd doesn't handle properly, we need to add a fix to setup-helper on the ISO, haven't taken the time to figure out what's best yet
<jaeger>
there was a suggestion to move /var/run to /var/run.36 or similar and let the user sort it out
<jaeger>
On a couple of my own systems I just moved everything in /var/run to /run, removed /var/run, created the symlink, and rebooted
<jaeger>
Little bit messy but oh well
<pitillo>
ummmmmm
<pitillo>
I'll try to do an Indiana Jones job with both dirs on the rpi3...
<pitillo>
same behavior with ps
<pitillo>
after rebooting
<jaeger>
bummer... well, was worth a try
<pitillo>
yeah, something more learnt
<pitillo>
thanks for the tip
<pitillo>
pitillo@lokalix_rpi3:~
<pitillo>
$ ps aux
<pitillo>
root 1 2.3 0.1 2188 1468 ? Ssps: Unable to get system boot time
<pitillo>
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
<jaeger>
I should do a new build for the rpi4 with all the latest changes
Stalevar has quit [Read error: Connection reset by peer]
Stalevar has joined #crux-arm
<pitillo>
there are two pending changes for optimized devices,,, one of them pretty important to add OPTIMIZED_DEVICED to make calls
<pitillo>
there is another one to the build release, buut that can be done by hand copying pkgmk.conf and prt-get.conf files to the rootfs-stage1 before creating the tarball
<jaeger>
ok
<jaeger>
I've already noticed the DEVICE_OPTIMIZATION define
<pitillo>
es, but it's not working alone
<pitillo>
it's used always that optimization if make isn't called on the Makefile passing the DEVICE_OPTIMIZATION option