Current Server Time: 11:27 (Central Europe)

#apertus IRC Channel Logs

2017/11/03

Timezone: UTC


00:40
BAndiT1983
changed nick to: BAndiT1983|away
00:45
BAndiT1983|away
changed nick to: BAndiT1983
00:59
rton
left the channel
01:02
BAndiT1983
changed nick to: BAndiT1983|away
01:11
dimaursu16
joined the channel
01:11
dimaursu16
left the channel
01:11
dimaursu16
joined the channel
01:24
jucar
joined the channel
02:07
dimaursu16
left the channel
02:07
dimaursu16
joined the channel
02:07
dimaursu16
left the channel
02:07
dimaursu16
joined the channel
02:20
jucar
left the channel
02:37
dimaursu16
left the channel
04:25
Bertl
off to bed now ... have a good one everyone!
04:25
Bertl
changed nick to: Bertl_zZ
04:51
BAndiT1983|away
changed nick to: BAndiT1983
07:36
se6astian|away
changed nick to: se6astian
07:39
davidak
joined the channel
07:46
davidak
left the channel
07:46
davidak
joined the channel
07:53
BAndiT1983
changed nick to: BAndiT1983|away
07:56
BAndiT1983|away
changed nick to: BAndiT1983
08:39
dimaursu16
joined the channel
08:59
dimaursu16
left the channel
09:14
dimaursu16
joined the channel
09:14
dimaursu16
left the channel
09:14
dimaursu16
joined the channel
09:34
davidak
left the channel
10:24
jarhab[m]
se6astian: could you invertigate the github webhook again?
10:24
jarhab[m]
it seems to be broken and the ci pipeline is not running :(
10:27
se6astian
I am happy to
10:27
se6astian
but I need some input from you
10:27
se6astian
https://lab.apertus.org/T697#12378
10:28
BAndiT1983
webhook gives 404
10:28
BAndiT1983
{"error":"404 Not Found"}
10:30
jarhab[m]
hm... ok wait
10:31
BAndiT1983
ah, now i see that a parameter is missing, already wondered
10:32
BAndiT1983
is ref_name a branch name?
10:33
jarhab[m]
jup
10:35
jarhab[m]
jus use master
10:35
se6astian
saved
10:36
se6astian
now can we test by manually triggering?
10:36
se6astian
or should I just push a temp file?
10:36
jarhab[m]
jup
10:37
se6astian
worked
10:37
se6astian
at least the response is good now
10:37
jarhab[m]
nice :)
10:38
se6astian
does it build now on gitlab?
10:38
jarhab[m]
yeah the trigger was triggered sucessfully
10:38
jarhab[m]
yup ti should (maybe with 5 minutes delay)
10:39
se6astian
perfect
10:39
jarhab[m]
yes! the badge in https://github.com/apertus-open-source-cinema/beta-software canged to running and when you click it, you can see the piplie going
10:40
BAndiT1983
nice one
10:42
BAndiT1983
hm, how do we prevent too much build runs? a "develop" branch and from time to time merge to master?
10:42
BAndiT1983
*too many
10:47
davidak
joined the channel
10:49
se6astian
I think gitlab can handle "many" build runs fine
10:54
jarhab[m]
yeah
10:54
jarhab[m]
they esentially dont care. Once I ran like 20 build jobs in paralell... its ok
10:55
BAndiT1983
just some racks were glowing in their data center ;)
10:56
jarhab[m]
ð
10:56
jarhab[m]
maybe we should link to https://etcher.io/ and the latest image?
10:57
jarhab[m]
hm... maybe its a bit to early for this. Lets first build a stable and mature image :)
10:58
BAndiT1983
could you check some possible integration of the daemon, please? just for the future, second one will be the REST server, but will follow later as developing and testing is still going on
11:01
BAndiT1983
the install script should already contain the important steps
11:06
jarhab[m]
hm... im currently afk and dont know anythin about the daemon
11:07
jarhab[m]
just try calling your install script in the build_tools/inner/install.sh
11:07
BAndiT1983
will look into it, thanks for the directions
11:07
jarhab[m]
then build the image with the build instructions and try wether it succeeds
11:07
jarhab[m]
ok, cools
11:09
BAndiT1983
testing is a bit difficult, don't have the hardware here, but maybe se6astian can assist with swapping the firmware on the microzed test board when the daemon is ready
11:11
jucar
joined the channel
11:12
jarhab[m]
bertl said yesterday, that the image is not yet booting on real hardware
11:12
davidak
left the channel
11:12
BAndiT1983
ah, good to know, what about QEMU, any news on that part?
11:14
davidak
joined the channel
11:14
jarhab[m]
you can build qemu (with build_tools/outer/build_qemu.sh) and then start it using the spawn command from build_tools/outer/run_qemu.expect
11:17
BAndiT1983
were the repos of QEMU scripts and beta-software merged?
11:26
se6astian
jarhab[m]: did you get an error log from the bootprocess from bertl
11:26
se6astian
I can also try in the evening with a new sd card
11:38
se6astian
if it makes sense
12:03
davidak1
joined the channel
12:04
davidak
left the channel
12:23
dimaursu16
left the channel
12:28
dimaursu16
joined the channel
12:28
jarhab[m]
BAndiT1983: no, I just added this single script, because it is needed for the test step in the ci pipeline
12:28
jarhab[m]
se6astian: bertl send me a log, yes
12:36
dimaursu16
left the channel
12:40
dimaursu16
joined the channel
12:40
Bertl_zZ
changed nick to: Bertl
12:40
Bertl
morning folks!
12:41
rton
joined the channel
12:52
jarhab[m]
Bertl: do you have a link to a working BOOT.bin?
12:54
Bertl
there is one in every .dd image we have online
12:55
Bertl
http://vserver.13thfloor.at/Stuff/AXIOM/BETA/beta_20170109.dd.xz
12:56
Bertl
did you see the various messages on startup trying to create namespaces, etc? this needs to be fixed in the Arch setup
13:01
jarhab[m]
could you upload a plain BOOT.bin to 13thfloor?
13:03
jucar
left the channel
13:05
jarhab[m]
do you mean errors like `File /usr/lib/systemd/system/systemd-journald.service:33 configures an IP firewall (IPAddressDeny=any), but the local system does not support BPF/cgroup based firewalling.
13:05
jarhab[m]
` ?
13:07
Bertl
yes
13:48
dimaursu16
left the channel
13:52
Bertl
jarhab[m]: can you build u-boot already?
13:53
jarhab[m]
i didnt try...
13:53
jarhab[m]
but i could
13:53
jarhab[m]
just master of xillinx?
13:53
Bertl
okay, that would be great
13:53
Bertl
we currently still use the digilent branch IIRC, but I haven't tested with the xilinx master branch
13:54
Bertl
but we should definitely use our configuration for the build
13:54
Bertl
(similar for the kernel)
13:54
dimaursu16
joined the channel
13:54
dimaursu16
left the channel
13:54
dimaursu16
joined the channel
13:55
Bertl
the last step is combining the FSBL and the u-boot elf into boot.bin
13:57
jarhab[m]
where do i find the configs for u-boot and the kernel?
13:59
Bertl
I'll upload them shortly
13:59
Bertl
but it would probably make sense to have them in a dedicated place on github
13:59
Bertl
similar to the devicetree source
13:59
Bertl
and uEnv file
14:04
Bertl
http://vserver.13thfloor.at/Stuff/JARHAB/
14:05
Bertl
and I lied, we are already using xilinx u-boot
14:05
jarhab[m]
yeah definitly
14:06
jarhab[m]
ok how do you merge the u-boot elf and the fsbl?
14:07
Bertl
http://www.wiki.xilinx.com/Prepare+boot+image
14:07
Bertl
this requires the bootgen utility which combines the elf files into boot.bin
14:08
jarhab[m]
thx
14:20
Bertl
when checking the image yesterday, I saw that scripts and binaries got thrown together in /usr/local/bin, prefixed with axiom- and stripped of any extension, why's that?
14:30
jarhab[m]
because i thought it is convenient to use
14:30
jarhab[m]
you could just type axiom-<tab> and see all the scripts
14:31
jarhab[m]
and because of https://lab.apertus.org/T690
14:32
Bertl
okay, we should probably break them up into /usr/local/bin and /usr/local/script and keep extensions like .sh and .py
14:32
jarhab[m]
yeah
14:32
jarhab[m]
makes sense
14:33
Bertl
we also need the *.func for now
14:33
Bertl
pathes need to be adjusted in the scripts, but that's another topic
14:34
jarhab[m]
the scripts are just symlinked to /usr/local/bin
14:35
jarhab[m]
and all the scripts, that need the funcs change into their real dir (see https://github.com/apertus-open-source-cinema/beta-software/commit/acb0cf2834b031a18a3d494053547df372ff08a7)
14:36
Bertl
except that they do not find the *.func there, for whatever reason (haven't checked in detail)
14:38
jarhab[m]
hm... ok
14:39
jarhab[m]
do you think we should add /usr/local/script to the PATH?
14:45
Bertl
probably
14:46
Bertl
hmm, what is 'processing_tools' ?
14:52
pusle
left the channel
14:53
pusle
joined the channel
14:53
Bertl
ah, bit.func seems to be missing, which is included from i2c.func
14:55
se6astian
off for now
14:55
se6astian
changed nick to: se6astian|away
15:36
dimaursu16
left the channel
15:51
BAndiT1983
changed nick to: BAndiT1983|away
16:29
se6astian|away
changed nick to: se6astian
17:10
niemand
joined the channel
17:14
se6astian
changed nick to: se6astian|away
17:21
slikdigit
joined the channel
17:42
jarhab[m]
Bertl: processing tools is the directory which contains tools that are not image sensor, but image pipeline related
17:45
Bertl
I see ...
18:27
BAndiT1983|away
changed nick to: BAndiT1983
18:32
davidak1
left the channel
18:36
se6astian|away
changed nick to: se6astian
18:47
BAndiT1983
changed nick to: BAndiT1983|away
19:19
niemand
left the channel
20:46
BAndiT1983|away
changed nick to: BAndiT1983
20:52
davidak
joined the channel
20:53
dimaursu16
joined the channel
20:53
dimaursu16
left the channel
20:53
dimaursu16
joined the channel
21:08
dimaursu16
left the channel
21:11
dimaursu16
joined the channel
21:11
dimaursu16
left the channel
21:11
dimaursu16
joined the channel
21:23
se6astian
off to bed
21:23
se6astian
good night
21:23
se6astian
changed nick to: se6astian|away
22:39
BAndiT1983
changed nick to: BAndiT1983|away