Current Server Time: 05:46 (Central Europe)

#apertus IRC Channel Logs

2021/03/22

Timezone: UTC


23:08
BAndiT1983
changed nick to: BAndiT1983|away
01:26
vnksnkr
joined the channel
02:53
futarisIRCcloud
joined the channel
03:05
vnksnkr
left the channel
03:32
vnksnkr
joined the channel
03:52
Bertl_oO
off to bed now ... have a good one everyone!
03:52
Bertl_oO
changed nick to: Bertl_zZ
04:59
futarisIRCcloud
left the channel
05:08
BAndiT1983|away
changed nick to: BAndiT1983
06:12
lexano
left the channel
06:54
mumptai
joined the channel
06:59
Maqs
left the channel
07:00
Maqs
joined the channel
07:59
illwieckz
left the channel
07:59
vnksnkr
left the channel
08:05
illwieckz
joined the channel
08:49
lexano
joined the channel
09:03
se6ast1an
good day
09:06
BAndiT1983
hi
09:44
SurajKMahto[m]
joined the channel
10:28
Bertl_zZ
changed nick to: Bertl
10:28
Bertl
morning folks!
11:27
se6ast1an
good day
13:26
Bertl
off for now ... bbl
13:26
Bertl
changed nick to: Bertl_oO
13:43
illwieckz
left the channel
14:03
illwieckz
joined the channel
16:00
se6ast1an
MEETINNG TIME!
16:00
se6ast1an
who is here?
16:00
Bertl_oO
changed nick to: Bertl
16:00
Bertl
is here ...
16:00
BAndiT1983
is here
16:00
satacker[m]
hi
16:00
eppisai
is here
16:01
Shohei
joined the channel
16:01
Shohei
left the channel
16:01
Shohei
joined the channel
16:02
Shohei
is here
16:02
se6ast1an
great
16:02
se6ast1an
eppisai: do you want to start?
16:02
eppisai
yea!
16:03
eppisai
https://user-images.githubusercontent.com/54789531/112026838-9cde5280-8b5c-11eb-9b02-ea9a70f50129.mp4
16:03
eppisai
I am trying to run visulizer in windows.. but last 2 days have been bit depressing.. above is the progress so far..
16:04
Bertl
windows as in Microsoft Windows (tm)?
16:05
eppisai
yea.. :D I know its not a good or of much use, since BAandiT1983 had added docker.. but still.. thought it might help some one..in future
16:06
Bertl
well, that is always depressing I guess ... :)
16:06
se6ast1an
I was about to ask why this was necessary yes :)
16:07
se6ast1an
it adds more work in the future to maintain
16:07
eppisai
I don;t know, to be honest.. before it was useful , because there was no docker container..
16:08
satacker[m]
Well I have been meaning to add the simplest checksum like char_+=some_char to complete task 1 but got a bit busy
16:08
eppisai
but since i have started it, will finish it by tommrrow, maybe someday it might be of use..
16:09
se6ast1an
thanks for the update eppisai, anything else to report? what will you work on next?
16:09
se6ast1an
btw whats the state with the finished tasks, keyboard for example?
16:11
eppisai
yea.. will write the doc for 2 bit image, and continue on transition task,I had finished keyborad,but BAndiT1983 was busy before, so couldn't check on real board..
16:11
BAndiT1983
it was crashing, but it is not related to the keyboard at first
16:11
BAndiT1983
we have some bloat which overwrites heap, as stack grows too big
16:11
se6ast1an
yes, the known heap/stack problem we are facing currently I assume
16:11
se6ast1an
yes
16:12
se6ast1an
Shohei: would you like to report next?
16:12
Shohei
Thank you, se6ast1an :)
16:13
Shohei
Iâm afraid I donât have finished tasks this week, but reviewing the designs of PCB holding brackets and Front plate. Itâs very exciting for me :)
16:13
Shohei
Iâm putting together my questions and comments in Google Docs so I think I can share them to you as early as tomorrow night.
16:14
Shohei
That's all from my side. Thank you:)
16:14
se6ast1an
great, thanks
16:15
se6ast1an
for reference: I created two mechanical parts (adaptions) for the AXIOM Beta compact enclosure
16:15
se6ast1an
I finished the design and sent them to Shohei for review last week
16:15
se6ast1an
BAndiT1983: would you like to report/share?
16:15
BAndiT1983
not much, but can do
16:16
se6ast1an
please :)
16:16
BAndiT1983
was mainly working on SAME70 tests, to check if i can port the firmware there, the advantage is that the toolchain is linux native, so docker container is much smaller than for pic32mz, and works very well
16:17
BAndiT1983
besides that was guiding the students, creating some prototypes, that would be it
16:20
se6ast1an
can you share some insight into the pic32 docker container?
16:20
BAndiT1983
yep
16:21
BAndiT1983
it contains mainly xc32 and pic32prog, so the user just has to download the remote repo and to open the folder in the container in vscode, but it would also work, if the folder is mounted from command line and any other text editor
16:21
Bertl
xc32 is already installed there?
16:21
BAndiT1983
advantage is, that people don't have install all the stuff on their machine, as we can correct and imrpove the container centrally
16:21
BAndiT1983
yes
16:22
Bertl
how does that work with the Microchip license agreement?
16:23
BAndiT1983
haven't found any restrictions there with unattended install and free license, many such containers already exist, but that's also why i want to move to free toolchain like ARM GCC
16:24
se6ast1an
what evaluations are already tested/working with same70 and what issues are still to be tested/evaluated?
16:26
BAndiT1983
UART works, USB is slowly progressing, as i'm a bit limited with time currently, wanted to test also SPI soon, as i have fitting display here, would like to test one with parallel interface, but maybe later
16:26
BAndiT1983
am in fact slowly porting some parts f pic32 firmware over, as agnostic code doesn't change
16:27
se6ast1an
ok, thanks!
16:27
se6ast1an
anything else to add?
16:28
BAndiT1983
next week ;)
16:28
BAndiT1983
we don't want to spoil too much
16:28
se6ast1an
right :)
16:28
se6ast1an
quick updates from me:
16:28
se6ast1an
CAD design progress as shared above
16:29
se6ast1an
when the review is done we would have those manufactured and assembly a V3 compact ennclosure
16:29
se6ast1an
*assemble
16:30
se6ast1an
tele hardware production is still delayed/going slow for "componnent sourcing reasons" - no specifics have been shared with me yet but I asked and have a video conference tomorrow morning where I will also raise the topic again
16:31
se6ast1an
thats it from my side, Bertl the closer?
16:32
Bertl
yeah ...
16:33
Bertl
well, I've been chasing ghosts and fighting windmills for the last few days trying to identify problems on the remote beta and hardware we got from felix
16:33
lexano
left the channel
16:34
lexano
joined the channel
16:34
Bertl
I replaced connectors and swapped out parts, finally swapped the sensor and it is still not working reliably
16:35
Bertl
note that there were definitely some problems on each and every changed aspect but I haven't found and fixed everything yet
16:35
Bertl
there has been slow progress on building new ZIF sockets (we only got one for testing at the moment)
16:36
Bertl
unfortunately my righ to reflow the sockets broke yesterday, so I'm currently fixing that (light-bulb ala malcolm-in-the-middle style :)
16:37
Bertl
we also got a new hdmi digitizer to test, a magewell USB device, but I haven't had much time yet to test it
16:37
Bertl
that's it from my side for this week, hopefully more positive news next week
16:37
se6ast1an
many thanks, fingers crossed!
16:38
se6ast1an
anyone else who wants to report progress/share/etc.?
16:39
satacker[m]
I am rewriting Kernel Challenge with kernel driver best practices link sent by BAndiT1983
16:40
se6ast1an
excellent satacker[m]!
16:40
BAndiT1983
?
16:40
BAndiT1983
coding guidelines are for C++, Bertl would kill me if i try to teach you that
16:41
satacker[m]
Sorry i meant this
16:41
satacker[m]
https://www.kernel.org/doc/html/v4.10/process/coding-style.html
16:41
Bertl
that's the right one :)
16:43
se6ast1an
ok then MEETING CONCLUDED!
16:43
se6ast1an
as usual, thanks to everyone contributing/reportinng!
16:44
lexano
left the channel
16:44
Bertl
thanks for the moderation, great work, as always!
16:44
eppisai
Thanks :)
16:44
Shohei
Thank you :)
16:45
satacker[m]
Thank you
16:45
lexano
joined the channel
16:45
Shohei
left the channel
17:41
illwieckz
left the channel
17:42
illwieckz
joined the channel
18:14
markusengsner
joined the channel
18:19
markusengsner
good evening!
18:20
BAndiT1983
hi
18:48
rassu_12[m]
what changes do i need to make in "Remote_Firmware_Visualizer' repo in order to run it locally on my system(windows 10) ,also i can not enable hyper-v vartualisation in my PC as i enabled virtualization through BIOS my PC ran into problem,even i tried to reset my PC but at last i had to disable virtualization mode to bring my PC back in normal mode ,working properly.
18:48
rassu_12[m]
I also tried to download virtual machine from Oracle, so that i can it virtually on linux through windows but that also showed failure by showing 'VirtualBox error with AMD Processors: AMD-V is disabled in the BIOS (or by the host OS) (VERR_SVM_DISABLED)', i tried many things, took me 3-4 days but i wasn't able to do anything productive.....
19:03
Bertl
sounds like you need a new computer or a better operating system
19:06
BAndiT1983
rassu_12[m]: why do you need virtualization there?
19:06
BAndiT1983
visualiser should be buildable with some adjustments, mostly regarding SDL2 on windows, you need to download it and unpack, then probably point to it in cmake script
19:12
eppisai
there is more to it, even after that unpacking SDL in compiler(mingw or any compiler), their are some things in code that require few things that do work with normal mingw , and also MSVC C++ doesn't work at all, (like std:thread) so need to use mingw with pois thread enabled,
19:25
BAndiT1983
mingw? std::thread is pretty standard and should also work under windows with MSVC++
19:34
eppisai
It was working for me yesterday.. :( so read online, why it was seen as undeclared by mingw. MSVC had other issues.. for example <unistd.h> is not there in MSVC..
19:35
eppisai
http://www.cplusplus.com/forum/general/267237/
19:35
eppisai
*wasn't == it wasn't working for me yesterday
19:42
BAndiT1983
what is unistd.h used for?
20:03
markusengsner
BAndiT1983: my thinking with the utils namespace was that it would be more consistent to move all the utility functions into the namespace
20:04
BAndiT1983
would certainly help
20:05
markusengsner
it's not a very big change, but for me who is new to the codebase, the split was a little confusing
20:05
BAndiT1983
yeah, some changes were rather workarounds, as we needed quick solutions at some points
20:06
BAndiT1983
feel free to ask questions regarding structure, am sure that it can be optimized, just had no time to start the analysis with tools, e.g. UML generators or so
20:21
BAndiT1983
eppisai: by the way, you can comment unistd.h out, as it is just belonging to some prototype code, which got pulled into the dev branch
20:23
BAndiT1983
for cross-platform build you don't have to do some strange voodoo, it just requires some analysis to find common features, structure the source files properly and adjust cmake script, have done same stuff for OpenCine back when it still was active and it worked under linux, windows and even macosx
20:31
BAndiT1983
markusengsner: this one looks interesting -> https://www.sourcetrail.com/
20:32
eppisai
Yea, will remove, I just was searching for ways to enable cross platform scripting without having the need to change the codebase, just cmake and found the compiler's right configuration for unistd, so didn't read more about unistd.h or analysised it..
20:33
BAndiT1983
but that's the first step one needs to do, check what affects the build, jumping to another compiler won't help as you need to understand the basics of the module first
20:33
BAndiT1983
also every compiler has some differences, so this adds another layer of problems, if not approaches properly
20:34
BAndiT1983
*approached
20:35
markusengsner
BAndiT1983: that looks really useful! I'll have to check it out immediately :)
20:36
BAndiT1983
am already on it ;) am trying to connect with vscode to the appimage server
20:42
BAndiT1983
yep, works, as soon as i select "Send location" the UI scrolls to the class etc. in the UI
20:42
BAndiT1983
*select from VSCode
20:48
BAndiT1983
and if someone wants to keep code clean, then cppcheck is very helpful
20:49
BAndiT1983
but usually i prefer clang to gcc, as it has pretty nice analysis built in
21:02
markusengsner
also got Sourcetrail up and running now, so much better than just using `Go to References` in VSCode :)
21:02
BAndiT1983
;)#
21:02
BAndiT1983
*;)
21:03
BAndiT1983
you have also to add Bootloader as source folder, was also done as workaround, as our BL hasn't been developed further yet
21:04
eppisai
BAndiT1983: can't remove uinstd.h as we are using read and write operation after all in visualizer, but we can use <io> or <process> .. but again, windows build is not something we should have been doing in the first, as docker is already here.. :)
21:04
BAndiT1983
but in the future it should be some sort of BIOS, which controls the hardware and e.g. firmware will ask it to draw to display
21:04
BAndiT1983
and how will docker help without opengl support under windows?
21:04
eppisai
https://github.com/apertus-open-source-cinema/AXIOM-Remote/blob/37c4abbc94b6f437ef396919e093d907cd2ce4f7/AXIOM_Remote_Firmware_Visualizer/Main.cpp#L115
21:05
BAndiT1983
next thing i have to check is regarding the licenses, which is to be honest not my favourite task at all
21:05
BAndiT1983
what about the line?
21:06
eppisai
I can work on docker with u, always wanted to learn.. this line is using uinstd.h
21:07
mumptai
left the channel
21:08
eppisai
And i know nothing of it, as i thought opengl can also be made to run in docker.. but if need i can learn as much as needed..
21:09
BAndiT1983
like i said before, this line was added for the prototype, which would use serial port/socket ot communicate to the programmer script or to some mock service written in python, which would simulate the service on the camera
21:11
BAndiT1983
e.g. it would not only send answers to the visualiser, like real camera would do for the real remote, but also on same or another socket send data to the virtual camera to change temperature, hue, gain or whatever we would like to have to allow people to implement firmware without real camera in first place
21:12
BAndiT1983
there is not much a problem with docker, but with unclear possibility to install the compiler there, either the install file will be in the container and installed on first startup, in text mode to show the license agreement to the user, just need to check how to persist the data, otherwise it will install at every startup
21:14
BAndiT1983
lets see what wsl gpu drivers will actually bring, but opengl is possible -> https://medium.com/@benjamin.botto/opengl-and-cuda-applications-in-docker-af0eece000f1
21:15
BAndiT1983
changed nick to: BAndiT1983|away
22:06
markusengsner
left the channel