Current Server Time: 00:04 (Central Europe)

#apertus IRC Channel Logs

2020/03/10

Timezone: UTC


05:27
metal_dent[m]
Hi BAndiT1983|away : I've written a test fro DrawIcon() but it's failing in the wrongDirection condition. I just checked out your commits in which you made some changes in PainterMod so should I update my repo or wait for you to finish doing the changes?
05:33
pratyush
joined the channel
05:35
pratyush
Bertl: we have to take input from user space and the we will generate the checksum of the data and write it byte by byte to the proc file
05:36
pratyush
and the process has to be repeated again using ioctl calls
05:37
Bertl_oO
character device ... so the input comes from there
05:37
Bertl_oO
checksum via procfs as well as IOCTL
05:37
Bertl_oO
so the checksum can be retrieved from procfs as well as via IOCTL to the character device
05:41
pratyush
i have created a character driver in which the user creates a driver and then a memory file using mknod command
05:42
Bertl_oO
what is a 'memory file'?
05:43
pratyush
the file which represents our driver
05:44
pratyush
i creat it using :mknod chrdev c 239 0
05:45
Bertl_oO
so you are referring to the character special file entry
05:45
pratyush
yes
05:45
Bertl_oO
okay
05:46
Bertl_oO
please go on
05:46
pratyush
So the user gives input by storing the data in a character buffer array which is written to character special file entry
05:47
pratyush
the data written can also be viewed by cat chrdev
05:47
Bertl_oO
like for example 'echo "hello character device" >/dev/char-dev-special
05:47
pratyush
yes
05:48
Bertl_oO
okay, the read back is a bonus, no problem there
05:50
pratyush
I am facing confusion regarding ioctl
05:50
Bertl_oO
please elaborate
05:54
pratyush
I have not worked with Ioctl before and I want to know which particular command would be useful for creating the checksum entry
05:55
Bertl_oO
as this is a challenge task, you can pick any IOCTL you consider appropriate and which allows you to communicate the checksum to userspace
05:57
pratyush
could we create multiple instances of character special file entry: the first one woul store the input data and the second one would store the checksum
05:58
Bertl_oO
while possible, that's not the task at hand
05:58
pratyush
okay
05:59
pratyush
I will update my github repo and try to implement the task using procfs by tonight(11 pm IST)
06:00
pratyush
I have also done some major modifications and will also share them
06:01
pratyush
*tonight not today night
06:03
Bertl_oO
great! just drop me a message here or send an email when I should check it
06:03
pratyush
okay
06:08
pratyush
left the channel
06:50
pratyush
joined the channel
06:53
pratyush
left the channel
07:29
BAndiT1983|away
changed nick to: BAndiT1983
07:33
BAndiT1983
hi metal_dent[m], yes, please update your repo, my changes are not final yet, but i've decided to commit nevertheless, as there were some base changes here and there, if you find problems then just ping me about them, will try to fix as soon as possible
08:47
metal_dent[m]
okay!
08:55
BAndiT1983
changed nick to: BAndiT1983|away
09:00
Bertl_oO
off to bed now ... have a good one everyone!
09:00
Bertl_oO
changed nick to: Bertl_zZ
09:00
apurvanandan[m]
Good night!
09:09
metal_dent[m]
BAndiT1983|away: I updated but the visualizer looks strange, many things are not in place
09:29
ATyagi_
left the channel
10:30
BAndiT1983|away
changed nick to: BAndiT1983
10:31
BAndiT1983
metal_dent[m]: that is expected, at the moment some things are not how they should be, but that is because of changes i have to make to straighten them out
10:31
BAndiT1983
but it should not affect DrawIcon() at all
10:32
metal_dent[m]
yes yes, that is okay
10:57
metal_dent[m]
BAndiT1983: I added a test for DrawIcon(), I didn't have any experience with unit tests so I referred to other tests, hope that's okay :)
11:01
BAndiT1983
how do you test correct output?
11:03
metal_dent[m]
i tried to draw another dummy image using DrawIcon()
11:04
BAndiT1983
i see that you also'Ve used wrongDirection, but that is not the goal of the test, as i epxect the test to draw correctly, drawing direction was added especially for old drawing routines which have drawn backwards and dragged performance down, also it waas rather counter-intuitive to implement back then
11:04
BAndiT1983
you should check against result
11:04
BAndiT1983
like input data, process it, check output data
12:04
HEGAZY
joined the channel
12:18
ashok_singh[m]
Hi BAndiT1983, I have pushed and squashed new changes, please review.
12:19
ashok_singh[m]
There is some changes in UI is it because of origin shift to top left in drawpixel?
12:20
ashok_singh[m]
*are
12:22
ashok_singh[m]
I have not added test in present commit because test build is failing at the moment. Once the upstream fixes I will check and push.
12:27
BAndiT1983
ashok_singh[m]: this was already discussed several times in the last days, that the origin at the top-left is the proper usage of display and MCU memory
12:28
BAndiT1983
before that it was at bottom-left and drawing backwards, which degrades performance because of cache misses and other hardware-related stuff
13:09
BAndiT1983
i won't have time to fix it right now, as i got sick and try to restore before doing big development again, so i would suggest that you push your code
13:44
se6ast1an
ashok_singh[m]: please also take a look at: https://lab.apertus.org/T1164
13:45
se6ast1an
with the new comments this should be trivial to fix I assume, please give it a shot to verify
14:02
HEGAZY
left the channel
14:02
pranabendra
joined the channel
14:06
pranabendra
Hello, BAndiT1983: For T1162, I added DrawIcon() with front color and background color parameters along with one test which I could think of. The test checks whether the icon is completely inside the 320x240 LCD
14:07
pranabendra
I pushed them here https://github.com/pranabendra/AXIOM-Remote/tree/dev
14:07
BAndiT1983
why are you testing the boundary, it is already covered by another method, wouldn't it be better to test input and output?
14:07
BAndiT1983
i mean given input data and expected output
14:09
BAndiT1983
reviewed and commented on github
14:11
pranabendra
By expected output, do you mean whether each DrawPixel() is being successfully executed or not?
14:11
pranabendra
Thank you!
14:11
BAndiT1983
nope, i mean which output is expected in the framebuffer
14:12
BAndiT1983
and framebuffer isn't fixed at 320x240, as it was done just for the convenience of the display specs, but size can be varied for tests also
14:27
pranabendra
Won't 1-bit depth data contain only two values (black and white)?
14:30
BAndiT1983
yes
14:30
BAndiT1983
how many does your contain per byte?
14:33
pranabendra
0 and 255
14:35
BAndiT1983
this doesn't sound like 1-bit depth
14:35
BAndiT1983
goal is to have 8 pixels encoded in one byte
14:42
Bertl_zZ
changed nick to: Bertl
14:42
Bertl
morning folks!
14:42
BAndiT1983
hi
14:56
pranabendra
I understood the 1-bit depth now, BAndiT1983: ... will work upon it. Thanks
15:04
Bertl
better work _on_ it :)
15:05
BAndiT1983
as brexit is already here, do we have to switch from british english to another variant?
15:10
HEGAZY
joined the channel
15:18
pranabendra
left the channel
15:18
Bertl
BAndiT1983: yes, British English is not allowed in the EU anymore
15:19
BAndiT1983
and what do we do with all the people infected by it?
15:19
Bertl
quarantine?
15:56
RexOrMatrix[m]
Yev tey'taak brockens, lest'ee be med dweeb-like in al yer gan's oen.
15:56
RexOrMatrix[m]
Plus like VW and Nissan manufacturing plants the project has to move to Britain.
16:03
Bertl
do they?
16:09
comradekingu
left the channel
16:10
RexOrMatrix[m]
Nissan and VW? Nissan has, VW is thinking about it.
16:13
RexOrMatrix[m]
It's about decline in demand on the continent, opening markets and the potential to save hundreds of millions, so... Looks likely.
16:14
BAndiT1983
doubt that VW will move that easily, as the supply chains are tied to germany
16:15
BAndiT1983
also the german market is not to be to underestimate,a s the country still has no speed limits on many highway sections
16:15
BAndiT1983
so customers will react accordingly
17:48
HEGAZY
left the channel
18:31
HEGAZY
joined the channel
19:19
Bertl
off for now ... bbl
19:19
Bertl
changed nick to: Bertl_oO
19:22
BAndiT1983
changed nick to: BAndiT1983|away
19:25
BAndiT1983|away
changed nick to: BAndiT1983
19:27
BAndiT1983
changed nick to: BAndiT1983|away
19:37
ashok_singh[m]
BAndiT1983: thank you for the explanation. I will push changes.
19:38
ashok_singh[m]
> ashok_singh: please also take a look at: https://lab.apertus.org/T1164
19:38
ashok_singh[m]
se6ast1an I am working on it.
19:42
BAndiT1983|away
changed nick to: BAndiT1983
19:52
se6ast1an
great
20:26
omar31
joined the channel
23:07
BAndiT1983
changed nick to: BAndiT1983|away
23:10
illwieckz
left the channel
23:11
aombk2
left the channel
23:12
aombk2
joined the channel
23:26
illwieckz
joined the channel
23:50
HEGAZY
left the channel