Posts
1758
Following
90
Followers
166
Riding horses, hacking computers, phones and smartwatch.
pavel

Pavel Machek

Reply to @datenwolf@chaos.social
@datenwolf I believe I'm limited to OpenGL ES 2.0. Presumably hardware can do more but our current drivers can not, so we are stuck there.

On the other hand... #librem5 main sensor can not do 10bpp at the moment, due to missing drivers. So maybe we can focus on 8bpp, first. Probably ineffecient conversion is "good enough" too, as GPU is a bit overpowered for this job.
2
0
0
pavel

Pavel Machek

Reply to @datenwolf@chaos.social
@datenwolf So linearization should be doable with functions, too. Small trouble is that it is sensor-dependent but ... we have enough mathematical tools to deal with that. Actual functions can be seen here: https://blog.brixit.nl/fixing-the-megapixels-sensor-linearization/
0
0
0
pavel

Pavel Machek

Reply to @dos@librem.one
Yep, recent bwtest shows that (extremely simple) debayer is feasible, and possibly more. So far I integrated debayer-gpu + gstreamer, and I'm meeting the deadlines.
0
0
0
pavel

Pavel Machek

Reply to @dos@librem.one
@dos Sorry, hero, that's dark magic behind my understanding. I see the words but don't understand the sentences. :-(

I'd need working example here. I got surprisingly far vibecoding this, but even robots have their limits.
1
0
0
pavel

Pavel Machek

I, once again, believe that GPU on Librem 5 can be useful for debayering. #librem5 #linuxphone https://gitlab.com/tui/debayer-gpu/-/tree/master/bwtest?ref_type=heads
0
0
1
pavel

Pavel Machek

Reply to @dos@librem.one
@dos If you have example of that, that would be welcome :-). That's not how megapixels work, at least.
1
0
0
pavel

Pavel Machek

Reply to @mkyral@mastodonczech.cz
@mkyral @Creep73 Jo jo, protoze chyba "kernel muze havarovat kdyz mu dojde pamet ktera dojit nemuze" je stejne zavazna jako "remote root diky chybe v parsovani IP". To dava smysl!
0
0
1
pavel

Pavel Machek

Reply to @mkyral@mastodonczech.cz
@mkyral @Creep73 A tak jasne, security je uzasna vymluva proc nekoho sikanovat.

Z duvodu bezpecnosti musite (dosadte zde). Napr. kdyz vas chcem spehovat, tak z duvodu bezpecnosti musite mit nasi aplikaci.

Mimochodem, vedeli jste ze CVE na kernel jsou generovane copy&paste a v podstate je to spam? :-(
1
0
0
pavel

Pavel Machek

Reply to @mkyral@mastodonczech.cz
@mkyral @Creep73 O "security" se mluvi. Je potreba ji porad zduraznovat... ale doopravdy se bezpecnost moc nedela. Takze mluvi o bezpecnosti, a pak pouzivaji Windows. Plna huba bezpecnosti, a pak posilaji cast mailu podepsanou, a cast ne. (mbank napr.)
1
0
0
pavel

Pavel Machek

Reply to @dos@librem.one
@dos It seems that ignoring half of green pixels is right thing to do at the moment: https://gitlab.com/tui/debayer-gpu/-/tree/master/bwtest?ref_type=heads "Normal" debayer is 40% too slow. (That's better than 5 times too slow, but still not good enough). If you can can get it to 24 loops in second, you'll become a hero :-).
1
0
0
pavel

Pavel Machek

Reply to @robertfoss@mastodon.social
@robertfoss I know about that one, see gitlab.com:tui/debayer-gpu.git . I could not get that to anywhere near the required performance.
0
0
0
pavel

Pavel Machek

Reply to @dos@librem.one
@dos As for copies... Yes, I'm currently doing more copies than needed. I measured Librem 5 at about 2GB/sec memory bandwidth, and stream is about 30MB/sec. At 1Mpix/24fps resolution, gstreamer should be able to encode it in real time.

Here's huge problem with v4l, which gives uncached memory buffers to userspace. That means one whole CPU core is dedicated to copying that to "normal" memory. If that is ever solved, yes, other optimalizations are possible. Currently, this means it is not even possible to copy anything bigger than 1Mpix out of the v4l.
1
0
0
pavel

Pavel Machek

Reply to @datenwolf@chaos.social
@datenwolf This is probably easiest to use (after pngtopnm): https://gitlab.com/tui/debayer-gpu/-/blob/master/test.png?ref_type=heads .
0
0
0
pavel

Pavel Machek

Reply to @datenwolf@chaos.social
@datenwolf Sorry. Example frame is here: https://gitlab.com/tui/debayer-gpu/-/blob/master/test.png?ref_type=heads (You probably want to run pngtopnm it, so that your code only works with uncompressed data).

Alternatively, I started file format for this. https://gitlab.com/tui/tui/-/tree/master/4cc?ref_type=heads dirgen.sh can generate example frames using gstreamer. You get raw data after 128 bytes header.
0
0
0
pavel

Pavel Machek

Reply to @datenwolf@chaos.social
@datenwolf Example of frame is here: https://gitlab.com/tui/tui/-/blob/master/ucam/bayer2rgb.rs?ref_type=heads (I also have frame generator and real frames captured from libobscura).

Anything that works on Librem 5 is fine, bonus points if I can understand it. Robot generated code using -lEGL -lGLESv2 -lm ... and that builds and does something. Librem 5 reports:

Vendor: etnaviv
Renderer: Vivante GC7000 rev 6214
OpenGL Version: OpenGL ES 2.0 Mesa 21.2.6
GLSL Version: OpenGL ES GLSL ES 1.0.16
2
0
0
pavel

Pavel Machek

Reply to @dos@librem.one
@dos @tizilogic I know. And I have "the rest" prototyped here: https://gitlab.com/tui/debayer-gpu/-/blob/master/isp.frg?ref_type=heads But I feel I need fast-enough naive debayering first, so that I can improve that.
0
0
0
pavel

Pavel Machek

Reply to @tizilogic@mastodon.gamedev.place
@tizilogic @dos I tried simd, https://gitlab.com/tui/tui/-/blob/master/ucam/bayer2rgb.rs?ref_type=heads , it did not have good enough performance. (I could not do 512x384 at 23fps).

GL versions are:
Vendor: etnaviv
Renderer: Vivante GC7000 rev 6214
OpenGL Version: OpenGL ES 2.0 Mesa 21.2.6
GLSL Version: OpenGL ES GLSL ES 1.0.16

Doing input in u8, with output in u8 and internal computation in u16 fixed point should be "good enough". Doing everything in u16 would be even better. Floats are okay, too.
0
0
0
pavel

Pavel Machek

Reply to @dos@librem.one
@dos That's problem for future Pavel :-). Right now, I'm storing frames on ramdisk, as "RGB3" basically.
0
0
0
pavel

Pavel Machek

Reply to @dos@librem.one
@dos Lets keep the example simple :-). Yes, g = G1+G2/2 is superior, and there are advanced debayer algorithms. I know them. Examples are at https://gitlab.com/tui/debayer-gpu/ . There's just one small problem: It takes minute and I need it to take 10 seconds.
0
0
0
pavel

Pavel Machek

Edited 21 days ago
edit: Thanks for all the heroes that chimed in. In meantime, I got help from entity that shall not be named, and currently have something fast enough. And I have great human experts on line, with patches to test. Thanks again! :-)

Can you program GPUs and do you want to become a HERO? #linuxphone
community needs your help.

We are trying record video, and have most pieces working, but one is
missing: fast enough debayering. That means about 23MB/sec on #librem5.

Debayering is not hard; camera images have subpixels split on two
lines, which need to be corrected. They also use different color
representation, but that's fixable by some table lookup and two matrix
multiplies.

Librem 5 has Vivante GPU, 4 in-order CPU cores and 3GB RAM. My feeling
is that it should be fast enough for that. If task is for some reason
impossible, that would be good to know, too.

Image data looks like this

RGRGRG...
xBxBxB...
.........
.........

Task is to turn that into usual rgbrgb.... format. rgb = RGB * color
matrix, with table lookups for better quality. I can fix that once I
get an example.

I'm looking for example code (#pinephone would work, too), reasons it
can not be done... and boosts if you have friends that can program
GPUs. #gpu #opensource
4
38
17
Show older
About social.kernel.org

Terms of service

  1. Please do not use this service in violation of the Linux Kernel Code of Conduct. Doing so will result in your account suspension with the referral of the matter to the CoC committee.

  2. "Repeating"/"boosting" someone else's status on this platform will be treated as endorsement and will fall under rule #1.

  3. You are encouraged to use this platform to promote your work on the Linux Kernel, but there is no restriction on permitted topics (with the exception of anything covered by #1 above).

  4. There is no requirement to post in English, but it should be considered the primary language of communication on this platform.

Privacy notice

The admins of this service have access to all posted statuses. They aren't looking, but if it's something they shouldn't know about, then you should not post it on this platform.

Please see the Linux Foundation Privacy Policy, which applies to this platform as well.

Getting your own account

If you would like an account on this instance, please check that the following applies to you:

  • You are listed in MAINTAINERS or CREDITS

  • OR: You have a kernel.org account or email address

  • OR: You have a long and established history of involvement with the Linux Kernel

If the above is true and you agree with the Terms of Service and Privacy Notice listed above, please use these instructions to request an account: