SOLO6x10 MPEG4/H.264 encoder driver
Hi Hans,
I'm trying to move to Linux 3.11 and I noticed you've made some significant changes to the SOLO6x10 driver. While I don't yet have the big picture, I can see some regressions here:
- the driver doesn't even try to work on big endian systems (I'm using IXP4xx-based system which is BE). For instance, you're now using bitfields for frame headers (struct vop_header) and this is well known to fail (unless you have a different struct for each endianness).
This is actually what I have fixed with commit c55564fdf793797dd2740a67c35a2cedc133c9ff in 2011, and you brought the old buggy version back with dcae5dacbce518513abf7776cb450b7bd95d722b.
- you removed my dynamic building of MPEG4/H.264 VOP headers (the same commit c55564fdf793797dd2740a67c35a2cedc133c9ff) and replaced it with precomputed static binary headers, one for each PAL/NTSC/D1/CIF combination. While I don't strictly object the precomputed data, perhaps you could consider adding some tool to optionally calculate them, as required by the license. For now, It seems it's practically impossible to make modifications to the header data, without, for example, extracting the code from older driver version.
- what was the motivation behind renaming all (C language) files in drivers/staging/media/solo6x10 to solo6x10-* (commits dad7fab933622ee67774a9219d5c18040d97a5e5 and 7bce33daeaca26a3ea3f6099fdfe4e11ea46cac6, essentially a reversion of my commit ae69b22c6ca7d1d7fdccf0b664dafbc777099abe)? I'm under impression that a driver file names don't need (and shouldn't) contain the driver name if the directory is already named after the driver.
This is also the case with b3c7d453a00b7dadc2a7435f68b012371ccc3a3e:
Quoted 4 lines
[media] solo6x10: rename
jpeg.htosolo6x10-jpeg.hThis header clashes with the
jpeg.hin gspca when doing a compatibility build using themedia_buildsystem.
What is this media_build system and why is it forcing code in different directories to have unique file names?
I appreciate the switch to VB2 and other improvements (though I can't test them yet), but perhaps it could be done without causing major breakage?
I'm thinking about a correct course of action now. I need the driver functional so I'll revert the struct vop_header thing again, any thoughts?
Signature · 4 lines
Krzysztof Halasa Research Institute for Automation and Measurements PIAP Al. Jerozolimskie 202, 02-486 Warsaw, Poland
Hi Krzysztof,
I've CC-ed Ismael, the solo maintainer.
Quoted 5 lines · Krzysztof Hałasa
Hi Hans,
I'm trying to move to Linux 3.11 and I noticed you've made some significant changes to the SOLO6x10 driver. While I don't yet have the big picture, I can see some regressions here:
The big picture behind all the changes is a push to get this driver out of staging. It turned out that the driver in the kernel had diverged substantially from the driver maintainer internally by bluecherry, and the work I did merged those two forks as best as I could, cleaned up the code and ensured that all the latest internal frameworks were used.
The only thing preventing the driver from being moved out of staging is the motion detection functionality. I've been working on that, but it didn't make 3.12.
Quoted 9 lines
- the driver doesn't even try to work on big endian systems (I'm using IXP4xx-based system which is BE). For instance, you're now using bitfields for frame headers (struct
vop_header) and this is well known to fail (unless you have a different struct for each endianness).This is actually what I have fixed with commit
c55564fdf793797dd2740a67c35a2cedc133c9ffin 2011, and you brought the old buggy version back withdcae5dacbce518513abf7776cb450b7bd95d722b.
The problem was that the bluecherry driver had many changes as well in the same area, and that I could not test it on a big endian system.
I hadn't realized the problems this would cause on big endian systems, if I had I would have attempted to merge your changes.
Quoted 8 lines
- you removed my dynamic building of MPEG4/H.264 VOP headers (the same commit
c55564fdf793797dd2740a67c35a2cedc133c9ff) and replaced it with precomputed static binary headers, one for each PAL/NTSC/D1/CIF combination. While I don't strictly object the precomputed data, perhaps you could consider adding some tool to optionally calculate them, as required by the license. For now, It seems it's practically impossible to make modifications to the header data, without, for example, extracting the code from older driver version.
I took the latest bluecherry code as the basis for the changes, merging what I could from the kernel code. Unfortunately this was very hard to do backport, so I decided to take bluecherry's code.
Quoted 7 lines
- what was the motivation behind renaming all (C language) files in
drivers/staging/media/solo6x10to solo6x10-* (commitsdad7fab933622ee67774a9219d5c18040d97a5e5and7bce33daeaca26a3ea3f6099fdfe4e11ea46cac6, essentially a reversion of my commitae69b22c6ca7d1d7fdccf0b664dafbc777099abe)? I'm under impression that a driver file names don't need (and shouldn't) contain the driver name if the directory is already named after the driver.
Two reasons: first of all this is the convention within drivers/media, secondly the backwards compatibility build system that allows people to compile the latest media code for older kernels requires unique filenames.
Quoted 9 lines
This is also the case with
b3c7d453a00b7dadc2a7435f68b012371ccc3a3e:Quoted 4 lines
[media] solo6x10: rename
jpeg.htosolo6x10-jpeg.hThis header clashes with the
jpeg.hin gspca when doing a compatibility build using themedia_buildsystem.What is this
media_buildsystem and why is it forcing code in different directories to have unique file names?
See http://git.linuxtv.org/media_build.git and here:
http://linuxtv.org/wiki/index.php/How_to_Obtain,_Build_and_Install_V4L-DVB_Device_Drivers
Quoted 5 lines
I appreciate the switch to VB2 and other improvements (though I can't test them yet), but perhaps it could be done without causing major breakage?
It's a staging driver, so by definition unstable. That said, I do apologize for missing the big-endian problems.
Quoted 3 lines
I'm thinking about a correct course of action now. I need the driver functional so I'll revert the struct
vop_headerthing again, any thoughts?
That sounds reasonable. Please post it to the mailinglist so it can be reviewed by Ismael and I'd be happy to take it in.
Regards,
Hans
Quoted 3 lines · Hans Verkuil
I took the latest bluecherry code as the basis for the changes, merging what I could from the kernel code. Unfortunately this was very hard to do backport, so I decided to take bluecherry's code.
I see, thanks for speedy explanation.
If I may suggest something (especially to Ismael), perhaps we do the further development here, I mean based on git.kernel.org sources, and not on (unsynced) bluecherry's.
I will fix this stuff again.
Signature · 4 lines
Krzysztof Halasa Research Institute for Automation and Measurements PIAP Al. Jerozolimskie 202, 02-486 Warsaw, Poland
Quoted 11 lines · On Mon 9 September 2013 13:49:53 Krzysztof Hałasa wrote:
Quoted 3 lines · Hans Verkuil
I took the latest bluecherry code as the basis for the changes, merging what I could from the kernel code. Unfortunately this was very hard to do backport, so I decided to take bluecherry's code.
I see, thanks for speedy explanation.
If I may suggest something (especially to Ismael), perhaps we do the further development here, I mean based on git.kernel.org sources, and not on (unsynced) bluecherry's.
That's the whole point of the exercise. Hopefully everything will stay in sync from now on.
Quoted 1 line
I will fix this stuff again.
Much appreciated! Sorry for the inconvenience.
Regards,
Hans