libav version in About prefs [patch]
So, few more lines showing our libav version (there is no differentiation between patched and unpatched ffmpeg yet, but at least you can see if build was done against specific API version)
mmm maybe even like second patch? (adds printing libav version to our startup banner) Another patch tries to add cinelerra-cv team copyright, I hope I did layout and dates themselves correctly! вс, 27 авг. 2023 г., 15:28 Andrew Randrianasulu <[email protected]>:
So, few more lines showing our libav version (there is no differentiation between patched and unpatched ffmpeg yet, but at least you can see if build was done against specific API version)
And forth patch adds Cinelerra-gg team woth copyright 2015-2023. вс, 27 авг. 2023 г., 17:18 Andrew Randrianasulu <[email protected]>:
mmm maybe even like second patch? (adds printing libav version to our startup banner)
Another patch tries to add cinelerra-cv team copyright, I hope I did layout and dates themselves correctly!
вс, 27 авг. 2023 г., 15:28 Andrew Randrianasulu <[email protected]>:
So, few more lines showing our libav version (there is no differentiation between patched and unpatched ffmpeg yet, but at least you can see if build was done against specific API version)
I tried to compile CinGG with the last 4 patches. I have an error in the compilation that I don't understand, but I don't think it is related to the patches. I attach cin5.log
apparently binutils 2.41 changed something. https://github.com/MythTV/mythtv/commit/9572dc5a124b46ac60c6cea34994ccc69ec5... so we need ffmpeg patch mentioned there... effadce6c756247ea8bae32dc13bb3e6f464f0eb . Author: Rémi Denis-Courmont <[email protected]> Date: Sun Jul 16 18:18:02 2023 +0300 avcodec/x86/mathops: clip constants used with shift instructions within inline assembly Fixes assembling with binutil as >= 2.41 can you try building with ffmpeg.git ? optionally we can try to add this patch to our tree https://git.ffmpeg.org/gitweb/ffmpeg.git/blobdiff_plain/ccc684993276248d64c3... вс, 27 авг. 2023 г., 19:08 Andrea paz <[email protected]>:
I tried to compile CinGG with the last 4 patches. I have an error in the compilation that I don't understand, but I don't think it is related to the patches. I attach cin5.log
On Sun, Aug 27, 2023 at 7:40 PM Andrea paz <[email protected]> wrote:
can you try building with ffmpeg.git ? Sorry to ask, but could you make me a script to compile ffmpeg.git? I have done it before, but I don't remember how anymore.
./configure --with-single-user --with-git-ffmpeg=https://git.ffmpeg.org/ffmpeg.git at least it was working day ago ...
Still a compilation error; although it doesn't look the same as before. I attach the cin5.log
On Sun, Aug 27, 2023 at 7:58 PM Andrea paz <[email protected]> wrote:
Still a compilation error; although it doesn't look the same as before. I attach the cin5.log
you still have this 0001-Fix-build-in-pluginfclient.C-with-ffmpeg-6.0.patch applied, right?
you still have this 0001-Fix-build-in-pluginfclient.C-with-ffmpeg-6.0.patch applied, right?
Ehm, No, I don't. Where do I find it again? Plus another question, in configure, don't I also need to use --without-thirdparty and disable some libraries? ( and maybe disable static-build as well)?
On Sun, Aug 27, 2023 at 8:06 PM Andrea paz <[email protected]> wrote:
you still have this 0001-Fix-build-in-pluginfclient.C-with-ffmpeg-6.0.patch applied, right?
Ehm, No, I don't. Where do I find it again?
attached. May be we really should add it to our git even if 6.0 not yet integrated (need to test on ffmpeg 5.1, specifically audio filters from ffmpeg)
Plus another question, in configure, don't I also need to use --without-thirdparty and disable some libraries? ( and maybe disable static-build as well)?
if building with ffmpeg.git ? No, unless some libs fails to build in this configuration ....
Now the compilation takes place without errors. However, I don't have the cin file inside bin: it is not possible to start CinGG. CIN_CONFIG=/home/paz/.bcast6 /home/paz/cinelerra5/cinelerra-5.1/bin/./cin bash: /home/paz/cinelerra5/cinelerra-5.1/bin/./cin: No such file or directory I attach yet another cin5.log
Testing libav and 001-004 "About" related patches for libav version and copyright notes on ffmpeg 5.1 (Not 6.0 like Andrea is doing). So far, so good. Why is the libav version so important to print out? isn't it only pertinent to compiler/builders? On Sun, Aug 27, 2023 at 8:54 AM Andrew Randrianasulu < [email protected]> wrote:
And forth patch adds Cinelerra-gg team woth copyright 2015-2023.
вс, 27 авг. 2023 г., 17:18 Andrew Randrianasulu <[email protected]>:
mmm maybe even like second patch? (adds printing libav version to our startup banner)
Another patch tries to add cinelerra-cv team copyright, I hope I did layout and dates themselves correctly!
вс, 27 авг. 2023 г., 15:28 Andrew Randrianasulu <[email protected]
:
So, few more lines showing our libav version (there is no differentiation between patched and unpatched ffmpeg yet, but at least you can see if build was done against specific API version)
вс, 27 авг. 2023 г., 22:39 Phyllis Smith <[email protected]>:
Testing libav and 001-004 "About" related patches for libav version and copyright notes on ffmpeg 5.1 (Not 6.0 like Andrea is doing). So far, so good.
Why is the libav version so important to print out? isn't it only pertinent to compiler/builders?
Um, I want to have reminder about what kind of binary I run ... (so speed/regressions can be spotted)
On Sun, Aug 27, 2023 at 8:54 AM Andrew Randrianasulu < [email protected]> wrote:
And forth patch adds Cinelerra-gg team woth copyright 2015-2023.
вс, 27 авг. 2023 г., 17:18 Andrew Randrianasulu <[email protected]
:
mmm maybe even like second patch? (adds printing libav version to our startup banner)
Another patch tries to add cinelerra-cv team copyright, I hope I did layout and dates themselves correctly!
вс, 27 авг. 2023 г., 15:28 Andrew Randrianasulu <[email protected]
:
So, few more lines showing our libav version (there is no differentiation between patched and unpatched ffmpeg yet, but at least you can see if build was done against specific API version)
With the new patch everything is fine. I'm not sure how to test the patch. I did a background render in jpg and everything is OK. I did a background render in tiff and got a crash when I tried to move the cursor. I attach the dump. I restarted and tried the background render in tiff again and everything worked fine and no crashes. Then I tried a render first as a sequence of jpeg images and then as a sequence of tiff images and everything is OK. I would say the dump was an exception (I probably mixed jpg and tiff in the same folder), but the patch works fine.
On Sun, Aug 27, 2023 at 11:07 PM Andrea paz <[email protected]> wrote:
With the new patch everything is fine. I'm not sure how to test the patch. I did a background render in jpg and everything is OK. I did a background render in tiff and got a crash when I tried to move the cursor. I attach the dump.
like this? bash-5.1$ bin/cin Cinelerra Infinity - built: Aug 27 2023 20:00:50 git://git.cinelerra-gg.org/goodguy/cinelerra.git (c) 2006-2019 Heroine Virtual Ltd. by Adam Williams 2007-2020 mods for Cinelerra-GG by W.P.Morrow aka goodguy 2003-2017 mods for Cinelerra-CV by CinelerraCV team 2015-2023 mods for Cinelerra-GG by Cinelerra-GG team Libav version: Lavc59.37.100 Cinelerra is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. There is absolutely no warranty for Cinelerra. RenderFarmClient::main_loop: client started BRenderThread::start 1 map=0 equivalent=0 brender_start=0 result=0 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 virtual int FileList::open_file(int, int): Error while opening "/dev/shm//tmp/brender000000" for writing. No such file or directory RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=0 equivalent=0 brender_start=0 result=0 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=653 equivalent=653 brender_start=0 result=653 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=653 equivalent=653 brender_start=0 result=653 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=653 equivalent=653 brender_start=0 result=653 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=653 equivalent=0 brender_start=0 result=0 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=1 equivalent=0 brender_start=0 result=0 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=1 equivalent=0 brender_start=0 result=0 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 BRenderThread::start 1 map=1 equivalent=0 brender_start=0 result=0 end=653 RenderFarmClientThread::run: Session finished. RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=1 equivalent=0 brender_start=0 result=0 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=67 equivalent=0 brender_start=0 result=0 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=1 equivalent=0 brender_start=0 result=0 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=1 equivalent=0 brender_start=0 result=0 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=78 equivalent=0 brender_start=0 result=0 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=1 equivalent=0 brender_start=0 result=0 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=1 equivalent=0 brender_start=0 result=0 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=1 equivalent=0 brender_start=0 result=0 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=653 equivalent=653 brender_start=0 result=653 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. BRenderThread::start 1 map=653 equivalent=653 brender_start=0 result=653 end=653 RenderFarmClient::main_loop: Session started from /tmp/cinelerra.64b245a1-8873-4e3a-88ec-8d0c47e481f9 RenderFarmClientThread::run: Session finished. FileTIFF: Not a TIFF or MDI file, bad magic number 55551 (0xd8ff). ** segv at 0x9e92bbf in pid 3376, tid 5139 writing debug data to /tmp/cinelerra_3376.dmp lock_items: 23 lock_frees: 12 ** dump complete Ошибка сегментирования it seems libtiff file reader does not check error codes on failure ... guess we need to harden it. Can you enter bug in Mants BT?
I restarted and tried the background render in tiff again and everything worked fine and no crashes. Then I tried a render first as a sequence of jpeg images and then as a sequence of tiff images and everything is OK. I would say the dump was an exception (I probably mixed jpg and tiff in the same folder), but the patch works fine.
In my opinion, it depends on putting the images in /tmp. When I changed the destination folder (/home/test) I had no more problems. Try setting up CinGG with a different brender destination folder; one that is not with root privileges. PS: I had not responded to this request of yours:
Are those errors with my patch and normal single-user build, or they only appear in appimage (with my patch?) All those warnings are from the build with your patches. But the vaapi decoder works fine.
The vaapi decoder failure errors only affect the appimage (from 2 months ago --> no patch). Those numerous warnings are not present here, though.
On Sun, Aug 27, 2023 at 11:29 PM Andrea paz <[email protected]> wrote:
In my opinion, it depends on putting the images in /tmp. When I changed the destination folder (/home/test) I had no more problems. Try setting up CinGG with a different brender destination folder; one that is not with root privileges.
I tried to runtime sitch brender type while using same /tmp (/dev/shm/tmp) dir. I got few crashes. Tried to fix two cases (tiff and png), filetga still crashes. Can you check this patch with normal file load and also bg renders?
PS: I had not responded to this request of yours:
Are those errors with my patch and normal single-user build, or they only appear in appimage (with my patch?) All those warnings are from the build with your patches. But the vaapi decoder works fine.
The vaapi decoder failure errors only affect the appimage (from 2 months ago --> no patch). Those numerous warnings are not present here, though.
Hopefully filetga check for invalid file (header bpp will be random) On Sun, Aug 27, 2023 at 11:33 PM Andrew Randrianasulu <[email protected]> wrote:
On Sun, Aug 27, 2023 at 11:29 PM Andrea paz <[email protected]> wrote:
In my opinion, it depends on putting the images in /tmp. When I changed the destination folder (/home/test) I had no more problems. Try setting up CinGG with a different brender destination folder; one that is not with root privileges.
I tried to runtime sitch brender type while using same /tmp (/dev/shm/tmp) dir.
I got few crashes.
Tried to fix two cases (tiff and png), filetga still crashes.
Can you check this patch with normal file load and also bg renders?
PS: I had not responded to this request of yours:
Are those errors with my patch and normal single-user build, or they only appear in appimage (with my patch?) All those warnings are from the build with your patches. But the vaapi decoder works fine.
The vaapi decoder failure errors only affect the appimage (from 2 months ago --> no patch). Those numerous warnings are not present here, though.
I tried loading and BRender adding png, tiff, jpg and tga images. I found several errors and I attach a long series of messages I found on the terminal. I will try the new patch tomorrow, now it is time to sleep....
пн, 28 авг. 2023 г., 00:21 Andrea paz <[email protected]>:
I tried loading and BRender adding png, tiff, jpg and tga images. I found several errors and I attach a long series of messages I found on the terminal.
I think for checking native file readers you should put them up in filetypes, above ffmpeg (or switch ffmpeg from early to late by main window button). Those undo messages looks like something trigger Dumps.../Undo menu item. May be some automatic keypress tool, like xdotool? I will try the new patch tomorrow, now it is time to sleep....
I put in the latest patch and, it seems to me, everything is fine. I did the following tests: A) appimage (without patches): 1- with "try ffmpeg first" 2- with "try ffmpeg last" I append tiff_appimage.txt B) build + patches + ffmpeg.git 1- with "try ffmpeg last" 2- with "try ffmpeg first" I append tiff_build_patch.txt My conclusion: there are far fewer messages in the terminal and there are not the errors of yesterday without the latest patch. It seems to me that you have done a very good job....
participants (3)
-
Andrea paz -
Andrew Randrianasulu -
Phyllis Smith