- Cin - Cinelerra-GG Mailing Lists

Fwd: Sparc32 "support" for cinelerra-gg
by Andrew Randrianasulu 21 Dec '25

21 Dec '25

---------- Forwarded message --------- От: Andrew Randrianasulu <randrianasulu(a)gmail.com> Date: пт, 19 дек. 2025 г., 19:36 Subject: Sparc32 "support" To: Cinelerra.GG <cin(a)lists.cinelerra-gg.org>, Phyllis Smith < phylsmith2017(a)gmail.com> Not really, but it compiled to completion, and shwed startup messages and died in GL library call (inside GL library, mesa) :) So, trying to rebuild on netBSD/sparc with opengl disabled Not even diffs, just bootstrap.c and Makefile from guicast folder. === Attached are git diffs against our git at https://git.cinelerra-gg.org/goodguy/cinelerra.git git commit 17fa80d7d099013c4ae4c8fb4ba5e5f359bfad31 please test them on "normal" targets like x86/arm. I only tested that cinelerra-gg builds in dynamic configuration on aarch64/termux and starts normally there, displaying small ogg/theora file. For NetBSD there is script in blds that should build cingg if you manually install needed packages (see Linux build files for names of deps, or pkgsrc-wip/cingg-netbsd . Need to update it and in particular PLIST there for newer files added recently ...) ==== Need more ram on compile machine, 256 M too small with 32 Mb of swap for affine.c :)

1 0

Re: Cinelerra-gg starts on NetBSD/sparc32 in qemu!
by Andrew Randrianasulu 21 Dec '25

21 Dec '25

On Sun, Dec 21, 2025 at 1:26 PM Andrew Randrianasulu <randrianasulu(a)gmail.com> wrote: > > On Sun, Dec 21, 2025 at 1:20 PM David Brownlee <abs(a)netbsd.org> wrote: > > > > On Sun, 21 Dec 2025 at 08:02, Andrew Randrianasulu > > <randrianasulu(a)gmail.com> wrote: > > > > > > https://mastodon.online/@Andrew_R/115756500366360503 - 8bit TCX, -smp > > > 3 as qemu parameter. > > > > Ooof, that brings back all the flipping colourmap joy I remember from > > cg6 based machines :) > > > > > Sadly, I see some Conversion from UTF-8 to UTF32BE is not supported > > > (from freetype?) so no text at all :) :( > > > > Does utf-32be work? > > This is not my locale setting, i tried to set it just to C (LOCALE, > LANG, LC_ALL ) but something still tries to convert to utf32BE > specifically ... https://mastodon.online/@Andrew_R/115758255150435130 qemu-system-ppc -cdrom ~/ISO/NetBSD-10.0-macppc.iso -hda ~/QEMU/Mac-netbsd-2gb.qcow2 -M mac99,via=pmu -boot c -prom-env 'boot-device=hd:,ofwboot.xcf;1' -net nic,macaddr=52:54:00:02:12:23,model=ne2k_pci -net vde,sock=/tmp/vde.qemux0 -g 1024x768x32 plus qemu-system-sparc -hda /dev/shm/sparc-netbsd-15g-compressed-r1.qcow2 -g 1024x768x24 -smp 1 -accel tcg,thread=multi,tb-size=256 -M SS-20 -m 0.25G -net nic,macaddr=52:54:00:01:12:24 -net vde,sock=/tmp/vde.qemux0 so, they both see each other, I launched X server on ppc side with "startx -- -listen tcp -ac" and then set DISPLAY variable on Sparc side. I think we 9Cinelerra-GG) work ok between LE and BE hosts, but qemu-system-sparc's TCX24 show swapped red-blue colors for some reason, on xine (launched from i586 host), mplayer (local and LE host), and few elements in qt5-based programs like Openshot-qt or Avidemux (those obviously launched from i586 host targeting qemu VMs) PPC qemu runs NetBSD 10.0 and generic framebuffer (and ne2k_pci, because GEM networking hw hang on configuation in this qemu/NetBSD kernel combo) I had a lot of fun, and some confusion about where bugs might be. But I am happy Cinelerra-gg at least does not crash on "strict" platforms like Sparc32. This gives me hope! > > > > > > Thanks > > > > David

1 0

Two more patches,one install script, second makes format gui button work for segmented profiles.
by Andrew Randrianasulu 20 Dec '25

20 Dec '25

I hope rebase did not mess them ... 0019 patch is from our failed libtheora 1.2.0 update, but may be if user now links to thirdparty theora set to 1.2.0 it will be needed anyway? Postprocessing takes some time but makes image less blocky, normally. There are other patches, like setting bitexact flag on swscale operations, but I am not sure how useful they are ....

2 5

Updated my rosa 2016 64bit spec file
by Andrew Randrianasulu 20 Dec '25

20 Dec '25

latest Cingg source release (with ffmpeg 8.0) builds fine in chroot https://cloud.mail.ru/public/2JbP/5f8QkfjsA

2 4

HTML documentation link lead to pdf version ?
by Andrew Randrianasulu 19 Dec '25

19 Dec '25

https://cinelerra-gg.org/download/CinelerraGG_Manual It downloads 71 Mb pdf, instead of dumping you into About.html or something for html browser proper ...

4 5

prores vulkan encoder at github.
by Andrew Randrianasulu 18 Dec '25

18 Dec '25

Probably at patchwork also, but GSoC page links to this https://github.com/raphaelthegreat/FFmpeg/tree/prores-vk-encode page itself: https://summerofcode.withgoogle.com/archive/2025/projects/PHTgRqGg so I pulled tree into separate folder, switched to branch, extracted 3 last commits, git am them to ffmpeg.git (there was host copy vulkan fix in meantime) And .... it seems to work. At least for HQ profile ... at 4.5 fps, but this is faster than ffv1 (but not lossless, obviously). ./ffmpeg -init_hw_device vulkan=vulkan -hwaccel vulkan -filter_hw_device vulkan -i /dev/shm/Nature072-A006C0035_230328_4U1Q.mov -vf libplacebo -c:v prores_ks_vulkan -f matroska -noautoscale -pix_fmt vulkan 4k.mov ffmpeg version N-120782-gc12bca52f8 Copyright (c) 2000-2025 the FFmpeg developers built with gcc 11.2.0 (GCC) configuration: --enable-opencl --disable-debug --enable-libx265 --enable-libx264 --enable-gpl --enable-libplacebo --enable-vulkan --enable-libshaderc --enable-libzimg --enable-libaom --enable-libdav1d --enable-libsoxr --enable-libfontconfig --enable-libfreetype --enable-libfribidi --enable-gnutls --enable-libass --enable-libbluray --enable-libcdio --enable-frei0r --enable-libgsm --enable-openal --enable-libopus --enable-librtmp --enable-libsnappy --enable-libspeex --enable-libssh --enable-libtheora --enable-libtwolame --enable-libv4l2 --enable-libvidstab --enable-libvorbis --enable-libvpx --enable-libwebp libavutil 60. 19.101 / 60. 19.101 libavcodec 62. 21.100 / 62. 21.100 libavformat 62. 6.103 / 62. 6.103 libavdevice 62. 2.100 / 62. 2.100 libavfilter 11. 10.101 / 11. 10.101 libswscale 9. 3.100 / 9. 3.100 libswresample 6. 2.100 / 6. 2.100 Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '/dev/shm/Nature072-A006C0035_230328_4U1Q.mov': Metadata: major_brand : qt minor_version : 512 compatible_brands: qt creation_time : 2023-04-07T18:41:32.000000Z encoder : Blackmagic Design DaVinci Resolve Studio Duration: 00:00:08.00, start: 0.000000, bitrate: 1787424 kb/s Stream #0:0[0x1]: Video: prores (HQ) (apch / 0x68637061), yuv422p10le(tv, bt709, progressive), 3840x2160, 1787418 kb/s, SAR 1:1 DAR 16:9, 60 fps, 60 tbr, 15360 tbn (default) Metadata: creation_time : 2023-04-07T18:41:32.000000Z handler_name : VideoHandler vendor_id : appl encoder : Apple ProRes 422 HQ timecode : 21:22:52:12 Stream #0:1[0x2](eng): Data: none (tmcd / 0x64636D74), 0 kb/s Metadata: creation_time : 2023-04-07T18:41:32.000000Z handler_name : TimeCodeHandler timecode : 21:22:52:12 File '4k.mov' already exists. Overwrite? [y/N] y Stream mapping: Stream #0:0 -> #0:0 (prores (native) -> prores (prores_ks_vulkan)) Press [q] to stop, [?] for help [prores_ks_vulkan @ 0xb299440] Autoselected HQ profile to keep best quality. It can be overridden through -profile option. Output #0, matroska, to '4k.mov': Metadata: major_brand : qt minor_version : 512 compatible_brands: qt encoder : Lavf62.6.103 Stream #0:0: Video: prores (HQ) (apch / 0x68637061), vulkan(tv, bt709, progressive), 3840x2160 [SAR 1:1 DAR 16:9], q=2-31, 200 kb/s, 60 fps, 1k tbn (default) Metadata: encoder : Lavc62.21.100 prores_ks_vulkan creation_time : 2023-04-07T18:41:32.000000Z handler_name : VideoHandler vendor_id : appl timecode : 21:22:52:12 [out#0/matroska @ 0xb29ddc0] video:1770722KiB audio:0KiB subtitle:0KiB other streams:0KiB global headers:0KiB muxing overhead: 0.001071% frame= 480 fps=4.5 q=-0.0 Lsize= 1770741KiB time=00:00:08.00 bitrate=1813239.1kbits/s speed=0.0755x elapsed=0:01:45.90 testing: bash-5.1$ mplayer 4k.mov MPlayer SVN-r38660-11.2.0 (C) 2000-2024 MPlayer Team 227 audio & 468 video codecs do_connect: could not connect to socket connect: No such file or directory Failed to open LIRC support. You will not be able to use your remote control. Playing 4k.mov. libavformat version 61.6.100 (internal) libavformat file format detected. [lavf] stream 0: video (prores), -vid 0 VIDEO: [apch] 3840x2160 0bpp 60.000 fps 0.0 kbps ( 0.0 kbyte/s) ========================================================================== Opening video decoder: [qtvideo] Quicktime Video decoder sh->ImageDesc not set, try -demuxer mov if this fails. QuickTime6.3 DLLs found QuickTime.qts patched!!! old entry=0x6693c3e0 theQuickTimeDispatcher catched -> 0x6693c3e0 WARNING! Invalid Ptr handle! theQuickTimeDispatcher catched -> 0x6693c3e0 theQuickTimeDispatcher catched -> 0x6693c3e0 Movie-Aspect is 1.78:1 - prescaling to correct movie aspect. VO: [vdpau] 3840x2160 => 3840x2160 Packed YUY2 DecompressSequenceBegin result=-8961 VDecoder init failed :( Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family libavcodec version 61.17.100 (internal) Selected video codec: [ffprores] vfm: ffmpeg (FFmpeg Apple ProRes) ========================================================================== Clip info: COMPATIBLE_BRANDS: qt MAJOR_BRAND: qt MINOR_VERSION: 512 ENCODER: Lavf62.6.103 Load subtitles in ./ Audio: no sound Starting playback... Could not find matching colorspace - retrying with -vf scale... Opening video filter: [scale] Movie-Aspect is 1.78:1 - prescaling to correct movie aspect. [swscaler @ 0x57a10180]bicubic scaler, from yuv422p10le to yuv420p using MMXEXT VO: [vdpau] 3840x2160 => 3840x2160 Planar YV12 [VD_FFMPEG] DRI failure. Movie-Aspect is 1.78:1 - prescaling to correct movie aspect. VO: [vdpau] 3840x2160 => 3840x2160 Planar YV12 V: 0.0 0/ 0 ??% ??% ??,?% 0 0 [VD_FFMPEG] DRI failure. V: 0.3 0/ 0 ??% ??% ??,?% 0 0 [ASPECT] Warning: No suitable new res found! [ASPECT] Warning: No suitable new res found! V: 1.9 0/ 0 47% 460% 0.0% 0 0 Exiting... (Quit) I wonder if Terje can replicate those results? Do not forgot to check if libplacebo was compiled in for your ffmpeg version ...

2 4

Cinelerra-GG Download Center - Updates
by Sam 18 Dec '25

18 Dec '25

Cinelerra-GG Download Center - Updates Screenshot-Cin-Donwload-Center ------------------------------------------------------------------------ * * *New Features & Improvements * *Search & Navigation* * Global Search – Search across all folders and subfolders * File Type Filters – Quick filter by PDF, AppImage, Archives, HTML, Images, Docs * Date Filter – Filter by Last 7/30/90 days or This Year * Keyboard Navigation – Navigate with arrow keys (↑/↓), open with Enter, download with D, copy link with C, show QR with Q * Sticky Toolbar – Search and filter bar stays visible while scrolling *Display & Views* * Grid/List Toggle – Switch between compact list and visual grid view * Image Thumbnails – Real image previews with lightbox for full-screen viewing * Dark Mode – Toggle between light and dark theme (auto-saves preference) * NEW Badges – Visual indicator for files uploaded in the last 7 days *File Actions* * Copy Link Button – One-click copy of download links to clipboard * QR Code Generator – Generate scannable QR codes for mobile downloads * View in Browser – HTML and PDF files can be viewed directly in browser * MD5 Checksums – Display file checksums where available *Feeds & Extras* * RSS Feed – Subscribe to latest uploads at /feed.php * Quick Links Bar – Direct access to Latest AppImages, Manual, and Source Code * Back to Top – Floating button for easy navigation * Toast Notifications – Visual feedback for actions *Technical* * Smart Sorting – Sort by date, name, or size (newest first by default) * Responsive Design – Optimized for mobile, tablet, and desktop * Security Hardened – Protected against XSS, path traversal, and header injection * Fast Loading – Lazy-loading images and optimized performance ------------------------------------------------------------------------ *Visit:* https://download.cinelerra-gg.org

2 1

Mailing List
by Andrea paz 17 Dec '25

17 Dec '25

I usually read the mailing list starting from the archive so that I always have all the monthly threads at a glance. It seems to me that since it was reopened, the archive does not show the most recent messages, continuing to show only the first three emails sent. Am I doing something wrong? Is the archive updated only monthly or at other intervals? The mailing list works, because I always see all emails as soon as they arrive in my email client. I am only referring to how the archive works. It is not even important, because I imagine I am the only one who views the mailing list from the archive; I am simply curious. The archive address I use is the following (it may not be the correct one): https://lists.cinelerra-gg.org/archives/list/[email protected]/

4 8

Forum
by Andrea paz 17 Dec '25

17 Dec '25

For a few days now, the layout of the “artist forum” has been confusing. Is this happening to you too?

3 3

-Wcast-align warnings
by Andrew Randrianasulu 17 Dec '25

17 Dec '25

Trying to see if there were more such warnings than simply in guicast/bootstrap.C (this one kills compiled program on Sparc32) I added CFLAGS += -Wcast-align to global_config after configure generated it Aaaand ... it seems that our python code really generates c files with loads of those cast-align warnings, but some core files have them too ... I honestly do not know how to properly fix this :( v make -j8 > MLOG 2>&1 grep warning MLOG | grep Wcast | wc -l 1921 bccapture.C:77:24: warning: cast from 'const char *' to 'const u_int32_t *' (aka 'const unsigned int *') increases required alignment from 1 to 4 [-Wcast-align] bcresources.C:417:20: warning: cast from 'const char *' to 'const u_int32_t *' (aka 'const unsigned int *') increases required alignment from 1 to 4 [-Wcast-align] bcresources.C:1842:19: warning: cast from 'char *' to 'uint32_t *' (aka 'unsigned int *') increases required alignment from 1 to 4 [-Wcast-align] bcresources.C:1843:19: warning: cast from 'char *' to 'uint32_t *' (aka 'unsigned int *') increases required alignment from 1 to 4 [-Wcast-align] bcresources.C:1850:19: warning: cast from 'char *' to 'uint32_t *' (aka 'unsigned int *') increases required alignment from 1 to 4 [-Wcast-align] bcresources.C:1851:19: warning: cast from 'char *' to 'uint32_t *' (aka 'unsigned int *') increases required alignment from 1 to 4 [-Wcast-align] bcwindowbase.C:475:25: warning: cast from 'const char *' to 'const u_int32_t *' (aka 'const unsigned int *') increases required alignment from 1 to 4 [-Wcast-align] units.C:584:7: warning: cast from 'const char *' to 'const uint32_t *' (aka 'const unsigned int *') increases required alignment from 1 to 4 [-Wcast-align] units.C:617:7: warning: cast from 'const char *' to 'const uint32_t *' (aka 'const unsigned int *') increases required alignment from 1 to 4 [-Wcast-align] bcwindowbase.C:4448:18: warning: cast from 'uint8_t *' (aka 'unsigned char *') to 'unsigned int *' increases required alignment from 1 to 4 [-Wcast-align] vframe.C:968:15: warning: cast from 'unsigned char *' to 'float *' increases required alignment from 1 to 4 [-Wcast-align] vframe.C:971:15: warning: cast from 'unsigned char *' to 'float *' increases required alignment from 1 to 4 [-Wcast-align] vframe.C:973:15: warning: cast from 'unsigned char *' to 'float *' increases required alignment from 1 to 4 [-Wcast-align] vframe.C:975:15: warning: cast from 'unsigned char *' to 'float *' increases required alignment from 1 to 4 [-Wcast-align] vframe.C:993:3: warning: cast from 'unsigned char *' to 'uint16_t *' (aka 'unsigned short *') increases required alignment from 1 to 2 [-Wcast-align] ==== and so on ... They are harmless on x86/arm, but surely kills program on mips64/sparc (and may be Alpha).

1 1