-
Notifications
You must be signed in to change notification settings - Fork 11
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[ Linux ] malloc(): invalid size (unsorted) #31
Comments
There must be an allocated memory over-addressing somewhere. Without other debug info unfortunately we cannot go any further. |
I had a similar issues ( I followed the guide (posix.rst) and used the I found that something was fishy, when I compiled 3.6.0 and it was also broken. I did build 3.6.0 and 3.7.0 to before, and they were both working. @pinterf let me know which kind of debug information might help on this and I'll try my best to get them. |
Thanks, it's not easy, the final error message unfortunately means nothing because the memory overwrite occurs somewhere else. |
With current master I am getting
I checked all the values being passed in the initialization of cArray and they are all valid. I've tried AI to redo this section but always end up with same error. Removing the throw condition just ends up with the same error for tbuffer later in TFM.cpp. |
Could you please check if this method works for you? |
This gave a ton of errors during build |
Than must be a different issue. including smmintrin.h gives you this huge error dump. Do you use cmake option Did you make any other changes in your build process or environment? The errors are not related to the changes made recently. |
Yeah I'm not sure what happened with that build, sorry about that. Anyways I started over completely and it built fine but I still get
Also can you fix I'm not using any cmake defines, I just started building exactly how the readme said because I was having problems. I am on a intel xeon w1290-t 64gb ecc ram debian testing machine using gcc 13.2.0-25. This is the script I've been using. I have changed and removed TFM's parameters to debug but nothing changed.
Removing TFM completely I get a perfectly fine video output. |
I was trying in minGW with an updated gcc (just for the syntax). Seems I have to resurrect my WSL Ubuntu... Stay tuned. |
I've already added it in the previous commit. I'm now on Linux and gcc 13.1 (instead of 9.4) and it builds just fine. Now I try to make a working avisynth environment as well and I hope I would see the magic error message. |
Awesome thanks. My build script auto adds the line so I didn't notice. Avisynth should be quick but debian and ubuntu should set |
I should note also that instead of rebuilding the entire ffmpeg binary to include avisynth support I just compiled avs2yuv and pipe it to ffmpeg. it can also just directly output to y4m (vlc compatible). ffmpeg in debian repo from what i remember doesnt include avs support so i suspect neither does ubuntu. |
Thanks, I'm using linux once a year and even forget what sudo and apt-get is, I always need the learning curve. I've installed a fresh 22.04, gcc, g++13, cmake (and saw your comment after there). Then got a stock 6.1 ffmpeg (22.04 was stuch with an earlier 4.x version) but it came w/o avs support, so I built it from qyot27's repo. Compiled a fresh avisynth and tivtc, and now the only problem I got (on my workbench) that at the very end of the encoding I get a segmentation fault. So I'm on the right trace, there must be a solution, I ask a little patience. |
Of course 😁. If u need any help/suggestions with linux just let me know. I noticed with a lot of the avisynth plugins if I compiled with optimizations such as O3 or LTO I would get segfaults. I just stopped trying and compile debug builds. |
Well, it works on my Linux, yeah. |
You are awesome, it builds and run perfect now. Thank you so much!
And this command to process a DVD video.
I just need nnedi3 for QTGMC to compile for linux and my entire conversion process will work on linux. I haven't experimented with this |
Your welcome, happy encoding :) |
off topic
You can try nnedi3cl (the latest QTGMC(p) versions can work with it). /off topic |
Thank you for the suggestion but I use mesa's amd driver and it doesn't include opencl support (listed as WIP) and amd's official driver has broken too many of my previous systems so I don't mess with it. Maybe I'll get back around to it. I'm waiting for AI to catch up and convert the code for me lol. And I haven't heard of qtgmcp, what is that? Update: 6/29 |
Thanks. I'm getting "Error: Script error: There is no function named 'ex_bs'.", so I guess requirements also changed for it. |
I'm trying to get QTGMC deinterlacing working on linux which requires a bit of work.
I've installed the arch linux package avisynth-plugin-tivtc-git and if I run the following script:
and then try to run it through ffplay:
Without
TDeint()
it works fine, so it seems to be aTDeint
issue, I'm not sure where to start on debugging it as it's my first time using AviSynth.The text was updated successfully, but these errors were encountered: