FFmpeg can't encode a raw 30720x30720 RGB24 input stream

[Rafael Saud]

I'm trying to encode a stream of multiple 30720x30720 images on a POWER8 machine but FFmpeg refuses to do anything.
FFmpeg was git clone'd, ./configure'd, and built just now.

How to reproduce:

$ ./example | ~/ffmpeg/ffmpeg -y -f rawvideo -pix_fmt rgb24 -s 30720x30720 -r 30 -i - -c:v libx264 -preset veryslow -s 7680x7680 output.h264
ffmpeg version N-114700-g0dbf45120a Copyright (c) 2000-2024 the FFmpeg developers
  built with gcc 13 (Gentoo 13.2.1_p20230826 p7)
  configuration: --enable-nonfree --enable-gpl --enable-version3 --enable-power8 --enable-altivec --enable-libx265 --enable-libx264 --enable-libwebp --enable-libvpx --enable-libopus --enable-librsvg --enable-libjxl --enable-libfdk-aac --disable-stripping
  libavutil      59. 13.100 / 59. 13.100
  libavcodec     61.  5.101 / 61.  5.101
  libavformat    61.  3.100 / 61.  3.100
  libavdevice    61.  2.100 / 61.  2.100
  libavfilter    10.  2.101 / 10.  2.101
  libswscale      8.  2.100 /  8.  2.100
  libswresample   5.  2.100 /  5.  2.100
  libpostproc    58.  2.100 / 58.  2.100
Splitting the commandline.
Reading option '-v' ... matched as option 'v' (set logging level) with argument '9'.
Reading option '-loglevel' ... matched as option 'loglevel' (set logging level) with argument '99'.
Reading option '-y' ... matched as option 'y' (overwrite output files) with argument '1'.
Reading option '-f' ... matched as option 'f' (force container format (auto-detected otherwise)) with argument 'rawvideo'.
Reading option '-pix_fmt' ... matched as option 'pix_fmt' (set pixel format) with argument 'rgb24'.
Reading option '-s' ... matched as option 's' (set frame size (WxH or abbreviation)) with argument '30720x30720'.
Reading option '-r' ... matched as option 'r' (override input framerate/convert to given output framerate (Hz value, fraction or abbreviation)) with argument '30'.
Reading option '-i' ... matched as input url with argument '-'.
Reading option '-c:v' ... matched as option 'c' (select encoder/decoder ('copy' to copy stream without reencoding)) with argument 'libx264'.
Reading option '-preset' ... matched as AVOption 'preset' with argument 'veryslow'.
Reading option '-s' ... matched as option 's' (set frame size (WxH or abbreviation)) with argument '7680x7680'.
Reading option 'output.h264' ... matched as output url.
Finished splitting the commandline.
Parsing a group of options: global .
Applying option v (set logging level) with argument 9.
Applying option y (overwrite output files) with argument 1.
Successfully parsed a group of options.
Parsing a group of options: input url -.
Applying option f (force container format (auto-detected otherwise)) with argument rawvideo.
Applying option pix_fmt (set pixel format) with argument rgb24.
Applying option s (set frame size (WxH or abbreviation)) with argument 30720x30720.
Applying option r (override input framerate/convert to given output framerate (Hz value, fraction or abbreviation)) with argument 30.
Successfully parsed a group of options.
Opening an input file: -.
[rawvideo @ 0x100173cf8f0] Opening 'fd:' for reading
[fd @ 0x100173d0300] Setting default whitelist 'crypto,data'
[rawvideo @ 0x100173cf8f0] [IMGUTILS @ 0x7fffefb6bb60] Picture size 30720x30720 is invalid
[AVIOContext @ 0x100173e04e0] Statistics: 65536 bytes read, 0 seeks
[in#0 @ 0x100173b5cb0] Error opening input: Invalid argument
Error opening input file -.
Error opening input files: Invalid argument
$

Before I manually removed the if statement on libavutil/imgutils.c (yes I'm aware this is a horrible practice) and encoded 30720x30720 PNGs but this trick doesn't work with raw video streams (it just silences the error message).

This simple C code could be used for reproducing the bug:

// gcc example.c -O3 -o example

#include <stdint.h>
#include <stdlib.h>
#include <stdio.h>

int main() {
    size_t X = 30720, Y = 30720;
    uint8_t* FrameBuffer = malloc(X * Y * 3);

    for (size_t f = 0; f < 30; f++) {
        for (size_t i = 0, p = 0; i < X * Y; i++, p = i * 3) {
            FrameBuffer[p + 0] = (i * f + 256) & 0xFF;
            FrameBuffer[p + 1] = (i * f + 512) & 0xFF;
            FrameBuffer[p + 2] = (i * f + 768) & 0xFF;
        }
        fwrite(FrameBuffer, 1, X * Y * 3, stdout);
    }

    free(FrameBuffer);
}

Here's the output of uname if it helps
Linux power8-gentoo 6.1.60-gentoo-dist #1 SMP Wed Oct 25 19:38:28 -00 2023 ppc64le POWER8 (architected), altivec supported CHRP IBM,8284-22A GNU/Linux