Skip to content
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

Segfault at exit involving multithreading and av_packet_free_side_data #381

Open
Octogonapus opened this issue Dec 20, 2022 · 0 comments
Open

Comments

@Octogonapus
Copy link
Contributor

I have been experiencing a segfault that occurs when Julia exits while a background thread is encoding a video.
I can't reproduce the segfault when exiting from the encoding thread; this happens only when exiting from one thread while encoding from another.
I also can't get you an rr trace since this requires two threads.

MWE:

using VideoIO, Mmap, ThreadPools

nframes = 500
resolution = [2000, 1500]

path, io = mktemp()
mat = Mmap.mmap(io, Array{UInt8,3}, (resolution..., nframes))
for i = 1:nframes
    mat[:, :, i] .= rand(UInt8, resolution...)
end
Mmap.sync!(mat)
close(io)
mat = nothing
GC.gc(true)

ThreadPools.@tspawnat 2 begin
    open(path) do io
        vidmat = Mmap.mmap(io, Array{UInt8,3}, (resolution..., nframes))
        open_video_out(
            "/tmp/out.mp4",
            view(vidmat, :, :, 1),
            framerate=55,
            encoder_options=(color_range=2, crf=0, preset="ultrafast"),
            scanline_major=true,
        ) do video_io
            i = 1
            while i <= nframes
                VideoIO.write(video_io, view(vidmat, :, :, i))
                i += 1
            end
        end
    end
end

sleep(5)
exit(0)
julia -t auto --project --startup-file=no reproduce.jl

signal (11): Segmentation fault
in expression starting at /home/salmon/Documents/code/reproduce_julia_segfault_with_tasks_at_exit/reproduce.jl:36
av_packet_free_side_data at /home/salmon/.julia/artifacts/b409c0eafb4254a980f9e730f6fbe56867890f6a/lib/libavcodec.so (unknown line)
unknown function (ip: (nil))
Allocations: 10866309 (Pool: 10859613; Big: 6696); GC: 44

VideoIO v1.0.3
ThreadPools v2.1.1

julia> versioninfo()
Julia Version 1.8.3
Commit 0434deb161e (2022-11-14 20:14 UTC)
Platform Info:
  OS: Linux (x86_64-linux-gnu)
  CPU: 16 × Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-13.0.1 (ORCJIT, skylake)
  Threads: 16 on 16 virtual cores

Linux 6.0.12-200.fc36.x86_64

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant