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.

Sign up for GitHub

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

Jump to bottom

Remove crashy unwrap() in on_error #1364

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Remove crashy unwrap() in on_error #1364

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 4 additions & 1 deletion src/decode.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4994,7 +4994,10 @@ pub fn rav1d_submit_frame(c: &Rav1dContext, state: &mut Rav1dState) -> Rav1dResu
) {
fc.task_thread.error.store(1, Ordering::Relaxed);
let _ = mem::take(&mut *fc.in_cdf.try_write().unwrap());
if f.frame_hdr.as_ref().unwrap().refresh_context != 0 {
if f.frame_hdr
.as_ref()
.is_some_and(|frame_hdr| frame_hdr.refresh_context != 0)
{
Comment on lines +4997 to +5000
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm not sure what the right behavior is in the None case, but crashing is certainly not it

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Places where there is an unwrap should have been unchecked derefs in C, so let me check what C was doing here.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

So C is doing the same thing as we are here:

Can you check if you get an error/crash on dav1d here or not? Thanks!

Copy link
Contributor Author

@emilk emilk Oct 4, 2024 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Using https://crates.io/crates/dav1d 0.10.3 I do not get a crash, but I get an Invalid argument error and some useful info printed:

Error parsing frame header
Error parsing OBU data

It should be noted though that dereferencing a null pointer is not necessarily a crash in C - it is Undefined Behavior. The compiler is free to assume the pointer is not null, and do whatever crazy shenanigans that implies.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hmm, it's possible UB could be doing something weird here, but I'd still expect a segfault for the most part. We likely diverged at some point, so we'll have to figure out where and see if that's something we need to fix or if it's okay to crash, just in a different place.

That said, this video should successfully decode with normal dav1d, but doesn't? Is that a bug with dav1d as well?

Copy link
Contributor Author

@emilk emilk Oct 7, 2024 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

or if it's okay to crash

I would prefer a rav1d that never crashes on any input, but returns errors.

That said, this video should successfully decode with normal dav1d, but doesn't? Is that a bug with dav1d as well?

I don't know if the video should successfully decode, or if it is actually a bad video.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I would prefer a rav1d that never crashes on any input, but returns errors.

Yup, that would definitely be preferable.

let _ = mem::take(&mut f.out_cdf);
}
for i in 0..7 {
Expand Down
Loading