Opened 3 years ago

Closed 3 years ago

Last modified 3 years ago

#9355 closed defect (fixed)

Fix for PES packets with too much padding

Reported by: Sergio Ammirata, Ph.D. Owned by:
Priority: normal Component: undetermined
Version: unspecified Keywords:
Cc: Blocked By:
Blocking: Reproduced by developer: no
Analyzed by developer: no

Description

Summary of the bug:
PES packet with too much padding trigger unlimited error messages "PES packet size mismatch" because the code that corrects the length is wrong.
Here is a sample file: ​http://99.93.62.129/smpte2038.ts
PID 300 is the one triggering the errors.
I am attaching a patch that fixes the problem.

Attachments (1)

0001-Fix-setting-the-correct-size-for-PES-packets-with-to.patch​ (1011 bytes ) - added by Sergio Ammirata, Ph.D. 3 years ago.
proposed patch

Download all attachments as: .zip

Change History (11)

by Sergio Ammirata, Ph.D., 3 years ago

Attachment: 0001-Fix-setting-the-correct-size-for-PES-packets-with-to.patch​ added

proposed patch

comment:1 by Balling, 3 years ago

What is #2392? That one has nothing to do with above.

in reply to:  1 ; comment:2 by Sergio Ammirata, Ph.D., 3 years ago

Replying to Balling:

What is #2392? That one has nothing to do with above.

I do not know what #2392 is ... that comment was there from before.

in reply to:  2 comment:3 by Balling, 3 years ago

Replying to Sergio Ammirata, Ph.D.:

Replying to Balling:

What is #2392? That one has nothing to do with above.

I do not know what #2392 is ... that comment was there from before.

Yet that comment should be deleted after your change or better clarified.

Version 0, edited 3 years ago by Balling (next)

comment:4 by Balling, 3 years ago

For example, does it fix that warning sample in #3309? What about my issue #8108? Also samples in #6152. And in #6026.

Last edited 3 years ago by Balling (previous) (diff)

comment:5 by Balling, 3 years ago

Status: new → open

Also this is a duplicate of #5860. P.S. Apparently not.

Last edited 3 years ago by Balling (previous) (diff)

in reply to:  2 comment:6 by Balling, 3 years ago

Replying to Sergio Ammirata, Ph.D.:

Replying to Balling:

What is #2392? That one has nothing to do with above.

I do not know what #2392 is ... that comment was there from before.

I found it! It is this issue, ​https://ffmpeg-issues.lscube.narkive.com/s0jecfqy/issue2392-memory-leak-in-mpegts-demuxer-on-samples-with-too-large-pes-packets so a different bug tracker. WTF.

Last edited 3 years ago by Balling (previous) (diff)

comment:7 by Balling, 3 years ago

What is even in that PID 0x012c in PES data?

comment:9 by Marton Balint, 3 years ago

Resolution: → fixed
Status: open → closed

Fixed in 9420f7e09560d82ebdb96ddc6724a734f4fe0b95.

comment:10 by Balling, 3 years ago

Yet it does not fix any other issues above, interesting!

Note: See TracTickets for help on using tickets.