mirror of
https://github.com/games-on-whales/wolf.git
synced 2026-10-15 21:12:08 +00:00
[GH-ISSUE #212] Connecting two clients, if one gets stuck and reconnected, it will cause Wolf to crash #143
Labels
No labels
🧠 1 - 🤖 999
AMD
bug
bug
cannot reproduce
documentation
duplicate
easy
enhancement
enhancement
enhancement
fixed?
fixed?
good first issue
hard
help wanted
Intel
invalid
LXC
medium
moonlight-protocol
Multi GPU
nixos
not a bug
nvidia
pull-request
question
wontfix
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
games-on-whales/wolf#143
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @HeiyuEP on GitHub (Jun 15, 2025).
Original GitHub issue: https://github.com/games-on-whales/wolf/issues/212
error msg:
I obtained the gdb stack at the time of the crash,
stack.txt
The code error occurred in the wait_for_ping function, and handler.unregister cannot return. I am not familiar with the code and cannot understand what happened here.
I'm not sure why the client often gets stuck, and this is a common issue for my computer
@HeiyuEP commented on GitHub (Jun 15, 2025):
stack when client stuck.txt
@ABeltramo commented on GitHub (Jun 15, 2025):
Thanks for the detailed report, I have a few more questions:
CTRL+ALT+SHIFT+S) when the stream works from that client?I'm deep in the woods with the zero copy pipeline PR, I'll probably get back to fixing this once I've merged that first, bear with me..
@HeiyuEP commented on GitHub (Jun 16, 2025):
4. I changed some code, complete log is below
wolf-log.txt
@HeiyuEP commented on GitHub (Jun 16, 2025):
I found a client that also triggers this problem, but the probability is lower.
@HeiyuEP commented on GitHub (Jun 17, 2025):
I got the gstreamer debug log and found a new discovery. Each call to cudaload3 is followed by cudaconvertscale3, but the last cudaload3 is not followed by cudaconvertscale3 and does not return ok. At the same time, cudaupload4 and cudaconvertscale4 of another pipeline are normal.
The error video pipeline is 12829664931820868236_video
gstreamer_log.txt
@HeiyuEP commented on GitHub (Jun 19, 2025):
There is no problem with wolf code. It is my PCIE that has a copy problem. A very simple code can also cause this problem.
@ABeltramo commented on GitHub (Jun 19, 2025):
@HeiyuEP sorry to hear about that, but I'm kind of glad you've found the actual root cause. I was scratching my head trying to think what could cause your issues in the Wolf codebase over the past few days 😅