-
Notifications
You must be signed in to change notification settings - Fork 747
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
vaDisplayIsValid random Segmentation fault #811
Comments
Can you add |
No
Depends. It will be slower but not always "significantly." |
@xlz Hi, I'm going to have to try this tomorrow, and will get back to you on it. |
libva is for RGB. If you don't need RGB you would use this instead
Yeah, but a bug is a bug. |
Oh, and you also need to use an environment variable to disable libva |
I'm away from the robot right now, but I can tell you that I'm |
Hi @xlz I rebuilt with the added line, but it doesn't print anything (I'm using the default logger). I've also disabled VAAPI whilst doing that. Should I try echoing on stdout? |
Essentially I want to find out what kind of value does display have and where exactly does it crash in libva. |
If I disable libva in CMAKE then I'm sure that display won't echo, right? |
If disabled at build time it won't print. |
ok, so I should re-enable and try again? Because right now I've built it without VA |
Yes. |
Hi, I've added the default logger, rebuilt using libva, added the line, but I still don't see that particular log entry... |
Then it's probably working as intended. |
I'm closing this as I'm going to assume its a bug in |
Overview Description:
I am getting random SEGFAULT from
vaDisplayIsValid
.Version, Platform, and Hardware Bug Found:
git log -1 --oneline
: cec93c7 Merge pull request Disable default copy and assignment constructors #765 from xlz/disable-default-constructorsSteps to Reproduce:
I was getting random SEGFAULT, so I kept running it under gdb, and at some stage I found out:
My system has the apt-get version of libva: 1.7.0-1 installed.
I understand this is quite likely a bug in libva, so I've opened an issue there as well.
However I'd like to know if this has been experienced before, and if removing libva will significantly slow down libfreenect2.
The text was updated successfully, but these errors were encountered: