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

igdaemon.exe stops when data is not consumed by client #9

Closed
IguanaBen opened this issue Jan 7, 2015 · 8 comments
Closed

igdaemon.exe stops when data is not consumed by client #9

IguanaBen opened this issue Jan 7, 2015 · 8 comments
Assignees

Comments

@IguanaBen
Copy link
Contributor

Reported by mortenab on 13 Dec 2010 10:36 UTC
If data is not consumed by client, the igdaemon.exe crashes.

Reproduce by:

  • Make a program that sends 0x12 command to start receiving IR data
  • Do not consume data from igdaemon (iguanaIR-0 named pipe)
@IguanaBen
Copy link
Contributor Author

Commented by jdunn on 16 Dec 2010 12:15 UTC
Thank you for the trio of bug reports. I just wanted to get a couple more details about them. First of all, I assume these were all triggered on the same system, so which version of Windows are you running and is it 32 or 64 bit. For this specific ticket, when does it crash? Right away or does it grow in memory use then crash?

@IguanaBen
Copy link
Contributor Author

Commented by mortenab on 16 Dec 2010 12:25 UTC
Windows 7 64-bit.

The program crashes after a couple of seconds after I stop consuming data (when I stop in a breakpoint in VS). I haven't watched memory usage. But it is persistant.

@IguanaBen
Copy link
Contributor Author

Commented by mortenab on 17 Dec 2010 08:03 UTC
It does not crash.. It just stops sending (ir-)data to the client. A restart of igdaemon corrects the problem.

@IguanaBen
Copy link
Contributor Author

Commented by jdunn on 29 Oct 2014 00:48 UTC
Looks like this is still a problem. I suspect that the main thread is stuck when it tries to write data back to the hung client and then never properly services other clients. This would not be triggerable with our igclient, but custom code could cause it so it needs to be fixed. I bet we can cause the same thing on Linux so I'll have to check it there.

@jdunn14
Copy link
Contributor

jdunn14 commented Mar 19, 2017

Looks like it works fine in Linux. The non-blocking code that makes it work was (re-)enabled way back in Dec 2008. I'll need to check this on Windows, but it seems like it should not be a problem.

@jdunn14
Copy link
Contributor

jdunn14 commented Mar 20, 2017

Well, it is definitely still a problem on Windows, but likely only on Windows.

@jdunn14
Copy link
Contributor

jdunn14 commented Aug 5, 2017

Checking in windows if there is a memory leak it is so slow as to be meaningless. The hardware does eventually crap out as well, possibly because the main thread is blocked.

@jdunn14
Copy link
Contributor

jdunn14 commented Aug 5, 2017

Fixed in a branch of my local repo. Closing this out.

@jdunn14 jdunn14 closed this as completed Aug 5, 2017
jdunn14 added a commit that referenced this issue Aug 5, 2017
jdunn14 added a commit that referenced this issue Aug 5, 2017
…ket. All this is in preparation for solving windows problem #9.
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

2 participants