Dereference device with a defer call to prevent memory leaks

[danielpaulus]

This PR fixes the memory leak mentioned in #97

gousb.OpenDevices currently only dereferences libusb device pointers when opening is unsuccessful or when the opener returns false.
This can cause a memory leak that becomes mostly visible when using multiple libusb contexts in a long running application.

The fix is to always call c.libusb.dereference(dev) so the device reference counter is decreased when OpenDevices is done. That is not a problem as libusb increases the reference counter internally on the c.libusb.open(dev) call and releases it when we call c.libusb.close(dev) again.

[google-cla]

Thanks for your pull request. It looks like this may be your first contribution to a Google open source project (if not, look below for help). Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

📝 Please visit https://cla.developers.google.com/ to sign.

Once you've signed (or fixed any issues), please reply here with @googlebot I signed it! and we'll verify it.

---

What to do if you already signed the CLA

Individual signers

• It's possible we don't have your GitHub username or you're using a different email address on your commit. Check your existing CLA data and verify that your email is set on your git commits.

Corporate signers

• Your company has a Point of Contact who decides which employees are authorized to participate. Ask your POC to be added to the group of authorized contributors. If you don't know who your Point of Contact is, direct the Google project maintainer to go/cla#troubleshoot (Public version).
• The email used to register you as an authorized contributor must be the email used for the Git commit. Check your existing CLA data and verify that your email is set on your git commits.
• The email used to register you as an authorized contributor must also be attached to your GitHub account.

ℹ️ Googlers: Go here for more info.

[coveralls]

Coverage increased (+0.1%) to 79.409% when pulling e40a201 on danielpaulus:fix-memleak-in-opendevices into 0eba1b1 on google:master.

[coveralls]

Coverage increased (+0.2%) to 79.431% when pulling 9cc7659 on danielpaulus:fix-memleak-in-opendevices into 0eba1b1 on google:master.

[danielpaulus]

@googlebot I signed it!

[danielpaulus]

Coverage increased (+0.1%) to 79.409% when pulling e40a201 on danielpaulus:fix-memleak-in-opendevices into 0eba1b1 on google:master.

wow I even increased code coverage ;-)

[zagrodzki]

could you also update a comment in https://github.com/google/gousb/blob/master/libusb.go#L212? it is misleading now as it is, since it implies that all devices returned on the list will be "closed" (and so also opened) at some point, but this is not the case.

[danielpaulus]

could you also update a comment in https://github.com/google/gousb/blob/master/libusb.go#L212? it is misleading now as it is, since it implies that all devices returned on the list will be "closed" (and so also opened) at some point, but this is not the case.

Done.
Should I also create a small unit test for OpenDevices while I am at it?

[zagrodzki]

Should I also create a small unit test for OpenDevices while I am at it?

sure, if you want. You can do that in a separate PR perhaps, this one I think is good to go.

[zagrodzki]

s/memoryleaks/memory leaks

[danielpaulus]

hey @zagrodzki ,
is there anything missing on the PR? If not, when do you think you can release it? I would like to move away from using my private fork :-)

[zagrodzki]

sorry, I assumed you will be able to merge yourself after my approval, but it seems this is not how GitHub flow works...

There is one comment left: typo in libusb.go (#98 (comment))