In D23849#534169, @ngraham wrote:

+1. Do you wanna commit this, or should I?

In D23849#534154, @romangg wrote:

Or let's just get it in. It's a small change and we still have the beta phase to test it.

In D23849#534126, @romangg wrote:

Thanks for the patch Jason. So the GSendStr(nullptr) is called on m_password.isNull() because this aborts the current PAM session? If this is correct, patch looks good to me.

In D23849#532568, @zzag wrote:

@jfranklin Ping me if you don't get any feedback in a day or two.

Friendly ping concerning this issue. Has anyone related to the screen locker project had some time to take a quick look at this?

I also didn't point this out in my original post, but I fixed a misuse of pam_end(). The status passed to pam_end() should be the returned status of the last PAM library call. This is noted in the documentation for Linux-PAM.

Problem exists in linux-pam. I will fix the issue there. Abandoning revision.

I investigated this further in the linux-pam repository, and I ended up creating this issue. The developers there acknowledged that the documentation was not correct and have asked for a PR to fix it.

In D23734#526285, @zzag wrote:

Hmm, "Unlocking failed" doesn't show up immediately if I type wrong password.

In D23734#526274, @zzag wrote:

We want to notify the user as soon as possible about authentication failure.

I'm working on a machine that uses kcheckpass with libpam-poldi for authentication. This allows for the usage of GPG smart cards at login.