Damien Miller

IMO this is a very niche feature and not useful to the majority of users. I don't want to assign a scarce option letter for it. Perhaps it could be...

This looks good to me and the better signal handling could allow us to make `cleanup_handler()` a regular call in future (i.e. no longer calling it in signal context). Would...

What if the user doesn't have faildelay enabled? sshd has no way of knowing...

Please reread what Darren [said above](https://github.com/openssh/openssh-portable/pull/250#issuecomment-841820674): > ssh-copy-id is a contributed file that OpenSSH distributes. Its upstream is https://git.hands.com/?p=ssh-copy-id.git;a=summary and any pull requests should be directed there. This PR will...

Sorry, we're not interested in enabling support for unsafe crypto. If you need to connect to such devices, then please compile your own ssh with SSH_RSA_MINIMUM_MODULUS_SIZE redefined.

We're not interested in taking a dependency on libsystemd, see https://bugzilla.mindrot.org/show_bug.cgi?id=2641 for the approach we are likely to take.

Commenting out the *FontOutline and *FontHinting methods gets it going for me on Ubuntu Precise with libsdl-ttf-2.09: ``` diff diff --git a/ttf/ttf.go b/ttf/ttf.go index 3401543..d1a4a74 100644 --- a/ttf/ttf.go +++ b/ttf/ttf.go...

> The implementation clears the TX FIFO just in case the user had been manually issuing commands / data with `enqueue_transaction` / `enqueue_data`. (The "if busy, return error" behavior should...

I was just doing a bunch of back-to-back spi.write(), something like: ```rust loop { let mut d: [u8; 4] = [ 0x12, 0x34, 0x56, 0x78 ]; spi.write(&mut d).ok(); let mut...

I think implementing it without backwards compatibility is fine, we could land this early in the release cycle and see if anyone finds breakage with non-OpenSSH clients. If there are...