Detect error at the end of echo(1) after performing an explicit
fflush().

Address jilles's comments.

Revert all diet changes.

Looks good to me as long as the capscium specific code were wrapped with #ifdef's as this is shared with other platforms.

In D25833#572370, @cem wrote:

Any idea why these were removed? There aren’t known attacks on correctly implemented CBC modes generally. Maybe it is specific to the openssh context. So, I am happy to defer to upstream defaults here.

ACK. I think this can be accomplished with a more intrusive change so let's do it in a later refactor instead.

Get rid of geom_ccd.c portion and focus on the main change.

Separate from D25738.

Eliminate some unrelated changes to make this smaller.

It seems that the lack of /dev/diskid nodes was not a regression so it's not a blocker.

@cem -- Note that D24968 needs to be amended due to conflict (I have it locally with an amendment to D25168 to make geli detach -l work again with aliases), I've uploaded these to https://github.com/delphij/freebsd/commits/scratch/geom .

Will use D25563.

Actually assert chunksize > MAXBPS;

Alternative solution: D25563 .

Minor style fix.

ministat graph for time savings.

Instead of padding the buffer, assert that they always fit as the specification requires
sector size to be one of 512, 1024, 2048 or 4096 bytes.

LGTM, but I wonder if it's possible to have AWSDNS/Route53 to do the resolution for the clients since we are already using it?

LGTM overall; do we support AVX in kernel code now, or are we just building these but giving kfpu_allowed a false value for now?

Minor typo fixes.

Were you able to find out why GPT partitions are not showing up in the /dev/diskid directory (e.g. /dev/diskid/DISK-BHYVE-2613-8EFD-BAF4p2), or are you not seeing similar behavior (since it might be specific to my usage, details in Message-ID: <cb21f4a0-f9d4-db5d-287a-0cb29c023b75@delphij.net>)?

I'm not 100% comfortable with the removal of these = NULL's but it appears that the code didn't cared about their value being non-zero, so the change as-is seemed Okay.

I think setting tdb_xform to NULL is no longer needed (as they are about to be done by the caller anyway).

Actually, my expectation was that the _Final methods shall destroy the context descriptor (and thus the bzero's are not necessary) like all other cryptographic hash methods, but apparently we are supporting the use of skein.h directly (!), and it doesn't do the bzero'ing.

LGTM.