r350160 brooks July 19, 2019, 11:58 p.m. 
This avoids reading past the end of the static strings.  On a system
with bounds checking these tests fault.

Reviewed by:	asomers
Obtained from:	CheriBSD
MFC after:	1 week
Sponsored by:	DARPA, AFRL
Differential Revision:	https://reviews.freebsd.org/D21004
ViewVC
r350159 brooks July 19, 2019, 11:39 p.m. 
Allow ABI to be over ridden to allow (with other changes) programs to be
built targeting ABIs other than the default.  This is used in CheriBSD.

Reviewed by:	imp
Obtained from:	CheriBSD
MFC after:	1 week
Sponsored by:	DARPA, AFRL
Differential Revision:	https://reviews.freebsd.org/D21001
ViewVC
r350158 brooks July 19, 2019, 11:10 p.m. 
There is no need to convert an intptr_t to a long just to cast it to a
(void *).

Obtained from:	CheriBSD
MFC after:	1 week
Sponsored by:	DARPA, AFRL
ViewVC
r350157 brooks July 19, 2019, 10:54 p.m. 
Obtained from:	CheriBSD
MFC after:	1 week
Sponsored by:	DARPA, AFRL
ViewVC
r350156 kib July 19, 2019, 8:51 p.m. 
When sendmsg(2) sucessfully internalized one SCM_RIGHTS control
message, but failed to process some other control message later, both
file references and filedescent memory needs to be freed. This was not
done, only mbuf chain was freed.

Noted, test case written, reviewed by:	markj
Sponsored by:	The FreeBSD Foundation
MFC after:	1 week
Differential revision:	https://reviews.freebsd.org/D21000
ViewVC
r350155 dougm July 19, 2019, 8:47 p.m. 
entry, combining code currently in vm_map_unwire and
vm_map_wire_locked into a single function, called by each of them for
entries in transition.

Discussed with: kib, markj
Reviewed by: alc
Approved by: kib, markj (mentors, implicit)
Tested by: pho
Differential Revision: https://reviews.freebsd.org/D20833
ViewVC
r350154 imp July 19, 2019, 8:04 p.m. 
If someone wants to subset, they can set MK_ICONV=no. We no longer use it for
the release.
ViewVC
r350153 imp July 19, 2019, 8:04 p.m. 
RELEASE_CRUNCH is no longer relevant.
ViewVC
r350152 imp July 19, 2019, 8:04 p.m. 
Remove documentation of RELEASE_CRUNCH here. It's obsolete and no longer a good
example.
ViewVC
r350151 imp July 19, 2019, 8:04 p.m. 
RELEASE_CRUNCH isn't used for releases any more. If someone wants to subset,
then they can set MK_JAIL=no instead.
ViewVC
r350150 mav July 19, 2019, 8:02 p.m. 
MFC after:	2 weeks
ViewVC
r350149 mav July 19, 2019, 7:15 p.m. 
AMA replaced HPA in ACS-3 specification.  It allows to limit size of the
disk alike to HPA, but declares inaccessible data as indeterminate.  One
of its practical use cases is to under-provision SATA SSDs for better
reliability and performance.

While there, fix HPA Security detection/reporting.

MFC after:	2 weeks
Relnotes:	yes
Sponsored by:	iXsystems, Inc.
ViewVC
r350148 brooks July 19, 2019, 6:47 p.m.
r350147 imp July 19, 2019, 6:39 p.m. 
While we print failure messages on the console, sometimes logs are lost or
overwhelmed. Keeping a count of how many times we've failed retriable commands
helps get a magnitude of the problem.
ViewVC
r350146 imp July 19, 2019, 6:39 p.m. 
Retried commands can indicate a performance degredation of an nvme drive. Keep
track of the number of retries and report it out via sysctl, just like number of
commands an interrupts.
ViewVC