Upstream: Add convenience wrappers for common uio usage

[lundman]

Use ZFS specific UIO macros for shared-code.
The first controversial PR!

Motivation and Context

Code that poke around inside uio, like uio->l_offset can not work on OsX due to opaqueness of the struct. The API call uio_offset() should be used. I went with this for readability, if people prefer #ifdef APPLE everywhere, that is also doable.

Description

The macOS uio struct is opaque and the API must be used, this
makes the smallest changes to the code for all platforms.

Most affected area is module/icp, which had a lot of code repetition.
It was tricky to make the for loop that skips to first non-empty
iov index look good, so it was moved into a macro. Technically, this
"skip code" is not required in ZFS, as all callers to crypto will not
create empty iov.

For those curious what the macOS versions of the macros look like, they are here:

https://github.com/openzfsonosx/openzfs/blob/sorted2/include/os/macos/zfs/sys/zfs_context_os.h#L69

How Has This Been Tested?

Unable to test, relying on CI.

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Performance enhancement (non-breaking change which improves efficiency)
• Code cleanup (non-breaking change which makes code smaller or more readable)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation (a change to man pages or other documentation)

Checklist:

• My code follows the ZFS on Linux code style requirements.
• I have updated the documentation accordingly.
• I have read the contributing document.
• I have added tests to cover my changes.
• I have run the ZFS Test Suite with this change applied.
• All commit messages are properly formatted and contain Signed-off-by.

[lundman]

Is FreeBSD not including zfs_context_os.h when compiling dmu.c?

[codecov]

Codecov Report

Merging #10412 into master will decrease coverage by 0.49%.
The diff coverage is 41.81%.

@@            Coverage Diff             @@
##           master   #10412      +/-   ##
==========================================
- Coverage   79.58%   79.09%   -0.50%     
==========================================
  Files         391      393       +2     
  Lines      123872   123859      -13     
==========================================
- Hits        98586    97963     -623     
- Misses      25286    25896     +610     

Flag	Coverage Δ
#kernel	79.97% <38.54%> (-0.06%)	⬇️
#user	63.57% <34.73%> (-2.55%)	⬇️

Impacted Files	Coverage Δ
module/icp/io/sha1_mod.c	2.43% <0.00%> (+0.06%)	⬆️
module/icp/io/sha2_mod.c	29.05% <0.00%> (+0.78%)	⬆️
module/icp/io/skein_mod.c	4.91% <0.00%> (+0.15%)	⬆️
module/zfs/sa.c	84.30% <ø> (ø)
module/zfs/zfs_sa.c	61.11% <0.00%> (ø)
include/os/linux/spl/sys/uio.h	100.00% <100.00%> (ø)
lib/libspl/include/sys/uio.h	100.00% <100.00%> (ø)
module/icp/algs/modes/modes.c	50.94% <100.00%> (-3.61%)	⬇️
module/icp/core/kcf_prov_lib.c	63.07% <100.00%> (-0.70%)	⬇️
module/zfs/dmu.c	86.19% <100.00%> (-0.24%)	⬇️
... and 64 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update e08b993...c31e037. Read the comment docs.

[behlendorf]

Just a couple last nits. Let's also run it by @mattmacy to get his feedback.

[mattmacy]

LGTM

[behlendorf]

You definitely got unlucky with the bots. But all the failures are known. If you could rebase this on master we should be able to get a clean run with only the FreeBSD head bot failing due to recent changes they made upstream.