• Incorporated imp@'s suggestions
• Mentioned that Xstosbt functions were introduced in 14.0

Incorporate markj@'s suggestions.

Addressed ziaee@'s comments.

Addressed ziaee@'s comments.

Looks good.

In D54526#1246414, @des wrote:

Your use of “Obtained from” is incorrect, unless “Hewlett Packard Enterprise” is the name of a piece of open-source software I've never heard of that has a copy of our libutil and already had this patch. You probably want to use “Sponsored by” instead.

In D54327#1241761, @kib wrote:

I suppose this is for stable/15, right?

Fix multi-line comments and wordings.

Within biodone(9):

• Translate bio_error_compat to bio_error, then clear the bio_error_compat and BIO_ERROR_COMPAT flag.
• Beforef calling bp->bio_done(), set the bio_error_compat field and BIO_ERROR_COMPAT flag if BIO_ERROR is set. Clear bio_error_compat and the BIO_ERROR_COMPAT flag after returning from bp->bio_done().

In D54327#1241511, @imp wrote:

So a few words about why this is useful. Our KBI commitments on 15.x is fuzzy at best. What does this enable?

.

g_io_request will empty bio_error_compat as well.

Only implement this for string stuff. mem* stuff will end up calling into itself again because of the use of __builtin_mem*.

Have to rethink the whole patch again.

Updated.

style(9).

Updated.

markj's suggestion.

In D51676#1180379, @markj wrote:

But that use of KSTACK_PAGES matches the definition of BOOTSTACK_SIZE in amd64/locore.S, which also doesn't include opt_kstack_pages.h. So both files should include it.

In D46537#1061028, @markj wrote:

In D46537#1060924, @khng wrote:

Or, I might just do it the way like how other debugger does - not using elf_file but doing adjustments with only ELF parsing libraries rather.

As discussed elsewhere, I'd prefer to have libkvm walk the linker_files list and locate the corresponding binaries, like kgdb does. This way we avoid growing the kernel ABI.

Addressed.

By the way, given that rpcbind and nfsd will install a default pidfile to /var/run even without specifying -P: shall we document that somewhere in RELNOTES and the commit message, or do we make the errno == EEXIST case a warn() call instead of errx()? I am thinking of that because some people might spin up multiple nfsd in their setup without using the rc.d/nfsd script, and with different bound IP for each nfsd spun.

In D50624#1158182, @rmacklem wrote:

I'll leave the minor nits up to you.
Are you a committer or do you need me
to do it?

Add comments.

Add more sysctls.

Changed commit message.