https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=193246

Bruce M Simpson <[email protected]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|---                         |Normal
                URL|                            |https://reviews.freebsd.org
                   |                            |/D55031
           Severity|Affects Many People         |Affects Some People

--- Comment #20 from Bruce M Simpson <[email protected]> ---
Just regained access after nearly a decade. There was some personal tragedy
involved.

The resolution for this is being tracked at: https://reviews.freebsd.org/D55031

With a minor errno correction at: https://reviews.freebsd.org/D55233

TL;DR This is purely an optional extension. It is not and never was a bug. The
OpenJDK and other software started relying on it over the years, but it really
doesn't fit with how the socket APIs involved were intended to operate, at all.
It largely stems from uncertainty about the extent of IPv4-mapped address
support, as tuexen has indicated.

The cheap fix is to just do what Apple did and just pass it through to netinet
for simple ASM-mode adds/joins, which is the most common multicast use case
anyway, and it's usually for local link scoped groups.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.

Reply via email to